The hysteresis modeling of ferromagnetic materials in electrical equipment such as Hermetically Sealed Electromagnetic Relay (HSER) is one of the fundamental theoretical studies in the field of electrical engineering. Long training time and poor accuracy are the main challenges faced by traditional hysteresis models. A Boltzmann Preisach hybrid model is proposed by combining the principle of minimizing magnetic domain energy with the Preisach model. The state equation of magnetic domain total energy is derived from the principle of minimizing magnetic domain energy, and the genetic algorithm is used to obtain the minimum value of magnetic domain total energy, corresponding magnetization vector, and magnetic domain occurrence probability. By minimizing the energy of magnetic domains, the distribution function of the Preisach model can be determined, which can compensate for the difficulty in determining unknown parameters and large computational complexity of the Preisach classical model distribution function. At the same time, it saves computation time and ensures the stability of the results.