Pub Date : 2022-07-21DOI: 10.1109/ICICCSP53532.2022.9862448
Krishna Mohan Boddapati, N. Patne, Ashwini D. Manchalwar
Energy trading in a smart microgrid based Peer to Peer (P2P) microgrid requires a novel framework. Where in a microgrid each prosumer equipped with Distributed Energy Resources (DER) is encouraged to trade energy. Energy is traded with peers in the microgrid or sold to power grid to generate revenue whenever there is a surplus of energy. As there is an economic benefit, more prosumers are attracted to install DER. The peers in the microgrid are incentivized to buy locally produced energy which is available at a cheaper cost compared to buying from power grid. This results in reducing energy consumption from power grid which in turn results in reducing power losses in transmission lines and results in reduction of $CO_{2}$ emissions. Three market paradigms are considered in this paper, i.e. Bill sharing, Supply to Demand Ratio method and Mid-market rate method. Each of the pricing models are elaborated with local energy trading price and prosumers energy costs. In this case, these methods are studied concerning a smart microgrid equipped with DER, Solar Photovoltaic (SPV) system in this case, and the effectiveness of the pricing mechanisms and their respective benefits are established.
{"title":"Energy Trading in Prosumer based Smart Grid Integrated with Distributed Energy Resources","authors":"Krishna Mohan Boddapati, N. Patne, Ashwini D. Manchalwar","doi":"10.1109/ICICCSP53532.2022.9862448","DOIUrl":"https://doi.org/10.1109/ICICCSP53532.2022.9862448","url":null,"abstract":"Energy trading in a smart microgrid based Peer to Peer (P2P) microgrid requires a novel framework. Where in a microgrid each prosumer equipped with Distributed Energy Resources (DER) is encouraged to trade energy. Energy is traded with peers in the microgrid or sold to power grid to generate revenue whenever there is a surplus of energy. As there is an economic benefit, more prosumers are attracted to install DER. The peers in the microgrid are incentivized to buy locally produced energy which is available at a cheaper cost compared to buying from power grid. This results in reducing energy consumption from power grid which in turn results in reducing power losses in transmission lines and results in reduction of $CO_{2}$ emissions. Three market paradigms are considered in this paper, i.e. Bill sharing, Supply to Demand Ratio method and Mid-market rate method. Each of the pricing models are elaborated with local energy trading price and prosumers energy costs. In this case, these methods are studied concerning a smart microgrid equipped with DER, Solar Photovoltaic (SPV) system in this case, and the effectiveness of the pricing mechanisms and their respective benefits are established.","PeriodicalId":326163,"journal":{"name":"2022 International Conference on Intelligent Controller and Computing for Smart Power (ICICCSP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116294514","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 : 2022-07-21DOI: 10.1109/ICICCSP53532.2022.9862422
B. Patle, N. Pagar, D. Parhi, S. Sanap
Recently, in the path planning of mobile robots, navigation in complex areas are still a challenging task using different AI techniques. One such problem of navigation is solved here using the firefly algorithm and the genetic algorithm as a hybrid approach. The proposed approach efficiently handles the sensory information and converts this into taking the accurate decision for solving the challenges of navigation such as obstacle avoidance and target seeking in a static environment. The proposed approach not only ensures path safety but also ensures path optimality on account of navigational parameters such as path length and navigational time. The developed approach has been tested in the simulation environment using the MATLAB software and in the real-time environment using the Khepera robot. The simulation and real-time results in presence of multiple obstacles are presented for the validation of the proposed FA-GA hybrid controller and obtained results are satisfactory in terms of path optimization.
{"title":"Hybrid FA-GA Controller for Path Planning of Mobile Robot","authors":"B. Patle, N. Pagar, D. Parhi, S. Sanap","doi":"10.1109/ICICCSP53532.2022.9862422","DOIUrl":"https://doi.org/10.1109/ICICCSP53532.2022.9862422","url":null,"abstract":"Recently, in the path planning of mobile robots, navigation in complex areas are still a challenging task using different AI techniques. One such problem of navigation is solved here using the firefly algorithm and the genetic algorithm as a hybrid approach. The proposed approach efficiently handles the sensory information and converts this into taking the accurate decision for solving the challenges of navigation such as obstacle avoidance and target seeking in a static environment. The proposed approach not only ensures path safety but also ensures path optimality on account of navigational parameters such as path length and navigational time. The developed approach has been tested in the simulation environment using the MATLAB software and in the real-time environment using the Khepera robot. The simulation and real-time results in presence of multiple obstacles are presented for the validation of the proposed FA-GA hybrid controller and obtained results are satisfactory in terms of path optimization.","PeriodicalId":326163,"journal":{"name":"2022 International Conference on Intelligent Controller and Computing for Smart Power (ICICCSP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121676011","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 : 2022-07-21DOI: 10.1109/ICICCSP53532.2022.9862506
Joseph Sanam, Yenimireddy Venkata Rajeswari, E. U. Sri, K. Spandhana, Chandu Bhavana Laxmi, Nandhiraju Gayathri
This paper presents the effect on Electrical Distribution Networks (EDN) due to the integration of shunt capacitors. A shunt capacitor is installed in all buses of an EDN, individually, and its effect on voltage profile and power loss of EDN is explored. An appropriate novel mathematical model of a shunt capacitor is derived to integrate it into the EDN. Exhaustive search-based FBSLFA (Forward-backward sweep load flow algorithm) is used to integrate the shunt capacitor in EDN and to optimize the voltage profile and power loss of the network respectively. The strategy of the proposed work is implemented in IEEE-52 bus EDN. voltage profile and power loss of the EDN are enhanced and reduced appreciably respectively with the integration of a shunt capacitor at a few buses.
{"title":"Impact on Electrical Distribution Networks with The Integration of Shunt Capacitor Model Using Exhaustive Search Based Load Flow Algorithm","authors":"Joseph Sanam, Yenimireddy Venkata Rajeswari, E. U. Sri, K. Spandhana, Chandu Bhavana Laxmi, Nandhiraju Gayathri","doi":"10.1109/ICICCSP53532.2022.9862506","DOIUrl":"https://doi.org/10.1109/ICICCSP53532.2022.9862506","url":null,"abstract":"This paper presents the effect on Electrical Distribution Networks (EDN) due to the integration of shunt capacitors. A shunt capacitor is installed in all buses of an EDN, individually, and its effect on voltage profile and power loss of EDN is explored. An appropriate novel mathematical model of a shunt capacitor is derived to integrate it into the EDN. Exhaustive search-based FBSLFA (Forward-backward sweep load flow algorithm) is used to integrate the shunt capacitor in EDN and to optimize the voltage profile and power loss of the network respectively. The strategy of the proposed work is implemented in IEEE-52 bus EDN. voltage profile and power loss of the EDN are enhanced and reduced appreciably respectively with the integration of a shunt capacitor at a few buses.","PeriodicalId":326163,"journal":{"name":"2022 International Conference on Intelligent Controller and Computing for Smart Power (ICICCSP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113985621","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 : 2022-07-21DOI: 10.1109/ICICCSP53532.2022.9862446
A. Bhuyan, B. Panigrahi, Kumaresh Pal, Subhendu Pati
Faults are becoming more common as the number of transmission lines grows progressively. The detection of faults must be quick and precise to do the least amount of harm to the power system. Convolutional Neural Networks (CNN) is one of the finest options for detecting faults in transmission lines. This paper presents a novel fault detection method based on Convolutional Neural Networks in which the current vs. time graph of all faults is used as input for the image classifier. For the input an image data has been generated with appropriate target values and given to the model. The model is trained and tested after it is created. The testing results reveal that the convolutional neural network performs well for all types of faults.
{"title":"Convolutional Neural Network Based Fault Detection for Transmission Line","authors":"A. Bhuyan, B. Panigrahi, Kumaresh Pal, Subhendu Pati","doi":"10.1109/ICICCSP53532.2022.9862446","DOIUrl":"https://doi.org/10.1109/ICICCSP53532.2022.9862446","url":null,"abstract":"Faults are becoming more common as the number of transmission lines grows progressively. The detection of faults must be quick and precise to do the least amount of harm to the power system. Convolutional Neural Networks (CNN) is one of the finest options for detecting faults in transmission lines. This paper presents a novel fault detection method based on Convolutional Neural Networks in which the current vs. time graph of all faults is used as input for the image classifier. For the input an image data has been generated with appropriate target values and given to the model. The model is trained and tested after it is created. The testing results reveal that the convolutional neural network performs well for all types of faults.","PeriodicalId":326163,"journal":{"name":"2022 International Conference on Intelligent Controller and Computing for Smart Power (ICICCSP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126364439","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 : 2022-07-21DOI: 10.1109/ICICCSP53532.2022.9862452
D. V. S. K. Rao, B. Prusty, Hareesh Myneni
The grid integration of solar photovoltaic (PV) systems has recently grabbed considerable research attention. Simultaneously, the grid has been subjected to disturbances due to PV generations' variability, uncertainty, and intermittency; therefore, accurately estimating the weather-dependent PV power is imperative. The daily global solar radiation, temperature, and sunshine duration of a location can reflect its weather condition; hence, they are used to estimate PV power output using artificial neural network (ANN). A sunshine duration index, “k,” has been introduced to classify a location's weather condition. Accordingly, two weather conditions are considered based on “k,” and solar PV power estimation models are developed for both cases (Condition-I: 0
{"title":"Bright Sunshine Duration Index-Based Prediction of Solar PV Power Using ANN Approach","authors":"D. V. S. K. Rao, B. Prusty, Hareesh Myneni","doi":"10.1109/ICICCSP53532.2022.9862452","DOIUrl":"https://doi.org/10.1109/ICICCSP53532.2022.9862452","url":null,"abstract":"The grid integration of solar photovoltaic (PV) systems has recently grabbed considerable research attention. Simultaneously, the grid has been subjected to disturbances due to PV generations' variability, uncertainty, and intermittency; therefore, accurately estimating the weather-dependent PV power is imperative. The daily global solar radiation, temperature, and sunshine duration of a location can reflect its weather condition; hence, they are used to estimate PV power output using artificial neural network (ANN). A sunshine duration index, “k,” has been introduced to classify a location's weather condition. Accordingly, two weather conditions are considered based on “k,” and solar PV power estimation models are developed for both cases (Condition-I: 0<k<0.8 and condition-II: 0.8k1). The performance of the proposed ANN-based models is evaluated using error metrics, namely, mean absolute percentage error (MAPE) and relative root mean square error (RRMSE). The temperature of the considered location has resulted in a minimum error in estimating PV power output. The ANN model for 0.8k1 has resulted in a MAPE of 1.888 % with temperature as input. The ANN model for 0<k<0.8 has resulted in a MAPE of 10.599 % with temperature as input. Excellent performance is noticed with the developed forecasting models in estimating PV power. These models are helpful for feasibility studies of PV power plant installations and economic scheduling.","PeriodicalId":326163,"journal":{"name":"2022 International Conference on Intelligent Controller and Computing for Smart Power (ICICCSP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125451990","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 : 2022-07-21DOI: 10.1109/ICICCSP53532.2022.9862367
Spoorthi B, P. Pradeepa
Battery Management System is very essential part in Electric Vehicle to ensure safe operation and to obtain maximum output of battery pack. The primary source of electricity in EV's are batteries. Lithium ion battery is extensively employed for energy storage in EV. BMS will monitor the parameters and determine battery state of charge, state of health and maintains the system in accurate, reliable state and also determines the life span of a battery. This paper provides a literature review on Battery Management System for safe and reliable operation of traction batteries in Electric Vehicle.
{"title":"Review on Battery Management System in EV","authors":"Spoorthi B, P. Pradeepa","doi":"10.1109/ICICCSP53532.2022.9862367","DOIUrl":"https://doi.org/10.1109/ICICCSP53532.2022.9862367","url":null,"abstract":"Battery Management System is very essential part in Electric Vehicle to ensure safe operation and to obtain maximum output of battery pack. The primary source of electricity in EV's are batteries. Lithium ion battery is extensively employed for energy storage in EV. BMS will monitor the parameters and determine battery state of charge, state of health and maintains the system in accurate, reliable state and also determines the life span of a battery. This paper provides a literature review on Battery Management System for safe and reliable operation of traction batteries in Electric Vehicle.","PeriodicalId":326163,"journal":{"name":"2022 International Conference on Intelligent Controller and Computing for Smart Power (ICICCSP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125616382","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 : 2022-07-21DOI: 10.1109/ICICCSP53532.2022.9862520
R. Karpagam, T. A. Dheeven
In the present development, deep learning has made incredible progress in many filed including computer vision and natural language processing. Contrasted with customary artificial intelligence techniques, deep learning has a solid learning capacity and can utilize datasets for highlight extraction. In view of its practicability, deep learning turns out to be increasingly more well known for some analytic investigation works. This paper, predominantly presented a few neural networking of deep learning in electrical grid codes that have been laid out with low voltage ride through (LVRT) capacity standard necessity for the network associated PVPPs that ought to be met. Thusly, for an effective LVRT control, the quick and exact hang recognition techniques are fundamental for the framework to change from typical activity to LVRT mode, pullout mode, grid mode of operation. Deep learning is an arising area of various hidden layers of artificial intelligence for automatic learning voltage dip features in microgrid research.
{"title":"Sag Rooting Out in Grid Connected Windfarm by Deploying Deep Learning","authors":"R. Karpagam, T. A. Dheeven","doi":"10.1109/ICICCSP53532.2022.9862520","DOIUrl":"https://doi.org/10.1109/ICICCSP53532.2022.9862520","url":null,"abstract":"In the present development, deep learning has made incredible progress in many filed including computer vision and natural language processing. Contrasted with customary artificial intelligence techniques, deep learning has a solid learning capacity and can utilize datasets for highlight extraction. In view of its practicability, deep learning turns out to be increasingly more well known for some analytic investigation works. This paper, predominantly presented a few neural networking of deep learning in electrical grid codes that have been laid out with low voltage ride through (LVRT) capacity standard necessity for the network associated PVPPs that ought to be met. Thusly, for an effective LVRT control, the quick and exact hang recognition techniques are fundamental for the framework to change from typical activity to LVRT mode, pullout mode, grid mode of operation. Deep learning is an arising area of various hidden layers of artificial intelligence for automatic learning voltage dip features in microgrid research.","PeriodicalId":326163,"journal":{"name":"2022 International Conference on Intelligent Controller and Computing for Smart Power (ICICCSP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115859289","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 : 2022-07-21DOI: 10.1109/ICICCSP53532.2022.9862471
B. Prasad, Nutan Saha, G. Panda
In this work, a novel parameter extraction technique for Photo Voltaic module known as Cosine Adapted Modified Whale Optmisation Algorithm is presented. For correct mathematical modeling and for further analysis, correct parameter estimation is necessary. Many parameter extraction techniques have been reported in literature. Still many more techniques that are capable of extracting global optimized parameter during varying weather states are required to explore in varying weather conditions. Cam WOA technique is the modified form of Whale Optimisation Technique. In this work a novel parameter extraction technique based on CamWOA is proposed for parameter extraction. The proposed Hybrid BESAS Technique is tested on PV module and also in different weather condition. The effect of these extracted parameter on PV module performance is also discussed. From the analysis of simulation results it is found that the proposed Cam Woabased parameter extraction scheme is more accurate as compared to WOA technique.
{"title":"Parameter Extraction of PV Module using CamWOA Technique","authors":"B. Prasad, Nutan Saha, G. Panda","doi":"10.1109/ICICCSP53532.2022.9862471","DOIUrl":"https://doi.org/10.1109/ICICCSP53532.2022.9862471","url":null,"abstract":"In this work, a novel parameter extraction technique for Photo Voltaic module known as Cosine Adapted Modified Whale Optmisation Algorithm is presented. For correct mathematical modeling and for further analysis, correct parameter estimation is necessary. Many parameter extraction techniques have been reported in literature. Still many more techniques that are capable of extracting global optimized parameter during varying weather states are required to explore in varying weather conditions. Cam WOA technique is the modified form of Whale Optimisation Technique. In this work a novel parameter extraction technique based on CamWOA is proposed for parameter extraction. The proposed Hybrid BESAS Technique is tested on PV module and also in different weather condition. The effect of these extracted parameter on PV module performance is also discussed. From the analysis of simulation results it is found that the proposed Cam Woabased parameter extraction scheme is more accurate as compared to WOA technique.","PeriodicalId":326163,"journal":{"name":"2022 International Conference on Intelligent Controller and Computing for Smart Power (ICICCSP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132360833","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 : 2022-07-21DOI: 10.1109/ICICCSP53532.2022.9862476
Aniruddha Mohanty, Alok R. Prusty, Ravindranath Cherukuri
Early detection of a disease is a crucial task because of unavailability of proper medical facilities. Cancer is one of the critical diseases that needs early detection for survival. A cancer tumor is caused due to thousands of genetic mutations. Understanding the genetic mutations of cancer tumor is a tedious and time-consuming task. A list of genetic variations is analysed manually by a molecular pathologist. The clinical strips of indication are of nine classes, but the classification is still unknown. The objective of this implementation is to suggest a multiclass classifier which classifies the genetic mutations with respect to the clinical signs. The clinical evidences are text-evidences of gene mutations and analysed by Natural Language Processing (NLP). Various machine learning concepts like Naive Bayes, Logistic Regression, Linear Support Vector Machine, Random Forest Classifier applied on the collected dataset which contain the evidence based on genetic mutations and other clinical evidences that pathology or specialists used to classify the gene mutations. The performances of the models are analysed to get the best results. The machine learning models are implemented and analyzed with the help of gene, variance and text features. Based on the variants of gene mutation, the risk of the cancer can be detected and the medications can be prescribed accordingly.
{"title":"Cancer Tumor Detection Using Genetic Mutated Data and Machine Learning Models","authors":"Aniruddha Mohanty, Alok R. Prusty, Ravindranath Cherukuri","doi":"10.1109/ICICCSP53532.2022.9862476","DOIUrl":"https://doi.org/10.1109/ICICCSP53532.2022.9862476","url":null,"abstract":"Early detection of a disease is a crucial task because of unavailability of proper medical facilities. Cancer is one of the critical diseases that needs early detection for survival. A cancer tumor is caused due to thousands of genetic mutations. Understanding the genetic mutations of cancer tumor is a tedious and time-consuming task. A list of genetic variations is analysed manually by a molecular pathologist. The clinical strips of indication are of nine classes, but the classification is still unknown. The objective of this implementation is to suggest a multiclass classifier which classifies the genetic mutations with respect to the clinical signs. The clinical evidences are text-evidences of gene mutations and analysed by Natural Language Processing (NLP). Various machine learning concepts like Naive Bayes, Logistic Regression, Linear Support Vector Machine, Random Forest Classifier applied on the collected dataset which contain the evidence based on genetic mutations and other clinical evidences that pathology or specialists used to classify the gene mutations. The performances of the models are analysed to get the best results. The machine learning models are implemented and analyzed with the help of gene, variance and text features. Based on the variants of gene mutation, the risk of the cancer can be detected and the medications can be prescribed accordingly.","PeriodicalId":326163,"journal":{"name":"2022 International Conference on Intelligent Controller and Computing for Smart Power (ICICCSP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133390493","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 : 2022-07-21DOI: 10.1109/ICICCSP53532.2022.9862328
E. Ramanujam, Arpit Sharma, J. Hussian, Thinagaran Perumal
Indoor Air Quality (IAQ) monitoring has been a significant research domain in energy conservation. Many energy resources are required to maintain the IAQ using airconditioning or other ventilation systems. Currently, the research works highly optimize an on-demand driven energy usage depending on the occupant present inside the building. In the last decade, numerous research works have evolved for such an optimization by installing sensors and predicting occupants using machine learning techniques. This research fails to deploy non-intrusive sensors and appropriate machine learning algorithms to predict the occupancy count. Advancement in neural network techniques termed deep learning has made significant performance in recognition and cognitive tasks. Thus, this paper proposes a hybrid deep learning model that stacks the convolutional neural network (CNN) and long short term memory (LSTM) to improve the prediction rate of the occupancy count. Experimentation has been carried out in real-time multivariate sensor data for the occupancy estimation and evaluated the performance in terms of accuracy, RMSE, MAPE, and coefficients of determinants.
{"title":"Improving Indoor occupancy estimation using a hybrid CNN-LSTM approach","authors":"E. Ramanujam, Arpit Sharma, J. Hussian, Thinagaran Perumal","doi":"10.1109/ICICCSP53532.2022.9862328","DOIUrl":"https://doi.org/10.1109/ICICCSP53532.2022.9862328","url":null,"abstract":"Indoor Air Quality (IAQ) monitoring has been a significant research domain in energy conservation. Many energy resources are required to maintain the IAQ using airconditioning or other ventilation systems. Currently, the research works highly optimize an on-demand driven energy usage depending on the occupant present inside the building. In the last decade, numerous research works have evolved for such an optimization by installing sensors and predicting occupants using machine learning techniques. This research fails to deploy non-intrusive sensors and appropriate machine learning algorithms to predict the occupancy count. Advancement in neural network techniques termed deep learning has made significant performance in recognition and cognitive tasks. Thus, this paper proposes a hybrid deep learning model that stacks the convolutional neural network (CNN) and long short term memory (LSTM) to improve the prediction rate of the occupancy count. Experimentation has been carried out in real-time multivariate sensor data for the occupancy estimation and evaluated the performance in terms of accuracy, RMSE, MAPE, and coefficients of determinants.","PeriodicalId":326163,"journal":{"name":"2022 International Conference on Intelligent Controller and Computing for Smart Power (ICICCSP)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133444421","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：