In this study, based on a magnetisation-controllable dynamic reluctance network (MC-DRN) model, an optimisation design method with an efficient modern evolution algorithm for the flux memory motor is proposed. To clarify the method conveniently, a doubly salient flux-control memory (DS-FM) motor is involved and taken as a design example. First, according to the motor topology, the detailed MC-DRN model of the DS-FM motor is built, in which the magnetisation state adjustment of the low coercivity force material is considered. During the optimisation design process, to meet the diverse design requirements of multiple operation modes, non-dominated sorting genetic II is employed to obtain the trade-off optimal design of the DS-FM motor under different magnetisation states. Moreover, to verify the feasibility of the optimisation method, the electromagnetic performances of the motor in different operation modes are investigated in detail, and the results are compared with that obtained by the finite-element method. Finally, a prototype is manufactured, and the experiments are carried out. The corresponding results not only validate the reasonability of the investigated DS-FM motor but also prove the accuracy and the effectiveness of the MC-DRN model and optimisation design method.