Abstract: This research was based on a large sludge anaerobic digestion project in Beijing, using a large number of engineering data. The multiple linear regression model, the neural network model, the classification and regression model and k nearest neighbor model to was adopted fit the system biogas production to simulate the biogas production of sluge anaerobic digestion system in practical engineering. Results show that the kNN model has the best fitting effect. For further kNN model analysis, cross validation error statistics selection method was used to determin the best k value. From the test results, it can be seen that with the increase of k value, the fitting degree of the training set first decreases and then tends to be stable, and the fitting degree of the test set was the opposite. Finally, when the k value was 5, the correlation between the model predictive value and the actual value was 0.862, which is better than the fitting effect of the system’s default parameters. The research shows that the data mining technology can be applied to the simulation of sludge anaerobic digestion very well, and has certain guiding significance for the application of mathematical simulation in the field of wastewater treatment.