基于多模型IMC-PID的三水箱液位控制算法研究

Research on the Algorithm of Three-tank Liquid Level Control Based on Multi-model IMC-PID

针对具有非线性、大惯性、时滞性的三容水箱液位过程系统,通过建立多个模型和设计对应的IMC-PID控制器来获取更好的系统性能指标。根据系统的非线性程度,在不同液位高度分别建立多个局部线性系统模型来拟合全局非线性系统模型;采用最小二乘算法和粒子群算法分别对模型参数进行辨识;采用内模算法对所辨识的多个局部线性系统模型分别设计相应的局部控制器,并转化为PID形式。通过Matlab软件仿真可以看出,基于多模型IMC-PID的三水箱液位过程控制算法改善了系统的性能指标。

For a three-tank liquid level process system with nonlinearity, large inertia, and time delay, to obtain better system performance indicators, establishment of multiple models and design of the corresponding IMC-PID controller are done. According to the degree of nonlinearity of the system, multiple local linear system models are established at different liquid level heights to fit the global nonlinear system model. The least squares algorithm and particle swarm algorithm are used to identify the model parameters. For the identified local linear system models, the internal model algorithm is used to design the corresponding local controllers which then converts into PID form. Matlab software simulation shows that the three-tank liquid level processcontrol algorithm based on multi-model IMC-PID improves the performance of the system.