Ozhan Gecgel, S. Ekwaro-Osire, J. Dias, Abdul Serwadda, Fisseha M. Alemayehu, Abraham Nispel
{"title":"Gearbox Fault Diagnostics Using Deep Learning with Simulated Data","authors":"Ozhan Gecgel, S. Ekwaro-Osire, J. Dias, Abdul Serwadda, Fisseha M. Alemayehu, Abraham Nispel","doi":"10.1109/ICPHM.2019.8819423","DOIUrl":"https://doi.org/10.1109/ICPHM.2019.8819423","url":null,"abstract":"Transmission components are prone to fatigue damage due to high and intermittent loading cycles, that cause premature failure of gearboxes. Recently, several vibration-based diagnostics approaches using Machine Learning (ML) and Deep Learning (DL) algorithms have been proposed to identify gearboxes faults. However, most of them rely on a large amount of training data collection from physical experiments, which is often associated with high costs. This paper offers an ML and DL classification performance comparison of several algorithms to diagnose faults in a gearbox based on realistic simulated vibration data. A dynamic model of a single-stage gearbox was developed to generate data for different health conditions. Generated datasets were fed to ML and DL algorithms and accuracy results were compared. Results revealed the superiority of Convolutional Neural Network compared to other classifiers. This research contributes to the prevention of catastrophic failures in gearboxes by early crack detection and maintenance schedule optimization.","PeriodicalId":113460,"journal":{"name":"2019 IEEE International Conference on Prognostics and Health Management (ICPHM)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130653934","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}
{"title":"Research on the Propagation of Defects in Assembly Process Based on SIR Epidemic Model","authors":"Mengyao Wu, W. Dai, Yu Zhao","doi":"10.1109/ICPHM.2019.8819418","DOIUrl":"https://doi.org/10.1109/ICPHM.2019.8819418","url":null,"abstract":"In the manufacturing field, the final product quality and cost are heavily determined by assembly process. As the evolution law of defect steams in assembly process has similarities with the disease propagation in SIR model, SIR model was applied to assembly process. Firstly, we give a brief introduction to the SIR model and analyze its applicability. Then the defect model of assembly process was established combined with SIR epidemic model. The extinction and persistence of defects when the basic reproduction number R0 ≤1 and R0 >1 were discussed. Next, we proofed the conclusion with theoretical derivation and simulation experiment. Finally, the disadvantages of exist model were discussed and the direction of future work was put forward.","PeriodicalId":113460,"journal":{"name":"2019 IEEE International Conference on Prognostics and Health Management (ICPHM)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132914705","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}
Saikath Bhattacharya, L. Fiondella, Saurabh Saxena, M. Pecht
{"title":"Quantifying the Impact of Prognostic Distance on Average Cost per Cycle","authors":"Saikath Bhattacharya, L. Fiondella, Saurabh Saxena, M. Pecht","doi":"10.1109/ICPHM.2019.8819408","DOIUrl":"https://doi.org/10.1109/ICPHM.2019.8819408","url":null,"abstract":"Prognostics and health management (PHM) is transforming reliability engineering with methods to enhance safety by accurately estimating end of useful life, thereby recommending maintenance of critical components and systems to manage cost. Previous studies emphasized degradation modeling and algorithms to improve state of health predictions. However, most of these techniques focused on improving the accuracy of predictions within a single maintenance interval, while fewer studies considered the effectiveness of alternative degradation models over multiple successive maintenance intervals. This paper develops a measure based on concepts from maintenance theory to provide a framework to objectively compare the effectiveness of existing and future battery degradation models over multiple maintenance intervals. The approach quantifies the impact of prognostic distance on average cost per cycle during the lifetime of a system. The approach is applied to state of health prediction for lithium-ion batteries, which are widely used in various mission-critical systems. The results indicate that the approach can be used to select a prognostic distance that minimizes average cost. The approach can thus evaluate models to select a suitable prognostic distance.","PeriodicalId":113460,"journal":{"name":"2019 IEEE International Conference on Prognostics and Health Management (ICPHM)","volume":"40 17","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114046958","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}
{"title":"Unsupervised Anomaly Detection Using Variational Auto-Encoder based Feature Extraction","authors":"Rong Yao, Chongdang Liu, Linxuan Zhang, Peng Peng","doi":"10.1109/ICPHM.2019.8819434","DOIUrl":"https://doi.org/10.1109/ICPHM.2019.8819434","url":null,"abstract":"Anomaly detection is a key task in Prognostics and Health Management (PHM) system. Specially, in most practical applications, the lack of labels often exists which makes the unsupervised anomaly detection very meaningful. Furthermore, unsupervised anomaly detection is also considered as a challenging task due to the diversity and information-lack of data. Variational Auto-Encoder (VAE) is a stochastic generative model which is designed to reconstruct input data as close as possible. In this paper, VAE is applied to extract valuable features for the unsupervised anomaly detection tasks. Comparison experiments are conducted on KDD CUP 99 dataset and MNIST dataset. Results show that features obtained by VAE can make unsupervised anomaly detection approaches perform better. Auto-Encoder (AE) and Kernel Principle Component Analysis (KPCA) were applied as comparisons. The result demonstrates that VAE gets best performance among them.","PeriodicalId":113460,"journal":{"name":"2019 IEEE International Conference on Prognostics and Health Management (ICPHM)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132698105","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}
{"title":"Diagnosis Method for Hydro-generator Rotor Fault Based on Stochastic Resonance","authors":"Junqing Li, Luo Wang, Yonggang Li","doi":"10.1109/ICPHM.2019.8819379","DOIUrl":"https://doi.org/10.1109/ICPHM.2019.8819379","url":null,"abstract":"The rotor of hydro-generator is in the state of rotary vibration. Rotor faults is a common fault in hydrogenerators. The fault is not easy to detect in the early stage, but with the development of the fault, it will pose a threat to the safe operation of hydro-generator. Many faults will change the vibration of generator rotor. In order to detect small fault signals, time frequency compression stochastic resonance method (FCSR) is proposed. This method uses vibration noise to enhance weak fault signal characteristics. The frequency range of stochastic resonance can be improved by the time-frequency compression algorithm. The algorithm eliminates the limitation of the system on the measurement signal frequency and extends the stochastic resonance system to the whole frequency band. In addition, according to the rotor vibration of the hydro-generator, the range of the relevant parameters of the method is improved. The stochastic resonance method is used to reduce the noise of hydrogenerator rotor vibration signal and improve the signal-to-noise ratio of the signal. This is conducive to the extraction of rotor fault feature vectors. The results show that the method can accurately identify the abnormal vibration of hydro-generator and has high rotor early fault diagnosis accuracy.","PeriodicalId":113460,"journal":{"name":"2019 IEEE International Conference on Prognostics and Health Management (ICPHM)","volume":"173 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134040137","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}
{"title":"A Novel Bearing Health Indicator Construction Method Based on Ensemble Stacked Autoencoder","authors":"Pengfei Lin, Jizhong Tao","doi":"10.1109/ICPHM.2019.8819405","DOIUrl":"https://doi.org/10.1109/ICPHM.2019.8819405","url":null,"abstract":"In the area of data-driven bearing prognostic, the construction of health indictor from condition monitoring data is important. This paper presents a novel bearing health indicator construction method based on ensemble stacked autoencoders. Firstly, the proposed ensemble stacked autoencoders extract features directly from the FFT results of raw vibration signals. Then, a deep neural network which serves as a non-linear transformation is trained to map the multi-dimensional learned features to a one-dimensional health indicator. Finally, the proposed method is validated using the IEEE PHM2012 Challenge dataset. To show the superiority of the proposed method, its performance is evaluated and compared with other methods. The results demonstrate that the proposed method can automatically and effectively build high-quality health indictor from raw data without any signal processing and manual feature engineering.","PeriodicalId":113460,"journal":{"name":"2019 IEEE International Conference on Prognostics and Health Management (ICPHM)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124983388","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}
{"title":"A Hybrid Deep Learning Based Approach for Remaining Useful Life Estimation","authors":"Khaled Akkad, D. He","doi":"10.1109/ICPHM.2019.8819435","DOIUrl":"https://doi.org/10.1109/ICPHM.2019.8819435","url":null,"abstract":"One of the most important aspects of PHM is remaining useful life (RUL) estimation. This paper proposes a hybrid deep learning-based approach for RUL estimation. The hybrid method is developed using a combination of long short-term memory and convolutional neural networks. The effectiveness of the hybrid method is validated using three engine fleets from turbofan engines simulation datasets.","PeriodicalId":113460,"journal":{"name":"2019 IEEE International Conference on Prognostics and Health Management (ICPHM)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131920970","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}
{"title":"Electronic Circuit Diagnosis with No Data","authors":"Varun Khemani, M. Azarian, M. Pecht","doi":"10.1109/ICPHM.2019.8819424","DOIUrl":"https://doi.org/10.1109/ICPHM.2019.8819424","url":null,"abstract":"Operational data from the target system is widely considered a pre-requisite for implementation of PHM, as it used as training data. Often this data is not available to PHM practitioners because health monitoring capabilities may not be installed in legacy systems. This research presents an approach in which fault diagnosis can be implemented without any operational data and is generic enough to be applied to any electronic circuit provided a simulation model of the system with acceptable fidelity can be developed. The research also employs the Space-Filling Design, which can be used to generate the training data in a systematic, statistically valid framework, and is especially valuable for complex circuit with a large number of components. This design provides sufficient coverage of the parametric design space to be representative of the unavailable operational data, as well as incorporating the effects of parameter interaction on the simulated response of the system. Most PHM studies in the literature ignore the effect of the degradation of interacting components. We show, how such an assumption can lead to incorrect fault diagnosis/RUL estimation and propose methods to screen for two-way and higher order interactions. Finally, we use various deep learning approaches to diagnose circuit faults. This simulation-based fusion approach is a holistic framework for all types of analog electronic circuits.","PeriodicalId":113460,"journal":{"name":"2019 IEEE International Conference on Prognostics and Health Management (ICPHM)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115060364","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}
{"title":"Early gear tooth crack detection based on singular value decomposition","authors":"Yuejian Chen, M. Zuo","doi":"10.1109/ICPHM.2019.8819417","DOIUrl":"https://doi.org/10.1109/ICPHM.2019.8819417","url":null,"abstract":"Detection of gear tooth crack fault through vibration analysis relies on extracting the fault induced periodic impulses. Singular value decomposition (SVD)-based methods have been used for periodic impulse extraction. Reported reweighted SVD-based method did not consider interferences from non-fault related vibration components on the periodic modulation intensity (PMI) criteria, leading to the selection of incorrect signal component(s) for reconstruction. This paper proposes an improved SVD-based method by adopting autoregression model-based baseline removal approach. SVD is applied to decompose the residual signal, instead of the raw signal. The interferences from non-fault related vibration components on the PMI are therefore eliminated. Simulation study has shown that the improved method outperforms the reported method in detecting early tooth crack fault.","PeriodicalId":113460,"journal":{"name":"2019 IEEE International Conference on Prognostics and Health Management (ICPHM)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116949231","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}
{"title":"Generating Real-valued Failure Data for Prognostics Under the Conditions of Limited Data Availability","authors":"G. Ranasinghe, A. Parlikad","doi":"10.1109/ICPHM.2019.8819392","DOIUrl":"https://doi.org/10.1109/ICPHM.2019.8819392","url":null,"abstract":"Data-driven prognostics solutions underperform under the conditions of limited failure data availability since the number of failure data samples is insufficient for training prognostics models effectively. In order to address this problem, we present a novel methodology for generating real-valued failure data which allows training datasets to be augmented so that the number of failure data samples is increased. In contrast to existing data generation techniques which duplicate or randomly generate data, the proposed methodology is capable of generating new and realistic failure data samples. To this end, we utilised the conditional generative adversarial network and auxiliary information pertaining to the failure modes. The proposed methodology is evaluated in a real-world case study involving the prediction of air purge valve failures in heavy trucks. Two prognostics models are developed using gradient boosting machine and random forest classifiers. It is shown that when these models are trained on the augmented training dataset, they outperform the best prognostics solution previously proposed in the literature for the case study by a large margin. More specifically, costs due to breakdowns and false alarms are reduced by 44%.","PeriodicalId":113460,"journal":{"name":"2019 IEEE International Conference on Prognostics and Health Management (ICPHM)","volume":"89 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129273468","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}