基于云模型和不确定推理的特高压输电塔损伤检测研究

In order to solve damage detection problem of ultra-high voltage transmission tower in Xinjiang electricity transmission, a damage detection method based on cloud model and uncertain reasoning is presented in this project. The aim is to build a kind of monitor system to alarm and detect the damage of ultra-high transmission tower. First, the multiple mode parameters are extracted from the wind-induced vibration response, cloud models of mode parameters are calculated and damage detection reasoning rules are built. Then, the multiple mode parameters are calculated by using different reasoning rules and the calculated results are processed again by using cloud model. Finally, backward cloud generator is utilized to obtain the damage detection results. The multiple mode parameters and multi-condition reasoning rules are used to eliminate the measured noise and error and to enhance the reliability of detection results. The damage detection method based on cloud model and uncertain reasoning is presented in this project. We haven't found the similar methods in published literatures. So, the proposed method has innovative significance. This regional characteristic of Xinjiang electricity transmission is also considered in this project. Therefore, the proposed project also has practical value.

针对"疆电外送"特高压线路中的输电塔损伤检测问题，本项目提出了基于云模型和不确定推理相结合的损伤检测方法。其主要研究目的是：建立一种特高压输电塔结构损伤的整体预警和检测机制。其学术思想为：利用特高压输电塔的风致响应提取结构的多模态特征数据，并建立相应的云模型特征值和损伤检测的推理规则，将多特征参数进行不确定推理运算，不同特征参数利用不同的损伤检测推理规则，获取相应的计算结果，再将计算结果进行云化处理，最后利用逆向云发生器生成最终的损伤检测结果。本项目利用多特征参数和多条件多规则推理相结合，可以有效的消减测量噪声和误差的影响，提高识别结果的可靠性。本项目所采用的基于云模型和不确定推理的损伤检测方法，目前尚无类似的研究报道，故研究方法具有创新性，项目结合了"疆电外送"这一区域特点，因此，本项目也具有地区特性和实用价值。

基于不确定方法研究输电塔等结构的损伤检测问题成为防灾减灾领域逐渐开展的课题，云模型可以较好地处理具有随机性与模糊性的不确定问题，云模型和不确定推理相结合，可以有效地应用于损伤检测领域。本项目系统研究了云模型的数字特征、处理算法以及隶属度与损伤检测技术相结合的方式方法，提出了基于模态应变能和云模型相结合的损伤检测方法以及基于等效隶属度的损伤识别技术。在此基础上，重点研究了基于云模型和不确定理论相结合的损伤检测机理，提出了云模型和不确定理论相结合的云推理方法，并进行了基于云模型和不确定推理的可靠性损伤检测的研究，利用了多源信息的云推理融合来提高损伤识别的可靠性。