基于支持向量机的中央空调节能控制技术研究

Research on Central Air Conditioning Energy-saving Control Technology based on Support Vector Machine

针对目前传统的中央空调节能控制方法和支持向量机(SVM)空调负荷预测方法各自的优缺点,提出了一种基于SVM的中央空调节能控制方法。SVM使用传统的中央空调节能控制系统采集的训练样本数据进行负荷预测,控制系统根据预测结果进行控制调整,SVM根据调整结果增加新的训练样本、或对训练样本进行修正,进而对回归函数进行修正。重复该迭代过程,可不断提高预测和控制的准确度和实时性,进而在保证基本需求(如舒适度)的同时达到最佳的节能效果。

According to the advantages and disadvantages of the traditional central air conditioning energy-saving control method and Support Vector Machine（SVM） air conditioning load prediction method, a central air conditioning system energy-saving control method based on SVM is presented： based on the training samples acquisited by traditional air conditioning system energy-saving control system, air conditioning load is predicted by SVM; and based on the prediction results, the control adjustion is made by the control system; then based on the adjustion results, the new training samples are added, or training samples are corrected,and the regression function is corrected. By repeating the iterative process,the accuracy and real time of prediction and control are improved, and the best energy-saving effect is reached under the guaranteed comfort.