基于余弦前馈的重复运动控制算法在水下仿生推进系统中的应用

Repetitive Motion Control Algorithm Based on Cosine Feedforward Method and Its Application to Underwater Bionic Propulsion System

常规比例-积分-微分(PID)控制算法在周期性正弦运动跟随上会产生幅值衰减和相位滞后,且负载越大其幅值衰减和相位滞后越严重,无法满足仿生推进运动的高精度跟随要求;针对周期性重复运动设计的重复控制器在受到外界扰动后,其产生的补偿输出量需滞后一个运动周期才能响应,无法适应水下变负载运行条件。基于此,文中针对某水下推进系统在变负载工况下对低频率正弦位置指令跟踪性能问题,在借鉴前馈控制思想的基础上,提出了余弦前馈补偿控制算法,与常规PID控制器一起,构成复合前馈控制来消除正弦跟随的幅值与相位误差,并进行了仿真与实验验证。仿真与实验结果表明,基于余弦前馈的重复运动控制算法较传统PID控制方法效果显著,在变负载工况下依然可以实现对给定正弦运动指令的精确跟随,相位差较补偿之前减小了60%左右。文中的研究可为实现仿生推进系统流畅精确的周期运动提供新的解决思路。

The conventional proportional integral derivative(PID) control algorithm will generate amplitude attenuation and phase lag in the periodic sinusoidal motion follow-up, and the larger the load is, the more serious the amplitude attenuation and the phase lag are, which makes the conventional PID control algorithm cannot meet the high-precision follow-up requirement of the bionic propulsion motion. And for a repetitive controller designed for periodic repetitive motion, its compensation output may be delayed for one motion cycle once external disturbance exists, which leads the controller not to be applicable for the underwater variable load operating conditions. The above controllers cannot meet the performance requirements for the underwater bionic propulsion system. Therefore, this paper proposes a cosine feedforward compensation control algorithm based on the feedforward control idea to combine the conventional PID controller for an underwater propulsion system in variable load conditions. A composite feedforward controller is hence constructed to eliminate the amplitude and phase errors of sinusoidal follow-up. Experimental results show that the repetitive motion control algorithm based on cosine feedforward method is more effective than the conventional PID control method, and under variable load conditions it can still achieve accurate follow-up of a given sinusoidal motion command with about 60% reduction in the phase difference compared with the results without compensation.