基于Bang-Bang控制的智能开关预测控制算法

Intelligent Bang-Bang Predictive Control Algorithm Based on Bang-Bang Control

为解决Bang-Bang控制原理具体应用中存在的如何纳入状态变量的附加约束条件、开关曲线的求取和最优轨线的确立等复杂问题,提出了一种智能开关预测控制方法.该方法根据离散卷积定理提炼出一组新的控制信号集,然后设计新的切换方程和减速规律,并在开关控制框架下,实施有阈值信号制约的最小拍增量控制.本文所提出的方法全面改善了预测控制的策略和算法,使控制系统动态性能良好,过渡过程时间小于理想开关控制过渡过程,且单调衰减,原理上稳态无静差,系统反应时间短,定位精度高,对输出扰动具有一定的鲁棒性,而且适用范围更加广泛.仿真结果表明了该控制算法的有效性.

In order to solve the complex problems of adding the additional constraints of state variables, getting the Bang-Bang curve and determining the optimal path in the application of the Bang-Bang control theory an intelligent Bang-Bang predictive control method is proposed. In this method, a new set of control signals is first presented according to the discrete convolution principle. Next, a new switching equation and a new deceleration law are constructed. Then, the deadbeat addition control restricted by threshold signals is accomplished with the frame of Bang-Bang control. The proposed method improves the strategy and algorithm of the predictive control, guarantees the dynamic performance of the control system, makes the transition time reduce to a value less than that in ideal Bang-Bang control, and enables the predictive control system to attenuate monotonously without static errors. Moreover, it is of the advantageous of quick reaction, high precision, good robustness to the output turbulence and wide scope of application. Its effectiveness is finally verified by simulated results.