Edinburgh Napier University

  This work presents a rigorous approach to simplify the design optimization process for Switched Reluctance Machines. First of all, the dimension of the Design Space is found to be equal to twelve, as the number of Independent Design Variables. Then, constraints and requirements in the design are represented as inequalities to determine the limit surfaces, which are nothing else than the boundaries of the Design Space. By applying constraints and requirements one by one, the paper shows how it is possible to find the Actual Design Space, which includes only feasible solutions to the design problem. A case study clearly shows how, by means of this methodology, the size of the Design Space can be decreased by orders of magnitude.