In this thesis, robust design optimization method was performed to minimize the cogging torque which cause vibration and noise of permanent magnet inserted in the rotor of interior permanent magnet type motor. This paper proposes a statistical probability approach to evaluate the quality loss function of interior permanent magnet type motor design problems, which is expressed in terms of the first two statistical moments, mean and variance.
Conventional optimal design methods does not consider uncertainties in design variables. Therefore, it is very important to consider uncertainty in the stage of design optimization. For this reason, robust design optimization method has been introduced to consider the uncertainty of design variables. Robust design optimization method(RDO) is designed to be insensitive for minimize of variation from uncertainty in product performance. Proposed method has compared the robustness by using Taguchi''s method and worst case scenario method.
To improve the performance quality, Taguchi''s method was applied the search method for the optimal level of the iterative design and solved the optimization problem with constraints applying the penalty function method for solving the inequality constraint. relative amount of output variations at worst case scenario is minimized under the assumption of a small change of system input. In other words, they cannot give any probabilistic information on how much the impact of the uncertainty concerned with system input is on system output at a nominal design. A univariate dimension reduction method is employed to calculate the statistical moments of a performance function and their sensitivities accurately and efficiently.
The proposed method is tested with a mathematical model and a BLDC motor, and its numerical accuracy and efficiency is examined by comparison with existing methods. The results are compared to the MCS of high accuracy and shown to be very effective and accurate.