Pub Date : 2023-05-26DOI: 10.1109/INCET57972.2023.10170437
Xiaoli Li, Xiaoye Pan
The research of commutator surface defect detection algorithm based on genetic algorithm is to use genetic algorithm to detect commutator surface defects. The main purpose of this study is to find a method to detect commutator defects. GA will be used to detect defects on the commutator surface by using data obtained from scanning electron microscopy (SEM). Based on this, we can easily detect and measure the defects on the surface of the commutator. This research works with genetic algorithms, which are developed to solve problems related to computer vision and image processing. To solve these problems, we use three basic rules: mutation, crossover and selection. This study can be used as an effective tool to analyze and repair the defective parts of motors, generators, transformers and so on.
{"title":"Commutator Surface Defect Detection Algorithm based on Genetic Algorithm","authors":"Xiaoli Li, Xiaoye Pan","doi":"10.1109/INCET57972.2023.10170437","DOIUrl":"https://doi.org/10.1109/INCET57972.2023.10170437","url":null,"abstract":"The research of commutator surface defect detection algorithm based on genetic algorithm is to use genetic algorithm to detect commutator surface defects. The main purpose of this study is to find a method to detect commutator defects. GA will be used to detect defects on the commutator surface by using data obtained from scanning electron microscopy (SEM). Based on this, we can easily detect and measure the defects on the surface of the commutator. This research works with genetic algorithms, which are developed to solve problems related to computer vision and image processing. To solve these problems, we use three basic rules: mutation, crossover and selection. This study can be used as an effective tool to analyze and repair the defective parts of motors, generators, transformers and so on.","PeriodicalId":403008,"journal":{"name":"2023 4th International Conference for Emerging Technology (INCET)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127025106","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 : 2023-05-26DOI: 10.1109/INCET57972.2023.10170142
Yingjie Li, Yitian Wang, Huaici Zhao
Aerial images are often plagued by background interference, and small targets with indistinct features, leading to low accuracy, high false detection rates, and high miss detection rates. To address these challenges, a small target detection algorithm based on YOLOv5, Coordinate-attention and Bidirectional-feature-pyramid-network YOLOv5 (CB-YOLOv5), is proposed in this paper. Considering the small number of pixels occupied by small targets and their indistinct features, a fourth target detection layer is added by concatenating the feature map from quadruple down-sampling during feature extraction with the feature map output from 8-fold up-sampling during feature fusion. Additionally, a coordinate attention mechanism is introduced during the feature extraction stage to improve small target localization and enhance detection accuracy. Finally, the original Path Aggregation Networks (PANet) structure is replaced with a weighted Bidirectional Feature Pyramid Network (BiFPN) structure during the feature fusion stage to improve the network’s ability to fuse feature maps of different scales. The simulation results demonstrate that the CB-YOLOv5 improves mAP50 by 9.4%, mAP75 by 9.7%, and mAP50:95 by 7.8% compared to the original YOLOv5s model. Thus, the effectiveness of the CB-YOLOv5 algorithm for detecting small targets in aerial images is validated.
{"title":"CB-YOLOv5 Algorithm for Small Target Detection in Aerial Images","authors":"Yingjie Li, Yitian Wang, Huaici Zhao","doi":"10.1109/INCET57972.2023.10170142","DOIUrl":"https://doi.org/10.1109/INCET57972.2023.10170142","url":null,"abstract":"Aerial images are often plagued by background interference, and small targets with indistinct features, leading to low accuracy, high false detection rates, and high miss detection rates. To address these challenges, a small target detection algorithm based on YOLOv5, Coordinate-attention and Bidirectional-feature-pyramid-network YOLOv5 (CB-YOLOv5), is proposed in this paper. Considering the small number of pixels occupied by small targets and their indistinct features, a fourth target detection layer is added by concatenating the feature map from quadruple down-sampling during feature extraction with the feature map output from 8-fold up-sampling during feature fusion. Additionally, a coordinate attention mechanism is introduced during the feature extraction stage to improve small target localization and enhance detection accuracy. Finally, the original Path Aggregation Networks (PANet) structure is replaced with a weighted Bidirectional Feature Pyramid Network (BiFPN) structure during the feature fusion stage to improve the network’s ability to fuse feature maps of different scales. The simulation results demonstrate that the CB-YOLOv5 improves mAP50 by 9.4%, mAP75 by 9.7%, and mAP50:95 by 7.8% compared to the original YOLOv5s model. Thus, the effectiveness of the CB-YOLOv5 algorithm for detecting small targets in aerial images is validated.","PeriodicalId":403008,"journal":{"name":"2023 4th International Conference for Emerging Technology (INCET)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127245390","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 : 2023-05-26DOI: 10.1109/INCET57972.2023.10170681
A. Shinde, Prajyot Bhoir, S. Shinde, Bushra Shaikh
It is now more important than ever to take action to lessen the negative consequences of private vehicles. If successfully implemented, mass transit is the ideal option, however because of its lack of door-to-door service, lengthier fixed routes, and unreliable timetable, many people do not appreciate it. Therefore, new facilities or services should be created to offer users a comfortable and dependable service and to lessen potentially dangerous environmental effects like pollution, congestion, etc. One of the cutting-edge technologies that is being used all over the world is ride sharing, in which users who have the same origin-destination and journey time are matched and share the transport. To help in implementation of ride sharing, a mobile application is being developed using machine learning techniques such as the Naïve Bayes algorithms to match users based on their travel preferences and habits. The app will provide a more personalized and convenient service to the users, ensuring that they are matched with the most suitable carpool partners.
{"title":"An Efficient Ridesharing Model using Machine Learning Based on Riders Reviews","authors":"A. Shinde, Prajyot Bhoir, S. Shinde, Bushra Shaikh","doi":"10.1109/INCET57972.2023.10170681","DOIUrl":"https://doi.org/10.1109/INCET57972.2023.10170681","url":null,"abstract":"It is now more important than ever to take action to lessen the negative consequences of private vehicles. If successfully implemented, mass transit is the ideal option, however because of its lack of door-to-door service, lengthier fixed routes, and unreliable timetable, many people do not appreciate it. Therefore, new facilities or services should be created to offer users a comfortable and dependable service and to lessen potentially dangerous environmental effects like pollution, congestion, etc. One of the cutting-edge technologies that is being used all over the world is ride sharing, in which users who have the same origin-destination and journey time are matched and share the transport. To help in implementation of ride sharing, a mobile application is being developed using machine learning techniques such as the Naïve Bayes algorithms to match users based on their travel preferences and habits. The app will provide a more personalized and convenient service to the users, ensuring that they are matched with the most suitable carpool partners.","PeriodicalId":403008,"journal":{"name":"2023 4th International Conference for Emerging Technology (INCET)","volume":"136 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127350155","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 : 2023-05-26DOI: 10.1109/INCET57972.2023.10170270
M. Bhattacharya, D. Datta
The research work in this manuscript is focused towards extraction of rules from decision tree classifier to predict the status of a patient suffering diabetic. Basic approach of machine learning algorithm to classify diabetic condition of a patient depends on various features such as glucose, blood pressure, insulin, skin thickness, body mass index (BMI), diabetic pedigree function and age. Decision trees are easily interpretable machine learning models as classifiers whose predictive accuracy is low. However, in comparison random forest machine learning tree ensembles show high predictive accuracy while being regarded as black-box models. In this work, we have developed an algorithm to extract decision rules from the corresponding tree in the form of human readable format (IF antecedent, THEN consequent). We have also provided logistic regression model and tree structure of random forest model to classify the diabetic condition. Experimental results of 768 women samples from PIMA Indian datasets of diabetic proves that the proposed rule extraction methodology outperform similar recently developed methods in terms of human comprehension and also limits the number of antecedents in the retained rules, while preserving the same level of accuracy. Performance of all machine learning classifier models are measured in terms of various metrics such as recall, precision, accuracy and F1-score via confusion matrix.
{"title":"Diabetes Prediction using Logistic Regression and Rule Extraction from Decision Tree and Random Forest Classifiers","authors":"M. Bhattacharya, D. Datta","doi":"10.1109/INCET57972.2023.10170270","DOIUrl":"https://doi.org/10.1109/INCET57972.2023.10170270","url":null,"abstract":"The research work in this manuscript is focused towards extraction of rules from decision tree classifier to predict the status of a patient suffering diabetic. Basic approach of machine learning algorithm to classify diabetic condition of a patient depends on various features such as glucose, blood pressure, insulin, skin thickness, body mass index (BMI), diabetic pedigree function and age. Decision trees are easily interpretable machine learning models as classifiers whose predictive accuracy is low. However, in comparison random forest machine learning tree ensembles show high predictive accuracy while being regarded as black-box models. In this work, we have developed an algorithm to extract decision rules from the corresponding tree in the form of human readable format (IF antecedent, THEN consequent). We have also provided logistic regression model and tree structure of random forest model to classify the diabetic condition. Experimental results of 768 women samples from PIMA Indian datasets of diabetic proves that the proposed rule extraction methodology outperform similar recently developed methods in terms of human comprehension and also limits the number of antecedents in the retained rules, while preserving the same level of accuracy. Performance of all machine learning classifier models are measured in terms of various metrics such as recall, precision, accuracy and F1-score via confusion matrix.","PeriodicalId":403008,"journal":{"name":"2023 4th International Conference for Emerging Technology (INCET)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129917067","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 : 2023-05-26DOI: 10.1109/INCET57972.2023.10170726
Priyanka Bhatele, Manas Dalvi, M. Kulkarni, Tejas Mali, Manthan Manalwar, Aman Manakshe
Rapid expansion in the quantity and variety of solid and hazardous waste as a consequence of continued economic development, urbanization, and industrialization poses a growing challenge for national and municipal governments to ensure efficient and long-term waste management. According to estimates, the total amount of municipal solid trash produced worldwide in 2006 was 2.02 billion tonnes, a rise of 7% annually since 2003. (Global Waste Management Market Report 2007). To reduce the risk to the patient and public health and safety, as well as an environmental hazard, waste management, transportation, and disposal must be carefully handled. There is currently no system in place for households to separate dry, moist, and metallic garbage. In order to send household waste directly for processing, an automated waste segregator is suggested in this study, which is an affordable, simple-to-use alternative. It is intended to separate the garbage into metallic, wet, and dry waste. The automated waste segregator uses a moisture sensor to discriminate between wet and dry trash and an inductive proximity sensor to detect metallic objects. According to experimental findings, the automated waste segregator has been effectively used to accomplish the classification of waste into dry, wet, and metallic waste.
{"title":"Smart Waste Segregation Using IoT","authors":"Priyanka Bhatele, Manas Dalvi, M. Kulkarni, Tejas Mali, Manthan Manalwar, Aman Manakshe","doi":"10.1109/INCET57972.2023.10170726","DOIUrl":"https://doi.org/10.1109/INCET57972.2023.10170726","url":null,"abstract":"Rapid expansion in the quantity and variety of solid and hazardous waste as a consequence of continued economic development, urbanization, and industrialization poses a growing challenge for national and municipal governments to ensure efficient and long-term waste management. According to estimates, the total amount of municipal solid trash produced worldwide in 2006 was 2.02 billion tonnes, a rise of 7% annually since 2003. (Global Waste Management Market Report 2007). To reduce the risk to the patient and public health and safety, as well as an environmental hazard, waste management, transportation, and disposal must be carefully handled. There is currently no system in place for households to separate dry, moist, and metallic garbage. In order to send household waste directly for processing, an automated waste segregator is suggested in this study, which is an affordable, simple-to-use alternative. It is intended to separate the garbage into metallic, wet, and dry waste. The automated waste segregator uses a moisture sensor to discriminate between wet and dry trash and an inductive proximity sensor to detect metallic objects. According to experimental findings, the automated waste segregator has been effectively used to accomplish the classification of waste into dry, wet, and metallic waste.","PeriodicalId":403008,"journal":{"name":"2023 4th International Conference for Emerging Technology (INCET)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129976611","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 : 2023-05-26DOI: 10.1109/INCET57972.2023.10169986
Huang Chao, D. Liang, Zhang Cheng, Rongtao Liao, Guo Yue, Dangdang Dai
Based on the improved k-means algorithm, this paper studies the identification of abnormal feature data of power equipment. Clustering according to the daily load curve can make a fine distinction between users. An accurate load pattern recognition model can also help grid workers to distinguish the load patterns of users, help power companies find their power laws, and provide a theoretical basis for load analysis, forecasting, decision-making and other work of the power system.
{"title":"Identification Method of Abnormal Characteristic Data of Power Equipment based on Improved K-Means Algorithm","authors":"Huang Chao, D. Liang, Zhang Cheng, Rongtao Liao, Guo Yue, Dangdang Dai","doi":"10.1109/INCET57972.2023.10169986","DOIUrl":"https://doi.org/10.1109/INCET57972.2023.10169986","url":null,"abstract":"Based on the improved k-means algorithm, this paper studies the identification of abnormal feature data of power equipment. Clustering according to the daily load curve can make a fine distinction between users. An accurate load pattern recognition model can also help grid workers to distinguish the load patterns of users, help power companies find their power laws, and provide a theoretical basis for load analysis, forecasting, decision-making and other work of the power system.","PeriodicalId":403008,"journal":{"name":"2023 4th International Conference for Emerging Technology (INCET)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129099972","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 : 2023-05-26DOI: 10.1109/INCET57972.2023.10170644
P. R., Srinag R, N. Rani
Smart farming techniques involve the use of plant identification and classification. Deep learning can be particularly useful for classifying low-light images because it can impulsively learn features from the data that can be relevant for classification. This is especially important in low light conditions where the image may be noisy or contain artefacts that are not relevant to the task. In the experiment, the plant seedlings and weedlings dataset consisting of low light images are subjected to a deep-learning model. Low-light images tend to have poor image quality due to the limited amount of available light. This results in a very low signal-to-noise ratio, making extracting beneficial information from the images extremely ambiguous. In the proposed work, a deep learning XceptionNet model is utilized to perform classification of plants using seedlings and weedlings that provides performance yielding an accuracy of 94.13% with 25 epochs.
{"title":"Classification of Plant Species based Seedlings and Weedlings in Low Lightening Conditions using Deep Convolution Neural Network","authors":"P. R., Srinag R, N. Rani","doi":"10.1109/INCET57972.2023.10170644","DOIUrl":"https://doi.org/10.1109/INCET57972.2023.10170644","url":null,"abstract":"Smart farming techniques involve the use of plant identification and classification. Deep learning can be particularly useful for classifying low-light images because it can impulsively learn features from the data that can be relevant for classification. This is especially important in low light conditions where the image may be noisy or contain artefacts that are not relevant to the task. In the experiment, the plant seedlings and weedlings dataset consisting of low light images are subjected to a deep-learning model. Low-light images tend to have poor image quality due to the limited amount of available light. This results in a very low signal-to-noise ratio, making extracting beneficial information from the images extremely ambiguous. In the proposed work, a deep learning XceptionNet model is utilized to perform classification of plants using seedlings and weedlings that provides performance yielding an accuracy of 94.13% with 25 epochs.","PeriodicalId":403008,"journal":{"name":"2023 4th International Conference for Emerging Technology (INCET)","volume":"220 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132372774","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 : 2023-05-26DOI: 10.1109/INCET57972.2023.10170686
R. Gound, B. Sundaram, S. B. V., Peerzada Anzar Azmat, Malik Najeeb Ul Habib, Avni Garg
Diabetic Retinopathy (DR) is the main cause of blindness in working-age adults around the world. Early detection and treatment of DR are critical for preventing vision loss. Image segmentation is a critical step in automated DR detection. UNET is a well-known convolutional neural network design for image segmentation. The typical UNET architecture, on the other hand, may not necessarily be appropriate for DR detection. This study introduces DRS UNET, an unique architecture specifically built for DR detection. DRS UNET incorporates residual blocks and attention mechanisms to improve feature extraction and segmentation performance. The proposed model is trained and tested using publically available datasets, yielding cutting-edge results.
{"title":"DRS-UNET: A Deep Learning Approach for Diabetic Retinopathy Detection and Segmentation from Fundus Images","authors":"R. Gound, B. Sundaram, S. B. V., Peerzada Anzar Azmat, Malik Najeeb Ul Habib, Avni Garg","doi":"10.1109/INCET57972.2023.10170686","DOIUrl":"https://doi.org/10.1109/INCET57972.2023.10170686","url":null,"abstract":"Diabetic Retinopathy (DR) is the main cause of blindness in working-age adults around the world. Early detection and treatment of DR are critical for preventing vision loss. Image segmentation is a critical step in automated DR detection. UNET is a well-known convolutional neural network design for image segmentation. The typical UNET architecture, on the other hand, may not necessarily be appropriate for DR detection. This study introduces DRS UNET, an unique architecture specifically built for DR detection. DRS UNET incorporates residual blocks and attention mechanisms to improve feature extraction and segmentation performance. The proposed model is trained and tested using publically available datasets, yielding cutting-edge results.","PeriodicalId":403008,"journal":{"name":"2023 4th International Conference for Emerging Technology (INCET)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127905995","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 : 2023-05-26DOI: 10.1109/INCET57972.2023.10170590
Guruprasad B, M. Veena, Usha Rani C M, Shalini M S
Limit of Detection (LOD) is the key parameter in chemical sensing instruments. The application of international union of pure and applied chemistry (IUPAC) recommended for univarient LOD estimations in the nonlinear semiconductor gas sensor. This methodology involves linearity, homoscedasticity and normality. This paper focuses on the analysis of LOD using omnicant instrument with selected Volatile Organic Compound (VOC) and the effect of flow rate of target VOC on detection sensitivity. For the experiment, acetone asVOC is selected as target gas, Polyvinylpyrrolidone (PVP) and honey are taken as coating materials on the fabricated sensor.Acetone, toluene, and isopropyl alcohol is considered as VOC which cause severe health hazardous to human life. The study about the concentration of acetone in the breathing environment is the necessity need, therefore acetone is selected for the study. This research work is targeted to check the effect of coating thickness and flow rate on the limit of detection of the testing instrument. As Honey and PVP are viscoelastic in nature, they exhibit more adhesive property. Hence these two materials are selected as coating materials for better sticking process. On the fabricated cantilever structure, PVP and honey are coated with a thickness of 600 nm, 900 nm and 1200 nm. The selected materials are tested against target gas acetone. From the LOD response of PVP and honey it is observed that, PVP shows a good sensitivity for lower thickness of coating whereas honey exhibits more sensitivity for higher thickness of coating .The effect on flow rate of the target VOCs viz., ethanol, toluene, acetone and isopropyl alcohol on detection sensitivity is studied and analyzed that increase in the flow rate of the target gas increases the detection sensitivity of the sensor.
{"title":"Estimation of the Limit of Detection and effect of flow rate on micro cantilever sensor","authors":"Guruprasad B, M. Veena, Usha Rani C M, Shalini M S","doi":"10.1109/INCET57972.2023.10170590","DOIUrl":"https://doi.org/10.1109/INCET57972.2023.10170590","url":null,"abstract":"Limit of Detection (LOD) is the key parameter in chemical sensing instruments. The application of international union of pure and applied chemistry (IUPAC) recommended for univarient LOD estimations in the nonlinear semiconductor gas sensor. This methodology involves linearity, homoscedasticity and normality. This paper focuses on the analysis of LOD using omnicant instrument with selected Volatile Organic Compound (VOC) and the effect of flow rate of target VOC on detection sensitivity. For the experiment, acetone asVOC is selected as target gas, Polyvinylpyrrolidone (PVP) and honey are taken as coating materials on the fabricated sensor.Acetone, toluene, and isopropyl alcohol is considered as VOC which cause severe health hazardous to human life. The study about the concentration of acetone in the breathing environment is the necessity need, therefore acetone is selected for the study. This research work is targeted to check the effect of coating thickness and flow rate on the limit of detection of the testing instrument. As Honey and PVP are viscoelastic in nature, they exhibit more adhesive property. Hence these two materials are selected as coating materials for better sticking process. On the fabricated cantilever structure, PVP and honey are coated with a thickness of 600 nm, 900 nm and 1200 nm. The selected materials are tested against target gas acetone. From the LOD response of PVP and honey it is observed that, PVP shows a good sensitivity for lower thickness of coating whereas honey exhibits more sensitivity for higher thickness of coating .The effect on flow rate of the target VOCs viz., ethanol, toluene, acetone and isopropyl alcohol on detection sensitivity is studied and analyzed that increase in the flow rate of the target gas increases the detection sensitivity of the sensor.","PeriodicalId":403008,"journal":{"name":"2023 4th International Conference for Emerging Technology (INCET)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121426342","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 : 2023-05-26DOI: 10.1109/INCET57972.2023.10170185
Mahesh Y. Pawar, Swarupanand Sewalkar, Ageda Guerra
Resolver is a widely used in the feedback loop of the PM traction drive to find exact rotary position of the permanent magnet. In real systems, position error is caused by various factors such as amplitude imbalance, imperfect quadrature, inductive harmonics, reference phase shift, excitation signal distortion or other disturbance signals. This has influence on motor torque production. So, it is crucial to monitor resolver performance so that failed sensor can be easily replaced. This also benefits supply chain to keep the parts ready.This paper demonstrates monitoring the health of the resolver sensor using a data driven approach. The algorithm developed is not only capable of classifying faulty/ healthy resolver, but it can also show the amount of degradation in the resolver sensor. The state-of-the-art developed neural network model is trained on the robust database covering all possible resolver degradations, partial and complete failures. This model is developed on a complete synthetic data tapped from the Simulink model and it is further optimized for the accuracy and size. The algorithm was initially tested on the standalone open-loop resolver model which later extended for the closed-loop version. It also supports commanded mode of prognostics which can detect and classify possible harness faults of the resolver sensor. The proposed algorithm has shown high confidence when it is tested offline on the actual hardware data.
{"title":"Position Sensor Fault Prognostic using Data Driven Approach","authors":"Mahesh Y. Pawar, Swarupanand Sewalkar, Ageda Guerra","doi":"10.1109/INCET57972.2023.10170185","DOIUrl":"https://doi.org/10.1109/INCET57972.2023.10170185","url":null,"abstract":"Resolver is a widely used in the feedback loop of the PM traction drive to find exact rotary position of the permanent magnet. In real systems, position error is caused by various factors such as amplitude imbalance, imperfect quadrature, inductive harmonics, reference phase shift, excitation signal distortion or other disturbance signals. This has influence on motor torque production. So, it is crucial to monitor resolver performance so that failed sensor can be easily replaced. This also benefits supply chain to keep the parts ready.This paper demonstrates monitoring the health of the resolver sensor using a data driven approach. The algorithm developed is not only capable of classifying faulty/ healthy resolver, but it can also show the amount of degradation in the resolver sensor. The state-of-the-art developed neural network model is trained on the robust database covering all possible resolver degradations, partial and complete failures. This model is developed on a complete synthetic data tapped from the Simulink model and it is further optimized for the accuracy and size. The algorithm was initially tested on the standalone open-loop resolver model which later extended for the closed-loop version. It also supports commanded mode of prognostics which can detect and classify possible harness faults of the resolver sensor. The proposed algorithm has shown high confidence when it is tested offline on the actual hardware data.","PeriodicalId":403008,"journal":{"name":"2023 4th International Conference for Emerging Technology (INCET)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126396645","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：