Pub Date : 2020-11-16DOI: 10.1109/CEFC46938.2020.9451418
Juliano R. da Silva, Pablo S. Paganoto, R. Graeff, C. M. D. Da Rocha, Marcelo L. Bernartt, Christiano W. dos Santos
The temperature rise in the transformers is caused by the dissipation of internal losses. The main sources of heat in the power transformers are those generated at full load, caused mainly by the applied current and those originated by the magnetic field. The consideration of the magnetic field in the loss estimate is necessary, as it can reach high values depending on the design and make the calculation of the internal temperature imprecise. To reduce the imprecision in the results, in this work the losses are calculated using the Finite Element Method (FEM) considering the temperature of the conductor, coupled with the dynamic analysis of fluids. 2D and 3D approximations with simple and sectioned cables were used to make comparisons with the calculation modes, analyzing the influence of the temperature effect on the additional losses created by eddy current and compared with the analytical form presented in the specialized literature. The methodology was applied to a 470 MVA power rate autotransformer with ODAF cooling mode. The results show that Foucault losses can reach high values, depending on the project, and the use of sectioned cables that allow the reduction of losses and the increase of the winding temperature must be considered.
{"title":"Analysis of Methods Eddy Current Loss Estimation in Power Transformer Windings with Multiphysical Consideration (Electromagnetic and Fluid Dynamic)","authors":"Juliano R. da Silva, Pablo S. Paganoto, R. Graeff, C. M. D. Da Rocha, Marcelo L. Bernartt, Christiano W. dos Santos","doi":"10.1109/CEFC46938.2020.9451418","DOIUrl":"https://doi.org/10.1109/CEFC46938.2020.9451418","url":null,"abstract":"The temperature rise in the transformers is caused by the dissipation of internal losses. The main sources of heat in the power transformers are those generated at full load, caused mainly by the applied current and those originated by the magnetic field. The consideration of the magnetic field in the loss estimate is necessary, as it can reach high values depending on the design and make the calculation of the internal temperature imprecise. To reduce the imprecision in the results, in this work the losses are calculated using the Finite Element Method (FEM) considering the temperature of the conductor, coupled with the dynamic analysis of fluids. 2D and 3D approximations with simple and sectioned cables were used to make comparisons with the calculation modes, analyzing the influence of the temperature effect on the additional losses created by eddy current and compared with the analytical form presented in the specialized literature. The methodology was applied to a 470 MVA power rate autotransformer with ODAF cooling mode. The results show that Foucault losses can reach high values, depending on the project, and the use of sectioned cables that allow the reduction of losses and the increase of the winding temperature must be considered.","PeriodicalId":439411,"journal":{"name":"2020 IEEE 19th Biennial Conference on Electromagnetic Field Computation (CEFC)","volume":"222 5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122883819","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 : 2020-11-16DOI: 10.1109/CEFC46938.2020.9451442
G. Lossa, O. Deblecker, Z. De Grève, C. Geuzaine
In this work, fast stochastic surrogate models are derived for extracting RL parameters of wound inductors using the Finite Element method. To this end, the Representative Volume Element (RVE) technique is employed to convert the geometrical uncertainties (e.g. due to conductor positions in the winding window) into material uncertainties (complex permeability and conductivity). The dimensionality of the stochastic input space is in that way reduced, thereby allowing the use of the Polynomial Chaos Expansion (PCE) technique for building the stochastic surrogate.
{"title":"Building Fast Stochastic Surrogate Models for Extracting RL Parameters of Wound Inductors Modeled Using FEM","authors":"G. Lossa, O. Deblecker, Z. De Grève, C. Geuzaine","doi":"10.1109/CEFC46938.2020.9451442","DOIUrl":"https://doi.org/10.1109/CEFC46938.2020.9451442","url":null,"abstract":"In this work, fast stochastic surrogate models are derived for extracting RL parameters of wound inductors using the Finite Element method. To this end, the Representative Volume Element (RVE) technique is employed to convert the geometrical uncertainties (e.g. due to conductor positions in the winding window) into material uncertainties (complex permeability and conductivity). The dimensionality of the stochastic input space is in that way reduced, thereby allowing the use of the Polynomial Chaos Expansion (PCE) technique for building the stochastic surrogate.","PeriodicalId":439411,"journal":{"name":"2020 IEEE 19th Biennial Conference on Electromagnetic Field Computation (CEFC)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124274533","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 : 2020-11-16DOI: 10.1109/CEFC46938.2020.9451308
Hiroto Tadano, Shota Ishikawa
Saddle point problems appear in many scientific and engineering applications. Hence, it is important to solve them fast with high accuracy. We have proposed the method for saddle point problems using block structure in order to solve them fast. In our proposed method, first, a linear system with multiple right-hand sides is solved instead of solving the saddle point problems directly. After that, a small linear system with a dense matrix is solved. In our previous work, it has been observed that our method is faster than the conventional approach, but the accuracy of the obtained approximate solutions is worse than the conventional one. In this paper, we propose to improve the accuracy of the approximate solutions of the saddle point problems by improving the accuracy of the solution of the small linear system using the mixed precision iterative refinement technique. Numerical experiments illustrate that the proposed approach improves the accuracy of the approximate solutions to the same extent as the conventional approach.
{"title":"Accuracy Improvement of Approximate Solutions Generated by the Method for Solving Saddle Point Problems Using Block Structure","authors":"Hiroto Tadano, Shota Ishikawa","doi":"10.1109/CEFC46938.2020.9451308","DOIUrl":"https://doi.org/10.1109/CEFC46938.2020.9451308","url":null,"abstract":"Saddle point problems appear in many scientific and engineering applications. Hence, it is important to solve them fast with high accuracy. We have proposed the method for saddle point problems using block structure in order to solve them fast. In our proposed method, first, a linear system with multiple right-hand sides is solved instead of solving the saddle point problems directly. After that, a small linear system with a dense matrix is solved. In our previous work, it has been observed that our method is faster than the conventional approach, but the accuracy of the obtained approximate solutions is worse than the conventional one. In this paper, we propose to improve the accuracy of the approximate solutions of the saddle point problems by improving the accuracy of the solution of the small linear system using the mixed precision iterative refinement technique. Numerical experiments illustrate that the proposed approach improves the accuracy of the approximate solutions to the same extent as the conventional approach.","PeriodicalId":439411,"journal":{"name":"2020 IEEE 19th Biennial Conference on Electromagnetic Field Computation (CEFC)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121590249","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 : 2020-11-16DOI: 10.1109/CEFC46938.2020.9451479
Wenjing Zhang, Yanliang Xu, Ming-lun Sun
Transverse flux permanent magnet motor (TFPM) and axial flux permanent magnet motor (AFPM) are both new types of highperformance motor and have been used in various applications. However, both of them are constrained by the cogging torque that leads to a large torque ripple. In view of a novel disk transverse flux motor (DTFM) which combines the characteristics of both transverse flux and axial flux permanent magnet motor, the research on its cogging torque is proceeded in this paper. First of all, the cogging torque of the DTFM is analyzed and expanded as the form of the Fourier series. In order to reduce its cogging torque, a method of unalignment of stator disks is introduced. Then, an analysis method based on Schwarz-Christoffel mapping (SC-based method) is extended to calculate the cogging torque of DTFM with unaligned stator disks. The cogging torque of the DTFM with unaligned stator disks can be reduced by 83% compared with the initial DTFM. Finally, a prototype DTFM with unaligned stator disks is manufactured and tested.
{"title":"Cogging Torque Reduction for a Novel Disk Transverse Flux Permanent Magnet Motor","authors":"Wenjing Zhang, Yanliang Xu, Ming-lun Sun","doi":"10.1109/CEFC46938.2020.9451479","DOIUrl":"https://doi.org/10.1109/CEFC46938.2020.9451479","url":null,"abstract":"Transverse flux permanent magnet motor (TFPM) and axial flux permanent magnet motor (AFPM) are both new types of highperformance motor and have been used in various applications. However, both of them are constrained by the cogging torque that leads to a large torque ripple. In view of a novel disk transverse flux motor (DTFM) which combines the characteristics of both transverse flux and axial flux permanent magnet motor, the research on its cogging torque is proceeded in this paper. First of all, the cogging torque of the DTFM is analyzed and expanded as the form of the Fourier series. In order to reduce its cogging torque, a method of unalignment of stator disks is introduced. Then, an analysis method based on Schwarz-Christoffel mapping (SC-based method) is extended to calculate the cogging torque of DTFM with unaligned stator disks. The cogging torque of the DTFM with unaligned stator disks can be reduced by 83% compared with the initial DTFM. Finally, a prototype DTFM with unaligned stator disks is manufactured and tested.","PeriodicalId":439411,"journal":{"name":"2020 IEEE 19th Biennial Conference on Electromagnetic Field Computation (CEFC)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129125410","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 : 2020-11-16DOI: 10.1109/CEFC46938.2020.9451363
Yuhe Fu, Shiyou Yang
The environmental electromagnetic field consists of different frequency components with very low energy densities. In this regard, an ideal miniature energy harvester should be a multi-resonance frequency, wide-frequency band, generally called multiband, high- gain one in order to collect enough electromagnetic energy from environmental electromagnetic fields. In this point of view, a microstrip multiband rectenna is proposed for miniature energy harvesters. Since the tuning of the physical parameters of the proposed rectenna is not an easy task considering the requirements on a multiband and a high gain, an improved particle swarm optimization (PSO) method is proposed. The numerical results have demonstrated that the optimized rectenna obtains high gains in multi-resonance frequency band.
{"title":"An Improved PSO for Design Optimizations of a Multiband Rectenna for Miniature Energy Harvester","authors":"Yuhe Fu, Shiyou Yang","doi":"10.1109/CEFC46938.2020.9451363","DOIUrl":"https://doi.org/10.1109/CEFC46938.2020.9451363","url":null,"abstract":"The environmental electromagnetic field consists of different frequency components with very low energy densities. In this regard, an ideal miniature energy harvester should be a multi-resonance frequency, wide-frequency band, generally called multiband, high- gain one in order to collect enough electromagnetic energy from environmental electromagnetic fields. In this point of view, a microstrip multiband rectenna is proposed for miniature energy harvesters. Since the tuning of the physical parameters of the proposed rectenna is not an easy task considering the requirements on a multiband and a high gain, an improved particle swarm optimization (PSO) method is proposed. The numerical results have demonstrated that the optimized rectenna obtains high gains in multi-resonance frequency band.","PeriodicalId":439411,"journal":{"name":"2020 IEEE 19th Biennial Conference on Electromagnetic Field Computation (CEFC)","volume":"252 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121410562","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 : 2020-11-16DOI: 10.1109/cefc46938.2020.9451409
Mingqiao Wang, Bin Yu, Chengde Tong, Guangyuan Oiao, Faliang Liu, Shijie Yang, P. Zheng
Variable-flux machine (VFM) is a promising candidate for wide-speed-range applications, such as electric vehicle, numerical control machine and railway traction. Magnetization-regulation range, which is the ratio of back electromotive forces at forward and reverse magnetization states, is an important performance of VFM. In this paper, the magnetization-regulation performance of a parallel VFM with flux barrier is analyzed, and the influences of magnetization current, split ratio, the geometries of PM pole, and the position and shape of flux barrier on the magnetization-regulation range of VFM are investigated. The best magnetization angle of VFM is explored, and a control method of forward magnetization is proposed. With the sample data obtained by finite element method, Kriging surrogate model of VFM is established to save optimization time, which is proven with good accuracy. The particle swarm optimization algorithm is utilized for optimizing the forward magnetization effect of VFM, and the optimal scheme is obtained, whose average flux density of AlNiCo PM at forward magnetization state is increased to 1.151T. The improvement measures and optimization method applied in this paper are proven effective in improving magnetization-regulation performance.
{"title":"Optimization on Magnetization-Regulation Performance of a Variable-Flux Machine with Parallel Permanent Magnets","authors":"Mingqiao Wang, Bin Yu, Chengde Tong, Guangyuan Oiao, Faliang Liu, Shijie Yang, P. Zheng","doi":"10.1109/cefc46938.2020.9451409","DOIUrl":"https://doi.org/10.1109/cefc46938.2020.9451409","url":null,"abstract":"Variable-flux machine (VFM) is a promising candidate for wide-speed-range applications, such as electric vehicle, numerical control machine and railway traction. Magnetization-regulation range, which is the ratio of back electromotive forces at forward and reverse magnetization states, is an important performance of VFM. In this paper, the magnetization-regulation performance of a parallel VFM with flux barrier is analyzed, and the influences of magnetization current, split ratio, the geometries of PM pole, and the position and shape of flux barrier on the magnetization-regulation range of VFM are investigated. The best magnetization angle of VFM is explored, and a control method of forward magnetization is proposed. With the sample data obtained by finite element method, Kriging surrogate model of VFM is established to save optimization time, which is proven with good accuracy. The particle swarm optimization algorithm is utilized for optimizing the forward magnetization effect of VFM, and the optimal scheme is obtained, whose average flux density of AlNiCo PM at forward magnetization state is increased to 1.151T. The improvement measures and optimization method applied in this paper are proven effective in improving magnetization-regulation performance.","PeriodicalId":439411,"journal":{"name":"2020 IEEE 19th Biennial Conference on Electromagnetic Field Computation (CEFC)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117331845","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 : 2020-11-16DOI: 10.1109/CEFC46938.2020.9451281
P. Cambareri, L. Di Rienzo
This paper proposes a method to calculate Joule and hysteresis losses of thin conductive and magnetic shields in two dimensions. A known BEM formulation enforcing the Thin Layer Impedance Boundary Conditions (TLIBCs) is extended to problems involving multiple closed shields. Analytical formulae for the losses are derived. A typical geometry of shields for a three-phase conductor system is used to show that a good accuracy can be achieved.
{"title":"2D BEM Computation of Power Losses in Multiple Thin Conductive Shields","authors":"P. Cambareri, L. Di Rienzo","doi":"10.1109/CEFC46938.2020.9451281","DOIUrl":"https://doi.org/10.1109/CEFC46938.2020.9451281","url":null,"abstract":"This paper proposes a method to calculate Joule and hysteresis losses of thin conductive and magnetic shields in two dimensions. A known BEM formulation enforcing the Thin Layer Impedance Boundary Conditions (TLIBCs) is extended to problems involving multiple closed shields. Analytical formulae for the losses are derived. A typical geometry of shields for a three-phase conductor system is used to show that a good accuracy can be achieved.","PeriodicalId":439411,"journal":{"name":"2020 IEEE 19th Biennial Conference on Electromagnetic Field Computation (CEFC)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125933221","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 : 2020-11-16DOI: 10.1109/CEFC46938.2020.9451362
Arbaaz Khan, M. H. Mohammadi, V. Ghorbanian, D. Lowther
This paper explores methods to extend a trained deep neural network for predicting efficiency maps to work on different motor drive topologies. This procedure reduces the computation cost associated with training deep networks by transferring knowledge over similar tasks handled by the deep networks. Two types of synchronous AC machines, including a flat-type interior and a surface-mounted permanent magnet motor are tested over their entire torque-speed profiles to validate the applicability of the proposed methodology. The obtained results demonstrate an improvement in both computation time required for training and a reduction in the size of the required dataset for transfer learning. Also, the performance is compared with conventional supervised learning on the same data and the same neural network architecture.
{"title":"Transfer Learning for Efficiency Map Prediction","authors":"Arbaaz Khan, M. H. Mohammadi, V. Ghorbanian, D. Lowther","doi":"10.1109/CEFC46938.2020.9451362","DOIUrl":"https://doi.org/10.1109/CEFC46938.2020.9451362","url":null,"abstract":"This paper explores methods to extend a trained deep neural network for predicting efficiency maps to work on different motor drive topologies. This procedure reduces the computation cost associated with training deep networks by transferring knowledge over similar tasks handled by the deep networks. Two types of synchronous AC machines, including a flat-type interior and a surface-mounted permanent magnet motor are tested over their entire torque-speed profiles to validate the applicability of the proposed methodology. The obtained results demonstrate an improvement in both computation time required for training and a reduction in the size of the required dataset for transfer learning. Also, the performance is compared with conventional supervised learning on the same data and the same neural network architecture.","PeriodicalId":439411,"journal":{"name":"2020 IEEE 19th Biennial Conference on Electromagnetic Field Computation (CEFC)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123553490","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 : 2020-11-16DOI: 10.1109/CEFC46938.2020.9451401
Ning Wang, Guoping Zou, Zhi Gong, Huifang Wang, Shiyou Yang
In order to quantitatively analyze the voltage and current overshoots caused by circuit stray parameters in the fast switching-on and -off transients of an inverter, a high-order distributed circuit model of the entire power electronics system is developed, and the numerical method for stray parameter computations is included. Moreover, a novel projection technique, i.e. an enhanced Structure- Preserving Reduced-order Interconnect Macromodeling (SPRIM) methodology by integrating Input-Output structure Preserving Order Reduction (IOPOR) methodology is proposed to reduce the order of the equation sets of the aforementioned full-order circuit model. To consider the influence of the skin effect of the electromagnetic phenomenon as well the stray parameters on the transient performance of an IGBT based inverter in small and extreme small time scales, a coupled 3D finite element-circuit model is developed and an its iterative solution methodology is proposed. The comparisons between the simulated and the tested results have validated the feasibilities and merits of the proposed work.
{"title":"A Coupled 3D FEM-Distribute Circuit Model for Numerical Analysis of Small Time Scale Transients of an IGBT Based Inverter","authors":"Ning Wang, Guoping Zou, Zhi Gong, Huifang Wang, Shiyou Yang","doi":"10.1109/CEFC46938.2020.9451401","DOIUrl":"https://doi.org/10.1109/CEFC46938.2020.9451401","url":null,"abstract":"In order to quantitatively analyze the voltage and current overshoots caused by circuit stray parameters in the fast switching-on and -off transients of an inverter, a high-order distributed circuit model of the entire power electronics system is developed, and the numerical method for stray parameter computations is included. Moreover, a novel projection technique, i.e. an enhanced Structure- Preserving Reduced-order Interconnect Macromodeling (SPRIM) methodology by integrating Input-Output structure Preserving Order Reduction (IOPOR) methodology is proposed to reduce the order of the equation sets of the aforementioned full-order circuit model. To consider the influence of the skin effect of the electromagnetic phenomenon as well the stray parameters on the transient performance of an IGBT based inverter in small and extreme small time scales, a coupled 3D finite element-circuit model is developed and an its iterative solution methodology is proposed. The comparisons between the simulated and the tested results have validated the feasibilities and merits of the proposed work.","PeriodicalId":439411,"journal":{"name":"2020 IEEE 19th Biennial Conference on Electromagnetic Field Computation (CEFC)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114989612","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：