论文部分内容阅读
针对变风量空调系统具有非线性和难以建立准确模型的问题,将神经网络与预测控制结合,采用了预测滚动优化控制算法训练多层前向神经网络,并将其作为优化反馈控制器来求解变风量空调系统的风量最优解。仿真结果表明,房间温度能够稳定在设定值,风量能跟随室温的变化调整,在加入扰动的情况下,系统表现出较好的稳定性和抗干扰性,实现了对房间温度的优化控制。
Aiming at the problem that the VAV air conditioning system is non-linear and difficult to establish an accurate model, the neural network is combined with the predictive control. The predictive rolling optimization control algorithm is used to train the multi-layer forward neural network, which is used as the optimal feedback controller to solve the problem Air volume of air conditioning system optimal solution. The simulation results show that the room temperature can be stabilized at the set value, and the air volume can be adjusted according to the change of room temperature. Under the condition of adding perturbation, the system shows good stability and anti-interference and realizes the optimal control of the room temperature.