Achmad Widodo

Intelligent prognostics for battery health monitoring based on sample entropy

In this paper, an intelligent prognostic for battery health based on sample entropy (SampEn) feature of discharge voltage is proposed. SampEn can provide computational means for assessing the predictability of a time series and also can quantity the regularity of a data sequence. Therefore, when it is applied to discharge voltage battery data, it could serve an indicator for battery health. In this work, the intelligent ability is introduced by utilizing machine learning methods namely support vector machine (SVM) and relevance vector machine (RVM). SampEn and estimated state of charge (SOH) are employed as data input and target vector of learning algorithms, respectively. The results show that the proposed method is plausible due to the good performance of SVM and RVM in SOH prediction. In our study, RVM outperforms SVM based battery health prognostics.

Machine health prognostics using survival probability and support vector machine

Prognostic of machine health estimates the remaining useful life of machine components. It deals with prediction of machine health condition based on past measured data from condition monitoring (CM). It has benefits to reduce the production downtime, spare-parts inventory, maintenance cost, and safety hazards. Many papers have reported the valuable models and methods of prognostics systems. However, it was rarely found the papers deal with censored data, which is common in machine condition monitoring practice. This work concerns with developing intelligent machine prognostics system using survival analysis and support vector machine (SVM). SA utilizes censored and uncensored data collected from CM routine and then estimates the survival probability of failure time of machine components. SVM is trained by data input from CM histories data that corresponds to target vectors of estimated survival probability. After validation process, SVM is employed to predict failure time of individual unit of machine component. Simulation and experimental bearing degradation data are employed to validate the proposed method. The result shows that the proposed method is promising to be a probability-based machine prognostics system.

Combined Probability Approach and Indirect Data-Driven Method for Bearing Degradation Prognostics

This study proposes an application of relevance vector machine (RVM), logistic regression (LR), and autoregressive moving average/generalized autoregressive conditional heteroscedasticity (ARMA/GARCH) models to assess failure degradation based on run-to-failure bearing simulating data. Failure degradation is calculated by using an LR model, and then regarded as the target vectors of the failure probability for training the RVM model. A multi-step-ahead method-based ARMA/GARCH is used to predict censored data, and its prediction performance is compared with one of Dempster-Shafer regression (DSR) method. Furthermore, RVM is selected as an intelligent system, and trained by run-to-failure bearing data and the target vectors of failure probability obtained from the LR model. After training, RVM is employed to predict the failure probability of individual units of bearing samples. In addition, statistical process control is used to analyze the variance of the failure probability. The result shows the novelty of the proposed method, which can be considered as a valid machine degradation prognostic model.

Application of relevance vector machine and survival probability to machine degradation assessment

Condition monitoring (CM) of machines health or industrial components and systems that can detect, classify and predict the impending faults is critical in reducing operating and maintenance cost. Many papers have reported the valuable models and methods of prognostic systems. However, it was rarely found the papers deal with censored data, which was common in machine condition monitoring practice. This work deals with development of machine degradation assessment system that utilizes censored and complete data collected from CM routine. Relevance vector machine (RVM) is selected as intelligent system then trained by input data obtained from run-to-failure bearing data and target vectors of survival probability estimated by Kaplan-Meier (KM) and probability density function estimators. After validation process, RVM is employed to predict survival probability of individual unit of machine component. The plausibility of the proposed method is shown by applying the proposed method to bearing degradation data in predicting survival probability of individual unit.

A Comparison of Classifier Performance for Fault Diagnosis of Induction Motor using Multi-type Signals

Fault detection and diagnosis is the most important technology in condition-based maintenance (CBM) systems, which typically starts from collecting signatures of running machines by multiple sensors for subsequent accurate analysis. Recently, there has been an increasing requirement of selecting special sensors, which are cheap, robust, easily installed, and good classifiers that have accurate classification, stable performance, and short calculating time. This article carries out a comparative study of various classification algorithms for fault diagnosis of electric motors using different types of signals. The authors evaluate experimentally the relative performances of five classifiers using five types of steady-state signals based on three kinds of performance evaluation strategies: training-test, cross-validation, and similar measure. First, the raw signals are collected and features are extracted from the collected signals. Then, the extracted features are classified using the five classification algorithms. Next, an overall comparison of the five classifiers is described, and experiment results are discussed. Finally, conclusions are summarized and suggestions are offered.

FAULT DIAGNOSIS OF LOW SPEED BEARING BASED ON ACOUSTIC EMISSION SIGNAL AND MULTI-CLASS RELEVANCE VECTOR MACHINE

This study presents an acoustic emission (AE) based fault diagnosis for low speed bearing using multi-class relevance vector machine (RVM). A low speed test rig was developed to simulate the various defects with shaft speeds as low as 10 rpm under several loading conditions. The data was acquired using an AE sensor with the test bearing operating at a constant loading (5 kN) and with a speed range from 20 to 80 rpm. This study is aimed at finding a reliable method/tool for low speed machines fault diagnosis based on AE signal. In the present study, component analysis was performed to extract the bearing feature and to reduce the dimensionality of original data feature. The result shows that multi-class RVM offers a promising approach for fault diagnosis of low speed machines.

Evaluation of thermography image data for machine fault diagnosis

A novel approach for fault diagnosis of rotating machine based on thermal image investigation using image histogram features is proposed in this paper. Herein, the machine learning and statistical approach are adopted along with thermal image signal to machine condition diagnosis. Using thermal images, the information of machine condition can be investigated more simply than other conventional methods of machine condition monitoring. In this work, the behaviour of thermal image is investigated with different conditions of machine. A test-rig that represents the machine in industry was set up to produce thermal image data in the experiment. Some significant features have been extracted and selected by means of principal component analysis, and independent component analysis, and other irrelevant features have been discarded. The aim of this study is to retrieve thermal images by means of selecting a proper feature to recognise the fault pattern of the machine. The result shows that the classification process of thermal image features by support vector machine and other classifiers can serve machine fault diagnosis.

Confirmation of Thermal Images and Vibration Signals for Intelligent Machine Fault Diagnostics

This paper deals with the maintenance technique for industrial machinery using the artificial neural network so-called self-organizing map (SOM). The aim of this work is to develop intelligent maintenance system for machinery based on an alternative way, namely, thermal images instead of vibration signals. SOM is selected due to its simplicity and is categorized as an unsupervised algorithm. Following the SOM training, machine fault diagnostics is performed by using the pattern recognition technique of machine conditions. The data used in this work are thermal images and vibration signals, which were acquired from machine fault simulator (MFS). It is a reliable tool and is able to simulate several conditions of faulty machine such as unbalance, misalignment, looseness, and rolling element bearing faults (outer race, inner race, ball, and cage defects). Data acquisition were conducted simultaneously by infrared thermography camera and vibration sensors installed in the MFS. The experimental data are presented as thermal image and vibration signal in the time domain. Feature extraction was carried out to obtain salient features sensitive to machine conditions from thermal images and vibration signals. These features are then used to train the SOM for intelligent machine diagnostics process. The results show that SOM can perform intelligent fault diagnostics with plausible accuracies.

Machine degradation prognostic based on RVM and ARMA/GARCH model for bearing fault simulated data

Recently, prognostics is an active area and growth rapidly. In this paper, bearing prognostic has been studied in viewpoint of failure degradation as an object of prediction. This study proposes the application of relevance vector machine (RVM), logistic regression (LR) and ARMA/GARCH in order to assess the failure degradation of run-to-failure bearing simulated data. Failure degradation is calculated using LR and then regarded as target vectors of failure probability for RVM training. ARMA/GARCH based on multi-step-ahead prediction is employed for censored data. Furthermore, RVM is selected as intelligent system then trained by using run-to-failure bearing data and target vectors of failure probability estimated by LR. After training process, RVM is employed to predict failure probability of individual unit of bearing sample. The result shows the novelty of the proposed method which can be considered as machine degradation prognostic model.

Degradation Trend Estimation and Prognosis of Large Low Speed Slewing Bearing Lifetime

In many applications, degradation of bearing conditions is usually monitored by changes in time-domain features. However, in low speed (< 10 rpm) slewing bearing, these changes are not easily detected because of the low energy and low frequency of the vibration. To overcome this problem, a combined low pass filter (LPF) and adaptive line enhancer (ALE) signal pre-conditioning method is used. Time-domain features such as root mean square (RMS), skewness and kurtosis are extracted from the output signal of the combined LPF and ALE method. The extracted features show accurate information about the incipient of fault as compared to extracted features from the original vibration signal. This information then triggers the prognostic algorithm to predict the remaining lifetime of the bearing. The algorithm used to determine the trend of the non-stationary data is auto-regressive integrated moving average (ARIMA).