Design aspects of high performance synchronous reluctance machines with and without permanent magnets
(2017)
 Author
 Mohamed Nabil Fathy Ibrahim (UGent)
 Promoter
 Peter Sergeant (UGent) and Essam Rashad
 Organization
 Project
 Abstract
 Recently, a growing interest in the efficiency and the cost of electrical machines has been observed. The efficiency of electric motors is important because electric motors consume about 40%45% of the produced electricity worldwide and about 70% of the industrial electricity1. Therefore, some types of electric motors have been classified in proposed standard classes1 based on their efficiency. By consequence, efficient and low cost electric motors are necessary on the market. Several types of electric motors are used in industrial applications such as permanent magnet synchronous motors (PMSMs), induction motors (IMs) and reluctance motors (RMs). Due to the high cost of PMSMs and due to the rotor losses of the IMs, the RMs can be considered as promising and attractive candidates. Moreover, they have a robust and simple structure, and a low cost as there are no cage, windings and magnets in the rotor. There are two main types of RMs: switched reluctance motors (SRMs) and synchronous reluctance motors (SynRMs). However, there are some disadvantages of these types of machines. On the one hand, the SRMs have problems of torque ripple, vibrations and noise. In addition, their control is more complicated than that of threephase conventional motor drives, a.o. because of the high nonlinearity of the inductance. On the other hand, the SynRMs have a low power factor, so that an inverter with a high VoltAmpère rating is required to produce a given motor output power. Therefore, adding a proper amount of low cost permanent magnet (PM) material  such as ferrite  may be a good option to boost the power factor. The PMs also increase the efficiency and torque density. These types of motors are  1Waide, P. and C. Brunner (2011),”EnergyEfficiency Policy Opportunities for Electric Motor Driven Systems”, IEA Energy Papers, No. 2011/07, OECD Publishing, Paris.xiv Summary called permanent magnetassisted synchronous reluctance motors (PMaSynRMs). In this thesis, both SynRMs and PMaSynRMs are investigated. The main focus is given to the rotor design, magnetic material grade and winding configuration. In addition, the modelling and control of SynRMs and PMaSynRMs is also investigated. First, parametrized models are made of the machines. The finite element method (FEM) is used to obtain the dqaxis fluxlinkages λd(id, iq, θr) and λq(id, iq, θr) of the SynRM in static 2D simulations, as a function of daxis current id, and qaxis current iq and rotor position θr. As known, the performance (output torque, power factor and efficiency) of SynRMs depends mainly on the ratio between the direct (d) and quadrature (q) axis inductances (Ld/Lq). This ratio is wellknown as the saliency ratio of the SynRM. As magnetic saturation causes significant changes in the inductances and by consequence in the saliency ratio during operation, a SynRM model based on constant inductances (Ld and Lq) is not good enough. It can lead to large deviations in the prediction of the torque capability compared with the real motor. How large these deviations are, is clarified in this thesis by comparing several models that do or do not take into account saturation, crosssaturation and rotor position effects. It is found that saturation and cross saturation must be included in the model for an accurate representation of the SynRM performance and control. This means the flux linkages should be function of id and iq. The rotor position needn’t be included. Apart from the currents, the FEM contains many parameters for the flux barrier geometry, which have a strong influence on the torque and torque ripple of the machine. Next to static simulations, also dynamic simulations are done. In these simulations, the fluxlinkages are stored in lookup tables, created a priori by FEM, to speed up the simulations. Based on the SynRM FEM model, the design of the SynRM rotor is investigated. Choosing the fluxbarrier geometry parameters is very complex because there are many parameters that play a role. Therefore, an optimization technique is always necessary to select the fluxbarrier parameters that optimize the SynRM performance indicators (maximize the saliency ratio and output torque and minimize the torque ripple). To gain insight in the relevant parameters, first a sensitivity analysis is done: the influence of the fluxbarrier parameters is studied on the SynRM performance indicators. These indicators are again saliency ratio, output torque and torque ripple. In addition, easytousexv parametrized equations are proposed to select the value of the two most crucial parameters of the rotor i.e. the fluxbarrier angle and width. The proposed equations are compared with three existing literature equations. At the end, an optimal rotor design is obtained based on an optimized technique coupled with FEM. The optimal rotor is checked mechanically for the robustness against mechanical stresses and deformations. Apart from the geometry, the electric steel grade plays a major role in the losses and efficiency of an electric machine. Therefore, several steel grades are compared with respect to the SynRM performance i.e. output torque, power factor, torque ripple, iron losses and efficiency. Four different steel grades NO20, M330P50A, M40050A and M600100A are considered. The steel grades differ in thickness and in the losses they produce. It was found that the “best” grade NO20 had in the rated operating point of the considered SynRM 9.0% point more efficiency than the “worst” grade M600100A. Next to energyefficiency, a large interest in recent research is dedicated to obtain a high torque density. One of the main techniques to improve the machine torque density is to increase the fundamental winding factor through an innovative winding layout. Among several configurations, the socalled combined stardelta winding layout was proposed in literature several years ago. In the PhD, the combined stardelta winding is compared with the conventional star winding in terms of output torque, torque ripple and efficiency. A simple method to calculate the equivalent winding factor of the different winding connections is proposed. In addition, the modelling of a SynRM with combined stardelta winding is given. Furthermore, the effect of different winding layouts on the performance of the SynRM is presented. To compare both windings experimentally, two stators are made, one with combined stardelta windings and one with conventional star windings, having the same copper volume. Measurements revealed a 5.2% higher output torque of the first machine at rated current and speed. In order to even further improve the power factor and the output torque of the SynRM, ferrite PMs are inserted in the center of the rotor fluxbarriers. The rotor geometry of the resulting PMaSynRM is the same as the conventional SynRM. Hence, two rotors with identical iron lamination stack were built: one with PMs and a second one without magnets. Having the two stators and two rotors, a comparison of fourxvi Summary prototype SynRMs is done in the PhD, each of 5.5 kW. Several validation measurements have been obtained. The combinedstar delta SynRM with PMs in the rotor had up to 1.5 % point more efficiency than the SynRM with star winding and rotor without magnets at the rated current and speed. As an application of SynRM, an efficient and low cost photovoltaic (PV) pumping system employing a SynRM is studied. The proposed system does not have a DCDC converter that is often used to maximize the PV output power, nor has it storage (battery). Instead, the system is controlled in such a way that both the PV output power is maximized and the SynRM works at the maximum torque per Ampère, using a conventional three phase pulse width modulated inverter. The design and the modelling of all the system components are given. The performance of the proposed PV pumping system is presented, showing the effectiveness of the system.
 Keywords
 SynRM
Downloads

PhD Ibrahim.pdf
 full text
 
 open access
 
 
 9.26 MB
Citation
Please use this url to cite or link to this publication: http://hdl.handle.net/1854/LU8540777
 MLA
 Ibrahim, Mohamed Nabil Fathy. “Design Aspects of High Performance Synchronous Reluctance Machines with and Without Permanent Magnets.” 2017 : n. pag. Print.
 APA
 Ibrahim, M. N. F. (2017). Design aspects of high performance synchronous reluctance machines with and without permanent magnets.
 Chicago authordate
 Ibrahim, Mohamed Nabil Fathy. 2017. “Design Aspects of High Performance Synchronous Reluctance Machines with and Without Permanent Magnets.”
 Chicago authordate (all authors)
 Ibrahim, Mohamed Nabil Fathy. 2017. “Design Aspects of High Performance Synchronous Reluctance Machines with and Without Permanent Magnets.”
 Vancouver
 1.Ibrahim MNF. Design aspects of high performance synchronous reluctance machines with and without permanent magnets. 2017.
 IEEE
 [1]M. N. F. Ibrahim, “Design aspects of high performance synchronous reluctance machines with and without permanent magnets,” 2017.
@phdthesis{8540777, abstract = {{Recently, a growing interest in the efficiency and the cost of electrical machines has been observed. The efficiency of electric motors is important because electric motors consume about 40%45% of the produced electricity worldwide and about 70% of the industrial electricity1. Therefore, some types of electric motors have been classified in proposed standard classes1 based on their efficiency. By consequence, efficient and low cost electric motors are necessary on the market. Several types of electric motors are used in industrial applications such as permanent magnet synchronous motors (PMSMs), induction motors (IMs) and reluctance motors (RMs). Due to the high cost of PMSMs and due to the rotor losses of the IMs, the RMs can be considered as promising and attractive candidates. Moreover, they have a robust and simple structure, and a low cost as there are no cage, windings and magnets in the rotor. There are two main types of RMs: switched reluctance motors (SRMs) and synchronous reluctance motors (SynRMs). However, there are some disadvantages of these types of machines. On the one hand, the SRMs have problems of torque ripple, vibrations and noise. In addition, their control is more complicated than that of threephase conventional motor drives, a.o. because of the high nonlinearity of the inductance. On the other hand, the SynRMs have a low power factor, so that an inverter with a high VoltAmpère rating is required to produce a given motor output power. Therefore, adding a proper amount of low cost permanent magnet (PM) material  such as ferrite  may be a good option to boost the power factor. The PMs also increase the efficiency and torque density. These types of motors are  1Waide, P. and C. Brunner (2011),”EnergyEfficiency Policy Opportunities for Electric Motor Driven Systems”, IEA Energy Papers, No. 2011/07, OECD Publishing, Paris.xiv Summary called permanent magnetassisted synchronous reluctance motors (PMaSynRMs). In this thesis, both SynRMs and PMaSynRMs are investigated. The main focus is given to the rotor design, magnetic material grade and winding configuration. In addition, the modelling and control of SynRMs and PMaSynRMs is also investigated. First, parametrized models are made of the machines. The finite element method (FEM) is used to obtain the dqaxis fluxlinkages λd(id, iq, θr) and λq(id, iq, θr) of the SynRM in static 2D simulations, as a function of daxis current id, and qaxis current iq and rotor position θr. As known, the performance (output torque, power factor and efficiency) of SynRMs depends mainly on the ratio between the direct (d) and quadrature (q) axis inductances (Ld/Lq). This ratio is wellknown as the saliency ratio of the SynRM. As magnetic saturation causes significant changes in the inductances and by consequence in the saliency ratio during operation, a SynRM model based on constant inductances (Ld and Lq) is not good enough. It can lead to large deviations in the prediction of the torque capability compared with the real motor. How large these deviations are, is clarified in this thesis by comparing several models that do or do not take into account saturation, crosssaturation and rotor position effects. It is found that saturation and cross saturation must be included in the model for an accurate representation of the SynRM performance and control. This means the flux linkages should be function of id and iq. The rotor position needn’t be included. Apart from the currents, the FEM contains many parameters for the flux barrier geometry, which have a strong influence on the torque and torque ripple of the machine. Next to static simulations, also dynamic simulations are done. In these simulations, the fluxlinkages are stored in lookup tables, created a priori by FEM, to speed up the simulations. Based on the SynRM FEM model, the design of the SynRM rotor is investigated. Choosing the fluxbarrier geometry parameters is very complex because there are many parameters that play a role. Therefore, an optimization technique is always necessary to select the fluxbarrier parameters that optimize the SynRM performance indicators (maximize the saliency ratio and output torque and minimize the torque ripple). To gain insight in the relevant parameters, first a sensitivity analysis is done: the influence of the fluxbarrier parameters is studied on the SynRM performance indicators. These indicators are again saliency ratio, output torque and torque ripple. In addition, easytousexv parametrized equations are proposed to select the value of the two most crucial parameters of the rotor i.e. the fluxbarrier angle and width. The proposed equations are compared with three existing literature equations. At the end, an optimal rotor design is obtained based on an optimized technique coupled with FEM. The optimal rotor is checked mechanically for the robustness against mechanical stresses and deformations. Apart from the geometry, the electric steel grade plays a major role in the losses and efficiency of an electric machine. Therefore, several steel grades are compared with respect to the SynRM performance i.e. output torque, power factor, torque ripple, iron losses and efficiency. Four different steel grades NO20, M330P50A, M40050A and M600100A are considered. The steel grades differ in thickness and in the losses they produce. It was found that the “best” grade NO20 had in the rated operating point of the considered SynRM 9.0% point more efficiency than the “worst” grade M600100A. Next to energyefficiency, a large interest in recent research is dedicated to obtain a high torque density. One of the main techniques to improve the machine torque density is to increase the fundamental winding factor through an innovative winding layout. Among several configurations, the socalled combined stardelta winding layout was proposed in literature several years ago. In the PhD, the combined stardelta winding is compared with the conventional star winding in terms of output torque, torque ripple and efficiency. A simple method to calculate the equivalent winding factor of the different winding connections is proposed. In addition, the modelling of a SynRM with combined stardelta winding is given. Furthermore, the effect of different winding layouts on the performance of the SynRM is presented. To compare both windings experimentally, two stators are made, one with combined stardelta windings and one with conventional star windings, having the same copper volume. Measurements revealed a 5.2% higher output torque of the first machine at rated current and speed. In order to even further improve the power factor and the output torque of the SynRM, ferrite PMs are inserted in the center of the rotor fluxbarriers. The rotor geometry of the resulting PMaSynRM is the same as the conventional SynRM. Hence, two rotors with identical iron lamination stack were built: one with PMs and a second one without magnets. Having the two stators and two rotors, a comparison of fourxvi Summary prototype SynRMs is done in the PhD, each of 5.5 kW. Several validation measurements have been obtained. The combinedstar delta SynRM with PMs in the rotor had up to 1.5 % point more efficiency than the SynRM with star winding and rotor without magnets at the rated current and speed. As an application of SynRM, an efficient and low cost photovoltaic (PV) pumping system employing a SynRM is studied. The proposed system does not have a DCDC converter that is often used to maximize the PV output power, nor has it storage (battery). Instead, the system is controlled in such a way that both the PV output power is maximized and the SynRM works at the maximum torque per Ampère, using a conventional three phase pulse width modulated inverter. The design and the modelling of all the system components are given. The performance of the proposed PV pumping system is presented, showing the effectiveness of the system.}}, author = {{Ibrahim, Mohamed Nabil Fathy}}, isbn = {{9789463550703}}, keywords = {{SynRM}}, language = {{eng}}, school = {{Ghent University}}, title = {{Design aspects of high performance synchronous reluctance machines with and without permanent magnets}}, year = {{2017}}, }