ABSTRACT:The optimal design of permanent magnet synchronous generator in the single-excitation PMSG system was mainly addressed in this thesis. The optimal machine geometry for minimum power losses and minimum power weight was provided. The studied structure of PMSG is 3-phase, 4-pole, and 12-slot machine with uniformly distributed stator windings. The waveform of output ac voltage is square wave because of the full pitch coil. The parallel coils for each phase was designed because the high current capability of loads. The Nd-Fe-B permanent magnet is selected because of its properties of high both Hc and (B´H)max, causing greatly withstanding the permanent magnet demagnetization even in the worst case of short circuit condition.
The machine parameters are solved by the use of a Matlab function, fmincon , an inequality constrained multi-objective optimization function. fmincon uses medium scale Sequential Quadratic Programming (SQP) , Quasi-Newton , Line-search algorithms. We had briefly discussed these algorithms. An insight on how the cost function (power loss and weight of the machine) and the various constraints are derived from the machine parameters was also provided.
We also had presented the fundamentals and mathematical formulations of developed modeling approach for permanent magnet machines (chapter–3). The simplicity and analytic properties of the magnetic circuit analysis make it most commonly used magnetic field approximation method. Also in this method geometry of the machine is related to the field distribution giving substantial design insight which helped in the design of the 42 V/3000W PMSG.
The model was based on developing equivalent reluctance model from the dimensions and material properties of the machine. Sectoral model of the machine were developed. A linear reluctance matrix was used to relate the flux and magnetomotive force throughout the different sectors of the machine. The saturation effects were evaluated according to the iron section properties and dimensions and then used to adjust the flux distribution.
We had initially presented the basics of reluctance modeling for some basic shapes which are rectangular where the direct substitution of formulae is possible. But in actual case the machine structure is complex with lots of curvature ie magnet, stator ,rotor etc are all curved . Hence some stress was laid on how to calculate reluctance of each part and then model stator, rotor etc. individually and then how to interface them and arrive at the complex reluctance model for the entire machine. A case study of the machine when projected on to 2-Dimension was studied first and then the permanent magnet motor was modeled. FEA was extensively used to validate the model developed.
The report has 163 pages.
If your company is a member of the Mechatronic Laboratory, please send the request to receive a copy of any technical report. If you are not a member please send a request to Ali Keyhani, Department of Electrical Engineering, Mechatronics Program at the following address: Ali Keyhani, Ohio State University, Electrical Engineering Department, Mechatronics Systems Laboratory, 2015 Neil Ave., 205 Dereese Lab., Columbus, OH 43210.