单神经元自适应PID在水下拖曳控制系统中的应用

The Application Simulation of Single Neuron Self-Adaptive PID Controller in Towed Systems

基于多参数拖曳式剖面测量系统的物理参数,从运动物体六自由度方程出发建立了深度控制模型。利用单神经元自适应PID,对拖体深度控制进行了仿真,并与常规PID控制方法进行了对比。结果表明:在3种不同的拖曳速度下,两种方法的响应时间随输入深度的变化曲线都存在一个阈值深度。大于阈值深度时,单神经元自适应PID比常规PID响应快得多,即对于大深度输入,前者的响应时间能够满足控制要求,而后者明显过长。说明单神经元自适应PID控制方法对于大深度输入响应快,不同拖曳速度下深度控制自适应性好。

Based on th trol model is construc e physical characters of Towed Multi-parameter Profile sampling System, depth conted from a six-degree freedom equation of the towed vehicle. The simulation of depth of the vehicle is done by using Single Neuron Self-Adaptive PID Controller, and was compared with the result by routine PID. The conclusion is that the curves of response time of the two methods with the change of depth exist a threshold depth under different towing velocities. When the depth inputs surpass the threshold depth, it was shown that the response of Single Neuron Self-Adaptive PID is much faster than that of routine PID. The response time of the former method can reach the control requirement for the large depth inputs,however, that of the latter method is too long. The results show that Single Neuron Self-Adaptive PID responds rapidly for large depth inputs, and self -adaptive ability of depth control is more efficient under different towed velocities.