非线性Thau观测下的水下机器人定速推进故障识别

Fault identification of constant speed propulsion of underwater vehicle under nonlinear Thau observation

水下机器人在水中运动时受水流、湍流和涡流等流体动力学的复杂影响,会对推进系统产生非线性影响,使得工作状态发生微小变化,难以准确估计其运行状态,增加了故障识别的难度。为此,提出一种非线性Thau观测下的水下机器人定速推进故障识别方法。利用无损卡尔曼滤波估计水下机器人定速推进器的状态量;通过非线性Thau观测算法,结合状态量估计结果建立非线性Thau观测器,识别定速推进故障;通过模糊神经网络逼近非线性Thau观测器内的有界非线性不确定性扰动,提升故障识别精度。实验结果证明:该方法可有效估计水下机器人定速推进器的状态量,并逼近非线性Thau观测器的有界非线性不确定性扰动;而且可有效识别定速推进故障,故障识别精度较高。

The complex influence of hydrodynamics such as water flow,turbulence and eddy current on the underwater robot in the water has a nonlinear effect on the propulsion system,which makes the working state change slightly,and it is difficult to accurately estimate the operating state and increase the difficulty of fault identification.Therefore,a fault identification method for constant speed propulsion of underwater vehicle under nonlinear Thau observation is proposed.The lossless Kalman filter is used to estimate the state quantity of constant speed propeller of underwater vehicle.Based on the nonlinear Thau observation algorithm and state estimation results,a nonlinear Thau observer is established to identify the faults of constant speed propulsion.The bounded nonlinear uncertainty disturbance in the nonlinear Thau observer is approximated by means of fuzzy neural network to improve the fault identification accuracy.The experimental results show that the proposed method can effectively estimate the state quantity of the constant speed propeller of underwater vehicle,and approximate the bounded nonlinear uncertainty disturbance of the nonlinear Thau observer.This method can effectively identify the fault of constant speed propulsion,and the fault identification accuracy is high.