工程结构损伤的集成神经网络识别研究

STUDY ON INTEGRATED NEURAL NETWORK FOR DAMAGE IDENTIFICATION OF ENGINEERING STRUCTURES

从结构损伤识别的实际出发,提出采用基于信息融合理论的集成神经网络技术对结构损伤状况进行识别,即通过结构损伤特征信息的有效组合,用各种子神经网络从不同侧面对结构损伤进行初步识别诊断,然后对识别结果进行决策融合。给出了系统的实现策略和子网络的组建原则。从识别实例中可以看出,此识别方法充分利用了各种特征信息,可以有效地提高识别率。

From the point of practice of damage identification of structures,the integrated neural network damage(identification) technology is put forward based on the information fusion theory.Taking the sub-neural networks as primary damage identification measures from different sides,the conclusions are gained through decision-making fusion.The realizable policy of the system identification and the established principle of the sub-neural networks are given in the paper.It can be educed from the examples that it takes full advantage of diversified characteristic informations,and(improves) the diagnosis rate.