In order to solve the problems of Flower Pollination Algorithm (FPA) in the optimization process easy to fall into local optimal solution and slow convergence speed, an improved FPA algorithm combined with catfish effect and uniform mutation operator is proposed. First of all catfish effect is introduced to avoid algorithm trapped in local optimal solution, then uniform mutation operator is introduced to improve convergence speed in later stage, the classical test functions is used to verify the improved FPA algorithm performance is better than other algorithms. Finally, the improved FPA algorithm is applied to the optimal design of circuit breakers, and is used to optimize and simulate the energy consumption models of two different types of universal circuit breakers, and the corresponding circuit breaker parameters are calculated. The results show that the improved FPA algorithm used to design the circuit breakers parameters is more reasonable and has lower energy consumption values, it has effectively improved the design accuracy and design efficiency of energy consumption parameters.