Mingjun Zhang

Thruster fault feature extraction for autonomous underwater vehicle in time-varying ocean currents based on single-channel blind source separation

This article presents a novel thruster fault feature extraction method for autonomous underwater vehicle in time-varying ocean currents. The novel method is based on the combination of ensemble empirical mode decomposition and independent component analysis, which is a universal single-channel blind source separation method. When using ensemble empirical mode decomposition to decompose the autonomous underwater vehicle surge speed signal into a set of intrinsic mode functions, the original number of intrinsic mode functions is big, making the independent components that are separated and reconstructed from the intrinsic mode functions based on independent component analysis contain many false components. To reduce the original number of intrinsic mode functions, a method combining wavelet decomposition with empirical mode decomposition is proposed. When extracting fault feature value from independent components based on modified Bayes’ classification algorithm, the results show that the difference of fault feature value and noise feature value is small, and the ratio of fault feature value and noise feature value is small as well, which make it difficult to distinguish fault feature value from noise feature value. To increase the difference and the ratio, a wavelet detail component–assisted feature extraction method is proposed. The method combining wavelet decomposition with empirical mode decomposition and the wavelet detail component–assisted feature extraction method together constitute a novel thruster fault feature extraction approach which is suitable for autonomous underwater vehicle in time-varying ocean currents. The effectiveness of the proposed methods is verified by the pool experiments of the experimental prototype.

Fault-tolerant control based on adaptive sliding mode for underwater vehicle with thruster fault

An adaptive sliding mode backstepping fault-tolerant control is proposed for autonomous underwater vehicle(AUV) with thruster faults. Thruster faults are treated as uncertainties. Gaussian radial-basis-function networks are used to approximate these general uncertainties. In addtion, no Fault Detection and Diagnosis unit is needed in the proposed method.The controller can guarantee closed-stability regardless of thruster fault occurs or not. Due to the controlled system is strict-feedback in the traditional backstepping scheme, an improved backstepping scheme integrating with adaptive sliding mode algorithm. Finally, simulations are carried out to verify the validity of the proposed strategy.

Weak thruster fault prediction method for autonomous underwater vehicles based on grey model

When adopting the conventional grey model (GM(1,1)) to predict weak thruster fault for autonomous underwater vehicles, the prediction error is not always satisfactory. In order to solve the problem, this article develops a new weak thruster fault prediction method based on an improved GM(1,1). In the developed GM(1,1) based fault prediction method, this article mainly makes improvement in the following aspects: construction of grey background value, solution of whiting differential equation and construction of predicted sequence. Specifically, the integral operation is used in range of the two adjacent steps to obtain the grey background value at first. Second, in the solving of whiting differential equation, the point corresponding to the least difference between the accumulated generation sequence and its predicted sequence is determined, and then this special point’s value in the original sequence is considered as the initial condition of the whiting differential equation. Third, in the construction of predicted sequence, another predicted value is obtained based on the error sequence between the accumulated generating operation sequence and its predicted sequence, and then the new predicted result is used to re-adjust the accumulated generating operation sequence, so as to guarantee the re-adjustability of the fault prediction result. Finally, experiments are performed on Beaver 2 autonomous underwater vehicle to evaluate the prediction performance of the developed method.

Fault degree identification method for thruster of autonomous underwater vehicle using homomorphic membership function and low frequency trend prediction

This article presents a novel thruster fault degree identification method for autonomous underwater vehicle. The novel method is developed from the fuzzy support vector domain description method, which establishes a fault identification model first, and then estimates fault degree according to the model. When establishing fault identification model for thruster based on fuzzy support vector domain description method, it is found that the relative fitting error of the model to the actual fault degree is large, making the model accuracy poor. To reduce the relative fitting error, a homomorphic membership function method is proposed. Different from fuzzy support vector domain description method, which calculates the fuzzy membership degree of fault sample in time domain, the proposed method calculates the fuzzy membership degree in log domain. On estimating thruster fault degree by fuzzy support vector domain description method, it is obtained that the estimated fault degree lags behind the actual fault degree. To shorten the lag time, a low frequency trend prediction method is proposed. Different from fuzzy support vector domain description method, which brings the fault feature extracted from the current surge speed and control voltage into the fault identification model to calculate fault degree, the proposed method firstly forward predicts surge speed and control voltage, and then takes the fault feature extracted from the predicted surge speed and control voltage into the model to acquire fault degree. The effectiveness of the proposed methods is verified by pool experiments of the experimental prototype.

Thruster fault identification based on fractal feature and multiresolution wavelet decomposition for autonomous underwater vehicle

There exist some problems when the fractal feature method is applied to identify thruster faults for autonomous underwater vehicles (AUVs). Sometimes it could not identify the thruster fault, or the identification error is large, even the identification results are not consistent for the repeated experiments. The paper analyzes the reasons resulting in these above problems according to the experiments on AUV prototype with thruster faults. On the basis of these analyses, in order to overcome the above deficiency, an improved fractal feature integrated with wavelet decomposition identification method is proposed for AUV with thruster fault. Different from the fractal feature method where the signal extraction and fault identification are completed in the time domain, the paper makes use of the time-domain and frequent-domain information to identify thruster faults. In the paper, the thruster fault could be mapped multisource and described redundantly by the fault feature matrix constructed based on the time-domain and frequent-domain information. In the process of identification, different from the fractal feature method where the fault is identified based on fault identification model, the fault sample bank is built at first in the paper, and then pattern recognition is achieved by calculating the relative coefficients between the constructed fault feature matrix and the elements in the fault sample bank. Finally, the online pool experiments are performed on an AUV prototype, and the effectiveness of the proposed method is demonstrated in comparison with the fractal feature method.

Weak thruster fault prediction method for autonomous underwater vehicle

Conventional grey GM(1,1) method might generate much large prediction errors when predicting the trend of weak thruster fault. To solve this problem, a prediction method based on improved grey prediction model is proposed. The proposed method selects the known sequence point with the minimum prediction error as the initial condition of grey derivative model, rather than first element of the known sequence in the conventional method, so as to improve the robustness of the method to external disturbance and obtain better prediction precision. Compared with the sequence generation process of the conventional grey GM(1,1) method, this paper constructs the grey background value of exponent curve to reduce the error of grey differential equation solution, so as to reduce the prediction errors. In the proposed method, the residual series between the forecasting sequence and original one is constructed to re-predict, and the result is used to modify the predictive value of thruster fault so as to improve prediction precision. Poolexperiments are performed on a prototype AUV by simulating thruster weak fault. The comparative experiment results based on the proposed method and the conventional grey GM(1,1) method demonstrate the effectiveness of the proposed method.