基于主成分分析的短期PV/T组件温度预测

Short-term PV/T component temperature prediction based on Principal Component Analysis

为进一步提高光伏/光热(photovoltaic/thermal,PV/T)综合利用系统中PV/T组件温度预测精度,该文采用主成分分析法对原始输入样本数据进行预处理,提取主成分,并利用遗传算法优化BP神经网络结构,对组件温度数据建立预测模型。仿真结果表明,相对于未经主成分提取的神经网络,该方法使得网络模型在晴天和多云天气条件下的预测精度分别提高了7.68%和4.97%,使得网络模型预测精度更高,泛化性能更强。

In order to further improve the prediction accuracy of PV/T module temperature in photovoltaic-thermal(PV/T)system,a principal component analysis method is used to preprocess the original input sample data to extract the main component.Then combined with the BP neural network theory and Genetic optimization algorithm,the prediction model of component temperature data is built.The simulation results show that the proposed method makes the prediction accuracy of the network model higher while the generalization performance stronger than that of the neural network without the main component extraction.The prediction accuracy is increased by 7.68%and 4.97%in sunny and cloudy weather.