Performance optimization method and device for gas generator set based on reinforcement learning

The invention relates to the technical field of gas generator electric control, and provides a gas generator set performance optimization method and device based on reinforcement learning, and the method comprises the steps: obtaining a one-dimensional simulation model corresponding to a gas generator set, setting model parameters corresponding to the one-dimensional simulation model, performing model calibration on the one-dimensional simulation model by using the model parameters to obtain a target gas model; collecting structure parameters and performance parameters by using the target gas model, and analyzing the influence degree of the performance parameters on the gas generator set according to the structure parameters; selecting a target performance parameter from the performance parameters based on the influence degree, and performing residual analysis and optimization on the target performance parameter to obtain an optimal performance parameter; matching parameters of a turbocharger corresponding to