Sensitivity analysis of output performance of free piston expander-linear generator based on artificial neural network

In order to optimize the output performance of the free-piston expander-linear generator for waste heat recovery, a method based on artificial neural network prediction is proposed. Taking output performance as an index, sensitivity analysis is carried out on five input parameters, namely intake pressure, intake temperature, intake duration, expansion duration and exhaust duration. The results of the range analysis of orthogonal experiment are compared and verified. The results show that the influence degree of input factors is intake pressure, intake temperature, intake duration, expansion duration and exhaust duration, respectively. The optimal input factor parameter configuration is intake pressure 0.3 MPa, intake duration 40 ms, expansion duration 70 ms, exhaust duration 70 ms and intake temperature 30 ℃, peak output voltage up to 8.96 V, peak output power up to 33.74 W. This method is suitable for studying the sensitivity of parameters and can provide a valuable reference for the subsequent parameter optimization and system improvement.