Pub Date : 2014-06-22DOI: 10.1109/ICPHM.2014.7036387
Jing Tian, M. Azarian, M. Pecht
Fault detection is a critical task in condition-based maintenance of rolling element bearings. In many applications unsupervised learning techniques are preferred in fault detection due to the lack of training data. Unsupervised learning techniques such as k-means clustering are most widely used in machinery health monitoring. These methods face two challenges: firstly, they cannot cluster non-convex data, which may have arbitrary shape; secondly, no rule has been established for these techniques to find a fault threshold. This paper introduces a fault detection methodology based on density clustering to address these challenges. This methodology assumes that data from healthy bearings is located in regions with a high density and data from faulty bearings is located in low density regions. By finding boundaries of these regions, which may be non-convex, data from faulty bearings can be identified. In this paper the value of the density for healthy bearings and faulty bearings is evaluated. The rate of change of the density from healthy to faulty is identified as a fault threshold. The methodology is validated by experimental data. This methodology can be applied to applications where faulty data are too difficult or costly to acquire. Also it can be used in applications where fault thresholds are difficult to determine.
{"title":"Rolling element bearing fault detection using density-based clustering","authors":"Jing Tian, M. Azarian, M. Pecht","doi":"10.1109/ICPHM.2014.7036387","DOIUrl":"https://doi.org/10.1109/ICPHM.2014.7036387","url":null,"abstract":"Fault detection is a critical task in condition-based maintenance of rolling element bearings. In many applications unsupervised learning techniques are preferred in fault detection due to the lack of training data. Unsupervised learning techniques such as k-means clustering are most widely used in machinery health monitoring. These methods face two challenges: firstly, they cannot cluster non-convex data, which may have arbitrary shape; secondly, no rule has been established for these techniques to find a fault threshold. This paper introduces a fault detection methodology based on density clustering to address these challenges. This methodology assumes that data from healthy bearings is located in regions with a high density and data from faulty bearings is located in low density regions. By finding boundaries of these regions, which may be non-convex, data from faulty bearings can be identified. In this paper the value of the density for healthy bearings and faulty bearings is evaluated. The rate of change of the density from healthy to faulty is identified as a fault threshold. The methodology is validated by experimental data. This methodology can be applied to applications where faulty data are too difficult or costly to acquire. Also it can be used in applications where fault thresholds are difficult to determine.","PeriodicalId":376942,"journal":{"name":"2014 International Conference on Prognostics and Health Management","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126019972","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 : 2014-06-22DOI: 10.1109/ICPHM.2014.7036372
Long Chen, Junyou Shi
Dependency matrix is a kind of description form of the dependency relationships between the faults and tests in a UUT (Unit Under Test). Testability analysis, testability prediction and fault diagnosis can be accomplished on the basis of dependency matrix. The traditional method of acquiring dependency matrix is to search the signal flow of the UUT's dependency graphical model. The difficulties in integrating the models together and the exponentially growing workload to search the signal flow are two major disadvantages of the traditional method. To overcome these disadvantages, a new method for dependency matrix combination based on port connection relationship is proposed. Definitions of port-extended dependency matrix and global port connection matrix are given. The principle of dependency matrix combination method is analyzed with the definitions of the two matrixes above given. The generation algorithms of a port-extended dependency matrix and a global port connection matrix are studied. Based on the port-extended dependency and the global port connection matrix, the algorithm flow of the dependency matrix combination method is given. A case of an avionic system is given and the above method is applied. The result shows that the method is feasible and effective.
依赖矩阵是UUT (Unit Under Test)中故障与测试之间依赖关系的一种描述形式。基于依赖矩阵可以完成测试性分析、测试性预测和故障诊断。传统的依赖矩阵获取方法是搜索UUT依赖图模型的信号流。传统方法的两个主要缺点是难以将模型整合在一起以及搜索信号流的工作量呈指数增长。为了克服这些缺点，提出了一种基于端口连接关系的依赖矩阵组合方法。给出了端口扩展依赖矩阵和全局端口连接矩阵的定义。分析了相关矩阵组合法的原理，给出了相关矩阵和相关矩阵的定义。研究了端口扩展依赖矩阵和全局端口连接矩阵的生成算法。基于端口扩展依赖关系和全局端口连接矩阵，给出了依赖矩阵组合法的算法流程。最后以某航电系统为例，对上述方法进行了应用。结果表明，该方法是可行和有效的。
{"title":"A method for dependency matrix combination based on port connection relationship","authors":"Long Chen, Junyou Shi","doi":"10.1109/ICPHM.2014.7036372","DOIUrl":"https://doi.org/10.1109/ICPHM.2014.7036372","url":null,"abstract":"Dependency matrix is a kind of description form of the dependency relationships between the faults and tests in a UUT (Unit Under Test). Testability analysis, testability prediction and fault diagnosis can be accomplished on the basis of dependency matrix. The traditional method of acquiring dependency matrix is to search the signal flow of the UUT's dependency graphical model. The difficulties in integrating the models together and the exponentially growing workload to search the signal flow are two major disadvantages of the traditional method. To overcome these disadvantages, a new method for dependency matrix combination based on port connection relationship is proposed. Definitions of port-extended dependency matrix and global port connection matrix are given. The principle of dependency matrix combination method is analyzed with the definitions of the two matrixes above given. The generation algorithms of a port-extended dependency matrix and a global port connection matrix are studied. Based on the port-extended dependency and the global port connection matrix, the algorithm flow of the dependency matrix combination method is given. A case of an avionic system is given and the above method is applied. The result shows that the method is feasible and effective.","PeriodicalId":376942,"journal":{"name":"2014 International Conference on Prognostics and Health Management","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128694387","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 : 2014-06-22DOI: 10.1109/ICPHM.2014.7036384
C. Iamsumang
This paper presents a new computational algorithm for reliability inference with dynamic hybrid Bayesian network. It features a component-based algorithm and structure to represent complex engineering systems characterized by discrete functional states (including degraded states), and models of underlying physics of failure, with continuous variables. The methodology is designed to be flexible and intuitive, and scalable from small localized functionality to large complex dynamic systems. Markov Chain Monte Carlo (MCMC) inference is optimized using pre-computation and dynamic programming for real-time monitoring of system health. The scope of this research includes new modeling approach, computation algorithm, and an example application for on-line System Health Management.
{"title":"Computational algorithm for dynamic hybrid Bayesian network in on-line system health management applications","authors":"C. Iamsumang","doi":"10.1109/ICPHM.2014.7036384","DOIUrl":"https://doi.org/10.1109/ICPHM.2014.7036384","url":null,"abstract":"This paper presents a new computational algorithm for reliability inference with dynamic hybrid Bayesian network. It features a component-based algorithm and structure to represent complex engineering systems characterized by discrete functional states (including degraded states), and models of underlying physics of failure, with continuous variables. The methodology is designed to be flexible and intuitive, and scalable from small localized functionality to large complex dynamic systems. Markov Chain Monte Carlo (MCMC) inference is optimized using pre-computation and dynamic programming for real-time monitoring of system health. The scope of this research includes new modeling approach, computation algorithm, and an example application for on-line System Health Management.","PeriodicalId":376942,"journal":{"name":"2014 International Conference on Prognostics and Health Management","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116335407","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 : 2014-06-22DOI: 10.1109/ICPHM.2014.7036376
Yangming Guo, Xiangtao Wang, Yafei Zheng, Chong Liu
Since it is difficult to establish precise physical model of complex systems, time series prediction is often used to predict their health trend and running state. Aiming at online prediction, we proposed a new scheme to fix the problems of time series online prediction, which is based on LS-SVR model and incremental learning algorithm. The scheme includes two aspects. Firstly, by replacing single kernel with new fixed kernel consisting of several basis kernels, a better information mapping in high dimension is obtained; secondly, by establishing new LS-SVR model without bias term b, the calculation process with incremental learning is simplified. Prediction experiment is performed via certain avionics application. The results indicate preliminarily that the proposed scheme is an effective prediction approach for its good prediction precision and less computing time. The method will be useful in actual application.
{"title":"Improved multi-kernel LS-SVR for time series online prediction with incremental learning","authors":"Yangming Guo, Xiangtao Wang, Yafei Zheng, Chong Liu","doi":"10.1109/ICPHM.2014.7036376","DOIUrl":"https://doi.org/10.1109/ICPHM.2014.7036376","url":null,"abstract":"Since it is difficult to establish precise physical model of complex systems, time series prediction is often used to predict their health trend and running state. Aiming at online prediction, we proposed a new scheme to fix the problems of time series online prediction, which is based on LS-SVR model and incremental learning algorithm. The scheme includes two aspects. Firstly, by replacing single kernel with new fixed kernel consisting of several basis kernels, a better information mapping in high dimension is obtained; secondly, by establishing new LS-SVR model without bias term b, the calculation process with incremental learning is simplified. Prediction experiment is performed via certain avionics application. The results indicate preliminarily that the proposed scheme is an effective prediction approach for its good prediction precision and less computing time. The method will be useful in actual application.","PeriodicalId":376942,"journal":{"name":"2014 International Conference on Prognostics and Health Management","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125231910","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 : 2014-06-22DOI: 10.1109/ICPHM.2014.7036379
Jamie L. Godwin, Peter C. Matthews, Christopher Watson
This paper presents a novel methodology for the identification and detection of faults on based upon high frequency bearing data collected from a 6205-2RS JEM SKF bearing. A robust derivative of the Mahalanobis distance is employed to accurately and precisely encapsulate varying fault behaviours, which can then be exploited for the purposes of fault detection and identification. Domain knowledge in the form of failure mode and effect analysis (FMEA) can be incorporated into the model, to determine potential failure modes. Seeded fault data was employed to derive the shape and location estimates to enable the use of a multivariate distance function. To reduce the computational complexity whilst simultaneously increasing sensitivity to the faults, the high frequency (48KHz) accelerometer data was pre-processed into a 4-tuple consisting of the Skewness, Kurtosis, Root mean square (RMS) and Shannon Entropy. This 4-tuple is shown to encapsulate and discriminate all fault modes identified through the FMEA, whilst reducing the data to 1Hz, allowing for the both exact, and meta-heuristic algorithms to be employed for robust analysis. Sensitivity to minimal fault development is demonstrated, with the technique accurately identifying 0.007" diameter inner race, outer race and roller element faults which had been seeded to the bearing through electro-discharge machining. To demonstrate the practicalities of the approach, the trained system is employed for analysis of an independent dataset, collected under different conditions. The technique is shown to accurately detect and identify the relevant fault mode pre-emptively, before catastrophic failure occurred, with 28.6% of bearing life remaining.
{"title":"Robust multivariate statistical ensembles for bearing fault detection and identification","authors":"Jamie L. Godwin, Peter C. Matthews, Christopher Watson","doi":"10.1109/ICPHM.2014.7036379","DOIUrl":"https://doi.org/10.1109/ICPHM.2014.7036379","url":null,"abstract":"This paper presents a novel methodology for the identification and detection of faults on based upon high frequency bearing data collected from a 6205-2RS JEM SKF bearing. A robust derivative of the Mahalanobis distance is employed to accurately and precisely encapsulate varying fault behaviours, which can then be exploited for the purposes of fault detection and identification. Domain knowledge in the form of failure mode and effect analysis (FMEA) can be incorporated into the model, to determine potential failure modes. Seeded fault data was employed to derive the shape and location estimates to enable the use of a multivariate distance function. To reduce the computational complexity whilst simultaneously increasing sensitivity to the faults, the high frequency (48KHz) accelerometer data was pre-processed into a 4-tuple consisting of the Skewness, Kurtosis, Root mean square (RMS) and Shannon Entropy. This 4-tuple is shown to encapsulate and discriminate all fault modes identified through the FMEA, whilst reducing the data to 1Hz, allowing for the both exact, and meta-heuristic algorithms to be employed for robust analysis. Sensitivity to minimal fault development is demonstrated, with the technique accurately identifying 0.007\" diameter inner race, outer race and roller element faults which had been seeded to the bearing through electro-discharge machining. To demonstrate the practicalities of the approach, the trained system is employed for analysis of an independent dataset, collected under different conditions. The technique is shown to accurately detect and identify the relevant fault mode pre-emptively, before catastrophic failure occurred, with 28.6% of bearing life remaining.","PeriodicalId":376942,"journal":{"name":"2014 International Conference on Prognostics and Health Management","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114726768","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 : 2014-06-22DOI: 10.1109/ICPHM.2014.7036398
Wooshik Kim, Jangbom Chai
We have developed a self-diagnostic monitoring system for an air operated valve system which produces arrow patterns according to the states of the system and makes a diagnosis whenever the system shows the corresponding symptom [1, 2]. In our first model, we have used a neural network and a simple comparison method for decision processor. In this paper, we modify and improve the decision processor module. We developed a logistic regression algorithm for the simple decision algorithm and modified the neural network algorithm. By changing the rule for translating arrow symbols into 2-D tuples, we could make unambiguous and rich training data set. With this, we performed some simulations and present a result.
{"title":"New algorithms for diagnosing defects of an air-operated valve for self diagnostic monitoring system","authors":"Wooshik Kim, Jangbom Chai","doi":"10.1109/ICPHM.2014.7036398","DOIUrl":"https://doi.org/10.1109/ICPHM.2014.7036398","url":null,"abstract":"We have developed a self-diagnostic monitoring system for an air operated valve system which produces arrow patterns according to the states of the system and makes a diagnosis whenever the system shows the corresponding symptom [1, 2]. In our first model, we have used a neural network and a simple comparison method for decision processor. In this paper, we modify and improve the decision processor module. We developed a logistic regression algorithm for the simple decision algorithm and modified the neural network algorithm. By changing the rule for translating arrow symbols into 2-D tuples, we could make unambiguous and rich training data set. With this, we performed some simulations and present a result.","PeriodicalId":376942,"journal":{"name":"2014 International Conference on Prognostics and Health Management","volume":"220 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120854572","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 : 2014-06-22DOI: 10.1109/ICPHM.2014.7036370
Qi Zhao, Wenfeng Zhang, Gan Zhou, XiuMei Guan
As requirements in diagnosis for hybrid systems increase, more and more researchers concentrate on hybrid models. However, common visual modeling methods such as GME (General Modeling Environment) lacks of flexibility. There is no appropriate modeling method for hybrid systems in cases containing plenty of complex components. This paper proposes a new hybrid hierarchy model description method, LLSM (Language for Large-Scale Modeling), based on concurrent probabilistic hybrid automata (cPHA) to make the process expediently. LLSM describes systems in the form of text. It settles the problem in three aspects: granularity, hierarchy and reusability. Component-oriented modeling of LLSM helps control granularity easily allowing users to create models in different scales. A special mark, which is employed to represent hierarchical relationship makes the system clearer and guides the accuracy of diagnosis. Reusability is achieved by C-style grammar which indicates component libraries for large-scale applications. In complex applications, LLSM creates models efficiently by existing libraries in the form of collaboration. Test on a switch demonstrates how it works.
随着对混合系统诊断要求的提高，越来越多的研究人员开始关注混合模型。然而，常见的可视化建模方法，如GME(通用建模环境)缺乏灵活性。对于包含大量复杂部件的混合动力系统，没有合适的建模方法。本文提出了一种新的基于并发概率混合自动机(cPHA)的混合层次模型描述方法——LLSM (Language for Large-Scale Modeling)。LLSM以文本的形式描述系统。它从粒度、层次和可重用性三个方面解决了这个问题。面向组件的LLSM建模有助于控制粒度，允许用户创建不同规模的模型。采用特殊的标记来表示层次关系，使系统更加清晰，并指导诊断的准确性。可重用性是通过c风格语法实现的，该语法为大规模应用程序指明了组件库。在复杂的应用程序中，LLSM以协作的形式通过现有的库有效地创建模型。在一个开关上测试它是如何工作的。
{"title":"A new hybrid hierarchy model description method","authors":"Qi Zhao, Wenfeng Zhang, Gan Zhou, XiuMei Guan","doi":"10.1109/ICPHM.2014.7036370","DOIUrl":"https://doi.org/10.1109/ICPHM.2014.7036370","url":null,"abstract":"As requirements in diagnosis for hybrid systems increase, more and more researchers concentrate on hybrid models. However, common visual modeling methods such as GME (General Modeling Environment) lacks of flexibility. There is no appropriate modeling method for hybrid systems in cases containing plenty of complex components. This paper proposes a new hybrid hierarchy model description method, LLSM (Language for Large-Scale Modeling), based on concurrent probabilistic hybrid automata (cPHA) to make the process expediently. LLSM describes systems in the form of text. It settles the problem in three aspects: granularity, hierarchy and reusability. Component-oriented modeling of LLSM helps control granularity easily allowing users to create models in different scales. A special mark, which is employed to represent hierarchical relationship makes the system clearer and guides the accuracy of diagnosis. Reusability is achieved by C-style grammar which indicates component libraries for large-scale applications. In complex applications, LLSM creates models efficiently by existing libraries in the form of collaboration. Test on a switch demonstrates how it works.","PeriodicalId":376942,"journal":{"name":"2014 International Conference on Prognostics and Health Management","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124682049","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 : 2014-06-22DOI: 10.1109/ICPHM.2014.7036373
G. Fagogenis, D. Flynn, D. Lane
This paper presents a novel, data-driven algorithm for the computation of the Remaining Useful Life (RUL) of an asset. The algorithm utilizes the asset's state history to learn a prognostic model from data. The prognostic model comprises an ensemble of Auto-Regressive (AR) models, together with a state-of-the-art classifier. The AR part of the algorithm is used to predict the system's state evolution. The classifier discriminates between healthy and faulty operation, given the asset's current state. The predicted state, as computed by the AR model, is fed to the classifier. The first time when the predicted state is classified as faulty is returned as the RUL of the system. The resulting prognostic algorithm was tested on the CMAPSS dataset as provided from NASA Ames Research Center. Cases of unknown future input trajectory as well as cases with multiple faults have been investigated.
{"title":"Novel RUL prediction of assets based on the integration of auto-regressive models and an RUSBoost classifier","authors":"G. Fagogenis, D. Flynn, D. Lane","doi":"10.1109/ICPHM.2014.7036373","DOIUrl":"https://doi.org/10.1109/ICPHM.2014.7036373","url":null,"abstract":"This paper presents a novel, data-driven algorithm for the computation of the Remaining Useful Life (RUL) of an asset. The algorithm utilizes the asset's state history to learn a prognostic model from data. The prognostic model comprises an ensemble of Auto-Regressive (AR) models, together with a state-of-the-art classifier. The AR part of the algorithm is used to predict the system's state evolution. The classifier discriminates between healthy and faulty operation, given the asset's current state. The predicted state, as computed by the AR model, is fed to the classifier. The first time when the predicted state is classified as faulty is returned as the RUL of the system. The resulting prognostic algorithm was tested on the CMAPSS dataset as provided from NASA Ames Research Center. Cases of unknown future input trajectory as well as cases with multiple faults have been investigated.","PeriodicalId":376942,"journal":{"name":"2014 International Conference on Prognostics and Health Management","volume":"48 17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129301213","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 : 2014-06-22DOI: 10.1109/ICPHM.2014.7036395
Limeng Guo, Yu Peng, Datong Liu, Yue Luo
Due to the high performance on state tracking and predicting, particle filter (PF) algorithm has been utilized for diagnosis and prognosis in a variety of areas. Especially, PF can provide uncertainty representation and management on estimating the remaining useful life (RUL) of components and systems. However, particle degeneracy phenomenon limits its performance and application in most of the situations. Therefore, several re-sampling algorithms are proposed to alleviate this problem. Thus, different re-sampling algorithms should be focused and studied for the adaptability and applicability in RUL estimation. This work aims to compare the capabilities of different re-sampling algorithms and evaluate the performance in lithium-ion battery RUL prediction. Four re-sampling algorithms including multinomial re-sampling, residual re-sampling stratified re-sampling and systematic re-sampling are involved and analyzed. Actual battery test data sets from NASA PCoE are used to conduct experiments for evaluation and comparison. Moreover, some quantitative analysis metrics are applied to compare the results of battery RUL estimation.
{"title":"Comparison of resampling algorithms for particle filter based remaining useful life estimation","authors":"Limeng Guo, Yu Peng, Datong Liu, Yue Luo","doi":"10.1109/ICPHM.2014.7036395","DOIUrl":"https://doi.org/10.1109/ICPHM.2014.7036395","url":null,"abstract":"Due to the high performance on state tracking and predicting, particle filter (PF) algorithm has been utilized for diagnosis and prognosis in a variety of areas. Especially, PF can provide uncertainty representation and management on estimating the remaining useful life (RUL) of components and systems. However, particle degeneracy phenomenon limits its performance and application in most of the situations. Therefore, several re-sampling algorithms are proposed to alleviate this problem. Thus, different re-sampling algorithms should be focused and studied for the adaptability and applicability in RUL estimation. This work aims to compare the capabilities of different re-sampling algorithms and evaluate the performance in lithium-ion battery RUL prediction. Four re-sampling algorithms including multinomial re-sampling, residual re-sampling stratified re-sampling and systematic re-sampling are involved and analyzed. Actual battery test data sets from NASA PCoE are used to conduct experiments for evaluation and comparison. Moreover, some quantitative analysis metrics are applied to compare the results of battery RUL estimation.","PeriodicalId":376942,"journal":{"name":"2014 International Conference on Prognostics and Health Management","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130271059","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 : 2014-06-22DOI: 10.1109/ICPHM.2014.7036383
K. Kumar, N. Arunachalam, L. Vijayaraghavan
In a machine tool, the cutting tool is mainly responsible for producing a component with good surface quality. With the time the cutting tool wear out and affects the surface quality. Hence it is very important to monitor the condition of the cutting tool to avoid the production of substandard parts. In this work the face milling cutter is made to interact with hardened steel components to manufacture the required surfaces with a specified amount of stock removal. The cutting conditions are selected and machining is done till the tool reaches its critical flank wear value. The captured surface images are analyzed using the statistical and spectral texture analysis methods. The flank wear of the cutting insert is measured at frequent intervals. The evaluated texture features are correlated with the flank wear using the multivariate correlation methods. The significant features are selected based on the correlation value and its mutual correlation value with other features. The selected texture features are plotted against machining time or the number of components. The developed regression model based on the selected parameters and the time is used to predict the flank wear.
{"title":"Prognostics model for tool life prediction in milling using texture features of surface image data","authors":"K. Kumar, N. Arunachalam, L. Vijayaraghavan","doi":"10.1109/ICPHM.2014.7036383","DOIUrl":"https://doi.org/10.1109/ICPHM.2014.7036383","url":null,"abstract":"In a machine tool, the cutting tool is mainly responsible for producing a component with good surface quality. With the time the cutting tool wear out and affects the surface quality. Hence it is very important to monitor the condition of the cutting tool to avoid the production of substandard parts. In this work the face milling cutter is made to interact with hardened steel components to manufacture the required surfaces with a specified amount of stock removal. The cutting conditions are selected and machining is done till the tool reaches its critical flank wear value. The captured surface images are analyzed using the statistical and spectral texture analysis methods. The flank wear of the cutting insert is measured at frequent intervals. The evaluated texture features are correlated with the flank wear using the multivariate correlation methods. The significant features are selected based on the correlation value and its mutual correlation value with other features. The selected texture features are plotted against machining time or the number of components. The developed regression model based on the selected parameters and the time is used to predict the flank wear.","PeriodicalId":376942,"journal":{"name":"2014 International Conference on Prognostics and Health Management","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114294286","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：