Electrolytic capacitor is widely selected as output capacitor in DC-DC converter, while it has limited reliability. Monitoring the equivalent series resistance (ESR) is an effective method to diagnostic the output capacitor. In this paper, an online monitoring scheme of ESR is present for discontinuous conduction mode (DCM) Buck converters. Based on the output ripple voltage, the model of ESR is established. By sampling the output ripple voltage, the ESR is calculated using the sampled values. The proposed method reduces the current measurement and avoids the change of converter topology. The simulation and experimental results verify the effectiveness of the method.
{"title":"An Online Monitoring Scheme of Output Capacitor’s ESR for DCM Buck","authors":"Xiaoxin Duan, Jian Zou, Dengyun Lei, B. Hou, Liwei Wang, Yun Huang","doi":"10.1109/phm-qingdao46334.2019.8942990","DOIUrl":"https://doi.org/10.1109/phm-qingdao46334.2019.8942990","url":null,"abstract":"Electrolytic capacitor is widely selected as output capacitor in DC-DC converter, while it has limited reliability. Monitoring the equivalent series resistance (ESR) is an effective method to diagnostic the output capacitor. In this paper, an online monitoring scheme of ESR is present for discontinuous conduction mode (DCM) Buck converters. Based on the output ripple voltage, the model of ESR is established. By sampling the output ripple voltage, the ESR is calculated using the sampled values. The proposed method reduces the current measurement and avoids the change of converter topology. The simulation and experimental results verify the effectiveness of the method.","PeriodicalId":259179,"journal":{"name":"2019 Prognostics and System Health Management Conference (PHM-Qingdao)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131744026","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 : 2019-10-01DOI: 10.1109/phm-qingdao46334.2019.8942843
Zhenteng Xu, Cheng Cheng, Yanjun Li
During the flight, the aircraft acquires the airspeed and altitude from the data collected by the pitot tube. Therefore, the static pressure source error of the pitot tube has a very large influence on the accuracy of the collected data. In order to correct the static source error, the static source error correction model was established based on Matlab & Simulink. Neural network and interpolation are used to build the error correction model. Firstly, the modified model collects the interface data of the atmospheric data computer (ADC), then it uses the neural network to make a preliminary forecast of the data, and displays the forecast results. Finally, the forecast results are modified by the cubic spline interpolation method, and the final modified results are output. This paper validates the model from both theory and practice, and proves that it can be used to correct the static source error.
{"title":"Static Source Error Correction Model Based on MATLAB and Simulink","authors":"Zhenteng Xu, Cheng Cheng, Yanjun Li","doi":"10.1109/phm-qingdao46334.2019.8942843","DOIUrl":"https://doi.org/10.1109/phm-qingdao46334.2019.8942843","url":null,"abstract":"During the flight, the aircraft acquires the airspeed and altitude from the data collected by the pitot tube. Therefore, the static pressure source error of the pitot tube has a very large influence on the accuracy of the collected data. In order to correct the static source error, the static source error correction model was established based on Matlab & Simulink. Neural network and interpolation are used to build the error correction model. Firstly, the modified model collects the interface data of the atmospheric data computer (ADC), then it uses the neural network to make a preliminary forecast of the data, and displays the forecast results. Finally, the forecast results are modified by the cubic spline interpolation method, and the final modified results are output. This paper validates the model from both theory and practice, and proves that it can be used to correct the static source error.","PeriodicalId":259179,"journal":{"name":"2019 Prognostics and System Health Management Conference (PHM-Qingdao)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131560178","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 : 2019-10-01DOI: 10.1109/phm-qingdao46334.2019.8942966
H. Chang
There are many ways to evaluate the performance of a five-axis machine tool, but an evaluation can be performed using a recognizable multi-type comparison, and it the most practical is the recognizable performance evaluation (RPE). The RPE is one of the current research methods that can derive accurate reference data in a quantitative and recognizable way and is one of the evaluation methods for multi-type five axis machine tool models. Therefore, based on the RPE and the interface of the IT level distribution in the general mechanical design change, this paper attempts to introduce fuzzy theory to obtain exceptional research results.This study calculates the attribution degree of the tested items. A direct discriminant defuzzification attribution degree drop interval is provided to manage the conflicts in the retested performance evaluation of various types of five-axis machine tools. It is possible to directly evaluate the predicted results. The experimental results show that the interval of the interval is 2σ. This result, for the quantifiable performance evaluation, further distinguishes the landing interval.
{"title":"Performance Evaluation of Multi-type Five-axis Machine Tool With Recognizable Performance Evaluation by Fuzzy Theory","authors":"H. Chang","doi":"10.1109/phm-qingdao46334.2019.8942966","DOIUrl":"https://doi.org/10.1109/phm-qingdao46334.2019.8942966","url":null,"abstract":"There are many ways to evaluate the performance of a five-axis machine tool, but an evaluation can be performed using a recognizable multi-type comparison, and it the most practical is the recognizable performance evaluation (RPE). The RPE is one of the current research methods that can derive accurate reference data in a quantitative and recognizable way and is one of the evaluation methods for multi-type five axis machine tool models. Therefore, based on the RPE and the interface of the IT level distribution in the general mechanical design change, this paper attempts to introduce fuzzy theory to obtain exceptional research results.This study calculates the attribution degree of the tested items. A direct discriminant defuzzification attribution degree drop interval is provided to manage the conflicts in the retested performance evaluation of various types of five-axis machine tools. It is possible to directly evaluate the predicted results. The experimental results show that the interval of the interval is 2σ. This result, for the quantifiable performance evaluation, further distinguishes the landing interval.","PeriodicalId":259179,"journal":{"name":"2019 Prognostics and System Health Management Conference (PHM-Qingdao)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129092908","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 : 2019-10-01DOI: 10.1109/phm-qingdao46334.2019.8942968
Hui Zeng, Zhinong Li, Zewen Zhou
The running time and the convergence between traditional parallel factor trilinear alternating least squares algorithm (TALS) algorithm and complex parallel factor (COMFAC) algorithm is compared by the experiment. The experiment result shows that both methods can obtain good separation performance. However, the traditional parallel factor separation algorithm has the higher complexity and the slower convergence. The complex parallel factor analysis can improve the convergence of the the traditional parallel factor analysis. The solution of complex parallel factor is usually very close to the least squares solution with only a few iterations.
{"title":"Comparative Study of Complex Parallel Factor Analysis and Parallel Factor Analysis","authors":"Hui Zeng, Zhinong Li, Zewen Zhou","doi":"10.1109/phm-qingdao46334.2019.8942968","DOIUrl":"https://doi.org/10.1109/phm-qingdao46334.2019.8942968","url":null,"abstract":"The running time and the convergence between traditional parallel factor trilinear alternating least squares algorithm (TALS) algorithm and complex parallel factor (COMFAC) algorithm is compared by the experiment. The experiment result shows that both methods can obtain good separation performance. However, the traditional parallel factor separation algorithm has the higher complexity and the slower convergence. The complex parallel factor analysis can improve the convergence of the the traditional parallel factor analysis. The solution of complex parallel factor is usually very close to the least squares solution with only a few iterations.","PeriodicalId":259179,"journal":{"name":"2019 Prognostics and System Health Management Conference (PHM-Qingdao)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134646265","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 : 2019-10-01DOI: 10.1109/phm-qingdao46334.2019.8942927
Xiaoyu Zhang, Lin Yue
Different from the operational modal analysis (OMA), the order based modal analysis (OBMA) is based on periodic sweep to obtain the dynamic behaviors of machinery. Therefore, its applicable condition is that the mechanical system generates periodic excitation force during the operational process. Due to the imbalance and misalignment of the rotating mechanical, it generates periodic excitation force whose frequency is proportional to the rotational speed during revolution. This paper utilizes OBMA to identify the resonances from the simulated signal with crossing order and white noise. First, the Vold-Kalman filter based order tracking (VK) method is utilized to extract harmonic response known as engine orders. Finally, the least-squares complex frequency-domain estimation method (PolyMAX) is applied to identify the resonance frequency, damping and modal shapes. Especially, two modes whose natural frequencies are close are successfully separated.
{"title":"Order Based Modal Analysis Using Vold-Kalman Filter","authors":"Xiaoyu Zhang, Lin Yue","doi":"10.1109/phm-qingdao46334.2019.8942927","DOIUrl":"https://doi.org/10.1109/phm-qingdao46334.2019.8942927","url":null,"abstract":"Different from the operational modal analysis (OMA), the order based modal analysis (OBMA) is based on periodic sweep to obtain the dynamic behaviors of machinery. Therefore, its applicable condition is that the mechanical system generates periodic excitation force during the operational process. Due to the imbalance and misalignment of the rotating mechanical, it generates periodic excitation force whose frequency is proportional to the rotational speed during revolution. This paper utilizes OBMA to identify the resonances from the simulated signal with crossing order and white noise. First, the Vold-Kalman filter based order tracking (VK) method is utilized to extract harmonic response known as engine orders. Finally, the least-squares complex frequency-domain estimation method (PolyMAX) is applied to identify the resonance frequency, damping and modal shapes. Especially, two modes whose natural frequencies are close are successfully separated.","PeriodicalId":259179,"journal":{"name":"2019 Prognostics and System Health Management Conference (PHM-Qingdao)","volume":"137 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131965270","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 : 2019-10-01DOI: 10.1109/phm-qingdao46334.2019.8942946
J. Mi, Xinyuan Wang, Yuhua Cheng, Songyi Zhang
Because of the measurement error and impact of other external factors, the experimentally measured fault information of rotary machinery equipment is with randomness and uncertainty. The diagnosis result gotten with uncertain information will not be accurate. Multi-source information fusion and fault identification based on cloud model and D-S evidence theory is studied in this paper. The rough set theory is used to screen and reduce the multiple fault attribute, then get the fewest fault features which also satisfy the diagnosis. The multi-source information are fused by the calculation of cloud parameters and evidence theory. At last, two kinds of rolling bearing fault databases from experiments are performed, and the diagnosis results have proved the validity and feasibility of the proposed method.
{"title":"Multi-Source Uncertain Information Fusion Method for Fault Diagnosis Based on Evidence Theory","authors":"J. Mi, Xinyuan Wang, Yuhua Cheng, Songyi Zhang","doi":"10.1109/phm-qingdao46334.2019.8942946","DOIUrl":"https://doi.org/10.1109/phm-qingdao46334.2019.8942946","url":null,"abstract":"Because of the measurement error and impact of other external factors, the experimentally measured fault information of rotary machinery equipment is with randomness and uncertainty. The diagnosis result gotten with uncertain information will not be accurate. Multi-source information fusion and fault identification based on cloud model and D-S evidence theory is studied in this paper. The rough set theory is used to screen and reduce the multiple fault attribute, then get the fewest fault features which also satisfy the diagnosis. The multi-source information are fused by the calculation of cloud parameters and evidence theory. At last, two kinds of rolling bearing fault databases from experiments are performed, and the diagnosis results have proved the validity and feasibility of the proposed method.","PeriodicalId":259179,"journal":{"name":"2019 Prognostics and System Health Management Conference (PHM-Qingdao)","volume":"101 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132640522","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}
According to the characteristics of the large transmission ratio of the helicopter main reducer in the future, the research on the characteristic distribution of the planetary gear train of the helicopter main reducer is carried out. Because of the characteristics that the heavy load condition in helicopter main reducer’s running and the changeable running state and complex and harsh climatic conditions and the increasing heat of the flow field in the reducer, by the way, the large transmission ratio increases the uncertainty of the planetary gear train operation analysis at the same time, the probability of causing a failure is greater; Especially the complex and variable structure of the helicopter which has large transmission ratio main reducer makes the analysis difficulty further. For ensuring the safety and enhancing the reliability of the helicopter, this paper make a comparative analysis of the time domain characteristics of the state signals under the multi-operating condition between the normal planetary gear and planetary gears with different degrees of failure, to explore the distribution regularity of the fault characteristics of the helicopter main reducer, so as to realize the research on fault diagnosis of helicopter large transmission ratio planetary gear.
{"title":"Fault Characteristics Analysis of Planetary Gear of Helicopter Main Reducer","authors":"Liang Cao, Yubin Xia, Yong Shen, Jinglin Wang, Tianmin Shan, Zeli Lin","doi":"10.1109/phm-qingdao46334.2019.8942930","DOIUrl":"https://doi.org/10.1109/phm-qingdao46334.2019.8942930","url":null,"abstract":"According to the characteristics of the large transmission ratio of the helicopter main reducer in the future, the research on the characteristic distribution of the planetary gear train of the helicopter main reducer is carried out. Because of the characteristics that the heavy load condition in helicopter main reducer’s running and the changeable running state and complex and harsh climatic conditions and the increasing heat of the flow field in the reducer, by the way, the large transmission ratio increases the uncertainty of the planetary gear train operation analysis at the same time, the probability of causing a failure is greater; Especially the complex and variable structure of the helicopter which has large transmission ratio main reducer makes the analysis difficulty further. For ensuring the safety and enhancing the reliability of the helicopter, this paper make a comparative analysis of the time domain characteristics of the state signals under the multi-operating condition between the normal planetary gear and planetary gears with different degrees of failure, to explore the distribution regularity of the fault characteristics of the helicopter main reducer, so as to realize the research on fault diagnosis of helicopter large transmission ratio planetary gear.","PeriodicalId":259179,"journal":{"name":"2019 Prognostics and System Health Management Conference (PHM-Qingdao)","volume":"263 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133928727","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 : 2019-10-01DOI: 10.1109/phm-qingdao46334.2019.8943006
Xiaochuan Li, D. Mba, Tianran Lin
In this work, a hybrid prognostic framework which interfaces a model-based prognostic method, namely particle filter, with a similarity-based prognostic method is proposed. The proposed framework consists of automatic determination of predication start point, sensor fusion, and prognostics steps that lead to accurate remaining useful life (RUL) estimations. This approach first applies the canonical variate analysis (CVA) approach for determining the prediction start time and constructing the prognostic health indicators (HIs). The similarity-based method is then employed together with the model-based particle filter (PF) algorithm to improve the predictive performance in terms of reducing the uncertainty of RUL and improving the prediction accuracy. The proposed framework can automatically construct HIs that are suitable for RUL prediction and offer higher prediction accuracy and lower uncertainty boundaries than traditional model-based PF methods. Our proposed approach is successfully applied on aircraft turbofan engines RUL prediction.
{"title":"A Similarity-based and Model-based Fusion Prognostics Framework for Remaining Useful Life Prediction","authors":"Xiaochuan Li, D. Mba, Tianran Lin","doi":"10.1109/phm-qingdao46334.2019.8943006","DOIUrl":"https://doi.org/10.1109/phm-qingdao46334.2019.8943006","url":null,"abstract":"In this work, a hybrid prognostic framework which interfaces a model-based prognostic method, namely particle filter, with a similarity-based prognostic method is proposed. The proposed framework consists of automatic determination of predication start point, sensor fusion, and prognostics steps that lead to accurate remaining useful life (RUL) estimations. This approach first applies the canonical variate analysis (CVA) approach for determining the prediction start time and constructing the prognostic health indicators (HIs). The similarity-based method is then employed together with the model-based particle filter (PF) algorithm to improve the predictive performance in terms of reducing the uncertainty of RUL and improving the prediction accuracy. The proposed framework can automatically construct HIs that are suitable for RUL prediction and offer higher prediction accuracy and lower uncertainty boundaries than traditional model-based PF methods. Our proposed approach is successfully applied on aircraft turbofan engines RUL prediction.","PeriodicalId":259179,"journal":{"name":"2019 Prognostics and System Health Management Conference (PHM-Qingdao)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121817283","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 : 2019-10-01DOI: 10.1109/phm-qingdao46334.2019.8942879
D. He, W. Guo, Mao He
Deep learning is the most attractive topic in the field of machine learning and relevant applications. Owing to the strong learning ability of the convolutional neural network (CNN), it integrates the feature extraction from raw data and classification as a complete learning process and makes the bearing fault diagnosis intelligent. In the published results, the inputs of the CNN may be the raw temporal waveform of vibration, its processed waveform or converted 2D images. In this paper, focusing on the diagnosis accuracy of rolling bearings, a comparative study is conducted among the inputs using the raw temporal waveform, the frequency spectrum, and the envelope spectrum of a vibration signal. First, an appropriate classification model based on the CNN is constructed. Then, experimental data from bearing with real damages are collected and then transformed and converted into some small gray pixel images for training and testing the CNN model. Finally, the classification accuracies using three signals are compared. The results indicate that the diagnosis performances using the above three signals are close when the trained CNN models are stable; among them the model using the frequency spectrum of the vibration signal is a little better than the models using the other two signals, which may be a reference for further investigating the deep learning used in the field of bearing diagnosis.
{"title":"Bearing Diagnosis Accuracy Comparison Using Convolutional Neural Network with Time/Frequency Domain Signals","authors":"D. He, W. Guo, Mao He","doi":"10.1109/phm-qingdao46334.2019.8942879","DOIUrl":"https://doi.org/10.1109/phm-qingdao46334.2019.8942879","url":null,"abstract":"Deep learning is the most attractive topic in the field of machine learning and relevant applications. Owing to the strong learning ability of the convolutional neural network (CNN), it integrates the feature extraction from raw data and classification as a complete learning process and makes the bearing fault diagnosis intelligent. In the published results, the inputs of the CNN may be the raw temporal waveform of vibration, its processed waveform or converted 2D images. In this paper, focusing on the diagnosis accuracy of rolling bearings, a comparative study is conducted among the inputs using the raw temporal waveform, the frequency spectrum, and the envelope spectrum of a vibration signal. First, an appropriate classification model based on the CNN is constructed. Then, experimental data from bearing with real damages are collected and then transformed and converted into some small gray pixel images for training and testing the CNN model. Finally, the classification accuracies using three signals are compared. The results indicate that the diagnosis performances using the above three signals are close when the trained CNN models are stable; among them the model using the frequency spectrum of the vibration signal is a little better than the models using the other two signals, which may be a reference for further investigating the deep learning used in the field of bearing diagnosis.","PeriodicalId":259179,"journal":{"name":"2019 Prognostics and System Health Management Conference (PHM-Qingdao)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121965145","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 : 2019-10-01DOI: 10.1109/phm-qingdao46334.2019.8943051
Kang Wu, S. Zhong, Xu-yun Fu, Changtsing Wei
In the process of aeroengine anomaly detection, there is always an unbalance distribution among the samples of gas path performance parameters, that is, the number of normal samples is much larger than the number of abnormal samples. In addition, this imbalance will worsen with time, which leads to the classifier paying too much attention to normal samples in the process of model training. Thus, the recognition rate of abnormal samples will reduce significantly. To solve the above problems, an adaptive decision threshold support vector machine (ADT-SVM) is proposed and applied to the anomaly detection of aeroengine. Firstly, this paper analyzes the influence of the unbalanced training data on the performance of the traditional classification model. Then the concept of decision threshold is introduced and introduced into support vector machine for anomaly detection. Finally, an adaptive method is proposed to calculate the decision threshold based on the equal expected number of samples, and the performance of the adaptive threshold and the traditional default threshold SVM is compared through experiments, which show that the adaptive threshold is effective in solving the problem of the classification performance degradation of unbalanced gas path performance parameters.
{"title":"An Aeroengine Gas Path Anomaly Detection Method in The Case of Sample Imbalance","authors":"Kang Wu, S. Zhong, Xu-yun Fu, Changtsing Wei","doi":"10.1109/phm-qingdao46334.2019.8943051","DOIUrl":"https://doi.org/10.1109/phm-qingdao46334.2019.8943051","url":null,"abstract":"In the process of aeroengine anomaly detection, there is always an unbalance distribution among the samples of gas path performance parameters, that is, the number of normal samples is much larger than the number of abnormal samples. In addition, this imbalance will worsen with time, which leads to the classifier paying too much attention to normal samples in the process of model training. Thus, the recognition rate of abnormal samples will reduce significantly. To solve the above problems, an adaptive decision threshold support vector machine (ADT-SVM) is proposed and applied to the anomaly detection of aeroengine. Firstly, this paper analyzes the influence of the unbalanced training data on the performance of the traditional classification model. Then the concept of decision threshold is introduced and introduced into support vector machine for anomaly detection. Finally, an adaptive method is proposed to calculate the decision threshold based on the equal expected number of samples, and the performance of the adaptive threshold and the traditional default threshold SVM is compared through experiments, which show that the adaptive threshold is effective in solving the problem of the classification performance degradation of unbalanced gas path performance parameters.","PeriodicalId":259179,"journal":{"name":"2019 Prognostics and System Health Management Conference (PHM-Qingdao)","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117293455","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：