论文部分内容阅读
提出了一种新颖的具有预测误差校正的多变量广义预测控制 (GPC)与动态矩阵控制 (DMC)混合集中优化控制算法。该算法直接对被控对象的整体模型性能指标进行优化 ,可获得最全面的反映过程未来变化趋势的动态信息量 ,以实现整个系统的最优控制。同时考虑了所有输入输出之间的交互影响 ,合理考虑各控制输入输出量在算法中的地位 ,避免常规多变量控制系统设计中因解耦而造成的不可实现性和解耦网络对系统模型的过分依赖性 ,且易于实施 ,对模型的变化有较强的鲁棒性。通过对 30 0MW单元机组协调控制系统进行仿真 ,结果表明 ,该算法具有控制性能好、鲁棒性强等优点
A novel multivariable generalized predictive control (GPC) and dynamic matrix control (DMC) hybrid centralized optimization control algorithm with predictive error correction is proposed. The algorithm directly optimizes the overall model performance index of the controlled object and obtains the most comprehensive dynamic information volume that reflects the trend of the process in the future so as to realize the optimal control of the whole system. At the same time, the interaction between all the inputs and outputs is taken into account, and the position of each control input and output in the algorithm is reasonably considered, so as to avoid the unrealizability and decoupling caused by decoupling in the design of the conventional multivariable control system. Overdependence, easy to implement and robust to model changes. Through the simulation of the coordinated control system of 30MW unit, the results show that this algorithm has the advantages of good control performance and robustness