论文部分内容阅读
以数据融合技术进行桁架结构的单损伤和多损伤识别。通过研究基于频率的结构损伤理论,分析归一化的频率和损伤位置的关系;利用小波概率神经网络的算法对决策融合进行修正,建立基于小波概率神经网络的数据融合结构损伤识别模型。运用结构计算软件计算了一典型桁架结构的频率,并融合为小波概率神经网络算法的输入特征向量,并对桁架算例模型结构进行损伤识别。通过桁架不同位置的损伤情况,验证该方法的有效性,并提出工程应用中应注意的问题。研究结果表明,基于小波概率神经网络算法的数据融合技术是一种比较可靠的损伤识别方法,具有良好的工程应用前景。
Data fusion technique for single and multiple damage identification of truss structure. By studying the theory of structural damage based on frequency, the relationship between normalized frequency and damage location is analyzed. The algorithm of wavelet probability neural network is used to correct decision fusion, and a data fusion structure damage identification model based on wavelet probability neural network is established. The frequency of a typical truss structure was calculated by using structural calculation software and the input eigenvectors of the algorithm were fused into the wavelet probabilistic neural network algorithm. The truss model structure was identified for damage identification. Through the damage of truss at different positions, the effectiveness of this method is verified and the problems that should be paid attention to in engineering application are put forward. The results show that the data fusion technology based on wavelet probability neural network algorithm is a reliable method of damage identification and has a good prospect of engineering application.