Conference Day Three | Thursday 27th September 2018

08:00

Welcome refreshments

08:45

Welcome back from the chair

09:00

Power Generation and Demand Forecasting Use Case – increasing forecasting accuracy through more sophisticated data sourcing and analysis in the transition to a more complex generation and consumption landscape

  • Using new insights about consumption in tandem with heightened grid awareness to achieve better demand foresight
  • Translating end-user behaviour data into meaningful insight about patterns of consumption based on a variety of factors
  • Building models to predict the impact of newer developments such as intermittent generation or EV uptake
  • Identifying and accounting for relevant external factors such as the impact of temperature and home insulation on heating/cooling requirements
  • Combining profiles for multiple sources of generation and changes to consumption patterns to build an overall picture of the demand landscape
  • Reducing penalties for load imbalance and increasing resilience of the network in the face of a changing energy mix

Jon Black, Load Forecasting Manager – ISO New England

09:45

Settlement Use Case – effectively processing data from a variety of energy market participants for a more accurate and timely settlement process

  • Using data from each grid domain to make the settlement process more precise and quick to complete
  • Developing the infrastructure to access and prepare data streams from multiple organisations
  • Combining the relevant data and performing calculations more quickly to shorten the timetable for final settlement and move to more precise settlement periods
  • Increasing the level of automation in the settlement process while improving accuracy and reducing costs for users

Liga Sadovica, Head of Data Analysis – Augstprieguma tīkls

10:30

Morning refreshments, exhibition and networking

11:00

Active Performance Enhancement Use Case – identifying the data sources and analytics models that produce KPIs which maximise performance efficiency and extend equipment lifespan

  • Leveraging more accurate asset data to identify critical metrics, actively monitor asset life-cycle, and respond accordingly to fluctuations in performance
  • Installing monitoring capabilities to generate precise data on more parameters pertaining to asset condition
  • Analysing data from a large sample of assets to minimise the impact of anomalous behaviour on predictions
  • Identifying the key conditions and predictors which govern asset life-cycle, degradation, and failure
  • Moving from routine to targeted maintenance, informing changes to increase asset life-cycle, and optimising grid upkeep

Borsu Shahnavaz, Innovation Analyst – UK Power Networks

11:45

Managing Data Gaps – leveraging AI driven tools to accurately interpolate incomplete data and produce accurate outcomes even with the most sensitive analytics models

  • Evaluating techniques to carry out accurate and reliable data analytics despite incomplete data sets
  • Making reasonable assumptions to decide when imputation of missing data is valid over deletion based on a variety of conditions
  • Comparing different methods for predicting missing data including statistical and machine learning tools such as spectral decomposition and clustering
  • Using more advanced machine learning and AI technology to optimise predictive techniques
  • Discussing cost-effective workflows to extend analysis to larger energy data sets

Chris Park, Advisor to Data Science Advisory Committee – UK Data Archive

Darren Bell, Repository Architect – UK Data Archive

12:30

Lunch, exhibition and networking

14:00

Machine Learning & AI – understanding the potential of AI for processing and interpreting large volumes of multi-dimensional data to support a more complex and dynamic future grid environment

  • Leveraging artificial intelligence tools with the ability to recognise complex patterns and give valuable insight into otherwise unpredictable behaviours
  • Creating deep learning algorithms capable of processing information vastly beyond the capabilities of human intelligence
  • Building large data sets to “teach” machine learning systems to recognise and evaluate the full range of factors which impact network performance
  • Safeguarding against unpredictable behaviour to ensure confidence in handing over critical decision-making functionality
  • Utilising AI technology to facilitate automated decision making in the grid while considering the ever-increasing range of variables impacting demand

Ana Filipa Ribeiro, Project Manager – EDP

14:45

Blockchain – leveraging blockchain technology to better manage distributed data sources inherent to a decentralised energy system

  • Storing data from behind the meter assets on distributed ledgers to assist DSO support for innovations such as EVs, microgeneration, and battery storage
  • Fully understanding blockchain technology and making informed decisions about its application in the smart utility big data infrastructure based on a full appraisal of its benefits and drawbacks
  • Facilitating the integration of a wide variety of peers into blockchain platforms through the development of APIs
  • Reducing vulnerabilities by ensuring the security of any legacy systems writing data onto blockchains
  • Navigating questions of compliance such as the immutability of distributed ledgers and the “right to be forgotten”
  • Providing real time information to DSOs about the consumption and behaviour of decentralised assets and facilitating innovations in the energy system

Prof David Shipworth, Professor of Energy and the Built Environment – UCL Energy Institute

15:30

Afternoon refreshments, exhibition and networking

16:00

Big Data in the Cloud – integrating cloud services with internal big data infrastructure to rapidly scale up use cases and flexibly meet future analytics requirements

During this 90-minute tutorial, the team from Microsoft will provide an in-depth insight into how Cloud services are being utilised within the smart utility environment, to maximise speed of deployment, take advantage of advanced analytics functionalities, and launch new use cases cost effectively. Key issues that will be addressed include:

  • Making the case for cloud platforms as a powerful means to store data and perform more powerful analytics in a cost-effective manner
  • Assessing different analytics challenges and choosing between on-premises, private cloud, third-party cloud, and hybrid solutions where appropriate
  • Choosing the right provider for cloud services to deliver the required technical capabilities while providing reliable support and adequate security measures
  • Integrating multiple systems across the business into the chosen platform in a time- and cost-efficient manner without impeding day-to-day operations
  • Foreseeing and mitigating potential data governance issues before they become critical problems
  • Utilising the cloud’s increased capacity for parallel processing to flexibly support advanced analytics tools as and when required

Tutorial Leaders:

Bas Van Dorst, Principal Solution Specialist, Data & Artificial Intelligence – Microsoft Advanced Analytics

Ebisa Negeri, Data Solution Architect – Microsoft Advanced Analytics

 

17:30

Close of conference day three

▲ Back to top