This research utilizes edge computing technology to analyze the signal by MPU-6050 three-axis accelerometer and Raspberry Pi. We collect the rotating equipment’s vibration signal through the fast Fourier transform to discover the relationship between the control signal and the corresponding frequency spectrum under different operating conditions. First, we take a general indoor fan as an example. Then we find a suitable location for installing the accelerometer on the motor housing, and then collect the vibration signal when the motor is running through the MPU-6050. Then, Python is used to transform the processed data and store it in the database. The neural algorithm retrieves the relationship between the frequency spectrum signal and the fan control signal. Finally, write the rules back to the Raspberry Pi to monitor the status.After performing the test of rotating equipment, the results express that several experiments are required to locate the best measurement position when collecting vibration signals. The signal is converted to the frequency spectrum and then calculated by the proposed algorithm, which effectively discovers the characteristic value related to the control signal and allows the Raspberry Pi to detect the operating state of the device quickly. As a result, the system can immediately respond to abnormal conditions, such as sending an alert e-mail or displaying warning lights.
{"title":"Develop an Abnormal Detection System of Rotating Equipment by Edge Computing","authors":"Nian-Ze Hu, Shang-Wei Liu, Kai-Hsun Hsu, Ruo-Wei Wu, Zheng-Han Shi, Jieh-Tsyr Chuang, You-Xing Zeng, Li-Chiuan Tung, Jeng-Dao Lee","doi":"10.1109/ECICE52819.2021.9645611","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645611","url":null,"abstract":"This research utilizes edge computing technology to analyze the signal by MPU-6050 three-axis accelerometer and Raspberry Pi. We collect the rotating equipment’s vibration signal through the fast Fourier transform to discover the relationship between the control signal and the corresponding frequency spectrum under different operating conditions. First, we take a general indoor fan as an example. Then we find a suitable location for installing the accelerometer on the motor housing, and then collect the vibration signal when the motor is running through the MPU-6050. Then, Python is used to transform the processed data and store it in the database. The neural algorithm retrieves the relationship between the frequency spectrum signal and the fan control signal. Finally, write the rules back to the Raspberry Pi to monitor the status.After performing the test of rotating equipment, the results express that several experiments are required to locate the best measurement position when collecting vibration signals. The signal is converted to the frequency spectrum and then calculated by the proposed algorithm, which effectively discovers the characteristic value related to the control signal and allows the Raspberry Pi to detect the operating state of the device quickly. As a result, the system can immediately respond to abnormal conditions, such as sending an alert e-mail or displaying warning lights.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123410392","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 : 2021-10-29DOI: 10.1109/ECICE52819.2021.9645615
Shigang Wang, Yue Hu, Guang-Xing Tan
Novel Coronavirus targets the lung posing a serious threat to human health and causing huge social and economic losses. Extraction of lung parenchyma from CT images is an important step in the diagnosis of Novel Coronavirus. Therefore, accurate segmentation of lung parenchyma is highly significant for the diagnosis of disease. A lung parenchyma segmentation method based on OTSU and morphological operation is proposed. First of all, according to the CT image noise type, bilateral filtering is selected as preprocessing to filter out image noise. Then, binary images are obtained by the OTSU-based algorithm. Secondly, the residual interference of the trachea and blood vessels in the image is removed by morphological operation, and connected areas are marked and holes are filled. Finally, the original image is multiplied by the mask to obtain the lung parenchyma image. Experimental results show that this method can accurately segment lung parenchyma.
{"title":"Lung Parenchyma Segmentation Based on CT Images","authors":"Shigang Wang, Yue Hu, Guang-Xing Tan","doi":"10.1109/ECICE52819.2021.9645615","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645615","url":null,"abstract":"Novel Coronavirus targets the lung posing a serious threat to human health and causing huge social and economic losses. Extraction of lung parenchyma from CT images is an important step in the diagnosis of Novel Coronavirus. Therefore, accurate segmentation of lung parenchyma is highly significant for the diagnosis of disease. A lung parenchyma segmentation method based on OTSU and morphological operation is proposed. First of all, according to the CT image noise type, bilateral filtering is selected as preprocessing to filter out image noise. Then, binary images are obtained by the OTSU-based algorithm. Secondly, the residual interference of the trachea and blood vessels in the image is removed by morphological operation, and connected areas are marked and holes are filled. Finally, the original image is multiplied by the mask to obtain the lung parenchyma image. Experimental results show that this method can accurately segment lung parenchyma.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121842812","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 : 2021-10-29DOI: 10.1109/ECICE52819.2021.9645614
Shukun Wu, Shigang Wang
In modern architecture, the glass wall has become a widely used element. Two-dimensional lidar is one of many sensors used in robots, but it cannot directly identify the glass wall due to the light transmission characteristics of the laser- facing glass. The ultrasonic sensor can detect the presence of glass due to its working principle, but the general single-line ultrasonic sensor can only detect one point when detecting the glass wall. The detection method mentioned in this article combines two sensors by using image processing-related methods. After the experimental verification of various glass wall occasions, the combination of the two sensors can not only judge whether there is glass in the front, but also identify the position and size of the glass wall, and filter out the point cloud behind the glass in the lidar detection results. The data is thus formed into a whole point cloud data that does not contain the item information behind the glass wall. This paper explores the identification method of the glass wall, which has a lower cost and wider application range.
{"title":"Method for Detecting Glass Wall with Lidar and Ultrasonic Sensor","authors":"Shukun Wu, Shigang Wang","doi":"10.1109/ECICE52819.2021.9645614","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645614","url":null,"abstract":"In modern architecture, the glass wall has become a widely used element. Two-dimensional lidar is one of many sensors used in robots, but it cannot directly identify the glass wall due to the light transmission characteristics of the laser- facing glass. The ultrasonic sensor can detect the presence of glass due to its working principle, but the general single-line ultrasonic sensor can only detect one point when detecting the glass wall. The detection method mentioned in this article combines two sensors by using image processing-related methods. After the experimental verification of various glass wall occasions, the combination of the two sensors can not only judge whether there is glass in the front, but also identify the position and size of the glass wall, and filter out the point cloud behind the glass in the lidar detection results. The data is thus formed into a whole point cloud data that does not contain the item information behind the glass wall. This paper explores the identification method of the glass wall, which has a lower cost and wider application range.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"285 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131466952","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 : 2021-10-29DOI: 10.1109/ECICE52819.2021.9645720
Ye Sun
According to statistics, in traffic accidents, injuries caused by frontal collisions of passenger cars account for the largest proportion of all traffic accidents, about 40%. Therefore, it is particularly important to improve the frontal collision of passenger cars. Therefore, this article uses the ANSYS/LSDYNA module to model the passenger car with finite element analysis and analyze the frontal rigid wall collision at a speed of 50km/h. According to the frontal collision process of the passenger car, the characteristics of deformation, energy absorption, and dummy posture during the collision process are studied. Through the extracted animation and curve results, the collision characteristics of the passenger car are analyzed. After optimizing the energy-absorbing box, through the analysis of the research results, the following conclusions are drawn. 5 layers of the energy-absorbing layer part of the energy-absorbing box give the best results, and the energy absorption capacity of the energy-absorbing box has reached twice the original structure. The impact resistance of the passenger car is greatly improved, the personnel in the cockpit are safer, and the vehicle is less likely to be damaged.
{"title":"Research and Analysis of Frontal Collision of Passenger Cars Based on ANSYS","authors":"Ye Sun","doi":"10.1109/ECICE52819.2021.9645720","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645720","url":null,"abstract":"According to statistics, in traffic accidents, injuries caused by frontal collisions of passenger cars account for the largest proportion of all traffic accidents, about 40%. Therefore, it is particularly important to improve the frontal collision of passenger cars. Therefore, this article uses the ANSYS/LSDYNA module to model the passenger car with finite element analysis and analyze the frontal rigid wall collision at a speed of 50km/h. According to the frontal collision process of the passenger car, the characteristics of deformation, energy absorption, and dummy posture during the collision process are studied. Through the extracted animation and curve results, the collision characteristics of the passenger car are analyzed. After optimizing the energy-absorbing box, through the analysis of the research results, the following conclusions are drawn. 5 layers of the energy-absorbing layer part of the energy-absorbing box give the best results, and the energy absorption capacity of the energy-absorbing box has reached twice the original structure. The impact resistance of the passenger car is greatly improved, the personnel in the cockpit are safer, and the vehicle is less likely to be damaged.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121276346","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 : 2021-10-29DOI: 10.1109/ECICE52819.2021.9645659
Jiye Li, Jianhui Ren, Longchen Ma, Simeng Zhang
This paper aims to study the correlation between the northern segment of the Yishu fault zone and the moderate-strong earthquake activities in the northeast region. The dynamic seduction of the western Pacific Plate and the tectonic environment of the northern segment of the Yishu fault zone are considered in this preliminary analysis. The seismic activities in this region play a role of a seismic window with the following characteristics: (1) Frequent occurrence of ML≥3.0 earthquakes in the window area is a signal indicating moderate - strong earthquakes in the northeast region, and the opening condition of the window is a frequency of 6. (2) The higher the frequency of ML>3.0 earthquakes in the window area, the further distance the moderate-strong earthquakes in Heilongjiang and its surrounding areas from the window boundary. (3) The higher the frequency of ML≥3.0 earthquakes, the larger the magnitude of moderate-strong earthquakes in Heilongjiang and its surrounding areas. In this paper, a retrospective study of earthquakes with ML≥5.0 in Northeast China since 1999 was carried out by using the above methods. It was found that when the opening condition of the "signal type" window was met, moderate and strong earthquakes in northeast China were more frequent.
{"title":"Seismic Signal Characteristics of Seismicity Data in the Northern Segment of the Yishu Fault Zone","authors":"Jiye Li, Jianhui Ren, Longchen Ma, Simeng Zhang","doi":"10.1109/ECICE52819.2021.9645659","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645659","url":null,"abstract":"This paper aims to study the correlation between the northern segment of the Yishu fault zone and the moderate-strong earthquake activities in the northeast region. The dynamic seduction of the western Pacific Plate and the tectonic environment of the northern segment of the Yishu fault zone are considered in this preliminary analysis. The seismic activities in this region play a role of a seismic window with the following characteristics: (1) Frequent occurrence of ML≥3.0 earthquakes in the window area is a signal indicating moderate - strong earthquakes in the northeast region, and the opening condition of the window is a frequency of 6. (2) The higher the frequency of ML>3.0 earthquakes in the window area, the further distance the moderate-strong earthquakes in Heilongjiang and its surrounding areas from the window boundary. (3) The higher the frequency of ML≥3.0 earthquakes, the larger the magnitude of moderate-strong earthquakes in Heilongjiang and its surrounding areas. In this paper, a retrospective study of earthquakes with ML≥5.0 in Northeast China since 1999 was carried out by using the above methods. It was found that when the opening condition of the \"signal type\" window was met, moderate and strong earthquakes in northeast China were more frequent.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133971280","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}
This research utilizes a camera, a collaborative arm, and a PLC-controlled conveyor belt to develop a system that automatically recognizes the angle of the workpiece and quickly loads and unloads the material. First, set up the camera on the gantry of the conveyor belt to accurately grasp the object’s position. Second, develop an algorithm to handle the collaborative arm’s multi-angle motion according to the object’s data. The captured image is analyzed through WebAPI technology to obtain the best and fast-moving path to grab the workpiece. Then, the AI algorithm determines the object’s type, shape, and angle on the conveyor belt. Finally, the recommended path is generated, and the coordinated arm can move to the target quickly and accurately. The experimental results show that the proposed platform accurately adjusts the angle of the gripper and quickly controls the arm to move the workpiece after acquiring the object’s position on the conveyor belt. Coupled with WebAPI technology and improving the efficiency of identification, it also enables a set of AI hosts to serve multiple production lines simultaneously, which is very suitable for use in manufacturing industrials.
{"title":"Design a Fast Loading and Unloading Robot Controlling System by WebAPI Technique","authors":"Nian-Ze Hu, Zheng-Han Shi, Kai-Hsun Hsu, Shang-Wei Liu, Ruo-Wei Wu, Jieh-Tsyr Chuang, You-Xing Zeng, Chia-Chen Kuo, Jeng-Dao Lee","doi":"10.1109/ECICE52819.2021.9645713","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645713","url":null,"abstract":"This research utilizes a camera, a collaborative arm, and a PLC-controlled conveyor belt to develop a system that automatically recognizes the angle of the workpiece and quickly loads and unloads the material. First, set up the camera on the gantry of the conveyor belt to accurately grasp the object’s position. Second, develop an algorithm to handle the collaborative arm’s multi-angle motion according to the object’s data. The captured image is analyzed through WebAPI technology to obtain the best and fast-moving path to grab the workpiece. Then, the AI algorithm determines the object’s type, shape, and angle on the conveyor belt. Finally, the recommended path is generated, and the coordinated arm can move to the target quickly and accurately. The experimental results show that the proposed platform accurately adjusts the angle of the gripper and quickly controls the arm to move the workpiece after acquiring the object’s position on the conveyor belt. Coupled with WebAPI technology and improving the efficiency of identification, it also enables a set of AI hosts to serve multiple production lines simultaneously, which is very suitable for use in manufacturing industrials.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115346956","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 : 2021-10-29DOI: 10.1109/ECICE52819.2021.9645619
Danni Zou, Jun Lang
Recently, researchers have made a lot of effort to reduce the cost of Synthetic Aperture Radar (SAR) imaging with the Compressed Sensing (CS) theory, of which sparse reconstruction is an important part. One method for solving sparse reconstruction problem is Smoothed L0 (SL0) algorithm, in which the L0 norm is approximated with a convex function. By minimizing the convex function, the feasible region constraint is satisfied. In this paper, we present the Improved SL0 (ISL0) algorithm by changing the optimization stage of the SL0 and performing it as Tabu Search (TS) algorithm. By replacing the steepest descent algorithm with the tabu search algorithm, we achieve a faster iteration speed with higher recovery quality using ISL0 compared to the SL0. Regarding the results of numerical experiments under the same test conditions, the number of iterations for the new algorithm compared to the original SL0 was about 20 times less. The efficiency of the ISL0 is evaluated for the reconstruction of spotlight SAR images through an experiment. The results of this simulation indicate that the quality of image reconstruction with ISL0 is better than SL0 for various SNR.
{"title":"Sparse Reconstruction Using Improved SL0 Algorithm by Tabu Search and Application to Spotlight SAR Imaging","authors":"Danni Zou, Jun Lang","doi":"10.1109/ECICE52819.2021.9645619","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645619","url":null,"abstract":"Recently, researchers have made a lot of effort to reduce the cost of Synthetic Aperture Radar (SAR) imaging with the Compressed Sensing (CS) theory, of which sparse reconstruction is an important part. One method for solving sparse reconstruction problem is Smoothed L0 (SL0) algorithm, in which the L0 norm is approximated with a convex function. By minimizing the convex function, the feasible region constraint is satisfied. In this paper, we present the Improved SL0 (ISL0) algorithm by changing the optimization stage of the SL0 and performing it as Tabu Search (TS) algorithm. By replacing the steepest descent algorithm with the tabu search algorithm, we achieve a faster iteration speed with higher recovery quality using ISL0 compared to the SL0. Regarding the results of numerical experiments under the same test conditions, the number of iterations for the new algorithm compared to the original SL0 was about 20 times less. The efficiency of the ISL0 is evaluated for the reconstruction of spotlight SAR images through an experiment. The results of this simulation indicate that the quality of image reconstruction with ISL0 is better than SL0 for various SNR.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117155135","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 : 2021-10-29DOI: 10.1109/ECICE52819.2021.9645641
Zhuqi Shi, Li Wei, Youyun Xu
With the rapid-growing demand for location-based services in indoor environments, fingerprint-based indoor positioning methods have attracted great interest owing to high accuracy and low online complexity. In this paper, we use the channel state information (CSI) of the massive MIMO (MaMIMO) system as the fingerprint to construct the fingerprint database. Different from the previous methods that only use CSI amplitude to construct fingerprints, phase information and angle of arrival (AOA) are added to the fingerprint to enhance the characteristics of fingerprint data. We modified the network according to the characteristics of fingerprint data based on Google Net and implemented a GoogleNet-like convolutional neural network(CNN) which uses skip connection and 1-D convolution kernel for fingerprint positioning. Experiment results show that with sufficient representative data sets, centimeter-level positioning can be achieved by using the proposed neural network, and the positioning accuracy can be further improved by 10% with the use of AOA information.
{"title":"CSI-based Fingerprinting for Indoor Localization with Multi-scale Convolutional Neural Network","authors":"Zhuqi Shi, Li Wei, Youyun Xu","doi":"10.1109/ECICE52819.2021.9645641","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645641","url":null,"abstract":"With the rapid-growing demand for location-based services in indoor environments, fingerprint-based indoor positioning methods have attracted great interest owing to high accuracy and low online complexity. In this paper, we use the channel state information (CSI) of the massive MIMO (MaMIMO) system as the fingerprint to construct the fingerprint database. Different from the previous methods that only use CSI amplitude to construct fingerprints, phase information and angle of arrival (AOA) are added to the fingerprint to enhance the characteristics of fingerprint data. We modified the network according to the characteristics of fingerprint data based on Google Net and implemented a GoogleNet-like convolutional neural network(CNN) which uses skip connection and 1-D convolution kernel for fingerprint positioning. Experiment results show that with sufficient representative data sets, centimeter-level positioning can be achieved by using the proposed neural network, and the positioning accuracy can be further improved by 10% with the use of AOA information.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123904234","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}
Based on the big data of solder joint degradation and failure under the coupling action of thermal, electrical, and mechanical stress, a three-parameter Weibull distribution failure probability model of SnAgCu solder joint is established to solve the mapping problem between different failure modes and different failure probability distribution of solder joint. The creep and degradation trend of solder joint reliability under three-stress coupling is analyzed by finite element analysis method, and the fatigue and minimum life model of solder joint with progressive excitation of interconnecting fault are presented. Using a multidimensional data-driven algorithm, the characterization and feature extraction of failure state of SnAgCu solder joint based on degradation sensitivity and salience model is presented to identify and excavate the internal laws and relations between the reliability level of the solder joint and the three-coupling stress. The semi-physical simulation shows that the reliability degradation model can effectively predict solder joint health.
{"title":"Research on Reliability Degradation Model of SnAgCu Solder Joint under Three-coupling Stress Based on Weibull Distribution and Multidimensional Data","authors":"Jie-Mei Zhang, Ningzhou Li, Danyu Zhang, Xiaojuan Wei, Xiaojuan Zhang, Mingming Chen","doi":"10.1109/ECICE52819.2021.9645634","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645634","url":null,"abstract":"Based on the big data of solder joint degradation and failure under the coupling action of thermal, electrical, and mechanical stress, a three-parameter Weibull distribution failure probability model of SnAgCu solder joint is established to solve the mapping problem between different failure modes and different failure probability distribution of solder joint. The creep and degradation trend of solder joint reliability under three-stress coupling is analyzed by finite element analysis method, and the fatigue and minimum life model of solder joint with progressive excitation of interconnecting fault are presented. Using a multidimensional data-driven algorithm, the characterization and feature extraction of failure state of SnAgCu solder joint based on degradation sensitivity and salience model is presented to identify and excavate the internal laws and relations between the reliability level of the solder joint and the three-coupling stress. The semi-physical simulation shows that the reliability degradation model can effectively predict solder joint health.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123978408","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 : 2021-10-29DOI: 10.1109/ECICE52819.2021.9645710
S. Koo, Yujin Lim
In the Industrial Internet of Things (IIoT), various types of tasks are processed for the small quantity batch production. But there are many challenges due to the limited battery lifespan and computational capabilities of devices. To overcome the limitations, Mobile Edge Computing (MEC) has been introduced. In MEC, a task offloading technique to execute the tasks attracts much attention. A MEC server (MECS) has limited computational capability, which increases the burden on the server and a cellular network if a larger number of tasks are offloaded to the server. It can reduce the quality of service for task execution. Thus, offloading between nearby devices through device-to-device (D2D) communication is drawing attention. We propose the optimal task offloading decision strategy in the MEC and D2D communication architecture. We aim to minimize the energy consumption of devices and task execution delay under delay constraints. To solve the problem, we adopt Q-learning algorithm as one of Reinforcement Learning (RL). Simulation results show that the proposed algorithm outperforms the other methods in terms of energy consumption of devices and task execution delay.
{"title":"Optimal Task Offloading Decision in IIoT Enviornments Using Reinforcement Learning","authors":"S. Koo, Yujin Lim","doi":"10.1109/ECICE52819.2021.9645710","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645710","url":null,"abstract":"In the Industrial Internet of Things (IIoT), various types of tasks are processed for the small quantity batch production. But there are many challenges due to the limited battery lifespan and computational capabilities of devices. To overcome the limitations, Mobile Edge Computing (MEC) has been introduced. In MEC, a task offloading technique to execute the tasks attracts much attention. A MEC server (MECS) has limited computational capability, which increases the burden on the server and a cellular network if a larger number of tasks are offloaded to the server. It can reduce the quality of service for task execution. Thus, offloading between nearby devices through device-to-device (D2D) communication is drawing attention. We propose the optimal task offloading decision strategy in the MEC and D2D communication architecture. We aim to minimize the energy consumption of devices and task execution delay under delay constraints. To solve the problem, we adopt Q-learning algorithm as one of Reinforcement Learning (RL). Simulation results show that the proposed algorithm outperforms the other methods in terms of energy consumption of devices and task execution delay.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125569092","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：