2022 International Conference on Sensing, Measurement & Data Analytics in the era of Artificial Intelligence (ICSMD)最新文献

{"title":"An Auto-adjusting Weight Model for Imbalanced Wafer Defects Recognition","authors":"Yu Chen, Xinjia Zhao, Meng Zhao, Meng Zhao, J. Ji","doi":"10.1109/ICSMD57530.2022.10058248","DOIUrl":"https://doi.org/10.1109/ICSMD57530.2022.10058248","url":null,"abstract":"Wafer defect recognition is popular research in the semiconductor industry. Generally, each defect pattern is related to a specific manufacturing problem. By identifying defect patterns correctly, manufacturing problems can be recognized and fixed in time, which improves the quality and production yield of wafers. However, due to the location, light and the increasing number of wafers, traditional recognition methods achieve unsatisfactory performance. Currently, convolutional neural network (CNN) based methods outperform traditional methods in accuracy and speed, but fail when training with imbalanced target classes. To address the imbalanced problem, a CNN-based knowledge distillation (KD) method is proposed. To improve the identification of different types of defects, a multi-head attention layer is applied to the proposed CNN model, which enriches local and global information of features. Besides, when training the CNN model, target features are constrained with Distillation Loss and Focal Loss, reducing the effect of dataset imbalance. Experiments on the public dataset WM-811K are conducted to verify the proposed methods, and experimental results showed that the accuracy, precision, recall, specificity, and F1 score of our method reached 97.7%, 96.9%, 97.2%, 99.7% and 97.0% respectively, and the classification accuracy of each class was above 93.0%, which indicates the proposed method was reasonable and effective on large-scale imbalanced wafer defect datasets.","PeriodicalId":396735,"journal":{"name":"2022 International Conference on Sensing, Measurement & Data Analytics in the era of Artificial Intelligence (ICSMD)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116862147","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":"Characteristic Analysis of the Stress Wave of Silicon MOSFET under Gate-Source Overvoltage Failure","authors":"Guangxin Wang, Yunze He, Xuefeng Geng, Longhai Tang, Songyuan Liu, Qiying Li","doi":"10.1109/ICSMD57530.2022.10058330","DOIUrl":"https://doi.org/10.1109/ICSMD57530.2022.10058330","url":null,"abstract":"Power metal-oxide-semiconductor field-effect transistors (MOSFETs) are the core components of power electronic systems, and it is of great significance to ensure their safe and reliable operation. As a real-time, online and non-invasive monitoring method, acoustic emission (AE) monitoring technology has a good application prospect in the condition monitoring and fault diagnosis of power MOSFETs. The stress wave will be generated when the power MOSFET is turned on and off. However, in the majority of the present research, stress waves are only detected and analyzed for normal devices, and no correlation between the characteristics of stress waves and specific failures inside the device has yet been established. As a result, the gate-source overvoltage failure experiment was conducted. The MOSFET's stress wave under different gate-source voltages was acquired. Besides, the stress wave that occurred when the chip failed was also recorded. Time domain analysis and wavelet analysis were performed on the stress wave signal, and the conclusion can be drawn that the peak-to-peak value in the time domain, the signal energy and the wavelet peak value of the stress wave at the time of failure are significantly different from those in the normal condition. This work aims to lay the foundation for establishing the correlation between the characteristics of stress waves and device failures.","PeriodicalId":396735,"journal":{"name":"2022 International Conference on Sensing, Measurement & Data Analytics in the era of Artificial Intelligence (ICSMD)","volume":"74 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114271341","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 domain adaptation for bearing fault diagnosis using nonlinear impact dynamics model under limited supervision","authors":"Wenzhen Xie, Te Han, Haidong Shao","doi":"10.1109/ICSMD57530.2022.10058222","DOIUrl":"https://doi.org/10.1109/ICSMD57530.2022.10058222","url":null,"abstract":"Rolling bearing is one of the crucial rotating parts of mechanical systems, which is usually exposed to high-load working conditions. The diagnosis of rolling bearing faults is significant for the health monitoring of the whole mechanical system. The deep learning method has been proven to be effective in many fault diagnosis occasions. However, sufficient labeled fault samples are unavailable in some practical industrial diagnosis tasks, which will lead to the serious performance degradation of traditional deep learning methods. Therefore, a rolling bearing dynamics model is established for generating sufficient simulation data for assisting the training process. Furthermore, to overcome the diagnostic performance degradation problem caused by the inconsistent feature distribution of simulation data and experimental data, adversarial learning is conducted to realize domain adaptation, thus capturing the generalized feature representation. The analysis results of an experimental rolling bearing dataset demonstrate the effectiveness of the proposed model, showing a potential industrial application value.","PeriodicalId":396735,"journal":{"name":"2022 International Conference on Sensing, Measurement & Data Analytics in the era of Artificial Intelligence (ICSMD)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127324541","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":"Optimized Acceleration of Single Shot Detection for Edge Computing Based-on FPGA","authors":"R. Wu, P. Fu, Lei Feng, Shan Sun, Shuyan Wang, Bing Liu","doi":"10.1109/ICSMD57530.2022.10058241","DOIUrl":"https://doi.org/10.1109/ICSMD57530.2022.10058241","url":null,"abstract":"Single Shot Detection (SSD) based Deep Neural Network (DNN) has been widely used in edge computing due to the properties of one-stage detection and high accuracy. Especially in Field Programmable Gate Array (FPGA) based acceleration, SSD can complete read-time object detection with higher efficiency. However, the acceleration of SSD has been suffered from constraints of system integration, size and power in strict scenarios. It requires all network inference to be done within single chip, as well as the acceleration of post-processing algorithm in SSD. To realize SSD acceleration in single chip, this paper proposes an optimized acceleration method of SSD for autonomous body detection scenario. The original post-processing algorithm is optimized through judging the probability threshold, which reduces the redundant computing operations of location box without losing detection accuracy. Meanwhile, an acceleration architecture is constructed to satisfy hardware constrained platform. In the experimental results, the optimized execution time is changed from 46.024ms to 9.277ms in software, and the accelerated time is further reduced to 1.117ms in hardware, which achieves performance improvement in 7.305 times. The proposed method can also be applied to other applications for the acceleration of post-processing algorithm.","PeriodicalId":396735,"journal":{"name":"2022 International Conference on Sensing, Measurement & Data Analytics in the era of Artificial Intelligence (ICSMD)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126908030","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":"Remaining Useful Life Prediction of Aero-Engine Based on Transformer with Tendency Retainment","authors":"Zhi Zhai, Jingcheng Wen, Fujin Wang, Zhibin Zhao, Yanjie Guo, Xuefeng Chen","doi":"10.1109/ICSMD57530.2022.10058242","DOIUrl":"https://doi.org/10.1109/ICSMD57530.2022.10058242","url":null,"abstract":"One of the essential technologies for prognostics and health management of aero-engines is remaining useful life (RUL) prediction. Many deep learning models have recently been presented to extract features adaptively and forecast RUL end-to-end. However, it is still a challenging task to model data of long-life cycles and retain the degradation information when extracting features. To overcome the problem, we present a Transformer-based method with tendency retainment to predict RUL. Convolutional neural network is first used to fuse data from different sensors. Then, the long-life cycle data is encoded by Transformer encoder followed by long short-term memory neural network to extract features and finally RUL is predicted. Moreover, a tendency retainment module is designed based on contrastive learning to maintain the degradation information. The proposed method's performance is validated using NASA's C-MAPSS aero-engine dataset. The experimental results reveal that the proposed method outperforms other state-of-the-art methods in terms of prediction accuracy.","PeriodicalId":396735,"journal":{"name":"2022 International Conference on Sensing, Measurement & Data Analytics in the era of Artificial Intelligence (ICSMD)","volume":"29 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124277385","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":"Effect of Pretreatment on Electrical Breakdown Performance of Insulation Oil in Submarine Oil-Filled Cable","authors":"Qingshuai Wu, Jie Deng, Weijia Zhang, Xiaowei Huang, Yang Rong, Junguo Gao, Lili Li","doi":"10.1109/ICSMD57530.2022.10058308","DOIUrl":"https://doi.org/10.1109/ICSMD57530.2022.10058308","url":null,"abstract":"For submarine oil-filled cable, the micro-water content and dielectric strength of insulation oil are important parameters to detect properties. In this paper, to explore variation trends of micro-water content and dielectric strength, the insulation oil of submarine oil-filled cable is pretreated in different ways by the oven, and the micro-water content and the breakdown performance of insulation oil after pretreatment is detected by Karl Fischer method and electrical breakdown theory. The experimental results showed that after comprehensive considering the time cost and electrical breakdown performance of pretreatment, when the pretreatment condition was 120 °C and 12 h, the micro-water content of insulation oil decreased by 65.73 % and the dielectric strength of insulation oil increased by 22.92 %. And then when the pretreatment temperature and time of insulation oil were different from the pre-treatment condition of 120 °C and 12h, micro-water content of insulation oil increased, or the dielectric strength decreased.","PeriodicalId":396735,"journal":{"name":"2022 International Conference on Sensing, Measurement & Data Analytics in the era of Artificial Intelligence (ICSMD)","volume":"150 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122631951","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 muti-dimension normalization layers improved ResNet based fault diagnosis method of rolling bearing","authors":"Changbo He, Yujie Cao, Yang Yang, Yongbin Liu, Xianzeng Liu, Zheng Cao","doi":"10.1109/ICSMD57530.2022.10058457","DOIUrl":"https://doi.org/10.1109/ICSMD57530.2022.10058457","url":null,"abstract":"CNN, a kind of deep learning method, has been widely used in fault diagnosis. It requires a large number of training samples, but it is difficult to obtain abundant samples under different conditions. Aiming at insufficient fault samples, an improved ResNet (IResNet) is proposed in this paper. Firstly, order spectrum is computed from raw data as pre-processed samples, which will be further augmented to improve the generalization ability of the model. Secondly, IResNet is constructed by several hybrid residual building blocks fused from multi-dimensional normalization layers, which can be adopted to enhance the feature extraction ability of the model. Then, the parameters of IResNet in the source domain are transferred to identify the health status of rolling bearing in the target domain. Finally, experimental data under different working conditions are used to verify the performance of the proposed method. The experimental results indicate that the recognition accuracy of the proposed method is higher than other methods and that the proposed method can identify the health status of rolling bearing with small training samples.","PeriodicalId":396735,"journal":{"name":"2022 International Conference on Sensing, Measurement & Data Analytics in the era of Artificial Intelligence (ICSMD)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121746353","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":"Vibration Transfer Analysis of Unbalance-Rubbing-Bearing Fault of Centrifugal Pump Considering Measuring Point of Machine Foot","authors":"Taiwei Zhang, Kaixuan Lv, K. Feng","doi":"10.1109/ICSMD57530.2022.10058320","DOIUrl":"https://doi.org/10.1109/ICSMD57530.2022.10058320","url":null,"abstract":"Marine centrifugal pump is usually installed vertically in order to save equipment space. The sensor which can obtain the vibration status of centrifugal pump in real time can't be installed directly in the pump body, but can only be installed in the foot of the machine, resulting in the real-time status monitoring and fault diagnosis of centrifugal pump difficult to carry out. In order to solve this problem, the vibration transmission of the rotor system of centrifugal pump was analyzed, and the response law of vibration signals of the foot of centrifugal pump to different faults was proposed in this paper. According to the internal structural constraints of the centrifugal pump, a dynamic model containing rotor unbalance-rubbing-bearing fault was established, which took into account the unbalance fault, rotor-static rubbing fault, clearance of rolling bearings, nonlinear Hertzian contact between ball and raceway, the VC vibration caused by the change of rolling bearing support stiffness force and the damage dynamic model of the rolling bearing outer ring. At the same time, the transfer path of pump body and foot is introduced. The numerical integration method is used to carry out dynamic simulation analysis, and the results show that the dynamic model of centrifugal pump unbalance-rubbing-bearing fault proposed in this paper is correct and effective. The vibration response of measuring points of pump body and feet to different faults is analyzed by simulation data, and verified by experimental data. The analysis results show that the unbalance fault will lead to the increase of the power frequency amplitude of the vibration response of the pump body and the machine foot of the centrifugal pump. The rubbing fault will cause the 2nd, 4th, and 6th frequency amplitudes of the power frequency to be prominent in the vibration response spectrum of the pump body and the machine foot. When there is a bearing fault, the signals of the centrifugal pump foot and the pump body can be enveloped demodulated to extract the characteristic frequency of the bearing fault.","PeriodicalId":396735,"journal":{"name":"2022 International Conference on Sensing, Measurement & Data Analytics in the era of Artificial Intelligence (ICSMD)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125658371","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":"Aero-engine Fan Acoustic Mode Detections via Orthogonal Matching Pursuit","authors":"Boyu Ma, Yanan Wang, Baijie Qiao, Bi Wen, Zepeng Li, Xuefeng Chen","doi":"10.1109/ICSMD57530.2022.10058329","DOIUrl":"https://doi.org/10.1109/ICSMD57530.2022.10058329","url":null,"abstract":"Azimuthal mode analysis (AMA) is one of the most commonly used approaches for comprehending the characteristics of the noise emitted from the aero-engine fans. This paper proposed a new azimuthal mode detection method based on compressive sensing, which breaks through the limitations of the Shannon-Nyquist sampling theorem and extends the range of mode detection. A $ell_{0}$ -norm regularized AMA method is proposed to reconstruct the spectrum of the tonal modes of aero-engine fans. Notably, the orthogonal matching pursuit (OMP) algorithm is implemented to effectively ameliorate the solution of the $ell_{0}$ -norm regularized problem. The feasibility of the proposed approach is verified by a series of simulations, of which the configurations are consistent with a practical case. Meanwhile, the performance of the $ell_{1}$ -norm regularized AMA method is compared with the proposed approach. The simulation results indicated that the $ell_{0}$ -norm regularized approach enhanced the sparsity of the estimations of the tonal noise mode spectrum. The stability and the robustness of the reconstruction results are notably improved, which leads to a higher accuracy of the amplitudes of the tonal acoustic modes and a noticeable reduction of the number of the microphones required by AMA.","PeriodicalId":396735,"journal":{"name":"2022 International Conference on Sensing, Measurement & Data Analytics in the era of Artificial Intelligence (ICSMD)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125659187","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":"Defect Detection Method of Wind Turbine Blades Based on Improved YOLOv4","authors":"Wenxiang Chen, Guoqiang Zhu, Ming Mao, Xuelei Xi, Weiqi Xiong, Lu Liu, Shuang Wang, Yu Chen","doi":"10.1109/ICSMD57530.2022.10058380","DOIUrl":"https://doi.org/10.1109/ICSMD57530.2022.10058380","url":null,"abstract":"A lightweight defect detection model for wind turbine blades is needed to meet the application in mobile devices and embedded devices. Though there are many kinds of research on Image Detection, designing a robust and effective defect detection model is still an open issue. Therefore, this paper proposes a lightweight target detection algorithm based on the regression-based YOLOv4 by simplifying the backbone network, pruning the model with channel attention, and simplifying the anchor box. From the perspective of backbone network simplification, we designed a novel framework named Tiny-GhostNet to replace the original CSPDarknet53 network. Channel attention-based model pruning mainly utilizes channel attention to remove those unimportant channels. The simplification of anchor boxes aims to simplify predefined anchor box settings and density distribution.","PeriodicalId":396735,"journal":{"name":"2022 International Conference on Sensing, Measurement & Data Analytics in the era of Artificial Intelligence (ICSMD)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115036333","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}