论文部分内容阅读
实际生产过程控制中,特别是多变量耦合的复杂过程中,对某些被控变量其实并没有苛刻的给定值要求,只是满足在给定区间内即可,从而保证整个生产过程控制的平稳。针对这一问题,提出一种基于模型预测控制理论的区间控制算法。该算法对预测控制的优化性能指标进行改进,将性能优化指标中的参考轨迹作为约束变量进行在线优化处理;参考轨迹自动跟踪被控变量输出,使预测控制针对输出在不同的区域范围采取不同的控制强度,从而在实现区间控制的前提下满足平稳控制要求。最后以Shell公司的典型重油分馏塔控制问题为例进行仿真,验证算法的有效性和可行性。
In actual production process control, especially in the complex process of multivariable coupling, there are actually no harsh given requirements for some controlled variables, just to meet the given interval, so as to ensure the smooth control of the entire production process . Aiming at this problem, an interval control algorithm based on model predictive control theory is proposed. The algorithm optimizes the optimal performance of predictive control and optimizes the reference trajectory in the performance optimization index as a constrained variable. The reference trajectory automatically tracks the output of the controlled variable so that the predictive control can take different output ranges in different regions Control strength, so as to achieve the control of the interval under the premise of a smooth control requirements. Finally, the simulation of a typical heavy oil fractionation tower in Shell Company is carried out to verify the effectiveness and feasibility of the algorithm.