为提高集中供暖系统的热效率，提出了一种基于神经网络的集中式供暖系统的水阀开合度控制策略，引入了障碍李雅普诺夫函数和一个特殊设计的辅助系统来解决控制系统中出现的输入输出约束的问题。采用了一种神经网络逼近器来在线估计系统模型，解决系统传递模型未知的问题。通过模拟仿真同时测试了所提出系统的性能，与传统PID控制系统进行对比，该系统具有良好的适应性和稳定性，有一定的实用价值。

In order to improve the thermal efficiency of central heating system,a water valve opening and closing control strategy based on neural network is proposed.An obstacle Lyapunov function and a specially designed auxiliary system are introduced to solve the problem of input and output constraints in the control system.A neural network approximator is used to estimate the system model online to solve any unknown problem in the system transfer model.The performance of the proposed system is also tested by simulations and experiments.Compared with the traditional PID control system,the proposed system has both great adaptability and stability,and in turn,has certain practical value.

TU832.1+2

国家重点研发计划(2017YFB0604000)；辽宁省自然科学基金（20170540747）