Pub Date : 2023-02-13DOI: 10.36001/ijphm.2023.v14i3.3129
Kaushal K. Jain, Kuo Yang, P. Meckl, Pingen Chen
This paper presents a diagnostics-oriented aging model for combined Selective Catalytic Reduction (SCR) and Ammonia Slip Catalyst (ASC) system, along with a model-based on-board diagnostic (OBD) method applied to both test-cell data and on-road data from commercial trucks. The key challenge with model development was unavailability of NOx and NH3 measurements between SCR and ASC. Since it would have been very difficult to calibrate both SCR and ASC dynamics without any measurements between SCR and ASC, therefore ASC was modeled using static look-up tables to determine ASC’s NH3 conversion efficiency and its selectivity to NOx and N2O as a function of temperature and flow rate. The traditional three-state single-cell ordinary differential equation (ODE) model was used for SCR. Hot Federal Test Procedure (hFTP) was used to calibrate the model. Cold FTP (cFTP) and Ramped Mode Cycle (RMC) were used for validation. Results show that the SCR-ASC model can capture the aging signatures in tailpipe NOx, NH3, and N2O reasonably well for cFTP, hFTP, and RMC cycles in the testcell data. After slight re-calibration and combining with a simple model for commercial NOx sensor’s cross-sensitivity to NH3, the model works reasonably well for on-road data from commercial trucks. A model-based on-board diagnostic (OBD) method has been presented with enable conditions designed to detect operating conditions suitable for detecting aging signatures, while minimizing false positives and false negatives. The OBD method is applied to both test-cell and real-world truck data with commercial NOx sensors. Results on test-cell data demonstrate the challenges of robust SCR monitoring even on the limited data set used in this work. The model-based enable conditions are shown to be robust but extremely restrictive as the OBD gets enabled at very few points in the test-cell data. Application on truck data showed that the proposed OBD method can be implemented on commercial trucks with limited sensors. In the truck data, the enable conditions were satisfied on many more points than the test-cell data. Results on truck data show encouraging trends between relative degradation level and the number of miles on four trucks. In future work, these trends will be validated using more data from commercial trucks with known aging levels.
{"title":"Diagnostics-oriented Model for Automotive SCR-ASC","authors":"Kaushal K. Jain, Kuo Yang, P. Meckl, Pingen Chen","doi":"10.36001/ijphm.2023.v14i3.3129","DOIUrl":"https://doi.org/10.36001/ijphm.2023.v14i3.3129","url":null,"abstract":"This paper presents a diagnostics-oriented aging model for combined Selective Catalytic Reduction (SCR) and Ammonia Slip Catalyst (ASC) system, along with a model-based on-board diagnostic (OBD) method applied to both test-cell data and on-road data from commercial trucks. The key challenge with model development was unavailability of NOx and NH3 measurements between SCR and ASC. Since it would have been very difficult to calibrate both SCR and ASC dynamics without any measurements between SCR and ASC, therefore ASC was modeled using static look-up tables to determine ASC’s NH3 conversion efficiency and its selectivity to NOx and N2O as a function of temperature and flow rate. The traditional three-state single-cell ordinary differential equation (ODE) model was used for SCR. Hot Federal Test Procedure (hFTP) was used to calibrate the model. Cold FTP (cFTP) and Ramped Mode Cycle (RMC) were used for validation. Results show that the SCR-ASC model can capture the aging signatures in tailpipe NOx, NH3, and N2O reasonably well for cFTP, hFTP, and RMC cycles in the testcell data. After slight re-calibration and combining with a simple model for commercial NOx sensor’s cross-sensitivity to NH3, the model works reasonably well for on-road data from commercial trucks. A model-based on-board diagnostic (OBD) method has been presented with enable conditions designed to detect operating conditions suitable for detecting aging signatures, while minimizing false positives and false negatives. The OBD method is applied to both test-cell and real-world truck data with commercial NOx sensors. Results on test-cell data demonstrate the challenges of robust SCR monitoring even on the limited data set used in this work. The model-based enable conditions are shown to be robust but extremely restrictive as the OBD gets enabled at very few points in the test-cell data. Application on truck data showed that the proposed OBD method can be implemented on commercial trucks with limited sensors. In the truck data, the enable conditions were satisfied on many more points than the test-cell data. Results on truck data show encouraging trends between relative degradation level and the number of miles on four trucks. In future work, these trends will be validated using more data from commercial trucks with known aging levels.","PeriodicalId":42100,"journal":{"name":"International Journal of Prognostics and Health Management","volume":null,"pages":null},"PeriodicalIF":2.1,"publicationDate":"2023-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43750800","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-24DOI: 10.36001/ijphm.2023.v14i1.3395
Mingyue Yu, Jingwen Su, Liqiu Liu, Yi Zhang
Aiming to precisely identify a compound fault of rolling bearing, the paper has contributed a fault characteristic enhancement method by combing entropy weight method (EWM) and intrinsic time scale decomposition (ITD). Firstly, to effectively segregate frequency components in vibration signals, proper rotation components (PRCs) were obtained by decomposing vibration signals based on ITD. Secondly, in view of the fact that amplitude, variance and correlation coefficient vary greatly in a bearing fault accompanied by impact components, parameter evaluation indexes were brought in to depict the fault characteristics of PRCs, including average, variance, correlation coefficient, margin factor, kurtosis, impulse factor, peak factor and so on. Thirdly, weight coefficient of each parameter index was calculated by entropy weight method and the characteristics of each PRC highlighted based on that. Finally, the signals were reconstructed according to the PRCs whose characteristics had been enhanced. Meanwhile reconstructed signals were denoised with singular differential spectrum (SDS) to reduce the influence of noise components, and then the type of compound fault was distinguished grounded on the frequency spectrum. To further prove the efficiency of proposed method, it is compared with other methods (SDS, ITD + entropy method). The result indicates that the proposed method can further highlight the characteristic information of compound faults of bearing and embody more exact identification and judgment on the type of faults.
{"title":"Composite Fault Feature Enhancement Approach for Rolling Bearings Grounded on ITD and Entropy-based Weight Method","authors":"Mingyue Yu, Jingwen Su, Liqiu Liu, Yi Zhang","doi":"10.36001/ijphm.2023.v14i1.3395","DOIUrl":"https://doi.org/10.36001/ijphm.2023.v14i1.3395","url":null,"abstract":"Aiming to precisely identify a compound fault of rolling bearing, the paper has contributed a fault characteristic enhancement method by combing entropy weight method (EWM) and intrinsic time scale decomposition (ITD). Firstly, to effectively segregate frequency components in vibration signals, proper rotation components (PRCs) were obtained by decomposing vibration signals based on ITD. Secondly, in view of the fact that amplitude, variance and correlation coefficient vary greatly in a bearing fault accompanied by impact components, parameter evaluation indexes were brought in to depict the fault characteristics of PRCs, including average, variance, correlation coefficient, margin factor, kurtosis, impulse factor, peak factor and so on. Thirdly, weight coefficient of each parameter index was calculated by entropy weight method and the characteristics of each PRC highlighted based on that. Finally, the signals were reconstructed according to the PRCs whose characteristics had been enhanced. Meanwhile reconstructed signals were denoised with singular differential spectrum (SDS) to reduce the influence of noise components, and then the type of compound fault was distinguished grounded on the frequency spectrum. To further prove the efficiency of proposed method, it is compared with other methods (SDS, ITD + entropy method). The result indicates that the proposed method can further highlight the characteristic information of compound faults of bearing and embody more exact identification and judgment on the type of faults.","PeriodicalId":42100,"journal":{"name":"International Journal of Prognostics and Health Management","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136252197","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-16DOI: 10.36001/ijphm.2023.v14i1.3377
H. Kazemi, Xinyu Du, Hossein Sadjadi
Brake rotors are essential parts of the disc brake systems. Brake rotor thickness variation (RTV) and corrosion are among top failure modes for brake rotors, which may lead to brake judder and pulsation, steering wheel oscillations and chassis vibration. To improve customer satisfaction, vehicle serviceability and availability, it is necessary to develop an onboard fault detection and isolation solution. This study presents a methodology to monitor the state-of-health of brake rotor system to reduce costs associated with scheduled inspection for autonomous fleet or corrective maintenance. We converted the vehicle signals from time-domain to angle-domain and determined health indicators to estimate the RTV level of the rotors. Variance, envelope and order analysis of the brake circuit pressure, longitudinal acceleration and wheel speed sensor signals in angle-domain were promising health indicators to differentiate healthy and faulty rotors. A classification model was developed to fuse the health indicators and estimate the state-of-health of the rotors to report the most degraded rotor with corner isolation. Results showed that using this concept we were able to detect failure levels of 20 microns and larger and meet the customer requirement. Robustness analysis showed that the concept is robust to the noise factors of tire type, tire pressure and vehicle weight. The sensitivity analysis showed that the algorithm is sensitive to two of the calibration parameters (i.e., brake pedal position gradient (BPPG) threshold and the filter order used to derive BPPG) used to determine the brake event and enable the algorithm.
{"title":"Method to Detect and Isolate Brake Rotor Thickness Variation and Corrosion","authors":"H. Kazemi, Xinyu Du, Hossein Sadjadi","doi":"10.36001/ijphm.2023.v14i1.3377","DOIUrl":"https://doi.org/10.36001/ijphm.2023.v14i1.3377","url":null,"abstract":"Brake rotors are essential parts of the disc brake systems. Brake rotor thickness variation (RTV) and corrosion are among top failure modes for brake rotors, which may lead to brake judder and pulsation, steering wheel oscillations and chassis vibration. To improve customer satisfaction, vehicle serviceability and availability, it is necessary to develop an onboard fault detection and isolation solution. This study presents a methodology to monitor the state-of-health of brake rotor system to reduce costs associated with scheduled inspection for autonomous fleet or corrective maintenance. We converted the vehicle signals from time-domain to angle-domain and determined health indicators to estimate the RTV level of the rotors. Variance, envelope and order analysis of the brake circuit pressure, longitudinal acceleration and wheel speed sensor signals in angle-domain were promising health indicators to differentiate healthy and faulty rotors. A classification model was developed to fuse the health indicators and estimate the state-of-health of the rotors to report the most degraded rotor with corner isolation. Results showed that using this concept we were able to detect failure levels of 20 microns and larger and meet the customer requirement. Robustness analysis showed that the concept is robust to the noise factors of tire type, tire pressure and vehicle weight. The sensitivity analysis showed that the algorithm is sensitive to two of the calibration parameters (i.e., brake pedal position gradient (BPPG) threshold and the filter order used to derive BPPG) used to determine the brake event and enable the algorithm.","PeriodicalId":42100,"journal":{"name":"International Journal of Prognostics and Health Management","volume":null,"pages":null},"PeriodicalIF":2.1,"publicationDate":"2023-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43881405","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-15DOI: 10.36001/ijphm.2022.v13i2.3284
Nathaniel DeVol, Christopher Saldaña, Katherine Fu
Accurate estimation of the remaining useful life (RUL) is a key component of condition-based maintenance (CBM) and prognosis and health management (PHM). Data-based models for the estimation of RUL are of particular interest because expert knowledge of systems is not always available, and physical modeling is often not feasible. Additionally, using data-based models, which make decisions based on raw sensor data, allow features to be learned instead of manually determined. In this work, deep convolutional neural network (CNN) architectures are investigated for their ability to estimate the RUL of turbofan engines. To improve the accuracy of the models, CNN architectures, which have proven successful in image classification, are implemented and tested. Specifically, the blocks used in the Visual Geometry Group (VGG) architecture, inception modules used in the GoogLeNet architecture, and residual blocks used in the ResNet architecture are incorporated. To account for varying flight lengths, the input to the models is a window of time series data collected from the engine under test. Window locations at the climb, cruise, and descent stages are considered. To further improve the RUL estimations, multiple overlapping windows at each location are used. This increases the amount of training data available and is found to increase the accuracy of the resulting RUL estimations by averaging the estimates from all overlapping segments. The model is trained and tested using the new Commercial Modular Aero-Propulsion System Simulation (N-CMAPSS) data set, and high prognosis accuracy was achieved. This work expands on the model developed and used in the 2021 PHM Society Data Challenge, which received second place.
{"title":"Evaluating Image Classification Deep Convolutional Neural Network Architectures for Remaining Useful Life Estimation of Turbofan Engines","authors":"Nathaniel DeVol, Christopher Saldaña, Katherine Fu","doi":"10.36001/ijphm.2022.v13i2.3284","DOIUrl":"https://doi.org/10.36001/ijphm.2022.v13i2.3284","url":null,"abstract":"Accurate estimation of the remaining useful life (RUL) is a key component of condition-based maintenance (CBM) and prognosis and health management (PHM). Data-based models for the estimation of RUL are of particular interest because expert knowledge of systems is not always available, and physical modeling is often not feasible. Additionally, using data-based models, which make decisions based on raw sensor data, allow features to be learned instead of manually determined. In this work, deep convolutional neural network (CNN) architectures are investigated for their ability to estimate the RUL of turbofan engines. To improve the accuracy of the models, CNN architectures, which have proven successful in image classification, are implemented and tested. Specifically, the blocks used in the Visual Geometry Group (VGG) architecture, inception modules used in the GoogLeNet architecture, and residual blocks used in the ResNet architecture are incorporated. To account for varying flight lengths, the input to the models is a window of time series data collected from the engine under test. Window locations at the climb, cruise, and descent stages are considered. To further improve the RUL estimations, multiple overlapping windows at each location are used. This increases the amount of training data available and is found to increase the accuracy of the resulting RUL estimations by averaging the estimates from all overlapping segments. The model is trained and tested using the new Commercial Modular Aero-Propulsion System Simulation (N-CMAPSS) data set, and high prognosis accuracy was achieved. This work expands on the model developed and used in the 2021 PHM Society Data Challenge, which received second place.","PeriodicalId":42100,"journal":{"name":"International Journal of Prognostics and Health Management","volume":null,"pages":null},"PeriodicalIF":2.1,"publicationDate":"2022-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46534827","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Breakdowns and unplanned shutdowns in industrial processes and equipment can lead to significant loss of availability and revenue. It is imperative to perform optimal maintenance of such systems when signs of abnormal behavior are detected and before they propagate and lead to catastrophic failure. This is particularly challenging in systems with interconnected multiple components as it is difficult to isolate the effect of one component on the operation of other components in the system. In this work, an ensemble approach based on Cascaded Convolutional neural network and Long Short-term Memory (CC-LSTM) network models is proposed for detecting and predicting the time of onset of faults in interconnected multicomponent systems. The performance of the ensemble CC-LSTM model was demonstrated on an industrial 4-component system and was found to improve the accuracy of onset time predictions by ~15% compared to individual CC-LSTM models and ~25-40% compared to commonly used deep learning techniques such as dense neural networks, convolutional neural networks and LSTMs. The CC-LSTM and the ensemble models also had the lowest missed detection rates and zero false positive rates making them ideal for real-time monitoring and fault detection in multicomponent systems.
{"title":"Ensemble Deep Learning for Detecting Onset of Abnormal Operation in Industrial Multi-component Systems","authors":"Balaji Selvanathan, S. Nistala, Venkataramana Runkana, Saurabh Jaywant Desai, Shashank Agarwal","doi":"10.36001/ijphm.2022.v13i2.3093","DOIUrl":"https://doi.org/10.36001/ijphm.2022.v13i2.3093","url":null,"abstract":"Breakdowns and unplanned shutdowns in industrial processes and equipment can lead to significant loss of availability and revenue. It is imperative to perform optimal maintenance of such systems when signs of abnormal behavior are detected and before they propagate and lead to catastrophic failure. This is particularly challenging in systems with interconnected multiple components as it is difficult to isolate the effect of one component on the operation of other components in the system. In this work, an ensemble approach based on Cascaded Convolutional neural network and Long Short-term Memory (CC-LSTM) network models is proposed for detecting and predicting the time of onset of faults in interconnected multicomponent systems. The performance of the ensemble CC-LSTM model was demonstrated on an industrial 4-component system and was found to improve the accuracy of onset time predictions by ~15% compared to individual CC-LSTM models and ~25-40% compared to commonly used deep learning techniques such as dense neural networks, convolutional neural networks and LSTMs. The CC-LSTM and the ensemble models also had the lowest missed detection rates and zero false positive rates making them ideal for real-time monitoring and fault detection in multicomponent systems.","PeriodicalId":42100,"journal":{"name":"International Journal of Prognostics and Health Management","volume":null,"pages":null},"PeriodicalIF":2.1,"publicationDate":"2022-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42272296","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-10-17DOI: 10.36001/ijphm.2022.v13i2.3267
Ayantha Senanayaka, Abdullah Al Mamun, Glenn Bond, Wenmeng Tian, Haifeng Wang, Sara Fuller, T.C. Falls, Shahram Rahimi, L. Bian
This study aims to develop an effective method of classification concerning time series signals for machine state prediction to advance predictive maintenance (PdM). Conventional machine learning (ML) algorithms are widely adopted in PdM, however, most existing methods assume that the training (source) and testing (target) data follow the same distribution, and that labeled data are available in both source and target domains. For real-world PdM applications, the heterogeneity in machine original equipment manufacturers (OEMs), operating conditions, facility environment, and maintenance records collectively lead to heterogeneous distribution for data collected from different machines. This will significantly limit the performance of conventional ML algorithms in PdM. Moreover, labeling data is generally costly and time-consuming. Finally, industrial processes incorporate complex conditions, and unpredictable breakdown modes lead to extreme complexities for PdM. In this study, similarity-based multi-source transfer learning (SiMuS-TL) approach is proposed for real-time classification of time series signals. A new domain, called "mixed domain," is established to model the hidden similarities among the multiple sources and the target. The proposed SiMuS-TL model mainly includes three key steps: 1) learning group-based feature patterns, 2) developing group-based pre-trained models, and 3) weight transferring. The proposed SiMuS-TL model is validated by observing the state of the rotating machinery using a dataset collected on the Skill boss manufacturing system, publicly available standard bearing datasets, Case Western Reserve University (CWRU), and Paderborn University (PU) bearing datasets. The results of the performance comparison demonstrate that the proposed SiMuS-TL method outperformed conventional Support Vector Machine (SVM), Artificial Neural Network (ANN), and Transfer learning with neural networks (TLNN) without similarity-based transfer learning methods.
{"title":"Similarity-based Multi-source Transfer Learning Approach for Time Series Classification","authors":"Ayantha Senanayaka, Abdullah Al Mamun, Glenn Bond, Wenmeng Tian, Haifeng Wang, Sara Fuller, T.C. Falls, Shahram Rahimi, L. Bian","doi":"10.36001/ijphm.2022.v13i2.3267","DOIUrl":"https://doi.org/10.36001/ijphm.2022.v13i2.3267","url":null,"abstract":"This study aims to develop an effective method of classification concerning time series signals for machine state prediction to advance predictive maintenance (PdM). Conventional machine learning (ML) algorithms are widely adopted in PdM, however, most existing methods assume that the training (source) and testing (target) data follow the same distribution, and that labeled data are available in both source and target domains. For real-world PdM applications, the heterogeneity in machine original equipment manufacturers (OEMs), operating conditions, facility environment, and maintenance records collectively lead to heterogeneous distribution for data collected from different machines. This will significantly limit the performance of conventional ML algorithms in PdM. Moreover, labeling data is generally costly and time-consuming. Finally, industrial processes incorporate complex conditions, and unpredictable breakdown modes lead to extreme complexities for PdM. In this study, similarity-based multi-source transfer learning (SiMuS-TL) approach is proposed for real-time classification of time series signals. A new domain, called \"mixed domain,\" is established to model the hidden similarities among the multiple sources and the target. The proposed SiMuS-TL model mainly includes three key steps: 1) learning group-based feature patterns, 2) developing group-based pre-trained models, and 3) weight transferring. The proposed SiMuS-TL model is validated by observing the state of the rotating machinery using a dataset collected on the Skill boss manufacturing system, publicly available standard bearing datasets, Case Western Reserve University (CWRU), and Paderborn University (PU) bearing datasets. The results of the performance comparison demonstrate that the proposed SiMuS-TL method outperformed conventional Support Vector Machine (SVM), Artificial Neural Network (ANN), and Transfer learning with neural networks (TLNN) without similarity-based transfer learning methods.","PeriodicalId":42100,"journal":{"name":"International Journal of Prognostics and Health Management","volume":null,"pages":null},"PeriodicalIF":2.1,"publicationDate":"2022-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41677389","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-29DOI: 10.36001/ijphm.2022.v13i2.3205
A. Anuj, Gurmeet Singh, Vallayil Narayana Achutha Naikan
Rotating machinery generally consist of a driver machine such as a motor and a driven machine or load such as a compressor or pump. Several condition monitoring (CM) techniques have been developed over the years for the predictive maintenance of rotating machinery. An appropriate selection of these techniques needs to be established for maximizing the ROI (Return on investment) of such systems. This paper proposes a methodology for the proper selection of CM techniques based on factors such as fault detectability, fault severity, cost, ease of data collection, noise, and system criticality. Effective techniques are recommended based on applicability in the industrial scenario and research done till now. A careful scoring system was adopted and weightage was given to each factor by expert opinion depending on its importance in the industrial environment. Multi-criteria decision-making (MCDM) was used to obtain comparable technique combination scores. The effectiveness of a single technique was found limited in rotating machinery, effective combinations were made and scored according to important factors. Final scores were obtained and top combinations were chosen for non-critical, sub-critical, and critical systems. A possible way of implementation is also shown for remote monitoring through literature.
{"title":"Methodology for Selection of Condition Monitoring Techniques for Rotating Machinery","authors":"A. Anuj, Gurmeet Singh, Vallayil Narayana Achutha Naikan","doi":"10.36001/ijphm.2022.v13i2.3205","DOIUrl":"https://doi.org/10.36001/ijphm.2022.v13i2.3205","url":null,"abstract":"Rotating machinery generally consist of a driver machine such as a motor and a driven machine or load such as a compressor or pump. Several condition monitoring (CM) techniques have been developed over the years for the predictive maintenance of rotating machinery. An appropriate selection of these techniques needs to be established for maximizing the ROI (Return on investment) of such systems. This paper proposes a methodology for the proper selection of CM techniques based on factors such as fault detectability, fault severity, cost, ease of data collection, noise, and system criticality. Effective techniques are recommended based on applicability in the industrial scenario and research done till now. A careful scoring system was adopted and weightage was given to each factor by expert opinion depending on its importance in the industrial environment. Multi-criteria decision-making (MCDM) was used to obtain comparable technique combination scores. The effectiveness of a single technique was found limited in rotating machinery, effective combinations were made and scored according to important factors. Final scores were obtained and top combinations were chosen for non-critical, sub-critical, and critical systems. A possible way of implementation is also shown for remote monitoring through literature.","PeriodicalId":42100,"journal":{"name":"International Journal of Prognostics and Health Management","volume":null,"pages":null},"PeriodicalIF":2.1,"publicationDate":"2022-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41679994","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-16DOI: 10.36001/ijphm.2022.v13i2.3138
T. Shibutani
Companies utilize highly accelerated limit testing (HALT) to ensure efficient product development by accelerating loading conditions in the qualification process. The aim of qualitative accelerated testing such as HALT is to properly identify early behavioral anomalies. To this end, this study utilizes machine learning techniques for detecting anomalies in servomotors in electronic products subjected to HALT. A case study was conducted using a programmable robot kit with 12 servomotors. HALT comprises five types of stress: thermal conditioning (cold and heat), rapid thermal change, vibration, and combined stresses. The anomalous behavior of a servomotor can be identified using a k-nearest neighbor algorithm and verified by inspection using the loading conditions and electrical responses. In addition, anomalous behaviors among servomotors and a control board are assessed using a Gaussian graph model approach. Changes in the Gaussian graph are assessed as anomaly scores using Kullback–Leibler divergence. The anomaly score increased earlier than the operating limit defined by inspection, i.e., the deviation from the initial position of the shaft. The machine learning algorithm successfully identified the failure precursor of the unit. The proposed approach of HALT with the machine learning algorithm supports prognostic health management of servomotors.
{"title":"Anomaly Detection of Servomotors Subject to Highly Accelerated Limit Testing","authors":"T. Shibutani","doi":"10.36001/ijphm.2022.v13i2.3138","DOIUrl":"https://doi.org/10.36001/ijphm.2022.v13i2.3138","url":null,"abstract":"Companies utilize highly accelerated limit testing (HALT) to ensure efficient product development by accelerating loading conditions in the qualification process. The aim of qualitative accelerated testing such as HALT is to properly identify early behavioral anomalies. To this end, this study utilizes machine learning techniques for detecting anomalies in servomotors in electronic products subjected to HALT. A case study was conducted using a programmable robot kit with 12 servomotors. HALT comprises five types of stress: thermal conditioning (cold and heat), rapid thermal change, vibration, and combined stresses. The anomalous behavior of a servomotor can be identified using a k-nearest neighbor algorithm and verified by inspection using the loading conditions and electrical responses. In addition, anomalous behaviors among servomotors and a control board are assessed using a Gaussian graph model approach. Changes in the Gaussian graph are assessed as anomaly scores using Kullback–Leibler divergence. The anomaly score increased earlier than the operating limit defined by inspection, i.e., the deviation from the initial position of the shaft. The machine learning algorithm successfully identified the failure precursor of the unit. The proposed approach of HALT with the machine learning algorithm supports prognostic health management of servomotors.","PeriodicalId":42100,"journal":{"name":"International Journal of Prognostics and Health Management","volume":null,"pages":null},"PeriodicalIF":2.1,"publicationDate":"2022-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48021671","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-18DOI: 10.36001/ijphm.2022.v13i2.3127
A. Trilla, Nenad Mijatovic, Xavier Vilasis-Cardona
This work explores how the causality inference paradigm may be applied to troubleshoot the root causes of failures through language processing and Deep Learning. To do so, the causality hierarchy has been taken for reference: associative, interventional, and retrospective levels of causality have thus been researched within textual data in the form of a failure analysis ontology and a set of written records on Return On Experience. A novel approach to extracting linguistic knowledge has been devised through the joint embedding of two contextualized Bag-Of-Words models, which defines both a probabilistic framework and a distributed representation of the underlying causal semantics. This method has been applied to the maintenance of rolling stock bogies, and the results indicate that the inference of causality has been partially attained with the currently available technical documentation (consensus over 70%). However, there is still some disagreement between root causes and problems that leads to confusion and uncertainty. In consequence, the proposed approach may be used as a strategy to detect lexical imprecision, make writing recommendations in the form of standard reporting guidelines, and ultimately help produce clearer diagnosis materials to increase the safety of the railway service.
{"title":"Towards Learning Causal Representations of Technical Word Embeddings for Smart Troubleshooting","authors":"A. Trilla, Nenad Mijatovic, Xavier Vilasis-Cardona","doi":"10.36001/ijphm.2022.v13i2.3127","DOIUrl":"https://doi.org/10.36001/ijphm.2022.v13i2.3127","url":null,"abstract":"This work explores how the causality inference paradigm may be applied to troubleshoot the root causes of failures through language processing and Deep Learning. To do so, the causality hierarchy has been taken for reference: associative, interventional, and retrospective levels of causality have thus been researched within textual data in the form of a failure analysis ontology and a set of written records on Return On Experience. A novel approach to extracting linguistic knowledge has been devised through the joint embedding of two contextualized Bag-Of-Words models, which defines both a probabilistic framework and a distributed representation of the underlying causal semantics. This method has been applied to the maintenance of rolling stock bogies, and the results indicate that the inference of causality has been partially attained with the currently available technical documentation (consensus over 70%). However, there is still some disagreement between root causes and problems that leads to confusion and uncertainty. In consequence, the proposed approach may be used as a strategy to detect lexical imprecision, make writing recommendations in the form of standard reporting guidelines, and ultimately help produce clearer diagnosis materials to increase the safety of the railway service.","PeriodicalId":42100,"journal":{"name":"International Journal of Prognostics and Health Management","volume":null,"pages":null},"PeriodicalIF":2.1,"publicationDate":"2022-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43841410","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jet engines are critical assets in aircraft, and their availability is crucial in the modern aircraft industry. Therefore, their maintenance scheduling is one of the major tasks an airline has to make during an engine’s lifetime. A proper engine maintenance schedule can significantly reduce maintenance costs without compromising the aircraft's reliability and safety. Different maintenance scheduling approaches have been used for jet engines, such as corrective, preventive, and predictive maintenance strategies. Regarding the safety demands in aircraft industries, preventive maintenance is a frequent maintenance method for jet engines. However, preventive maintenance schedules are often use fixed maintenance intervals, which is usually suboptimal. This paper focuses on minimizing a jet engine's overall maintenance cost by optimizing its preventive maintenance schedule based on an engine’s comprehensive reliability model. A hierarchical optimization framework including the golden section search and genetic algorithms is applied to find the optimal set of preventive maintenance number and their times and the components to be replaced at those times during the jet engine's overall lifetime. The Monte Carlo simulation is used to estimate the engine’s failure times using their lifetime distributions from the reliability model. The estimated failure times are then used to determine the engine's overall corrective and preventive maintenance costs during its lifetime. Finally, an optimal preventive maintenance schedule is proposed for an RB 211 jet engine using the presented method. In the end, comparing the proposed method's overall maintenance cost with two other maintenance methods demonstrates the proposed schedule's effectiveness. The method presented in this paper is generic, and it can be used for other similar engineering systems.
{"title":"Jet Engine Optimal Preventive Maintenance Scheduling Using Golden Section Search and Genetic Algorithm","authors":"Sajad Saraygord Afshari, Bhargava Jonnadula, Xiangyang Xu, Xihui Liang, Zhaohui Yang","doi":"10.22215/jphm.v2i1.3321","DOIUrl":"https://doi.org/10.22215/jphm.v2i1.3321","url":null,"abstract":"Jet engines are critical assets in aircraft, and their availability is crucial in the modern aircraft industry. Therefore, their maintenance scheduling is one of the major tasks an airline has to make during an engine’s lifetime. A proper engine maintenance schedule can significantly reduce maintenance costs without compromising the aircraft's reliability and safety. Different maintenance scheduling approaches have been used for jet engines, such as corrective, preventive, and predictive maintenance strategies. Regarding the safety demands in aircraft industries, preventive maintenance is a frequent maintenance method for jet engines. However, preventive maintenance schedules are often use fixed maintenance intervals, which is usually suboptimal. This paper focuses on minimizing a jet engine's overall maintenance cost by optimizing its preventive maintenance schedule based on an engine’s comprehensive reliability model. A hierarchical optimization framework including the golden section search and genetic algorithms is applied to find the optimal set of preventive maintenance number and their times and the components to be replaced at those times during the jet engine's overall lifetime. The Monte Carlo simulation is used to estimate the engine’s failure times using their lifetime distributions from the reliability model. The estimated failure times are then used to determine the engine's overall corrective and preventive maintenance costs during its lifetime. Finally, an optimal preventive maintenance schedule is proposed for an RB 211 jet engine using the presented method. In the end, comparing the proposed method's overall maintenance cost with two other maintenance methods demonstrates the proposed schedule's effectiveness. The method presented in this paper is generic, and it can be used for other similar engineering systems.","PeriodicalId":42100,"journal":{"name":"International Journal of Prognostics and Health Management","volume":null,"pages":null},"PeriodicalIF":2.1,"publicationDate":"2022-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72539634","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：