Optimal Design of Electric Machine with Efficient Handling of Constraints and Surrogate Assistance

Published: 22 April 2022| Version 1 | DOI: 10.17632/j57d4jrf2j.1
Bhuvan Khoshoo


This dataset includes data for reproducing results presented in the mentioned paper. There are results obtained from 3 different optimization methods, NSGA-II, NSGA-II-WR, and NSGA-II-WR-SA (details of methods included in the paper). For NSGA-II and NSGA-II-WR, there are 5 different excel files which are the output data of five optimization runs with both these methods. For NSGA-II-WR-SA, there are 12 excel files and each excel file corresponds to a particular parameter configuration (explained in the paper) and holds the data for five optimization runs completed for that configuration. In each dataset, the first two columns represent objective functions F1 (average torque) and F2 (torque pulsations). The next ten columns represent corresponding optimization variables (x1 – x10). It should be noted that values for F1 (average torque) are multiplied by -1 since the optimization problem was formulated as a minimization problem.



Michigan State University


Multi-Objective Optimization, Electric Machine