基函数自适应选择的预测函数控制方法在混凝投加中的应用

摘要

The traditional algorithm of predictive functional control is fixed. When the control environment is not stable, the control quality of the basis function is not ideal. This paper proposes an adaptive selection algorithm of the basis function to improve the control quality. According to the errors between the output value and the expectation, the base function is selected. All the errors can be divided into several zones, with each zone corresponding to a basis function respectively. Through the calculation of the errors to the areas, the corresponding basis function is selected. For delay reason, this paper establishes a model of coagulant dosage based on soft measurement method in the process of coagulant dosage. The model is applied to forecast the output of the process. Matlab simulation shows that this algorithm has better stability and fast response ability. It has better control quality than the traditional predictive functional control algorithm.%传统预测函数控制算法的基函数是固定的，在控制环境不稳定的情况下控制品质不理想，本文提出一种通过对基函数的自适应选择来改善控制品质。本文根据过程输出值与期望值的误差来选择基函数，将全体误差分为几个区域，每个区域分别对应一个基函数，通过计算误差落选区域来选择对应的基函数。针对混凝投药过程的时滞原因，本文利用软测量方法建立一个混凝投药模型，应用该模型预测过程的输出。Matlab仿真表明此算法具有更好的平稳性和快速响应能力，比传统预测函数控制算法具有更好的控制品质。