Knowledge Based Permanent Magnet Machine Design – Literature Survey
Ari Zachas, Yao Duan, Ronald G. Harley, and Thomas G. Habetler at Georgia Institute of Technology, Atlanta, GA
This report is a summary of a literature evaluation completed from 1 July, 2006, to 30 June, 2007, under a grant from the Grainger Center for Electrical Machinery and Electromechanics.
The use of permanent magnet (PM) machines continues to grow, and as we strive for machines with higher efficiencies and larger power densities, designs must be optimized. A beginning machine designer requires an enhanced set of tools that would allow efficient use of experience and insight gained by previous machine designers. These tools should allow the designer to determine a set of machine performance requirements (weight, rated speed, output power, etc.), set suitable ranges for certain key parameters (maximum tooth flux density, realizable airgap length, range of lamination thickness, etc.) and then, together with prior experience and insight, provide appropriate machine parameters. Without such tools engineers and designers would find themselves “recreating the wheel” and in many cases creating designs that could have been completed more efficiently and with better performance. Seasoned designers often rely on their experience to efficiently design electrical machines. As these professionals retire, there is not a sufficiently large pool of young engineers with the necessary knowledge to jump in and produce similar results. The long-term goal of this research is to develop a knowledge-based PM machine design tool that uses the current best practice and past design knowledge in order to improve the effectiveness of new designers in this field and, where possible, to optimize performance. This goal can be reached in three distinct, yet dependent phases: (1) creation of a knowledge base of information from a detailed literature evaluation; (2) establishment of links between the various areas of the machine design; and (3) development of optimization tools that will utilize the established links to develop an optimized machine design that meets the application requirements.
As a first step towards this long-term goal, a literature evaluation of 210 published papers in journals and conference proceedings has been carried out. The focus has been on brushless PM machines and several subtopics have been identified, although many of the principles can be equally applied to interior PM machines. A full report of this work includes chapters on analytical modeling, parameter calculation, pulsating torque, iron losses, materials, performance, machine design, application-specific design, and optimization techniques. Although these topics are by no means exclusive, they form a foundation point for further research.
Figure 4 shows a PM machine configuration.