Performance Prediction of a Stirling heat engine using Artificial Neural Network model

Résumé fourni par la source

Global energy use has increased significantly over the past few years. This increase is as a result of several factors which include growth in population, improved living standards and the development of the trade and commercial industry. With the world's increased reliance on fossil fuels, various environmental issues have surfaced. Several scholars in energy-related research have recommended the adoption of renewable energy as an alternative energy source. However, Stirling engines are among the devices developed by engineers to counter some of the environmental and social implications of fossil fuels. In this study, artificial neural network (ANN) model has been implemented to predict a Stirling heat engine system power and torque. The ANN model used a sigmoid activation transfer function to obtain the optimum architecture for this prediction problem. Python is used to build and train the ANN model and the performance of the algorithm was adjudged using the root mean square error and the coefficient of determination R2, Based on the analysis, it was observed that a 3-10-1 ANN model gave a good prediction of the engine's torque and power.