Elman and back propagation neural networks based working fluid side energy level analysis of shell-and-tube evaporator in organic Rankine cycle (ORC) system

Xu Ping,Fubin Yang,Hongguang Zhang,Jian Zhang,Wujie Zhang
DOI: https://doi.org/10.1016/j.aej.2022.01.006
IF: 6.626
2022-09-01
Alexandria Engineering Journal
Abstract:As the heat exchange component of the organic Rankine cycle (ORC) system, the evaporator directly affects the overall operation performance of the system. In this paper, an analytical method for energy level on the working fluid side of evaporator is proposed based on the energy level and enerty theory. The reliability, validity, and correlation of the proposed analytical method are studied by means of theoretical analysis, experimental evaluation, and Elman neural network (ElmanNN). The bilinear interpolation algorithm is used to analyze the non-linear relationship between the system parameters and the energy level on the working fluid side. In addition, the correlation between the heat exchange efficiency of the evaporator and the operating performance of the system is compared. Based on the back propagation neural network (BPNN), the high energy level area on working fluid side is accurately evaluated and verified, the direction and degree of the non-linear mapping relationship between the working fluid side energy level and the system performance are quantitatively evaluated, and the sensitivity of thermodynamic cycle parameters in the high energy level area on working fluid side is quantitatively evaluated. This study provides a novel approach to evaluate the operation of the evaporator in the ORC system. In addition, this study also provides guidance on how to keep the evaporator operating continuously in the high energy level area throughout the experiment.
engineering, multidisciplinary
What problem does this paper attempt to address?