Abstract: Aiming at the problems of low braking energy feedback efficiency and large braking torque ripple coefficient when using traditional optimization algorithm to optimize regenerative braking control parameters of switched reluctance motor, a dual-objective non-dominated sorting genetic algorithm based on progressive constraint rule is proposed. First, against the traditional dual target non dominated sorting genetic algorithm optimization of switched reluctance motor regenerative braking control parameters easily plunged into local optimal solution and other deficiencies. A gradual control rule is proposed, then based on the gradual constraint dominate the double goal of the law of non dominated sorting genetic algorithm is applied to the switched reluctance motor regenerative braking control parameter optimization. The results are verified by simulation and compared with