信阳师范学院学报(自然科学版)
信暘師範學院學報(自然科學版)
신양사범학원학보(자연과학판)
JOURNAL OF XINYANG NORMAL UNIVERSITY(NATURAL SCIENCE EDITION)
2013年
3期
428-431
,共4页
刘道华%张礼涛%曾召霞%孙文萧
劉道華%張禮濤%曾召霞%孫文蕭
류도화%장례도%증소하%손문소
正交最小二乘法%高斯函数%径向基函数神经网络%网络模型
正交最小二乘法%高斯函數%徑嚮基函數神經網絡%網絡模型
정교최소이승법%고사함수%경향기함수신경망락%망락모형
orthogonal least squares%Gaussian function%radial basis function ( RBF) neural network%network model
为提高神经网络模型的预测精度以及提高模型的计算效率,减少获得高精度模型的计算量,构建了基于正交最小二乘法的高斯径向基神经网络模型结构,给出了最小二乘法高斯径向基神经网络的递归模型。依据样本点序列信息,给出了高斯径向基函数中心参数的确定方法,并采用正交最小二乘法回归迭代,从而获得隐层同输出层间的连接权参数值。采用混沌 Lorenz 时间序列预测问题对该设计的网络模型进行验证,并同其他文献对该序列预测的精度以及迭代所需的时间作对比。结果表明,采用该设计方法获得的网络模型具有时间预测精度高及计算效率高等优点。
為提高神經網絡模型的預測精度以及提高模型的計算效率,減少穫得高精度模型的計算量,構建瞭基于正交最小二乘法的高斯徑嚮基神經網絡模型結構,給齣瞭最小二乘法高斯徑嚮基神經網絡的遞歸模型。依據樣本點序列信息,給齣瞭高斯徑嚮基函數中心參數的確定方法,併採用正交最小二乘法迴歸迭代,從而穫得隱層同輸齣層間的連接權參數值。採用混沌 Lorenz 時間序列預測問題對該設計的網絡模型進行驗證,併同其他文獻對該序列預測的精度以及迭代所需的時間作對比。結果錶明,採用該設計方法穫得的網絡模型具有時間預測精度高及計算效率高等優點。
위제고신경망락모형적예측정도이급제고모형적계산효솔,감소획득고정도모형적계산량,구건료기우정교최소이승법적고사경향기신경망락모형결구,급출료최소이승법고사경향기신경망락적체귀모형。의거양본점서렬신식,급출료고사경향기함수중심삼수적학정방법,병채용정교최소이승법회귀질대,종이획득은층동수출층간적련접권삼수치。채용혼돈 Lorenz 시간서렬예측문제대해설계적망락모형진행험증,병동기타문헌대해서렬예측적정도이급질대소수적시간작대비。결과표명,채용해설계방법획득적망락모형구유시간예측정도고급계산효솔고등우점。
In order to improve the forecasting accuracy of the neural network model and the computational efficien -cy, the structure of Gaussian radial basis neural network based on orthogonal least squares was constructed and the re -gression models of neural network was given .The center parameters of Gaussian function were determined by the se -quence information of the sample point and the connection weights between the hidden layer and output layer was deter -mined by the recursive computation of the orthogonal least squares .The performances of this method and the other liter -ature method used to forecast the model based on chaotic Lorenz time series were compared in terms of forecasting accu -racy and the recursive time required .The results indicated that the designed model has many advantages such as higher forecasting accuracy and higher computational efficiency .