Dropout Deep Neural Network Assisted Transfer Learning for Bi-Objective Pareto AGC Dispatch

To balance the unexpected power disturbances, an independent system operator (ISO) should assign the dynamic power regulation commands to all the regulation resources via an automatic generation control (AGC) dispatch. It can be described as a bi-objective Pareto optimization by considering the minimizations of total power deviation and total regulation mileage payment. In this work, a novel dropout deep neural network assisted transfer learning (DDNN-TL) is proposed to rapidly approximate the high-quality Pareto optimal solutions for AGC dispatch. Firstly, the training data is generated from the Pareto optimal solutions and fronts obtained by various multi-objective optimization algorithms according to the anticipated total power regulation commands. Secondly, the network parameters of DDNN will be updated via an off-line training with these data at each frequency regulation service period. Finally, based on an efficient transfer learning with a correction of infeasible solutions, DDNN-TL can directly approximate the high-quality Pareto optimal solutions for on-line decision of AGC dispatch. Case studies of DDNN-TL for bi-objective Pareto AGC dispatch are carried out on a two-area load frequency control model and Hainan power grid of China Southern Power Grid (CSG), which demonstrates its superior performance on optimization speed and stability compared with other multi-objective optimization algorithms.

Automated summary

To balance power disturbances, an independent system operator (ISO) assigns power regulation commands to resources via AGC dispatch. This paper proposes a novel DDNN-TL for rapid approximation of high-quality Pareto optimal solutions in AGC dispatch. Training data is generated from Pareto optimal solutions obtained by various algorithms. DDNN parameters are updated through offline training with this data. The results demonstrate superior performance in optimization speed and stability compared to other algorithms.