Pub Date : 2021-10-15DOI: 10.1109/PHM-Nanjing52125.2021.9612779
Zhu Xiao, Zhe Cheng, Yuehao Li
As one of the important methods in the field of artificial intelligence, machine learning plays a crucial role in the promoting engineering applications and the academic research. In recent years, with the rapid development of the field of artificial intelligence, other fields using artificial intelligence as a means has also made great breakthroughs, such as fault diagnosis. The traditional fault diagnosis method is based on a variety of different signal acquisition, signal processing, signal analysis means for equipment fault diagnosis and detection, while the fault diagnosis method based on machine learning has made a great breakthrough in recent years, and plays an important role in the field of fault diagnosis. This paper first describes the basic concepts of machine learning and fault diagnosis, and then describes several common machine learning methods, and summarizes and analyzes the development status in recent years. Finally, the author puts forward some of his own views and summarizes.
{"title":"A Review of Fault Diagnosis Methods Based on Machine Learning Patterns","authors":"Zhu Xiao, Zhe Cheng, Yuehao Li","doi":"10.1109/PHM-Nanjing52125.2021.9612779","DOIUrl":"https://doi.org/10.1109/PHM-Nanjing52125.2021.9612779","url":null,"abstract":"As one of the important methods in the field of artificial intelligence, machine learning plays a crucial role in the promoting engineering applications and the academic research. In recent years, with the rapid development of the field of artificial intelligence, other fields using artificial intelligence as a means has also made great breakthroughs, such as fault diagnosis. The traditional fault diagnosis method is based on a variety of different signal acquisition, signal processing, signal analysis means for equipment fault diagnosis and detection, while the fault diagnosis method based on machine learning has made a great breakthrough in recent years, and plays an important role in the field of fault diagnosis. This paper first describes the basic concepts of machine learning and fault diagnosis, and then describes several common machine learning methods, and summarizes and analyzes the development status in recent years. Finally, the author puts forward some of his own views and summarizes.","PeriodicalId":436428,"journal":{"name":"2021 Global Reliability and Prognostics and Health Management (PHM-Nanjing)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115386548","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}
Aiming at realizing the abnormal detection for spacecraft components based on the monitoring during environmental testing, this paper proposes a novel strategy containing principal component analysis (PCA), one class support vector machine (OCSVM), and integrated learning. Firstly, product features are extracted from the raw data. Then, PCA is utilized to reduce the feature dimension and standardize the data. After that, sub-datasets are generated through resampling and utilized to train the individual OCSVM models. Finally, the decision results of these models are averaged to obtain the final classification results. A case study based on a thruster simulation dataset shows that the proposed strategy can obtain accurate detection results.
{"title":"The Abnormal Detection Strategy for Spacecraft Components with Multi-dimension Parameters","authors":"Shouwen Liu, Taichun Qin, Shouqing Huang, Yunfei Jia, Guangyuan Zheng, Wanning Yao, Baohui Wang","doi":"10.1109/PHM-Nanjing52125.2021.9612759","DOIUrl":"https://doi.org/10.1109/PHM-Nanjing52125.2021.9612759","url":null,"abstract":"Aiming at realizing the abnormal detection for spacecraft components based on the monitoring during environmental testing, this paper proposes a novel strategy containing principal component analysis (PCA), one class support vector machine (OCSVM), and integrated learning. Firstly, product features are extracted from the raw data. Then, PCA is utilized to reduce the feature dimension and standardize the data. After that, sub-datasets are generated through resampling and utilized to train the individual OCSVM models. Finally, the decision results of these models are averaged to obtain the final classification results. A case study based on a thruster simulation dataset shows that the proposed strategy can obtain accurate detection results.","PeriodicalId":436428,"journal":{"name":"2021 Global Reliability and Prognostics and Health Management (PHM-Nanjing)","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115400203","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}
Integrated system health management (ISHM) technology is the extension of fault diagnosis and fault tolerant control and also is an important research direction of satellite engineering in the future. The ISHM of satellite, which can support satellite self-sufficient, aims to improve the reliability, safety and economy. In this paper, the overall architecture of the ISHM of on-orbit satellite is firstly constructed, which is divided into several layers, including system-level layer, subsystem layer, reasoning and diagnosis layer and single machine layer. The functions and tasks of each layer are analyzed, and the differences between ISHM and traditional fault diagnosis methods are proposed. Secondly, according to the analysis of history fault data, the subsystems who are prone to fault are studied and the ISHM schemes for these subsystems are designed. Finally, key technologies for achieving ISHM of satellite, such as the use of advanced distributed sensor system, improvement of software and hardware on board, and the application of intelligent algorithms, are illustrated and discussed.
{"title":"Integrated System Health Management Research and Application on Satellite","authors":"Yufei Xu, Xianglong Kong, Wei Ma, Zhu Zhu, Jianqiao Zhang","doi":"10.1109/PHM-Nanjing52125.2021.9612928","DOIUrl":"https://doi.org/10.1109/PHM-Nanjing52125.2021.9612928","url":null,"abstract":"Integrated system health management (ISHM) technology is the extension of fault diagnosis and fault tolerant control and also is an important research direction of satellite engineering in the future. The ISHM of satellite, which can support satellite self-sufficient, aims to improve the reliability, safety and economy. In this paper, the overall architecture of the ISHM of on-orbit satellite is firstly constructed, which is divided into several layers, including system-level layer, subsystem layer, reasoning and diagnosis layer and single machine layer. The functions and tasks of each layer are analyzed, and the differences between ISHM and traditional fault diagnosis methods are proposed. Secondly, according to the analysis of history fault data, the subsystems who are prone to fault are studied and the ISHM schemes for these subsystems are designed. Finally, key technologies for achieving ISHM of satellite, such as the use of advanced distributed sensor system, improvement of software and hardware on board, and the application of intelligent algorithms, are illustrated and discussed.","PeriodicalId":436428,"journal":{"name":"2021 Global Reliability and Prognostics and Health Management (PHM-Nanjing)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115827045","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 : 2021-10-15DOI: 10.1109/PHM-Nanjing52125.2021.9612945
Yi-Lin Wang, Yuanxiang Li, Yuxuan Zhang, Yongsheng Yang, Lei Liu
The maintenance decision models based on Prognostic and Health Management (PHM) technology have significantly improved complex equipment submission reliability and economy. One of the essential techniques of PHM is predicting the remaining useful life (RUL) of the system or the system components. Compared to other RUL prediction methods, deep learning has become a research hotspot due to its automatic feature extraction capability, big data process efficiency, powerful representation of complex mappings, and “end-to-end” learning process. However, deep learning (DL) models are with high complexity, huge parameter quantity, and low interpretability, namely black box models. Lack of interpretability limits their application and development in “high-risk” fields such as aviation maintenance decision-making. To solve this problem, we propose a universal RUL interpretation method for DL named as RUL Shapley Additive explanation (RUSHAP). RUSHAP uses the input and output of the DL model to calculate the Shapley value and then obtain the interpretation from three different hierarchies, i.e., time level, feature level, and component level. With RUSHAP, it is possible to go from only knowing the RUL of the system to locating fault state points, observing the declining trend of sensor data, and evaluating the health status of subsystem, achieving partial white-boxing of the RUL prediction DL model. RUSHAP can also compare the advantages and disadvantages between different DL models, giving references for model debugging and ideas for model design.
{"title":"RUSHAP: A Unified approach to interpret Deep Learning model for Remaining Useful Life Estimation","authors":"Yi-Lin Wang, Yuanxiang Li, Yuxuan Zhang, Yongsheng Yang, Lei Liu","doi":"10.1109/PHM-Nanjing52125.2021.9612945","DOIUrl":"https://doi.org/10.1109/PHM-Nanjing52125.2021.9612945","url":null,"abstract":"The maintenance decision models based on Prognostic and Health Management (PHM) technology have significantly improved complex equipment submission reliability and economy. One of the essential techniques of PHM is predicting the remaining useful life (RUL) of the system or the system components. Compared to other RUL prediction methods, deep learning has become a research hotspot due to its automatic feature extraction capability, big data process efficiency, powerful representation of complex mappings, and “end-to-end” learning process. However, deep learning (DL) models are with high complexity, huge parameter quantity, and low interpretability, namely black box models. Lack of interpretability limits their application and development in “high-risk” fields such as aviation maintenance decision-making. To solve this problem, we propose a universal RUL interpretation method for DL named as RUL Shapley Additive explanation (RUSHAP). RUSHAP uses the input and output of the DL model to calculate the Shapley value and then obtain the interpretation from three different hierarchies, i.e., time level, feature level, and component level. With RUSHAP, it is possible to go from only knowing the RUL of the system to locating fault state points, observing the declining trend of sensor data, and evaluating the health status of subsystem, achieving partial white-boxing of the RUL prediction DL model. RUSHAP can also compare the advantages and disadvantages between different DL models, giving references for model debugging and ideas for model design.","PeriodicalId":436428,"journal":{"name":"2021 Global Reliability and Prognostics and Health Management (PHM-Nanjing)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124541575","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 : 2021-10-15DOI: 10.1109/PHM-Nanjing52125.2021.9612927
Ming Han, Han Liu
To solve the problem of low edge protection index in traditional enhancement algorithm, a single image super-resolution enhancement algorithm based on deep learning is proposed. The super-resolution feature of a single image is extracted by the sinusoidal two-dimensional transform function modulated by Gaussian function. The local Laplacian filter is used to preprocess the super-resolution of a single image, and the deep learning method is introduced to enhance the super-resolution of a single image. The experimental results show that the improved method has higher edge protection index, can effectively improve the enhancement accuracy, and has certain advantages.
{"title":"Research on single image super resolution enhancement algorithm based on deep learning","authors":"Ming Han, Han Liu","doi":"10.1109/PHM-Nanjing52125.2021.9612927","DOIUrl":"https://doi.org/10.1109/PHM-Nanjing52125.2021.9612927","url":null,"abstract":"To solve the problem of low edge protection index in traditional enhancement algorithm, a single image super-resolution enhancement algorithm based on deep learning is proposed. The super-resolution feature of a single image is extracted by the sinusoidal two-dimensional transform function modulated by Gaussian function. The local Laplacian filter is used to preprocess the super-resolution of a single image, and the deep learning method is introduced to enhance the super-resolution of a single image. The experimental results show that the improved method has higher edge protection index, can effectively improve the enhancement accuracy, and has certain advantages.","PeriodicalId":436428,"journal":{"name":"2021 Global Reliability and Prognostics and Health Management (PHM-Nanjing)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114436770","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}
The reliability and security of analog circuits are becoming increasingly significant. Fault diagnosis methods can identify fault classes of analog circuits, thus locating the fault components. However, the fault diagnosis methods based on multi-classification learning framework suffer from the problem of desirable classification effect in the case of lack of fault samples. To address these issues, a fault detection method based on multiple features for analog circuits is proposed in this paper. By learning only normal samples to obtain control limits, the proposed fault detection method can effectively determine the health states of analog circuits. First, features of the output signals of the circuit under test (CUT) in the time domain, frequency domain, and time-frequency domain are calculated to comprehensively reflect its states. In addition, the construction method of related similarity (RS) features is introduced to achieve feature enhancement, which further explores the essential information in the features. Then, to remove redundant features, the feature selection is adaptively performed by using the quantum particle swarm optimization (QPSO) algorithm, where the fitness is the improved Wilks statistic (IWS). Finally, the feature vectors are transmitted to the fault detection model based on kernel principal component analysis (KPCA) to identify the health states of CUT. The experimental results indicate that the proposed method exhibits excellent detection performance for analog circuits in the case of lack of fault samples.
{"title":"A Novel Fault Detection Method Based on Multiple Features for Analog Circuits","authors":"Tianyu Gao, Jingli Yang, Jianfeng Wang, Shouda Jiang","doi":"10.1109/PHM-Nanjing52125.2021.9612952","DOIUrl":"https://doi.org/10.1109/PHM-Nanjing52125.2021.9612952","url":null,"abstract":"The reliability and security of analog circuits are becoming increasingly significant. Fault diagnosis methods can identify fault classes of analog circuits, thus locating the fault components. However, the fault diagnosis methods based on multi-classification learning framework suffer from the problem of desirable classification effect in the case of lack of fault samples. To address these issues, a fault detection method based on multiple features for analog circuits is proposed in this paper. By learning only normal samples to obtain control limits, the proposed fault detection method can effectively determine the health states of analog circuits. First, features of the output signals of the circuit under test (CUT) in the time domain, frequency domain, and time-frequency domain are calculated to comprehensively reflect its states. In addition, the construction method of related similarity (RS) features is introduced to achieve feature enhancement, which further explores the essential information in the features. Then, to remove redundant features, the feature selection is adaptively performed by using the quantum particle swarm optimization (QPSO) algorithm, where the fitness is the improved Wilks statistic (IWS). Finally, the feature vectors are transmitted to the fault detection model based on kernel principal component analysis (KPCA) to identify the health states of CUT. The experimental results indicate that the proposed method exhibits excellent detection performance for analog circuits in the case of lack of fault samples.","PeriodicalId":436428,"journal":{"name":"2021 Global Reliability and Prognostics and Health Management (PHM-Nanjing)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114844367","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 : 2021-10-15DOI: 10.1109/PHM-Nanjing52125.2021.9613115
Peng Liu, G. Wang, Yi-xiong Zhang
When a non-renewing warranty period expires, the customer can know the age of the product surviving warranty period. In this paper, by making use of the surviving product’s age information, an improved post-warranty maintenance strategy is proposed. Based on renewal process theory, the product life cycle cost rate is derived explicitly. Numerical results show that the proposed post-warranty maintenance strategy can achieve a lower cost rate than the one without considering the surviving product’s age information.
{"title":"A Post-Warranty Maintenance Strategy Considering Product Age","authors":"Peng Liu, G. Wang, Yi-xiong Zhang","doi":"10.1109/PHM-Nanjing52125.2021.9613115","DOIUrl":"https://doi.org/10.1109/PHM-Nanjing52125.2021.9613115","url":null,"abstract":"When a non-renewing warranty period expires, the customer can know the age of the product surviving warranty period. In this paper, by making use of the surviving product’s age information, an improved post-warranty maintenance strategy is proposed. Based on renewal process theory, the product life cycle cost rate is derived explicitly. Numerical results show that the proposed post-warranty maintenance strategy can achieve a lower cost rate than the one without considering the surviving product’s age information.","PeriodicalId":436428,"journal":{"name":"2021 Global Reliability and Prognostics and Health Management (PHM-Nanjing)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114870290","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 : 2021-10-15DOI: 10.1109/PHM-Nanjing52125.2021.9612944
Yang Dong, Xiao DingYang, Jiayi Liang, Qin Zhou Lin
The role of aircraft power supply system on flight safety is irreplaceable, and the lack of system-level assessment makes it necessary to assess its health status. The paper puts forward the concept of failure mode degradation state value, unifies the research route of equipment health assessment based on failure and degradation, proposes a calculation method for the importance of model nodes, and establishes a hierarchical model of aircraft power supply system health assessment. According to the comprehensive evaluation characteristics of the multi-attribute indicators that are coupled to the status of each component of the power supply system, the cloud model, entropy weight method and TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) are used to comprehensively calculate the risk ranking of failure modes. And the proportion, the influence weight of each layer model node on the upper layer node is obtained, and the comprehensive risk evaluation is completed. Then, according to the degradation degree of the failure mode degradation state value, the entropy weight method is used to improve the weight of the failure mode degradation state value, and the effectiveness of the evaluation algorithm is evaluated and improved. Finally, the health status of each subsystem and the health of the equipment are displayed with a radar chart, evaluation result. It provides an effective implementation path for the health assessment of the aircraft power supply system.
飞机供电系统对飞行安全的作用是不可替代的，缺乏系统级评估，有必要对其健康状态进行评估。提出了失效模式退化状态值的概念，统一了基于失效退化的设备健康评估的研究路线，提出了模型节点重要性的计算方法，建立了飞机供电系统健康评估的分层模型。根据与供电系统各部件状态耦合的多属性指标的综合评价特点，采用云模型、熵权法和TOPSIS (Order Preference Technique for Similarity to an Ideal Solution)综合计算故障模式的风险等级。并得到各层模型节点对上层节点的影响权重的比例，完成综合风险评价。然后，根据失效模式退化状态值的退化程度，采用熵权法对失效模式退化状态值的权值进行改进，对评价算法的有效性进行评价和改进。最后以雷达图和评估结果显示各子系统的健康状态和设备的健康状况。为飞机供电系统健康评估提供了有效的实施路径。
{"title":"Research on Health Assessment System of Aircraft Power Supply System","authors":"Yang Dong, Xiao DingYang, Jiayi Liang, Qin Zhou Lin","doi":"10.1109/PHM-Nanjing52125.2021.9612944","DOIUrl":"https://doi.org/10.1109/PHM-Nanjing52125.2021.9612944","url":null,"abstract":"The role of aircraft power supply system on flight safety is irreplaceable, and the lack of system-level assessment makes it necessary to assess its health status. The paper puts forward the concept of failure mode degradation state value, unifies the research route of equipment health assessment based on failure and degradation, proposes a calculation method for the importance of model nodes, and establishes a hierarchical model of aircraft power supply system health assessment. According to the comprehensive evaluation characteristics of the multi-attribute indicators that are coupled to the status of each component of the power supply system, the cloud model, entropy weight method and TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) are used to comprehensively calculate the risk ranking of failure modes. And the proportion, the influence weight of each layer model node on the upper layer node is obtained, and the comprehensive risk evaluation is completed. Then, according to the degradation degree of the failure mode degradation state value, the entropy weight method is used to improve the weight of the failure mode degradation state value, and the effectiveness of the evaluation algorithm is evaluated and improved. Finally, the health status of each subsystem and the health of the equipment are displayed with a radar chart, evaluation result. It provides an effective implementation path for the health assessment of the aircraft power supply system.","PeriodicalId":436428,"journal":{"name":"2021 Global Reliability and Prognostics and Health Management (PHM-Nanjing)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123571259","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 : 2021-10-15DOI: 10.1109/PHM-Nanjing52125.2021.9613013
R. Jiang
Wear is one of major failure causes of cutting tools. Monitoring the wear process of a cutting tool and predicting its residual life have attracted wide attentions. A stochastic wear process model that relates the wear amount to cumulative cutting time is needed so as to make the inspection and replacement decisions of the cutting tool. The wear amount as a function of cutting time is often inverse S-shaped. That is, the wear rate curve is bathtub-shaped. The works that explicitly model inverse S-shaped wear processes are rare. This paper presents a two-stage approach for modeling this type of wear processes. The proposed approach divides the process into two stages with the inflection point of the wear curve as the boundary of stages. The task in the first stage is to collect data and the tasks in the second stage are to predict residual life and make inspection and replacement decisions. The stochastic wear process model obtained from the proposed approach is simple and realistic, and does not need many data. A real-world example is included to illustrate the simplicity and appropriateness of the proposed approach.
{"title":"A two-stage approach for modeling inverse S-shaped wear processes of cutting tools","authors":"R. Jiang","doi":"10.1109/PHM-Nanjing52125.2021.9613013","DOIUrl":"https://doi.org/10.1109/PHM-Nanjing52125.2021.9613013","url":null,"abstract":"Wear is one of major failure causes of cutting tools. Monitoring the wear process of a cutting tool and predicting its residual life have attracted wide attentions. A stochastic wear process model that relates the wear amount to cumulative cutting time is needed so as to make the inspection and replacement decisions of the cutting tool. The wear amount as a function of cutting time is often inverse S-shaped. That is, the wear rate curve is bathtub-shaped. The works that explicitly model inverse S-shaped wear processes are rare. This paper presents a two-stage approach for modeling this type of wear processes. The proposed approach divides the process into two stages with the inflection point of the wear curve as the boundary of stages. The task in the first stage is to collect data and the tasks in the second stage are to predict residual life and make inspection and replacement decisions. The stochastic wear process model obtained from the proposed approach is simple and realistic, and does not need many data. A real-world example is included to illustrate the simplicity and appropriateness of the proposed approach.","PeriodicalId":436428,"journal":{"name":"2021 Global Reliability and Prognostics and Health Management (PHM-Nanjing)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123654923","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 : 2021-10-15DOI: 10.1109/PHM-Nanjing52125.2021.9612777
Jing Huang, Rongxi Wang, Zhiyong Gao, Jianmin Gao, W. Deng, Zhen Wang
The monitoring variables of offshore wind turbines are widely distributed and coupled with each other, and the relationship of information transfer is fuzzy, which brings great challenges to the research of performance characterization and fault diagnosis of wind turbines. Therefore, an information transmission model for offshore wind turbines was proposes based on multi-variable coupling relationships. First, based on the DBSCAN unsupervised clustering method, the different states corresponding to the performance of the wind turbine are obtained. Second, the general symbolic parameters of the monitoring time series are determined, and the adaptive symbolic processing of the monitoring time series is performed. Finally, the transfer entropy of symbolic sequence pair is calculated and the information transmission model is established. By analyzing the change of information transmission between nodes, the performance is characterized, which provides a good model basis for fault traceability.
{"title":"An Information Transmission Model For Offshore Wind Turbines Based On Multi-variable Coupling Relationship","authors":"Jing Huang, Rongxi Wang, Zhiyong Gao, Jianmin Gao, W. Deng, Zhen Wang","doi":"10.1109/PHM-Nanjing52125.2021.9612777","DOIUrl":"https://doi.org/10.1109/PHM-Nanjing52125.2021.9612777","url":null,"abstract":"The monitoring variables of offshore wind turbines are widely distributed and coupled with each other, and the relationship of information transfer is fuzzy, which brings great challenges to the research of performance characterization and fault diagnosis of wind turbines. Therefore, an information transmission model for offshore wind turbines was proposes based on multi-variable coupling relationships. First, based on the DBSCAN unsupervised clustering method, the different states corresponding to the performance of the wind turbine are obtained. Second, the general symbolic parameters of the monitoring time series are determined, and the adaptive symbolic processing of the monitoring time series is performed. Finally, the transfer entropy of symbolic sequence pair is calculated and the information transmission model is established. By analyzing the change of information transmission between nodes, the performance is characterized, which provides a good model basis for fault traceability.","PeriodicalId":436428,"journal":{"name":"2021 Global Reliability and Prognostics and Health Management (PHM-Nanjing)","volume":"1094 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120870287","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：