Data Driven Surrogate Model Based Operation Quality Optimization Algorithm of Urban Transmission Network

In this paper, through the topology reconfiguration of the 110kV high-voltage distribution network, an optimization algorithm for the urban transmission network operation quality improvement based on the data-driven surrogate model is proposed. Firstly, a large number of system operation scenarios are generated by Markov chain Monte Carlo sampling (MCMC) and the corresponding urban power grid operation quality for each scenario is computed. Then the deep neural network (DNN) is used to fit the nonlinear relationship between operating parameters and operation quality. Finally, the DNN is embedded into the non-dominated sorting genetic algorithm (NSGA-II) as the fast evaluator of operation quality and the operation quality is improved by iteratively searching the optimal topology structure based on the proposed algorithm. The effectiveness of the algorithm is demonstrated in an urban transmission network.