Pub Date : 2023-05-01DOI: 10.1109/PHM58589.2023.00054
Liuxing Bai
Developing as servo drive technology is, permanent-magnet synchronous motor is gradually replacing DC motor and stepper motor and become the development direction of servo drive. Because the permanent-magnet synchronous servo system is affected by the motor parameter change, external load disturbance and other factors to obtain good performance and wide speed range of permanent magnet synchronous servo system, we must study advanced control strategy and control means, so that the adaptability and strong anti-interference ability of the system are strong. In this paper, the vector control of permanent-magnet synchronous motor is simulated in MATLAB.
{"title":"Research on the vector of permanent magnet synchronous motor based on MATLAB simulation","authors":"Liuxing Bai","doi":"10.1109/PHM58589.2023.00054","DOIUrl":"https://doi.org/10.1109/PHM58589.2023.00054","url":null,"abstract":"Developing as servo drive technology is, permanent-magnet synchronous motor is gradually replacing DC motor and stepper motor and become the development direction of servo drive. Because the permanent-magnet synchronous servo system is affected by the motor parameter change, external load disturbance and other factors to obtain good performance and wide speed range of permanent magnet synchronous servo system, we must study advanced control strategy and control means, so that the adaptability and strong anti-interference ability of the system are strong. In this paper, the vector control of permanent-magnet synchronous motor is simulated in MATLAB.","PeriodicalId":196601,"journal":{"name":"2023 Prognostics and Health Management Conference (PHM)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130633356","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}
Remaining Useful Life (RUL) prognostics and pre-failure warning for complex industrial systems enables the timely detection of hidden problems and effectively avoids multiple accidents. Therefore, highly accurate and reliable RUL prediction is crucial. Bayesian neural networks can model the uncertainty in the process of equipment degradation while effectively assessing RUL, which helps to implement reliable risk analysis and maintenance decisions. In this paper, we propose a Convolutional Bayesian Long Short-Term Memory neural network (CB-LSTM)-based RUL prediction algorithm, which uses a Convolutional Neural Network (CNN) to implicitly extract features from training data, to generate an abstract representation of the input signal, and combine it with a Bayesian Long Short-Term Memory neural network (B-LSTM) to build a multivariate time series prediction model. The method is validated on the C-MAPSS dataset by NASA. The experimental results show that the method has good prediction accuracy and uncertainty quantification ability.
{"title":"Remaining Useful Life Prognostics and Uncertainty Quantification for Aircraft Engines Based on Convolutional Bayesian Long Short-Term Memory Neural Network","authors":"Shaowei Chen, Jiawei He, Pengfei Wen, Jing Zhang, Dengshan Huang, Shuai Zhao","doi":"10.1109/PHM58589.2023.00052","DOIUrl":"https://doi.org/10.1109/PHM58589.2023.00052","url":null,"abstract":"Remaining Useful Life (RUL) prognostics and pre-failure warning for complex industrial systems enables the timely detection of hidden problems and effectively avoids multiple accidents. Therefore, highly accurate and reliable RUL prediction is crucial. Bayesian neural networks can model the uncertainty in the process of equipment degradation while effectively assessing RUL, which helps to implement reliable risk analysis and maintenance decisions. In this paper, we propose a Convolutional Bayesian Long Short-Term Memory neural network (CB-LSTM)-based RUL prediction algorithm, which uses a Convolutional Neural Network (CNN) to implicitly extract features from training data, to generate an abstract representation of the input signal, and combine it with a Bayesian Long Short-Term Memory neural network (B-LSTM) to build a multivariate time series prediction model. The method is validated on the C-MAPSS dataset by NASA. The experimental results show that the method has good prediction accuracy and uncertainty quantification ability.","PeriodicalId":196601,"journal":{"name":"2023 Prognostics and Health Management Conference (PHM)","volume":"105 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124096476","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 : 2023-05-01DOI: 10.1109/PHM58589.2023.00027
A. Mosallam, Fares Ben Youssef, Karolina Sobczak-Oramus, Jinlong Kang, Vikrant Gupta, Nannan Shen, L. Laval
This paper presents a novel data-driven approach for modeling degradation of the neutron generator component in logging-while-drilling tools. The study begins by identifying the incipient failure modes of the neutron generator and constructing a health indicator (HI) that serves as a quantitative measure of the component’s health status. The resulting HI can be used for additional analysis and decision-making. Then, a random forest classifier is trained to establish the relationship between the extracted HI values and the corresponding degradation level labels. The proposed method is validated using actual data collected from oil well drilling operations. The experimental results demonstrate its effectiveness in accurately classifying the health state of the neutron generator component. The study is part of a long-term project aimed at developing a digital fleet management system for drilling tools.
{"title":"Data-Driven Degradation Modeling Approach for Neutron Generators in Multifunction Logging-While-Drilling Service","authors":"A. Mosallam, Fares Ben Youssef, Karolina Sobczak-Oramus, Jinlong Kang, Vikrant Gupta, Nannan Shen, L. Laval","doi":"10.1109/PHM58589.2023.00027","DOIUrl":"https://doi.org/10.1109/PHM58589.2023.00027","url":null,"abstract":"This paper presents a novel data-driven approach for modeling degradation of the neutron generator component in logging-while-drilling tools. The study begins by identifying the incipient failure modes of the neutron generator and constructing a health indicator (HI) that serves as a quantitative measure of the component’s health status. The resulting HI can be used for additional analysis and decision-making. Then, a random forest classifier is trained to establish the relationship between the extracted HI values and the corresponding degradation level labels. The proposed method is validated using actual data collected from oil well drilling operations. The experimental results demonstrate its effectiveness in accurately classifying the health state of the neutron generator component. The study is part of a long-term project aimed at developing a digital fleet management system for drilling tools.","PeriodicalId":196601,"journal":{"name":"2023 Prognostics and Health Management Conference (PHM)","volume":"456 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124322070","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 : 2023-05-01DOI: 10.1109/PHM58589.2023.00053
Lu Zhang, C. Delpha, D. Diallo
This work proposes a method for estimating fault severity in the presence of noise using the measured currents for a 7-phase electrical machine. The method is based on analytical models in stationary reference frames and analysis of the DC and fundamental components in the four fictitious machines. The slope of the decision function from the CUSUM algorithm, which will be noticeably different depending on the fault severity, is used to assess the performance of the fault severity estimation rapidly. The effects on the decision function’s slope of the fault severity estimation for different noise levels are evaluated. The simulation results show that even in presence of high noise levels, the decision function is an efficient fault estimation indicator. When the noise level is high, the decision function and its slope are noisier. Conversely, the decision function and its slope are less noisy when the noise level is low. The results also show that for the three fault types under study (gain fault, phase shift fault, and mean value fault), the current components of the fictitious machines in the stationary frames have distinct robustness to noise.
{"title":"Performance of Fault Severity Estimation in 7-Phase Electrical Machines under Noisy Conditions","authors":"Lu Zhang, C. Delpha, D. Diallo","doi":"10.1109/PHM58589.2023.00053","DOIUrl":"https://doi.org/10.1109/PHM58589.2023.00053","url":null,"abstract":"This work proposes a method for estimating fault severity in the presence of noise using the measured currents for a 7-phase electrical machine. The method is based on analytical models in stationary reference frames and analysis of the DC and fundamental components in the four fictitious machines. The slope of the decision function from the CUSUM algorithm, which will be noticeably different depending on the fault severity, is used to assess the performance of the fault severity estimation rapidly. The effects on the decision function’s slope of the fault severity estimation for different noise levels are evaluated. The simulation results show that even in presence of high noise levels, the decision function is an efficient fault estimation indicator. When the noise level is high, the decision function and its slope are noisier. Conversely, the decision function and its slope are less noisy when the noise level is low. The results also show that for the three fault types under study (gain fault, phase shift fault, and mean value fault), the current components of the fictitious machines in the stationary frames have distinct robustness to noise.","PeriodicalId":196601,"journal":{"name":"2023 Prognostics and Health Management Conference (PHM)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124743078","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 : 2023-05-01DOI: 10.1109/PHM58589.2023.00061
Yi Sun, Canyu Cai, Hongli Gao, Zhichao You
Tool condition monitoring in high-speed cutting machining is essential to ensure the machining surface accuracy requirements, improve the tool utilization and extend the machine tool life. However, it is challenging to screen and process the data of each stage of feed-path. Moreover, how to utilize the massive unlabeled data of different machining parameters in the actual machining process is an open problem. To address these challenges, this paper proposes the TCM-U2PL model, comprising a teacher model and a student model, which can adaptively extract the data of cutting stages with tool condition features and improve model performance using unlabeled data. First, the teacher model consists of two independent classifiers in a multi-branch classification model, which can adaptively extract and classify the tool condition features in the cutting stage and can label part of the unlabeled data as positive samples and negative samples. Then, the student model identifies the tool condition with high accuracy by minimizing the marginal distribution discrepancy and maximizing the conditional distribution alignment. The model was validated on the tool condition dataset, and TCM-U2PL achieved a classification accuracy of 85.7%, significantly outperforming CNN, DA-DBN, and NSVDD models.
{"title":"Online Tool Condition Monitoring Using Unreliable Pseudo-Labels","authors":"Yi Sun, Canyu Cai, Hongli Gao, Zhichao You","doi":"10.1109/PHM58589.2023.00061","DOIUrl":"https://doi.org/10.1109/PHM58589.2023.00061","url":null,"abstract":"Tool condition monitoring in high-speed cutting machining is essential to ensure the machining surface accuracy requirements, improve the tool utilization and extend the machine tool life. However, it is challenging to screen and process the data of each stage of feed-path. Moreover, how to utilize the massive unlabeled data of different machining parameters in the actual machining process is an open problem. To address these challenges, this paper proposes the TCM-U2PL model, comprising a teacher model and a student model, which can adaptively extract the data of cutting stages with tool condition features and improve model performance using unlabeled data. First, the teacher model consists of two independent classifiers in a multi-branch classification model, which can adaptively extract and classify the tool condition features in the cutting stage and can label part of the unlabeled data as positive samples and negative samples. Then, the student model identifies the tool condition with high accuracy by minimizing the marginal distribution discrepancy and maximizing the conditional distribution alignment. The model was validated on the tool condition dataset, and TCM-U2PL achieved a classification accuracy of 85.7%, significantly outperforming CNN, DA-DBN, and NSVDD models.","PeriodicalId":196601,"journal":{"name":"2023 Prognostics and Health Management Conference (PHM)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126566006","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}
Most data-driven fault diagnosis methods for analog circuits achieve good results when the data satisfies the assumption of independent and equal distribution, which is difficult to achieve in real-world scenarios. To solve this problem, a fault diagnosis method for analog circuits based on Deep Subdomain Adaptation Network is presented. By incorporating the optimization of Local Maximum Mean Discrepancy loss into the training of One-dimensional Convolutional Neural Network, this method can adaptively align the feature representation of the source and target domains without labeling in the target domain. The simulation experiments of Sallen-Key band-pass filter and four-opamp biquad high-pass filter are designed. Two groups of different component parameters are selected as the data sources of source domain and target domain, noise and random offset are added to the target domain data to simulate the actual scene. Through comparative experiments, it is verified that the analog circuit fault diagnosis method presented in this paper has steady training and high accuracy.
{"title":"A Transfer Learning Method for Fault Diagnosis of Analog Circuit Using Deep Subdomain Adaptation Network","authors":"Weizheng Chen, Xu Han, Guangquan Zhao, Xiyuan Peng","doi":"10.1109/PHM58589.2023.00056","DOIUrl":"https://doi.org/10.1109/PHM58589.2023.00056","url":null,"abstract":"Most data-driven fault diagnosis methods for analog circuits achieve good results when the data satisfies the assumption of independent and equal distribution, which is difficult to achieve in real-world scenarios. To solve this problem, a fault diagnosis method for analog circuits based on Deep Subdomain Adaptation Network is presented. By incorporating the optimization of Local Maximum Mean Discrepancy loss into the training of One-dimensional Convolutional Neural Network, this method can adaptively align the feature representation of the source and target domains without labeling in the target domain. The simulation experiments of Sallen-Key band-pass filter and four-opamp biquad high-pass filter are designed. Two groups of different component parameters are selected as the data sources of source domain and target domain, noise and random offset are added to the target domain data to simulate the actual scene. Through comparative experiments, it is verified that the analog circuit fault diagnosis method presented in this paper has steady training and high accuracy.","PeriodicalId":196601,"journal":{"name":"2023 Prognostics and Health Management Conference (PHM)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132978090","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}
I am honored to host the DSD/SEAA event in Verona, one of the most important cities in North-Eastern Italy. Verona is a splendid city of art, well-known through the Shakespearean tragedy of Romeo and Juliet. Roman ruins, medieval vestiges, Venetian and Austrian traces can be seen all across the city, as well as antique palaces, bridges and churches. For these reasons Verona is the fourth Italian city for the number of tourists and it is recognized as a UNESCO World Heritage Site. Close to Verona, you can also visit Lake Garda, the greatest Italian lake, impressive mountains, and lovely hills full of vineyards.
{"title":"Message from the General Chair","authors":"Mark A. Gondree","doi":"10.1109/ICPADS.2006.61","DOIUrl":"https://doi.org/10.1109/ICPADS.2006.61","url":null,"abstract":"I am honored to host the DSD/SEAA event in Verona, one of the most important cities in North-Eastern Italy. Verona is a splendid city of art, well-known through the Shakespearean tragedy of Romeo and Juliet. Roman ruins, medieval vestiges, Venetian and Austrian traces can be seen all across the city, as well as antique palaces, bridges and churches. For these reasons Verona is the fourth Italian city for the number of tourists and it is recognized as a UNESCO World Heritage Site. Close to Verona, you can also visit Lake Garda, the greatest Italian lake, impressive mountains, and lovely hills full of vineyards.","PeriodicalId":196601,"journal":{"name":"2023 Prognostics and Health Management Conference (PHM)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133771989","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 : 2023-05-01DOI: 10.1109/PHM58589.2023.00039
Yaoxiang Yu, Liang Guo, Hongli Gao
The axle-box bearing (ABB) makes crucial influence on the operation of urban rail vehicles through supporting the weight of the vehicle and load, lubricating the axle neck, and reducing friction. However, wheel-polygonal wear (WPW) can compromise the stability of the vehicle by aggravating the axle-box vibration. This work aims to study the dynamic characteristics of ABB in the presence of WPW. On one hand, a vehicle-track coupled dynamics model with ABB and flexible wheelset is established. Onsite tests are implemented to validated the effectiveness of this model, and the significance of the first flexible mode are also researched. On other hand, the study also analyzes the influence of WPW amplitude and order on ABB by inputting WPW into the model at different vehicle speeds. The results indicate that the amplitude of WPW influences the axle-box vibration amplitude, with an increase in amplitude leading to an increase in vibration amplitude. However, the influence of the order of WPW is more complex due to the existence of resonance phenomenon. The findings of this study can guide the maintenance of wheel machining and repair in urban rail vehicles, providing reference and guidance for future research in this area.
{"title":"A Study on the Effect of Wheel-polygonal Wear on Dynamic Vibration Characteristics of Urban Rail Vehicle Axle-box Bearings","authors":"Yaoxiang Yu, Liang Guo, Hongli Gao","doi":"10.1109/PHM58589.2023.00039","DOIUrl":"https://doi.org/10.1109/PHM58589.2023.00039","url":null,"abstract":"The axle-box bearing (ABB) makes crucial influence on the operation of urban rail vehicles through supporting the weight of the vehicle and load, lubricating the axle neck, and reducing friction. However, wheel-polygonal wear (WPW) can compromise the stability of the vehicle by aggravating the axle-box vibration. This work aims to study the dynamic characteristics of ABB in the presence of WPW. On one hand, a vehicle-track coupled dynamics model with ABB and flexible wheelset is established. Onsite tests are implemented to validated the effectiveness of this model, and the significance of the first flexible mode are also researched. On other hand, the study also analyzes the influence of WPW amplitude and order on ABB by inputting WPW into the model at different vehicle speeds. The results indicate that the amplitude of WPW influences the axle-box vibration amplitude, with an increase in amplitude leading to an increase in vibration amplitude. However, the influence of the order of WPW is more complex due to the existence of resonance phenomenon. The findings of this study can guide the maintenance of wheel machining and repair in urban rail vehicles, providing reference and guidance for future research in this area.","PeriodicalId":196601,"journal":{"name":"2023 Prognostics and Health Management Conference (PHM)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131790975","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 : 2023-05-01DOI: 10.1109/PHM58589.2023.00019
Gil-hyun Kang, Hwi-Jin Kwon, In Soo Chung, Chul-Su Kim
Most of the maintenance and training of the railway vehicle of the Korean urban railway operator was conducted in the form of document-based manuals or internet-based e-learning. This training method is inefficient due to restrictions such as time, space, and human resource operation. This study is about the development of high-definition augmented reality content for innovation in existing education and training in accordance with the recent smart maintenance transition and digitalization of maintenance. To this end, the realistic contents for maintenance and training of commuter rail vehicle air compressors that can increase immersion and realism for railway vehicle maintenance workers were developed. In addition, a questionnaire evaluation was conducted on field applicability. Rail vehicle maintenance workers can receive maintenance support by efficiently accessing work information at the workplace using mobile devices. In order to evaluate the usability of the developed air compressor maintenance augmented reality content, a usability evaluation survey was conducted on 100 college students majoring in railway vehicles. The overall average score of the 6 questionnaire items for the content was 4.12 out of 5 points, which was very good. Therefore, this content is very useful for beginners in maintenance of railway vehicles and is considered to be very effective in using it for maintenance and training of air compressors in the workplace.
{"title":"A Study on the Development of Augmented Reality Contents for Air Compressor of Railway Vehicles","authors":"Gil-hyun Kang, Hwi-Jin Kwon, In Soo Chung, Chul-Su Kim","doi":"10.1109/PHM58589.2023.00019","DOIUrl":"https://doi.org/10.1109/PHM58589.2023.00019","url":null,"abstract":"Most of the maintenance and training of the railway vehicle of the Korean urban railway operator was conducted in the form of document-based manuals or internet-based e-learning. This training method is inefficient due to restrictions such as time, space, and human resource operation. This study is about the development of high-definition augmented reality content for innovation in existing education and training in accordance with the recent smart maintenance transition and digitalization of maintenance. To this end, the realistic contents for maintenance and training of commuter rail vehicle air compressors that can increase immersion and realism for railway vehicle maintenance workers were developed. In addition, a questionnaire evaluation was conducted on field applicability. Rail vehicle maintenance workers can receive maintenance support by efficiently accessing work information at the workplace using mobile devices. In order to evaluate the usability of the developed air compressor maintenance augmented reality content, a usability evaluation survey was conducted on 100 college students majoring in railway vehicles. The overall average score of the 6 questionnaire items for the content was 4.12 out of 5 points, which was very good. Therefore, this content is very useful for beginners in maintenance of railway vehicles and is considered to be very effective in using it for maintenance and training of air compressors in the workplace.","PeriodicalId":196601,"journal":{"name":"2023 Prognostics and Health Management Conference (PHM)","volume":"110 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121210037","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 : 2023-05-01DOI: 10.1109/PHM58589.2023.00050
Abdelilah Hammou, Jianwen Meng, D. Diallo, R. Petrone, H. Gualous
State of health monitoring for batteries is of utmost importance for efficient and secured operations. This work proposes a hybrid approach to forecast battery’s performance losses. Particularly, the proposed method combines the Kalman filter (KF) and Gaussian Process Regression (GPR) techniques to predict the battery capacity evolution with aging. The effectiveness of the approach is validated based on experimental data. Data are obtained testing four cells of lithium nickel manganese cobalt oxide. These cells are cycled using a dynamic current profile derived from the Worldwide Harmonized Light Vehicles Test Cycle (WLTC) under controlled temperature conditions. The proposed method is validated by comparing the actual End of Life (EoL) with the predicted, one obtained with different sections of the training dataset; 30%, 50% and 70%. The results show that the best average prediction error is obtained when the training data set is larger, and the aging trend is uniform. The results also show that the dispersion around the estimated EoL is lower when the training data set is larger. For seven of the twelve case studies, the estimated EoL is lower than the actual one, which is a conservative but good scenario for safety reasons.
{"title":"State-of-health prediction of Li-ion NMC Batteries Using Kalman Filter and Gaussian Process Regression","authors":"Abdelilah Hammou, Jianwen Meng, D. Diallo, R. Petrone, H. Gualous","doi":"10.1109/PHM58589.2023.00050","DOIUrl":"https://doi.org/10.1109/PHM58589.2023.00050","url":null,"abstract":"State of health monitoring for batteries is of utmost importance for efficient and secured operations. This work proposes a hybrid approach to forecast battery’s performance losses. Particularly, the proposed method combines the Kalman filter (KF) and Gaussian Process Regression (GPR) techniques to predict the battery capacity evolution with aging. The effectiveness of the approach is validated based on experimental data. Data are obtained testing four cells of lithium nickel manganese cobalt oxide. These cells are cycled using a dynamic current profile derived from the Worldwide Harmonized Light Vehicles Test Cycle (WLTC) under controlled temperature conditions. The proposed method is validated by comparing the actual End of Life (EoL) with the predicted, one obtained with different sections of the training dataset; 30%, 50% and 70%. The results show that the best average prediction error is obtained when the training data set is larger, and the aging trend is uniform. The results also show that the dispersion around the estimated EoL is lower when the training data set is larger. For seven of the twelve case studies, the estimated EoL is lower than the actual one, which is a conservative but good scenario for safety reasons.","PeriodicalId":196601,"journal":{"name":"2023 Prognostics and Health Management Conference (PHM)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115615091","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：