非线性系统的蚁群优化预测PID控制

针对非线性、时变及大惯性系统的控制问题,提出了一种基于蚁群算法的预测PID控制算法。该算法以神经网络作为预测模型,将预测控制和PID控制相结合,并用蚁群算法在线优化控制器参数,其中以常规的Ziegler-N ichols方法整定的控制器参数为基础,选取蚁群优化变量的动态搜索区间。该算法考虑了控制能量受限情况下,非线性系统的预测控制问题。计算机仿真结果表明,该非线性控制方案具有较好的鲁棒性,相对传统PID控制策略还表现出了良好的动态性能,能够满足对再热汽温对象的控制要求。

PredictiVe PID Control Based on Ant Colony Algorithm Optimization for Nonlinear System

For the control of nonlinear,time-varying and big inertia system,a predictive PID control strategy based on ant colony algorithm(ACA) is presented.Artificial neural network(ANN) is used as the predictive model and ACA is adopted to optimize the controller parameters online by combining the predictive control structure with PID control.The variable searching region is set on the basis of the parameters based on Z-N methods.The control strategy can be used to the nonlinear systems with control energy constraint.The computer simulation result shows that the nonlinear control strategy has more favorable dynamic characteristics and strong robustness than traditional PID control in reheated steam temperature system.