论文部分内容阅读
本文介绍了最小二乘支持向量机的原理,并针对热舒适性指标建立了最小二乘支持向量机预测模型,以人的新陈代谢率、衣服热阻、空气温度、相对湿度、平均辐射温度和空气流速作为输入变量,以PMV指标作为输出。该模型计算结果与Fanger方程的计算结果吻合很好,与BP神经网络模型及传统的支持向量回归机模型进行分析比较的结果表明最小二乘支持向量机模型具有较高的拟合精度和泛化能力,可以满足PMV指标作为被控参数对空调系统进行实时控制的要求。
In this paper, the principle of least square support vector machine is introduced. And the least squares support vector machine prediction model is established for the thermal comfort index. The least squares support vector machine is used to predict the metabolic rate, the thermal resistance of the clothes, the air temperature, the relative humidity, the average radiation temperature and the air The flow rate is used as the input variable, with the PMV indicator as output. The results of the model are in good agreement with those of the Fanger equation. The results of comparison with the BP neural network model and the traditional support vector regression model show that the least squares support vector machine model has higher fitting accuracy and generalization Ability to meet the PMV indicators as a controlled parameter of the air-conditioning system for real-time control requirements.