摘要
集中供热系统存在大惯性、非线性以及时变性等问题,为了保证集中供热系统末端用户的采暖质量,针对集中供热二次网系统,设计了二次网回水温度的预测模型及智能控制策略,以实现二次网回水温度的准确控制,满足末端用户的采暖需求。通过飞升曲线建立了温控系统的数学模型。设计了一个RBF神经网络预测模型,模型的输出作为温控系统的二次网回水温度给定值。在控制算法上,设计了三层前向型神经网络与PID相结合的智能控制器,实现对二次网回水温度的闭环控制。基于所建立的数学模型、预测模型及控制器进行了仿真实验。仿真结果表明,所采用的控制方法与常规PID控制相比,具有调节时间短,超调量小的优点。验证了所采用控制方法的可行性和有效性。
The central heating system has characteristics such as large inertia, nonlinearity and time varying. In order to ensure the quality of users at the end of the heating system, the article designs a prediction model and an intelligent control strategy for secondary net return water temperature control system. A mathematical model of temperature control system is established according to the rising curve. A RBF neural network prediction model is designed and the output of the model is regarded as input of return water temperature. An intelligent controller based on the combination of neural network and PID is designed. The simulation experiment is carried out based on the established mathematical model, prediction model and intelligent controller. The results show that this control method has shorter settling time and smaller overshoot compared with conventional PID control and verify the feasibility and effectiveness of the adopted control method.
出处
《控制工程》
CSCD
北大核心
2015年第2期291-295,共5页
Control Engineering of China
基金
吉林省科技支撑重大项目(20126040)
吉林省自然科学基金(201115151)
关键词
集中供热
回水温度
神经网络预测
智能控制
Central heating
return water temperature
prediction of neural network
intelligent control