Chuan Li

Fault Diagnosis for Rotating Machinery Using Vibration Measurement Deep Statistical Feature Learning

Fault diagnosis is important for the maintenance of rotating machinery. The detection of faults and fault patterns is a challenging part of machinery fault diagnosis. To tackle this problem, a model for deep statistical feature learning from vibration measurements of rotating machinery is presented in this paper. Vibration sensor signals collected from rotating mechanical systems are represented in the time, frequency, and time-frequency domains, each of which is then used to produce a statistical feature set. For learning statistical features, real-value Gaussian-Bernoulli restricted Boltzmann machines (GRBMs) are stacked to develop a Gaussian-Bernoulli deep Boltzmann machine (GDBM). The suggested approach is applied as a deep statistical feature learning tool for both gearbox and bearing systems. The fault classification performances in experiments using this approach are 95.17% for the gearbox, and 91.75% for the bearing system. The proposed approach is compared to such standard methods as a support vector machine, GRBM and a combination model. In experiments, the best fault classification rate was detected using the proposed model. The results show that deep learning with statistical feature extraction has an essential improvement potential for diagnosing rotating machinery faults.

Gearbox Fault Identification and Classification with Convolutional Neural Networks

Vibration signals of gearbox are sensitive to the existence of the fault. Based on vibration signals, this paper presents an implementation of deep learning algorithm convolutional neural network (CNN) used for fault identification and classification in gearboxes. Different combinations of condition patterns based on some basic fault conditions are considered. 20 test cases with different combinations of condition patterns are used, where each test case includes 12 combinations of different basic condition patterns. Vibration signals are preprocessed using statistical measures from the time domain signal such as standard deviation, skewness, and kurtosis. In the frequency domain, the spectrum obtained with FFT is divided into multiple bands, and the root mean square (RMS) value is calculated for each one so the energy maintains its shape at the spectrum peaks. The achieved accuracy indicates that the proposed approach is highly reliable and applicable in fault diagnosis of industrial reciprocating machinery. Comparing with peer algorithms, the present method exhibits the best performance in the gearbox fault diagnosis.

A statistical comparison of neuroclassifiers and feature selection methods for gearbox fault diagnosis under realistic conditions

Gearboxes are crucial devices in rotating power transmission systems with applications in a variety of industries. Gearbox faults can cause catastrophic physical consequences, long equipment downtimes, and severe production costs. Several artificial neural networks, learning algorithms, and feature selection methods have been used in the diagnosis of the gearbox healthy state. Given a specific gearbox, this study investigates how these approaches compare with each other in terms of the typical fault classification accuracy but also in terms of the area under curve (AUC), where the curve refers to the precision-recall curve otherwise known as receiver operating characteristic (ROC) curve. In particular, the comparison aims at identifying whether there are statistically significant (dis)similarities among six feature selection methods, and seven pairs of neural nets with different learning rules. Genetic algorithm based, entropy based, linear discriminants, principal components, most neighbors first, and non-negative matrix factorization are the studied feature selection methods. Feed forward perceptrons, cascade forward, probabilistic nets, and radial basis function neural nets are evaluated. Six supervised and one unsupervised learning rules are considered. Both parametric and nonparametric statistical tests are employed. A ranking process is defined to elect the best approach, when available. An experimental setup was especially prepared to ensure operating conditions as realistic as possible.

A multi-pattern deep fusion model for short-term bus passenger flow forecasting

Abstract Short-term passenger flow forecasting is one of the crucial components in transportation systems with data support for transportation planning and management. For forecasting bus passenger flow, this paper proposes a multi-pattern deep fusion (MPDF) approach that is constructed by fusing deep belief networks (DBNs) corresponding to multiple patterns. The dataset of the short-term bus passenger flow is first segmented into different clusters by an affinity propagation algorithm. The passenger flow distribution of these clusters is subsequently analyzed for identifying different patterns. In each pattern, a DBN is developed as a deep representation for the passenger flow. The outputs of the DBNs are finally fused by chronological order rearrangement. Taking a bus line in Guangzhou city of China as an example, the present MPDF approach is modeled. Five approaches, non-parametric and parametric models, are applied to the same case for comparison. The results show that, the proposed model overwhelms all the peer methods in terms of mean absolute percentage error, root-mean-square error, and determination coefficient criteria. In addition, there exists significant difference between the addressed model and the comparison models. It is recommended from the present study that the deep learning technique incorporating the pattern analysis is promising in forecasting the short-term passenger flow.

Fuzzy determination of informative frequency band for bearing fault detection

Detecting early faults in rolling element bearings is a crucial measure for the health maintenance of rotating machinery. As faulty features of bearings are usually demodulated into a high-frequency band, determining the informative frequency band (IFB) from the vibratory signal is a challenging task for weak fault detection. Existing approaches for IFB determination often divide the frequency spectrum of the signal into even partitions, one of which is regarded as the IFB by an individual selector. This work proposes a fuzzy technique to select the IFB with improvements in two aspects. On the one hand, an IFB-specific fuzzy clustering method is developed to segment the frequency spectrum into meaningful sub-bands. Considering the shortcomings of the individual selectors, on the other hand, three commonly-used selectors are combined using a fuzzy comprehensive evaluation method to guide the clustering. Among all the meaningful sub-bands, the one with the minimum comprehensive cost is determined as the IFB. The bearing faults, if any, can be detected from the demodulated envelope spectrum of the IFB. The proposed fuzzy technique was evaluated using both simulated and experimental data, and then compared with the state-of-the-art peer method. The results indicate that the proposed fuzzy technique is capable of generating a better IFB, and is suitable for detecting bearing faults.

Deep Feature Learning Architectures for Daily Reservoir Inflow Forecasting

Inflow forecasting applies data supports for the operations and managements of reservoirs. To better accommodate the sophisticated characteristics of the daily reservoir inflow, two deep feature learning architectures, i.e., deep restricted Boltzmann machine (DRBM) and stack Autoencoder (SAE), respectively, are introduced in this paper. This study sheds light on the application of deep learning architectures for daily reservoir inflow forecasting, which has been attracting much attention in various areas for its ability to extract and learn useful features from a large number of data. Evaluations are made comparing the basic feed forward neural network (FFNN), the autoregressive integrated moving average (ARIMA), and two categories deep neural networks (DNNs) constructed by the integrations the FFNN with two deep feature learning architectures, named DRBM-based NN and stack SAE-based NN, respectively. Two daily inflow series of the Three Gorges reservoir (1/1/2000–31/12/2014) and the Gezhouba reservoir (1/1/1992–31/12/2014), China, are applied for four modeling exercises, respectively. The results show that, the two DNN models overwhelm the FFNN and the ARIMA models in terms of mean absolute percentage error, normalized root-mean-square error, and threshold statistic criteria.

On-Line Monitoring of Air Cooler in Transformer by Using Fiber Bragg Grating Temperature Sensors

The air cooler reduce the working temperature of the transformer, high temperature will reduce the transformers life, and even make insulation overheating, aging, then the transformer will be damaged. The air cooler which is forced oil circulation in transformer outdoor in substation has been monitored. The real-time monitoring of 24-hours indicates that the temperature changes in the range of 3°C. The ambient temperature is lower than the temperature of fan about 1°C. At the long-term of 479 days, the average daily temperature range of fan is 35.32°C, the maximum temperature is 50.25°C on August 18,2011, and the maximum temperature is 14.93°C on January 17, 2011. The daily average of ambient temperature range is 37.59°C, the maximum temperature is 51.16°C on August 17, 2011,and the minimum temperature is 13.57°C on January 17, 2011. The maximum difference between the temperature of fan and the ambient temperature is 12.48°C on September 15, 2011. According to the relevant standards and monitoring results, the maximum threshold of fan temperature is defined to 80°C, and the threshold of temperature rise is 20°C.

The Real-Time Strain Monitoring of Tunnel Supporting Structure and Nonlinear Regression Analysis

Concrete embedded fiber Bragg grating strain sensors are used to monitor real-time strain of each section in the tunnel surrounding rock supporting structure, and the nonlinear regression analysis method is adopted to analysis the real-time measurement data. By using nonlinear regression analysis method, the strain development status of surrounding rock supporting structure can be grasped timely and the variation trend of the strain value of monitoring point can be predicted, which can provides foundations for judging the stability of tunnel supporting structure.

Control network for monitoring deformation of slope by using optical measurements and FBG sensors

The ground deformation observation is the method that estimates the ground deformation according to the displacement change of work spot measured periodically. By using the repeatability of the mechanical transducer and the reliability of the wavelength modulation, the differential fiber Bragg grating settlement gauge reduces availably the disturbance of the artificial causation and the weather. In this scheme, the drive rod converts the displacement of settlement block into the strain of beam, and causes the Bragg wavelength shifts of the sensing gratings mounted on the top and bottom surfaces of the beam, on which the temperature-compensated is achieved by the differential operation of the shifts of Bragg wavelengths. Because the collapse of slope is caused by the cutting slippage as the shearing strength in the soil body of slope exceeded the anti-shearing intensity, the slide prevention pier is the regular retaining structure for treating slope. The differential fiber Bragg grating settlement gauge is developed and fixed on the pier, which converts the settlement of the observation point into the Bragg wavelength shifts of the sensing gratings. Applied in Bai Ni-jing Slope, the measurement precision between the settlement gauges and Wild N3 precise level is 0.03, which is satisfied with the engineering order measure precision 0.05~0.1. By subtract the displacement-induced settlement deviation, the corrected precision between the settlement gauges compensated by Leica J2 theodolite and the precise level is improved to 0.007, which is satisfied with the scientific research order measure precision 0.01~0.05.

Intelligent Fault Diagnosis of Delta 3D Printers Using Attitude Sensors Based on Support Vector Machines

Health condition is a vital factor affecting printing quality for a 3D printer. In this work, an attitude monitoring approach is proposed to diagnose the fault of the delta 3D printer using support vector machines (SVM). An attitude sensor was mounted on the moving platform of the printer to monitor its 3-axial attitude angle, angular velocity, vibratory acceleration and magnetic field intensity. The attitude data of the working printer were collected under different conditions involving 12 fault types and a normal condition. The collected data were analyzed for diagnosing the health condition. To this end, the combination of binary classification, one-against-one with least-square SVM, was adopted for fault diagnosis modelling by using all channels of attitude monitoring data in the experiment. For comparison, each one channel of the attitude monitoring data was employed for model training and testing. On the other hand, a back propagation neural network (BPNN) was also applied to diagnose fault using the same data. The best fault diagnosis accuracy (94.44%) was obtained when all channels of the attitude monitoring data were used with SVM modelling. The results indicate that the attitude monitoring with SVM is an effective method for the fault diagnosis of delta 3D printers.