Science Summary:
\nTraditionally, electric power systems were operated with large stability margins. The main task of control room operators was to maintain the balance of generation and demand by controlling large centralized power plants.\n
\nNowadays, ambitious goals to reduce greenhouse gas emissions drives the transition from fossil fuel based electricity generation to renewable energy sources (RES), such as wind and photovoltaics (PV). In addition, the electrification of transportation introduces additional large loads mostly connected at the lower voltage levels, e.g. electric vehicles (EVs) connected at residential homes. As the power output of RES cannot be controlled and large-scale deployment of EVs radically changes power flows in distribution but also transmission grids, the operation of electric power systems becomes more challenging.\n
\nIn order to keep up with these changes, existing tools that support control room operators in operating the power system need to be enhanced and additional tools need to be developed to complement them. Having the right tools to monitor the power system, and to support the control room operator when making important operational decisions is a fundamental requirement for appropriate situation awareness.\n
\nThe dissertation focused on three selected aspects to increase the situation awareness and improve decision support of control room operators.\n
\nThe thesis proposes:\n
\n a tool for preventive re-scheduling of generators to maintain the power system in a state with sufficient security margins to be able to withstand possible contingencies.\n
\n a data-driven neural network approach for voltage estimation in distribution grids with high penetration of dispersed RES-based generation.\n
\n a model which allows to aggregate large numbers of EVs and to harness their flexibility, i.e. controlling the charging behavior by shifting the demand in time.\n
\nAll three covered aspects are aimed at improving the situation awareness of the operator, and to enable advanced controllability of distributed resources. On the one hand, this will give the operator more knowledge about the state of the power system leading to better decisions. On the other hand, extended controllability will allow enhanced operation.\n
Supervisors:
\nPrincipal supervisor: Senior Scientist Henrik W. Bindner, DTU Elektro CEE, DTU\n
\nCo supervisor: Associate Professor Mattia Marinelli, DTU Elektro CEE, DTU\n
\nCo supervisor: Assistant Professor Kai Heussen, DTU Elektro CEE, DTU\n
\n
\nExaminers:\n
\nAssociate Professor Chresten Træholt, DTU Elektro CEE, DTU\n
\nAssociate Professor Roberto Turri, Università di Padova\n
\nAndrew Keane, Head of UCD Energy Institute, University College Dublin\n
\n
\nChairperson at defense:\n
\nAssociate Professor Nenad Mijatovic, DTU Elektro CEE, DTU\n
\n
\n
Science Summary:
\nTraditionally, electric power systems were operated with large stability margins. The main task of control room operators was to maintain the balance of generation and demand by controlling large centralized power plants.\n
\nNowadays, ambitious goals to reduce greenhouse gas emissions drives the transition from fossil fuel based electricity generation to renewable energy sources (RES), such as wind and photovoltaics (PV). In addition, the electrification of transportation introduces additional large loads mostly connected at the lower voltage levels, e.g. electric vehicles (EVs) connected at residential homes. As the power output of RES cannot be controlled and large-scale deployment of EVs radically changes power flows in distribution but also transmission grids, the operation of electric power systems becomes more challenging.\n
\nIn order to keep up with these changes, existing tools that support control room operators in operating the power system need to be enhanced and additional tools need to be developed to complement them. Having the right tools to monitor the power system, and to support the control room operator when making important operational decisions is a fundamental requirement for appropriate situation awareness.\n
\nThe dissertation focused on three selected aspects to increase the situation awareness and improve decision support of control room operators.\n
\nThe thesis proposes:\n
\n a tool for preventive re-scheduling of generators to maintain the power system in a state with sufficient security margins to be able to withstand possible contingencies.\n
\n a data-driven neural network approach for voltage estimation in distribution grids with high penetration of dispersed RES-based generation.\n
\n a model which allows to aggregate large numbers of EVs and to harness their flexibility, i.e. controlling the charging behavior by shifting the demand in time.\n
\nAll three covered aspects are aimed at improving the situation awareness of the operator, and to enable advanced controllability of distributed resources. On the one hand, this will give the operator more knowledge about the state of the power system leading to better decisions. On the other hand, extended controllability will allow enhanced operation.\n
Supervisors:
\nPrincipal supervisor: Senior Scientist Henrik W. Bindner, DTU Elektro CEE, DTU\n
\nCo supervisor: Associate Professor Mattia Marinelli, DTU Elektro CEE, DTU\n
\nCo supervisor: Assistant Professor Kai Heussen, DTU Elektro CEE, DTU\n
\n
\nExaminers:\n
\nAssociate Professor Chresten Træholt, DTU Elektro CEE, DTU\n
\nAssociate Professor Roberto Turri, Università di Padova\n
\nAndrew Keane, Head of UCD Energy Institute, University College Dublin\n
\n
\nChairperson at defense:\n
\nAssociate Professor Nenad Mijatovic, DTU Elektro CEE, DTU\n
\n
\n