五邑大学学报(自然科学版)
五邑大學學報(自然科學版)
오읍대학학보(자연과학판)
JOURNAL OF WUYI UNIVERSITY(NATURAL SCIENCE EDITION)
2014年
3期
55-60
,共6页
应变电测法%径向基函数神经网络%正则化%最小二乘法
應變電測法%徑嚮基函數神經網絡%正則化%最小二乘法
응변전측법%경향기함수신경망락%정칙화%최소이승법
strain measurement methods%radial basis function neural network%regularization%least squares method
针对空调管路的应变测试数据会随环境温度而产生漂移的现象,根据应变电测法,提出了基于径向基函数(RBF)神经网络拟合逼近管路真实应变值的方法,即结合正则化正交最小二乘法,逼近现实中由温度引起的漂移,间接求出空调管路较为真实的交变应变值。实验结果表明该方法是可靠的。
針對空調管路的應變測試數據會隨環境溫度而產生漂移的現象,根據應變電測法,提齣瞭基于徑嚮基函數(RBF)神經網絡擬閤逼近管路真實應變值的方法,即結閤正則化正交最小二乘法,逼近現實中由溫度引起的漂移,間接求齣空調管路較為真實的交變應變值。實驗結果錶明該方法是可靠的。
침대공조관로적응변측시수거회수배경온도이산생표이적현상,근거응변전측법,제출료기우경향기함수(RBF)신경망락의합핍근관로진실응변치적방법,즉결합정칙화정교최소이승법,핍근현실중유온도인기적표이,간접구출공조관로교위진실적교변응변치。실험결과표명해방법시가고적。
In light of test data of air conditioning pipes drifting with environmental temperature, based on the strain measuring method, an approach based on the RBF neural network is proposed to fit the true strain values of pipes. The method adopts the regularization orthogonal least squares method, approaches the reality drift caused by temperature change, and derives indirectly more real alternating strain value. Experimental results show that the method is reliable.