Search

The working state of machinery can be reflected by vibration signals. Accurate classification of these vibration signals is helpful for the machinery fault diagnosis. A novel classification method for vibration signals, named Transform Domain Sparse Representation-based Classification (TDSRC), is proposed. The method achieves high classification accuracy by three steps. Firstly, time-domain vibration signals, including training samples and test samples, are transformed to another domain, e.g. frequency-domain, wavelet-domain etc. Then, the transform coefficients of the training samples are combined as a dictionary and the transform coefficients of the test samples are sparsely coded on the dictionary. Finally, the class label of the test samples is identified by their minimal reconstruction errors. Although the proposed method is very similar to the Sparse Representation-based Classification (SRC), experimental results illustrates its performance is far superior to SRC in the classification of vibration signals. These experiments include: frequency-domain classification of bearing vibration data from the Case Western Reserve University (CWRU) Bearing Data Center and wavelet-domain classification of six fault-types gearbox vibration data from our rotating machinery experimental platform.

In order to extract fault impulse feature of large-scale rotating machinery from strong background noise, a sparse feature extraction method based on sparse decomposition combined multiresolution generalized S transform is proposed in this paper. In this method, multiresolution generalized S transform is employed to find the optimal atom for every iteration, which firstly takes in to account the generalized S transform with discretized adjustment factors, then builds an atom corresponding to the maximum energy. The multiresolution generalized S transform has better accuracy compared to generalized S transform and faster searching speed compared to the orthogonal matching pursuit method in selecting the optimal atom. Then, the orthogonal matching pursuit method is used to decompose the signal into several optimal atoms. The proposed method is applied to analyze the simulated signal and vibration signals collected from experimental failure rolling bearings. The results prove that the proposed method has better performances such as high precision and fast decomposition speed than the traditional orthogonal matching pursuit method method and local mean decomposition method.

Empirical mode decomposition (EMD) is an adaptive nonlinear, non-stationary signal analysis and processing method. Engineering applications of EMD are primarily restricted by “mode mixing,” or information coupling between IMFs(Intrinsic mode function, IMF) such that the result of EMD decomposition cannot correctly reflect the real physical process. After summarizing existing analysis and processing methods, this paper proposes an algorithm of integrating the least square mutual information (with grid search for CEEMD decomposition), and the aliasing frequency to be amended (ensuring orthogonality between various IMF components to further suppress mode mixing, reduce the number of IMF components, and improve computational efficiency). The validity of the proposed method is verified through simulation experiments. The method is then used to extract the fault characteristic frequency of a rolling bearing micro-fault signal. Experimental results demonstrate that the proposed algorithm can obtain more accurate fault frequency characteristics with a lower total lumped number, fewer IMF components, and lower computational costs relative to traditional methods.

As the core component and the most vulnerable part of mechanical equipment, The rolling bearing is prone to failure during the operation of the process. In order to ensure the safe operation of the entire system, it is important to extract fault features of early and weak fault signal which has a very lower signal-noise ratio. Aiming at this problem, a hybrid method that based on Higher-Density Dual-Tree Discrete Wavelet Packet(HDDWP), Improved Hilbert Huang Transform(IHHT) is proposed. First, using HDDWP to eliminate the high frequency components of the original signal to improve signal-noise radio. Then the original signal is decomposed into a set of the intrinsic mode functions(IMFs) through the Complete Ensemble Empirical Mode Decomposition(CEEMD), some useful IMFs were collected by using Cross-Correlation function, and then calculated the Marginal Spectrum by Hilbert Transform. The experimental results indicate that the proposed method can effectively extract the early and weak fault features, and thereby to show that the proposed methods are feasible and effective.

Aiming at the end effect in Empirical Mode Decomposition (EMD) of vibration signals, Shape-Preserving Piecewise Cubic Spline (SPPCS) interpolation method and Homotopy Least Squares Support Vector Double-Regression (HLSSVDR) algorithm is proposed in this paper which can restrain the end effect. In order to obtain valid upper and lower envelopes, based on the studies of end effect mechanism and the existing research results, the SPPCS used to eliminate the fitting overshoot/undershoot problems by structuring the upper and lower envelope curves of the signal. Then the HLSSVDR algorithm is introduced to predict and replace the left and right values at both ends of the mean values of the upper and lower envelopes. The proposed method is analyzed and tested using the actual rolling bearing vibration signals. The experimental results indicate that the proposed method can effectively inhibit the fitting overshoot and undershoot, as well as the end effect, with higher accuracy and less distortion decomposition.

In the initial fault of rolling bearing, the useful weak impulses reflecting fault feature in measured vibration signal are usually corrupted by strong background noise. Sparse representation over learned dictionary is taken to extract the initial fault feature. Firstly, K-SVD learning algorithm is employed to obtain an adaptive dictionary matching the impulses. Then Batch Orthogonal Matching Pursuit (Batch-OMP) is utilized in sparse-coding stage, and kurtosis is introduced to determine the iteration stop condition in sparse approximation. The simulate data and real bearing data tests validate the proposed method.

The static and dynamic performance of speed loop is usually greatly influenced by the nonlinear friction in linear servo system. To solve the problem, an adaptive compensation scheme for the purpose of reducing friction effects on linear speed control is present. Firstly, the mathematical model is analyzed in detail according to components of linear servo system, and the LuGre friction model is simplified. Secondly, to improve the model identification accuracy and ensure the feasibility of identification algorithm simultaneously, a high-effective and stable just-in-time learning (JITL) algorithm is used to estimate the friction model parameter. Finally, based on the identification results, an adaptive feed-forward component is applied into linear servo system to compensate the friction. Experimental results show that linear servo system using the proposed method has better speed control performance than using a traditional compensation method.

Aimed at the problem of extraction for week impulse signal in rolling bearings, a method of week impulse fault signal extraction based on sparse decomposition is proposed. The over-complete dictionary is divided into a number of sub dictionaries, and calculate the coherent coefficient accumulation with the fault vibration signal to get the wave of coherent coefficient accumulation and displacement. According to this wave, the shift factor and frequency factor of the optimal atom can be confirm, then search further to confirm the scale factor and phase factor. Repeat the steps to gain a set of atoms which can represent the fault vibration signal sparsely. This method can extract the week impulses obviously and restrain the low-frequency component and noise effectively with the application of the over-complete dictionary, and can improve the efficiency by analyzing the relevance between the wave of coherent coefficient accumulation and the fault signals. The results of simulations show this method is effective and superior.

As one of the key issues of users' mobility management in next generation of heterogeneous wireless network, vertical handover has received extensive attention in academic and industrial circles. With the development of mobile terminals, mobile users can simultaneously run different types of applications, which have different requirements on network properties. Therefore, this article proposes an adaptive weight vertical handover algorithm, which can select the most suitable network as target handover network, according to the type of working application and the adaptive calculating weight vector to user's preference. The simulation results show that the presented algorithm can not only automatically select access points that suit mobile users best, but also reduce the unnecessary handovers.

As the interrelationship of different measuring points in rotating machinery fault monitoring. A blind source separation method combined with multi-doubling is presented to detect independent vibration signal among the measured signals. In this method, the separation matrix in blind source separation is replaced with relative gain array which is calculated through multi-doubling technique. As the multi-doubling is realized through hardware circuit, so the calculation of the relative gain array is simplied and the effiency of the algorithm is improved. The example shows that this method has better performance on separating signals and more accurate in fault location.

By making the best of the wireless channel's broadcasting property, opportunistic routing can greatly improve the performance of wireless multi-hop networks. Traditional routing security mechanism is no longer suitable for the design of routing protocol in Ad hoc network. So this thesis introduces trust, i.e. the concept of security and credibility into opportunistic routing, and put forward an opportunistic routing algorithm based on trust. According to the simulation result, compared with ExOR algorithm of traditional opportunistic routing, TOR algorithm can prevent attacks of malicious node, improve the robustness of network and has an advantage in aspects such as handling capacity and end-to-end time delay.

Chaos and fractal theory has its particular characteristic in posting essences of things and it's sensitive to complex machine system forepart faults. The fault diagnosis and prediction of complicated mechanical systems are researched by using the fractal and chaos theory. Judging by the kinetic characteristics of mechanical equipment have changed to judge whether the fault exists or not. The frequency spectrum is analyzed to determine the reason and the locality of the fault, further, in high-speed wire rolling mill of fault detection and diagnosis of achieved good effect.