Investigation of Support Vector Machine and Back Propagation Artificial Neural Network for performance prediction of the organic Rankine cycle system

Low temperature power generation system based on organic Rankine cycle (ORC) has been a popular candidate for low grade heat utilization and recovery. To find a way to predict the performance of the ORC system, the exploration and analyses of the Support Vector Machine (SVM) and Back Propagation Artificial Neural Network (BP-ANN) were carried out. For comparison, both Gauss Radial Basis kernel function (SVM-RBF) and linear function (SVM-LF) have been employed in SVM. Additionally, for the sake of comprehensiveness, two division methods for data set called “random division method” and “blocked division method” were studied. Finally, SVM-LF and BP-ANN demonstrated better stability and higher accuracy for both two division methods and for different testing sets while SVM-RBF showed good results for random division method and disappointing results for blocked division method.