基于GPC的增量PID算法改进及仿真研究

Improvement and Simulation of GPC-based Augmented PID Algorithm

针对广义预测控制(GPC)难以在DCS平台上应用的现状,有文献提出了一种基于GPC的增量PID算法。该算法利用多个常规DCS模块的组合实现了类似GPC的控制效果。然而该文献只讨论了针对一阶和二阶系统的增量PID算法,并且仿真对象的纯迟延较小,对于大迟延的高阶对象并未进行算法设计和仿真。基于此,笔者对其进行了改进,提出了针对三阶系统的增量PID算法,并对某600MW超临界机组过热汽温被控对象进行了仿真。仿真结果表明,改进后的算法仍具有类似GPC的控制效果,其控制品质明显优于常规PID,并且鲁棒性较强。另外,该改进算法完全利用常规DCS模块实现,因此具有较高的应用价值。

Since the Generalized Predictive Control （GPC） is difficult to applied to the DCS level at present, the GPC- based augmented PID algorithm is proposed. This algorithm utilizes multiple DCS blocks augmented to mimic GPC. The algorithm is designed only for the first-order and second-order processes, and has short time delay. As for the large time delay and high-order process, this algorithm is not involved. The augmented PID algorithm for the third-order process is proposed, and is applied to the simulation of one supercritical 600MW boiler＇s superheated steam temperature plant. The simulation results show that the control performance of this improved algorithm is similar to that of GPC, and is obviously better than that of the conventional PID control, and has strong robustness. The algorithm has high value of application for it is totally realized by the normal DCS blocks.