Pub Date : 2014-07-24DOI: 10.1109/AICERA.2014.6908216
S. Devadethan, Geevarghese Titus, S. Purushothaman
Detection of human face from an image is a very difficult process. There are many reasons that affect the detection process such as lighting condition, shadows, facial expression etc. Thus facial feature extraction itself becomes a difficult task. In order to propose an efficient method for facial feature extraction, we used the characteristic features of nostrils, eyes lips etc. In our method we assume that frontal face image is readily available. At first face regions detected by detecting the eye regions. After detecting the face region other feature points such as nostril, corners of eyes, corners of lips etc are extracted. At first eye pairs are obtained by finding and verifying possible eye regions. After detecting the eye regions, the distance between the eyes is used to find a possible face candidate. Next, the face is divided into different regions and facial features are extracted from the corresponding regions.
{"title":"Face detection and facial feature extraction based on a fusion of knowledge based method and morphological image processing","authors":"S. Devadethan, Geevarghese Titus, S. Purushothaman","doi":"10.1109/AICERA.2014.6908216","DOIUrl":"https://doi.org/10.1109/AICERA.2014.6908216","url":null,"abstract":"Detection of human face from an image is a very difficult process. There are many reasons that affect the detection process such as lighting condition, shadows, facial expression etc. Thus facial feature extraction itself becomes a difficult task. In order to propose an efficient method for facial feature extraction, we used the characteristic features of nostrils, eyes lips etc. In our method we assume that frontal face image is readily available. At first face regions detected by detecting the eye regions. After detecting the face region other feature points such as nostril, corners of eyes, corners of lips etc are extracted. At first eye pairs are obtained by finding and verifying possible eye regions. After detecting the eye regions, the distance between the eyes is used to find a possible face candidate. Next, the face is divided into different regions and facial features are extracted from the corresponding regions.","PeriodicalId":425226,"journal":{"name":"2014 Annual International Conference on Emerging Research Areas: Magnetics, Machines and Drives (AICERA/iCMMD)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126161952","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 : 2014-07-24DOI: 10.1109/AICERA.2014.6908251
Krishnapriya, C. K. Sakker Hussain
Solar energy and wind energy are highly depended on the climate. So, the available power is discrete. Hence two different renewable energy sources could be combined to get a demanded load power. Hence the system is known as a hybrid power system. Instead of several independent converter, in hybrid system multiple input converter have simple circuit and low cost. But multiple input converters have many modes of operation. So the control circuit is complicated. This paper explains an OCC (one cycle control) method for MIC (Multiple Input Converter) which is simple compared to other control methods. The working of OCC is explained in detail. In one cycle control no current regulator is needed and the design in different mode is simple. Thus, control design is simple.
{"title":"One cycle control for hybrid power system","authors":"Krishnapriya, C. K. Sakker Hussain","doi":"10.1109/AICERA.2014.6908251","DOIUrl":"https://doi.org/10.1109/AICERA.2014.6908251","url":null,"abstract":"Solar energy and wind energy are highly depended on the climate. So, the available power is discrete. Hence two different renewable energy sources could be combined to get a demanded load power. Hence the system is known as a hybrid power system. Instead of several independent converter, in hybrid system multiple input converter have simple circuit and low cost. But multiple input converters have many modes of operation. So the control circuit is complicated. This paper explains an OCC (one cycle control) method for MIC (Multiple Input Converter) which is simple compared to other control methods. The working of OCC is explained in detail. In one cycle control no current regulator is needed and the design in different mode is simple. Thus, control design is simple.","PeriodicalId":425226,"journal":{"name":"2014 Annual International Conference on Emerging Research Areas: Magnetics, Machines and Drives (AICERA/iCMMD)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114277072","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 : 2014-07-24DOI: 10.1109/AICERA.2014.6908168
Swati Padmanabhan, Suraj Rajamani
In most earth leakage applications, the detection of a current leakage to the earth is carried out using a current transducer, called a core balance current transformer (CBCT). The phases and neutral conductors pass through a toroidal CBCT. The design procedure of the CBCT is thus a very important aspect in accurately determining an earth leakage. This paper describes an analytical model to design a CBCT for such an application. The algorithm predicts the operating permeability of the core using an iterative method removing the ambiguity brought on by assuming a constant permeability value. A comparison of its effectiveness with a finite element analysis (FEA) and actual test results for a specific material is also compiled.
{"title":"A novel analytical approach to design a core balance current transformer for earth leakage application","authors":"Swati Padmanabhan, Suraj Rajamani","doi":"10.1109/AICERA.2014.6908168","DOIUrl":"https://doi.org/10.1109/AICERA.2014.6908168","url":null,"abstract":"In most earth leakage applications, the detection of a current leakage to the earth is carried out using a current transducer, called a core balance current transformer (CBCT). The phases and neutral conductors pass through a toroidal CBCT. The design procedure of the CBCT is thus a very important aspect in accurately determining an earth leakage. This paper describes an analytical model to design a CBCT for such an application. The algorithm predicts the operating permeability of the core using an iterative method removing the ambiguity brought on by assuming a constant permeability value. A comparison of its effectiveness with a finite element analysis (FEA) and actual test results for a specific material is also compiled.","PeriodicalId":425226,"journal":{"name":"2014 Annual International Conference on Emerging Research Areas: Magnetics, Machines and Drives (AICERA/iCMMD)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126949698","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 : 2014-07-24DOI: 10.1109/AICERA.2014.6908181
A. Sah, Kalnana Chaudhary, V. V. Ratnam
Since energy conservation is one of the important issue now days and making our planet pollution free. For these purposes researchers are suggesting alternatives. Battery fed motor vehicles is one of the emerging option rather than conventional fuel vehicles. Bidirectional DC-DC converters are now mostly used in electric vehicles. The main reason behind this is to operate motor in two quadrants as motoring and regenerative for making efficient operation. Bidirectional DC-DC converter consists of buck and boost converter. During motoring mode energy is supplied through a battery and in regenerative mode battery is charged through a DC link created. This paper primarily gives attention on control strategy used for operation. In this gate complimentary control used to trigger initially turned off switch and divert current through anti parallel connected diode of initially active switch so that main switch can be triggered under zero voltage switching.
{"title":"Non-isolated multiphase buck-boost converter design for electric vehicle applications","authors":"A. Sah, Kalnana Chaudhary, V. V. Ratnam","doi":"10.1109/AICERA.2014.6908181","DOIUrl":"https://doi.org/10.1109/AICERA.2014.6908181","url":null,"abstract":"Since energy conservation is one of the important issue now days and making our planet pollution free. For these purposes researchers are suggesting alternatives. Battery fed motor vehicles is one of the emerging option rather than conventional fuel vehicles. Bidirectional DC-DC converters are now mostly used in electric vehicles. The main reason behind this is to operate motor in two quadrants as motoring and regenerative for making efficient operation. Bidirectional DC-DC converter consists of buck and boost converter. During motoring mode energy is supplied through a battery and in regenerative mode battery is charged through a DC link created. This paper primarily gives attention on control strategy used for operation. In this gate complimentary control used to trigger initially turned off switch and divert current through anti parallel connected diode of initially active switch so that main switch can be triggered under zero voltage switching.","PeriodicalId":425226,"journal":{"name":"2014 Annual International Conference on Emerging Research Areas: Magnetics, Machines and Drives (AICERA/iCMMD)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130593875","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 : 2014-07-24DOI: 10.1109/AICERA.2014.6908161
R. Nirmalkumar, B. Sreedhar, G. Padmakumar, K. Rajan
Centrifugal Pumps are used in the primary and secondary heat transport systems of fast reactors for pumping liquid sodium. Lubrication oil leakage from the conventional bearings used in these pumps is a potential threat to cause reactivity changes which could result in extended reactor shut down. Actively controlled magnetic bearings which do not require lubrication is an excellent alternative to conventional bearings in overcoming these problems. The present work deals with the development of “Thrust and Radial Active” magnetic bearings for a small centrifugal sodium pump of 50 m3/h capacity, which controls both the axial and radial movements of rotor. The developed active magnetic bearing takes a thrust load of around 100kg, and a nominal radial load for the existing vertically-configured shaft system of the centrifugal pump. Performance of the active magnetic bearings has been tested successfully by running the pump up to the full operating speed of 2900 rpm and the measured vibration levels were well within the allowable limits of ISO 14839.
{"title":"Active magnetic bearings for sodium pump","authors":"R. Nirmalkumar, B. Sreedhar, G. Padmakumar, K. Rajan","doi":"10.1109/AICERA.2014.6908161","DOIUrl":"https://doi.org/10.1109/AICERA.2014.6908161","url":null,"abstract":"Centrifugal Pumps are used in the primary and secondary heat transport systems of fast reactors for pumping liquid sodium. Lubrication oil leakage from the conventional bearings used in these pumps is a potential threat to cause reactivity changes which could result in extended reactor shut down. Actively controlled magnetic bearings which do not require lubrication is an excellent alternative to conventional bearings in overcoming these problems. The present work deals with the development of “Thrust and Radial Active” magnetic bearings for a small centrifugal sodium pump of 50 m3/h capacity, which controls both the axial and radial movements of rotor. The developed active magnetic bearing takes a thrust load of around 100kg, and a nominal radial load for the existing vertically-configured shaft system of the centrifugal pump. Performance of the active magnetic bearings has been tested successfully by running the pump up to the full operating speed of 2900 rpm and the measured vibration levels were well within the allowable limits of ISO 14839.","PeriodicalId":425226,"journal":{"name":"2014 Annual International Conference on Emerging Research Areas: Magnetics, Machines and Drives (AICERA/iCMMD)","volume":"113 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128712252","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 : 2014-07-24DOI: 10.1109/AICERA.2014.6908167
Y. Raut, V. Katkar, S. Sarode
In a Vehicular Ad Hoc Network (VANET), before reaching a potentially dangerous zone on road, wireless Collision Avoidance (CA) system issues warnings to drivers. Global positioning system can be used to get the vehicle position and which can be shared with other vehicles in the network. In case of emergency not all the vehicle get affected so broadcasting of alert packet is not feasible rather it has to be multicast. Calculating the list of relative vehicles position is depending on the travelling direction, bearing angle and the distance. Recent studies have shown that sparse vehicle traffic leads to network fragmentation which poses crucial research challenge for safety application. Hence in this paper we propose a system which will find out the relative position between multiple vehicles by using Great Circle Algorithm. The improvement in VANET connectivity is made by road side unit which will manage all the vehicle information and detect the failure vehicle and calculate the detail of the vehicles that get affected by the failure vehicle and multicast alert packet to identified vehicles. This will avoid the broadcasting problem.
{"title":"Early alert system using relative positioning in Vehicular Ad-hoc Network","authors":"Y. Raut, V. Katkar, S. Sarode","doi":"10.1109/AICERA.2014.6908167","DOIUrl":"https://doi.org/10.1109/AICERA.2014.6908167","url":null,"abstract":"In a Vehicular Ad Hoc Network (VANET), before reaching a potentially dangerous zone on road, wireless Collision Avoidance (CA) system issues warnings to drivers. Global positioning system can be used to get the vehicle position and which can be shared with other vehicles in the network. In case of emergency not all the vehicle get affected so broadcasting of alert packet is not feasible rather it has to be multicast. Calculating the list of relative vehicles position is depending on the travelling direction, bearing angle and the distance. Recent studies have shown that sparse vehicle traffic leads to network fragmentation which poses crucial research challenge for safety application. Hence in this paper we propose a system which will find out the relative position between multiple vehicles by using Great Circle Algorithm. The improvement in VANET connectivity is made by road side unit which will manage all the vehicle information and detect the failure vehicle and calculate the detail of the vehicles that get affected by the failure vehicle and multicast alert packet to identified vehicles. This will avoid the broadcasting problem.","PeriodicalId":425226,"journal":{"name":"2014 Annual International Conference on Emerging Research Areas: Magnetics, Machines and Drives (AICERA/iCMMD)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128932393","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 : 2014-07-24DOI: 10.1109/AICERA.2014.6908277
Jagendra Singh, Swati Rani, Rohini, Amrita Parida
Bones are the integral part of human body that shows piezoelectric properties. It means that when mechanical stress or mechanical excitation is applied to human body, this mechanical excitation directly affects the bones in body. Due to this excitation bones produce a current within itself which is a human bone property called as piezoelectric effect also known piezoelectricity. It is ability of certain materials for generating AC voltage when it is subjected to mechanical excitation or vibration. Human bones are made up of piezoelectric material thereby when human comes under mechanical excitation, AC voltage sets up in whole body due to piezoelectric effect in bones. A current starts flowing in all parts of body thereby the aim of our research is to determine the electrical conductivity in human body to establish the maximum voltage of 30-60 volts so that this produced piezoelectric charge could be stored in our designed human electricity sensor device. The major goal of our paper is to find out nature and amplitude of produced electric voltage signal waveform from human body and make it storable.
{"title":"Generation of piezoelectricity from the human body","authors":"Jagendra Singh, Swati Rani, Rohini, Amrita Parida","doi":"10.1109/AICERA.2014.6908277","DOIUrl":"https://doi.org/10.1109/AICERA.2014.6908277","url":null,"abstract":"Bones are the integral part of human body that shows piezoelectric properties. It means that when mechanical stress or mechanical excitation is applied to human body, this mechanical excitation directly affects the bones in body. Due to this excitation bones produce a current within itself which is a human bone property called as piezoelectric effect also known piezoelectricity. It is ability of certain materials for generating AC voltage when it is subjected to mechanical excitation or vibration. Human bones are made up of piezoelectric material thereby when human comes under mechanical excitation, AC voltage sets up in whole body due to piezoelectric effect in bones. A current starts flowing in all parts of body thereby the aim of our research is to determine the electrical conductivity in human body to establish the maximum voltage of 30-60 volts so that this produced piezoelectric charge could be stored in our designed human electricity sensor device. The major goal of our paper is to find out nature and amplitude of produced electric voltage signal waveform from human body and make it storable.","PeriodicalId":425226,"journal":{"name":"2014 Annual International Conference on Emerging Research Areas: Magnetics, Machines and Drives (AICERA/iCMMD)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133284039","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 : 2014-07-24DOI: 10.1109/AICERA.2014.6908247
D. Sai Krishna Kanth, M. Padma Lalitha
Now a days the usage of electrical power is increasing enormously and it is also becoming highly impossible in imagining the world without the electricity. So here is one of the techniques for maintaining the continuous supply i.e., by generating the electrical power in the distribution premises. Distribution Generation has been growing rapidly in power systems due to their capable solution for the issues like power system deregulation, to meet the demand of electric power and shortages in the electric power transmission capacities. Concept of Distributed generation (DG) plays an effective key role in the distribution networks. Among many of their merits, real power loss reduction, THD reduction voltage sag and voltage swell reduction and voltage profile improvement can be the salient merits of DG. Improper location of DG sources in to the electrical power system would lead to the increase energy or power loss, and also jeopardize the system operation. In this paper, determination of DG location is done by using Sensitivity Analysis and corresponding DG size is found by using Particle Swarm Optimization (PSO) to mitigate the real power losses in the distribution networks. Standard IEEE 15 bus test system is chosen to test the proposed PSO - Sensitivity Analysis approach. The results so obtained show the enhancement of voltage profile, reduction of real power loss using MATLAB.
{"title":"Mitigation of real power loss, THD & enhancement of voltage profile with optimal DG allocation using PSO & sensitivity analysis","authors":"D. Sai Krishna Kanth, M. Padma Lalitha","doi":"10.1109/AICERA.2014.6908247","DOIUrl":"https://doi.org/10.1109/AICERA.2014.6908247","url":null,"abstract":"Now a days the usage of electrical power is increasing enormously and it is also becoming highly impossible in imagining the world without the electricity. So here is one of the techniques for maintaining the continuous supply i.e., by generating the electrical power in the distribution premises. Distribution Generation has been growing rapidly in power systems due to their capable solution for the issues like power system deregulation, to meet the demand of electric power and shortages in the electric power transmission capacities. Concept of Distributed generation (DG) plays an effective key role in the distribution networks. Among many of their merits, real power loss reduction, THD reduction voltage sag and voltage swell reduction and voltage profile improvement can be the salient merits of DG. Improper location of DG sources in to the electrical power system would lead to the increase energy or power loss, and also jeopardize the system operation. In this paper, determination of DG location is done by using Sensitivity Analysis and corresponding DG size is found by using Particle Swarm Optimization (PSO) to mitigate the real power losses in the distribution networks. Standard IEEE 15 bus test system is chosen to test the proposed PSO - Sensitivity Analysis approach. The results so obtained show the enhancement of voltage profile, reduction of real power loss using MATLAB.","PeriodicalId":425226,"journal":{"name":"2014 Annual International Conference on Emerging Research Areas: Magnetics, Machines and Drives (AICERA/iCMMD)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124510575","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 : 2014-07-24DOI: 10.1109/AICERA.2014.6908262
Irfan Habeeb, E. K. Ismail
Linear motors is a class of motors which produces linear motion. In general we deal with rotational type motors. In these paper position control of linear motors is discussed. The linear motors are very much expensive and difficult to control. To deal with such problems a different type of linear motor is discussed here namely linear variable reluctance motor. Linear variable reluctance motor comprises the characters of linear synchronous and linear switched type motors. Linear variable reluctance motors is easy to control compared to linear motors. Another aspect of these paper is that the problems associated with conventional sliding mode. The major drawback of conventional sliding mode controller is that the system is sensitive to external disturbances prior reaching the sliding surface. Also the output position is affected by chattering when the system reach the sliding surface. To eliminate these two drawbacks here introduce a controller called adaptive sliding mode controller. Also here a comparison is made between sliding mode and model predictive controller.
{"title":"Adaptive sliding mode and model predictive control of linear variable reluctance motor","authors":"Irfan Habeeb, E. K. Ismail","doi":"10.1109/AICERA.2014.6908262","DOIUrl":"https://doi.org/10.1109/AICERA.2014.6908262","url":null,"abstract":"Linear motors is a class of motors which produces linear motion. In general we deal with rotational type motors. In these paper position control of linear motors is discussed. The linear motors are very much expensive and difficult to control. To deal with such problems a different type of linear motor is discussed here namely linear variable reluctance motor. Linear variable reluctance motor comprises the characters of linear synchronous and linear switched type motors. Linear variable reluctance motors is easy to control compared to linear motors. Another aspect of these paper is that the problems associated with conventional sliding mode. The major drawback of conventional sliding mode controller is that the system is sensitive to external disturbances prior reaching the sliding surface. Also the output position is affected by chattering when the system reach the sliding surface. To eliminate these two drawbacks here introduce a controller called adaptive sliding mode controller. Also here a comparison is made between sliding mode and model predictive controller.","PeriodicalId":425226,"journal":{"name":"2014 Annual International Conference on Emerging Research Areas: Magnetics, Machines and Drives (AICERA/iCMMD)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122915821","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 : 2014-07-24DOI: 10.1109/AICERA.2014.6908217
Ketan R. Gandhi, K. Badgujar
Deformations in windings can be diagnosed by a reliable and powerful method called sweep frequency response analysis (SFRA). In this work the deviation in the frequency response plots is derived in terms of statistical indicators. Nine statistical indicators have been used for the purpose. These indicators, then, complemented using artificial neural network approach, to derive a useful conclusion regarding the deviation based on the frequency responses. Winding deformation case data along with healthy transformer case data have been used to train a multilayer feed-forward neural network with the backpropagation algorithm. The trained neural network can help an expert to analyse statistical indicators to verify the level of deviation and in turn the level of deformation.
{"title":"Artificial neural network based identification of deviation in frequency response of power transformer windings","authors":"Ketan R. Gandhi, K. Badgujar","doi":"10.1109/AICERA.2014.6908217","DOIUrl":"https://doi.org/10.1109/AICERA.2014.6908217","url":null,"abstract":"Deformations in windings can be diagnosed by a reliable and powerful method called sweep frequency response analysis (SFRA). In this work the deviation in the frequency response plots is derived in terms of statistical indicators. Nine statistical indicators have been used for the purpose. These indicators, then, complemented using artificial neural network approach, to derive a useful conclusion regarding the deviation based on the frequency responses. Winding deformation case data along with healthy transformer case data have been used to train a multilayer feed-forward neural network with the backpropagation algorithm. The trained neural network can help an expert to analyse statistical indicators to verify the level of deviation and in turn the level of deformation.","PeriodicalId":425226,"journal":{"name":"2014 Annual International Conference on Emerging Research Areas: Magnetics, Machines and Drives (AICERA/iCMMD)","volume":"113 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121246963","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：