Method for predicting outlet temperature of disk-type solar heat collector based on GA-RBF

Abstract

The invention provides a method for predicting an outlet temperature of a disk-type solar heat collector based on GA-RBF. The method comprises the steps that historical data is normalized according to the historical data of atmospheric temperature, humidity, solar irradiation and air flow rates, and a training sample is established; an RBF artificial neural network is established; a genetic algorithm is used to optimize initial values of parameters of the RBF artificial neural network; a gradient descent method is used to learn the training sample, and a network weight W, a Gaussian function center vector C and a sound stage width vector B of the RBF artificial neural network can be obtained through training; and a trained RBF artificial neural network prediction model is used to predict the outlet temperature of the disk-type solar heat collector. The method provided by the invention has the advantages that related factor data of the outlet temperature of the disk-type solar heat collector is acquired effectively; the genetic algorithm and the gradient descend method are used to optimize the parameters of the RBF artificial neural network; and learning efficiency and prediction accuracy are high.

Description

technical field [0001] The invention relates to a GA-RBF-based method for predicting the outlet temperature of a dish solar heat collector. Background technique [0002] The outlet temperature of the collector is an important factor affecting the stability and efficiency of the dish solar thermal power generation system, and its correct prediction can well grasp the performance of the dish system. Accurately predicting the outlet temperature of dish collectors is a arduous and complicated task, and there are few domestic studies on it at present. Since the outlet temperature of dish solar collectors is affected by air temperature, air pressure, solar radiation and other weather conditions, there is a high degree of nonlinearity, randomness and complexity. Therefore, it is difficult to model the outlet temperature of the dish solar collector, and the inaccurate prediction will bring difficulties to the dish solar power generation. [0003] The above problems should be consi...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a GA-RBF-based method for predicting the outlet temperature of dish solar collectors, which has higher precision and better generalization ability, and solves the problem of outlet temperature of dish solar collectors in the prior art. Modeling is difficult and prediction is inaccurate, which brings difficulties to dish solar power generation

Images