Pub Date : 2022-12-22DOI: 10.1109/ISMODE56940.2022.10180918
Israt Zarin, Nagib Mahfuz, Sarnali Bashik, Ahsan Ul Islam, Mehrab Mustafy Rahman, Kazi Sazzad Hosen
Edge detection or segmentation is a rudiment innovation as it can evaluate sharpness and analyze object boundaries. That’s why it carries an influential Figure in the image processing era. However, the approach of partitioning an image into discontinuous parts is called edge detection. It defines the change of intensity associated with the image boundary. Edge detection can be done using a variety of approaches. This research proposed an innovative method to measure performance of four edge detection techniques using quality assessment metrics on satellite images and Gaussian noise-influenced satellite images. This paper comprises well-known edge detection technologies like Canny, Prewitt, Scharr, and Robert operators. Furthermore, the Image Quality Assessment (IQA) metric is an image’s essential characteristic for measuring image quality. For evaluating image quality, we mainly consider SSIM, MSE, PSNR, and RMSE. The execution of the Canny and Prewitt methods on the satellite dataset has been experimentally validated. However, Canny edge detection achieves better results when the Gaussian Noise effect is applied to the same dataset.
{"title":"Execution Examination of Distinctive Edge Detection Algorithms","authors":"Israt Zarin, Nagib Mahfuz, Sarnali Bashik, Ahsan Ul Islam, Mehrab Mustafy Rahman, Kazi Sazzad Hosen","doi":"10.1109/ISMODE56940.2022.10180918","DOIUrl":"https://doi.org/10.1109/ISMODE56940.2022.10180918","url":null,"abstract":"Edge detection or segmentation is a rudiment innovation as it can evaluate sharpness and analyze object boundaries. That’s why it carries an influential Figure in the image processing era. However, the approach of partitioning an image into discontinuous parts is called edge detection. It defines the change of intensity associated with the image boundary. Edge detection can be done using a variety of approaches. This research proposed an innovative method to measure performance of four edge detection techniques using quality assessment metrics on satellite images and Gaussian noise-influenced satellite images. This paper comprises well-known edge detection technologies like Canny, Prewitt, Scharr, and Robert operators. Furthermore, the Image Quality Assessment (IQA) metric is an image’s essential characteristic for measuring image quality. For evaluating image quality, we mainly consider SSIM, MSE, PSNR, and RMSE. The execution of the Canny and Prewitt methods on the satellite dataset has been experimentally validated. However, Canny edge detection achieves better results when the Gaussian Noise effect is applied to the same dataset.","PeriodicalId":335247,"journal":{"name":"2022 2nd International Seminar on Machine Learning, Optimization, and Data Science (ISMODE)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125768127","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 : 2022-12-22DOI: 10.1109/ISMODE56940.2022.10180957
Gagah Ghaniswara Khaesarrahman, C. Setianingsih, R. E. Saputra, Dyka Khairullah Zamhari, Erwan Maulana, R. E. Saputra, Raka Zia Insani
The current measurement of electricity consumption use a device called kWh meter that logs the total consumption of electricity, unfortunately to record the data to the electricity provider in Indonesia the employee of the provider still need to come and check the eletrical usage manually. In this paper we created a Deep Learning model based on Faster R-CNN to reads the digit from an analog electricity meter using dataset from the UFPR-AMR Dataset From the training we achieved the best model with the configurations of 90:10 for the data partition split, batch size of 3, learning rate of 0.04, and epoch of 7000 and gained results with accuracy of 99.67%, recall of 98.04%, and precision of 98.04%
{"title":"Analog Digit Electricity Meter Recognition Using Faster R-CNN","authors":"Gagah Ghaniswara Khaesarrahman, C. Setianingsih, R. E. Saputra, Dyka Khairullah Zamhari, Erwan Maulana, R. E. Saputra, Raka Zia Insani","doi":"10.1109/ISMODE56940.2022.10180957","DOIUrl":"https://doi.org/10.1109/ISMODE56940.2022.10180957","url":null,"abstract":"The current measurement of electricity consumption use a device called kWh meter that logs the total consumption of electricity, unfortunately to record the data to the electricity provider in Indonesia the employee of the provider still need to come and check the eletrical usage manually. In this paper we created a Deep Learning model based on Faster R-CNN to reads the digit from an analog electricity meter using dataset from the UFPR-AMR Dataset From the training we achieved the best model with the configurations of 90:10 for the data partition split, batch size of 3, learning rate of 0.04, and epoch of 7000 and gained results with accuracy of 99.67%, recall of 98.04%, and precision of 98.04%","PeriodicalId":335247,"journal":{"name":"2022 2nd International Seminar on Machine Learning, Optimization, and Data Science (ISMODE)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115012702","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 : 2022-12-22DOI: 10.1109/ISMODE56940.2022.10180978
Muhammad Riefky, S. W. Purnami, Nur Iriawan, W. Islamiyah
The brain is one of the most important and complex organs in the human body which controls all the activities of the organs, and daily activities, and plays a role in determining emotions. If there is interference in the brain, control of the body’s activities will be disrupted. Therefore, a tool is needed, namely the electroencephalograph (EEG) which plays a role in detecting brain wave movements. This study aims to detect patterns of brain wave movement in children with autism accompanied by seizures to monitor their activities. Using the Nihon Kohden application, Channel 5 (T5 – Cz) data as part of the T lobe brain, was used in this study. This data indicated stationary but non-linear pattern. Using the Markov switching process monitoring (MSPM) method, the ARL value for regime 1 (defined as abnormal brain wave movements) is 45.91 milliseconds which is greater than the ARL for regime 2 (defined as normal brain wave movements) is 26.91 milliseconds, so autism children have abnormal brain wave movements in the temporal lobe brain.
{"title":"Markov Switching Process Monitoring Brain Wave Movement in Autism Children","authors":"Muhammad Riefky, S. W. Purnami, Nur Iriawan, W. Islamiyah","doi":"10.1109/ISMODE56940.2022.10180978","DOIUrl":"https://doi.org/10.1109/ISMODE56940.2022.10180978","url":null,"abstract":"The brain is one of the most important and complex organs in the human body which controls all the activities of the organs, and daily activities, and plays a role in determining emotions. If there is interference in the brain, control of the body’s activities will be disrupted. Therefore, a tool is needed, namely the electroencephalograph (EEG) which plays a role in detecting brain wave movements. This study aims to detect patterns of brain wave movement in children with autism accompanied by seizures to monitor their activities. Using the Nihon Kohden application, Channel 5 (T5 – Cz) data as part of the T lobe brain, was used in this study. This data indicated stationary but non-linear pattern. Using the Markov switching process monitoring (MSPM) method, the ARL value for regime 1 (defined as abnormal brain wave movements) is 45.91 milliseconds which is greater than the ARL for regime 2 (defined as normal brain wave movements) is 26.91 milliseconds, so autism children have abnormal brain wave movements in the temporal lobe brain.","PeriodicalId":335247,"journal":{"name":"2022 2nd International Seminar on Machine Learning, Optimization, and Data Science (ISMODE)","volume":"2012 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114550389","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 : 2022-12-22DOI: 10.1109/ISMODE56940.2022.10180421
Agung Septiadi, Erwin Nashrullah, Muhammad Arief, Junanto Prihantoro, Jemie Muliadi, Fandy Harahap, Kusnanda Supriatna, Aris Suwarjono
One of the most important devices in cyber security is Intrusion Detection System (IDS). It is a device that is required to be able to monitor network traffic and detect the possibility of intrusion. Anomaly-based IDS is a type of IDS that works by detecting an anomaly in network traffic. The method that is starting to be widely used for detection is machine learning. In this work, the performance of five machine learning algorithm architectures—Decision Tree, ANN, Random Forest, SVM, and Naive Bayes—in an anomaly-based intrusion detection system will be evaluated. Two datasets—KDD Cup 1999 and UNSW-NB15—have been utilized. Before being used, data pre-processing is carried out to reduce the number of features. Our experiment results demonstrate that Random Forest surpassed other algorithms in accuracy, precision and recall on the KDD Cup 1999 dataset, while for the UNSW-NB15 dataset, SVM provides the best performance for all aspects measured.
入侵检测系统(IDS)是网络安全中最重要的设备之一。它是一种需要能够监控网络流量并检测入侵可能性的设备。基于异常的IDS是一种通过检测网络流量中的异常来工作的IDS。开始被广泛用于检测的方法是机器学习。在这项工作中，五种机器学习算法架构——决策树、人工神经网络、随机森林、支持向量机和朴素贝叶斯——在基于异常的入侵检测系统中的性能将被评估。使用了kdd Cup 1999和unsw - nb15两个数据集。在使用前对数据进行预处理，减少特征的数量。实验结果表明，在KDD Cup 1999数据集上，随机森林在准确率、精密度和召回率方面优于其他算法，而在UNSW-NB15数据集上，SVM在各方面都表现最佳。
{"title":"A Comparative Study of Five Machine Learning Algorithms for Anomaly-based IDS","authors":"Agung Septiadi, Erwin Nashrullah, Muhammad Arief, Junanto Prihantoro, Jemie Muliadi, Fandy Harahap, Kusnanda Supriatna, Aris Suwarjono","doi":"10.1109/ISMODE56940.2022.10180421","DOIUrl":"https://doi.org/10.1109/ISMODE56940.2022.10180421","url":null,"abstract":"One of the most important devices in cyber security is Intrusion Detection System (IDS). It is a device that is required to be able to monitor network traffic and detect the possibility of intrusion. Anomaly-based IDS is a type of IDS that works by detecting an anomaly in network traffic. The method that is starting to be widely used for detection is machine learning. In this work, the performance of five machine learning algorithm architectures—Decision Tree, ANN, Random Forest, SVM, and Naive Bayes—in an anomaly-based intrusion detection system will be evaluated. Two datasets—KDD Cup 1999 and UNSW-NB15—have been utilized. Before being used, data pre-processing is carried out to reduce the number of features. Our experiment results demonstrate that Random Forest surpassed other algorithms in accuracy, precision and recall on the KDD Cup 1999 dataset, while for the UNSW-NB15 dataset, SVM provides the best performance for all aspects measured.","PeriodicalId":335247,"journal":{"name":"2022 2nd International Seminar on Machine Learning, Optimization, and Data Science (ISMODE)","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123541128","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 : 2022-12-22DOI: 10.1109/ISMODE56940.2022.10180414
D. Pau, Randriatsimiovalaza, Alessandro Carra, Marco Garzola
Tiny machine learning has proven its capabilities and applicability in several research fields such as IoT and Automotive applications. The introduction of the deeply quantized neural network has been a game changer as it allowed to reduce dramatically the memory footprint. The challenge is to achieve a marginal accuracy drop low enough while quantizing 32 bits floating point neural networks. In case of mice studies, by acquiring the appropriate images per each use case, with the neural networks proposed by this work, it is possible to classify the objects inside the mice’ cages and if they drink or not. The outcomes are important to indicate the health status of the rodents. In that context, pBottleNet, pFoodNet, pCageNet have been introduced to classify the presence of the bottle, the food level and the presence of the cage while pDrinkingNet was designed to identify if the rodent was drinking when the bottle was present in the cage. The accuracies of the above cited four deeply quantized neural networks were between 95.70% and 99.9%. The entire process, from the image capture to the inference’s execution, have been deployed on microcontrollers. The design of the networks, therefore, shall respect the memory constraints of the STM32H7 and of the STM32L4 microcontrollers in which the models have been analyzed and tested. The inference times on the STM32H7 for each pico model were 1. 912ms, 12.579ms, 2. 263ms and 2. 264ms respectively.
{"title":"Deep Tiny Quantization for Fish-Eye Distorted Object Classification","authors":"D. Pau, Randriatsimiovalaza, Alessandro Carra, Marco Garzola","doi":"10.1109/ISMODE56940.2022.10180414","DOIUrl":"https://doi.org/10.1109/ISMODE56940.2022.10180414","url":null,"abstract":"Tiny machine learning has proven its capabilities and applicability in several research fields such as IoT and Automotive applications. The introduction of the deeply quantized neural network has been a game changer as it allowed to reduce dramatically the memory footprint. The challenge is to achieve a marginal accuracy drop low enough while quantizing 32 bits floating point neural networks. In case of mice studies, by acquiring the appropriate images per each use case, with the neural networks proposed by this work, it is possible to classify the objects inside the mice’ cages and if they drink or not. The outcomes are important to indicate the health status of the rodents. In that context, pBottleNet, pFoodNet, pCageNet have been introduced to classify the presence of the bottle, the food level and the presence of the cage while pDrinkingNet was designed to identify if the rodent was drinking when the bottle was present in the cage. The accuracies of the above cited four deeply quantized neural networks were between 95.70% and 99.9%. The entire process, from the image capture to the inference’s execution, have been deployed on microcontrollers. The design of the networks, therefore, shall respect the memory constraints of the STM32H7 and of the STM32L4 microcontrollers in which the models have been analyzed and tested. The inference times on the STM32H7 for each pico model were 1. 912ms, 12.579ms, 2. 263ms and 2. 264ms respectively.","PeriodicalId":335247,"journal":{"name":"2022 2nd International Seminar on Machine Learning, Optimization, and Data Science (ISMODE)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133670730","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 : 2022-12-22DOI: 10.1109/ISMODE56940.2022.10180988
Parjito, F. Ulum, K. Muludi, Z. Abidin, Risa Meidiana Alma, Permata
Indonesia has two seasons, and the potential as a producer of superior products in the plantation sector is tremendous. Coverage in the plantation sector has ornamental plant species. Ornamental plants are plants that can be used as decorations indoors or outdoors. Each form of the plant is diverse and has its charm. Some Indonesian people still do not know the types of ornamental plants, so one of the efforts is to introduce ornamental plants to the public. In this case, with conditions that are currently digital, computer applications can be used to introduce ornamental plants. Therefore, there is a technology with the Deep Learning method using Convolutional Neural Networks. Using the dataset obtained, there are 1554 images with five categories of ornamental plants divided by a ratio of 80% train data and 20% test data. Then using the Pareto principle, the train data will be divided into 80% train data and 20% data validation. After the training and testing, the accuracy results are 75% for train data and 67% for data validation. Several experiments were conducted to find the parameters that get the model with the best accuracy, namely by experimenting with the MobilenetV2 model.
{"title":"Classification of Ornamental Plants with Convolutional Neural Networks and MobileNetV2 Approach","authors":"Parjito, F. Ulum, K. Muludi, Z. Abidin, Risa Meidiana Alma, Permata","doi":"10.1109/ISMODE56940.2022.10180988","DOIUrl":"https://doi.org/10.1109/ISMODE56940.2022.10180988","url":null,"abstract":"Indonesia has two seasons, and the potential as a producer of superior products in the plantation sector is tremendous. Coverage in the plantation sector has ornamental plant species. Ornamental plants are plants that can be used as decorations indoors or outdoors. Each form of the plant is diverse and has its charm. Some Indonesian people still do not know the types of ornamental plants, so one of the efforts is to introduce ornamental plants to the public. In this case, with conditions that are currently digital, computer applications can be used to introduce ornamental plants. Therefore, there is a technology with the Deep Learning method using Convolutional Neural Networks. Using the dataset obtained, there are 1554 images with five categories of ornamental plants divided by a ratio of 80% train data and 20% test data. Then using the Pareto principle, the train data will be divided into 80% train data and 20% data validation. After the training and testing, the accuracy results are 75% for train data and 67% for data validation. Several experiments were conducted to find the parameters that get the model with the best accuracy, namely by experimenting with the MobilenetV2 model.","PeriodicalId":335247,"journal":{"name":"2022 2nd International Seminar on Machine Learning, Optimization, and Data Science (ISMODE)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116065363","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 : 2022-12-22DOI: 10.1109/ISMODE56940.2022.10180412
Fitriani, Z. Zainuddin, Syafaruddin
Limited land makes agriculture increasingly squeezed by settlements, trade, and industry, and this can be seen in unresolved human growth. The existence of hydroponic technology is a solution for farming on narrow land. Hydroponics is the cultivation of plants by utilizing water as a planting medium, so it doesn’t need to use a large area. The cultivation of hydroponic planting requires a particular method. The nutritional needs and pH of hydroponic plants must be maintained so that a nutrient control system can facilitate the controlling and monitoring of nutrients so that they remain according to to plant needs. In this research, an automatic control system for nutrition and pH was created in the Nutrient Film Technique (NFT) hydroponic model. The control system process uses a microcontroller with the Convolutional Neural Network (CNN) method. Overall the system can carry out the nutrition control process automatically without using a laptop. The system runs entirely within the microcontroller. The control system uses the CNN method with the input parameters pH, nutrition, and time as well as output the duration of the flame up, pH down, food, and water pump to reach the set target value. The results of the research that has been done show that the error value for healthy control is 3.35% and 0.98% for pH control.
{"title":"Nutrition Control System In Nutrient Film Technique (NFT) Hydroponics With Convolutional Neural Network (CNN) Method","authors":"Fitriani, Z. Zainuddin, Syafaruddin","doi":"10.1109/ISMODE56940.2022.10180412","DOIUrl":"https://doi.org/10.1109/ISMODE56940.2022.10180412","url":null,"abstract":"Limited land makes agriculture increasingly squeezed by settlements, trade, and industry, and this can be seen in unresolved human growth. The existence of hydroponic technology is a solution for farming on narrow land. Hydroponics is the cultivation of plants by utilizing water as a planting medium, so it doesn’t need to use a large area. The cultivation of hydroponic planting requires a particular method. The nutritional needs and pH of hydroponic plants must be maintained so that a nutrient control system can facilitate the controlling and monitoring of nutrients so that they remain according to to plant needs. In this research, an automatic control system for nutrition and pH was created in the Nutrient Film Technique (NFT) hydroponic model. The control system process uses a microcontroller with the Convolutional Neural Network (CNN) method. Overall the system can carry out the nutrition control process automatically without using a laptop. The system runs entirely within the microcontroller. The control system uses the CNN method with the input parameters pH, nutrition, and time as well as output the duration of the flame up, pH down, food, and water pump to reach the set target value. The results of the research that has been done show that the error value for healthy control is 3.35% and 0.98% for pH control.","PeriodicalId":335247,"journal":{"name":"2022 2nd International Seminar on Machine Learning, Optimization, and Data Science (ISMODE)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121075930","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 : 2022-12-22DOI: 10.1109/ISMODE56940.2022.10180965
K. Roy, L. M. Lopez-Ramos, B. Beferull-Lozano
Discovery of causal dependencies among time series has been tackled in the past either by using linear models, or using kernel- or deep learning-based nonlinear models, the latter ones entailing great complexity. This paper proposes a nonlinear modelling technique for multiple time series that has a complexity similar to that of linear vector autoregressive (VAR), but it can account for nonlinear interactions for each sensor variable. The modelling assumption is that the time series are generated in two steps: i) a VAR process in a latent space, and ii) a set of invertible nonlinear mappings applied component-wise, mapping each sensor variable into a latent space. Successful identification of the support of the VAR coefficients reveals the topology of the interconnected system. The proposed method enforces sparsity on the VAR coefficients and models the component-wise nonlinearities using invertible neural networks. To solve the estimation problem, a technique combining proximal gradient descent (PGD) and projected gradient descent is designed. Experiments conducted on real and synthetic data sets show that the proposed algorithm provides an improved identification of the support of the VAR coefficients, while improving also the prediction capabilities.
{"title":"Joint Learning of Topology and Invertible Nonlinearities from Multiple Time Series","authors":"K. Roy, L. M. Lopez-Ramos, B. Beferull-Lozano","doi":"10.1109/ISMODE56940.2022.10180965","DOIUrl":"https://doi.org/10.1109/ISMODE56940.2022.10180965","url":null,"abstract":"Discovery of causal dependencies among time series has been tackled in the past either by using linear models, or using kernel- or deep learning-based nonlinear models, the latter ones entailing great complexity. This paper proposes a nonlinear modelling technique for multiple time series that has a complexity similar to that of linear vector autoregressive (VAR), but it can account for nonlinear interactions for each sensor variable. The modelling assumption is that the time series are generated in two steps: i) a VAR process in a latent space, and ii) a set of invertible nonlinear mappings applied component-wise, mapping each sensor variable into a latent space. Successful identification of the support of the VAR coefficients reveals the topology of the interconnected system. The proposed method enforces sparsity on the VAR coefficients and models the component-wise nonlinearities using invertible neural networks. To solve the estimation problem, a technique combining proximal gradient descent (PGD) and projected gradient descent is designed. Experiments conducted on real and synthetic data sets show that the proposed algorithm provides an improved identification of the support of the VAR coefficients, while improving also the prediction capabilities.","PeriodicalId":335247,"journal":{"name":"2022 2nd International Seminar on Machine Learning, Optimization, and Data Science (ISMODE)","volume":"130 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129586365","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 : 2022-12-22DOI: 10.1109/ISMODE56940.2022.10180980
Ahmad Masry Bin Zainol, Nurul Rawaida Ain Burhani
This study identifies the most influential factors of rotating machines or induction motors that are commonly used in the industry such as oil and gas industry. The data gathered in the industry, especially for High Voltage Induction Motors (HVIM) in the plant commonly does not fully utilize to detect the most influential maintenance factor. This required advance and latest technology to convert the data into more usable and estimate the probability of failure efficiently. A predictive maintenance solution can be used to solve data-driven problems like the complexity to compute, taking longer time to analyze, and difficulty in using big data features. Logistic Regression Analysis (LRA) methods can be used to determine the most influencing factor (MIF) of maintenance of high voltage induction motors. The MIF of HVIM maintenance obtained in the study is vibration, temperature, and power factor with an R2 value of 0.9993888. It is shown that the R2 value was high and significant. The HVIM maintenance prediction based on MIF is suited for industrial applications due to its fitness in using industrial data for analysis. As a result, it suited the existing industrial maintenance for predictive purposes and can be further developed in the future.
{"title":"Detecting Most Influential Parameters in High Voltage Induction Motor Failure using Logistic Regression Analysis","authors":"Ahmad Masry Bin Zainol, Nurul Rawaida Ain Burhani","doi":"10.1109/ISMODE56940.2022.10180980","DOIUrl":"https://doi.org/10.1109/ISMODE56940.2022.10180980","url":null,"abstract":"This study identifies the most influential factors of rotating machines or induction motors that are commonly used in the industry such as oil and gas industry. The data gathered in the industry, especially for High Voltage Induction Motors (HVIM) in the plant commonly does not fully utilize to detect the most influential maintenance factor. This required advance and latest technology to convert the data into more usable and estimate the probability of failure efficiently. A predictive maintenance solution can be used to solve data-driven problems like the complexity to compute, taking longer time to analyze, and difficulty in using big data features. Logistic Regression Analysis (LRA) methods can be used to determine the most influencing factor (MIF) of maintenance of high voltage induction motors. The MIF of HVIM maintenance obtained in the study is vibration, temperature, and power factor with an R2 value of 0.9993888. It is shown that the R2 value was high and significant. The HVIM maintenance prediction based on MIF is suited for industrial applications due to its fitness in using industrial data for analysis. As a result, it suited the existing industrial maintenance for predictive purposes and can be further developed in the future.","PeriodicalId":335247,"journal":{"name":"2022 2nd International Seminar on Machine Learning, Optimization, and Data Science (ISMODE)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129474218","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 : 2022-12-22DOI: 10.1109/ISMODE56940.2022.10180426
M. Rais, M. Djalal, V. A. Tandirerung, Rosihan Aminuddin, Irwan Syarif, Rosmiati
The stability of a generator has an important function in the continuity of electricity production. A multimachine electric power system has many generators connected. The Sulselrabar system consists of several interconnected power plants. Proper coordination between generating centres can support the performance of the electric power system, especially when disturbances can disrupt system stability. Sudden load changes are one of the electric power system’s disturbances, which can impact the generator’s stability. In generator operation, the controller is assigned to the generator excitation equipment. However, the dynamics of the electric power system continue to evolve, causing the generator excitation equipment to reach its limit when a large disturbance occurs. Control equipment such as PID and Power System Stabilizer (PSS) produce good performance on the system. The use of these controls requires optimal coordination in finding the right parameters and locations. In this study, an approach is proposed in coordinating PID and PSS controllers for multi-engine generators in the Sulselrabar system. The Ant Colony optimization (ACO) algorithm is a smart algorithm that adopts the behavior of ants in finding food sources. ACO is used for precise PID-PSS parameter optimization. A case study was used in Sengkang generators that were subjected to load change disturbances. From the simulation results, it is obtained that the performance of the Sengkang generator is optimal in terms of speed overshoot response and minimum rotor angle. The application of PID-PSS also increases the damping system so that the oscillations generated due to disturbances can be properly attenuated.
{"title":"Coordination PID-PSS Control Based on Ant Colony optimization In Sulselrabar System","authors":"M. Rais, M. Djalal, V. A. Tandirerung, Rosihan Aminuddin, Irwan Syarif, Rosmiati","doi":"10.1109/ISMODE56940.2022.10180426","DOIUrl":"https://doi.org/10.1109/ISMODE56940.2022.10180426","url":null,"abstract":"The stability of a generator has an important function in the continuity of electricity production. A multimachine electric power system has many generators connected. The Sulselrabar system consists of several interconnected power plants. Proper coordination between generating centres can support the performance of the electric power system, especially when disturbances can disrupt system stability. Sudden load changes are one of the electric power system’s disturbances, which can impact the generator’s stability. In generator operation, the controller is assigned to the generator excitation equipment. However, the dynamics of the electric power system continue to evolve, causing the generator excitation equipment to reach its limit when a large disturbance occurs. Control equipment such as PID and Power System Stabilizer (PSS) produce good performance on the system. The use of these controls requires optimal coordination in finding the right parameters and locations. In this study, an approach is proposed in coordinating PID and PSS controllers for multi-engine generators in the Sulselrabar system. The Ant Colony optimization (ACO) algorithm is a smart algorithm that adopts the behavior of ants in finding food sources. ACO is used for precise PID-PSS parameter optimization. A case study was used in Sengkang generators that were subjected to load change disturbances. From the simulation results, it is obtained that the performance of the Sengkang generator is optimal in terms of speed overshoot response and minimum rotor angle. The application of PID-PSS also increases the damping system so that the oscillations generated due to disturbances can be properly attenuated.","PeriodicalId":335247,"journal":{"name":"2022 2nd International Seminar on Machine Learning, Optimization, and Data Science (ISMODE)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124391110","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：