ENGI.AI

The AI platform for the Net Zero

The Net Zero companies have a Data Science Problem

Hi! I’m Dave and I’m here to share my work on ENGI.AI.

• The energy market has a data science problem.

Data is fueling a Clean Tech Boom

“The majority of buildings in the UK have a smart meter” - BEIS

Reporting

Generation

Automation

EV Charging

• As smart meter uptake increases, so does the amount of data available
• Energy management and reporting
• Home Automation and IoT
• Electric Vehicle Charging
• Domestic Solar PV and Battery Storage
  • This information is being actively used to achieve Net-Zero by energy management and reporting companies
  • Most of the products need some form of forecasting the expected energy demand so that it can be met in the greenest possible way
  • This is also true for the people who run the grid.

Demand for skills > Supply of people

40% of companies find it “Very Difficult” to hire Data Scientists with the right skills - Energy Systems Catapult

• Very difficult to hire data scientists, no offence to the JavaScript/PHP/Ruby devs, but we are pretty out-numbered
  • Hard to find with correct skills
    • Tend to have a minimum of an Undergraduate Degree in Engineering or Mathematics
    • Important to have Energy Market domain knowledge AND general software development skills in addition to Data Science Experience
• This is what is making it hard to scale data science in the energy market
  • And therefore also the products that rely on them

Forecasting is foundational

“Forecasting using Time Series is the most common data science technique” - Energy Systems Catapult

• Forecasting is the most common data science objective
  • according to 95% of data scientists in the energy market
• The reasons for forecasting are quite common
  • Trading
  • Management
  • Efficiency
  • Storage
  • Generation
• All of these are reliant on, or can be enhanced by, good forecasting

From Network…

Operate and maintain

• OFGEM is pushing for Distribution Network Operators to become Distribution System Operators
• Distribution Network Operators are responsible for operation and maintenance of the distribution network

… to System

Operate and maintain

Active Management

Distributed Energy Resources

Enable Customers

• A DSO’s responsibility expands on a DNO’s
  • As a DSO they need to become more actively involved
    • Monitor, control and actively manage the power flows
    • Enable Distributed Energy Resources on distribution networks
    • Enable customers to be producers, consumers and storers of energy
  • In order to control and actively manage the power flows, they are going to need an accurate forecast of the demand, at a more granular level than they have now
• This is need is also shared by virtual power plant operators
  • Who need to be able to forecast the demand of their aggregated assets

ENGI.AI

Benefits of ENGI.AI

Drop in integration

Modern techniques

Scales with demand

No survey required

• Consumption Demand Forecasting for each building
• No expensive, time consuming survey required
• Learns from the buildings as they improve
• Easy to integrate into IoT, SaaS, BMS, and the National Grid
• Not reliant on surveys
• Stays up to date with the buildings changes in construction and operation
• Informed by deep domain knowledge
  • Using the most accurate techniques available
• Scales better than
  • a simulation per building
  • an excel sheet per employee
  • a data science team per company

ENGI.AI

Artificial Intelligence Platform for Net Zero

ENGI.AI

Artificial Intelligence Platform for Net Zero

ENGI.AI

Artificial Intelligence Platform for Net Zero

• Energy forecasting as a service
  • Plug and play into your existing software
• AI doesn’t need a description of the building
  • Uses the most current Machine Learning tools and techniques
  • Learns from the buildings smart meter data
• Designed to be integrated
  • Apps
  • IoT/BMS
  • Network mechanisms

Platform for Net Zero Innovation

Demand Side Response Systems

Grid Connected Energy Storage

Time of Use Flexible Energy Tariffs

Electric Vehicle Charging Control

• This integrated system is valuable for multiple Net Zero innovations
  • Sophisticated demand side response to help support load balancing becomes possible
    • Solves peak power carbon intensity
  • Smart storage control algorithms to improve utilisation of stored energy becomes widely available
    • Enhances the value of battery storage and renewable generation
  • Increase the confidence of consumers with flexible energy products
    • Improving adoption of time of use tariffs
  • Electric vehicles can be charged in a distributed manner more effectively
    • Improving the value of EVs and mitigating for the impact of them on grid infrastructure

ENGI.AI accelerates the transition

• And because it’s designed to be integrated, it’s ready to go when you are with minimal effort through our REST based API
  • Because the climate crisis isn’t waiting for anyone

• Eventually moving into revenue generation there should be various suitable models for pricing. Depending on where and what type of market demand can be captured.

Where is ENGI.AI now?

Progress & Status

• Starting in 2023, Engi.AI has moved rapidly through a deployed Alpha product and is about to launch for general release

Contact

• engi.ai

• dave@engi.ai

• engi-ai

• @engi_ai

• engi-ai.github.io/engi-ai-pitch

• Here’s a lot of information on where to find me in one slide.

Sources + Notes

Speaker notes can be read by pressing the ‘s’ key.

• Data Science Skills in the Energy Sector - Energy Systems Catapult
• Smart Meters in Great Britain - Department for Business, Energy & Industrial Strategy
• The Data Science Venn Diagram - Drew Conway
• UK Data Science Salary - Glassdoor.com
• OFGEM Glossary