Future Energy Systems Embracing AI
Find Similar History 11 Claim Ownership Request Data Change Add FavouriteTitle
CoPED ID
Status
Value
Start Date
End Date
Description
The UK has adopted a target to become carbon neutral by 2050, this is in line with broader EU targets which aim to reduce carbon emissions across much of the EU. A significant proportion of carbon emissions in the UK and Ireland comes from the energy sector, in order to achieve these ambitious targets it is expected that more renewable energy sources will be installed. However, the variability and uncertainty associated with this typically non-synchronous generation and the nature of the technology create many operational challenges including harmonics, low inertia, reversed power flows and islanding.
The project intends to explore Artificial Intelligence applied to future energy systems. The latest Machine Learning techniques will be applied to historic and simulated Synchrophasor data of known faults to determine more information about how and why these faults occur and the measures taken by the system operator in response to these faults. This will create a clearer picture surrounding events and allow better analysis and development of operational procedures for faults or planned maintenance. This project will investigate what measures can be taken to mitigate the problems introduced by renewable generation allowing more renewable capacity to be installed on the grid.
The objectives of the project are:
1. Investigation into concept and problems associated with power system events, monitoring and protection
2. To become familiar with various simulated and real scenarios, models and software (e.g. DigSilent)
3. Data acquisition using real time measurements
4. Acquire skill in the use of Machine Learning software such as TensorFlow, Scikit-learn, Pandas and apply deep learning analysis on Synchrophasor data
5. Optimise AI method for speed to allow reporting in real time with minimal lag
6. Investigate a methodology to compile and reports for operational presentation
7. Develop graphical user interface useful for system operators and method for integration with existing energy system monitoring technologies and software
8. Development of advanced event detection and diagnosis techniques
9. Publish novel research outcome in prestigious journals
| Queen's University Belfast | LEAD_ORG |
| Xueqin Liu | SUPER_PER |
Subjects by relevance
- Renewable energy sources
- Machine learning
- Artificial intelligence
- Emissions
- Energy production (process industry)
- Optimisation
- Simulation
Extracted key phrases
- Future Energy Systems
- Broad EU target
- AI method
- Carbon emission
- Carbon neutral
- Energy system
- Power system event
- Renewable energy source
- UK
- Ambitious target
- Late Machine Learning technique
- System operator
- Machine Learning software
- Energy sector
- Simulated Synchrophasor datum