Dendrite neural network scheme for estimating output power and efficiency for a class of solar free-piston Stirling engine

This paper presents the dendrite neural network (DNN) scheme for predicting two challenging parameters, namely the output power and efficiency of the solar free-piston Stirling engine (FPSE). This paper utilizes vital characteristics to predict the output power and efficiency accurately. MSE and regression analysis are employed to investigate the reliability of the proposed DNN structure. As a result, not only was the MSE close to zero, but the regression outcomes were also very close to 1, which confirmed the reliability of the DNN. In the next step, several experimental case studies are used to study the effectiveness of the proposed DNN more thoroughly. Therefore, a maximum error of 5% was observed, confirming good agreement between the empirical data and the predictions made by the DNN. Lastly, through the simulation and experimental investigations, it can be concluded that the method is reliable and can be used for other types of FPSEs.