GT-SUITE software was used to build the ORC system simulation model. Based on the simulation model, the effects of single screw expander speed and multistage centrifugal pump speed on system performance were studied. Based on the data obtained by Design of experiment (DOE), using machine learning, a system performance prediction model is constructed. At the same time, the system performance is predicted. The optimal operating performance of the system is determined according to the prediction results. The results show that in the common speed range of single screw expander and multistage centrifugal pump, there is an optimal matching value so that the net power output of the system reaches the maximum, and the optimal expander speed and the optimal pump speed are maintained at lower speed.
Keywords Organic Rankine cycle,Design of experiment,Operating performance,Machine learning,Optimization;