Smart and flexible energy networks

 

Modelling and characterisation of “Active” buildings in the residential built environment

Camille Griton

In the context of an ever-increasing demand for electricity, with in particular the electrification of heat and transport, the grid is more and more stressed. With properly managed energy management schedules, 'Active' buildings can play a significant role to support this grid, making good use of their on-site renewable generation, electric storage, EVs... Today, there are many efficient Energy Management Systems at the scale of a building but not enough at the scale of a district. This paper aims to assess the impact of uncertainty in demand forecasting on the total cost of electricity supply for a district of Active buildings.

Supervisor:

  • Dr Edward O’Dwyer, Department of Chemical Engineering, Imperial College London

 

Enabling Sustainability and Resilience in Energy Systems via Flexibility in Design and Real Options

Guillermina Valenzuela

Designing energy systems with flexibility has been shown in many studies to improve significantly the economic performance of such systems. Using real options allows the system to adapt, evolve, and reconfigure in the face of changing conditions.

Supervisor:

  • Dr Michel-Alexandre Cardin, Dyson School of Design Engineering

 

Allocation of Electrical Load Disconnection in Highly Renewable Power Systems

María José Parajeles Herrera

Dealing with uncertainty in highly renewable power systems is one of the most important tasks of current energy engineering. For example, variable energy generation makes the planning and operation of the electricity grids more complex. This project contributes to dealing with uncertainty in the grid’s operation when facing generation and demand unbalance events. A new method for allocating load disconnection among distribution feeders was developed. It optimizes the selection of feeders to disconnect based on the uncertainty of the forecast of their net load; and, the fairness of the frequency of selection for disconnection.

Supervisors:

  • Dr Luis Badesa, Department of Electrical and Electronic Engineering, Imperial College London
  • Mr Cormac O’Malley, Department of Electrical and Electronic Engineering, Imperial College London
  • Prof Goran Strbac, Department of Electrical and Electronic Engineering, Imperial College London

 

Flexible Power Generation via Thermal Energy Storage Integration

Michael Chinonso Azih

A whole system assessment on the value of thermal energy storage integration to produce a flexible nuclear power fleet in South Africa

Supervisors:

  • Prof Christos Markides, Department of Chemical Engineering, Imperial College London
  • Dr. Marko Aunedi, Department of Electrical and Electronic Engineering, Imperial College London
  • Dr. Antonio Marco Pantaleo, Department of Chemical Engineering, Imperial College London
  • Abdullah Al Kindi, Department of Chemical Engineering, Imperial College London

 

System Analysis of Compressed Air Energy Storage for Grid-scale Electricity Storage

Nadia Assad

In the context of the ongoing race to decarbonise electricity grids, solving the issue of affordable and efficient large scale energy storage is becoming more critical than ever, in order for renewable energy to take up a large share of the energy mix without the risk of destabilising grids. This thesis focuses on improving the performance of compressed air energy storage systems for grid-scale electricity storage. By employing thermodynamic models of four Adiabatic CAES systems, the effect of various operating parameters on system performance was assessed to inform high-performance system designs.

Supervisors:

  • Professor Christos Markides, Department of Chemical Engineering, Imperial College London
  • Dr Paul Sapin, Department of Chemical Engineering, Imperial College London
  • Mr Matthias Mersch, Department of Chemical Engineering, Imperial College London

 

Artificial Intelligence for Electricity Price Forecasting: An Ensemble Approach

Owen O'Connor

Electricity grid stability requires a constant balance between supply and demand. This balance is becoming more volatile with the continuous penetration of renewable energy. An efficient solution to this challenge requires improved forecasting accuracy of grid imbalance and its causes. In this context, artificial intelligence (AI) models are being leveraged to improve predictions in electricity markets. Despite their success, no single AI model performs best at all times within the highly dynamic short-term markets. This research takes a collective approach, where top performing models are ensembled together to enhance individual models' strengths, resulting in superior forecasts.

Supervisors:

  • Dr. Sam Cooper, Dyson School of Design Engineering, Imperial College London
  • Steve Kench, Dyson School of Design Engineering, Imperial College London