Pub Date : 2022-05-01DOI: 10.1109/PHM2022-London52454.2022.00043
Jianwen Meng, Meiling Yue, D. Diallo
The first priority of battery predictive maintenance is to estimate its end-of-life (EOL) cycle and assess the uncertainty associated with the predicted values. In this paper, a hybrid method combining empirical mode decomposition (EMD) and particle filter (PF) is applied to an open source database of NASA Ames Prognostics Center of Excellence for the early EOL prediction of four battery cells. The results show a clear decreasing trend of EOL prediction uncertainty when the prediction starts from later operation cycles. However, the distance between the true EOL and the mean predicted EOL has no obvious decrease when more operation data is available. Interestingly, the mean predicted EOL is lower than the true EOL with more available operation data, which is meaningful for reliability engineering and system safety. For instance, the final EOL prediction results from the 80-th cycle are 17 cycles, 7 cycles, 33 cycles and 16 cycles earlier than the real values, respectively.
电池预测维护的首要任务是估算电池的寿命终止周期，并评估与预测值相关的不确定性。本文将经验模态分解(EMD)和粒子滤波(PF)相结合的混合方法应用于NASA Ames Prognostics Center of Excellence的开源数据库，对4个电池单体的早期EOL进行了预测。结果表明，从较晚的运行周期开始，EOL预测的不确定性有明显的下降趋势。然而，当有更多的操作数据时，真实EOL与平均预测EOL之间的距离没有明显减小。有趣的是，当可用的运行数据更多时，平均预测EOL低于真实EOL，这对可靠性工程和系统安全具有重要意义。例如，第80周期的最终EOL预测结果分别比实际值早17、7、33和16个周期。
{"title":"Battery Early End-Of-Life Prediction and Its Uncertainty Assessment with Empirical Mode Decomposition and Particle Filter","authors":"Jianwen Meng, Meiling Yue, D. Diallo","doi":"10.1109/PHM2022-London52454.2022.00043","DOIUrl":"https://doi.org/10.1109/PHM2022-London52454.2022.00043","url":null,"abstract":"The first priority of battery predictive maintenance is to estimate its end-of-life (EOL) cycle and assess the uncertainty associated with the predicted values. In this paper, a hybrid method combining empirical mode decomposition (EMD) and particle filter (PF) is applied to an open source database of NASA Ames Prognostics Center of Excellence for the early EOL prediction of four battery cells. The results show a clear decreasing trend of EOL prediction uncertainty when the prediction starts from later operation cycles. However, the distance between the true EOL and the mean predicted EOL has no obvious decrease when more operation data is available. Interestingly, the mean predicted EOL is lower than the true EOL with more available operation data, which is meaningful for reliability engineering and system safety. For instance, the final EOL prediction results from the 80-th cycle are 17 cycles, 7 cycles, 33 cycles and 16 cycles earlier than the real values, respectively.","PeriodicalId":269605,"journal":{"name":"2022 Prognostics and Health Management Conference (PHM-2022 London)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115514996","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 : 2022-05-01DOI: 10.1109/PHM2022-London52454.2022.00096
Yingshun Li, Hongda Kan, Aina Wang, Zhannan Guo
In order to save the high cost of tank maintenance, reduce the redundant input of manpower and material resources for tank maintenance, and improve the reliability of tank performance, a fault diagnosis method based on NRS and WOA-SVM is proposed. Taking the fire control computer and sensor subsystem of a certain type of tank fire control system as the research object, the NRS algorithm is used to reduce the properties of the performance parameters of the fire control computer, and the most important performance index is selected. Then, a novel meta-heuristic algorithm, WOA, is used to optimize the parameters of the SVM, and the fault data classification model is constructed according to the global best fitness function value. Finally, the attribute-reduced dataset is input into the WOA-SVM fault classification model to realize the fault diagnosis of the system. The experimental results show that the method can effectively evaluate the health status and fault diagnosis of the fire control system, achieve the purpose of precise maintenance, repair and replacement, and improve the reliability of the equipment.
{"title":"Fault Diagnosis of Tank Fire Control System Based on NRS and WOA-SVM","authors":"Yingshun Li, Hongda Kan, Aina Wang, Zhannan Guo","doi":"10.1109/PHM2022-London52454.2022.00096","DOIUrl":"https://doi.org/10.1109/PHM2022-London52454.2022.00096","url":null,"abstract":"In order to save the high cost of tank maintenance, reduce the redundant input of manpower and material resources for tank maintenance, and improve the reliability of tank performance, a fault diagnosis method based on NRS and WOA-SVM is proposed. Taking the fire control computer and sensor subsystem of a certain type of tank fire control system as the research object, the NRS algorithm is used to reduce the properties of the performance parameters of the fire control computer, and the most important performance index is selected. Then, a novel meta-heuristic algorithm, WOA, is used to optimize the parameters of the SVM, and the fault data classification model is constructed according to the global best fitness function value. Finally, the attribute-reduced dataset is input into the WOA-SVM fault classification model to realize the fault diagnosis of the system. The experimental results show that the method can effectively evaluate the health status and fault diagnosis of the fire control system, achieve the purpose of precise maintenance, repair and replacement, and improve the reliability of the equipment.","PeriodicalId":269605,"journal":{"name":"2022 Prognostics and Health Management Conference (PHM-2022 London)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123008244","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 : 2022-05-01DOI: 10.1109/PHM2022-London52454.2022.00037
Shushuai Xie, Wei Cheng, Zelin Nie, Xuefeng Chen
Intelligent fault diagnosis of bearing based on data drive has been a hot research field in recent years and achieved lots of results. However, current research mainly faces: 1) It is a great challenge to develop an effective intelligent diagnosis method in practical industrial scenarios because of the lack of fault signals in small samples; 2) It has poor adaptability to intelligent fault diagnosis under variable working conditions. Aiming at the above problems, an intelligent fault diagnosis method for bearings under variable working conditions and small samples based on generative adversarial network is proposed. Firstly, the signal highly similar to the actual fault signal is generated through generative adversarial network training and this part of the signal can be used as training data to solve the problem of deficient small sample fault dataset. Then, the similar fault characteristics learned from the data of a certain working condition through domain confrontation training are transferred to the target working condition. Finally, fault diagnosis is realized on the target domain data by the classifier trained on the fault features. The proposed method is evaluated through the Case Western Reserve University (CWRU) bearing dataset with the result show that it has high fault classification accuracy and transferability under the condition of small samples and variable working conditions.
{"title":"Intelligent Fault Diagnosis of Bearings under Variable Working Conditions and Small Samples with Generative Adversarial Network","authors":"Shushuai Xie, Wei Cheng, Zelin Nie, Xuefeng Chen","doi":"10.1109/PHM2022-London52454.2022.00037","DOIUrl":"https://doi.org/10.1109/PHM2022-London52454.2022.00037","url":null,"abstract":"Intelligent fault diagnosis of bearing based on data drive has been a hot research field in recent years and achieved lots of results. However, current research mainly faces: 1) It is a great challenge to develop an effective intelligent diagnosis method in practical industrial scenarios because of the lack of fault signals in small samples; 2) It has poor adaptability to intelligent fault diagnosis under variable working conditions. Aiming at the above problems, an intelligent fault diagnosis method for bearings under variable working conditions and small samples based on generative adversarial network is proposed. Firstly, the signal highly similar to the actual fault signal is generated through generative adversarial network training and this part of the signal can be used as training data to solve the problem of deficient small sample fault dataset. Then, the similar fault characteristics learned from the data of a certain working condition through domain confrontation training are transferred to the target working condition. Finally, fault diagnosis is realized on the target domain data by the classifier trained on the fault features. The proposed method is evaluated through the Case Western Reserve University (CWRU) bearing dataset with the result show that it has high fault classification accuracy and transferability under the condition of small samples and variable working conditions.","PeriodicalId":269605,"journal":{"name":"2022 Prognostics and Health Management Conference (PHM-2022 London)","volume":"209 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121083580","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 : 2022-05-01DOI: 10.1109/PHM2022-London52454.2022.00064
Su Caiyu, Dong Jie, Mo Yi, Wu Shanyun
The booming socio-economic development has led to great progress in the Internet of Things (IoT) and computer technology, which are gradually applied in all aspects of society. The Internet of Health Things (IoH) has emerged to meet the new era of higher demands placed on medical institutions. Machine learning is beginning to be used in the medical service system and is achieving significant results in driving related services. This paper analyses sensor data from several elderly people to analyse their postural status and text reports on classification metrics. The performance of the support vector machine on this problem is evaluated using information such as accuracy, recall, and F1 value. The study achieves a more accurate judgement of the health status of the elderly and provides some help to medical institutions in developing relevant treatment plans, as well as providing a reference for related academic research. abstract
{"title":"A Classification Application Based on Support Vector Machine for Health IoT","authors":"Su Caiyu, Dong Jie, Mo Yi, Wu Shanyun","doi":"10.1109/PHM2022-London52454.2022.00064","DOIUrl":"https://doi.org/10.1109/PHM2022-London52454.2022.00064","url":null,"abstract":"The booming socio-economic development has led to great progress in the Internet of Things (IoT) and computer technology, which are gradually applied in all aspects of society. The Internet of Health Things (IoH) has emerged to meet the new era of higher demands placed on medical institutions. Machine learning is beginning to be used in the medical service system and is achieving significant results in driving related services. This paper analyses sensor data from several elderly people to analyse their postural status and text reports on classification metrics. The performance of the support vector machine on this problem is evaluated using information such as accuracy, recall, and F1 value. The study achieves a more accurate judgement of the health status of the elderly and provides some help to medical institutions in developing relevant treatment plans, as well as providing a reference for related academic research. abstract","PeriodicalId":269605,"journal":{"name":"2022 Prognostics and Health Management Conference (PHM-2022 London)","volume":"118 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129201065","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 : 2022-05-01DOI: 10.1109/PHM2022-London52454.2022.00072
Yong Zhang, Guangjun He, Yuanyuan Yu, Guanjian Li
In data analysis, the analysis efficiency and accuracy can be significantly improved after preprocessing the original data. And Symbolic Aggregate approXimation(SAX) is an effective data compression analysis method. Because of its simple, intuitive and effective characteristics, it has become the most typical symbolic feature representation method. However, in the approximate data compression of segmented aggregation, this method adopts a unified average method regardless of the characteristics of the data, which weakens the prominent characteristics of the data and causes the loss of effective information, which has a negative impact on the accuracy of data mining and analysis. Aiming at this problem, a local gradient search method (LGS) is proposed, which is the LGS-SAX method for piecewise aggregated symbol approximation. It can use gradient transformation to perceive the angle to prevent the loss of feature information, so as to achieve the effect of efficiently compressing data and retaining feature information. Through error analysis and comparison, the method has small error, complete information retention, and the method is efficient and feasible.
{"title":"A Data Processing Method of Symbolic Approximation","authors":"Yong Zhang, Guangjun He, Yuanyuan Yu, Guanjian Li","doi":"10.1109/PHM2022-London52454.2022.00072","DOIUrl":"https://doi.org/10.1109/PHM2022-London52454.2022.00072","url":null,"abstract":"In data analysis, the analysis efficiency and accuracy can be significantly improved after preprocessing the original data. And Symbolic Aggregate approXimation(SAX) is an effective data compression analysis method. Because of its simple, intuitive and effective characteristics, it has become the most typical symbolic feature representation method. However, in the approximate data compression of segmented aggregation, this method adopts a unified average method regardless of the characteristics of the data, which weakens the prominent characteristics of the data and causes the loss of effective information, which has a negative impact on the accuracy of data mining and analysis. Aiming at this problem, a local gradient search method (LGS) is proposed, which is the LGS-SAX method for piecewise aggregated symbol approximation. It can use gradient transformation to perceive the angle to prevent the loss of feature information, so as to achieve the effect of efficiently compressing data and retaining feature information. Through error analysis and comparison, the method has small error, complete information retention, and the method is efficient and feasible.","PeriodicalId":269605,"journal":{"name":"2022 Prognostics and Health Management Conference (PHM-2022 London)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126066291","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 : 2022-01-13DOI: 10.1109/PHM2022-London52454.2022.00059
Rui Wang
With the fast development of machine learning technologies, deep learning models have been deployed in almost every aspect of everyday life. However, the privacy and security of these models are threatened by adversarial attacks. Among which black-box attack is closer to reality, where limited knowledge can be acquired from the model. In this paper, we provided basic background knowledge about adversarial attack and analyzed four black-box attack algorithms: Bandits, NES, Square Attack and ZOsignSGD comprehensively. We also explored the newly proposed Square Attack method with respect to square size, hoping to improve its query efficiency.
{"title":"Evaluation of Four Black-box Adversarial Attacks and Some Query-efficient Improvement Analysis","authors":"Rui Wang","doi":"10.1109/PHM2022-London52454.2022.00059","DOIUrl":"https://doi.org/10.1109/PHM2022-London52454.2022.00059","url":null,"abstract":"With the fast development of machine learning technologies, deep learning models have been deployed in almost every aspect of everyday life. However, the privacy and security of these models are threatened by adversarial attacks. Among which black-box attack is closer to reality, where limited knowledge can be acquired from the model. In this paper, we provided basic background knowledge about adversarial attack and analyzed four black-box attack algorithms: Bandits, NES, Square Attack and ZOsignSGD comprehensively. We also explored the newly proposed Square Attack method with respect to square size, hoping to improve its query efficiency.","PeriodicalId":269605,"journal":{"name":"2022 Prognostics and Health Management Conference (PHM-2022 London)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115815181","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：