The invention relates to a power generation control method, belongs to the field of power control, and particularly relates to a new energy output predication error based electric and gas combined system power output control method. The method uses a plurality of anemometer towers in a wind power plant to measure the real time data of the wind velocity and the active wind power output in real timeto fit a wind power characteristic curve, the MLE technology is used to determine the consistency between the distribution of the predication error of the wind power and the candidate distribution, agame model is established to reflect the internal cooperation game relation of a power generation side of an electric and gas combined system, and a differential evolution algorithm is used to resolve an optimal game strategy taking the wind power predication error indeterminacy into consideration. Therefore, the new energy output predication error based electric and gas combined system power output control method has the following advantages: 1, the distribution characteristic of the predication error of the new energy power output is simulated, and the optimal game strategy for the power generation side taking the predication error indeterminacy into consideration is determined through the game theory method; 2, the new energy output predication error based electric and gas combined system power output control method can be applied to making a wind power predication output scheme conforming to the characteristic of the wind power plant, and making the reasonable game strategy for the power generation side.