摘要
基于BP神经网络建立了地板辐射供暖系统的预测控制模型,以实验实测数据训练预测控制模型,通过实验对训练好的预测控制模型进行在线修正预测控制,在线修正预测控制输出的室内温度与实验实测结果的相对误差最大为-6.2%,证明预测控制模型的预测控制效果良好。预测控制模型的应用可提高对地板辐射供暖系统的控制精度,使室内温度控制在设定范围内,保证了室内的热舒适性。
A model for predictive control of radiant floor heating system is established on the basis of BP neural network. The model is trained with the measured data of the experiment, and the on-line correction predictive control of the trained model is conducted through experiment. The maximum relative error between the indoor temperature given by the on-line correction predictive control and the measured result in the experiment is -6.2%, proving that the predictive control result of the model is good. The application of the model can improve the control accuracy of radiant floor heating system, and control indoor temperature within a given range, thus ensure indoor thermal comfortableness.
出处
《煤气与热力》
2006年第7期65-69,共5页
Gas & Heat
关键词
地板辐射供暖系统
间歇运行模式
BP神经网络
单步预测控制
在线修正
radiant floor heating system
intermittent running mode
BP neural network
single-step predictive control
on-line correction
