Optimal design of composite achromatic wave plate based on the improvement with initial point selection of intelligent algorithm

In order to optimize the design of three-element achromatic quarter-wave plate at 1000–1500 nm, the random initial point selection method of traditional intelligent algorithms is improved with the fixed initial points. The fixed initial point nears the local or global minimum is obtained by analyzing the objective function. The design based on random and fixed initial points is compared. The results show that the method of the fixed initial points is better than the random initial points. Among the four intelligent algorithms for designing achromatic wave plate, particle swarm optimization is best, followed by ant colony optimization, simulated annealing algorithm and genetic algorithm. The average relative phase retardation deviation of achromatic wave plate based on particle swarm optimization is 0.87 %, which meets the design requirements of the device. The selection of the fixed initial points can overcome instability and low accuracy of the traditional intelligent algorithms to a great extent. This method is of great reference value for designing composite achromatic wave plate with different phase retardation.