Francesco Fatone, Professore Ordinario di Impianti Chimici presso Dipartimento di Scienze ed Ingegneria della Materia, dell'ambiente ed Urbanistica - Università Politecnica delle Marche IWS Italian Water Tour Live Digital - 5° tappa 14 ottobre 2021, Acque Veronesi

1. Digital Water City
Francesco Fatone, PhD, IWA Fellow
Professore Ordinario di Impianti Chimici Ambientali
Co-leader Cluster Value in Water – Water Europe
Action Group Business Models ICT4WATER Cluster
Università Politecnica delle Marche, Italy

2. Contenuti
→ Digital Water City : soluzioni digitali per innovare il servizio idrico
integrato in Europa
→ IUWMS: il sistema integrato - dalla raccolta al riutilizzo irriguo
→ Digitalizzazione nel Water Reuse Risk Management Plans
→(Big) data analytics e l’incertezza dei dati
→Digital twin e come colmare il gap
→Health risk analysis and assessment
→ Water-Energy-Food-Climatic Nexus: l’impronta ed il serious game
→ Trasferimento e replicazione dei risultati: casi studio in centro Italia

3. Digital-Water.City
Leading urban water management to its digital future
H2020 innovation action | 5 M€ funding
Project start: June 2019 | Duration: 3.5 years
5 cities
→ Large scale assessment of the benefits provided by the digital solutions
→ Lighthouse to raise awareness of other cities and accelerate market uptake
→ Create linkages between the physical and digital worlds
→ Develop and demonstrate 15 advanced digital solutions
to address water-related challenges
→ Leverage the potential of data and digital technologies
→ Boost the water management in 5 EU cities
→ Promote the value of the digital solutions for the tech providers
→ Achieve a new step in the integration of digital solutions in EU,
in particular regarding cybersecurity, interoperability and governance

5. │ Full potential of digital technologies
Sensors, modelling, AI, Digital Twin, AR, Drones…
│ Assess the benefits of 15 digital solutions along the water value chain
Paris, Copenhagen, Berlin, Milan and Sofia
Digital-Water.City

6. Soluzioni Digitali
Sensori per misura
della contaminazione
batterica
Sistema di Early
Warning per il
riutilizzo in sicurezza
Monitoraggio da remoto
dei fabbisogni e degli
stress idrici
Piattaforma di Match making
tra fabbisogno e disponibilità
della risorsa idrica
Sviluppo di un serous game
basato sui dati reali per l’analisi
del nesso acqua-energia-
agricoltura-ecosistema
#Milano
→ Riutilizzo in sicurezza di acqua reflua depurata a uso irriguo in
agricoltura
→ Sensibilizzare la centralità del nesso tra acqua-energia-agricoltura-ecosistema

7. Integrated Urban Wastewater Management
Scheme
INDUSTRIALI COMMERCIALI RESIDENZIALI
CSO
PLUVIOMETRIA VASCHE LAMINAZIONE
Integrazione delle soluzioni digitali per il riutilizzo in sicurezza delle acque reflue in agricoltura
MONITORAGGIO E CONTROLLO DEGLI APPORTI E
DEGLI SCARICHI ALLA/DALLA RETE FOGNARIA
PIATTAFORMA INTEGRATA PER LA QUALITÀ DELLE ACQUE REFLUE
SISTEMA DI GESTIONE DELLA RETE FOGNARIA
MATCH-MAKING TRA DISPONIBILITÀ E DOMANDA
MONITORAGGIO DEL
FABBISOGNO IDRICO E DELLE
CONDIZIONI DEL TERRENO
RETE DI MONITORAGGIO INTEGRATA CON SENSORI DI PROCESSO E
SPECIFICI PER MISURA DELLA CONTAMINAZIONE BATTERICA
SISTEMA DI EARLY WARNING E DSS

8. CATCHMENT
SEWER
WWTP
DISTRIBUTION
IRRIGATION CROP AND
SOIL
The integrated wastewater and reuse system
PERI-URBAN AREA
PESCHIERA BORROMEO
INNOVATIVE SENSORS
IRRIGATED AREA
PESCHIERA BORROMEO WWTP
MONITORING NETWORK FIELD SENSORS

9. WWTP
9
Unconventional treatment Line 2:
→ Screening
→ Sand – oil removal and lamellar settler
→ Biologic filtration
→ UV disinfection
Existing sensors:
• Effluent: TSS, N-NH4, N-NOx, P-PO4, Q (flow rate)
• BioProcesses: T, DO, REDOX,
• Disinfection: UV transmittance
New sensors:
• Effluent: Conductivity, pH, TOC, UV, E.coli
• Influent: Conductivity, pH, TSS, N-NH4, P-PO4

10. Uncertainty in planning, designing and
modelling of urban water infrastructure
Source: IWA Digital Water

11. Need of meta data
Source: IWA Digital Water

12. Early Warning System for safe water reuse
Tool for process monitoring and control Tool to support decision making Tool to support risk management
Monitoring and
supervision
Data elaborator and integrator to predict
water quality
Green light for
water reuse
Provide warnings if quality requirements for
water reuse are at risk of non-achievement
Decision
Support
Integration in digital twin providing
data/scenarios supporting decisions to optimize
cost-benefit of plants and processes in terms of
(waste)water-health nexus
Risk
minimization
Integration of EWS in risk management,
together with online sensor control from
remote, data elaborations and periodic analysis
(QMRA) as control measures to reduce risk.

13. Digital twin
Source: IWA Digital Water

14. Esempio:
IWA
digital
water
program
Source: IWA Digital Water

15. Influent:
- Q
- N-NH4
- P-PO4
- pH
- BOD5
- COD
- TSS
Biological process:
- DO
- T
Simulated data
Effluent:
- Q
- N-NH4
- N-NOx
- P-PO4
- pH
- BOD5
- COD
- TSS (sensor
not reliable)
Sensor data (blue line) compared to the
laboratory analyses (red points) for
a) NH4
+,
b) NO3-N,
c) PO4-P
d) TSS
Online data + Offline data
Simulated Malfunction Note
Aeration interruption/reduction Different durations of air interruption/reduction
Error in the recirculation of the mixed liquor Simulation performed with different Qr
Q backwash reduction or interruption
Industrial discharge pH 5 or pH 11; High COD load; Nutrient
deficiency
Rain event Different durations/intensity
Error in external carbon dosage
Error in sludge extraction
Generation of data related to extra-ordinary events
Methodology for water reuse risk management: data analysis
and model integration

16. Architecture of real-time data processing
Data from Model Sim
1
Needed for ANN generation
(Lambda function) 1. Real-time evaluation of TSS,
BOD, COD (parameters that are
not measured by sensors)
2. Predictive WWTP performance
(time series ANN to predict
water quality parameters in the
effluent)
Early warning for
water reuse
Digital twin: identification of
malfunction and selection of
the best preventive action

17. Input parameters at time t min std mean max
Influent Flow 17881 26717 76721 252000
Influent pH 7.100 0.181 7.674 8.000
Influent NNH4 4.225 5.537 16.101 37.315
Influent PPO4 0.288 0.420 1.237 3.610
Biofor DN Temperature 3.714 3.690 18.880 25.982
Biofor CN - Dissolved oxygen 0.000 1.474 5.238 7.750
Effluent Flow 16386 26717 75226 250504
Effluent pH 6.792 0.084 6.992 7.488
Effluent NNH4 0.073 2.217 1.042 19.956
Effluent NNOx 0.000 2.628 6.538 15.289
Effluent PPO4 0.001 0.401 0.627 2.134
Output parameters at time t+24 min std mean max
Effluent TSS 4.421 1.338 6.284 10.997
Parameters and their range – ANN network
Network type Train function Divisions Number of inputs and outputs
Feedforwardnet
(Feed forward neural network)
'trainlm'
Train ratio = 60/100
Validation ratio = 15/100
Test ratio = 25/100
Inputs = 11
Output = 1 (Effluent TSS)
ANN for TSS prediction

18. Data
type
CC RMSE SI BIAS
Train 0.993 0.159 0.025 -0.562
Test 0.992 0.175 0.028 -3.951
Full 0.992 0.165 0.026 -15.488
ANN for TSS prediction

19. Risk management approach
SYSTEM DESCRIPTION ACTORS AND ROLES HAZARD IDENTIFICATION EXPOSURE TARGETS RISK ASSESSMENT
SPECIFIC REQUIREMENTS PREVENTIVE MEASURES QUALITY CONTROL SYSTEMS
ENVIRONMENTAL MONITORING
SYSTEMS
EMERGENCY MANAGEMENT COORDINATION
SEMI-QUANTITATIVE APPROACH  Check List and Risk Matrix
→ Risk Matrix helps to identify priorities, but evaluations are
affected by score attribution
→ Online data can support risk minimization as control
measures to minimize risks

20. + Monte Carlo Analysis
Fitting
Measured E.coli concentrations
Risk management approach
Probability distribution
functions
1.00E-09
1.00E-08
1.00E-07
1.00E-06
1.00E-05
1.00E-04
1.00E-03
1.00E-02
1.00E-01
1.00E+00
campy DALY cripto DALY rota DALY
→ QRMA can be performed periodically,
updating data trends and distributions
→ Monte Carlo simulations increase the
available database, but assumptions
on distributions are needed
→ Different scenarios can be evaluated
using additional barriers (e.g., drip
irrigation, produce processing)
→ Online tools are available to evaluate
QMRA and flexible to modify
distributions and scenarios
Preliminary Results

21. Peschiera Borromeo Demo Site
(season 2020-2021)
21
Drip irrigated field
Border irrigated
field
Weather station
2 Suction Lysimeters
Multilevel Water
Content probe
Piezometric well
equipped with a
pressure transducer
2 Suction Lysimeters
Piezometric well
equipped with a
pressure transducer

22. RGB Image Moisture Index Nutrient Stress Map
• Imagery from Satellites – Sentinel S2
• Flights authorized by ENAC
(we are close to Linate airport)
• VIS + NIR + thermal
Peschiera Borromeo Demo Site
(season 2020-2021)

23. Mapping field status from UAV
23
Thermal data
Nutrient stress Index
Peschiera Borromeo Demo Site
(season 2020-2021)

24. Match Making Front-end
24

25. 25
From Models (crop + weather - meteograms)
Match Making Front-end

26. PUBLISHED RESULTS
Serious Game

27. https://www.seriousgame4dwc.eu/
OPTIMIZED FOR NOTEBOOK / TABLET

30. INQUADRAMENTO NORMATIVO, LINEE GUIDA E MIGLIORI TECNICHE DISPONIBILI
CASI STUDIO
FLUSSHYGIENE - GERMANIA
SISTEMA DI CONTROLLO E GESTIONE DEI CSO
BERLINO
PIANO DI SALVAGUARDIA DELLA BALNEAZIONE DI
RIMINI
BATHING WATER FORECAST SYSTEM (BWF) - DANIMARCA

31. METHODOLOGICAL APPROACH: PLANNING – DESIGN – MANAGEMENT - TREATMENT

32. MITIGATION SCENARIOS
On-site treatment with UV disinfection (85, 128 and 256 mJ/cm²) coupled with sand filtration unit provided no microbial
indicators in the effluent; PFA (2,4 and 6 mg/l) treatment also yielded high removal efficiencies.
Two mitigation scenarios:
• HYDRAULIC SCENARIO  critical overflows collected to the WWTP;
• TREATMENT SCENARIO  critical overflows locally treated with filtration and UV disinfection
The E. Coli removal obtained are 28% in
the Hydraulic Scenario and 73% in the
Treatment Scenario.

33. GRAZIE
Francesco Fatone, PhD, IWA Fellow
Professore Ordinario di Impianti Chimici Ambientali
Co-leader Cluster Value in Water – Water Europe
Action Group Business Models ICT4WATER Cluster
Università Politecnica delle Marche, Italy