基于小波除噪和神经网络理论的气固循环流化床颗粒浓度预测

Prediction of Solids Concentration in Gas-Solid Circulating Fluidized Bed Based on Wavelet and Neural Networks

提出了一种将混沌的相空间重构、小波包分析和神经网络相结合的新方法用于预测气-固循环流化床的颗粒浓度。首先利用小波包进行数据的消噪,然后用混沌方法重构相空间吸引子,用径向基神经网络拟合吸引子上的全局整体映射,构成混合预测模型。实验结果表明,将此混合模型用于预测气-固循环流化床的颗粒浓度,能达到较好的预测效果,预测精度比奇异值分解和傅里叶变换除噪高。

A new hybrid method for prediction of solids holdup in gas-solid circulating fluidized bed is proposed based on chaos phase reconstruction and wavelet package as well as neural networks.This article presents the use of the chaos method to reconstruct attractors in phase spaces and a radial basis networks to fit the attractor's global map after the use of wavelet packet analysis to reduce noise,to construct a new hybrid prediction model.Experimental results show that the model provides good predictions and has promising applications.