基于自回归/自回归条件异方差模型转换距离的输电塔非线性损伤识别及试验研究  

Nonlinear Damage Identification of Transmission Towers Based on AR/ARCH Model Conversion Distance and Its Experimental Research

作　　者：冯衡 高彬 郭惠勇[2] FENG Heng;GAO Bin;GUO Huiyong(Central Southern China Electric Power Design Institute of China Power Engineering Consulting Group Corporation,Wuhan 430071,China;School of Civil Engineering,Chongqing University,Chongqing 400045,China)

机构地区：[1]中国电力工程顾问集团中南电力设计院有限公司,武汉430071 [2]重庆大学土木工程学院,重庆400045

出　　处：《工业建筑》2022年第8期57-63,共7页Industrial Construction

基　　金：中国能源建设股份有限公司重大科技专项(CEEC2020-KJZX-05)。

摘　　要：输电塔等工程钢结构的安全常易受到裂纹等损伤的影响,该类损伤在结构振动中常表现出变刚度的非线性特征。为了解决该类非线性损伤检测问题,提出了一种基于自回归/自回归条件异方差AR/ARCH模型转换距离的非线性损伤识别方法。首先描述了AR/ARCH混合模型的基本组成原理,给出了模型建模的定阶和参数估计方法。然后描述了非线性损伤的时变刚度特性,给出了基于二阶方差指标的非线性损伤识别策略。在此基础上,分析了AR/ARCH模型残差和条件方差与非线性损伤的关联关系,并提出了一种基于AR/ARCH模型的转换距离指标。最后采用三层框架试验验证了指标的有效性,并对输电塔模型进行了损伤识别试验。试验和计算结果表明:基于AR/ARCH模型的转换距离指标明显优于传统的二阶方差指标,其具有更强的非线性损伤识别能力和可靠性。The safety of engineering steel structures such as transmission towers is often affected by cracks and other damages. Such damages often exhibit nonlinear characteristics of variable stiffness in structural vibration. In order to solve this kind of nonlinear damage detection problem, a nonlinear damage identification method based on AR/ARCH model conversion distance was presented. Firstly, the basic theory of AR/ARCH hybrid model was described, and the order determination and parameter estimation methods of modeling were given. Then, the time-varying stiffness characteristics of nonlinear damage were described, and the nonlinear damage identification strategy based on second-order variance index was given. On this basis, the relations between AR/ARCH model residual and condition variance and nonlinear damage was analyzed, and a conversion distance index based on AR/ARCH model was proposed. Finally, the related experimental research was carried out, and the three-layer frame experiment was used to verify the effectiveness of the index, and the damage identification experiment of the transmission tower model was also conducted. The experimental and calculation results showed that the conversion distance index based on AR/ARCH model was obviously superior to the traditional second-order variance index, and it had stronger nonlinear damage identification capability and reliability.

关 键 词：输电塔 损伤识别 自回归/自回归条件异方差模型 非线性 二阶方差 

分 类 号：TU317[建筑科学—结构工程] TM75[电气工程—电力系统及自动化]