Deep neural network approach to estimation of power production for an organic Rankine cycle system

Abstract

In this study, the possibility of using Stepwise multilinear regression and deep learning models to estimate the behaviour of the organic Rankine cycle (ORC) has been investigated. It was found that a number of parameters affects the performance of the turbine and hence the amount of power obtained by the ORC. Therefore, limited and simulative parameters might not be sufficient to obtain the best prediction expression. In the present study, the data obtained from a 10 kW ORC system was used as the basis for deep learning models. To this end, the variable selection, which represents the inputs to the neural network, is included in the first steps of a stepwise multilinear regression (SMLR). The aim of the deep learning (DL) models is to use the capabilities of dense layers, and then to strengthen SMLR contributions. The main aim here was to estimate the power generation of the expander, which has an important role in deciding the ORC's performance. The present study is intended to act as a crucial resource for defining an active estimation procedure for the ORC system through the use of DL. Therefore, an interoperability framework is proposed to estimate ORC power production using SMLR and DL as a new approach in this study. The interoperability approach for the proposed models (SMLR and DL) was found to be successful.