Pub Date : 2021-07-13DOI: 10.1109/ICIAS49414.2021.9642508
Noor Azmiya Bt Sirajun Noor, I. Elamvazuthi, N. Yahya
Diabetes Mellitus (DM) is one of the most prevalent diseases in the world today which is associated by having high glucose levels in the body either due to inadequate production of insulin or the body cell’s not responding towards the produced insulin. Data mining and machine learning techniques can be extremely useful in classification of DM considering the need to have a shift from current traditional methods which use sharp needles to draw blood towards a non - invasive method. The objective of this study is to perform DM classification using various machine learning algorithms. In this paper, individual classifiers such as Support Vector Machine, Naïve Bayes, Bayes Net, Decision Stump, k - Nearest Neighbors, Logistic Regression, Multilayer Perceptron and Decision Tree are experimented. Apart from that, ensemble methods such as bagging, boosting, hybrid classifier using combinations of Random Forest with other base classifiers and ensemble algorithm which is the Random Forest has also been studied. Proposed DM classification model is chosen based on an optimized model reflected by their accuracy and performance of the model. In this research, it was found that performance of ensemble method using hybrid classifier of Random Forest - Bayes Net model has proven to be the best DM classification model with an accuracy of 83.91% and AUC of 0.904 using the Pima Indian Diabetes Dataset (PIDD).
{"title":"Classification of Diabetes Mellitus using Ensemble Algorithms","authors":"Noor Azmiya Bt Sirajun Noor, I. Elamvazuthi, N. Yahya","doi":"10.1109/ICIAS49414.2021.9642508","DOIUrl":"https://doi.org/10.1109/ICIAS49414.2021.9642508","url":null,"abstract":"Diabetes Mellitus (DM) is one of the most prevalent diseases in the world today which is associated by having high glucose levels in the body either due to inadequate production of insulin or the body cell’s not responding towards the produced insulin. Data mining and machine learning techniques can be extremely useful in classification of DM considering the need to have a shift from current traditional methods which use sharp needles to draw blood towards a non - invasive method. The objective of this study is to perform DM classification using various machine learning algorithms. In this paper, individual classifiers such as Support Vector Machine, Naïve Bayes, Bayes Net, Decision Stump, k - Nearest Neighbors, Logistic Regression, Multilayer Perceptron and Decision Tree are experimented. Apart from that, ensemble methods such as bagging, boosting, hybrid classifier using combinations of Random Forest with other base classifiers and ensemble algorithm which is the Random Forest has also been studied. Proposed DM classification model is chosen based on an optimized model reflected by their accuracy and performance of the model. In this research, it was found that performance of ensemble method using hybrid classifier of Random Forest - Bayes Net model has proven to be the best DM classification model with an accuracy of 83.91% and AUC of 0.904 using the Pima Indian Diabetes Dataset (PIDD).","PeriodicalId":212635,"journal":{"name":"2020 8th International Conference on Intelligent and Advanced Systems (ICIAS)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128756559","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-13DOI: 10.1109/ICIAS49414.2021.9642631
N. Gurusamy, I. Elamvazuthi, N. Yahya
Design analysis has become an essential tool to ensure the outcome of the biomechanical harvesting device met the expected goal. Biomechanical energy harvesting has become a prominent research topic to substitute the huge and heavy commercialized batteries. Hence, a lot of limitations and challenges faced during the design stage of harvesting device. The major considerations are on the increased metabolic rate to carry the device and the electrical power harvested in given duration. The additional effort needed to carry the device at different joint locations is essential in ergonomics and it’s a key factor to design wearable device. Analysis was carried out to derive the scope of human’s metabolic rate with the motion speed and walking duration. This paper discusses the outcome which shows that longer duration with higher speed and lower device mass is essential for better performance.
{"title":"Design Analysis of Energy Harvester: The Impact of Speed of Motion and Mass of Device","authors":"N. Gurusamy, I. Elamvazuthi, N. Yahya","doi":"10.1109/ICIAS49414.2021.9642631","DOIUrl":"https://doi.org/10.1109/ICIAS49414.2021.9642631","url":null,"abstract":"Design analysis has become an essential tool to ensure the outcome of the biomechanical harvesting device met the expected goal. Biomechanical energy harvesting has become a prominent research topic to substitute the huge and heavy commercialized batteries. Hence, a lot of limitations and challenges faced during the design stage of harvesting device. The major considerations are on the increased metabolic rate to carry the device and the electrical power harvested in given duration. The additional effort needed to carry the device at different joint locations is essential in ergonomics and it’s a key factor to design wearable device. Analysis was carried out to derive the scope of human’s metabolic rate with the motion speed and walking duration. This paper discusses the outcome which shows that longer duration with higher speed and lower device mass is essential for better performance.","PeriodicalId":212635,"journal":{"name":"2020 8th International Conference on Intelligent and Advanced Systems (ICIAS)","volume":"190 5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123237209","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-13DOI: 10.1109/ICIAS49414.2021.9642587
R. Kannan, Sundar Ramakrishnan, K. Porkumaran, Srinivasan Prabakar, R. Karthikeyan
In this paper Electro Magnetic Pulse (EMP) which is used to destroy the electrical and electronic equipment operation in defence is proposed. The EMP is a small burst of electro magnetism. It’s a highly disruptive signal which has a property of high magnetic transients. A nuclear weapon actually generate an EMP regardless of altitude when detonated, however the effects of the EMP only becomes useful at high altitudes where the energy released is not so rapidly dispersed as a surface discharged. Temporary or permanent damages may occur to unshielded electronics devices due to EMP. This EMP also causes power failure in all unshielded devices and disabling them. The only way to suppress the EMP is to use faraday cage, but faraday cage alone has some capacity. EMP operated at 400 kV, 10-12 MHz, which can detect and deactivate electronics devices over a range of 10-15 meters, is designed for defence purpose.
{"title":"Modeling of Scattering the Electromagnetic Pulses by High Voltage Converters","authors":"R. Kannan, Sundar Ramakrishnan, K. Porkumaran, Srinivasan Prabakar, R. Karthikeyan","doi":"10.1109/ICIAS49414.2021.9642587","DOIUrl":"https://doi.org/10.1109/ICIAS49414.2021.9642587","url":null,"abstract":"In this paper Electro Magnetic Pulse (EMP) which is used to destroy the electrical and electronic equipment operation in defence is proposed. The EMP is a small burst of electro magnetism. It’s a highly disruptive signal which has a property of high magnetic transients. A nuclear weapon actually generate an EMP regardless of altitude when detonated, however the effects of the EMP only becomes useful at high altitudes where the energy released is not so rapidly dispersed as a surface discharged. Temporary or permanent damages may occur to unshielded electronics devices due to EMP. This EMP also causes power failure in all unshielded devices and disabling them. The only way to suppress the EMP is to use faraday cage, but faraday cage alone has some capacity. EMP operated at 400 kV, 10-12 MHz, which can detect and deactivate electronics devices over a range of 10-15 meters, is designed for defence purpose.","PeriodicalId":212635,"journal":{"name":"2020 8th International Conference on Intelligent and Advanced Systems (ICIAS)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125319348","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-13DOI: 10.1109/ICIAS49414.2021.9642706
Saeed S. Ba Hashwan, M. H. M. Md Khir, Y. Al-Douri, A. Yousif, H. Ramza, S. Arjo
The analytical modeling, design, and simulation of micromachined MEMS resonator for ammonia gas detection is presented in this paper. The MEMS resonator is designed to be vibrated electrostatically using interdigitated comb fingers. The demonstrated device is designed to be capable to carry micro-ring resonator and vibrated in-plane laterally to enhance the sensitivity of the gas detection. This MEMS resonator working principle is based on the changes in the output signal wavelength due to the change in the effective refractive index introduced by the ammonia gas. The resonant frequency of the actuator and the pull-in voltage have been calculated theoretically and found to be 11.15 kHz and 79.7 V respectively. The design and simulation of the micromachined micro-resonator has been carried out using CoventorWare software. Furthermore, the mathematically modeled results were verified using the finite element analysis software and the result shows a good agreement within 1.06% error between the modeled and simulated frequencies where the modeled and the simulated frequencies are found to be 11.15 kHz and 11.27 kHz respectively.
{"title":"Design and Simulation of MEMS Electrostatic Resonator for Ammonia Gas Detection Based on SOIMUMPs","authors":"Saeed S. Ba Hashwan, M. H. M. Md Khir, Y. Al-Douri, A. Yousif, H. Ramza, S. Arjo","doi":"10.1109/ICIAS49414.2021.9642706","DOIUrl":"https://doi.org/10.1109/ICIAS49414.2021.9642706","url":null,"abstract":"The analytical modeling, design, and simulation of micromachined MEMS resonator for ammonia gas detection is presented in this paper. The MEMS resonator is designed to be vibrated electrostatically using interdigitated comb fingers. The demonstrated device is designed to be capable to carry micro-ring resonator and vibrated in-plane laterally to enhance the sensitivity of the gas detection. This MEMS resonator working principle is based on the changes in the output signal wavelength due to the change in the effective refractive index introduced by the ammonia gas. The resonant frequency of the actuator and the pull-in voltage have been calculated theoretically and found to be 11.15 kHz and 79.7 V respectively. The design and simulation of the micromachined micro-resonator has been carried out using CoventorWare software. Furthermore, the mathematically modeled results were verified using the finite element analysis software and the result shows a good agreement within 1.06% error between the modeled and simulated frequencies where the modeled and the simulated frequencies are found to be 11.15 kHz and 11.27 kHz respectively.","PeriodicalId":212635,"journal":{"name":"2020 8th International Conference on Intelligent and Advanced Systems (ICIAS)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122726392","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-13DOI: 10.1109/ICIAS49414.2021.9642622
A. Elashmawy, I. Elamvazuthi, Syed Saad Azhar Ali, Elango Natarajan, S. Paramasivam
Tuberculosis (TB), a disease that targets the individual’s lungs and can cause fatalities can be cured if detected and treated early. Computer Aided Diagnosis (CAD) systems could be utilized to detect the presence of TB in Chest X-Ray Images (CXR). This paper proposes to investigate a hybridized pre-processing method for Convolutional Neural Network (CNN) CAD system for detecting TB in CXR images. The aim of this research is to improve the performance of CNNs by combining two different pre-processing methods and to further multi-classify different manifestation of TB. In this research, the experimental design is to apply augmentation and segmentation to CXR images as pre-processing and use a pretrained CNN model to classify the pre-processed images. It is hypothesized that the research would improve the accuracy and Area Under Curve (AUC) of detection of TB in CXR images.
{"title":"A Hybridized Pre-Processing Method for Detecting Tuberculosis using Deep Learning","authors":"A. Elashmawy, I. Elamvazuthi, Syed Saad Azhar Ali, Elango Natarajan, S. Paramasivam","doi":"10.1109/ICIAS49414.2021.9642622","DOIUrl":"https://doi.org/10.1109/ICIAS49414.2021.9642622","url":null,"abstract":"Tuberculosis (TB), a disease that targets the individual’s lungs and can cause fatalities can be cured if detected and treated early. Computer Aided Diagnosis (CAD) systems could be utilized to detect the presence of TB in Chest X-Ray Images (CXR). This paper proposes to investigate a hybridized pre-processing method for Convolutional Neural Network (CNN) CAD system for detecting TB in CXR images. The aim of this research is to improve the performance of CNNs by combining two different pre-processing methods and to further multi-classify different manifestation of TB. In this research, the experimental design is to apply augmentation and segmentation to CXR images as pre-processing and use a pretrained CNN model to classify the pre-processed images. It is hypothesized that the research would improve the accuracy and Area Under Curve (AUC) of detection of TB in CXR images.","PeriodicalId":212635,"journal":{"name":"2020 8th International Conference on Intelligent and Advanced Systems (ICIAS)","volume":"129 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133263367","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-13DOI: 10.1109/ICIAS49414.2021.9642538
Muhammad Kirana Baiduri, D. Perdana, S. Sussi
Hypertension is a condition when blood pressure values exceed a certain number range. Hypertension can occur complications that can threaten life. Handling and fast detection can be done with the help of the Internet of Things. In this study, designed a device that can detect the risk of hypertension in patients by monitoring the patient’s blood pressure. The accuracy of the systolic value is 96.83 percent and the diastole value of 90.1 percent. The average delay from the device to the firebase for 77.72 ms, 78.89 ms, and 79.34 ms. Then from firebase to the application for 240.23 ms, 241.86 ms, and 242.47 ms. The farther the distance, the higher the delay. The jitter is 124.49 ms from device to Firebase, while from Firebase to the application is 125.62 ms. The throughput is 4,194 bps from Firebase to the application, while from the device to Firebase is 12,543 bps.
{"title":"Blood Pressure Measuring Device for Hypertension Monitoring Based on Internet of Things with E-KTP Authentication","authors":"Muhammad Kirana Baiduri, D. Perdana, S. Sussi","doi":"10.1109/ICIAS49414.2021.9642538","DOIUrl":"https://doi.org/10.1109/ICIAS49414.2021.9642538","url":null,"abstract":"Hypertension is a condition when blood pressure values exceed a certain number range. Hypertension can occur complications that can threaten life. Handling and fast detection can be done with the help of the Internet of Things. In this study, designed a device that can detect the risk of hypertension in patients by monitoring the patient’s blood pressure. The accuracy of the systolic value is 96.83 percent and the diastole value of 90.1 percent. The average delay from the device to the firebase for 77.72 ms, 78.89 ms, and 79.34 ms. Then from firebase to the application for 240.23 ms, 241.86 ms, and 242.47 ms. The farther the distance, the higher the delay. The jitter is 124.49 ms from device to Firebase, while from Firebase to the application is 125.62 ms. The throughput is 4,194 bps from Firebase to the application, while from the device to Firebase is 12,543 bps.","PeriodicalId":212635,"journal":{"name":"2020 8th International Conference on Intelligent and Advanced Systems (ICIAS)","volume":"153 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134372356","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-13DOI: 10.1109/ICIAS49414.2021.9642563
Kalaimany Arumuggam, A. D. Saliang, M. Fuad, M. Yusof Abu Bakar, Adrian Buayeh Tadam, M. Naim
The generator and step-up transformer system is one of the most important equipment in a liquefied natural gas (LNG) plant. Reliability of this equipment is important in ensuring stable production of LNG. Thus, a rigorous, “fit for purpose” and comprehensive condition based maintenance strategy is required to ensure these equipment are properly maintained; otherwise, premature failures are possible to occur. A good condition based maintenance program will be able to assess the equipment healthiness and predict failures before happening. This paper presents the results, findings and analysis of condition based maintenance program for generators and step-up transformers system in the plant. Rectifications performed based on the findings are explained, which will serve as a guideline to maintenance engineers.
{"title":"Condition based maintenance for generator and transformer system","authors":"Kalaimany Arumuggam, A. D. Saliang, M. Fuad, M. Yusof Abu Bakar, Adrian Buayeh Tadam, M. Naim","doi":"10.1109/ICIAS49414.2021.9642563","DOIUrl":"https://doi.org/10.1109/ICIAS49414.2021.9642563","url":null,"abstract":"The generator and step-up transformer system is one of the most important equipment in a liquefied natural gas (LNG) plant. Reliability of this equipment is important in ensuring stable production of LNG. Thus, a rigorous, “fit for purpose” and comprehensive condition based maintenance strategy is required to ensure these equipment are properly maintained; otherwise, premature failures are possible to occur. A good condition based maintenance program will be able to assess the equipment healthiness and predict failures before happening. This paper presents the results, findings and analysis of condition based maintenance program for generators and step-up transformers system in the plant. Rectifications performed based on the findings are explained, which will serve as a guideline to maintenance engineers.","PeriodicalId":212635,"journal":{"name":"2020 8th International Conference on Intelligent and Advanced Systems (ICIAS)","volume":"98 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121586793","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-13DOI: 10.1109/ICIAS49414.2021.9642625
Q. J. Liew, H. F. Hawari, L. H. Wah, M. Khir, S. Sulaiman
In this study, flexible inkjet-printed capacitive humidity sensors were developed using Fujifilm Dimatix 2850 printer. The interdigitated electrodes (IDE) of the sensors were printed with the commercial silver nano-particles ink (AgNPs ink) on the polyethylene terephthalate (PET). To develop the sensing material on the humidity sensors, two methodologies were investigated. The former was a drop-cast method while the latter was an inkjet-printing method. Both utilized graphene oxide (GO) as the sensing material mixed with polyacrylic acid (PAA). The results in capacitance were obtained from experiments conducted in the humidity range from 30 % RH to 80 % RH. It was found that the former to be more suitable working in humidity range of 30 % RH up to 70% RH while the latter more suitable working in humidity range of 30 % RH up to 60% RH. The methodology using the drop-casted method presented higher sensitivity 9.766 pF/% RH compared to the inkjet printing method which was 0.441 pF/% RH. The sensors also exhibited good stabilities during short-term and long-term stability tests under predefined humidity and temperature.
{"title":"A Comparative Study on Performance of Drop Casted and Inkjet Printed Humidity Sensor","authors":"Q. J. Liew, H. F. Hawari, L. H. Wah, M. Khir, S. Sulaiman","doi":"10.1109/ICIAS49414.2021.9642625","DOIUrl":"https://doi.org/10.1109/ICIAS49414.2021.9642625","url":null,"abstract":"In this study, flexible inkjet-printed capacitive humidity sensors were developed using Fujifilm Dimatix 2850 printer. The interdigitated electrodes (IDE) of the sensors were printed with the commercial silver nano-particles ink (AgNPs ink) on the polyethylene terephthalate (PET). To develop the sensing material on the humidity sensors, two methodologies were investigated. The former was a drop-cast method while the latter was an inkjet-printing method. Both utilized graphene oxide (GO) as the sensing material mixed with polyacrylic acid (PAA). The results in capacitance were obtained from experiments conducted in the humidity range from 30 % RH to 80 % RH. It was found that the former to be more suitable working in humidity range of 30 % RH up to 70% RH while the latter more suitable working in humidity range of 30 % RH up to 60% RH. The methodology using the drop-casted method presented higher sensitivity 9.766 pF/% RH compared to the inkjet printing method which was 0.441 pF/% RH. The sensors also exhibited good stabilities during short-term and long-term stability tests under predefined humidity and temperature.","PeriodicalId":212635,"journal":{"name":"2020 8th International Conference on Intelligent and Advanced Systems (ICIAS)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127680846","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-13DOI: 10.1109/ICIAS49414.2021.9642641
Chew Kia Yuan Zerahny, L. Yun, W. Raymond, K. Mei
A long transmission line was simulated to collect fault data from one end of the line, which makes it a cost-efficient approach. Using Discrete Wavelet Transform (DWT), the essential characteristics of the fault type and its location can be extracted. Twelve types of mother wavelets with decomposition levels up to level 9 were compared and Haar wavelet was found to be most suitable. The resulting output was used as features to train several machine learning models for location and classification of faults. Fault estimation was carried out using the features extracted. By relying on the fault estimation, the search area for the fault can be reduced, thus decreasing the time needed to locate the actual fault. The artificial neural network (ANN) performed very well for fault classification having up to 100% accuracy. Another ANN was used for fault zone location and the accuracy obtained was 95.9%. Other machine learning models perform slightly poorer than ANN but had acceptable accuracy for fault location and classification. The results obtained considered single-phase to ground, two-phase, two-phase to ground and three-phase to ground faults. The faults occurred at various fault inception angles. The faults also included low and high fault impedances. The results indicate that this approach managed to detect and locate the fault zone with reasonable accuracy on a long transmission line model using data measured from one end only.
{"title":"Fault Classification and Location in Three-Phase Transmission Lines Using Wavelet-based Machine Learning","authors":"Chew Kia Yuan Zerahny, L. Yun, W. Raymond, K. Mei","doi":"10.1109/ICIAS49414.2021.9642641","DOIUrl":"https://doi.org/10.1109/ICIAS49414.2021.9642641","url":null,"abstract":"A long transmission line was simulated to collect fault data from one end of the line, which makes it a cost-efficient approach. Using Discrete Wavelet Transform (DWT), the essential characteristics of the fault type and its location can be extracted. Twelve types of mother wavelets with decomposition levels up to level 9 were compared and Haar wavelet was found to be most suitable. The resulting output was used as features to train several machine learning models for location and classification of faults. Fault estimation was carried out using the features extracted. By relying on the fault estimation, the search area for the fault can be reduced, thus decreasing the time needed to locate the actual fault. The artificial neural network (ANN) performed very well for fault classification having up to 100% accuracy. Another ANN was used for fault zone location and the accuracy obtained was 95.9%. Other machine learning models perform slightly poorer than ANN but had acceptable accuracy for fault location and classification. The results obtained considered single-phase to ground, two-phase, two-phase to ground and three-phase to ground faults. The faults occurred at various fault inception angles. The faults also included low and high fault impedances. The results indicate that this approach managed to detect and locate the fault zone with reasonable accuracy on a long transmission line model using data measured from one end only.","PeriodicalId":212635,"journal":{"name":"2020 8th International Conference on Intelligent and Advanced Systems (ICIAS)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115951119","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-13DOI: 10.1109/ICIAS49414.2021.9642637
A. E. Ibrahim, N. M. Nor, M. Romlie, K. Hasan
Partial Shaded Condition (PSC) is a process of non-optimal power capture in PV systems; this phenomenon occurs when one or more PV solar cells are shaded by external factors. It generates one global peak and many local peaks in the Power voltage curve. Thus, it makes a sudden change in cell irradiance and leads to non-optimal power capture in PV systems. In this work, a novel MPPT technique was proposed to solve the problem of tracking speed and accuracy in the shaded PV systems. FLC and GA-FLC with Power Scanning Technique (PST) algorithms were designed, developed and simulated using MATLAB/Simulink. The performance of FLC and GAFLC with PST was figured out under varied operating conditions. Proposed FLC tracked the PV power under PSC in 0. 4s with 92% accuracy, meanwhile, GA-FLC with PST achieved 0. 4s as tracking speed with 99% accuracy using Buck converter. Proposed GA-FLC with PST showed high capability in terms of tracking speed and accuracy compared to MPPT techniques. Finally, the dynamic performance of GA-FLC with the PST algorithm was compared to other techniques reported in the works of literature. Improved tracking speed and accuracy were obtained compared to other methods. For future work, experimental work in terms of tracking speed and accuracy should be carried out to validate the performance of the proposed algorithm.
{"title":"New Technique for Tracking the Global Peaks of PV System under Partial Shaded Conditions","authors":"A. E. Ibrahim, N. M. Nor, M. Romlie, K. Hasan","doi":"10.1109/ICIAS49414.2021.9642637","DOIUrl":"https://doi.org/10.1109/ICIAS49414.2021.9642637","url":null,"abstract":"Partial Shaded Condition (PSC) is a process of non-optimal power capture in PV systems; this phenomenon occurs when one or more PV solar cells are shaded by external factors. It generates one global peak and many local peaks in the Power voltage curve. Thus, it makes a sudden change in cell irradiance and leads to non-optimal power capture in PV systems. In this work, a novel MPPT technique was proposed to solve the problem of tracking speed and accuracy in the shaded PV systems. FLC and GA-FLC with Power Scanning Technique (PST) algorithms were designed, developed and simulated using MATLAB/Simulink. The performance of FLC and GAFLC with PST was figured out under varied operating conditions. Proposed FLC tracked the PV power under PSC in 0. 4s with 92% accuracy, meanwhile, GA-FLC with PST achieved 0. 4s as tracking speed with 99% accuracy using Buck converter. Proposed GA-FLC with PST showed high capability in terms of tracking speed and accuracy compared to MPPT techniques. Finally, the dynamic performance of GA-FLC with the PST algorithm was compared to other techniques reported in the works of literature. Improved tracking speed and accuracy were obtained compared to other methods. For future work, experimental work in terms of tracking speed and accuracy should be carried out to validate the performance of the proposed algorithm.","PeriodicalId":212635,"journal":{"name":"2020 8th International Conference on Intelligent and Advanced Systems (ICIAS)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125915277","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：