LI Bin, DENG Yu, BIAN Yuming, et al. Prediction of limestone-gypsum wet flue gas desulfurization efficiency based on T-S fuzzy neural network[J]. Thermal power generation, 2016, 45(6): 116-119.
LI Bin, DENG Yu, BIAN Yuming, et al. Prediction of limestone-gypsum wet flue gas desulfurization efficiency based on T-S fuzzy neural network[J]. Thermal power generation, 2016, 45(6): 116-119.DOI:
基于T-S模糊神经网络的湿法脱硫效率预测
摘要
由于影响脱硫效率的因素较多
且相互之间均具有关联性
造成脱硫效率难以实时监测
且测量结果不精准。本文运用T-S模糊神经网络建立了脱硫效率的预测模型
基于某电厂DCS采集的湿法脱硫系统原始数据
在MATLAB平台上进行训练与检验
得到较精准的脱硫效率预测模型。模型验证结果显示:采用T-S模糊神经网络模型预测脱硫效率
85%的样本点相对误差分布在-1.0%~0.5%之间
最大误差不超过1.5%
说明该模型的预测精度较高
能较好地满足工程实际的需求。
Abstract
There are many relevant factors affecting the desulfurization efficiency
leading to difficulties in measuring the desulfurization efficiency.To solve this problem
the T-S fuzzy neural network was applied to establish prediction model of the desulfurization efficiency
on the basis of the original data of desulfurization system collected by apower plant's DCS in Jiangxi province.The original collected data were tested and trained to be accurate on the platform of MATLAB.The results show that
when predicting the desulfurization efficiency by using the T-S fuzzy neural network prediction model
relative error of 85%samples is between-1.0% and 0.5%
and the maximum error is no higher than 1.5%.It illustrates that the proposed model has higher prediction precision
which can better meet the actual needs of engineering.