Pub Date : 2021-03-05DOI: 10.1109/ICEPE50861.2021.9404534
Mamata Debbarma, P. Krishna, M. A. Rani
Owing to the increasing demand for electric power there exist an increased penetration of renewable energy sources with the grid. However, this is now the major reason for grid instability. Thus, for precise power control and grid integration of Wind Energy Conversion Systems (WECS), accurate angular grid frequency and speed signal is necessary. In this regard, a sensorless speed estimation method is proposed for wind driven SCIGs with reduced number of current sensors. Unlike the other methods reported so far, this method brings forth a sensorless method which is capable of computing all the variations in speed signal either under dynamic speed variations or ripples in speed in case of distorted grid conditions. Here, the rotor speed signal is derived from the measured stator voltages and a single stator current signal, thus facilitating high reliability and robustness in speed computation. The efficacy of the proposed scheme is shown using simulation results on a 2.2 kW SCIG in MATLAB Simulink.
{"title":"A Robust Sensorless Speed Estimation Method for SCIG based WECS with Reduced Current Sensors","authors":"Mamata Debbarma, P. Krishna, M. A. Rani","doi":"10.1109/ICEPE50861.2021.9404534","DOIUrl":"https://doi.org/10.1109/ICEPE50861.2021.9404534","url":null,"abstract":"Owing to the increasing demand for electric power there exist an increased penetration of renewable energy sources with the grid. However, this is now the major reason for grid instability. Thus, for precise power control and grid integration of Wind Energy Conversion Systems (WECS), accurate angular grid frequency and speed signal is necessary. In this regard, a sensorless speed estimation method is proposed for wind driven SCIGs with reduced number of current sensors. Unlike the other methods reported so far, this method brings forth a sensorless method which is capable of computing all the variations in speed signal either under dynamic speed variations or ripples in speed in case of distorted grid conditions. Here, the rotor speed signal is derived from the measured stator voltages and a single stator current signal, thus facilitating high reliability and robustness in speed computation. The efficacy of the proposed scheme is shown using simulation results on a 2.2 kW SCIG in MATLAB Simulink.","PeriodicalId":250203,"journal":{"name":"2020 3rd International Conference on Energy, Power and Environment: Towards Clean Energy Technologies","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130218600","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 : 2021-03-05DOI: 10.1109/ICEPE50861.2021.9404488
Pushpanjali Shadangi, S. D. Swain, P. K. Ray
In this paper, a Modified Instantaneous Reactive Power Theory (MIRPT) is proposed for DSTATCOM to abolish harmonic current and indemnify the reactive power under various load perturbations. The DSTATCOM is the building blocks of voltage source inverter with a common DC-link capacitor and a hysteresis band current controller for switching pattern generation. In order to proof the efficacy of th proposed controller based DSTATCOM for three different case studies are considered and then simulated. All simulations are executed by using MATLAB 2016. DSTATCOM performance has been found more effective to compensate harmonics under all scenarios of load disorder.
{"title":"Experimental realization of DSTATCOM for Power Quality Enhancement Under Various Load Perturbations","authors":"Pushpanjali Shadangi, S. D. Swain, P. K. Ray","doi":"10.1109/ICEPE50861.2021.9404488","DOIUrl":"https://doi.org/10.1109/ICEPE50861.2021.9404488","url":null,"abstract":"In this paper, a Modified Instantaneous Reactive Power Theory (MIRPT) is proposed for DSTATCOM to abolish harmonic current and indemnify the reactive power under various load perturbations. The DSTATCOM is the building blocks of voltage source inverter with a common DC-link capacitor and a hysteresis band current controller for switching pattern generation. In order to proof the efficacy of th proposed controller based DSTATCOM for three different case studies are considered and then simulated. All simulations are executed by using MATLAB 2016. DSTATCOM performance has been found more effective to compensate harmonics under all scenarios of load disorder.","PeriodicalId":250203,"journal":{"name":"2020 3rd International Conference on Energy, Power and Environment: Towards Clean Energy Technologies","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133965695","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 : 2021-03-05DOI: 10.1109/ICEPE50861.2021.9404445
S. Chowdhury, N. Haque, S. Chatterjee
This paper puts forward a novel approach to evaluate an equivalent circuit of oil paper insulation used in transformers from frequency domain spectroscopy measurements. In recent years, there has been considerable attention on the development of diagnostic tools for condition monitoring of power transformer insulation. Frequency Domain Spectroscopy is one such tool. A proper modelling will ensure that the parameters are influenced by insulation deterioration. Hence, if FDS data can be used for such modelling, it may lead to better interpretation of FDS data from the perspective of insulation monitoring. Considering the above-mentioned facts, a method has been proposed in this work that can model the FDS data through an equivalent circuit comprising of resistances and capacitances. The influence of moisture content on the circuit parameters have been investigated. From the investigation, a relation between the DC conductivity and moisture content of the insulation has also been proposed.
{"title":"Equivalent Circuit Modelling of Transformer Oil-Paper Insulation from Frequency Domain Spectroscopy Measurements","authors":"S. Chowdhury, N. Haque, S. Chatterjee","doi":"10.1109/ICEPE50861.2021.9404445","DOIUrl":"https://doi.org/10.1109/ICEPE50861.2021.9404445","url":null,"abstract":"This paper puts forward a novel approach to evaluate an equivalent circuit of oil paper insulation used in transformers from frequency domain spectroscopy measurements. In recent years, there has been considerable attention on the development of diagnostic tools for condition monitoring of power transformer insulation. Frequency Domain Spectroscopy is one such tool. A proper modelling will ensure that the parameters are influenced by insulation deterioration. Hence, if FDS data can be used for such modelling, it may lead to better interpretation of FDS data from the perspective of insulation monitoring. Considering the above-mentioned facts, a method has been proposed in this work that can model the FDS data through an equivalent circuit comprising of resistances and capacitances. The influence of moisture content on the circuit parameters have been investigated. From the investigation, a relation between the DC conductivity and moisture content of the insulation has also been proposed.","PeriodicalId":250203,"journal":{"name":"2020 3rd International Conference on Energy, Power and Environment: Towards Clean Energy Technologies","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133332019","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 : 2021-03-05DOI: 10.1109/icepe50861.2021.9404420
ICEPE-2020 Reviewers.
ICEPE-2020 评审员。
{"title":"ICEPE-2020 Reviewers","authors":"","doi":"10.1109/icepe50861.2021.9404420","DOIUrl":"https://doi.org/10.1109/icepe50861.2021.9404420","url":null,"abstract":"ICEPE-2020 Reviewers.","PeriodicalId":250203,"journal":{"name":"2020 3rd International Conference on Energy, Power and Environment: Towards Clean Energy Technologies","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127428841","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 : 2021-03-05DOI: 10.1109/ICEPE50861.2021.9404487
Dipak P. Bhapkar, A. Maity
This paper presents a thermoelectic energy harvesting technique by sensing the temperature gradient between the human body and the ambient condition. With a temperature difference of 2-3 °C, the voltage available at the output of the thermoelectric harvester is merely 85 mV. To make it usable to CMOS circuits, this voltage needs to be boosted using a DC-DC boost converter. The key objective of this paper is to replace a motion-activated off-chip mechanical switch that is conventionally used in the boost converter with an integrated NMOS switch for starting-up the voltage conversion process. During start-up operation, the fully integrated control circuit produces a suitable level-converted ON-OFF signal with a swing of nearly 0–300 mV for the NMOS switch from the input voltage of 85 mV. This is achieved by using a start-up assisted ring oscillator topology that produces oscillations at a very low voltage. Subsequently, a series of clock booster stages, switch capacitor-based voltage booster stages and level shifter circuit are used to increase the voltage swing of the gate-drive signal for periodic switching of the NMOS switch. The proposed solution implemented in 180 nm CMOS process is capable of boosting the very low harvested voltage to nearly 1 V without requiring any external power supply such as battery or an off-chip mechanical switch.
{"title":"Power Extraction From an Ultra-Low Input Voltage Source in a Battery-less Thermoelectric Harvester","authors":"Dipak P. Bhapkar, A. Maity","doi":"10.1109/ICEPE50861.2021.9404487","DOIUrl":"https://doi.org/10.1109/ICEPE50861.2021.9404487","url":null,"abstract":"This paper presents a thermoelectic energy harvesting technique by sensing the temperature gradient between the human body and the ambient condition. With a temperature difference of 2-3 °C, the voltage available at the output of the thermoelectric harvester is merely 85 mV. To make it usable to CMOS circuits, this voltage needs to be boosted using a DC-DC boost converter. The key objective of this paper is to replace a motion-activated off-chip mechanical switch that is conventionally used in the boost converter with an integrated NMOS switch for starting-up the voltage conversion process. During start-up operation, the fully integrated control circuit produces a suitable level-converted ON-OFF signal with a swing of nearly 0–300 mV for the NMOS switch from the input voltage of 85 mV. This is achieved by using a start-up assisted ring oscillator topology that produces oscillations at a very low voltage. Subsequently, a series of clock booster stages, switch capacitor-based voltage booster stages and level shifter circuit are used to increase the voltage swing of the gate-drive signal for periodic switching of the NMOS switch. The proposed solution implemented in 180 nm CMOS process is capable of boosting the very low harvested voltage to nearly 1 V without requiring any external power supply such as battery or an off-chip mechanical switch.","PeriodicalId":250203,"journal":{"name":"2020 3rd International Conference on Energy, Power and Environment: Towards Clean Energy Technologies","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115846074","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 : 2021-03-05DOI: 10.1109/ICEPE50861.2021.9404503
S. Samanta, Subir Datta, Satyajit Das, B. K. Roy, Amrita Ganguly
Proper detection of islanding of a microgrid and its resynchronisation is very important for the plug and play operation. A simple passive islanding detection and resynchronisation technique is proposed in this paper. The change in the voltage magnitude, frequency and phase are taken into the consideration to avoid any false triggering during the detection process. In addition, the proposed technique is designed in such a way that it can be implemented in a field-programmable gate array (FPGA) hardware using FPGA-in-Loop (FIL) platform. The complete controller is converted into VHDL, a hardware description language code. Finally, the bitstream generated from the code is implemented in FPGA hardware, Zedboard. The results obtained in terms of both islanding detection and resynchronisation using the FPGA-in-Loop simulation are found to be satisfactory.
{"title":"Design and FPGA Implementation of an Islanding Detection cum Re-synchronisation Technique for a Grid Connected Inverter in a DC Microgrid","authors":"S. Samanta, Subir Datta, Satyajit Das, B. K. Roy, Amrita Ganguly","doi":"10.1109/ICEPE50861.2021.9404503","DOIUrl":"https://doi.org/10.1109/ICEPE50861.2021.9404503","url":null,"abstract":"Proper detection of islanding of a microgrid and its resynchronisation is very important for the plug and play operation. A simple passive islanding detection and resynchronisation technique is proposed in this paper. The change in the voltage magnitude, frequency and phase are taken into the consideration to avoid any false triggering during the detection process. In addition, the proposed technique is designed in such a way that it can be implemented in a field-programmable gate array (FPGA) hardware using FPGA-in-Loop (FIL) platform. The complete controller is converted into VHDL, a hardware description language code. Finally, the bitstream generated from the code is implemented in FPGA hardware, Zedboard. The results obtained in terms of both islanding detection and resynchronisation using the FPGA-in-Loop simulation are found to be satisfactory.","PeriodicalId":250203,"journal":{"name":"2020 3rd International Conference on Energy, Power and Environment: Towards Clean Energy Technologies","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114481316","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 : 2021-03-05DOI: 10.1109/ICEPE50861.2021.9404479
Prashant Kumar, D. Bhaskar, R. Behera
A sensorless field-oriented control of PMBLDCM based on sliding mode observer is presented. FOC is a technique for regulating torque in variable frequency drive. A modified sliding-mode observer has been designed for field-oriented control of PMBLDCM powered by a voltage-source inverter in order to accomplish speed control of drives. The disturbances have been estimated taking the load torque and viscous friction coefficient into account. This article demonstrates the PMBLDCM drive sensorless FOC in two-phase operating mode to achieve torque control. Provided with the proposed algorithms, such observer is beneficial in reducing the rotor position estimation error for variable speed drives. The implemented sensorless control algorithms are implemented and evaluated in a hardware prototype. The simulation and experimental findings are presented to demonstrate the efficacy of the strategy developed.
{"title":"Sliding Mode Observer based Rotor Position Estimation with Field Oriented Control of PMBLDC Motor Drive","authors":"Prashant Kumar, D. Bhaskar, R. Behera","doi":"10.1109/ICEPE50861.2021.9404479","DOIUrl":"https://doi.org/10.1109/ICEPE50861.2021.9404479","url":null,"abstract":"A sensorless field-oriented control of PMBLDCM based on sliding mode observer is presented. FOC is a technique for regulating torque in variable frequency drive. A modified sliding-mode observer has been designed for field-oriented control of PMBLDCM powered by a voltage-source inverter in order to accomplish speed control of drives. The disturbances have been estimated taking the load torque and viscous friction coefficient into account. This article demonstrates the PMBLDCM drive sensorless FOC in two-phase operating mode to achieve torque control. Provided with the proposed algorithms, such observer is beneficial in reducing the rotor position estimation error for variable speed drives. The implemented sensorless control algorithms are implemented and evaluated in a hardware prototype. The simulation and experimental findings are presented to demonstrate the efficacy of the strategy developed.","PeriodicalId":250203,"journal":{"name":"2020 3rd International Conference on Energy, Power and Environment: Towards Clean Energy Technologies","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114855966","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 : 2021-03-05DOI: 10.1109/ICEPE50861.2021.9404463
Madan Kumar Das, S. Chauhan, Parusharamulu Buduma, K. Jana, Sukumar Ishara
This paper shows a novel hybrid Multilevel inverter configuration with less number of power electronics (8 IGBT) switches and 4 dc-link voltage source can be modified with 7 level based model (6 switeches and 2 dc sources) series connected with half h-bride to produced output voltage 21-levels, which is nesar to sinusoidal. The carrier-based level shifted PWM (LSPWM) switching pulses method are used. The proposed MLI is finally verified with the MATLAB/SIMULINK platform and developed to prototype hardware setup with RL load with symmetrical and asymmetrical configuration. In addition, the corresponding experimental results are demonstrated for the different values of modulation index with real time dspace ds1103.
{"title":"A Hybrid Novel Cascaded Asymmetrical 21-level Inverter with Reduced Switches","authors":"Madan Kumar Das, S. Chauhan, Parusharamulu Buduma, K. Jana, Sukumar Ishara","doi":"10.1109/ICEPE50861.2021.9404463","DOIUrl":"https://doi.org/10.1109/ICEPE50861.2021.9404463","url":null,"abstract":"This paper shows a novel hybrid Multilevel inverter configuration with less number of power electronics (8 IGBT) switches and 4 dc-link voltage source can be modified with 7 level based model (6 switeches and 2 dc sources) series connected with half h-bride to produced output voltage 21-levels, which is nesar to sinusoidal. The carrier-based level shifted PWM (LSPWM) switching pulses method are used. The proposed MLI is finally verified with the MATLAB/SIMULINK platform and developed to prototype hardware setup with RL load with symmetrical and asymmetrical configuration. In addition, the corresponding experimental results are demonstrated for the different values of modulation index with real time dspace ds1103.","PeriodicalId":250203,"journal":{"name":"2020 3rd International Conference on Energy, Power and Environment: Towards Clean Energy Technologies","volume":"930 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116423416","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 : 2021-03-05DOI: 10.1109/ICEPE50861.2021.9404456
Thangavel Subbaiyan, Yuvaraju Venkatachalam
Phasor measurement unit (PMU) is a device that gives a fast and accurate measurement of both current and voltage along with time synchronization. This PMU data is useful for effective monitoring and control of power systems in smart grid. To minimize the system cost, the number of PMUs required should be reduced and must be placed optimally in a power system. To achieve this objective, particle swarm optimization (PSO) based technique is presented in this paper for maximum power system observability. Also, the zero injection bus (ZIB) case is considered for further reduction in the PMU count which in turn reduces system cost. The selection of appropriate PMU location set is identified based on the system observability redundancy index (SORI). In this paper, the PSO algorithm is tested on various test systems like IEEE 9-bus, 14-bus, and 24-bus systems for normal and ZIB cases. The simulation results are compared with various MATLAB solver-based optimization techniques. The simulation results show that PSO based technique is more effective in terms of observability.
{"title":"PSO Based Optimal Phasor Measurement Unit Placement Considering Measurement Redundancy","authors":"Thangavel Subbaiyan, Yuvaraju Venkatachalam","doi":"10.1109/ICEPE50861.2021.9404456","DOIUrl":"https://doi.org/10.1109/ICEPE50861.2021.9404456","url":null,"abstract":"Phasor measurement unit (PMU) is a device that gives a fast and accurate measurement of both current and voltage along with time synchronization. This PMU data is useful for effective monitoring and control of power systems in smart grid. To minimize the system cost, the number of PMUs required should be reduced and must be placed optimally in a power system. To achieve this objective, particle swarm optimization (PSO) based technique is presented in this paper for maximum power system observability. Also, the zero injection bus (ZIB) case is considered for further reduction in the PMU count which in turn reduces system cost. The selection of appropriate PMU location set is identified based on the system observability redundancy index (SORI). In this paper, the PSO algorithm is tested on various test systems like IEEE 9-bus, 14-bus, and 24-bus systems for normal and ZIB cases. The simulation results are compared with various MATLAB solver-based optimization techniques. The simulation results show that PSO based technique is more effective in terms of observability.","PeriodicalId":250203,"journal":{"name":"2020 3rd International Conference on Energy, Power and Environment: Towards Clean Energy Technologies","volume":"3 12","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120836949","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 : 2021-03-05DOI: 10.1109/ICEPE50861.2021.9404519
Tejasri Kari, L. N, Sayeera Banu A, DhanuShree R, K. Jagannatha, S. Natarajan
Ensemble learning techniques adopt comprehensive learning methodologies that produce optimized predictions with reduced variance and bias. The structured Random Forest ensemble learning technique equips a set of weak and diverse decision trees, resulting in an active hybrid learning ensemble. Plagued with high computational complexity, Random Forest Ensemble continues to be the preferred technique when accuracy is of primary importance for learners. Efforts to accelerate the Random Forest Ensembles are in place, however failing to efficiently utilize the data transmission bandwidth between the host and the accelerator hardware. This paper provides an architectural overview of a reconfigurable accelerator based architecture of the Random Forest Ensemble with an efficient data path model for data streaming. The paper also derives the need for an accelerated parallel ensemble method by deriving the results from equivalent sequential software implementations of the algorithm. The validation of the results have been done on healthcare application involving breast cancer classification and environmental applications involving temperature prediction and fuel consumption.
{"title":"An Accelerated Approach to Parallel Ensemble Techniques Targeting Healthcare and Environmental Applications","authors":"Tejasri Kari, L. N, Sayeera Banu A, DhanuShree R, K. Jagannatha, S. Natarajan","doi":"10.1109/ICEPE50861.2021.9404519","DOIUrl":"https://doi.org/10.1109/ICEPE50861.2021.9404519","url":null,"abstract":"Ensemble learning techniques adopt comprehensive learning methodologies that produce optimized predictions with reduced variance and bias. The structured Random Forest ensemble learning technique equips a set of weak and diverse decision trees, resulting in an active hybrid learning ensemble. Plagued with high computational complexity, Random Forest Ensemble continues to be the preferred technique when accuracy is of primary importance for learners. Efforts to accelerate the Random Forest Ensembles are in place, however failing to efficiently utilize the data transmission bandwidth between the host and the accelerator hardware. This paper provides an architectural overview of a reconfigurable accelerator based architecture of the Random Forest Ensemble with an efficient data path model for data streaming. The paper also derives the need for an accelerated parallel ensemble method by deriving the results from equivalent sequential software implementations of the algorithm. The validation of the results have been done on healthcare application involving breast cancer classification and environmental applications involving temperature prediction and fuel consumption.","PeriodicalId":250203,"journal":{"name":"2020 3rd International Conference on Energy, Power and Environment: Towards Clean Energy Technologies","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125079213","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：