粉煤灰
粉煤灰
분매회
COAL ASH CHINA
2012年
6期
1-3
,共3页
氯离子扩散系数%网格法%减法聚类法%粉煤灰混凝土
氯離子擴散繫數%網格法%減法聚類法%粉煤灰混凝土
록리자확산계수%망격법%감법취류법%분매회혼응토
Chloride ion diffusion coefficient%Grid method%Subtractive clustering method%Fly Ash Concrete
采用网格法和减法聚类法划分网络,分别建立了基于 ANFIS (Adaptive Neuron-Fuzzy Inference System)的两种粉煤灰混凝土氯离子扩散系数预测模型,并用NEL法试验中的数据作为样本来训练和测试两种模型,比较其预测精度.结果表明,采用减法聚类法建立的 ANFIS 模型预测精度更高,更能满足工程应用的要求.
採用網格法和減法聚類法劃分網絡,分彆建立瞭基于 ANFIS (Adaptive Neuron-Fuzzy Inference System)的兩種粉煤灰混凝土氯離子擴散繫數預測模型,併用NEL法試驗中的數據作為樣本來訓練和測試兩種模型,比較其預測精度.結果錶明,採用減法聚類法建立的 ANFIS 模型預測精度更高,更能滿足工程應用的要求.
채용망격법화감법취류법화분망락,분별건립료기우 ANFIS (Adaptive Neuron-Fuzzy Inference System)적량충분매회혼응토록리자확산계수예측모형,병용NEL법시험중적수거작위양본래훈련화측시량충모형,비교기예측정도.결과표명,채용감법취류법건립적 ANFIS 모형예측정도경고,경능만족공정응용적요구.
Grid method and subtractive clustering method are used to divide the network of ANFIS(Adaptive Neuron-Fuzzy Inference System).Two kinds of chloride ion diffusion coefficient prediction models for fly ash concrete are respectively established. The data obtained from NEL test method are used as samples to train and test two models, and to compare their prediction accuracy. The results show that ANFIS model gets higher prediction accuracy by using subtractive clustering method and can be better to meet the requirements of engineering applications.
