基于RBF的碟式太阳能集热器温度预测研究  

Research on Temperature Prediction of Dish-Solar Collectors Based on RBF

作　　者：朱正林[1] 吴昊[1] 郑健[1] 

机构地区：[1]南京工程学院能源与动力工程学院,江苏南京211167

出　　处：《南京工程学院学报（自然科学版）》2017年第2期67-71,共5页Journal of Nanjing Institute of Technology(Natural Science Edition)

基　　金：南京工程学院科研基金(CKJB201207)

摘　　要：影响碟式太阳能集热器出口温度的因素较多,采用传统数学方法模拟较为复杂.利用RBF神经网络建立碟式太阳能集热器出口温度预测模型,为提高RBF的预测精度和学习效率,采用最近邻聚类算法选取基函数的中心,应用实际数据进行网络训练,网络预测结果较为准确.将本算法与传统的RBF神经网络进行仿真预测对比,本算法的结果和算法学习效率都要好于传统的RBF神经网络,验证了该算法的可行性和有效性.It is not easy to use traditional mathematical method for simulation in that there are many factors affecting the temperature of disc-solar collectors. This paper, therefore, proposes a model for predicting the temperature of dish-solar collectors by using RBF neural network. In order to improve the prediction accuracy of RBF, this paper uses the nearest neighbor-clustering algorithm to select the basis function center. Finally, actual data are selected for training the network and the result is accurate. The comparison between simulation prediction from this algorithm and that from traditional RBF neural network shows that the results and efficiency from the former are better than those from the latter. This algorithm is thus proved to be feasible and effective.

关 键 词：碟式太阳能 RBF神经网络 最近邻聚类算法 

分 类 号：TK519[动力工程及工程热物理—热能工程]