OUTLET TEMPERATURE PREDICTION OF PARABOLIC TROUGH SOLAR FIELD BASED ON HYBRID NEURAL NETWORK

Abstract

In order to solve the problem that the actual field conditions of the trough solar collector field cannot meet the existing thermal performance test standards，a method of predicting the outlet temperature of the trough solar collector field based on hybrid neural network（ALSTM）model is proposed in this paper. Firstly，the experimental data and meteorological data of dynamic thermal performance of trough solar heat collecting fields in Institute of Electrical Engineering Chinese Academy of Sciences- Yanqing ，Changzhou Longteng Wulat Middle Qi，Delingha of China General Nuclear Power Corporation and Wulat Middle Qi of China Ship of Electric Power are preprocessed and correlation analysis is carried out to form training sample data and verification sample data. Secondly，the network model is trained and optimized，and finally verified. It is concluded that the maximum relative errors between the actual and predicted outlet temperatures of the four heat collection fields are 3.00%，0.31%，1.45% and 1.95%，respectively. It is proved that the prediction accuracy of this model is high，and it provides a new method for the actual field thermal performance prediction and evaluation of trough solar heat collecting field.