基于支持向量机算法的空调负荷预测及实验研究

Prediction and Experimental Investigation of Air Conditioning Load based on Support Vector Machine Algorithm

实行负荷预测是空气调节系统优化运行的基础,如何选择工程应用切实可行的方法,仍然是一个值得探讨和研究的问题。支持向量机(SVM)算法在解决小样本、非线性及高维模式识别中表现出许多特有的优势。本文将支持向量机算法引入空调负荷预测中,对深圳市夏季六、七月份的逐时空调负荷,分别用SVM模型和armax模型进行了训练和预测,结果表明SVM模型适用于空调负荷预测,具有很好的泛化能力。

Air conditioning load prediction is the basis of the optimal operation of the air conditioning system. It is still an important problem that how to choose the method of practical engineering application. Support vector machine (SVM) algorithm shows many unique advantages in tackling small sample, nonlinear and high dimensional pattern recognition. In the present study, the support vector machine (SVM) algorithm was introduced into the air conditioning load prediction, an SVM model and armax model were used for predicting the hourly air conditioning load in June and July in Shenzen city. The results show that, the SVM model is effective for predicting the air conditioning load, and it possesses generalization ability.