Abstract
Electrical distribution networks are facing an energy transition which entails an increase of decentralised renewable energy sources and electric vehicles. The resulting temporal and spatial uncertainty in the generation/load patterns challenges the operations of an infrastructure not designed for such a transition. In this situation, Optimal Power Flow methods can play a key role in identifying system weak points and supporting efficient management of the electrical networks, including the distribution level. In this work, to support distribution system operators’ decision-making process, we aim at attaining a quasi-optimal solution in the shortest time possible in an electrical network experiencing a large growth of distributed energy sources. We propose an optimisation method based on a modified version of a genetic algorithm and the Python pandapower package. The method is tested on a model of a real urban meshed network of a large Czech city. The optimisation method minimises the total operating costs of the distribution network by controlling selected network components and parameters, namely the transformer tap changers and the active power demand at consumption nodes. The results of our method are compared with the exact solution showing that a close-to-optimal solution of the observed problem can be reached in a relatively short time.
Highlights
Conceived several decades ago to supply electrical power generated by few large power plants to end consumers, power systems have recently undergone radical changes
By running 100 random instances of the problem, we were able to reach 90% precision of results for all the tests, meaning that all the results were close to the optimal solution with an absolute error of less than 10%
It is worth mentioning that our intuition to use both information from the feasible and the non-feasible domain in the genetic algorithm (GA) is quite effective
Summary
Conceived several decades ago to supply electrical power generated by few large power plants to end consumers, power systems (mainly transmission and distribution systems) have recently undergone radical changes. Apart from managing outages, the existence of local markets where prosumers can make bids/offers to exchange electrical power is becoming increasingly possible, which makes real-time operation of the DS necessary To cope with these tasks, DSOs will need to be able to perform OPF analyses in the shortest possible time. We present a method for solving an AC OPF problem, optimising the operation of a real electrical distribution network (powering a part of a Czech city). Our approach, based on the genetic algorithm (GA), allows us to find an optimal solution effectively and efficiently To ensure this fact, the GA was compared to the particle swarm optimisation (PSO) algorithm, which is commonly used to solve OPF problems [6].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
