动态矩阵控制算法的仿真研究及PLC应用

Simulation Research and PLC Application of Dynamic Matrix Control Algorithm

针对具有大迟延、大惯性、时变和非线性等特性的主汽温系统,传统PID已无法满足工业生产的需要。采用动态矩阵控制(DMC)算法,对主蒸汽温度系统进行控制仿真,并利用西门子S7-300 PLC的S7-SCL语言,将DMC算法封装成可供用户调用的功能块,完成了DMC算法的PLC实现。为了增强主蒸汽温度控制系统的抗干扰性,将DMC算法与PID控制策略相结合,设计了DMC-PID串级控制结构,以充分发挥DMC对大延迟对象适应能力强和PID抗干扰能力强的优点。针对主汽温系统时变和非线性的特性,在DMC算法中添加了模型切换策略,完成了对主汽温系统多模型控制的仿真,实现了不同模型之间平滑切换。引入性能指标,通过粒子群算法对DMC控制参数进行优化,结合国电智深EDPF-NT DCS系统,对荥阳电厂主汽温系统进行控制试验。试验结果表明,DMC算法在主汽温控制系统中具有着良好的控制效果,提高了系统动静态性能指标。

For the main steam temperature system which features large delay, large inertia, time varying and nonlinear characteristics, the traditional PID control is unable to meet the demand for the industrial production. The dynamic matrix control （DMC） algorithm is adopted to implement the simulation of control. By using SIEMENS S7-SCL language of S7-300 PLC,the DMC algorithm is encapsulated into function block （ FB） function blocks which users can call, and the PLC implementation of DMC algorithm is completed. In order to enhance the anti-interference performance, a DMC-PID cascade control structure is designed by combining the DMC algorithm with the PID control strategy, thus advantages of both DMC and PID are fully played, i. e. ,the adaptability to large delay objects and the strong anti-interference capability. In view of the time-varying and nonlinear characteristics of main steam temperature system, the model switching strategy is added to the DMC algorithm, the simulation of multi-model control is completed, and the smooth switching between different models is realized. Particle swarm optimization algorithm is used to optimize the DMC control parameters. With the Guodian Zhishen EDPF-NT DCS as the object, control experiment is carried out for main steam temperature system of Xingyang power plant. The results show that the DMC algorithm has good control effect for the main steam temperature control system, and improves the dynamic and static performance indexes of the system.