REAL-TIME ELECTROMAGNETIC MODELING AND ANALYSIS TECHNIQUES FOR ELECTRIC MACHINES
Interim project report
Submitted by: Dr. Mahesh Krishnamurthy (ECE) and Dr. Shuwang Li (Applied Math)
An electric machine is an electromechanical transducer in which electrical energy is converted into mechanical energy. Over
the past several decades, development of control of electric machines has typically been viewed using a “black box” approach which
typically involves manipulation of voltage and current to generate the desired characteristics. In this approach, not much attention is
paid to optimizing the intermediate magnetic energy stored inside the machine. A thorough understanding of the local magnetic field
distribution inside the machine plays a significant role in the development of accurate models and control techniques that utilize
electromagnetic signatures.
0.4
0.8
0.6
Normal Flux Density (T)
0.4
0.2
0
-0.2
Figure 1: Magnetic field distribution in an
PMSM using finite element analysis
0.2
0.1
0
-0.1
-0.4
-0.2
-0.6
-0.3
-0.8
0 7 12 18 23
37 42 48 53
67 72 78 83
97102 108113
Position(deg)
127132 138143
157162 168173 180
2.7ms
2.9ms
3.1ms
3.3ms
3.5ms
3.7ms
3.9ms
4.1ms
4.3ms
4.5ms
4.7ms
4.9ms
5.1ms
5.3ms
5.5ms
0.3
Tangential Flux Density (T)
2.7ms
2.9ms
3.1ms
3.3ms
3.5ms
3.7ms
3.9ms
4.1ms
4.3ms
4.5ms
4.7ms
4.9ms
5.1ms
5.3ms
5.5ms
-0.4
0 7 12 18 23
37 42 48 53
67 72 78 83
97102108113
Position(deg)
127132138143
157162 168173 180
Figure 2: Distribution of normal and tangential components of field distribution in
different layers of the airgap
In the first phase of this project, a three-phase, full pitch, distributed winding, radially magnetized surface mount permanent
magnet synchronous motor (PMSM) has been designed. The objective has been to analyze the electromagnetic behavior of the
machine and calculate force distribution in different parts of the machine using Finite Element Analysis (FEA). Figures 1 and 2 show
the FEA model of the PMSM and the field distribution in different parts of the machine. It can be seen that while the force density
enclosed by each layer is the same, there is a difference in the concentration of flux lines in different parts. The two main objectives
that have been targeted in this phase of the project are as follows:
• Profiling force distribution: After obtaining the distribution of magnetic flux density, Maxwell stress method was used to
calculate the distribution of electromagnetic force density and analyze the dynamic process of force generation within the PMSM.
From the calculated results, normal and tangential force distribution, generated torque and radial vibration in the motor can be
obtained, which is valuable in developing high precision control techniques that can strategically optimize specific attributes of the
machine. These results can also be used to explain the operation of the machine under different control strategies and develop more
efficient machine geometries. This concept will be submitted to the 2011 IEEE- International Electric Machines and Drives
Conference to be held in Niagara Falls, Canada in May 2011, which is a forum for presenting innovative concepts in the area of
electric machines and drives.
• Identifying electromagnetic signatures: Force produced in the machine is a function of local field distribution inside the
machine. Therefore any change in the operating condition results in a corresponding change in the field distribution. By applying
appropriate methods to calculate the field in different parts of the machine, this technique can be used to determine faults in the
machine. This concept has been accepted for presentation at the 2010 IEEE- Vehicle Power and Propulsion Conference (VPPC) to be
held in September 2010. The paper has received very high reviews and has been considered eligible for an award at the conference by
the reviewers.