模型预测控制在网络闭环控制中的应用研究

The Research of Network Closed-loop Control Based on the Model Predictive Control

在基于Internet的远程控制系统中,通过Internet进行信息传输的时间延时的不确定性对实时闭环控制系统的设计带来较大的困难.研究了网络传输延时对网络闭环控制系统的影响,提出了通过基于神经网络的预测控制方法解决存在网络随机延时的控制系统闭环控制的可行性.仿真结果表明该方法能够反映并预测该测量数据所代表的网络路径之间的时延特性,并能在基于Internet的闭环控制系统的设计中有效地替代实际网络进行研究;而所用的方法具有快速、准确的特点,能用于在线学习网络模型并对网络的时延值进行预测,为基于Internet的远程闭环控制提供了新的思路.

Because of the uncertainty of network time-delay in the remote control system based on the lnternet, which will bring greater difficulty to the design of the close＆loop control system. The influence of the network transmission delay on the closed loop control system is researched, the predictive control method based on the neural network is proposed in the paper. The method provide the feasibility to solve the random delay of network control system. The simulation results show that the model can effectively substitute the actual network in Internet-based closed loop control system research under time-delay condition; the methods used in the modeling, due to the virtue of speediness and exactness, is capable of learn and predict the network time-delay on line, this gives a new idea in Internet-based remote close＆loop control.