Capacity Optimization Configuration of Aircraft Energy Storage Units Based on Particle Swarm Optimization and Neural Network Algorithm

In order to reduce the problem of excessive capacity allocation of energy storage units and achieve comprehensive optimization of aircraft power supply system volume, weight, and price, this paper proposes a capacity optimization configuration method that combines particle swarm optimization (PSO) with BP neural network algorithm. This method first establishes a capacity allocation strategy, and on this basis, uses PSO algorithm to provide the capacity optimization configuration results of the energy storage unit. Part of the optimization results are used to train the BP neural network model, while the other part is used to verify the optimization results of the BP neural network. Then, the BP neural network model is trained using the data from the PSO algorithm to obtain the mapping relationship between the energy storage unit parameters and the optimization configuration results. When optimizing new energy storage units, this mapping can be directly utilized without the need for intermediate capacity allocation strategies and particle swarm optimization algorithm steps. Compared with PSO algorithm, this method can avoid the problem of non-convergence of calculation results, improve calculation accuracy, reduce optimization iteration process, and significantly shorten calculation time. The simulation results verify the correctness of the method.