研究采用BP神经网络和模糊神经网络(FNN)模型对逐步提高有机负荷的半连续式餐厨垃圾和猪粪混合厌氧消化试验进行日产气量预测。结果表明,BP神经网络模型的预测准确率为77.63%,FNN模型为82.33%,2种模型均可用于产气预测,但FNN模型在传统神经网络模型基础上加入了模糊控制,可提高其准确率,更适用于混合厌氧消化产气量预测。 In this study, BP neural network and fuzzy neural network (FNN) model were used to predict the daily production of mixed semi-continuous anaerobic digestion of food waste and pig manure with gradually increasing organic load. The results show that the prediction accuracy of the BP neural network model is 77.63% and that of the FNN model is 82.33%. Both models can be used for gas production prediction. However, the FNN model adds fuzzy control to the traditional neural network model to improve its accuracy Rate, more suitable for mixed anaerobic digestion gas production forecast.