Pub Date : 2022-02-24DOI: 10.1109/icaeee54957.2022.9836434
Towkir Ahmed, M. Alam, R. Paul, M. T. Hasan, Raqeebir Rab
Music genre classification is extremely important for both music recommendation and acquisition of music data, as well as for music discovery. There have already been a vast amount of researches conducted on the classification of music genres using various machine learning algorithms. Despite the fact that Bangla music is extremely diverse in terms of its own style, there has been little notable work done to date to categorize song genres in Bangla music using machine learning approaches. There are numerous varieties and modes of Bangla music, all of which may be categorised into different classes by their musical compositions. The dataset we use contains six different Bangla music genres. There are several unique attributes for each song which is included in the dataset, including zero crossing value, delta, chroma frequency, spectral roll-off, spectral bandwidth, and many others. Several machine learning models, as well as a deep learning technique, are proposed in this paper for classi-fying Bangla musics into multi-class classification. To train the supervised learning models, we used dimentionality reduction and feature scaling to increase the performance. Finally, our models are evaluated using f'l-score, recall, accuracy and precision. As can be observed, the implemented deep neural network model was able to reach an accuracy of 77.68 percent.
{"title":"Machine Learning and Deep Learning Techniques For Genre Classification of Bangla Music","authors":"Towkir Ahmed, M. Alam, R. Paul, M. T. Hasan, Raqeebir Rab","doi":"10.1109/icaeee54957.2022.9836434","DOIUrl":"https://doi.org/10.1109/icaeee54957.2022.9836434","url":null,"abstract":"Music genre classification is extremely important for both music recommendation and acquisition of music data, as well as for music discovery. There have already been a vast amount of researches conducted on the classification of music genres using various machine learning algorithms. Despite the fact that Bangla music is extremely diverse in terms of its own style, there has been little notable work done to date to categorize song genres in Bangla music using machine learning approaches. There are numerous varieties and modes of Bangla music, all of which may be categorised into different classes by their musical compositions. The dataset we use contains six different Bangla music genres. There are several unique attributes for each song which is included in the dataset, including zero crossing value, delta, chroma frequency, spectral roll-off, spectral bandwidth, and many others. Several machine learning models, as well as a deep learning technique, are proposed in this paper for classi-fying Bangla musics into multi-class classification. To train the supervised learning models, we used dimentionality reduction and feature scaling to increase the performance. Finally, our models are evaluated using f'l-score, recall, accuracy and precision. As can be observed, the implemented deep neural network model was able to reach an accuracy of 77.68 percent.","PeriodicalId":383872,"journal":{"name":"2022 International Conference on Advancement in Electrical and Electronic Engineering (ICAEEE)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131109046","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 : 2022-02-24DOI: 10.1109/icaeee54957.2022.9836588
Tafsir Ahmed Khan, Syed Abdullah-Al-Nahid, Md. Abu Taseen, S. Tasnim, T. Aziz
Generation Expansion Planning (GEP) is determining the type, location and number of new generating stations (GSs). In this paper, a GEP problem is formed by considering three types of GSs and then their possible combinations are sorted. Infeasible combinations are screened out based on the capacity limit and maximum allowable budget. The best solution with minimum cost is recognized by optimizing the feasible combinations using Genetic Algorithm (GA). Share of fuel mix (gas and oil) for winter and other seasons are considered as the constraints. In simulation, 14 out of 75 combinations came out feasible. GA was used to find the best combination which had an optimized amount of gas and oil usage. The results display the superiority of proposed methodology in contrast with other studies in finding the best solution of the GEP problem with minimum iteration.
{"title":"Generation Expansion Planning Optimized by Genetic Algorithm Considering Seasonal Impact and Fuel Price","authors":"Tafsir Ahmed Khan, Syed Abdullah-Al-Nahid, Md. Abu Taseen, S. Tasnim, T. Aziz","doi":"10.1109/icaeee54957.2022.9836588","DOIUrl":"https://doi.org/10.1109/icaeee54957.2022.9836588","url":null,"abstract":"Generation Expansion Planning (GEP) is determining the type, location and number of new generating stations (GSs). In this paper, a GEP problem is formed by considering three types of GSs and then their possible combinations are sorted. Infeasible combinations are screened out based on the capacity limit and maximum allowable budget. The best solution with minimum cost is recognized by optimizing the feasible combinations using Genetic Algorithm (GA). Share of fuel mix (gas and oil) for winter and other seasons are considered as the constraints. In simulation, 14 out of 75 combinations came out feasible. GA was used to find the best combination which had an optimized amount of gas and oil usage. The results display the superiority of proposed methodology in contrast with other studies in finding the best solution of the GEP problem with minimum iteration.","PeriodicalId":383872,"journal":{"name":"2022 International Conference on Advancement in Electrical and Electronic Engineering (ICAEEE)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133676779","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 : 2022-02-24DOI: 10.1109/icaeee54957.2022.9836370
Robi Paul
The aggressive reduction of FET devices predicted in Moore's law has escorted us to an exponential decrease in device performance. Shifting from existing FET devices to Tunneling Field-Effect Transistor (TFET) has demonstrated higher performance while maintaining a significantly lower transistor gate size. It offers a steep subthreshold swing slope with a substantially lower leakage current, resulting in competitively lower power absorption from ordinary FETs. However, to increase the control over the TFET device even further, a slight variation in a design known as the Double Gate Tunneling Field-Effect Transistor (DG- TFET) is implicated. In this study, I have investigated and adjusted the performance of an N-type DG-TFET by altering several parameters such as device materials, high-k dielectric as oxide layers, and oxide thickness. In the end, Tungsten Ditelluride (WTe2) a 2-D material is used as the device material, while Niobium pentoxide (Nb2O5) is used as the high-k dielectric material according to the optimization process of the DG-TFET. The device has achieved a subthreshold swing of 18.37 mv/Dec and an Ion/Ioff of 1011. Finally, I have also conducted a comparative analysis between DG-TFET and a Single Gate Tunneling Field-Effect Transistor (SG-TFET) device with identical specifications.
{"title":"Performance Investigation and Optimization of 2-D Material based Double Gate Tunneling Field-Effect Transistor (DG-TFET)","authors":"Robi Paul","doi":"10.1109/icaeee54957.2022.9836370","DOIUrl":"https://doi.org/10.1109/icaeee54957.2022.9836370","url":null,"abstract":"The aggressive reduction of FET devices predicted in Moore's law has escorted us to an exponential decrease in device performance. Shifting from existing FET devices to Tunneling Field-Effect Transistor (TFET) has demonstrated higher performance while maintaining a significantly lower transistor gate size. It offers a steep subthreshold swing slope with a substantially lower leakage current, resulting in competitively lower power absorption from ordinary FETs. However, to increase the control over the TFET device even further, a slight variation in a design known as the Double Gate Tunneling Field-Effect Transistor (DG- TFET) is implicated. In this study, I have investigated and adjusted the performance of an N-type DG-TFET by altering several parameters such as device materials, high-k dielectric as oxide layers, and oxide thickness. In the end, Tungsten Ditelluride (WTe2) a 2-D material is used as the device material, while Niobium pentoxide (Nb2O5) is used as the high-k dielectric material according to the optimization process of the DG-TFET. The device has achieved a subthreshold swing of 18.37 mv/Dec and an Ion/Ioff of 1011. Finally, I have also conducted a comparative analysis between DG-TFET and a Single Gate Tunneling Field-Effect Transistor (SG-TFET) device with identical specifications.","PeriodicalId":383872,"journal":{"name":"2022 International Conference on Advancement in Electrical and Electronic Engineering (ICAEEE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129667911","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 : 2022-02-24DOI: 10.1109/icaeee54957.2022.9836354
Zabedur Rahman, Mahfuzulhoq Chowdhury, Abu Bakkar Siddique
Car parking is one of the most significant issues in today's world. Parking cars on surrounding roads and pathways can cause unfair traffic jams and thus hampers people's daily activity. To avoid these problems, the development of a smart car parking system is a major concern for several developed countries. At present, most of the previous studies on car parking systems suffer from several limitations such as lack of security, wastage of time, huge money expenses, and lack of user interest-aware car parking system. To overcome existing challenges, this paper presents a user interest and payment-aware automated car parking system using Internet-of-things (IoT) technology. In this paper, an android application for smart car parking is developed for Bangladeshi people that allow users to choose emergency or non-emergency parking slots based on their interest and payment verification. For anti-theft purposes, this system offers an early alert and notification feature. The experimental test results by investigating several use cases depict the suitability of the proposed system.
{"title":"An User Interest and Payment-aware Automated Car Parking System for the Bangladeshi People Using Android Application","authors":"Zabedur Rahman, Mahfuzulhoq Chowdhury, Abu Bakkar Siddique","doi":"10.1109/icaeee54957.2022.9836354","DOIUrl":"https://doi.org/10.1109/icaeee54957.2022.9836354","url":null,"abstract":"Car parking is one of the most significant issues in today's world. Parking cars on surrounding roads and pathways can cause unfair traffic jams and thus hampers people's daily activity. To avoid these problems, the development of a smart car parking system is a major concern for several developed countries. At present, most of the previous studies on car parking systems suffer from several limitations such as lack of security, wastage of time, huge money expenses, and lack of user interest-aware car parking system. To overcome existing challenges, this paper presents a user interest and payment-aware automated car parking system using Internet-of-things (IoT) technology. In this paper, an android application for smart car parking is developed for Bangladeshi people that allow users to choose emergency or non-emergency parking slots based on their interest and payment verification. For anti-theft purposes, this system offers an early alert and notification feature. The experimental test results by investigating several use cases depict the suitability of the proposed system.","PeriodicalId":383872,"journal":{"name":"2022 International Conference on Advancement in Electrical and Electronic Engineering (ICAEEE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114347975","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 : 2022-02-24DOI: 10.1109/icaeee54957.2022.9836541
Tasmima Noushiba Mahbub, M. Yousuf, M.N. Uddin
Every year a significant number of women dies because of suffering from breast cancer all over the world. The rate of mortality due to breast cancer can be decreased if the cancer and the stage is early detected. Early Diagnosis is not possible in every corner of all countries over the world because of the lack of experienced consultant or doctor. A novel approach is presented in this study based on convolutional neural network and fuzzy analytical hierarchy process for diagnosis of breast cancer along with stage identification. The proposed model detects breast cancer from mammographic images using modified convolutional neural network. Then identifies the stage using fuzzy analytical hierarchy process model which is comprised of 3 layers (goal, criteria and alternative). Proposed modified convolutional neural network model achieves 98.75% validation accuracy on detecting breast cancer from mammograms as well as the fuzzy AHP model efficiently identifies the stage of the cancer.
{"title":"A Modified CNN And Fuzzy AHP Based Breast Cancer Stage Detection System","authors":"Tasmima Noushiba Mahbub, M. Yousuf, M.N. Uddin","doi":"10.1109/icaeee54957.2022.9836541","DOIUrl":"https://doi.org/10.1109/icaeee54957.2022.9836541","url":null,"abstract":"Every year a significant number of women dies because of suffering from breast cancer all over the world. The rate of mortality due to breast cancer can be decreased if the cancer and the stage is early detected. Early Diagnosis is not possible in every corner of all countries over the world because of the lack of experienced consultant or doctor. A novel approach is presented in this study based on convolutional neural network and fuzzy analytical hierarchy process for diagnosis of breast cancer along with stage identification. The proposed model detects breast cancer from mammographic images using modified convolutional neural network. Then identifies the stage using fuzzy analytical hierarchy process model which is comprised of 3 layers (goal, criteria and alternative). Proposed modified convolutional neural network model achieves 98.75% validation accuracy on detecting breast cancer from mammograms as well as the fuzzy AHP model efficiently identifies the stage of the cancer.","PeriodicalId":383872,"journal":{"name":"2022 International Conference on Advancement in Electrical and Electronic Engineering (ICAEEE)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125316958","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 : 2022-02-24DOI: 10.1109/icaeee54957.2022.9836364
Merajur Rahman Mollah, Muhammad Asad Rahman, Md. Shohanur Rahman Shohan
A multi-band Sierpinski carpet fractal antenna with a modified ground plane is designed. Fractal shapes are applied on the both sides of the antenna to achieve multi-band characteristics. Sierpinski carpet fractal with iteration-3 is applied to the rectangular-shaped radiating patch. Here, the novelty of the proposed design is the modified ground plane. The ground is modified through the same fractal shape of the patch (i.e., Sierpinski here) up to 2nd iteration as defected ground structure (DGS) on a partial ground to get more resonant bands over the range of 4 GHz to 12 GHz. Moreover, partial ground helps to get better input impedance matching at the resonance frequencies. The overall dimension of the proposed structure is 45 mm x 60 mm x 1.60 mm. The proposed antenna operates at six resonant frequencies (6 GHz, 6.42 GHz, 7.09 GHz, 7.63 GHz, 9.15 GHz, and 10.11 GHz) over the range of 4 to 12 GHz with good impedance matching, good gain and efficiency. The design is suitable for different applications of C- and X-bands.
{"title":"Design of a Multi-band Sierpinski Carpet Fractal Antenna With Modified Ground Plane","authors":"Merajur Rahman Mollah, Muhammad Asad Rahman, Md. Shohanur Rahman Shohan","doi":"10.1109/icaeee54957.2022.9836364","DOIUrl":"https://doi.org/10.1109/icaeee54957.2022.9836364","url":null,"abstract":"A multi-band Sierpinski carpet fractal antenna with a modified ground plane is designed. Fractal shapes are applied on the both sides of the antenna to achieve multi-band characteristics. Sierpinski carpet fractal with iteration-3 is applied to the rectangular-shaped radiating patch. Here, the novelty of the proposed design is the modified ground plane. The ground is modified through the same fractal shape of the patch (i.e., Sierpinski here) up to 2nd iteration as defected ground structure (DGS) on a partial ground to get more resonant bands over the range of 4 GHz to 12 GHz. Moreover, partial ground helps to get better input impedance matching at the resonance frequencies. The overall dimension of the proposed structure is 45 mm x 60 mm x 1.60 mm. The proposed antenna operates at six resonant frequencies (6 GHz, 6.42 GHz, 7.09 GHz, 7.63 GHz, 9.15 GHz, and 10.11 GHz) over the range of 4 to 12 GHz with good impedance matching, good gain and efficiency. The design is suitable for different applications of C- and X-bands.","PeriodicalId":383872,"journal":{"name":"2022 International Conference on Advancement in Electrical and Electronic Engineering (ICAEEE)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114260054","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 : 2022-02-24DOI: 10.1109/icaeee54957.2022.9836477
M. Alam, Moin Uddin Siddique, H. Sakib, Imtiaz Hossain, Iftekher Alam Rahat
A hybrid energy system combines two or more renewable energy sources to improve system efficiency and supply balance. Vehicles will be a huge source of power. Among all renewable energy sources, solar and wind are the most efficient to attach to a car. The Hybrid Renewable Energy Vehicle System (HREVS) proposes charging the vehicle's battery with hybrid renewable energy sources. This work's major goals are to minimize vehicle dependence on fossil fuels, increase reliance on renewable energy sources, and lower fuel costs. Development of a full battery charging system each photovoltaic and wind model is designed separately, then combined with a charge controller and a battery. A maximum power point tracking system and code have been developed for solar tracking using the Perturb and Observe (P & O) method. State of Charge (SOC) controls the battery's charging and draining. We tried to address the issues raised above. A desired output result from a hybrid energy configuration has also been explored. All simulation and setup are done in MATLAB-SIMULINK. Blender creates a 3D model of a hybrid car. A minor expansion created and controlled using Arduino-UNO is also included. The results of the experiments and simulations suggest that the proposed system can generate power and reduce fuel usage.
{"title":"Reducing Fuel Dependency of Electric Vehicles using Hybrid Renewable Energy System","authors":"M. Alam, Moin Uddin Siddique, H. Sakib, Imtiaz Hossain, Iftekher Alam Rahat","doi":"10.1109/icaeee54957.2022.9836477","DOIUrl":"https://doi.org/10.1109/icaeee54957.2022.9836477","url":null,"abstract":"A hybrid energy system combines two or more renewable energy sources to improve system efficiency and supply balance. Vehicles will be a huge source of power. Among all renewable energy sources, solar and wind are the most efficient to attach to a car. The Hybrid Renewable Energy Vehicle System (HREVS) proposes charging the vehicle's battery with hybrid renewable energy sources. This work's major goals are to minimize vehicle dependence on fossil fuels, increase reliance on renewable energy sources, and lower fuel costs. Development of a full battery charging system each photovoltaic and wind model is designed separately, then combined with a charge controller and a battery. A maximum power point tracking system and code have been developed for solar tracking using the Perturb and Observe (P & O) method. State of Charge (SOC) controls the battery's charging and draining. We tried to address the issues raised above. A desired output result from a hybrid energy configuration has also been explored. All simulation and setup are done in MATLAB-SIMULINK. Blender creates a 3D model of a hybrid car. A minor expansion created and controlled using Arduino-UNO is also included. The results of the experiments and simulations suggest that the proposed system can generate power and reduce fuel usage.","PeriodicalId":383872,"journal":{"name":"2022 International Conference on Advancement in Electrical and Electronic Engineering (ICAEEE)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124320375","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The high levels of blood sugar (or glucose) that occur in diabetes can damage organs such as the heart, blood vessels, eyes, kidneys, and nerves in time. Type 2 diabetes typically affects adults and is most prevalent in adults due to an insufficient supply of insulin. On the other hand, Diabetes type 1, also known as juvenile diabetes or insulin-dependent diabetes, is a chronic disease in which the body cannot produce insulin on its own. Diabetes prevalence has increased over the past three decades at every income level. Affordable treatment is vital for those with diabetes. Several cost-effective interventions can improve patient outcomes. However, a diagnosis of this disease can be costly and difficult. The aim of this research is, therefore, to demonstrate a comparative analysis and improved performance using deep learning to classify diabetic and non-diabetic patients that will provide a feasible way to diagnose this chronic disease. In this work, we used a neural network model with very low variance applying the synthetic minority oversampling technique to augment and improve the variety of data. By removing imbalances and classifying diabetes based on different features, our model achieved an accuracy of approximately 99 % for training and 98 % for validation.
{"title":"Diabetes Complication Prediction using Deep Learning-Based Analytics","authors":"Takrim Rahman Albi, Md Nakhla Rafi, Tasfia Anika Bushra, Dewan Ziaul Karim","doi":"10.1109/icaeee54957.2022.9836401","DOIUrl":"https://doi.org/10.1109/icaeee54957.2022.9836401","url":null,"abstract":"The high levels of blood sugar (or glucose) that occur in diabetes can damage organs such as the heart, blood vessels, eyes, kidneys, and nerves in time. Type 2 diabetes typically affects adults and is most prevalent in adults due to an insufficient supply of insulin. On the other hand, Diabetes type 1, also known as juvenile diabetes or insulin-dependent diabetes, is a chronic disease in which the body cannot produce insulin on its own. Diabetes prevalence has increased over the past three decades at every income level. Affordable treatment is vital for those with diabetes. Several cost-effective interventions can improve patient outcomes. However, a diagnosis of this disease can be costly and difficult. The aim of this research is, therefore, to demonstrate a comparative analysis and improved performance using deep learning to classify diabetic and non-diabetic patients that will provide a feasible way to diagnose this chronic disease. In this work, we used a neural network model with very low variance applying the synthetic minority oversampling technique to augment and improve the variety of data. By removing imbalances and classifying diabetes based on different features, our model achieved an accuracy of approximately 99 % for training and 98 % for validation.","PeriodicalId":383872,"journal":{"name":"2022 International Conference on Advancement in Electrical and Electronic Engineering (ICAEEE)","volume":"46 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129383404","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 : 2022-02-24DOI: 10.1109/icaeee54957.2022.9836359
S. Haque, Gobinda Chandra Sarker, Kazi Md Sadat
Power generation is increasing worldwide every year to cope with ever-increasing energy demand. Therefore, a significant necessity exists for forecasting the load demand to manage and increase electricity production capacity. Short-term load forecasting (STLF) using artificial neural network has become one of the most efficient and widely popular methods. This paper proposes a hybrid network of Long Short-Term Memory (LSTM) network and Convolutional Neural Network (CNN) to predict demand for seven days into the future. The proposed CNN-LSTM method is compared with various deep learning techniques such as vanilla neural network and gated recurrent unit (GRU). Power Grid Company of Bangladesh (PGCB) has the responsibility of reliable power transmission all over the country. Each model is trained and tested on multivariate historical data collected from the daily report section of PGCB website for the Mymensingh Division in Bangladesh. Various input features such as temperature, peak generation at evening, maximum generation, month and the season of the year are used to aid the prediction. It is found that the proposed CNN-LSTM method outperforms the other models with a MAPE error rate of 2.8992%, which is less than the MAPE error of 5.5554% for demand estimation of seven days used by PGCB.
{"title":"Short-Term Electrical Load Prediction for Future Generation Using Hybrid Deep Learning Model","authors":"S. Haque, Gobinda Chandra Sarker, Kazi Md Sadat","doi":"10.1109/icaeee54957.2022.9836359","DOIUrl":"https://doi.org/10.1109/icaeee54957.2022.9836359","url":null,"abstract":"Power generation is increasing worldwide every year to cope with ever-increasing energy demand. Therefore, a significant necessity exists for forecasting the load demand to manage and increase electricity production capacity. Short-term load forecasting (STLF) using artificial neural network has become one of the most efficient and widely popular methods. This paper proposes a hybrid network of Long Short-Term Memory (LSTM) network and Convolutional Neural Network (CNN) to predict demand for seven days into the future. The proposed CNN-LSTM method is compared with various deep learning techniques such as vanilla neural network and gated recurrent unit (GRU). Power Grid Company of Bangladesh (PGCB) has the responsibility of reliable power transmission all over the country. Each model is trained and tested on multivariate historical data collected from the daily report section of PGCB website for the Mymensingh Division in Bangladesh. Various input features such as temperature, peak generation at evening, maximum generation, month and the season of the year are used to aid the prediction. It is found that the proposed CNN-LSTM method outperforms the other models with a MAPE error rate of 2.8992%, which is less than the MAPE error of 5.5554% for demand estimation of seven days used by PGCB.","PeriodicalId":383872,"journal":{"name":"2022 International Conference on Advancement in Electrical and Electronic Engineering (ICAEEE)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130553704","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 : 2022-02-24DOI: 10.1109/icaeee54957.2022.9836405
Mohammad Saiful Islam, Md. Rashidul Islam, M. Shafiullah, Md. Samiul Azam
Low-frequency oscillation (LFO) is a significant problem for Multi-machine power system (MPS) networks. It makes the power system networks unstable. In this article, a new Power system stabilizer (PSS) design method is demonstrated using the Dragonfly algorithm (DA). To enhance system damping, a damping ratio-based objective function is used, and a typical lead-lag type PSS (CPSS) structure is considered. In this case, the algorithm's ability to provide the best PSS design regardless of the starting guess demonstrates its robustness. This method is tested on two separate multi-machine networks exposed to a 3-Φ fault, and compared with two well-known optimization algorithms called PSO and BSA. The optimization results show that the DA technique provides better system damping than PSO and BSA.
{"title":"Dragonfly Algorithm for Robust Tuning of Power System Stabilizers in Multimachine Networks","authors":"Mohammad Saiful Islam, Md. Rashidul Islam, M. Shafiullah, Md. Samiul Azam","doi":"10.1109/icaeee54957.2022.9836405","DOIUrl":"https://doi.org/10.1109/icaeee54957.2022.9836405","url":null,"abstract":"Low-frequency oscillation (LFO) is a significant problem for Multi-machine power system (MPS) networks. It makes the power system networks unstable. In this article, a new Power system stabilizer (PSS) design method is demonstrated using the Dragonfly algorithm (DA). To enhance system damping, a damping ratio-based objective function is used, and a typical lead-lag type PSS (CPSS) structure is considered. In this case, the algorithm's ability to provide the best PSS design regardless of the starting guess demonstrates its robustness. This method is tested on two separate multi-machine networks exposed to a 3-Φ fault, and compared with two well-known optimization algorithms called PSO and BSA. The optimization results show that the DA technique provides better system damping than PSO and BSA.","PeriodicalId":383872,"journal":{"name":"2022 International Conference on Advancement in Electrical and Electronic Engineering (ICAEEE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130663928","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：