SlideShare a Scribd company logo
1 of 6
Download to read offline
tmcgrail@doble.com
21, rue d’Artois, F-75008 PARIS CIGRE US National Committee
http://www.cigre.org 2019 Grid of the Future Symposium
Challenges of Data Management & Analytics in the Future Grid
T. MCGRAIL, B. GWYN J. BEARDSALL, M. ROWBOTTOM
Doble Engineering Drax Power
USA UK
SUMMARY
With a growth in interest in ‘big data’ as electric grids evolve and data sources become more
common and more productive, there needs to be a discussion of the management of data in a
secure manner, and the role of analytics to provide information and have ‘meaning’. This paper
looks at a number of challenges that are beginning to be faced, and opportunities to ensure that the
Future Grid is secure. Challenge 1 is the management of ‘big data’, which may provide value if
appropriately viewed and analyzed; Challenge 2 is the management of security, for both data and
systems which use the data; Challenge 3 is the need for appropriate urgency in analysis and action;
Challenge 4 is to understand the meaning of the data and associated analyses, but also to understand
the limits of our understanding.
KEYWORDS
Big data, Analytics, Sense-Making
2
CHALLENGE 1: DATA MANAGEMENT
There has been over the last few years a growth in interest in the idea of ‘big data’ [1].
Occasionally the characteristics of ‘big data’ have been summarized in terms of the ‘V’s of
‘big data, some of which are collated in Table 1.
Characteristic Implication
Velocity High rates of data generation, collation and analysis
Volume ‘incredible amounts’ of data to be managed
Variety Structured v unstructured: think photo content
Value The value may be virtual: unrealizable
Veracity Trustworthiness - verisimilitude
Table 1 Some of the various V’s of ‘big data’
We collect data to support decisions – through analytics such as level checking,
presence/absence and other, often more exotic approaches [2][3]. With increasing pressure on
assets to perform continuously, to reduce outages for maintenance or through unplanned
issues, and within the context of Industry 4.0 bringing condition/operation sensor ubiquity and
autonomy, the status quo is insufficient: we need to know not only where we are now, but
where we will be, and when [4][5]. A CIGRE Technical brochure indicates some of the key
approaches to improving knowledge of substation asset condition, and optimizing asset usage
– the approaches it describes are ‘sensible’, from Partial Discharge (PD) and Electro-
Magnetic Interference (EMI) surveys to increased visual inspection and appropriate condition
monitoring [6].
Analysis of data brings challenges – not least in understanding new and exotic techniques
related to Machine Learning (ML), deep-ML, and Artificial Intelligence (AI) – but also in
terms of how to apply them: what is needed to make the techniques valuable? Google
provides a glossary of terms, but not everyone will agree with the Google definitions [7]. It
has also been noted that the application of exotic analytics requires very ‘clean’ and well
presented data – to the point where 95% of the possible value can be derived from data
through basic statistics and meaningful organization of the data [8].
The application of condition monitoring is, at its core, an asset management activity
[9][10][11]. Many organizations are embracing the approaches and structures described in the
international standard for asset management, ISO 55000 [12]. Originally published as three
documents in the 55000 range, including management system requirements and application,
further documents are in development, including, for example, ‘alignment of asset
management, finance and accounting’ in ISO 55010 [13]. The development and evolution of
the Common Information Model (CIM) has been instrumental in enabling organizations to
both structure their own data and to allow communication of data in a standard format with
other organizations [14]. Even here there is possible cause for confusion: does data relate to
an asset or to a location in a grid system, or both? The load on a transformer is of interest to
the grid system management, but also for the ageing of the individual asset.
3
A big focus of an ISO 55000 asset management approach has been the balance of cost, risk
and performance [13][15]. These have been augmented with the concept of ‘sustainability’ in
revised versions – indicating a desire for an asset management which plans for the future. In
terms of risk management, condition monitoring is a decision support tool to reduce the
likelihood of catastrophic events and thus reduce risk – condition monitoring addresses the
cause of hazards, rather than the consequence. Analysis of risk is a key element of asset
management, and thus also must be addressed at the condition monitoring level: how does the
data generated help us review the asset condition and the consequent probability of failure?
CHALLENGE 2: SECURITY
Cyber security is a ‘hot topic’ in the industry at present, and rightly so. The consequences of
cyber attacks can be devastating. Security applies not only to the data itself, and access to the
data, but to management of the software systems and applications which generate, analyze
and present the data and resulting information. For example, a Dissolved Gas Analysis (DGA)
detector on a generator station auxiliary transformer may indicate the presence and severity of
a possible problem, but a diagnosis of possible cause requires further analysis tools or sensors.
Figure 1 shows a DGA detector in action, rising from a nominal 30 ppm to almost 80 ppm in
about a week. The question is: do we believe the data and the resulting need for action? This
is a very simple case, but if the plan is a de-energization of a power transformer as a result of
such a rise, that has an impact on the grid as a whole: the recent events in the UK show the
effects of a single lightning strike, which could be exacerbated by deliberate weakening of the
grid’s resilience through cyber attacks on monitoring to necessitate the removal of highly
inertial equipment, as discussed in the Financial Times of London [16].
Figure 1 Detector Type Monitoring of DGA Level
CHALLENGE 3: TIME IS OF THE ESSENCE
If assets did not deteriorate over time, they would not need maintenance, except to address
external influences, and they would not need replacing except to address failures caused by
external agents [17] [18]. But deterioration takes place through use/application and through
interaction with the working environment over time: transformer paper insulation ages and
becomes brittle with temperature, say, leading to replacement of a transformer before failure
through mechanical forces.
4
We need to understand timescales of operation for failure modes – and the determination of
parameters that indicate the presence and possible severity of a failure mode so that we can
plan intervention in a timely manner, as per Figure 2 [18]. An analogy, using an automobile
(car) can be useful: tire pressure reduces over time, eventually leading to the failure of the
tire, and possible requirement for a new car.
Figure 2 Modeling the Failure Mode Analysis Process
The process may take weeks to reach a conclusion. Detection of low pressure can therefore be
an indication of a problem to be addressed within a number of weeks - we know that as the
pressure reduces the likelihood/probability of tire failure increases, even if we do not have
exact values: we understand the urgency [18]. Power transformers are much more complex
assets than tires, and with many more failure modes and more parameters to indicate
symptoms, the overall condition assessment is much more complex [19].
In a world of ‘big data’ more and more operational and condition data may be available, but
the collation of that data is complex and machine learning and artificial intelligence only
provide clues as to what may be a true cause for concern: a true failure mode in operation.
Combining data from multiple measurements, each of which may be relevant to multiple
failure modes, and then combining failure modes to assess a power transformer is not a simple
process. To ensure that a sense of urgency is not lost, assessments must be calibrated for time
and combined to reflect the true condition and intervention requirements of the asset. The
means of reducing data to allow for a concise assessment for comparative purposes, includes
the process of generating an Asset Health Index: and a health index which loses the sense of
urgency is of little value in practice [20][21][22].
5
CHALLENGE 4: ANALYTICS AND ACTION
An Asset Health Index is a useful place to start discussion of ‘meaning’ in data. Often a
numeric value, an AHI is apparently simple to understand. But what does the number mean?
We believe we understand numbers – we know which are bigger, and which are smaller and
have some grasp of relative size – so how hard can it be? We are at risk of succumbing to the
Dunning-Kruger Effect [23]: our estimation of our capability is not in proportion to our actual
knowledge and capability. Hence the old saying: “A little learning is a dangerous thing.” The
health index may take a vast amount of data, from the asset itself, from sensors, from
maintenance history and produce a number which should be calibrated for time, monotonic in
scale and urgency, auditable in terms of having a clear path from raw data to condition/health,
justifiable in terms of the analysis used [19]. And yet many indices are none of those things –
they just seem to be a way to collate a lot of data and produce a useful number: and we
understand numbers, so we think we understand the analysis and the derivation… the result is
often a health index which is disappointing and an asset management approach which is
unjustified.
Machine Learning is growing in popularity [7][8]. But we may suffer from the same
Dunning-Kruger effect: the analysis of data leads to a conclusion, but is the conclusion valid?
The application of machine learning to PD/EMI spectra has been beneficial – even to the
point of surpassing experts in their analyses of raw data [24], and as has been noted, much of
the benefit of Machine Learning can be derived from good data clean up and standard
statistical approaches [8].
It is important to have a plan to respond to anomalous or unusual data: in condition
monitoring a response plan is a ‘best practice’: a response to a fire alarm should be to follow
the plan in place at the particular location, not have a discussion and work out an approach
‘on the fly’. But plans need to address the variability in data sources and applications – in
monitoring a human heart rate, for example, a value of 140 may be considered very high,
unless there was strenuous exercise being undertaken, in which case we may be more
concerned about the duration at this level.
CONCLUSION
Ultimately, we must be aware of the ‘meaning´ inherent in the data we use for decisions: and
also be aware of our own limitations in giving meaning to data, or to resulting parameters
used for decision.
ACKNOWLEDGMENTS
The authors would like to thank their many colleagues for conversations and discussion on
matters related to the content of this paper; you know who you are.
BIBLIOGRAPHY
[1] https://www.xsnet.com/blog/bid/205405/the-v-s-of-big-data-velocity-volume-value-
variety-and-veracity
[2] T. McGrail, D. Angell & K. Elkinson, “Asset Management - Data, Analysis & Decisions”,
Electric Light & Power magazine, Volume 18, Issue 7, July 2013
[3] Tony McGrail, “Asset Management: Data and Decisions”, IEEE International Smart Grid
Conference, Perth Western Australia, 2011
6
[4] “Machine Condition Monitoring Market”, B2B Market Research,
www.marketsandmarkets.com
[5] “Industry 4.0: Opportunities and Challenges of the Industrial Internet”, Price Waterhouse
Coopers, https://www.pwc.com/gx/en/technology/pdf/industrial-internet-of-things.pdf
[6] CIGRE Technical Brochure 660 “Saving Through Optimized Maintenance in Air
Insulated Substations”, 2016
[7] https://developers.google.com/machine-learning/glossary/
[8] https://www.kdnuggets.com/2018/04/dirty-little-secret-data-scientist.html
[9] T. McGrail, R. Heywood, “Intelligent Transformers: Monitoring, smart grid integration
and asset management”, Coil Winding Insulation and Electrical Manufacturing
Exhibition, Berlin, Germany, 2014
[10] T. McGrail, K. Elkinson, M. Lawrence et al, “Aspects of Power Transformer Asset
Management”, IEEE PES T&D Conference, Orlando, USA, 2012
[11] K. Wyper, G. MacKay, TransGrid Australia & T. McGrail “Condition Monitoring in
the Real World” 80th International Conference of Doble Clients, Boston, USA, 2013
[12] ISO 55000 “Asset Management – Overview, Principles and Terminology”, 2014
[13] https://en.wikipedia.org/wiki/ISO_55000
[14] Common Information Model: https://www.dmtf.org/standards/cim
[15] T. McGrail, "Practical Steps to Improve Asset Management Applications", TrafoTech
Conference, New Delhi, India, 2016
[16] https://www.ft.com/content/bcb9f3b2-c352-11e9-a8e9-296ca66511c9
[17] https://www.through-life-engineering-
services.org/downloads/27302_White_paper_amended.pdf
[18] "Condition Monitoring: Data Driven & Knowledge Based Analyses", T. McGrail & K.
Elkinson, CIGRE Canada, Vancouver, 2016
[19] “Developing & Using Justifiable Asset Health Indices For Tactical & Strategic Risk
Management”, McGrail, Rhoads, White, CEPSI Conference, 2018, Kuala Lumpur,
Malaysia
[20] “Transformer Health Indices & Probability of Failure”, McGrail, Kennedy, Lawrence,
TechCon Australia/NZ, Sydney, Australia, 2017
[21] “Turning an Asset Health Index into a Probability of Failure”, McGrail, 85th
International Conference of Doble Clients, Boston, USA, 2018
[22] “Translating an Asset Health Index into a Probability of Failure”, CEATI report
30113, 2018
[23] "Unskilled and Unaware of It”, Kruger & Dunning, Journal of Personality and Social
Psychology. 77 (6): 1121–1134, 1999
[24] “AI approaches to interpretation of electrical CM data”, Dr. I. Mitiche, BINDT 16th
International Conference on Condition Monitoring & Asset Management, Glasgow, UK,
2019

More Related Content

What's hot

Dynamic Rule Base Construction and Maintenance Scheme for Disease Prediction
Dynamic Rule Base Construction and Maintenance Scheme for Disease PredictionDynamic Rule Base Construction and Maintenance Scheme for Disease Prediction
Dynamic Rule Base Construction and Maintenance Scheme for Disease Predictionijsrd.com
 
Energy Event Index: Weather risk forecasts for power utilities
Energy Event Index: Weather risk forecasts for power utilitiesEnergy Event Index: Weather risk forecasts for power utilities
Energy Event Index: Weather risk forecasts for power utilitiesSchneider Electric
 
CS309A Final Paper_KM_DD
CS309A Final Paper_KM_DDCS309A Final Paper_KM_DD
CS309A Final Paper_KM_DDDavid Darrough
 
Enabling Big Data with Data-Level Security:The Cloud Analytics Reference Arch...
Enabling Big Data with Data-Level Security:The Cloud Analytics Reference Arch...Enabling Big Data with Data-Level Security:The Cloud Analytics Reference Arch...
Enabling Big Data with Data-Level Security:The Cloud Analytics Reference Arch...Booz Allen Hamilton
 
Compegence: Dr. Abhinanda Sarkar - Energy Analytics_IISC_2012_Oct
Compegence: Dr. Abhinanda Sarkar - Energy Analytics_IISC_2012_OctCompegence: Dr. Abhinanda Sarkar - Energy Analytics_IISC_2012_Oct
Compegence: Dr. Abhinanda Sarkar - Energy Analytics_IISC_2012_OctCOMPEGENCE
 
Data Mining and Offshore Maintenance Management
Data Mining and Offshore Maintenance ManagementData Mining and Offshore Maintenance Management
Data Mining and Offshore Maintenance ManagementBilly Buckenham
 
Should this loan be approved or denied
Should this loan be approved or deniedShould this loan be approved or denied
Should this loan be approved or deniedPOOJA PATIL
 

What's hot (7)

Dynamic Rule Base Construction and Maintenance Scheme for Disease Prediction
Dynamic Rule Base Construction and Maintenance Scheme for Disease PredictionDynamic Rule Base Construction and Maintenance Scheme for Disease Prediction
Dynamic Rule Base Construction and Maintenance Scheme for Disease Prediction
 
Energy Event Index: Weather risk forecasts for power utilities
Energy Event Index: Weather risk forecasts for power utilitiesEnergy Event Index: Weather risk forecasts for power utilities
Energy Event Index: Weather risk forecasts for power utilities
 
CS309A Final Paper_KM_DD
CS309A Final Paper_KM_DDCS309A Final Paper_KM_DD
CS309A Final Paper_KM_DD
 
Enabling Big Data with Data-Level Security:The Cloud Analytics Reference Arch...
Enabling Big Data with Data-Level Security:The Cloud Analytics Reference Arch...Enabling Big Data with Data-Level Security:The Cloud Analytics Reference Arch...
Enabling Big Data with Data-Level Security:The Cloud Analytics Reference Arch...
 
Compegence: Dr. Abhinanda Sarkar - Energy Analytics_IISC_2012_Oct
Compegence: Dr. Abhinanda Sarkar - Energy Analytics_IISC_2012_OctCompegence: Dr. Abhinanda Sarkar - Energy Analytics_IISC_2012_Oct
Compegence: Dr. Abhinanda Sarkar - Energy Analytics_IISC_2012_Oct
 
Data Mining and Offshore Maintenance Management
Data Mining and Offshore Maintenance ManagementData Mining and Offshore Maintenance Management
Data Mining and Offshore Maintenance Management
 
Should this loan be approved or denied
Should this loan be approved or deniedShould this loan be approved or denied
Should this loan be approved or denied
 

Similar to Challenges of Data Management & Analytics in the Future Grid

Data Science for Building Energy Management a reviewMigue.docx
Data Science for Building Energy Management a reviewMigue.docxData Science for Building Energy Management a reviewMigue.docx
Data Science for Building Energy Management a reviewMigue.docxrandyburney60861
 
Mitigating Climate Change using Artificial Intelligence
Mitigating Climate Change using Artificial IntelligenceMitigating Climate Change using Artificial Intelligence
Mitigating Climate Change using Artificial IntelligenceIRJET Journal
 
DIFFERENCES OF CLOUD-BASED SERVICES AND THEIR SAFETY RENEWAL IN THE HEALTH CA...
DIFFERENCES OF CLOUD-BASED SERVICES AND THEIR SAFETY RENEWAL IN THE HEALTH CA...DIFFERENCES OF CLOUD-BASED SERVICES AND THEIR SAFETY RENEWAL IN THE HEALTH CA...
DIFFERENCES OF CLOUD-BASED SERVICES AND THEIR SAFETY RENEWAL IN THE HEALTH CA...IRJET Journal
 
DIFFERENCES OF CLOUD-BASED SERVICES AND THEIR SAFETY RENEWAL IN THE HEALTH CA...
DIFFERENCES OF CLOUD-BASED SERVICES AND THEIR SAFETY RENEWAL IN THE HEALTH CA...DIFFERENCES OF CLOUD-BASED SERVICES AND THEIR SAFETY RENEWAL IN THE HEALTH CA...
DIFFERENCES OF CLOUD-BASED SERVICES AND THEIR SAFETY RENEWAL IN THE HEALTH CA...IRJET Journal
 
A survey of big data and machine learning
A survey of big data and machine learning A survey of big data and machine learning
A survey of big data and machine learning IJECEIAES
 
IRJET- Accident Information Mining and Insurance Dispute Resolution
IRJET- Accident Information Mining and Insurance Dispute ResolutionIRJET- Accident Information Mining and Insurance Dispute Resolution
IRJET- Accident Information Mining and Insurance Dispute ResolutionIRJET Journal
 
A FALL DETECTION SMART WATCH USING IOT AND DEEP LEARNING
A FALL DETECTION SMART WATCH USING IOT AND DEEP LEARNINGA FALL DETECTION SMART WATCH USING IOT AND DEEP LEARNING
A FALL DETECTION SMART WATCH USING IOT AND DEEP LEARNINGIRJET Journal
 
EMB Briefings_technology_risk management final
EMB Briefings_technology_risk management finalEMB Briefings_technology_risk management final
EMB Briefings_technology_risk management finalMike Wilkinson
 
The Internet of Things - IoT and EHS - Alex Lima - 3 15 16
The Internet of Things - IoT and EHS - Alex Lima - 3 15 16The Internet of Things - IoT and EHS - Alex Lima - 3 15 16
The Internet of Things - IoT and EHS - Alex Lima - 3 15 16Alex Lima
 
A Novel Approach for Forecasting Disease Using Machine Learning
A Novel Approach for Forecasting Disease Using Machine LearningA Novel Approach for Forecasting Disease Using Machine Learning
A Novel Approach for Forecasting Disease Using Machine LearningIRJET Journal
 
Full Paper: Analytics: Key to go from generating big data to deriving busines...
Full Paper: Analytics: Key to go from generating big data to deriving busines...Full Paper: Analytics: Key to go from generating big data to deriving busines...
Full Paper: Analytics: Key to go from generating big data to deriving busines...Piyush Malik
 
Excel In Managing Spreadsheet Risk Presentation
Excel In Managing Spreadsheet Risk PresentationExcel In Managing Spreadsheet Risk Presentation
Excel In Managing Spreadsheet Risk Presentationgreghawes
 
MITIGATING RISK THROUGH AUTOMATIC TEMPERATURE SENSING
MITIGATING RISK THROUGH AUTOMATIC TEMPERATURE SENSINGMITIGATING RISK THROUGH AUTOMATIC TEMPERATURE SENSING
MITIGATING RISK THROUGH AUTOMATIC TEMPERATURE SENSINGIRJET Journal
 
Big Data Blackout: Are Utilities Powering up their Data Analytics
Big Data Blackout: Are Utilities Powering up their Data AnalyticsBig Data Blackout: Are Utilities Powering up their Data Analytics
Big Data Blackout: Are Utilities Powering up their Data AnalyticsRick Bouter
 
Big Data BlackOut: Are Utilities Powering Up Their Data Analytics?
Big Data BlackOut: Are Utilities Powering Up Their Data Analytics?Big Data BlackOut: Are Utilities Powering Up Their Data Analytics?
Big Data BlackOut: Are Utilities Powering Up Their Data Analytics?Capgemini
 
IRJET-Financial Distress Prediction of a Company using Data Mining
IRJET-Financial Distress Prediction of a Company using Data MiningIRJET-Financial Distress Prediction of a Company using Data Mining
IRJET-Financial Distress Prediction of a Company using Data MiningIRJET Journal
 
What is Digital Twin Technology
What is Digital Twin TechnologyWhat is Digital Twin Technology
What is Digital Twin TechnologyCIO Talk Network
 
Keys to extract value from the data analytics life cycle
Keys to extract value from the data analytics life cycleKeys to extract value from the data analytics life cycle
Keys to extract value from the data analytics life cycleGrant Thornton LLP
 
Anomalies detection for smart-home energy forecasting using moving average
Anomalies detection for smart-home energy forecasting using  moving averageAnomalies detection for smart-home energy forecasting using  moving average
Anomalies detection for smart-home energy forecasting using moving averageIJECEIAES
 

Similar to Challenges of Data Management & Analytics in the Future Grid (20)

Data Science for Building Energy Management a reviewMigue.docx
Data Science for Building Energy Management a reviewMigue.docxData Science for Building Energy Management a reviewMigue.docx
Data Science for Building Energy Management a reviewMigue.docx
 
Mitigating Climate Change using Artificial Intelligence
Mitigating Climate Change using Artificial IntelligenceMitigating Climate Change using Artificial Intelligence
Mitigating Climate Change using Artificial Intelligence
 
DIFFERENCES OF CLOUD-BASED SERVICES AND THEIR SAFETY RENEWAL IN THE HEALTH CA...
DIFFERENCES OF CLOUD-BASED SERVICES AND THEIR SAFETY RENEWAL IN THE HEALTH CA...DIFFERENCES OF CLOUD-BASED SERVICES AND THEIR SAFETY RENEWAL IN THE HEALTH CA...
DIFFERENCES OF CLOUD-BASED SERVICES AND THEIR SAFETY RENEWAL IN THE HEALTH CA...
 
DIFFERENCES OF CLOUD-BASED SERVICES AND THEIR SAFETY RENEWAL IN THE HEALTH CA...
DIFFERENCES OF CLOUD-BASED SERVICES AND THEIR SAFETY RENEWAL IN THE HEALTH CA...DIFFERENCES OF CLOUD-BASED SERVICES AND THEIR SAFETY RENEWAL IN THE HEALTH CA...
DIFFERENCES OF CLOUD-BASED SERVICES AND THEIR SAFETY RENEWAL IN THE HEALTH CA...
 
A survey of big data and machine learning
A survey of big data and machine learning A survey of big data and machine learning
A survey of big data and machine learning
 
IRJET- Accident Information Mining and Insurance Dispute Resolution
IRJET- Accident Information Mining and Insurance Dispute ResolutionIRJET- Accident Information Mining and Insurance Dispute Resolution
IRJET- Accident Information Mining and Insurance Dispute Resolution
 
A FALL DETECTION SMART WATCH USING IOT AND DEEP LEARNING
A FALL DETECTION SMART WATCH USING IOT AND DEEP LEARNINGA FALL DETECTION SMART WATCH USING IOT AND DEEP LEARNING
A FALL DETECTION SMART WATCH USING IOT AND DEEP LEARNING
 
Cloud complexity: the need for resilience
Cloud complexity: the need for resilienceCloud complexity: the need for resilience
Cloud complexity: the need for resilience
 
EMB Briefings_technology_risk management final
EMB Briefings_technology_risk management finalEMB Briefings_technology_risk management final
EMB Briefings_technology_risk management final
 
The Internet of Things - IoT and EHS - Alex Lima - 3 15 16
The Internet of Things - IoT and EHS - Alex Lima - 3 15 16The Internet of Things - IoT and EHS - Alex Lima - 3 15 16
The Internet of Things - IoT and EHS - Alex Lima - 3 15 16
 
A Novel Approach for Forecasting Disease Using Machine Learning
A Novel Approach for Forecasting Disease Using Machine LearningA Novel Approach for Forecasting Disease Using Machine Learning
A Novel Approach for Forecasting Disease Using Machine Learning
 
Full Paper: Analytics: Key to go from generating big data to deriving busines...
Full Paper: Analytics: Key to go from generating big data to deriving busines...Full Paper: Analytics: Key to go from generating big data to deriving busines...
Full Paper: Analytics: Key to go from generating big data to deriving busines...
 
Excel In Managing Spreadsheet Risk Presentation
Excel In Managing Spreadsheet Risk PresentationExcel In Managing Spreadsheet Risk Presentation
Excel In Managing Spreadsheet Risk Presentation
 
MITIGATING RISK THROUGH AUTOMATIC TEMPERATURE SENSING
MITIGATING RISK THROUGH AUTOMATIC TEMPERATURE SENSINGMITIGATING RISK THROUGH AUTOMATIC TEMPERATURE SENSING
MITIGATING RISK THROUGH AUTOMATIC TEMPERATURE SENSING
 
Big Data Blackout: Are Utilities Powering up their Data Analytics
Big Data Blackout: Are Utilities Powering up their Data AnalyticsBig Data Blackout: Are Utilities Powering up their Data Analytics
Big Data Blackout: Are Utilities Powering up their Data Analytics
 
Big Data BlackOut: Are Utilities Powering Up Their Data Analytics?
Big Data BlackOut: Are Utilities Powering Up Their Data Analytics?Big Data BlackOut: Are Utilities Powering Up Their Data Analytics?
Big Data BlackOut: Are Utilities Powering Up Their Data Analytics?
 
IRJET-Financial Distress Prediction of a Company using Data Mining
IRJET-Financial Distress Prediction of a Company using Data MiningIRJET-Financial Distress Prediction of a Company using Data Mining
IRJET-Financial Distress Prediction of a Company using Data Mining
 
What is Digital Twin Technology
What is Digital Twin TechnologyWhat is Digital Twin Technology
What is Digital Twin Technology
 
Keys to extract value from the data analytics life cycle
Keys to extract value from the data analytics life cycleKeys to extract value from the data analytics life cycle
Keys to extract value from the data analytics life cycle
 
Anomalies detection for smart-home energy forecasting using moving average
Anomalies detection for smart-home energy forecasting using  moving averageAnomalies detection for smart-home energy forecasting using  moving average
Anomalies detection for smart-home energy forecasting using moving average
 

More from Power System Operation

Thermography test of electrical panels
Thermography test of electrical panelsThermography test of electrical panels
Thermography test of electrical panelsPower System Operation
 
Big Data Analytics for Power Grid Operations
Big Data Analytics for Power Grid OperationsBig Data Analytics for Power Grid Operations
Big Data Analytics for Power Grid OperationsPower System Operation
 
SPS to RAS Special Protection Scheme Remedial Action Scheme
SPS to RAS Special Protection Scheme  Remedial Action SchemeSPS to RAS Special Protection Scheme  Remedial Action Scheme
SPS to RAS Special Protection Scheme Remedial Action SchemePower System Operation
 
SVC PLUS Frequency Stabilizer Frequency and voltage support for dynamic grid...
SVC PLUS Frequency Stabilizer Frequency and voltage support for  dynamic grid...SVC PLUS Frequency Stabilizer Frequency and voltage support for  dynamic grid...
SVC PLUS Frequency Stabilizer Frequency and voltage support for dynamic grid...Power System Operation
 
Principles & Testing Methods Of Earth Ground Resistance
Principles & Testing Methods Of Earth Ground ResistancePrinciples & Testing Methods Of Earth Ground Resistance
Principles & Testing Methods Of Earth Ground ResistancePower System Operation
 
Gas Insulated Switchgear? Gas-Insulated High-Voltage Switchgear (GIS)
Gas Insulated Switchgear?  Gas-Insulated High-Voltage Switchgear (GIS)Gas Insulated Switchgear?  Gas-Insulated High-Voltage Switchgear (GIS)
Gas Insulated Switchgear? Gas-Insulated High-Voltage Switchgear (GIS)Power System Operation
 
Electrical Transmission Tower Types - Design & Parts
Electrical Transmission Tower  Types - Design & PartsElectrical Transmission Tower  Types - Design & Parts
Electrical Transmission Tower Types - Design & PartsPower System Operation
 
The Need for Enhanced Power System Modelling Techniques & Simulation Tools
The Need for Enhanced  Power System  Modelling Techniques  &  Simulation Tools The Need for Enhanced  Power System  Modelling Techniques  &  Simulation Tools
The Need for Enhanced Power System Modelling Techniques & Simulation Tools Power System Operation
 
Power Quality Trends in the Transition to Carbon-Free Electrical Energy System
Power Quality  Trends in the Transition to  Carbon-Free Electrical Energy SystemPower Quality  Trends in the Transition to  Carbon-Free Electrical Energy System
Power Quality Trends in the Transition to Carbon-Free Electrical Energy SystemPower System Operation
 

More from Power System Operation (20)

ENERGY TRANSITION OUTLOOK 2021
ENERGY TRANSITION OUTLOOK  2021ENERGY TRANSITION OUTLOOK  2021
ENERGY TRANSITION OUTLOOK 2021
 
Thermography test of electrical panels
Thermography test of electrical panelsThermography test of electrical panels
Thermography test of electrical panels
 
What does peak shaving mean
What does peak shaving meanWhat does peak shaving mean
What does peak shaving mean
 
What's short circuit level
What's short circuit levelWhat's short circuit level
What's short circuit level
 
Power System Restoration Guide
Power System Restoration Guide  Power System Restoration Guide
Power System Restoration Guide
 
Big Data Analytics for Power Grid Operations
Big Data Analytics for Power Grid OperationsBig Data Analytics for Power Grid Operations
Big Data Analytics for Power Grid Operations
 
SPS to RAS Special Protection Scheme Remedial Action Scheme
SPS to RAS Special Protection Scheme  Remedial Action SchemeSPS to RAS Special Protection Scheme  Remedial Action Scheme
SPS to RAS Special Protection Scheme Remedial Action Scheme
 
Substation Neutral Earthing
Substation Neutral EarthingSubstation Neutral Earthing
Substation Neutral Earthing
 
SVC PLUS Frequency Stabilizer Frequency and voltage support for dynamic grid...
SVC PLUS Frequency Stabilizer Frequency and voltage support for  dynamic grid...SVC PLUS Frequency Stabilizer Frequency and voltage support for  dynamic grid...
SVC PLUS Frequency Stabilizer Frequency and voltage support for dynamic grid...
 
Principles & Testing Methods Of Earth Ground Resistance
Principles & Testing Methods Of Earth Ground ResistancePrinciples & Testing Methods Of Earth Ground Resistance
Principles & Testing Methods Of Earth Ground Resistance
 
Gas Insulated Switchgear? Gas-Insulated High-Voltage Switchgear (GIS)
Gas Insulated Switchgear?  Gas-Insulated High-Voltage Switchgear (GIS)Gas Insulated Switchgear?  Gas-Insulated High-Voltage Switchgear (GIS)
Gas Insulated Switchgear? Gas-Insulated High-Voltage Switchgear (GIS)
 
Electrical Transmission Tower Types - Design & Parts
Electrical Transmission Tower  Types - Design & PartsElectrical Transmission Tower  Types - Design & Parts
Electrical Transmission Tower Types - Design & Parts
 
What is load management
What is load managementWhat is load management
What is load management
 
What does merit order mean
What does merit order meanWhat does merit order mean
What does merit order mean
 
What are Balancing Services ?
What are  Balancing Services ?What are  Balancing Services ?
What are Balancing Services ?
 
The Need for Enhanced Power System Modelling Techniques & Simulation Tools
The Need for Enhanced  Power System  Modelling Techniques  &  Simulation Tools The Need for Enhanced  Power System  Modelling Techniques  &  Simulation Tools
The Need for Enhanced Power System Modelling Techniques & Simulation Tools
 
Power Quality Trends in the Transition to Carbon-Free Electrical Energy System
Power Quality  Trends in the Transition to  Carbon-Free Electrical Energy SystemPower Quality  Trends in the Transition to  Carbon-Free Electrical Energy System
Power Quality Trends in the Transition to Carbon-Free Electrical Energy System
 
Power Purchase Agreement PPA
Power Purchase Agreement PPA Power Purchase Agreement PPA
Power Purchase Agreement PPA
 
Harmonic study and analysis
Harmonic study and analysisHarmonic study and analysis
Harmonic study and analysis
 
What is leakage current testing
What is leakage current testingWhat is leakage current testing
What is leakage current testing
 

Recently uploaded

VIP Call Girls Service Kondapur Hyderabad Call +91-8250192130
VIP Call Girls Service Kondapur Hyderabad Call +91-8250192130VIP Call Girls Service Kondapur Hyderabad Call +91-8250192130
VIP Call Girls Service Kondapur Hyderabad Call +91-8250192130Suhani Kapoor
 
Processing & Properties of Floor and Wall Tiles.pptx
Processing & Properties of Floor and Wall Tiles.pptxProcessing & Properties of Floor and Wall Tiles.pptx
Processing & Properties of Floor and Wall Tiles.pptxpranjaldaimarysona
 
AKTU Computer Networks notes --- Unit 3.pdf
AKTU Computer Networks notes ---  Unit 3.pdfAKTU Computer Networks notes ---  Unit 3.pdf
AKTU Computer Networks notes --- Unit 3.pdfankushspencer015
 
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...ranjana rawat
 
Java Programming :Event Handling(Types of Events)
Java Programming :Event Handling(Types of Events)Java Programming :Event Handling(Types of Events)
Java Programming :Event Handling(Types of Events)simmis5
 
Introduction to Multiple Access Protocol.pptx
Introduction to Multiple Access Protocol.pptxIntroduction to Multiple Access Protocol.pptx
Introduction to Multiple Access Protocol.pptxupamatechverse
 
result management system report for college project
result management system report for college projectresult management system report for college project
result management system report for college projectTonystark477637
 
UNIT-II FMM-Flow Through Circular Conduits
UNIT-II FMM-Flow Through Circular ConduitsUNIT-II FMM-Flow Through Circular Conduits
UNIT-II FMM-Flow Through Circular Conduitsrknatarajan
 
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service Nashik
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service NashikCall Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service Nashik
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service NashikCall Girls in Nagpur High Profile
 
KubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghlyKubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghlysanyuktamishra911
 
(MEERA) Dapodi Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Escorts
(MEERA) Dapodi Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Escorts(MEERA) Dapodi Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Escorts
(MEERA) Dapodi Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Escortsranjana rawat
 
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...Dr.Costas Sachpazis
 
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...ranjana rawat
 
(ANJALI) Dange Chowk Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(ANJALI) Dange Chowk Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...(ANJALI) Dange Chowk Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(ANJALI) Dange Chowk Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...ranjana rawat
 
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur EscortsHigh Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escortsranjana rawat
 
Software Development Life Cycle By Team Orange (Dept. of Pharmacy)
Software Development Life Cycle By  Team Orange (Dept. of Pharmacy)Software Development Life Cycle By  Team Orange (Dept. of Pharmacy)
Software Development Life Cycle By Team Orange (Dept. of Pharmacy)Suman Mia
 
UNIT-III FMM. DIMENSIONAL ANALYSIS
UNIT-III FMM.        DIMENSIONAL ANALYSISUNIT-III FMM.        DIMENSIONAL ANALYSIS
UNIT-III FMM. DIMENSIONAL ANALYSISrknatarajan
 
College Call Girls Nashik Nehal 7001305949 Independent Escort Service Nashik
College Call Girls Nashik Nehal 7001305949 Independent Escort Service NashikCollege Call Girls Nashik Nehal 7001305949 Independent Escort Service Nashik
College Call Girls Nashik Nehal 7001305949 Independent Escort Service NashikCall Girls in Nagpur High Profile
 
Introduction and different types of Ethernet.pptx
Introduction and different types of Ethernet.pptxIntroduction and different types of Ethernet.pptx
Introduction and different types of Ethernet.pptxupamatechverse
 

Recently uploaded (20)

VIP Call Girls Service Kondapur Hyderabad Call +91-8250192130
VIP Call Girls Service Kondapur Hyderabad Call +91-8250192130VIP Call Girls Service Kondapur Hyderabad Call +91-8250192130
VIP Call Girls Service Kondapur Hyderabad Call +91-8250192130
 
Processing & Properties of Floor and Wall Tiles.pptx
Processing & Properties of Floor and Wall Tiles.pptxProcessing & Properties of Floor and Wall Tiles.pptx
Processing & Properties of Floor and Wall Tiles.pptx
 
AKTU Computer Networks notes --- Unit 3.pdf
AKTU Computer Networks notes ---  Unit 3.pdfAKTU Computer Networks notes ---  Unit 3.pdf
AKTU Computer Networks notes --- Unit 3.pdf
 
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
The Most Attractive Pune Call Girls Budhwar Peth 8250192130 Will You Miss Thi...
 
Java Programming :Event Handling(Types of Events)
Java Programming :Event Handling(Types of Events)Java Programming :Event Handling(Types of Events)
Java Programming :Event Handling(Types of Events)
 
Introduction to Multiple Access Protocol.pptx
Introduction to Multiple Access Protocol.pptxIntroduction to Multiple Access Protocol.pptx
Introduction to Multiple Access Protocol.pptx
 
result management system report for college project
result management system report for college projectresult management system report for college project
result management system report for college project
 
UNIT-II FMM-Flow Through Circular Conduits
UNIT-II FMM-Flow Through Circular ConduitsUNIT-II FMM-Flow Through Circular Conduits
UNIT-II FMM-Flow Through Circular Conduits
 
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service Nashik
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service NashikCall Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service Nashik
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service Nashik
 
KubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghlyKubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghly
 
(MEERA) Dapodi Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Escorts
(MEERA) Dapodi Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Escorts(MEERA) Dapodi Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Escorts
(MEERA) Dapodi Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Escorts
 
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
 
Roadmap to Membership of RICS - Pathways and Routes
Roadmap to Membership of RICS - Pathways and RoutesRoadmap to Membership of RICS - Pathways and Routes
Roadmap to Membership of RICS - Pathways and Routes
 
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
 
(ANJALI) Dange Chowk Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(ANJALI) Dange Chowk Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...(ANJALI) Dange Chowk Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(ANJALI) Dange Chowk Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
 
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur EscortsHigh Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
 
Software Development Life Cycle By Team Orange (Dept. of Pharmacy)
Software Development Life Cycle By  Team Orange (Dept. of Pharmacy)Software Development Life Cycle By  Team Orange (Dept. of Pharmacy)
Software Development Life Cycle By Team Orange (Dept. of Pharmacy)
 
UNIT-III FMM. DIMENSIONAL ANALYSIS
UNIT-III FMM.        DIMENSIONAL ANALYSISUNIT-III FMM.        DIMENSIONAL ANALYSIS
UNIT-III FMM. DIMENSIONAL ANALYSIS
 
College Call Girls Nashik Nehal 7001305949 Independent Escort Service Nashik
College Call Girls Nashik Nehal 7001305949 Independent Escort Service NashikCollege Call Girls Nashik Nehal 7001305949 Independent Escort Service Nashik
College Call Girls Nashik Nehal 7001305949 Independent Escort Service Nashik
 
Introduction and different types of Ethernet.pptx
Introduction and different types of Ethernet.pptxIntroduction and different types of Ethernet.pptx
Introduction and different types of Ethernet.pptx
 

Challenges of Data Management & Analytics in the Future Grid

  • 1. tmcgrail@doble.com 21, rue d’Artois, F-75008 PARIS CIGRE US National Committee http://www.cigre.org 2019 Grid of the Future Symposium Challenges of Data Management & Analytics in the Future Grid T. MCGRAIL, B. GWYN J. BEARDSALL, M. ROWBOTTOM Doble Engineering Drax Power USA UK SUMMARY With a growth in interest in ‘big data’ as electric grids evolve and data sources become more common and more productive, there needs to be a discussion of the management of data in a secure manner, and the role of analytics to provide information and have ‘meaning’. This paper looks at a number of challenges that are beginning to be faced, and opportunities to ensure that the Future Grid is secure. Challenge 1 is the management of ‘big data’, which may provide value if appropriately viewed and analyzed; Challenge 2 is the management of security, for both data and systems which use the data; Challenge 3 is the need for appropriate urgency in analysis and action; Challenge 4 is to understand the meaning of the data and associated analyses, but also to understand the limits of our understanding. KEYWORDS Big data, Analytics, Sense-Making
  • 2. 2 CHALLENGE 1: DATA MANAGEMENT There has been over the last few years a growth in interest in the idea of ‘big data’ [1]. Occasionally the characteristics of ‘big data’ have been summarized in terms of the ‘V’s of ‘big data, some of which are collated in Table 1. Characteristic Implication Velocity High rates of data generation, collation and analysis Volume ‘incredible amounts’ of data to be managed Variety Structured v unstructured: think photo content Value The value may be virtual: unrealizable Veracity Trustworthiness - verisimilitude Table 1 Some of the various V’s of ‘big data’ We collect data to support decisions – through analytics such as level checking, presence/absence and other, often more exotic approaches [2][3]. With increasing pressure on assets to perform continuously, to reduce outages for maintenance or through unplanned issues, and within the context of Industry 4.0 bringing condition/operation sensor ubiquity and autonomy, the status quo is insufficient: we need to know not only where we are now, but where we will be, and when [4][5]. A CIGRE Technical brochure indicates some of the key approaches to improving knowledge of substation asset condition, and optimizing asset usage – the approaches it describes are ‘sensible’, from Partial Discharge (PD) and Electro- Magnetic Interference (EMI) surveys to increased visual inspection and appropriate condition monitoring [6]. Analysis of data brings challenges – not least in understanding new and exotic techniques related to Machine Learning (ML), deep-ML, and Artificial Intelligence (AI) – but also in terms of how to apply them: what is needed to make the techniques valuable? Google provides a glossary of terms, but not everyone will agree with the Google definitions [7]. It has also been noted that the application of exotic analytics requires very ‘clean’ and well presented data – to the point where 95% of the possible value can be derived from data through basic statistics and meaningful organization of the data [8]. The application of condition monitoring is, at its core, an asset management activity [9][10][11]. Many organizations are embracing the approaches and structures described in the international standard for asset management, ISO 55000 [12]. Originally published as three documents in the 55000 range, including management system requirements and application, further documents are in development, including, for example, ‘alignment of asset management, finance and accounting’ in ISO 55010 [13]. The development and evolution of the Common Information Model (CIM) has been instrumental in enabling organizations to both structure their own data and to allow communication of data in a standard format with other organizations [14]. Even here there is possible cause for confusion: does data relate to an asset or to a location in a grid system, or both? The load on a transformer is of interest to the grid system management, but also for the ageing of the individual asset.
  • 3. 3 A big focus of an ISO 55000 asset management approach has been the balance of cost, risk and performance [13][15]. These have been augmented with the concept of ‘sustainability’ in revised versions – indicating a desire for an asset management which plans for the future. In terms of risk management, condition monitoring is a decision support tool to reduce the likelihood of catastrophic events and thus reduce risk – condition monitoring addresses the cause of hazards, rather than the consequence. Analysis of risk is a key element of asset management, and thus also must be addressed at the condition monitoring level: how does the data generated help us review the asset condition and the consequent probability of failure? CHALLENGE 2: SECURITY Cyber security is a ‘hot topic’ in the industry at present, and rightly so. The consequences of cyber attacks can be devastating. Security applies not only to the data itself, and access to the data, but to management of the software systems and applications which generate, analyze and present the data and resulting information. For example, a Dissolved Gas Analysis (DGA) detector on a generator station auxiliary transformer may indicate the presence and severity of a possible problem, but a diagnosis of possible cause requires further analysis tools or sensors. Figure 1 shows a DGA detector in action, rising from a nominal 30 ppm to almost 80 ppm in about a week. The question is: do we believe the data and the resulting need for action? This is a very simple case, but if the plan is a de-energization of a power transformer as a result of such a rise, that has an impact on the grid as a whole: the recent events in the UK show the effects of a single lightning strike, which could be exacerbated by deliberate weakening of the grid’s resilience through cyber attacks on monitoring to necessitate the removal of highly inertial equipment, as discussed in the Financial Times of London [16]. Figure 1 Detector Type Monitoring of DGA Level CHALLENGE 3: TIME IS OF THE ESSENCE If assets did not deteriorate over time, they would not need maintenance, except to address external influences, and they would not need replacing except to address failures caused by external agents [17] [18]. But deterioration takes place through use/application and through interaction with the working environment over time: transformer paper insulation ages and becomes brittle with temperature, say, leading to replacement of a transformer before failure through mechanical forces.
  • 4. 4 We need to understand timescales of operation for failure modes – and the determination of parameters that indicate the presence and possible severity of a failure mode so that we can plan intervention in a timely manner, as per Figure 2 [18]. An analogy, using an automobile (car) can be useful: tire pressure reduces over time, eventually leading to the failure of the tire, and possible requirement for a new car. Figure 2 Modeling the Failure Mode Analysis Process The process may take weeks to reach a conclusion. Detection of low pressure can therefore be an indication of a problem to be addressed within a number of weeks - we know that as the pressure reduces the likelihood/probability of tire failure increases, even if we do not have exact values: we understand the urgency [18]. Power transformers are much more complex assets than tires, and with many more failure modes and more parameters to indicate symptoms, the overall condition assessment is much more complex [19]. In a world of ‘big data’ more and more operational and condition data may be available, but the collation of that data is complex and machine learning and artificial intelligence only provide clues as to what may be a true cause for concern: a true failure mode in operation. Combining data from multiple measurements, each of which may be relevant to multiple failure modes, and then combining failure modes to assess a power transformer is not a simple process. To ensure that a sense of urgency is not lost, assessments must be calibrated for time and combined to reflect the true condition and intervention requirements of the asset. The means of reducing data to allow for a concise assessment for comparative purposes, includes the process of generating an Asset Health Index: and a health index which loses the sense of urgency is of little value in practice [20][21][22].
  • 5. 5 CHALLENGE 4: ANALYTICS AND ACTION An Asset Health Index is a useful place to start discussion of ‘meaning’ in data. Often a numeric value, an AHI is apparently simple to understand. But what does the number mean? We believe we understand numbers – we know which are bigger, and which are smaller and have some grasp of relative size – so how hard can it be? We are at risk of succumbing to the Dunning-Kruger Effect [23]: our estimation of our capability is not in proportion to our actual knowledge and capability. Hence the old saying: “A little learning is a dangerous thing.” The health index may take a vast amount of data, from the asset itself, from sensors, from maintenance history and produce a number which should be calibrated for time, monotonic in scale and urgency, auditable in terms of having a clear path from raw data to condition/health, justifiable in terms of the analysis used [19]. And yet many indices are none of those things – they just seem to be a way to collate a lot of data and produce a useful number: and we understand numbers, so we think we understand the analysis and the derivation… the result is often a health index which is disappointing and an asset management approach which is unjustified. Machine Learning is growing in popularity [7][8]. But we may suffer from the same Dunning-Kruger effect: the analysis of data leads to a conclusion, but is the conclusion valid? The application of machine learning to PD/EMI spectra has been beneficial – even to the point of surpassing experts in their analyses of raw data [24], and as has been noted, much of the benefit of Machine Learning can be derived from good data clean up and standard statistical approaches [8]. It is important to have a plan to respond to anomalous or unusual data: in condition monitoring a response plan is a ‘best practice’: a response to a fire alarm should be to follow the plan in place at the particular location, not have a discussion and work out an approach ‘on the fly’. But plans need to address the variability in data sources and applications – in monitoring a human heart rate, for example, a value of 140 may be considered very high, unless there was strenuous exercise being undertaken, in which case we may be more concerned about the duration at this level. CONCLUSION Ultimately, we must be aware of the ‘meaning´ inherent in the data we use for decisions: and also be aware of our own limitations in giving meaning to data, or to resulting parameters used for decision. ACKNOWLEDGMENTS The authors would like to thank their many colleagues for conversations and discussion on matters related to the content of this paper; you know who you are. BIBLIOGRAPHY [1] https://www.xsnet.com/blog/bid/205405/the-v-s-of-big-data-velocity-volume-value- variety-and-veracity [2] T. McGrail, D. Angell & K. Elkinson, “Asset Management - Data, Analysis & Decisions”, Electric Light & Power magazine, Volume 18, Issue 7, July 2013 [3] Tony McGrail, “Asset Management: Data and Decisions”, IEEE International Smart Grid Conference, Perth Western Australia, 2011
  • 6. 6 [4] “Machine Condition Monitoring Market”, B2B Market Research, www.marketsandmarkets.com [5] “Industry 4.0: Opportunities and Challenges of the Industrial Internet”, Price Waterhouse Coopers, https://www.pwc.com/gx/en/technology/pdf/industrial-internet-of-things.pdf [6] CIGRE Technical Brochure 660 “Saving Through Optimized Maintenance in Air Insulated Substations”, 2016 [7] https://developers.google.com/machine-learning/glossary/ [8] https://www.kdnuggets.com/2018/04/dirty-little-secret-data-scientist.html [9] T. McGrail, R. Heywood, “Intelligent Transformers: Monitoring, smart grid integration and asset management”, Coil Winding Insulation and Electrical Manufacturing Exhibition, Berlin, Germany, 2014 [10] T. McGrail, K. Elkinson, M. Lawrence et al, “Aspects of Power Transformer Asset Management”, IEEE PES T&D Conference, Orlando, USA, 2012 [11] K. Wyper, G. MacKay, TransGrid Australia & T. McGrail “Condition Monitoring in the Real World” 80th International Conference of Doble Clients, Boston, USA, 2013 [12] ISO 55000 “Asset Management – Overview, Principles and Terminology”, 2014 [13] https://en.wikipedia.org/wiki/ISO_55000 [14] Common Information Model: https://www.dmtf.org/standards/cim [15] T. McGrail, "Practical Steps to Improve Asset Management Applications", TrafoTech Conference, New Delhi, India, 2016 [16] https://www.ft.com/content/bcb9f3b2-c352-11e9-a8e9-296ca66511c9 [17] https://www.through-life-engineering- services.org/downloads/27302_White_paper_amended.pdf [18] "Condition Monitoring: Data Driven & Knowledge Based Analyses", T. McGrail & K. Elkinson, CIGRE Canada, Vancouver, 2016 [19] “Developing & Using Justifiable Asset Health Indices For Tactical & Strategic Risk Management”, McGrail, Rhoads, White, CEPSI Conference, 2018, Kuala Lumpur, Malaysia [20] “Transformer Health Indices & Probability of Failure”, McGrail, Kennedy, Lawrence, TechCon Australia/NZ, Sydney, Australia, 2017 [21] “Turning an Asset Health Index into a Probability of Failure”, McGrail, 85th International Conference of Doble Clients, Boston, USA, 2018 [22] “Translating an Asset Health Index into a Probability of Failure”, CEATI report 30113, 2018 [23] "Unskilled and Unaware of It”, Kruger & Dunning, Journal of Personality and Social Psychology. 77 (6): 1121–1134, 1999 [24] “AI approaches to interpretation of electrical CM data”, Dr. I. Mitiche, BINDT 16th International Conference on Condition Monitoring & Asset Management, Glasgow, UK, 2019