基于人工神经网络的居民生活垃圾可燃成分热值预测

Prediction of household waste combustible component calorific value based on artificial neural network

研究采用BP、RBF和自适应神经模糊推理系统(ANFIS)对生活垃圾可燃成分的热值进行预测。结果表明,BP神经网络模型的预测准确率为93.36%,RBF模型为96.87%,ANFIS模型为91.06%,3种模型均可用于可燃成分热值预测,但RBF模型的预测准确率相对较高,更适用于可燃垃圾的热值预测。

This study established combustible-component calorific-value-prediction models based on a back propagation（ BP） neural network,radical basis function neural network（ RBF）,and adaptive neural fuzzy inference system（ ANFIS） for resident waste. The results showed that the prediction accuracy of the BP neural network model,the RBF model,and the ANFIS model were 93. 36%,96. 87%,and 91. 06%,respectively.Each model can be used to predict calorific value,but the RBF model has the highest prediction accuracy,and is thus most suitable for estimating the calorific value of combustible waste.