Increasing the sustainability of manufacturing processes
The Epsilon project supports manufacturing companies in effective and efficient energy management by developing a digital and analytics solution. Our modular and scalable AI solution empowers energy managers to:
Increase energy independence
Reduce energy costs
Reduce environmental footprint
Increase compliance with regulations
Context
Tackling grid congestion and increasing energy costs
Energy is a key resource for many companies to maintain processes running. As the world is facing an energy crisis highlighted by the ongoing grid congestion, organizations are being forced to take a more active role in their energy management.
The increased demand, supply volatility and high energy costs emphasize the need for a transition towards more sustainable energy sources. At the same time, stay in control of energy resources to help stay competitive.
Our approach
To address these challenges, we are developing in the Epsilon project a energy optimization solution.
With several European partners, including a leading steel manufacturer, we are developing a digital and analytics solution tailored for manufacturing companies. Our approach includes energy mix optimization, peak energy load shifting and shaving, and smart energy storage allocation.
The Epsilon project is made possible with funding from:
Impact of grid congestion on manufacturing
The manufacturing sector, particularly the energy-intensive industries, faces challenges in increased energy costs and operations expansions, caused by grid congestion. In order to expand operations, it could take of up to five years or more for new heavy grid connections.
Energy-intensive industries have a substantial impact on both final energy consumption and greenhouse gas emissions. Combined, these subsectors account for:
of total energy consumption in industry
share of total industrial GHG emissions
Epsilon supports manufacturing companies with complex energy management problems by balancing energy supply and demand
Added value and high-level solution architecture of the digital and analytics solution
Visualisation layer
Analytics layer
Energy production prediction
Machine Learning model that accurately forecasts generation from local renewable sources.
Energy Consumption Estimation
Machine Learning model that accurately estimates energy consumption production facilities.
Scheduling Engine
Mathematical optimization model that balances energy supply, demand and storage.
Data layer
Achieving positive economic impact...
- Energy cost reduction
- Additional revenue generation
- Improved energy efficiency
- CO2 reduction
... through
- Energy mix optimization
- Improved production and energy scheduling
- Smart allocation of energy storage
- Peak energy load shifting / shaving
Achieving postive economic impact...
- Energy cost reduction
- Additional revenue generation
- Improved energy efficiency
- CO2 reduction
... through
- Energy mix optimization
- Improved production and energy scheduling
- Smart allocation of energy storage
- Peak energy load shifting / shaving
Visualisation layer
Analytics layer
Energy production prediction
Machine Learning model that accurately forecasts generation from local renewable sources.
Energy Consumption Estimation
Machine Learning model that accurately estimates energy consumption production facilities.
Scheduling Engine
Mathematical optimization model that balances energy supply, demand and storage.
Data layer
Use case
Energy mix optimization results suggest 15% - 17% annual cost saving potential
01.
Challenge
To support Voestalpine on their ambitious roadmap to carbon neutrality, we leveraged data science techniques to identify opportunities to optimize the energy supply mix (grid, PV, storage).
02.
Solution approach
Build a modular energy orchestration solution providing the following functionality:
-
PV generation forecast Built machine learning model trained on weather data measurements to accurately predict renewable energy production from photovoltaic (PV) systems.
-
Energy mix optimization Built mathematical model to determine the optimal balance between energy supply (grid and local PV energy forecast), energy consumption, and local battery storage.
-
Energy peak shaving Identified scenarios to reduce demand of grid electricity and achieve cost savings through smart allocation of PV energy and battery storage.
03.
Simulation of energy mix optimization
Optimization results reveal 15% - 17% cost saving potential for electricity usage on a annual basis.
04.
Benefits
Increase energy independence
Reduce energy costs
Reduce environmental footprint
Get in touch
Do you have questions?
Please reach out.
We would like to hear your questions or suggestions about our solutions. Just schedule a meeting with one of our expert!