SlideShare uma empresa Scribd logo
1 de 70
Baixar para ler offline
Industrial Standards, Computer Algebra,
                               and Formal Verification

                    Dominik Dietrich              Lutz Schr¨der
                                                           o             Ewaryst Schulz

                                               DFKI Bremen, Germany
                                                ewaryst.schulz@dfki.de


          20th International Workshop on Algebraic Development Techniques
                              Schloss Etelsen, Germany
                                    4th July 2010




Industrial Standards, and Formal Verification                                  German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                                         for Artificial Intelligence
The Flange


  A CAD design of a flange-bolt-gasket system.




Industrial Standards, and Formal Verification    German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                           for Artificial Intelligence
The Industrial Standard EN 1591


  A standard for gasketed circular
  flange connections
  The standard consists of
    Applicability and basic
    assumptions
      Nomenclature
      Calculation method
  The calculation method assures the
  impermeability and mechanical
  strength of the flange-bolt-gasket
  system.


Industrial Standards, and Formal Verification   German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                          for Artificial Intelligence
The Industrial Standard EN 1591


  A standard for gasketed circular
  flange connections
  The standard consists of
    Applicability and basic
    assumptions
      Nomenclature
      Calculation method
  The calculation method assures the
  impermeability and mechanical
  strength of the flange-bolt-gasket
  system.


Industrial Standards, and Formal Verification   German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                          for Artificial Intelligence
The Industrial Standard EN 1591


  A standard for gasketed circular
  flange connections
  The standard consists of
    Applicability and basic
    assumptions
      Nomenclature
      Calculation method
  The calculation method assures the
  impermeability and mechanical
  strength of the flange-bolt-gasket
  system.


Industrial Standards, and Formal Verification   German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                          for Artificial Intelligence
The Industrial Standard EN 1591


  A standard for gasketed circular
  flange connections
  The standard consists of
    Applicability and basic
    assumptions
      Nomenclature
      Calculation method
  The calculation method assures the
  impermeability and mechanical
  strength of the flange-bolt-gasket
  system.


Industrial Standards, and Formal Verification   German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                          for Artificial Intelligence
The Industrial Standard EN 1591
    cont.

  The input parameters to the calculation method
      Flange data, e.g., dimensions and material constants




      Mounting data such as screw tightening method
      Data for operating states such as pressure and temperature
Industrial Standards, and Formal Verification           German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                  for Artificial Intelligence
The Industrial Standard EN 1591
    cont.

  The input parameters to the calculation method
      Flange data, e.g., dimensions and material constants




      Mounting data such as screw tightening method
      Data for operating states such as pressure and temperature
Industrial Standards, and Formal Verification           German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                  for Artificial Intelligence
The Industrial Standard EN 1591
    cont.

  The input parameters to the calculation method
      Flange data, e.g., dimensions and material constants




      Mounting data such as screw tightening method
      Data for operating states such as pressure and temperature
Industrial Standards, and Formal Verification           German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                  for Artificial Intelligence
Calculation Method and Iteration




Industrial Standards, and Formal Verification   German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                          for Artificial Intelligence
Calculation Method and Iteration




Industrial Standards, and Formal Verification   German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                          for Artificial Intelligence
Calculation Method and Iteration




Industrial Standards, and Formal Verification   German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                          for Artificial Intelligence
Calculation Method and Iteration




Industrial Standards, and Formal Verification   German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                          for Artificial Intelligence
Calculation Method and Maximize




Industrial Standards, and Formal Verification   German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                          for Artificial Intelligence
Calculation Method and Maximize




Industrial Standards, and Formal Verification   German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                          for Artificial Intelligence
Calculation Method and Maximize




Industrial Standards, and Formal Verification   German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                          for Artificial Intelligence
Calculation Method and Maximize




Industrial Standards, and Formal Verification   German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                          for Artificial Intelligence
Calculation Method
    and Computer Algebra

  The formulas occurring in the standard can be calculated using
      Standard real arithmetic
                                               √
      Real functions such as cos,              n   , etc.
      Special functions such as maximize
      Control structures such as conditional statements and iteration
  Use a computer algebra system for the calculations.




Industrial Standards, and Formal Verification                German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                       for Artificial Intelligence
Calculation Method
    and Computer Algebra

  The formulas occurring in the standard can be calculated using
      Standard real arithmetic
                                               √
      Real functions such as cos,              n   , etc.
      Special functions such as maximize
      Control structures such as conditional statements and iteration
  Use a computer algebra system for the calculations.




Industrial Standards, and Formal Verification                German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                       for Artificial Intelligence
Calculation Method
    and Computer Algebra

  The formulas occurring in the standard can be calculated using
      Standard real arithmetic
                                               √
      Real functions such as cos,              n   , etc.
      Special functions such as maximize
      Control structures such as conditional statements and iteration
  Use a computer algebra system for the calculations.




Industrial Standards, and Formal Verification                German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                       for Artificial Intelligence
Calculation Method
    and Computer Algebra

  The formulas occurring in the standard can be calculated using
      Standard real arithmetic
                                               √
      Real functions such as cos,              n   , etc.
      Special functions such as maximize
      Control structures such as conditional statements and iteration
  Use a computer algebra system for the calculations.




Industrial Standards, and Formal Verification                German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                       for Artificial Intelligence
Calculation Method
    and Computer Algebra

  The formulas occurring in the standard can be calculated using
      Standard real arithmetic
                                               √
      Real functions such as cos,              n   , etc.
      Special functions such as maximize
      Control structures such as conditional statements and iteration
  Use a computer algebra system for the calculations.




Industrial Standards, and Formal Verification                German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                       for Artificial Intelligence
Formal Verification


  Correctness of calculations crucial for application to safety critical
  environments
      CASs do not provide justifications of calculations
      x
      x   simplifies to 1 in the Reduce CAS
  Results of the CAS can be formally verified
      One can generate lemmas from CAS result to be proved
      Checking is easier than finding




Industrial Standards, and Formal Verification              German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                     for Artificial Intelligence
Formal Verification


  Correctness of calculations crucial for application to safety critical
  environments
      CASs do not provide justifications of calculations
      x
      x   simplifies to 1 in the Reduce CAS
  Results of the CAS can be formally verified
      One can generate lemmas from CAS result to be proved
      Checking is easier than finding




Industrial Standards, and Formal Verification              German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                     for Artificial Intelligence
Formal Verification


  Correctness of calculations crucial for application to safety critical
  environments
      CASs do not provide justifications of calculations
      x
      x   simplifies to 1 in the Reduce CAS
  Results of the CAS can be formally verified
      One can generate lemmas from CAS result to be proved
      Checking is easier than finding




Industrial Standards, and Formal Verification              German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                     for Artificial Intelligence
Formal Verification


  Correctness of calculations crucial for application to safety critical
  environments
      CASs do not provide justifications of calculations
      x
      x   simplifies to 1 in the Reduce CAS
  Results of the CAS can be formally verified
      One can generate lemmas from CAS result to be proved
      Checking is easier than finding




Industrial Standards, and Formal Verification              German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                     for Artificial Intelligence
Hets- the Heterogeneous Tool Set




Industrial Standards, and Formal Verification   German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                          for Artificial Intelligence
Specification Language CSL


  Design goals of CSL
      Formal specification of the calculation method
      Specification of assignments in an arbitrary order, but:
      We require assignments to be unique and sortable w.r.t. the
      dependency order
      Generic interface to CAS

  Translation to CAS
      Suitably ordered assignments together with control structures form an
      imperative program
      Constants depending on constants which were modified are recomputed
      Executing the program using CAS yields a symbolic valuation

Industrial Standards, and Formal Verification            German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                   for Artificial Intelligence
Specification Language CSL


  Design goals of CSL
      Formal specification of the calculation method
      Specification of assignments in an arbitrary order, but:
      We require assignments to be unique and sortable w.r.t. the
      dependency order
      Generic interface to CAS

  Translation to CAS
      Suitably ordered assignments together with control structures form an
      imperative program
      Constants depending on constants which were modified are recomputed
      Executing the program using CAS yields a symbolic valuation

Industrial Standards, and Formal Verification            German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                   for Artificial Intelligence
Specification Language CSL


  Design goals of CSL
      Formal specification of the calculation method
      Specification of assignments in an arbitrary order, but:
      We require assignments to be unique and sortable w.r.t. the
      dependency order
      Generic interface to CAS

  Translation to CAS
      Suitably ordered assignments together with control structures form an
      imperative program
      Constants depending on constants which were modified are recomputed
      Executing the program using CAS yields a symbolic valuation

Industrial Standards, and Formal Verification            German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                   for Artificial Intelligence
Specification Language CSL


  Design goals of CSL
      Formal specification of the calculation method
      Specification of assignments in an arbitrary order, but:
      We require assignments to be unique and sortable w.r.t. the
      dependency order
      Generic interface to CAS

  Translation to CAS
      Suitably ordered assignments together with control structures form an
      imperative program
      Constants depending on constants which were modified are recomputed
      Executing the program using CAS yields a symbolic valuation

Industrial Standards, and Formal Verification            German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                   for Artificial Intelligence
Specification Language CSL


  Design goals of CSL
      Formal specification of the calculation method
      Specification of assignments in an arbitrary order, but:
      We require assignments to be unique and sortable w.r.t. the
      dependency order
      Generic interface to CAS

  Translation to CAS
      Suitably ordered assignments together with control structures form an
      imperative program
      Constants depending on constants which were modified are recomputed
      Executing the program using CAS yields a symbolic valuation

Industrial Standards, and Formal Verification            German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                   for Artificial Intelligence
Specification Language CSL


  Design goals of CSL
      Formal specification of the calculation method
      Specification of assignments in an arbitrary order, but:
      We require assignments to be unique and sortable w.r.t. the
      dependency order
      Generic interface to CAS

  Translation to CAS
      Suitably ordered assignments together with control structures form an
      imperative program
      Constants depending on constants which were modified are recomputed
      Executing the program using CAS yields a symbolic valuation

Industrial Standards, and Formal Verification            German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                   for Artificial Intelligence
A Little CSL Example


  Calculating a root of cos using Newton’s Method

     The CSL specification
       y := cos(x) %(A)%
       z := sin(x) %(B)%
       x := 10 %(C)%
       repeat
         x := x + y/z %(D)%
       until abs(y) < 0.001



  The translation yields this program:
  C;A;B;repeat D;A;B; until abs(y) < 0.001

Industrial Standards, and Formal Verification        German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                               for Artificial Intelligence
A Little CSL Example


  Calculating a root of cos using Newton’s Method

     The CSL specification                      Building the Dependency Graph
       y := cos(x) %(A)%
       z := sin(x) %(B)%
       x := 10 %(C)%                                         x
       repeat                                            A
         x := x + y/z %(D)%                          y
       until abs(y) < 0.001



  The translation yields this program:
  C;A;B;repeat D;A;B; until abs(y) < 0.001

Industrial Standards, and Formal Verification                     German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                            for Artificial Intelligence
A Little CSL Example


  Calculating a root of cos using Newton’s Method

     The CSL specification                      Building the Dependency Graph
       y := cos(x) %(A)%
       z := sin(x) %(B)%
       x := 10 %(C)%                                         x
       repeat                                            A         B
         x := x + y/z %(D)%                          y                  z
       until abs(y) < 0.001



  The translation yields this program:
  C;A;B;repeat D;A;B; until abs(y) < 0.001

Industrial Standards, and Formal Verification                     German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                            for Artificial Intelligence
A Little CSL Example


  Calculating a root of cos using Newton’s Method

     The CSL specification                      Building the Dependency Graph
       y := cos(x) %(A)%
       z := sin(x) %(B)%                                         C
       x := 10 %(C)%                                         x
       repeat                                            A             B
         x := x + y/z %(D)%                          y                      z
       until abs(y) < 0.001



  The translation yields this program:
  C;A;B;repeat D;A;B; until abs(y) < 0.001

Industrial Standards, and Formal Verification                         German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                                for Artificial Intelligence
A Little CSL Example


  Calculating a root of cos using Newton’s Method

     The CSL specification                      Building the Dependency Graph
       y := cos(x) %(A)%
       z := sin(x) %(B)%                                         C
       x := 10 %(C)%                                         x
       repeat                                            A             B
         x := x + y/z %(D)%                          y                      z
       until abs(y) < 0.001
                                                                 D


  The translation yields this program:
  C;A;B;repeat D;A;B; until abs(y) < 0.001

Industrial Standards, and Formal Verification                         German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                                for Artificial Intelligence
A Little CSL Example


  Calculating a root of cos using Newton’s Method

     The CSL specification                      Building the Dependency Graph
       y := cos(x) %(A)%
       z := sin(x) %(B)%                                         C
       x := 10 %(C)%                                         x
       repeat                                            A             B
         x := x + y/z %(D)%                          y                      z
       until abs(y) < 0.001
                                                                 D


  The translation yields this program:
  C;A;B;repeat D;A;B; until abs(y) < 0.001

Industrial Standards, and Formal Verification                         German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                                for Artificial Intelligence
Verified CAS


  Verification Points in CSL
      are positions of subterms of CSL statements
      Evaluating a such marked term produces a verification condition
      The CAS result is extended by a list of verification conditions
      Use Hets to prove verification conditions

  Specifying CAS program semantics in HasCASL
      Standard interpretation of programs as state transformers
      Properties of algorithms specified in CSL can be verified




Industrial Standards, and Formal Verification            German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                   for Artificial Intelligence
Verified CAS


  Verification Points in CSL
      are positions of subterms of CSL statements
      Evaluating a such marked term produces a verification condition
      The CAS result is extended by a list of verification conditions
      Use Hets to prove verification conditions

  Specifying CAS program semantics in HasCASL
      Standard interpretation of programs as state transformers
      Properties of algorithms specified in CSL can be verified




Industrial Standards, and Formal Verification            German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                   for Artificial Intelligence
Verified CAS


  Verification Points in CSL
      are positions of subterms of CSL statements
      Evaluating a such marked term produces a verification condition
      The CAS result is extended by a list of verification conditions
      Use Hets to prove verification conditions

  Specifying CAS program semantics in HasCASL
      Standard interpretation of programs as state transformers
      Properties of algorithms specified in CSL can be verified




Industrial Standards, and Formal Verification            German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                   for Artificial Intelligence
Verified CAS


  Verification Points in CSL
      are positions of subterms of CSL statements
      Evaluating a such marked term produces a verification condition
      The CAS result is extended by a list of verification conditions
      Use Hets to prove verification conditions

  Specifying CAS program semantics in HasCASL
      Standard interpretation of programs as state transformers
      Properties of algorithms specified in CSL can be verified




Industrial Standards, and Formal Verification            German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                   for Artificial Intelligence
Verified CAS


  Verification Points in CSL
      are positions of subterms of CSL statements
      Evaluating a such marked term produces a verification condition
      The CAS result is extended by a list of verification conditions
      Use Hets to prove verification conditions

  Specifying CAS program semantics in HasCASL
      Standard interpretation of programs as state transformers
      Properties of algorithms specified in CSL can be verified




Industrial Standards, and Formal Verification            German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                   for Artificial Intelligence
Verified CAS


  Verification Points in CSL
      are positions of subterms of CSL statements
      Evaluating a such marked term produces a verification condition
      The CAS result is extended by a list of verification conditions
      Use Hets to prove verification conditions

  Specifying CAS program semantics in HasCASL
      Standard interpretation of programs as state transformers
      Properties of algorithms specified in CSL can be verified




Industrial Standards, and Formal Verification            German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                   for Artificial Intelligence
Example


  Verifying a result from the CAS

   A CAS program                               We set verification point at maximize
                                               position → maximize(t, x) is marked
      .
      . Environment = σ
      .                                        CAS computes this expression in context σ
   y := maximize(t, x)                         and retuns result r
      .
      .
      .                                        Apply substitution σ to t and obtain t
                                               We produce the verification condition
                                               maximize(t , x) = r
                                               Translate this equality to HasCASL for
                                               proving



Industrial Standards, and Formal Verification                         German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                                for Artificial Intelligence
Example


  Verifying a result from the CAS

   A CAS program                               We set verification point at maximize
                                               position → maximize(t, x) is marked
      .
      . Environment = σ
      .                                        CAS computes this expression in context σ
   y := maximize(t, x)                         and retuns result r
      .
      .
      .                                        Apply substitution σ to t and obtain t
                                               We produce the verification condition
                                               maximize(t , x) = r
                                               Translate this equality to HasCASL for
                                               proving



Industrial Standards, and Formal Verification                         German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                                for Artificial Intelligence
Example


  Verifying a result from the CAS

   A CAS program                               We set verification point at maximize
                                               position → maximize(t, x) is marked
      .
      . Environment = σ
      .                                        CAS computes this expression in context σ
   y := maximize(t, x)                         and retuns result r
      .
      .
      .                                        Apply substitution σ to t and obtain t
                                               We produce the verification condition
                                               maximize(t , x) = r
                                               Translate this equality to HasCASL for
                                               proving



Industrial Standards, and Formal Verification                         German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                                for Artificial Intelligence
Example


  Verifying a result from the CAS

   A CAS program                               We set verification point at maximize
                                               position → maximize(t, x) is marked
      .
      . Environment = σ
      .                                        CAS computes this expression in context σ
   y := maximize(t, x)                         and retuns result r
      .
      .
      .                                        Apply substitution σ to t and obtain t
                                               We produce the verification condition
                                               maximize(t , x) = r
                                               Translate this equality to HasCASL for
                                               proving



Industrial Standards, and Formal Verification                         German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                                for Artificial Intelligence
Example


  Verifying a result from the CAS

   A CAS program                               We set verification point at maximize
                                               position → maximize(t, x) is marked
      .
      . Environment = σ
      .                                        CAS computes this expression in context σ
   y := maximize(t, x)                         and retuns result r
      .
      .
      .                                        Apply substitution σ to t and obtain t
                                               We produce the verification condition
                                               maximize(t , x) = r
                                               Translate this equality to HasCASL for
                                               proving



Industrial Standards, and Formal Verification                         German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                                for Artificial Intelligence
Example


  Verifying a result from the CAS

   A CAS program                               We set verification point at maximize
                                               position → maximize(t, x) is marked
      .
      . Environment = σ
      .                                        CAS computes this expression in context σ
   y := maximize(t, x)                         and retuns result r
      .
      .
      .                                        Apply substitution σ to t and obtain t
                                               We produce the verification condition
                                               maximize(t , x) = r
                                               Translate this equality to HasCASL for
                                               proving



Industrial Standards, and Formal Verification                         German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                                for Artificial Intelligence
CSL, CAS and Hets


  CSL and the Hets Logic Graph

                                                   Logic Graph
   Isabelle Prover                             Isabelle

                                                          HasCASL




Industrial Standards, and Formal Verification                        German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                               for Artificial Intelligence
CSL, CAS and Hets


  CSL and the Hets Logic Graph

                                                   Logic Graph
   Isabelle Prover                             Isabelle

                                                          HasCASL

                                                CSL




Industrial Standards, and Formal Verification                        German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                               for Artificial Intelligence
CSL, CAS and Hets


  CSL and the Hets Logic Graph

                                                   Logic Graph
   Isabelle Prover                             Isabelle

                                                          HasCASL
          Reduce
                                                CSL
         Maxima

     Mathematica                                  CAS Interface


Industrial Standards, and Formal Verification                        German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                               for Artificial Intelligence
CSL, CAS and Hets


  CSL and the Hets Logic Graph

                                                   Logic Graph
   Isabelle Prover                             Isabelle

                                                          HasCASL
          Reduce
                                                CSL
         Maxima

     Mathematica                                  CAS Interface


Industrial Standards, and Formal Verification                        German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                               for Artificial Intelligence
CSL, CAS and Hets cont.


  The CSL institution
      Signatures are collections of real constants and functions over the reals
      Sentences are program statements or first order formulas in an extended
      theory of the reals augmented by the signature
      Models are program states, i.e., symbolic valuations
      A state satisfies a program if it terminates successfully
      A state satisfies a formula φ if φ holds under this valuation




Industrial Standards, and Formal Verification              German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                     for Artificial Intelligence
CSL, CAS and Hets cont.


  The CSL institution
      Signatures are collections of real constants and functions over the reals
      Sentences are program statements or first order formulas in an extended
      theory of the reals augmented by the signature
      Models are program states, i.e., symbolic valuations
      A state satisfies a program if it terminates successfully
      A state satisfies a formula φ if φ holds under this valuation




Industrial Standards, and Formal Verification              German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                     for Artificial Intelligence
CSL, CAS and Hets cont.


  The CSL institution
      Signatures are collections of real constants and functions over the reals
      Sentences are program statements or first order formulas in an extended
      theory of the reals augmented by the signature
      Models are program states, i.e., symbolic valuations
      A state satisfies a program if it terminates successfully
      A state satisfies a formula φ if φ holds under this valuation




Industrial Standards, and Formal Verification              German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                     for Artificial Intelligence
CSL, CAS and Hets cont.


  The CSL institution
      Signatures are collections of real constants and functions over the reals
      Sentences are program statements or first order formulas in an extended
      theory of the reals augmented by the signature
      Models are program states, i.e., symbolic valuations
      A state satisfies a program if it terminates successfully
      A state satisfies a formula φ if φ holds under this valuation




Industrial Standards, and Formal Verification              German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                     for Artificial Intelligence
CSL, CAS and Hets cont.


  The CSL institution
      Signatures are collections of real constants and functions over the reals
      Sentences are program statements or first order formulas in an extended
      theory of the reals augmented by the signature
      Models are program states, i.e., symbolic valuations
      A state satisfies a program if it terminates successfully
      A state satisfies a formula φ if φ holds under this valuation




Industrial Standards, and Formal Verification              German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                     for Artificial Intelligence
Summary and Outlook


      Specification language CSL for industrial standards
      Synthesis of programs for generic CAS interface
      Verification Points for local verification of CAS result
      Integration of CSL and CAS interface in Hets
      Specification of CSL semantics in HasCASL
      Relating CSL to HasCASL by theoroidal comorphism
  Benefit from symbolic character of CAS computations
      Using CAS to simplify CSL specifications for partial instantiations or
      given set of additional assumptions
      Replace special functions by closed solutions found by the CAS
      Finding instantiations for underspecified specifications, e.g., number of
      bolts needed for flange to satisfy standard
Industrial Standards, and Formal Verification             German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                    for Artificial Intelligence
Summary and Outlook


      Specification language CSL for industrial standards
      Synthesis of programs for generic CAS interface
      Verification Points for local verification of CAS result
      Integration of CSL and CAS interface in Hets
      Specification of CSL semantics in HasCASL
      Relating CSL to HasCASL by theoroidal comorphism
  Benefit from symbolic character of CAS computations
      Using CAS to simplify CSL specifications for partial instantiations or
      given set of additional assumptions
      Replace special functions by closed solutions found by the CAS
      Finding instantiations for underspecified specifications, e.g., number of
      bolts needed for flange to satisfy standard
Industrial Standards, and Formal Verification             German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                    for Artificial Intelligence
Summary and Outlook


      Specification language CSL for industrial standards
      Synthesis of programs for generic CAS interface
      Verification Points for local verification of CAS result
      Integration of CSL and CAS interface in Hets
      Specification of CSL semantics in HasCASL
      Relating CSL to HasCASL by theoroidal comorphism
  Benefit from symbolic character of CAS computations
      Using CAS to simplify CSL specifications for partial instantiations or
      given set of additional assumptions
      Replace special functions by closed solutions found by the CAS
      Finding instantiations for underspecified specifications, e.g., number of
      bolts needed for flange to satisfy standard
Industrial Standards, and Formal Verification             German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                    for Artificial Intelligence
Summary and Outlook


      Specification language CSL for industrial standards
      Synthesis of programs for generic CAS interface
      Verification Points for local verification of CAS result
      Integration of CSL and CAS interface in Hets
      Specification of CSL semantics in HasCASL
      Relating CSL to HasCASL by theoroidal comorphism
  Benefit from symbolic character of CAS computations
      Using CAS to simplify CSL specifications for partial instantiations or
      given set of additional assumptions
      Replace special functions by closed solutions found by the CAS
      Finding instantiations for underspecified specifications, e.g., number of
      bolts needed for flange to satisfy standard
Industrial Standards, and Formal Verification             German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                    for Artificial Intelligence
Summary and Outlook


      Specification language CSL for industrial standards
      Synthesis of programs for generic CAS interface
      Verification Points for local verification of CAS result
      Integration of CSL and CAS interface in Hets
      Specification of CSL semantics in HasCASL
      Relating CSL to HasCASL by theoroidal comorphism
  Benefit from symbolic character of CAS computations
      Using CAS to simplify CSL specifications for partial instantiations or
      given set of additional assumptions
      Replace special functions by closed solutions found by the CAS
      Finding instantiations for underspecified specifications, e.g., number of
      bolts needed for flange to satisfy standard
Industrial Standards, and Formal Verification             German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                    for Artificial Intelligence
Summary and Outlook


      Specification language CSL for industrial standards
      Synthesis of programs for generic CAS interface
      Verification Points for local verification of CAS result
      Integration of CSL and CAS interface in Hets
      Specification of CSL semantics in HasCASL
      Relating CSL to HasCASL by theoroidal comorphism
  Benefit from symbolic character of CAS computations
      Using CAS to simplify CSL specifications for partial instantiations or
      given set of additional assumptions
      Replace special functions by closed solutions found by the CAS
      Finding instantiations for underspecified specifications, e.g., number of
      bolts needed for flange to satisfy standard
Industrial Standards, and Formal Verification             German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                    for Artificial Intelligence
Summary and Outlook


      Specification language CSL for industrial standards
      Synthesis of programs for generic CAS interface
      Verification Points for local verification of CAS result
      Integration of CSL and CAS interface in Hets
      Specification of CSL semantics in HasCASL
      Relating CSL to HasCASL by theoroidal comorphism
  Benefit from symbolic character of CAS computations
      Using CAS to simplify CSL specifications for partial instantiations or
      given set of additional assumptions
      Replace special functions by closed solutions found by the CAS
      Finding instantiations for underspecified specifications, e.g., number of
      bolts needed for flange to satisfy standard
Industrial Standards, and Formal Verification             German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                    for Artificial Intelligence
Summary and Outlook


      Specification language CSL for industrial standards
      Synthesis of programs for generic CAS interface
      Verification Points for local verification of CAS result
      Integration of CSL and CAS interface in Hets
      Specification of CSL semantics in HasCASL
      Relating CSL to HasCASL by theoroidal comorphism
  Benefit from symbolic character of CAS computations
      Using CAS to simplify CSL specifications for partial instantiations or
      given set of additional assumptions
      Replace special functions by closed solutions found by the CAS
      Finding instantiations for underspecified specifications, e.g., number of
      bolts needed for flange to satisfy standard
Industrial Standards, and Formal Verification             German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                    for Artificial Intelligence
Summary and Outlook


      Specification language CSL for industrial standards
      Synthesis of programs for generic CAS interface
      Verification Points for local verification of CAS result
      Integration of CSL and CAS interface in Hets
      Specification of CSL semantics in HasCASL
      Relating CSL to HasCASL by theoroidal comorphism
  Benefit from symbolic character of CAS computations
      Using CAS to simplify CSL specifications for partial instantiations or
      given set of additional assumptions
      Replace special functions by closed solutions found by the CAS
      Finding instantiations for underspecified specifications, e.g., number of
      bolts needed for flange to satisfy standard
Industrial Standards, and Formal Verification             German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                    for Artificial Intelligence
Summary and Outlook


      Specification language CSL for industrial standards
      Synthesis of programs for generic CAS interface
      Verification Points for local verification of CAS result
      Integration of CSL and CAS interface in Hets
      Specification of CSL semantics in HasCASL
      Relating CSL to HasCASL by theoroidal comorphism
  Benefit from symbolic character of CAS computations
      Using CAS to simplify CSL specifications for partial instantiations or
      given set of additional assumptions
      Replace special functions by closed solutions found by the CAS
      Finding instantiations for underspecified specifications, e.g., number of
      bolts needed for flange to satisfy standard
Industrial Standards, and Formal Verification             German Research Center
D. Dietrich, L. Schr¨der, E. Schulz
                    o                                    for Artificial Intelligence

Mais conteúdo relacionado

Destaque

Category management in farmacia
Category management in farmaciaCategory management in farmacia
Category management in farmaciaAlberto Barlassina
 
Visual marketing e social media con pinterest e instagram
Visual marketing e social media con pinterest e instagramVisual marketing e social media con pinterest e instagram
Visual marketing e social media con pinterest e instagramCino Wang Platania
 
4f14.4012.file
4f14.4012.file4f14.4012.file
4f14.4012.fileimartini
 
Turn audience into customers with social contests
Turn audience into customers with social contestsTurn audience into customers with social contests
Turn audience into customers with social contestsAndrea D'Ottavio
 
Fairs and exhibitions.. with Veroni
Fairs and exhibitions.. with VeroniFairs and exhibitions.. with Veroni
Fairs and exhibitions.. with VeroniGianfranco Delfini
 
Presentazione Nardelli Luxury Nv 1
Presentazione Nardelli Luxury Nv 1Presentazione Nardelli Luxury Nv 1
Presentazione Nardelli Luxury Nv 1THE GHOST TEAM
 
Category management (by Fabio Bullita)
Category management (by Fabio Bullita)Category management (by Fabio Bullita)
Category management (by Fabio Bullita)Fabio Bullita
 
Space management e Visual Merchandising in ambito Retail (by Fabio Bullita)
Space management e Visual Merchandising in ambito Retail (by Fabio Bullita)Space management e Visual Merchandising in ambito Retail (by Fabio Bullita)
Space management e Visual Merchandising in ambito Retail (by Fabio Bullita)Fabio Bullita
 
Il Visual Merchandising nella distribuzione specializzata.
Il Visual Merchandising nella distribuzione specializzata.Il Visual Merchandising nella distribuzione specializzata.
Il Visual Merchandising nella distribuzione specializzata.DORA BINNELLA studio
 

Destaque (14)

Presentazione myfruitmail l'ortofrutta nel dettaglio
Presentazione myfruitmail l'ortofrutta nel dettaglioPresentazione myfruitmail l'ortofrutta nel dettaglio
Presentazione myfruitmail l'ortofrutta nel dettaglio
 
Inbound marketing (Marchet)
Inbound marketing (Marchet)Inbound marketing (Marchet)
Inbound marketing (Marchet)
 
Category management in farmacia
Category management in farmaciaCategory management in farmacia
Category management in farmacia
 
Visual marketing e social media con pinterest e instagram
Visual marketing e social media con pinterest e instagramVisual marketing e social media con pinterest e instagram
Visual marketing e social media con pinterest e instagram
 
4f14.4012.file
4f14.4012.file4f14.4012.file
4f14.4012.file
 
Regole di visual merchandising
Regole di visual merchandisingRegole di visual merchandising
Regole di visual merchandising
 
Lezione corso Visual merchandising
Lezione corso Visual merchandisingLezione corso Visual merchandising
Lezione corso Visual merchandising
 
Turn audience into customers with social contests
Turn audience into customers with social contestsTurn audience into customers with social contests
Turn audience into customers with social contests
 
Fairs and exhibitions.. with Veroni
Fairs and exhibitions.. with VeroniFairs and exhibitions.. with Veroni
Fairs and exhibitions.. with Veroni
 
Presentazione Nardelli Luxury Nv 1
Presentazione Nardelli Luxury Nv 1Presentazione Nardelli Luxury Nv 1
Presentazione Nardelli Luxury Nv 1
 
Category management (by Fabio Bullita)
Category management (by Fabio Bullita)Category management (by Fabio Bullita)
Category management (by Fabio Bullita)
 
Space management e Visual Merchandising in ambito Retail (by Fabio Bullita)
Space management e Visual Merchandising in ambito Retail (by Fabio Bullita)Space management e Visual Merchandising in ambito Retail (by Fabio Bullita)
Space management e Visual Merchandising in ambito Retail (by Fabio Bullita)
 
Il Visual Merchandising nella distribuzione specializzata.
Il Visual Merchandising nella distribuzione specializzata.Il Visual Merchandising nella distribuzione specializzata.
Il Visual Merchandising nella distribuzione specializzata.
 
LinkedIn per giornalisti
LinkedIn per giornalistiLinkedIn per giornalisti
LinkedIn per giornalisti
 

Semelhante a Industrial Standards, Computer Algebra, and Formal Veri cation

FASE 2011 - Formalizing and Operationalizing Industrial Standards
FASE 2011 - Formalizing and Operationalizing Industrial StandardsFASE 2011 - Formalizing and Operationalizing Industrial Standards
FASE 2011 - Formalizing and Operationalizing Industrial StandardsEwaryst Schulz
 
A C Product Design
A C Product DesignA C Product Design
A C Product DesignJoe Harsany
 
Career guidance in engineering lecture #2 ETC, ELE, Instru
Career guidance in engineering lecture #2 ETC, ELE, InstruCareer guidance in engineering lecture #2 ETC, ELE, Instru
Career guidance in engineering lecture #2 ETC, ELE, InstruDr. BP Joshi
 
CV SMT PROCESS ENG JOSE G VELEZ RESUME-LINKEDIN
CV SMT PROCESS ENG JOSE G VELEZ RESUME-LINKEDINCV SMT PROCESS ENG JOSE G VELEZ RESUME-LINKEDIN
CV SMT PROCESS ENG JOSE G VELEZ RESUME-LINKEDINJOSE GERARDO VELEZ
 
Automatic ICD-10 Code Assignment to Consultations
Automatic ICD-10 Code Assignment to ConsultationsAutomatic ICD-10 Code Assignment to Consultations
Automatic ICD-10 Code Assignment to ConsultationsDatabricks
 
Common Criteria service overview for Developers - jtsec a CC consultancy company
Common Criteria service overview for Developers - jtsec a CC consultancy companyCommon Criteria service overview for Developers - jtsec a CC consultancy company
Common Criteria service overview for Developers - jtsec a CC consultancy companyJavier Tallón
 
Engineering: a key knowledge for boosting the competitiveness of biotechnolog...
Engineering: a key knowledge for boosting the competitiveness of biotechnolog...Engineering: a key knowledge for boosting the competitiveness of biotechnolog...
Engineering: a key knowledge for boosting the competitiveness of biotechnolog...Francisco J. Menéndez González
 
MayorTek Technical Services - Overview
MayorTek Technical Services - OverviewMayorTek Technical Services - Overview
MayorTek Technical Services - OverviewMarco Mayor
 
Bid scoin ohbm2019-poster
Bid scoin ohbm2019-posterBid scoin ohbm2019-poster
Bid scoin ohbm2019-posterMarcel Zwiers
 
Engineering Assignment
Engineering AssignmentEngineering Assignment
Engineering AssignmentNeilOw87
 
Slides4lucid
Slides4lucidSlides4lucid
Slides4lucidroysolay
 
Slides4lucid
Slides4lucidSlides4lucid
Slides4lucidroysolay
 
Cigdem gencel persistence in poor estimating in software engineering- whys an...
Cigdem gencel persistence in poor estimating in software engineering- whys an...Cigdem gencel persistence in poor estimating in software engineering- whys an...
Cigdem gencel persistence in poor estimating in software engineering- whys an...oxwocs
 
Engineering drawing (introduction of engineering drawing) lesson 1
Engineering drawing (introduction of engineering drawing) lesson 1Engineering drawing (introduction of engineering drawing) lesson 1
Engineering drawing (introduction of engineering drawing) lesson 1hermiraguilar
 
chap01_computer_programming.ppt
chap01_computer_programming.pptchap01_computer_programming.ppt
chap01_computer_programming.pptssuser1ffb13
 

Semelhante a Industrial Standards, Computer Algebra, and Formal Veri cation (20)

FASE 2011 - Formalizing and Operationalizing Industrial Standards
FASE 2011 - Formalizing and Operationalizing Industrial StandardsFASE 2011 - Formalizing and Operationalizing Industrial Standards
FASE 2011 - Formalizing and Operationalizing Industrial Standards
 
A C Product Design
A C Product DesignA C Product Design
A C Product Design
 
Career guidance in engineering lecture #2 ETC, ELE, Instru
Career guidance in engineering lecture #2 ETC, ELE, InstruCareer guidance in engineering lecture #2 ETC, ELE, Instru
Career guidance in engineering lecture #2 ETC, ELE, Instru
 
CURRICULUM VITAE
CURRICULUM VITAECURRICULUM VITAE
CURRICULUM VITAE
 
CV SMT PROCESS ENG JOSE G VELEZ RESUME-LINKEDIN
CV SMT PROCESS ENG JOSE G VELEZ RESUME-LINKEDINCV SMT PROCESS ENG JOSE G VELEZ RESUME-LINKEDIN
CV SMT PROCESS ENG JOSE G VELEZ RESUME-LINKEDIN
 
Automatic ICD-10 Code Assignment to Consultations
Automatic ICD-10 Code Assignment to ConsultationsAutomatic ICD-10 Code Assignment to Consultations
Automatic ICD-10 Code Assignment to Consultations
 
Common Criteria service overview for Developers - jtsec a CC consultancy company
Common Criteria service overview for Developers - jtsec a CC consultancy companyCommon Criteria service overview for Developers - jtsec a CC consultancy company
Common Criteria service overview for Developers - jtsec a CC consultancy company
 
Industrial Sensory Data Analysis
Industrial Sensory Data AnalysisIndustrial Sensory Data Analysis
Industrial Sensory Data Analysis
 
Engineering: a key knowledge for boosting the competitiveness of biotechnolog...
Engineering: a key knowledge for boosting the competitiveness of biotechnolog...Engineering: a key knowledge for boosting the competitiveness of biotechnolog...
Engineering: a key knowledge for boosting the competitiveness of biotechnolog...
 
Industrial Sensory Data Analytics
Industrial Sensory Data AnalyticsIndustrial Sensory Data Analytics
Industrial Sensory Data Analytics
 
MayorTek Technical Services - Overview
MayorTek Technical Services - OverviewMayorTek Technical Services - Overview
MayorTek Technical Services - Overview
 
Bid scoin ohbm2019-poster
Bid scoin ohbm2019-posterBid scoin ohbm2019-poster
Bid scoin ohbm2019-poster
 
Engineering Assignment
Engineering AssignmentEngineering Assignment
Engineering Assignment
 
Slides4lucid
Slides4lucidSlides4lucid
Slides4lucid
 
Slides4lucid
Slides4lucidSlides4lucid
Slides4lucid
 
Cigdem gencel persistence in poor estimating in software engineering- whys an...
Cigdem gencel persistence in poor estimating in software engineering- whys an...Cigdem gencel persistence in poor estimating in software engineering- whys an...
Cigdem gencel persistence in poor estimating in software engineering- whys an...
 
Engineering drawing (introduction of engineering drawing) lesson 1
Engineering drawing (introduction of engineering drawing) lesson 1Engineering drawing (introduction of engineering drawing) lesson 1
Engineering drawing (introduction of engineering drawing) lesson 1
 
CST 2014
CST 2014CST 2014
CST 2014
 
Technology Scouting
Technology ScoutingTechnology Scouting
Technology Scouting
 
chap01_computer_programming.ppt
chap01_computer_programming.pptchap01_computer_programming.ppt
chap01_computer_programming.ppt
 

Industrial Standards, Computer Algebra, and Formal Veri cation

  • 1. Industrial Standards, Computer Algebra, and Formal Verification Dominik Dietrich Lutz Schr¨der o Ewaryst Schulz DFKI Bremen, Germany ewaryst.schulz@dfki.de 20th International Workshop on Algebraic Development Techniques Schloss Etelsen, Germany 4th July 2010 Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 2. The Flange A CAD design of a flange-bolt-gasket system. Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 3. The Industrial Standard EN 1591 A standard for gasketed circular flange connections The standard consists of Applicability and basic assumptions Nomenclature Calculation method The calculation method assures the impermeability and mechanical strength of the flange-bolt-gasket system. Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 4. The Industrial Standard EN 1591 A standard for gasketed circular flange connections The standard consists of Applicability and basic assumptions Nomenclature Calculation method The calculation method assures the impermeability and mechanical strength of the flange-bolt-gasket system. Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 5. The Industrial Standard EN 1591 A standard for gasketed circular flange connections The standard consists of Applicability and basic assumptions Nomenclature Calculation method The calculation method assures the impermeability and mechanical strength of the flange-bolt-gasket system. Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 6. The Industrial Standard EN 1591 A standard for gasketed circular flange connections The standard consists of Applicability and basic assumptions Nomenclature Calculation method The calculation method assures the impermeability and mechanical strength of the flange-bolt-gasket system. Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 7. The Industrial Standard EN 1591 cont. The input parameters to the calculation method Flange data, e.g., dimensions and material constants Mounting data such as screw tightening method Data for operating states such as pressure and temperature Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 8. The Industrial Standard EN 1591 cont. The input parameters to the calculation method Flange data, e.g., dimensions and material constants Mounting data such as screw tightening method Data for operating states such as pressure and temperature Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 9. The Industrial Standard EN 1591 cont. The input parameters to the calculation method Flange data, e.g., dimensions and material constants Mounting data such as screw tightening method Data for operating states such as pressure and temperature Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 10. Calculation Method and Iteration Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 11. Calculation Method and Iteration Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 12. Calculation Method and Iteration Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 13. Calculation Method and Iteration Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 14. Calculation Method and Maximize Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 15. Calculation Method and Maximize Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 16. Calculation Method and Maximize Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 17. Calculation Method and Maximize Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 18. Calculation Method and Computer Algebra The formulas occurring in the standard can be calculated using Standard real arithmetic √ Real functions such as cos, n , etc. Special functions such as maximize Control structures such as conditional statements and iteration Use a computer algebra system for the calculations. Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 19. Calculation Method and Computer Algebra The formulas occurring in the standard can be calculated using Standard real arithmetic √ Real functions such as cos, n , etc. Special functions such as maximize Control structures such as conditional statements and iteration Use a computer algebra system for the calculations. Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 20. Calculation Method and Computer Algebra The formulas occurring in the standard can be calculated using Standard real arithmetic √ Real functions such as cos, n , etc. Special functions such as maximize Control structures such as conditional statements and iteration Use a computer algebra system for the calculations. Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 21. Calculation Method and Computer Algebra The formulas occurring in the standard can be calculated using Standard real arithmetic √ Real functions such as cos, n , etc. Special functions such as maximize Control structures such as conditional statements and iteration Use a computer algebra system for the calculations. Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 22. Calculation Method and Computer Algebra The formulas occurring in the standard can be calculated using Standard real arithmetic √ Real functions such as cos, n , etc. Special functions such as maximize Control structures such as conditional statements and iteration Use a computer algebra system for the calculations. Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 23. Formal Verification Correctness of calculations crucial for application to safety critical environments CASs do not provide justifications of calculations x x simplifies to 1 in the Reduce CAS Results of the CAS can be formally verified One can generate lemmas from CAS result to be proved Checking is easier than finding Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 24. Formal Verification Correctness of calculations crucial for application to safety critical environments CASs do not provide justifications of calculations x x simplifies to 1 in the Reduce CAS Results of the CAS can be formally verified One can generate lemmas from CAS result to be proved Checking is easier than finding Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 25. Formal Verification Correctness of calculations crucial for application to safety critical environments CASs do not provide justifications of calculations x x simplifies to 1 in the Reduce CAS Results of the CAS can be formally verified One can generate lemmas from CAS result to be proved Checking is easier than finding Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 26. Formal Verification Correctness of calculations crucial for application to safety critical environments CASs do not provide justifications of calculations x x simplifies to 1 in the Reduce CAS Results of the CAS can be formally verified One can generate lemmas from CAS result to be proved Checking is easier than finding Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 27. Hets- the Heterogeneous Tool Set Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 28. Specification Language CSL Design goals of CSL Formal specification of the calculation method Specification of assignments in an arbitrary order, but: We require assignments to be unique and sortable w.r.t. the dependency order Generic interface to CAS Translation to CAS Suitably ordered assignments together with control structures form an imperative program Constants depending on constants which were modified are recomputed Executing the program using CAS yields a symbolic valuation Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 29. Specification Language CSL Design goals of CSL Formal specification of the calculation method Specification of assignments in an arbitrary order, but: We require assignments to be unique and sortable w.r.t. the dependency order Generic interface to CAS Translation to CAS Suitably ordered assignments together with control structures form an imperative program Constants depending on constants which were modified are recomputed Executing the program using CAS yields a symbolic valuation Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 30. Specification Language CSL Design goals of CSL Formal specification of the calculation method Specification of assignments in an arbitrary order, but: We require assignments to be unique and sortable w.r.t. the dependency order Generic interface to CAS Translation to CAS Suitably ordered assignments together with control structures form an imperative program Constants depending on constants which were modified are recomputed Executing the program using CAS yields a symbolic valuation Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 31. Specification Language CSL Design goals of CSL Formal specification of the calculation method Specification of assignments in an arbitrary order, but: We require assignments to be unique and sortable w.r.t. the dependency order Generic interface to CAS Translation to CAS Suitably ordered assignments together with control structures form an imperative program Constants depending on constants which were modified are recomputed Executing the program using CAS yields a symbolic valuation Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 32. Specification Language CSL Design goals of CSL Formal specification of the calculation method Specification of assignments in an arbitrary order, but: We require assignments to be unique and sortable w.r.t. the dependency order Generic interface to CAS Translation to CAS Suitably ordered assignments together with control structures form an imperative program Constants depending on constants which were modified are recomputed Executing the program using CAS yields a symbolic valuation Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 33. Specification Language CSL Design goals of CSL Formal specification of the calculation method Specification of assignments in an arbitrary order, but: We require assignments to be unique and sortable w.r.t. the dependency order Generic interface to CAS Translation to CAS Suitably ordered assignments together with control structures form an imperative program Constants depending on constants which were modified are recomputed Executing the program using CAS yields a symbolic valuation Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 34. A Little CSL Example Calculating a root of cos using Newton’s Method The CSL specification y := cos(x) %(A)% z := sin(x) %(B)% x := 10 %(C)% repeat x := x + y/z %(D)% until abs(y) < 0.001 The translation yields this program: C;A;B;repeat D;A;B; until abs(y) < 0.001 Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 35. A Little CSL Example Calculating a root of cos using Newton’s Method The CSL specification Building the Dependency Graph y := cos(x) %(A)% z := sin(x) %(B)% x := 10 %(C)% x repeat A x := x + y/z %(D)% y until abs(y) < 0.001 The translation yields this program: C;A;B;repeat D;A;B; until abs(y) < 0.001 Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 36. A Little CSL Example Calculating a root of cos using Newton’s Method The CSL specification Building the Dependency Graph y := cos(x) %(A)% z := sin(x) %(B)% x := 10 %(C)% x repeat A B x := x + y/z %(D)% y z until abs(y) < 0.001 The translation yields this program: C;A;B;repeat D;A;B; until abs(y) < 0.001 Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 37. A Little CSL Example Calculating a root of cos using Newton’s Method The CSL specification Building the Dependency Graph y := cos(x) %(A)% z := sin(x) %(B)% C x := 10 %(C)% x repeat A B x := x + y/z %(D)% y z until abs(y) < 0.001 The translation yields this program: C;A;B;repeat D;A;B; until abs(y) < 0.001 Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 38. A Little CSL Example Calculating a root of cos using Newton’s Method The CSL specification Building the Dependency Graph y := cos(x) %(A)% z := sin(x) %(B)% C x := 10 %(C)% x repeat A B x := x + y/z %(D)% y z until abs(y) < 0.001 D The translation yields this program: C;A;B;repeat D;A;B; until abs(y) < 0.001 Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 39. A Little CSL Example Calculating a root of cos using Newton’s Method The CSL specification Building the Dependency Graph y := cos(x) %(A)% z := sin(x) %(B)% C x := 10 %(C)% x repeat A B x := x + y/z %(D)% y z until abs(y) < 0.001 D The translation yields this program: C;A;B;repeat D;A;B; until abs(y) < 0.001 Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 40. Verified CAS Verification Points in CSL are positions of subterms of CSL statements Evaluating a such marked term produces a verification condition The CAS result is extended by a list of verification conditions Use Hets to prove verification conditions Specifying CAS program semantics in HasCASL Standard interpretation of programs as state transformers Properties of algorithms specified in CSL can be verified Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 41. Verified CAS Verification Points in CSL are positions of subterms of CSL statements Evaluating a such marked term produces a verification condition The CAS result is extended by a list of verification conditions Use Hets to prove verification conditions Specifying CAS program semantics in HasCASL Standard interpretation of programs as state transformers Properties of algorithms specified in CSL can be verified Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 42. Verified CAS Verification Points in CSL are positions of subterms of CSL statements Evaluating a such marked term produces a verification condition The CAS result is extended by a list of verification conditions Use Hets to prove verification conditions Specifying CAS program semantics in HasCASL Standard interpretation of programs as state transformers Properties of algorithms specified in CSL can be verified Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 43. Verified CAS Verification Points in CSL are positions of subterms of CSL statements Evaluating a such marked term produces a verification condition The CAS result is extended by a list of verification conditions Use Hets to prove verification conditions Specifying CAS program semantics in HasCASL Standard interpretation of programs as state transformers Properties of algorithms specified in CSL can be verified Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 44. Verified CAS Verification Points in CSL are positions of subterms of CSL statements Evaluating a such marked term produces a verification condition The CAS result is extended by a list of verification conditions Use Hets to prove verification conditions Specifying CAS program semantics in HasCASL Standard interpretation of programs as state transformers Properties of algorithms specified in CSL can be verified Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 45. Verified CAS Verification Points in CSL are positions of subterms of CSL statements Evaluating a such marked term produces a verification condition The CAS result is extended by a list of verification conditions Use Hets to prove verification conditions Specifying CAS program semantics in HasCASL Standard interpretation of programs as state transformers Properties of algorithms specified in CSL can be verified Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 46. Example Verifying a result from the CAS A CAS program We set verification point at maximize position → maximize(t, x) is marked . . Environment = σ . CAS computes this expression in context σ y := maximize(t, x) and retuns result r . . . Apply substitution σ to t and obtain t We produce the verification condition maximize(t , x) = r Translate this equality to HasCASL for proving Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 47. Example Verifying a result from the CAS A CAS program We set verification point at maximize position → maximize(t, x) is marked . . Environment = σ . CAS computes this expression in context σ y := maximize(t, x) and retuns result r . . . Apply substitution σ to t and obtain t We produce the verification condition maximize(t , x) = r Translate this equality to HasCASL for proving Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 48. Example Verifying a result from the CAS A CAS program We set verification point at maximize position → maximize(t, x) is marked . . Environment = σ . CAS computes this expression in context σ y := maximize(t, x) and retuns result r . . . Apply substitution σ to t and obtain t We produce the verification condition maximize(t , x) = r Translate this equality to HasCASL for proving Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 49. Example Verifying a result from the CAS A CAS program We set verification point at maximize position → maximize(t, x) is marked . . Environment = σ . CAS computes this expression in context σ y := maximize(t, x) and retuns result r . . . Apply substitution σ to t and obtain t We produce the verification condition maximize(t , x) = r Translate this equality to HasCASL for proving Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 50. Example Verifying a result from the CAS A CAS program We set verification point at maximize position → maximize(t, x) is marked . . Environment = σ . CAS computes this expression in context σ y := maximize(t, x) and retuns result r . . . Apply substitution σ to t and obtain t We produce the verification condition maximize(t , x) = r Translate this equality to HasCASL for proving Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 51. Example Verifying a result from the CAS A CAS program We set verification point at maximize position → maximize(t, x) is marked . . Environment = σ . CAS computes this expression in context σ y := maximize(t, x) and retuns result r . . . Apply substitution σ to t and obtain t We produce the verification condition maximize(t , x) = r Translate this equality to HasCASL for proving Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 52. CSL, CAS and Hets CSL and the Hets Logic Graph Logic Graph Isabelle Prover Isabelle HasCASL Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 53. CSL, CAS and Hets CSL and the Hets Logic Graph Logic Graph Isabelle Prover Isabelle HasCASL CSL Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 54. CSL, CAS and Hets CSL and the Hets Logic Graph Logic Graph Isabelle Prover Isabelle HasCASL Reduce CSL Maxima Mathematica CAS Interface Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 55. CSL, CAS and Hets CSL and the Hets Logic Graph Logic Graph Isabelle Prover Isabelle HasCASL Reduce CSL Maxima Mathematica CAS Interface Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 56. CSL, CAS and Hets cont. The CSL institution Signatures are collections of real constants and functions over the reals Sentences are program statements or first order formulas in an extended theory of the reals augmented by the signature Models are program states, i.e., symbolic valuations A state satisfies a program if it terminates successfully A state satisfies a formula φ if φ holds under this valuation Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 57. CSL, CAS and Hets cont. The CSL institution Signatures are collections of real constants and functions over the reals Sentences are program statements or first order formulas in an extended theory of the reals augmented by the signature Models are program states, i.e., symbolic valuations A state satisfies a program if it terminates successfully A state satisfies a formula φ if φ holds under this valuation Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 58. CSL, CAS and Hets cont. The CSL institution Signatures are collections of real constants and functions over the reals Sentences are program statements or first order formulas in an extended theory of the reals augmented by the signature Models are program states, i.e., symbolic valuations A state satisfies a program if it terminates successfully A state satisfies a formula φ if φ holds under this valuation Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 59. CSL, CAS and Hets cont. The CSL institution Signatures are collections of real constants and functions over the reals Sentences are program statements or first order formulas in an extended theory of the reals augmented by the signature Models are program states, i.e., symbolic valuations A state satisfies a program if it terminates successfully A state satisfies a formula φ if φ holds under this valuation Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 60. CSL, CAS and Hets cont. The CSL institution Signatures are collections of real constants and functions over the reals Sentences are program statements or first order formulas in an extended theory of the reals augmented by the signature Models are program states, i.e., symbolic valuations A state satisfies a program if it terminates successfully A state satisfies a formula φ if φ holds under this valuation Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 61. Summary and Outlook Specification language CSL for industrial standards Synthesis of programs for generic CAS interface Verification Points for local verification of CAS result Integration of CSL and CAS interface in Hets Specification of CSL semantics in HasCASL Relating CSL to HasCASL by theoroidal comorphism Benefit from symbolic character of CAS computations Using CAS to simplify CSL specifications for partial instantiations or given set of additional assumptions Replace special functions by closed solutions found by the CAS Finding instantiations for underspecified specifications, e.g., number of bolts needed for flange to satisfy standard Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 62. Summary and Outlook Specification language CSL for industrial standards Synthesis of programs for generic CAS interface Verification Points for local verification of CAS result Integration of CSL and CAS interface in Hets Specification of CSL semantics in HasCASL Relating CSL to HasCASL by theoroidal comorphism Benefit from symbolic character of CAS computations Using CAS to simplify CSL specifications for partial instantiations or given set of additional assumptions Replace special functions by closed solutions found by the CAS Finding instantiations for underspecified specifications, e.g., number of bolts needed for flange to satisfy standard Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 63. Summary and Outlook Specification language CSL for industrial standards Synthesis of programs for generic CAS interface Verification Points for local verification of CAS result Integration of CSL and CAS interface in Hets Specification of CSL semantics in HasCASL Relating CSL to HasCASL by theoroidal comorphism Benefit from symbolic character of CAS computations Using CAS to simplify CSL specifications for partial instantiations or given set of additional assumptions Replace special functions by closed solutions found by the CAS Finding instantiations for underspecified specifications, e.g., number of bolts needed for flange to satisfy standard Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 64. Summary and Outlook Specification language CSL for industrial standards Synthesis of programs for generic CAS interface Verification Points for local verification of CAS result Integration of CSL and CAS interface in Hets Specification of CSL semantics in HasCASL Relating CSL to HasCASL by theoroidal comorphism Benefit from symbolic character of CAS computations Using CAS to simplify CSL specifications for partial instantiations or given set of additional assumptions Replace special functions by closed solutions found by the CAS Finding instantiations for underspecified specifications, e.g., number of bolts needed for flange to satisfy standard Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 65. Summary and Outlook Specification language CSL for industrial standards Synthesis of programs for generic CAS interface Verification Points for local verification of CAS result Integration of CSL and CAS interface in Hets Specification of CSL semantics in HasCASL Relating CSL to HasCASL by theoroidal comorphism Benefit from symbolic character of CAS computations Using CAS to simplify CSL specifications for partial instantiations or given set of additional assumptions Replace special functions by closed solutions found by the CAS Finding instantiations for underspecified specifications, e.g., number of bolts needed for flange to satisfy standard Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 66. Summary and Outlook Specification language CSL for industrial standards Synthesis of programs for generic CAS interface Verification Points for local verification of CAS result Integration of CSL and CAS interface in Hets Specification of CSL semantics in HasCASL Relating CSL to HasCASL by theoroidal comorphism Benefit from symbolic character of CAS computations Using CAS to simplify CSL specifications for partial instantiations or given set of additional assumptions Replace special functions by closed solutions found by the CAS Finding instantiations for underspecified specifications, e.g., number of bolts needed for flange to satisfy standard Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 67. Summary and Outlook Specification language CSL for industrial standards Synthesis of programs for generic CAS interface Verification Points for local verification of CAS result Integration of CSL and CAS interface in Hets Specification of CSL semantics in HasCASL Relating CSL to HasCASL by theoroidal comorphism Benefit from symbolic character of CAS computations Using CAS to simplify CSL specifications for partial instantiations or given set of additional assumptions Replace special functions by closed solutions found by the CAS Finding instantiations for underspecified specifications, e.g., number of bolts needed for flange to satisfy standard Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 68. Summary and Outlook Specification language CSL for industrial standards Synthesis of programs for generic CAS interface Verification Points for local verification of CAS result Integration of CSL and CAS interface in Hets Specification of CSL semantics in HasCASL Relating CSL to HasCASL by theoroidal comorphism Benefit from symbolic character of CAS computations Using CAS to simplify CSL specifications for partial instantiations or given set of additional assumptions Replace special functions by closed solutions found by the CAS Finding instantiations for underspecified specifications, e.g., number of bolts needed for flange to satisfy standard Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 69. Summary and Outlook Specification language CSL for industrial standards Synthesis of programs for generic CAS interface Verification Points for local verification of CAS result Integration of CSL and CAS interface in Hets Specification of CSL semantics in HasCASL Relating CSL to HasCASL by theoroidal comorphism Benefit from symbolic character of CAS computations Using CAS to simplify CSL specifications for partial instantiations or given set of additional assumptions Replace special functions by closed solutions found by the CAS Finding instantiations for underspecified specifications, e.g., number of bolts needed for flange to satisfy standard Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence
  • 70. Summary and Outlook Specification language CSL for industrial standards Synthesis of programs for generic CAS interface Verification Points for local verification of CAS result Integration of CSL and CAS interface in Hets Specification of CSL semantics in HasCASL Relating CSL to HasCASL by theoroidal comorphism Benefit from symbolic character of CAS computations Using CAS to simplify CSL specifications for partial instantiations or given set of additional assumptions Replace special functions by closed solutions found by the CAS Finding instantiations for underspecified specifications, e.g., number of bolts needed for flange to satisfy standard Industrial Standards, and Formal Verification German Research Center D. Dietrich, L. Schr¨der, E. Schulz o for Artificial Intelligence