Simulation of AC drive control for supercapacitor trams based on high-order neural network pattern discrimination algorithm

Abstract

This paper presents an in-depth study and analysis of the AC drive control simulation of a supercapacitor tram using a high-order neural network pattern discrimination algorithm. Firstly, the line conditions and shunting locomotive operation conditions of a freight coal loading station are analyzed, the capacity of the onboard supercapacitor energy storage device is estimated using traction simulation calculations, the supercapacitor is formed by the series–parallel connection of supercapacitor monoliths group module, and the mathematical model of the supercapacitor energy storage system is obtained. In response to the existence of fuzzy controllers with human-set parameters and potential errors when writing fuzzy rules, the self-learning capability possessed by the adaptive neuro-fuzzy network (ANFIS) is used to improve the accuracy of the prediction model by training and learning to the model a large amount of experimental data with the ANFIS controller, and the simulation model of the dynamic setting of the supercapacitor voltage threshold of the urban rail energy storage system is built using the software, and the comparison and analysis show that the traction network voltage amplitude under the adaptive neuro-fuzzy network control strategy is smaller than the traction network voltage amplitude under the fuzzy control strategy. The global searchability of the optimization algorithm is ensured. In the middle of the search, a medium-precision simplex is used for parallel local search to prevent the searcher from skipping the global optimum in the process of rapid convergence. In the later stage of the search, the optimization range is already within the global optimal range. Subsequently, the advantages and disadvantages of various locomotive topologies are analyzed by using software, and the topology of the supercapacitor shunting locomotive is analyzed according to the design requirements and the actual situation, and the supercapacitor power system scheme of the shunting locomotive is designed, and the traction electric drive system of the supercapacitor shunting locomotive is designed, and its main structural performance parameters are determined. Finally, the overall modeling of supercapacitor shunting locomotive is carried out, including the whole vehicle dynamics model, automatic charging model, supercapacitor model, and motor model, and then, simulation analysis is carried out based on building a relatively perfect model, followed by further stimulation through software to verify the rationality of the design.