In practical engineering applications, the multivariate signal contains more fault feature information than the single channel signal. How to realize synchronous extraction of fault features from the multivariate signal is of great significance in fault diagnosis of rotary machinery. Dynamic mode decomposition (DMD) has attracted much attention due to its excellent dynamic feature extraction ability. However, DMD lacks mode aliasing property when dealing with the multivariate signal, which may lead to the loss of critical fault feature information. Cater to this problem, this paper proposed a multivariate dynamic mode decomposition (MDMD) algorithm which is the multivariate extension of DMD. First, snapshot tensors are defined to convert multivariate signals to tensor format. Then, MDMD algorithm is proposed by introducing tensor operations into the original DMD algorithm, where a tensor low tubal rank component extraction framework is constructed to enable simultaneous extraction of bearing fault features from the multivariate signal, to enhance the robustness and effectiveness of the algorithm. Finally, both numerical simulations and experiments verify the proposed MDMD has higher fault diagnosis accuracy than other multivariate signal processing methods.
Monitoring and identifying the health condition of rolling bearings can reduce the risk of mechanical equipment failure. This paper proposes a novel intelligent diagnosis method of rolling bearings: First, the vibration signals are decomposed into band-limited instinct mode functions (BLIMFs) by variational mode decomposition (VMD). Then, the proposed high-dimensional common spatial pattern (hdCSP) filter is used to generate the high-dimensional eigenvectors representing the decomposed BLIMFs. Finally, the random forests classifier is used to classify the eigenvectors and obtain the diagnosis results. The performance of the proposed VMD-hdCSP method is evaluated on the Case Western Reserve University dataset. The experimental results show the proposed method can automatically classify different health states of rolling bearings and obtain precise diagnosis results.
27 February 2024 in IEEE Transactions on Instrumentation and Measurement
Variational mode decomposition (VMD) gained popularity due to its excellent performance in rolling bearing fault diagnosis. To obtain accurate diagnosis results depend on proper parameter selection, an improved VMD is proposed to achieve adaptive optimal parameter selection. This algorithm is based on a variable bandwidth control parameter strategy and a center frequency adaptive convergence strategy. First, a variable-bandwidth strategy is constructed according to the frequency distribution difference of each component. Next, the convergence property of the signal is analyzed by a self-convergent strategy based on the variable bandwidth control parameters. Then, the optimal initial center frequencies are discriminated to generate the optimal parameters. Finally, the optimal parameters for the improved VMD are used to obtain the decomposed modes. The validity of the proposed method is demonstrated by one simulation research and two application case analyses of faulty bearings. The performance comparisons indicate that the proposed method provides more accurate, robust, and efficient results.
Multivariate variational mode decomposition (MVMD) is a novel extension of variational mode decomposition (VMD) for multi-channel data sets. It decomposes multi-component and multi-channel signals into multivariate modulated oscillations crossing different center frequencies and limited bandwidths with sparse characteristics. MVMD inherits all the limitations of VMD and faces challenges in processing mechanical failure signals. The pre-selected values of the mode number K and balance parameters α still have the most significant impact on the decomposition results. Although the parameter-optimization method solves the problem of parameter selection to a certain extent, the result is often not optimal, and it is difficult to deal with multi-fault signals. A new multi-fault diagnosis method is proposed in this paper to solve these problems. Firstly, a new index, called the weighted combined fault index, is proposed to evaluate the fault information contained in each mode decomposed by MVMD, which is the criterion for selecting the optimal mode. Secondly, an iterative decomposition algorithm based on MVMD is proposed to iteratively decompose different fault components into the optimal modes to extract all potential fault information. Benefiting from these algorithms, this method applies MVMD to multi-fault diagnosis with adaptive parameter selection. Through simulations and experiments, the effectiveness and superiority of the proposed method are verified.
For autonomous underwater vehicles (AUVs), autonomous navigation in an unknown underwater environment is still a difficult problem. In recent years, people have proposed some machine learning-based methods to solve this problem, but the existing methods still cannot meet the complex and changeable underwater environment. This paper conducts technical research on the path planning of autonomous underwater vehicles, combines deep learning and reinforcement learning, uses WL interpolation surface to model the seabed, and proposes a path planning model for autonomous underwater vehicles based on deep reinforcement learning. And train the path planning model in the simulation environment, and finally achieve the goal of path planning for the underwater robot in the complex and changeable underwater environment.