Performance Degradation Assessment Research Articles

Performance degradation assessment of rolling bearing based on convolutional neural network and deep long-short term memory network

Many traditional approaches for performance degradation assessment of rolling bearings, using sensor data, make assumptions about how they degrade or fault evolve. However, the sequential sensor data cannot be directly taken as input in the traditional models since the data always contain noise and change in length. To solve these problems, a convolutional neural network and deep long-short term memory (CNN-DLSTM) based architecture is proposed to obtain an unsupervised H-statistic for performance degradation assessment of rolling bearing using sensor time-series data. Firstly, a CNN is applied to extract local abstract features from raw sensor data. Secondly, a deep LSTM is explored to extract temporal features. CNN-DLSTM is trained to reconstruct the time-series sensor signal reflecting the health condition of rolling bearing. The D- and Q-statistic are used to compute H-statistic which is then used for performance degradation assessment. The proposed approach is evaluated on an experiment with rolling bearings and the results are presented on a public dataset of rolling bearing, verifying that the proposed approach outperforms several state-of-the-art methods.

Performance degradation assessment of slewing bearings based on deep auto-encoder and optimized particle filtering

Due to the low speed and heavy load conditions of slewing bearings, extracting of effective features for fault diagnosis and prediction is difficult but crucial. Moreover, challenges such as large data volumes, unlabeled and multi-source bring more difficulties for advanced prognosis and health management methods. To solve these problems, a novel method for performance degradation assessment of bearings based on raw signals is proposed. In this methodology, a combination of deep auto-encoder (DAE) algorithm and particle filter algorithm is utilized for feature extraction and remaining useful life (RUL) prediction. First, the raw vibration signal is employed to train parameters of a restricted Boltzmann machine to build the DAE model. Through encoding and decoding multi-source data, root mean square error of reconstruction error between the raw signal and reconstructed signal is employed to detect incipient faults of slewing bearings. Then, degradation trend model is established by particle filtering to predict RUL of bearings. The effectiveness of proposed method is validated using simulated and experimental vibration signals. Results illustrate that proposed method can evaluate the performance degradation process and RUL of slewing bearings.

Degradation assessment of rolling bearing towards safety based on random matrix single ring machine learning

Aiming at problems of feature extraction of monitoring data may lose useful information and different feature extraction methods have great influence on the results in rolling bearing performance degradation assessment. A method which is based on random matrix theory to assess the degradation of rolling bearing is proposed to improve production safety. Firstly, the observation matrix was constructed by the raw data of rolling bearing health monitoring, and the random matrix model of rolling bearing operating states was established by the steps of normalization, singular matrix singular value decomposition and standardization. Then, three performance degradation indicators, the average spectral radius, the maximum eigenvalue, and the number of random points in the inner ring are proposed by constructing the linear eigenvalue statistic of the random matrix, combined with the single ring curve of the bearing operating states, to describe the fluctuation discipline of the random matrix elements and the variation law of the characteristic root distribution, thus characterize the statistical random characteristics of the operating states of the rolling bearing as a whole. Finally, the method was applied by the rolling bearing life test vibration data. And four different operation states of the bearing are selected to build the random matrix model and analysis. The analysis results are calculated by single ring curves and quantitative indicators of different states and proved that the method proposed in this paper can effectively identify different degradation states and realize safety states of rolling bearing, meanwhile, it can quantitatively assess the degree of performance degradation.

Assessing the Performance Degradation of Lithium-Ion Batteries Using an Approach Based on Fusion of Multiple Feature Parameters

A method based on fusion of multiple features is proposed to assess and accurately describe the performance degradation of lithium-ion batteries in this paper. First, the discharge voltage signal of lithium-ion batteries under real-time monitoring is analyzed from the perspective of time domain and complexity to obtain the values of multiple features. Then, the multi-feature parameters undergo a spectral regression process to reduce the number of dimensions and to eliminate redundancy, and on the basis of this regression, a Gaussian mixture model is established to model the health state of batteries. Thus, the degree of lithium-ion battery performance degradation can be quantitatively assessed using the Bayesian inference-based distance metric. A case calculation experiment is carried out to verify the effectiveness of the method proposed in this paper. The experimental results demonstrate that, compared with other assessment methods, the performance degradation assessment method proposed in this paper can be used to monitor the degradation process of lithium-ion batteries more effectively and to improve the accuracy of condition monitoring of batteries, thereby providing powerful support for making maintenance decisions.

A Data-Driven, Statistical Feature-Based, Neural Network Method for Rotary Seal Prognostics

Failure of the rotary seal is one of the foremost causes of breakdown in rotary machinery, and such a failure can be catastrophic, resulting in costly downtime and large expenses. Assessing the performance degradation of the rotary seal is very important for maintenance decision-making. Although significant progress has been made over the last 5 years to understand the degradation of seals using experimental verification and numerical simulation, there is a research gap on the data-driven-based tools and methods to assess the health condition of rotary seals. In this paper, we propose a data-driven-based performance degradation assessment approach to classify the running/health condition of rotary seals, which was not considered in the previous studies. Statistical time domain features are extracted from friction torque-based degradation signals collected from a rotary setup. Wrapper-based feature selection approach is used to select relevant features, with multilayer perceptron neural network utilized as a classification technique. To validate the proposed methodology, an accelerated aging and testing procedure is developed to capture the performance of rotary seals. The study findings indicate that multilayer perceptron (MLP) classifier built using features related to the amplitude of torque signal has a better classification accuracy for unseen data when compared with logistic regression and random forest classifiers.

Performance degradation assessment of a wind turbine gearbox based on multi-sensor data fusion

Gearbox is a critical transmission component in the drivetrain of wind turbine having a dominant failure rate and a highest downtime loss among all wind turbine subsystems. Ensemble empirical mode decomposition and principal component analysis have been extensively investigated for signal decomposition and fault feature extraction from the signals of a wind turbine gearbox. However, presence of background noise in wind turbine signals restricts the applicability of these algorithms in real scenarios. To solve this problem, a novel performance degradation assessment method based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and kernel principal component analysis (KPCA) was proposed to de-noise and fuse vibration signals. A comparison is conducted between CEEMDAN-KPCA and other five cross combinations. After feature extraction, extreme learning machine (ELM) optimized by fruit fly of algorithm (FOA) is employed to predict the remaining use life (RUL) of wind turbine gearbox. To avoid the tedious hand-crafting hidden nodes, a dynamic adjustment strategy is presented for the improvement of search efficiency. The effectiveness of the proposed method is validated using simulated and experimental vibration signals. The results illustrate that the CEEMDAN-KPCA gets better result in signal de-noising. Compared with different swarm intelligence algorithms, the RUL prediction rates of wind turbine gearbox are improved by using the FOA-ELM prediction model with multiple parameters.

Roller Bearing Performance Degradation Assessment Based on Fusion of Multiple Features of Electrostatic Sensors.

This paper presents a new method to assess the performance degradation of roller bearings based on the fusion of multiple features, with the aim of improving the early degradation detection ability of the electrostatic monitoring system. At first, a set of feature parameters of the electrostatic monitoring system indicating the normal state of the bearings are extracted from the perspective of the time domain, frequency domain and complexity. Then, the parameter set is processed to reduce the dimensions and eliminate the redundancy using spectral regression. With the processed features, a Gaussian mixed model is established to gauge the health of the bearing, providing the distance value obtained using Bayesian inference as a quantitative indicator for assessing the performance degradation. The method is applied to access the life of a bearing in which the mechanic fatigue is artificially accelerated. The test results show that the proposed method can better reflect the degradation process of the bearing compared to other evaluation methods. This enables the electrostatic monitoring technique to detect the degradation of the bearing earlier than the vibration monitoring, providing a powerful tool for the condition monitoring of roller bearings.

Subspace-based MVE for performance degradation assessment of aero-engine bearings with multimodal features

Abstract Performance degradation assessment is an important concept in prognostics and health management (PHM) of complex engineering systems. In this study, a novel method utilizing subspace-based minimum volume ellipsoid (SMVE) for bearing performance degradation assessment is proposed. The statistical features extracted from vibration signals are seen as multimodal homologous features in time, frequency and wavelet modalities, thus it needs to consider the differences in variance for the homologous features as well as covariance among them. The subspaces are used to model the homologous features of different frequency bands since they can capture the dynamic information. Based on subspaces, the proposed SMVE model covering most or all of the normal data by a unique ellipsoid with minimum volume considers variance of each dimension adaptively. A performance degradation index is designed based on the SMVE. The proposed method is applied on the vibration signals acquired in run-to-failure tests of aero-engine bearings and common bearings. By comparison with the commonly used time-domain features and traditional minimum volume ellipsoid method, the results demonstrate the effectiveness and superiority of the proposed method.

Smartphone’s Sensing Capabilities for On‐Board Railway Track Monitoring: Structural Performance and Geometrical Degradation Assessment

Railway infrastructure managers run dedicated inspection vehicles to monitor the geometric quality of the track (among other aspects) to detect irregularities and ensure safe running conditions of railway lines, in accordance with specific regulations. Unfortunately, these inspections disturb the normal traffic operation, especially in networks with intense traffic; are generally carried out only a few times per year; and, consequently, do not provide prompt identification of critical situations. Considering the recent developments and cost reduction in sensing capabilities of smartphones, the authors present an approach to use these technologies to perform constant acceleration measurements inside in‐service trains to complement the assessment of the structural performance and geometrical degradation of the tracks. Cross‐correlation values above 0.85 were obtained between the standard deviations of the longitudinal level and the experimental vertical accelerations measured on‐board a passenger train on an 11‐km railway stretch. The results showed that the approach can be used to identify critical situations that affect the performance of the track, regarding passenger comfort, degradation rates, and risk of derailment. It may comprise a low‐cost and crowdsourced complement to the general current practice of track geometric inspection by dedicated vehicles and contribute to an earlier detection of track malfunctions, consequently, to a more efficient maintenance planning and infrastructure management.

A Bearing Performance Degradation Modeling Method Based on EMD‐SVD and Fuzzy Neural Network

Bearing performance degradation assessment has great significance to condition‐based maintenance (CBM). A novel degradation modeling method based on EMD‐SVD and fuzzy neural network (FNN) was proposed to identify and evaluate the degradation process of bearings in the whole life cycle accurately. Firstly, the vibration signals of bearings in known states were decomposed by empirical mode decomposition (EMD) to obtain the intrinsic mode functions (IMFs) containing feature information. Then, the selected key IMFs which contain the main features were decomposed by singular value decomposition (SVD). And the decomposed results were used as the training samples of FNN. At last, the output results of the tested data were normalized to the health index (HI) through learning and training of FNN, and then the performance degradation degree could be described by the distance between the test sample and the normal one. According to the case study, this modeling method could evaluate the performance degradation of bearings effectively and identify the early fault features accurately. This method also provided an important maintenance strategy for the CBM of bearings.

Rolling Bearing Performance Degradation Assessment Based on Convolutional Sparse Combination Learning

A novel bearing performance degradation assessment method based on convolutional sparse coding and combination learning is proposed in this paper, which can avoid the impact of the traditional features on the assessment results under complex operation conditions. The vibration signal can be decomposed into the convolution of the kernel sets and their corresponding sparse solutions using convolutional sparse coding. The learned kernel sets based on the training signal samples are the comprehensive embodiment of the information related to the operational complexity and the bearing healthy state, and the corresponding sparse solutions indicate the energy of the kernel activation. Asimulation experiment of bearing signal proves that the activation energy of the kernels, which are more related to bearing healthy, will rise with the increase of the bearing degradation degree; meanwhile, the error of reconstructed signal based on the learned kernel sets and original signal will decrease since the description effect of convolutional sparse coding to the signal will be enhanced with the periodic strengthen of signal caused by the fault. Thus, an index based on the kernel sparse norms and the errors between the original and reconstructed signals has been proposed to evaluate the degradation degree of the bearing in this paper. On the other hand, combination learning will be fused into the convolutional sparse coding to improve the real-time performance of the index calculation and obtain the best assessment result in the testing part by dividing the kernel dictionary set into multiple sub-dictionaries. A bearing run-to-fail experiment is analyzed to verify the validity of the proposed method. The testing results show that the proposed assessment index can clearly detect the initial fault, serve fault and failure of the bearing in the whole life, and the proposed method is more self-adaptive and sensitive to the fault degree. In addition, it is verified that the time consumption of the combination dictionaries is less than that of a single dictionary set.

Rolling Bearing Initial Fault Detection Using Long Short-Term Memory Recurrent Network

The complete failure of the rolling bearing is a deterioration process from the initial minor fault to the serious fault, it is meaningless for guiding maintenance when the serious fault is alarmed. This work presents a novel initial fault diagnosis framework based on sliding window stacked denoising auto-encoder (SDAE) and long short-term memory (LSTM) model. In this approach, multiple vibration value of the rolling bearings are entered into SDAE by sliding window processing. Then, multiple vibration value of the rolling bearings of the next period is predicted from the signal reconstructed by the trained SDAE in the previous period using LSTM. For the given input data, the reconstruction errors between the next period data and the output data generated by trained LSTM are used to detect initial anomalous conditions. The proposed method not only utilizes the ability of SDAE to learn the inherent distribution of data, but also ensures that LSTM can extract timing relationships between data cycles, and the model is built using only normal data. The initial fault detection as a key difficulty in the operating condition monitoring and performance degradation assessment of the rolling bearing is effectively solved. Experimental and classic rotating machinery datasets have been employed to testify the effectiveness of the proposed method and its preponderance over some state-of-the-art methods. The experiment results indicate that the proposed method can effectively detect the initial anomalies of the rolling bearing and accurately describe the deterioration trend with strong robustness, and have high significance for maintenance guiding.

Bearing Performance Degradation Assessment Based on Ensemble Empirical Mode Decomposition and Affinity Propagation Clustering

As key components in a rotating machinery system, bearings affect the safety of the entire mechanical system. Hence, early-stage monitor of bearing degradation is critical to avoid abrupt mechanical system failure. In this paper, a novel bearing performance assessment model is constructed based on ensemble empirical mode decomposition (EEMD) and affinity propagation (AP) clustering. Unlike most clustering methods, AP clustering, which automatically finds the center of all available clusters, can determine the bearing degradation status without an experience-based selection of the number of degradation states. The original bearing vibration signal is first decomposed by EEMD and its degradation fault features are extracted from the singular-value decomposition of intrinsic mode functions. Then, the degradation features are selected as the input of AP clustering to find the cluster centers of different bearing health statuses: “normal”, “slight”, and “severe”. Last, a health evaluation indicator, referred to as the confidence value, which is obtained from the dissimilarity between actual samples and the various cluster centers, is used to evaluate the bearing health status. To prove the superiority of the approach, the proposed model is compared to various popular clustering methods, including, k-means, k-medoids, fuzzy c-means, Gustafson-Kessel, and Gath-Geva, and commonly used time-domain indicators such as root mean square and kurtosis. The experimental results show that the proposed method outperforms the above time-domain indicators and clustering methods in monitoring early-stage degradation, without presetting the number of clusters.

Bearing performance degradation assessment using long short-term memory recurrent network

Bearing is commonly used in rotating machinery, and it is significant to monitor bearing running states to ensure machine safety. Performance degradation assessment is an important work in bearing condition-based maintenance (CBM) and predictive maintenance. In this paper, a data-driven bearing performance degradation assessment method based on long short-term memory (LSTM) recurrent neural network (RNN) is proposed to comprehensively utilize the fault propagation information. Firstly, universal degradation simulation model based on vibration response mechanism is constructed for feature verification. A new proposed indicator "waveform entropy (WFE)" is developed and validated by this simulation model. Then waveform entropy and other conventional indicators are input into the LSTM RNN to identify the bearing running state, while particle swarm optimization method is applied to optimize the network structure parameters simultaneously. Experimental results demonstrate that LSTM RNN can effectively identify the bearing degradation states and accurately predict the remaining useful life.

Degradation assessment of ball bearings utilizing curvilinear component analysis

The performance degradation assessment of ball bearings is of great importance to increase the efficiency and the reliability of rotating mechanical systems. The large dimensionality of feature space introduces a lot of noise and buries the potential information about faults hidden in the feature data. This paper proposes a novel health assessment method facilitated with two compatible methods, namely curvilinear component analysis and self-organizing map network. The novelty lies in the implementation of a vector quantization approach for the sub-manifolds in the feature space and to extract the fault signatures through nonlinear mapping technique. Curvilinear component analysis is a nonlinear mapping tool that can effectively represent the average manifold of the highly folded information and further preserves the local topology of the data. To answer the complications and to accomplish reliability and accuracy in bearing performance degradation assessment, the work is carried out with following steps; first, ensemble empirical mode decomposition is used to decompose the vibration signals into useful intrinsic mode functions; second, two fault features i.e. singular values and energy entropies are extracted from the envelopes of the intrinsic mode function signals; third, the extracted feature vectors under healthy conditions, further reduced with curvilinear component analysis are used to train the self-organizing map model; finally, the reduced test feature vectors are supplied to the trained self-organizing map and the confidence value is obtained. The effectiveness of the proposed technique is validated on three run-to-failure test signals with the different type of defects. The results indicate that the proposed technique detects the weak degradation earlier than the widely used indicators such as root mean square, kurtosis, self-organizing map-based minimum quantization error, and minimum quantization error-based on the principal component analysis.

An Electro-Mechanical Actuator Motor Voltage Estimation Method with a Feature-Aided Kalman Filter.

Electro-Mechanical Actuators (EMA) have attracted growing attention with their increasing incorporation in More Electric Aircraft. The performance degradation assessment of EMA needs to be studied, in which EMA motor voltage is an essential parameter, to ensure its reliability and safety of EMA. However, deviation exists between motor voltage monitoring data and real motor voltage due to electromagnetic interference. To reduce the deviation, EMA motor voltage estimation generally requires an accurate voltage state equation which is difficult to obtain due to the complexity of EMA. To address this problem, a Feature-aided Kalman Filter (FAKF) method is proposed, in which the state equation is substituted by a physical model of current and voltage. Consequently, voltage state data can be obtained through current monitoring data and a current–voltage model. Furthermore, voltage estimation can be implemented by utilizing voltage state data and voltage monitoring data. To validate the effectiveness of the FAKF-based estimation method, experiments have been conducted based on the published data set from NASA’s Flyable Electro-Mechanical Actuator (FLEA) test stand. The experiment results demonstrate that the proposed method has good performance in EMA motor voltage estimation.

Fault diagnosis of rotating machinery using Gaussian process and EEMD‐treelet

SummaryFault detection of rotating machinery is very important for its performance degradation assessment. In this work, an effective feature learning and detecting method based on the ensemble empirical mode decomposition (EEMD) and Gaussian process classifier (GPC) is put forward. Compared with the traditional parameter optimization methods of GPC, this work proposed a bacterial foraging optimization as the optimal solution of the hyperparameters of GP model. To find a valid feature vector, this work also utilized EEMD to decompose the vibration signals and get some time‐frequency features. Then, treelet transform is proposed to reduce the feature dimension. The results of some applications indicate that the EEMD has stronger processing capability of the status signals of rotating machinery. Treelet can transform the high‐dimensional vector to low‐dimensional space, which is used as the input of the proposed BFO‐GP model. The proposed diagnosis method can identify not only the optimal feature vector but also the fault locations.

HVSRMS localization formula and localization law: Localization diagnosis of a ball bearing outer ring fault

Abstract The angle position of the outer ring fault has a significant influence on the residual life and the operation performance of ball bearing. Therefore, the realization of the localization diagnosis of the outer ring fault is of great importance to the bearing performance degradation assessment and the life prediction. In this paper, a novel localization diagnosis method, named horizontal–vertical synchronized root mean square (HVSRMS) localization law and localization formula, is developed to diagnose the angle position of outer ring fault accurately. Firstly, based on the established nonlinear contact model of the bearing system, the detailed change law of static contact force when the angle position of the outer ring fault is on both sides of 270° is studied, which lays a theoretical foundation for the discovery of localization law. Subsequently, the vibration acceleration signal and dynamic contact force of the outer ring fault are synchronously analyzed through theoretical research, and the new feature for localization diagnosis is presented as: the prime impact direction of horizontal vibration acceleration (IPDoHVA). Therefore, the outer ring fault localization law of ball bearing is proposed. On this basis, the HVSRMS localization formula is improved by combining it with the localization law. Thus the one-to-one mapping relationship between HVSRMS and outer ring fault angle position is determined. Finally, experiments are carried out to validate the accuracy and feasibility of the proposed localization law and localization formula. The research results provide a new approach and method for localization diagnosis and the residual life prediction of a defective bearing.

Performance Assessment based on Health Baseline and MML for Hydraulic System

In recent years, with the wide application of hydraulic system, the performance assessment for hydraulic system has gained significant attention. However, a few studies focus on health assessment and traditional methods such as distance metric functions have limitation because different metric functions are only suitable for specific requirements. A scheme of performance degradation assessment based on health baseline and metric learning is proposed in this study. First, General regression neural network (GRNN) based observer is employed as health baseline to generate the estimated output of hydraulic system. The residual is obtained by calculating difference between the actual and estimated output. Then, time domain features are extracted from residual error. After that, apply the Mahalanobis metric learning (MML) to find a suitable metric adaptively for the training data set regarding distance. Finally, the distance between current status and normal status is normalized into confidence value (CV) to quantize the performance. A simulation model of hydraulic system is established based on HyPneu and Simulink, then gradual fault are injected to validate the proposed method. The results of experimental analysis demonstrate the effectiveness and adaptability of the performance degradation assessment method.

An integrated approach to bearing prognostics based on EEMD-multi feature extraction, Gaussian mixture models and Jensen-Rényi divergence

This paper proposes an integrated bearing prognostics approach that is divided into two components: The first component develops a new health indicator (HI) for performance degradation assessment (PDA) of bearings. Many existing HIs suffer from certain shortcomings such as insensitiveness to incipient defects and a highly fluctuating behavior with the increase in degradation severity. To overcome these disadvantages, a HI based on Gaussian mixture models (GMM) and Jensen-Rényi Divergence (JRD) is suggested, which retains its monotonicity as the bearing condition deteriorates. Firstly, the acquired vibration signals are decomposed by ensemble empirical mode decomposition (EEMD) to extract the fault features i.e. singular values and energy moment entropies of the intrinsic mode functions (IMFs). Secondly, the feature vectors under healthy conditions are used to train the GMM. Thirdly, the test feature vectors are supplied to the trained GMM to calculate the components’ posterior probabilities. Fourthly, JRD measure is used to discriminate the defective posterior probability distributions from the healthy ones. Finally, the JRD is converted into confidence value (CV) to realize bearing PDA. The second component utilizes the CV values to train the support vector regression (SVR) model. The particle swarm optimization (PSO) technique is implemented to attain the optimal values of SVR hyperparameters. The optimal SVR model is then used to forecast the CV upto predefined failure threshold and evaluate the bearing remaining useful life (RUL). The experimental results verify that the proposed CV performs better than the GMM-negative log likelihood probability (NLLP) for RUL estimation of bearings.