In the inversion problem of particle sizing by ultrasonic attenuation spectroscopy,the inversion algorithm plays an extremely important role.As the genetic algorithm is a global optimization algorithm,it's highly pivotal to optimize the parameters in genetic algorithm for the inversion effects of particle size distribution.After optimizing the parameters of maximum generation number,population size and genetic operators (crossover fraction and mutation fraction),a series of particle systems with three typical distribution functions were numerically calculated.The numerical simulation reveals that a relatively anastomotic inversion result can be obtained using ultrasonic spectroscopy,as compared to the input particle size distribution,when the maximum generation number is 300,the value of population size is 60 and the ranges for crossover fraction and mutation fraction are 0.4~0.85 and 0.045~0.08 respectively.Finally,the stability and noise immunity of the genetic algorithm based on the optimization of parameters were validated through the numerical and experimental examples of two kinds of micron-sized glass beads-glycerol suspension samples.