In the railway container yard, there are few mature intelligent lifting prevention solutions available for train flatbed loading and unloading operations due to the poor detection accuracy or speed of traditional detection methods. This paper designs a train Flatbed Twist Rail (F-TR) lock anti-lifting detection method based on an improved BP neural network. The system collects weight and laser distance measurement data from the four locks of the hoist, establishes a flatbed lifting detection model based on the BP neural network, and optimizes the model's performance by incorporating a momentum factor and adaptive learning rate during weight adjustment. In practical tests, this system demonstrates a high detection rate and fast detection speed, offering intelligent safety protection for automated rail mounted gantry in the railway container yard.
在铁路集装箱堆场，由于传统检测方法的检测精度或速度较差，目前针对列车平板装卸作业的成熟智能防抬升解决方案还很少。本文设计了一种基于改进型 BP 神经网络的列车平板扭轨（F-TR）锁防起重检测方法。该系统收集了提升机四个锁的重量和激光测距数据，建立了基于 BP 神经网络的平板提升检测模型，并在重量调整过程中加入了动量因子和自适应学习率，优化了模型的性能。在实际测试中，该系统表现出较高的检测率和较快的检测速度，为铁路集装箱堆场的自动化轨道龙门架提供了智能安全保护。
{"title":"A train F-TR lock anti-lifting detection method based on improved BP neural network","authors":"Jun Jiang","doi":"10.21595/jme.2023.23638","DOIUrl":"https://doi.org/10.21595/jme.2023.23638","url":null,"abstract":"In the railway container yard, there are few mature intelligent lifting prevention solutions available for train flatbed loading and unloading operations due to the poor detection accuracy or speed of traditional detection methods. This paper designs a train Flatbed Twist Rail (F-TR) lock anti-lifting detection method based on an improved BP neural network. The system collects weight and laser distance measurement data from the four locks of the hoist, establishes a flatbed lifting detection model based on the BP neural network, and optimizes the model's performance by incorporating a momentum factor and adaptive learning rate during weight adjustment. In practical tests, this system demonstrates a high detection rate and fast detection speed, offering intelligent safety protection for automated rail mounted gantry in the railway container yard.","PeriodicalId":42196,"journal":{"name":"Journal of Measurements in Engineering","volume":"53 4","pages":""},"PeriodicalIF":1.6,"publicationDate":"2024-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139446079","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}
With the continuous development of the energy industry, photovoltaic power generation is gradually becoming one of the main power generation methods. However, detecting hot spot defects in photovoltaic power stations is challenging. Therefore, enhancing detection efficiency using information technology has become a crucial aspect. The study presents a defect detection model for PV power stations using the YOLOv3 (You Only Look Once v3) algorithm. The model incorporates coordinate attention module (CAM) and self-attention module (SAM) to improve feature extraction in low-resolution conditions. The Multi objective Sparrow is employed to achieve multiple objectives. It is very contributing in the detection of low-resolution features. It shows that the research method can reduce the loss value to 0.009 after 400 iterations of the loss curve test. The precision-recall (P-R) curve generated by the research method only starts to drop sharply when the Recall value reaches 0.96. The number of parameters generated by the research method is 3.46×106. The detection accuracy of the research method reaches 98.86 % when there are five defective fault types. The results indicate that the proposed research method offers improved detection speed and higher accuracy in identifying hot spot defects in PV power stations. This technology provides valuable support for hot spot defect detection and presents new opportunities for the field.
随着能源产业的不断发展，光伏发电逐渐成为主要的发电方式之一。然而，检测光伏发电站的热点缺陷是一项挑战。因此，利用信息技术提高检测效率已成为一个重要方面。本研究利用 YOLOv3（You Only Look Once v3）算法提出了一种光伏电站缺陷检测模型。该模型结合了协调注意模块（CAM）和自我注意模块（SAM），以改进低分辨率条件下的特征提取。多目标麻雀（Multi objective Sparrow）用于实现多个目标。它对低分辨率特征的检测非常有帮助。结果表明，经过 400 次迭代损失曲线测试后，该研究方法可将损失值降至 0.009。研究方法生成的精度-召回（P-R）曲线在召回值达到 0.96 时才开始急剧下降。研究方法生成的参数数为 3.46×106。当有五种缺陷故障类型时，研究方法的检测准确率达到 98.86 %。结果表明，所提出的研究方法在识别光伏电站热点缺陷方面具有更快的检测速度和更高的准确度。该技术为热点缺陷检测提供了有价值的支持，并为该领域带来了新的机遇。
{"title":"YOLOv3-MSSA based hot spot defect detection for photovoltaic power stations","authors":"Kaiming Gu, Yong Chen","doi":"10.21595/jme.2023.23418","DOIUrl":"https://doi.org/10.21595/jme.2023.23418","url":null,"abstract":"With the continuous development of the energy industry, photovoltaic power generation is gradually becoming one of the main power generation methods. However, detecting hot spot defects in photovoltaic power stations is challenging. Therefore, enhancing detection efficiency using information technology has become a crucial aspect. The study presents a defect detection model for PV power stations using the YOLOv3 (You Only Look Once v3) algorithm. The model incorporates coordinate attention module (CAM) and self-attention module (SAM) to improve feature extraction in low-resolution conditions. The Multi objective Sparrow is employed to achieve multiple objectives. It is very contributing in the detection of low-resolution features. It shows that the research method can reduce the loss value to 0.009 after 400 iterations of the loss curve test. The precision-recall (P-R) curve generated by the research method only starts to drop sharply when the Recall value reaches 0.96. The number of parameters generated by the research method is 3.46×106. The detection accuracy of the research method reaches 98.86 % when there are five defective fault types. The results indicate that the proposed research method offers improved detection speed and higher accuracy in identifying hot spot defects in PV power stations. This technology provides valuable support for hot spot defect detection and presents new opportunities for the field.","PeriodicalId":42196,"journal":{"name":"Journal of Measurements in Engineering","volume":"4 1","pages":""},"PeriodicalIF":1.6,"publicationDate":"2024-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139451344","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}
Foundation pit excavation can cause settlement and displacement of surrounding existing buildings and roads. In order to study the influence of soil unloading on the surrounding buildings during pit foundation excavation, the application of a pile-anchor retaining structure in a deep foundation pit was studied, with the deep foundation pit project of Anhui Bright Pearl Mall as the research subject. Through theoretical analysis, field measurements, and FLAC3D numerical simulations, the supporting structure was comprehensively analyzed. A comparison was made between the measured displacement data and the numerical simulation results of the supporting structure and the surrounding environment during the excavation process of the foundation pit. The results indicate that the model results, obtained through the use of the FLAC3D software for numerical simulations, generally align with the field data. This approach can more accurately reflect the evolutionary laws of soil pressure and deformation during the excavation of the foundation pit. The maximum displacement of the horizontal displacement monitoring point in this project's foundation pit is 25.96 mm, which is less than the monitoring alarm value of 30 mm. The horizontal displacement monitoring of the sidewall of the foundation pit is crucial among them. An analysis of the three major causes of numerical deviation provides valuable insights for the design of deep foundation pit supporting structures.
{"title":"Displacement analysis and numerical simulation of pile-anchor retaining structure in deep foundation pit","authors":"Xupeng Yin, Hongmei Ni","doi":"10.21595/jme.2023.23635","DOIUrl":"https://doi.org/10.21595/jme.2023.23635","url":null,"abstract":"Foundation pit excavation can cause settlement and displacement of surrounding existing buildings and roads. In order to study the influence of soil unloading on the surrounding buildings during pit foundation excavation, the application of a pile-anchor retaining structure in a deep foundation pit was studied, with the deep foundation pit project of Anhui Bright Pearl Mall as the research subject. Through theoretical analysis, field measurements, and FLAC3D numerical simulations, the supporting structure was comprehensively analyzed. A comparison was made between the measured displacement data and the numerical simulation results of the supporting structure and the surrounding environment during the excavation process of the foundation pit. The results indicate that the model results, obtained through the use of the FLAC3D software for numerical simulations, generally align with the field data. This approach can more accurately reflect the evolutionary laws of soil pressure and deformation during the excavation of the foundation pit. The maximum displacement of the horizontal displacement monitoring point in this project's foundation pit is 25.96 mm, which is less than the monitoring alarm value of 30 mm. The horizontal displacement monitoring of the sidewall of the foundation pit is crucial among them. An analysis of the three major causes of numerical deviation provides valuable insights for the design of deep foundation pit supporting structures.","PeriodicalId":42196,"journal":{"name":"Journal of Measurements in Engineering","volume":"133 11","pages":""},"PeriodicalIF":1.6,"publicationDate":"2024-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139453539","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}
A. Czakó, K. Řehák, A. Prokop, Jakub Rekem, Daniel Láštic, Miroslav Trochta
Transmission error (TE) is a significant parameter related to gears vibration widely investigated by many authors using different approaches. However, in previous studies, spur and helical gears were mainly examined. There is a lack of studies addressed to double helical and herringbone gears and a comparison among several types of gearing with parallel axes. In this paper, spur, helical, double helical, and herringbone gears are analyzed in terms of static transmission error (STE), contact pressure and tooth root stress. Static contact analyses were conducted using the finite element method (FEM) which is often considered a tool for validating other methods and approaches. Moreover, three variants of boundary conditions of each gear type are introduced, including flexible shafts and the effect of a tip relief modification at sole gears, without shafts, was analyzed. In addition, a concept of a compact test rig intended for STE measurements at low loads was presented. The results have shown, among other things, significant influence of the shaft stiffness and boundary conditions on meshing characteristics.
{"title":"Static transmission error measurement of various gear-shaft systems by finite element analysis","authors":"A. Czakó, K. Řehák, A. Prokop, Jakub Rekem, Daniel Láštic, Miroslav Trochta","doi":"10.21595/jme.2023.23843","DOIUrl":"https://doi.org/10.21595/jme.2023.23843","url":null,"abstract":"Transmission error (TE) is a significant parameter related to gears vibration widely investigated by many authors using different approaches. However, in previous studies, spur and helical gears were mainly examined. There is a lack of studies addressed to double helical and herringbone gears and a comparison among several types of gearing with parallel axes. In this paper, spur, helical, double helical, and herringbone gears are analyzed in terms of static transmission error (STE), contact pressure and tooth root stress. Static contact analyses were conducted using the finite element method (FEM) which is often considered a tool for validating other methods and approaches. Moreover, three variants of boundary conditions of each gear type are introduced, including flexible shafts and the effect of a tip relief modification at sole gears, without shafts, was analyzed. In addition, a concept of a compact test rig intended for STE measurements at low loads was presented. The results have shown, among other things, significant influence of the shaft stiffness and boundary conditions on meshing characteristics.","PeriodicalId":42196,"journal":{"name":"Journal of Measurements in Engineering","volume":"282 1","pages":""},"PeriodicalIF":1.6,"publicationDate":"2023-12-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139152814","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}
Aling Zhang, Qianmiao Bu, Wen Zhang, Guomeng He, Yong Deng
In this study, movable steel barrier with grade SB light composite corrugated beam is designed, which addresses the problems of the prior central partition belt portable guardrail in terms of easy mobility, local safety, easy construction, and other indications. This guardrail employs explicit algorithms to conduct a dynamic finite element simulation analysis and a real vehicle crash test, and verifies the guardrails' blocking, guiding, and buffering functions in accordance with the SB level collision conditions listed in the Standard for Safety Performance Evaluation of Highway Barriers (JTG B05-01-2013). According to the results, the safety performance of SB grade lightweight composite corrugated beam movable steel guardrail meets the requirements of the Standard for Safety Performance Evaluation of Highway Barriers (JTG B05-01-2013). In addition, the guardrail can be opened for 12 meters in 1 minute and returned to close in 2 minutes. The opening and restoration of the movable guardrail is superior to the previous central divider movable guardrail. This guardrail has been tried for some high-speed and its safety performance has been verified again in actual high-speed vehicle collisions.
{"title":"Test and application of movable steel barrier with grade SB light composite corrugated beam","authors":"Aling Zhang, Qianmiao Bu, Wen Zhang, Guomeng He, Yong Deng","doi":"10.21595/jme.2023.23386","DOIUrl":"https://doi.org/10.21595/jme.2023.23386","url":null,"abstract":"In this study, movable steel barrier with grade SB light composite corrugated beam is designed, which addresses the problems of the prior central partition belt portable guardrail in terms of easy mobility, local safety, easy construction, and other indications. This guardrail employs explicit algorithms to conduct a dynamic finite element simulation analysis and a real vehicle crash test, and verifies the guardrails' blocking, guiding, and buffering functions in accordance with the SB level collision conditions listed in the Standard for Safety Performance Evaluation of Highway Barriers (JTG B05-01-2013). According to the results, the safety performance of SB grade lightweight composite corrugated beam movable steel guardrail meets the requirements of the Standard for Safety Performance Evaluation of Highway Barriers (JTG B05-01-2013). In addition, the guardrail can be opened for 12 meters in 1 minute and returned to close in 2 minutes. The opening and restoration of the movable guardrail is superior to the previous central divider movable guardrail. This guardrail has been tried for some high-speed and its safety performance has been verified again in actual high-speed vehicle collisions.","PeriodicalId":42196,"journal":{"name":"Journal of Measurements in Engineering","volume":"33 10","pages":""},"PeriodicalIF":1.6,"publicationDate":"2023-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139162051","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 study presents a new solution to address challenges encountered in additive manufacturing, specifically in the context of 3D printing, where failures can occur due to complications associated with the nozzle or filament. The proposed solution in this research involves using a time-domain feature extraction method that leverages sound and vibration patterns. By implementing sensors to capture these signals in a controlled and noise-free environment, and then utilizing a Multi-Layer Perceptron (MLP) model trained accurately to predict upcoming signals and vibrations, proactive anticipation of printing outcomes is facilitated, including potential failures. Simulation results obtained using MATLAB for the MLP showcase the effectiveness of this approach, demonstrating remarkably low error rates. Furthermore, through rigorous data validation, the proposed method's ability to accurately identify sound and vibration signals is confirmed. As a result, the likelihood of failures is significantly reduced, thereby preventing defects in the filament. The implications of this solution hold great promise in substantially enhancing the reliability and efficiency of additive manufacturing processes.
{"title":"Utilizing a knowledge-based training algorithm and time-domain extraction for pattern recognition in cylindrical features through vibration and sound signals","authors":"M. Dirhamsyah, Hammam Riza, M. S. Rizal","doi":"10.21595/jme.2023.23452","DOIUrl":"https://doi.org/10.21595/jme.2023.23452","url":null,"abstract":"This study presents a new solution to address challenges encountered in additive manufacturing, specifically in the context of 3D printing, where failures can occur due to complications associated with the nozzle or filament. The proposed solution in this research involves using a time-domain feature extraction method that leverages sound and vibration patterns. By implementing sensors to capture these signals in a controlled and noise-free environment, and then utilizing a Multi-Layer Perceptron (MLP) model trained accurately to predict upcoming signals and vibrations, proactive anticipation of printing outcomes is facilitated, including potential failures. Simulation results obtained using MATLAB for the MLP showcase the effectiveness of this approach, demonstrating remarkably low error rates. Furthermore, through rigorous data validation, the proposed method's ability to accurately identify sound and vibration signals is confirmed. As a result, the likelihood of failures is significantly reduced, thereby preventing defects in the filament. The implications of this solution hold great promise in substantially enhancing the reliability and efficiency of additive manufacturing processes.","PeriodicalId":42196,"journal":{"name":"Journal of Measurements in Engineering","volume":"9 1","pages":""},"PeriodicalIF":1.6,"publicationDate":"2023-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139001455","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 study aims to design efficient and reliable artificial intelligence vision detection models to improve detection efficiency and accuracy. The study filters defect-free images by image preprocessing and region of interest detection techniques. AlexNet network is enhanced by introducing attention mechanism modules, deep separable convolutions, and more to effectively boost the network's feature extraction capacity. An area convolutional neural network is developed to rapidly identify and locate defects on steel plate surfaces, utilizing an enhanced AlexNet network for feature extraction. Results demonstrated that the algorithm attained an average detection rate of 98 % and can identify defects in a minimal time of only 0.0011 seconds. For the detection of six types of steel plate defects, the average accuracy of the optimized fast regional convolutional neural network reached more than 0.9, especially for the detection of small-size defects with excellent performance. This improved AlexNet network has a great advantage in F1 value. The conclusion of the study shows that the designed artificial intelligence vision detection model has high detection accuracy, speed, and performance stability in steel plate surface defect detection and has a wide range of application prospects.
{"title":"Application of AI intelligent vision detection technology using deep learning algorithm","authors":"Yan Huang","doi":"10.21595/jme.2023.23506","DOIUrl":"https://doi.org/10.21595/jme.2023.23506","url":null,"abstract":"This study aims to design efficient and reliable artificial intelligence vision detection models to improve detection efficiency and accuracy. The study filters defect-free images by image preprocessing and region of interest detection techniques. AlexNet network is enhanced by introducing attention mechanism modules, deep separable convolutions, and more to effectively boost the network's feature extraction capacity. An area convolutional neural network is developed to rapidly identify and locate defects on steel plate surfaces, utilizing an enhanced AlexNet network for feature extraction. Results demonstrated that the algorithm attained an average detection rate of 98 % and can identify defects in a minimal time of only 0.0011 seconds. For the detection of six types of steel plate defects, the average accuracy of the optimized fast regional convolutional neural network reached more than 0.9, especially for the detection of small-size defects with excellent performance. This improved AlexNet network has a great advantage in F1 value. The conclusion of the study shows that the designed artificial intelligence vision detection model has high detection accuracy, speed, and performance stability in steel plate surface defect detection and has a wide range of application prospects.","PeriodicalId":42196,"journal":{"name":"Journal of Measurements in Engineering","volume":"24 10","pages":""},"PeriodicalIF":1.6,"publicationDate":"2023-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138600954","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}
Zhihong Wang, Chaoying Wang, Yonggang Chen, Jianxin Li
An improved generative adversarial network model is adopted to improve the resolution of remote sensing images and the target detection algorithm for color remote sensing images. The main objective is to solve the problem of training super-resolution reconstruction algorithms and missing details in reconstructed images, aiming to achieve high-precision detection of medium and low-resolution color remote sensing targets. First, a lightweight image super-resolution reconstruction algorithm based on an improved generative adversarial network (GAN) is proposed. This algorithm combines the pixel attention mechanism and up-sampling method to restore image details. It further integrates edge-oriented convolution modules into traditional convolution to reduce model parameters and achieve better feature collection. Then, to further enhance the feature collection ability of the model, the YOLOv4 object detection algorithm is also improved. This is achieved by introducing the Focus structure into the backbone feature extraction network and integrating multi-layer separable convolutions to improve the feature extraction ability. The experimental results show that the improved target detection algorithm based on super resolution has a good detection effect on remote sensing image targets. It can effectively improve the detection accuracy of remote sensing images, and have a certain reference significance for the realization of small target detection in remote sensing images.
{"title":"Target detection algorithm based on super- resolution color remote sensing image reconstruction","authors":"Zhihong Wang, Chaoying Wang, Yonggang Chen, Jianxin Li","doi":"10.21595/jme.2023.23510","DOIUrl":"https://doi.org/10.21595/jme.2023.23510","url":null,"abstract":"An improved generative adversarial network model is adopted to improve the resolution of remote sensing images and the target detection algorithm for color remote sensing images. The main objective is to solve the problem of training super-resolution reconstruction algorithms and missing details in reconstructed images, aiming to achieve high-precision detection of medium and low-resolution color remote sensing targets. First, a lightweight image super-resolution reconstruction algorithm based on an improved generative adversarial network (GAN) is proposed. This algorithm combines the pixel attention mechanism and up-sampling method to restore image details. It further integrates edge-oriented convolution modules into traditional convolution to reduce model parameters and achieve better feature collection. Then, to further enhance the feature collection ability of the model, the YOLOv4 object detection algorithm is also improved. This is achieved by introducing the Focus structure into the backbone feature extraction network and integrating multi-layer separable convolutions to improve the feature extraction ability. The experimental results show that the improved target detection algorithm based on super resolution has a good detection effect on remote sensing image targets. It can effectively improve the detection accuracy of remote sensing images, and have a certain reference significance for the realization of small target detection in remote sensing images.","PeriodicalId":42196,"journal":{"name":"Journal of Measurements in Engineering","volume":"66 1","pages":""},"PeriodicalIF":1.6,"publicationDate":"2023-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139262103","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}
In order to evaluate the comprehensive dynamic performance of probability screen and select the appropriate working conditions, a dynamic model of probability screen vibration system is established. Then, the calculation method of the dynamic characteristic parameters, based on the time series Auto Regression (AR) model of vibration test, is used. The relationship among the comprehensive dynamic characteristics, the screening efficiency and the box dimension of probability screen vibration system is analyzed, and Least Square Support Vector Machine (LSSVM), Generalized Regression Neural Network (GRNN) and Back Propagation Neural Network (BPNN) are used to predict the screening efficiency with box dimension. The analysis result shows that the screening efficiency, the stability, the response rapidity and the comprehensive dynamic characteristic of the system are all related to the box dimension of time series. As for the complexity of probability screen vibration system, it affects the comprehensive dynamic performance, and ultimately touches the screening efficiency of the probability screen; The best working conditions for the system are selected by the curve between box dimension and the working condition parameter; Taking box dimension as the only input variable, the prediction accuracy of the screening efficiency is high by using LSSVM,GRNN and BPNN methods, the prediction results are stable and reliable, and the box dimension can be used as a single input variable to predict the screening efficiency, it has the advantages of fewer input parameters, high prediction efficiency, and high prediction accuracy, which has great potential for expanding application space and further research value.
{"title":"Prediction of comprehensive dynamic performance for probability screen based on AR model-box dimension","authors":"Qingtang Chen, Yijian Huang","doi":"10.21595/jme.2023.23522","DOIUrl":"https://doi.org/10.21595/jme.2023.23522","url":null,"abstract":"In order to evaluate the comprehensive dynamic performance of probability screen and select the appropriate working conditions, a dynamic model of probability screen vibration system is established. Then, the calculation method of the dynamic characteristic parameters, based on the time series Auto Regression (AR) model of vibration test, is used. The relationship among the comprehensive dynamic characteristics, the screening efficiency and the box dimension of probability screen vibration system is analyzed, and Least Square Support Vector Machine (LSSVM), Generalized Regression Neural Network (GRNN) and Back Propagation Neural Network (BPNN) are used to predict the screening efficiency with box dimension. The analysis result shows that the screening efficiency, the stability, the response rapidity and the comprehensive dynamic characteristic of the system are all related to the box dimension of time series. As for the complexity of probability screen vibration system, it affects the comprehensive dynamic performance, and ultimately touches the screening efficiency of the probability screen; The best working conditions for the system are selected by the curve between box dimension and the working condition parameter; Taking box dimension as the only input variable, the prediction accuracy of the screening efficiency is high by using LSSVM,GRNN and BPNN methods, the prediction results are stable and reliable, and the box dimension can be used as a single input variable to predict the screening efficiency, it has the advantages of fewer input parameters, high prediction efficiency, and high prediction accuracy, which has great potential for expanding application space and further research value.","PeriodicalId":42196,"journal":{"name":"Journal of Measurements in Engineering","volume":"53 12","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136281913","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The quality of rolling bearings determines the safety of mechanical equipment operation, and bearings with more precise structures are prone to damage due to excessive operation. Therefore, cross domain fault diagnosis of bearings has become a research hotspot. To better improve the accuracy of bearing cross domain fault diagnosis, this study proposes two models. One is a cross domain feature extraction model constructed using a mixed attention mechanism, which recognizes and extracts high-level features of bearing faults through channel attention and spatial attention mechanisms. The other is a bearing cross domain fault diagnosis model based on multi-layer perception mechanism. This model takes the feature signals collected by the attention mechanism model as input to identify and align the differences between the source and target domain features, facilitating cross domain transfer of features. The experimental results show that the mixed attention mechanism model has a maximum accuracy of 97.3 % for feature recognition of different faults, and can successfully recognize corresponding signal values. The multi-layer perception model can achieve the highest recognition accuracy of 99.5 % in bearing fault diagnosis, and it can reach a stable state when it iterates to 26, and the final stable loss value is 0.28. Therefore, the two models proposed in this study have good application value.
{"title":"Cross domain fault diagnosis method based on MLP-mixer network","authors":"Xiaodong Mao","doi":"10.21595/jme.2023.23460","DOIUrl":"https://doi.org/10.21595/jme.2023.23460","url":null,"abstract":"The quality of rolling bearings determines the safety of mechanical equipment operation, and bearings with more precise structures are prone to damage due to excessive operation. Therefore, cross domain fault diagnosis of bearings has become a research hotspot. To better improve the accuracy of bearing cross domain fault diagnosis, this study proposes two models. One is a cross domain feature extraction model constructed using a mixed attention mechanism, which recognizes and extracts high-level features of bearing faults through channel attention and spatial attention mechanisms. The other is a bearing cross domain fault diagnosis model based on multi-layer perception mechanism. This model takes the feature signals collected by the attention mechanism model as input to identify and align the differences between the source and target domain features, facilitating cross domain transfer of features. The experimental results show that the mixed attention mechanism model has a maximum accuracy of 97.3 % for feature recognition of different faults, and can successfully recognize corresponding signal values. The multi-layer perception model can achieve the highest recognition accuracy of 99.5 % in bearing fault diagnosis, and it can reach a stable state when it iterates to 26, and the final stable loss value is 0.28. Therefore, the two models proposed in this study have good application value.","PeriodicalId":42196,"journal":{"name":"Journal of Measurements in Engineering","volume":"522 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136018966","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：