The invention discloses a method for achieving shielding mortar performance optimization based on a BP neural network algorithm and a genetic algorithm. The method comprises the following steps that multiple influencing parameters influencing mortar performance in the mortar mixing process and performance observation parameters capable of reflecting the mortar performance are selected; multiple groups of mortar samples are prepared according to different influencing parameters, performance observation parameters of the mortar samples are acquired, and normalization processing is performed to obtain sample matrixes X and performance sample matrixes Y respectively; the sample matrixes X are used as inputs, the performance sample matrixes Y are used as outputs, the BP neural network algorithm is applied to train the samples, a mixing process evolution model is established; the genetic algorithm is utilized to optimize the model, and an optimal range of the influencing parameters is found.