Prediction of the temperature in a Chinese solar greenhouse based on LSSVM optimized by improved PSO

Abstract

Predictions of the Chinese solar greenhouse temperatures are important because they play a vital role in greenhouse cultivation, with solar greenhouse crops susceptible to potential losses because of cold and hot temperatures. Therefore, it is important to set up a precise predictive model of temperature that can predict the occurrence of temperatures several hours before head to reduce financial losses. This paper presents a novel temperature prediction model based on a least squares support vector machine (LSSVM) model with parameters optimized by improved particle swarm optimization (IPSO). The IPSO with probability of mutation was employed to optimize the required hyper parameters of the LSSVM model. The performance of the IPSO–LSSVM model was compared with traditional modeling approaches by applying it to predict solar greenhouse temperatures, and the results showed that its predictions of the maximum and minimum temperature were more accurate than those of the standard support vector machine (SVM) and Back propagation neural network (BPNN). Therefore, it is a suitable and effective method for predicting the Chinese solar greenhouse temperatures.