Optimization of the thermal lag Stirling engine performance

Abstract

In this paper, neural network and genetic algorithm is used to obtain the optimal output power of thermal lag Stirling engine. A neural network is trained and developed using the theoretical data of previous literatures in order to predict the performance of Stirling engine. Input parameters to neural network include angular velocity, thermal resistance, stroke length radius, piston diameter, the volume of heat buffer chamber and the volume of gas chamber, and output parameter includes output power. The accuracy of neural network is evaluated by average square error and regression analysis. Also, genetic algorithm is used for the optimization of the output power of the Stirling engine. The results of present study show that the neural network can be used as an appreciate tool to predict the output power of the thermal lag Stirling engine with a high precision and speed. The main deficiency of thermal lag type of Stirling engines is low output power. However, the optimization of design parameters of thermal lag Stirling engine causes an increase of 86.9% in output power.