Abstract: At present, the difficulty in formulating the operation plan of booster station is how to adjust the inlet pressure accurately and quickly according to the change of downstream gas consumption without exceeding the maximum compression power and how to prepare the multi-unit combined operation plan in advance according to the specific requirements. In this paper, 6RDSA-1 compressor unit in Tayu booster station of Daniudi Gasfield was taken as the study object to establish the compressor unit operation optimization model by using the BP neural network algorithm. Three existing basic compressor parameters (inlet temperature, exhaust pressure and exhaust rate) were selected as the input values of the model, and the appropriate inlet pressure and the shaft power of the unit were calculated. Based on the comparison of multiple data in different conditions, the relative error between the predicted inlet pressure and the actual value is less than 2.75%. It is verified that the compressor unit operation optimization model based on BP neutral network algorithm is reliable, and it is conducive to the preparation of multi-unit online operation plan in advance, the improvement of compressor unit operation efficiency and the reduction of energy consumption and operation & maintenance cost.