首页> 外国专利> SYSTEM FOR DETECTING ABNORMAL BEHAVIOR AND EVALUATING SAFETY OF STRUCTURE FOR MERGING NON-PERIODIC SMALL-SCALE DATA, AND METHOD THEREOF

SYSTEM FOR DETECTING ABNORMAL BEHAVIOR AND EVALUATING SAFETY OF STRUCTURE FOR MERGING NON-PERIODIC SMALL-SCALE DATA, AND METHOD THEREOF

机译:非周期性小规模数据融合的异常行为检测和结构安全性评估系统及其方法

摘要

Provided are a system for detecting an abnormal behavior and evaluating the safety of a structure merging non-periodic small-scale data, and a method thereof. The system according to the present invention can minimize generable noise by merging the data to give continuity to the non-periodic data, can simply perform safety evaluation and abnormal behavior detection of the structure by generating a data set matched with the minimum condition using the non-periodic small-scale data, and can implement hardware with lost costs by merging the non-periodic small-scale data, thereby reducing operating costs for the safety evaluation and abnormal behavior detection of the structure. The system according to the present invention includes a non-periodic small-scale data collection unit which cumulatively collects the non-periodic small-scale data of the structure, a non-periodic small-scale data merging unit which inputs the non-periodic small-scale data by merging and aligning the non-periodic small-scale data, an input data feature extracting unit which extracts the feature of the input data, a unit for detecting the abnormal behavior of the structure and damage to the structure, a damage index comparing unit which checks a sensor node in which the abnormal behavior is generated, a damage index pattern analyzing unit which analyzes a damage index pattern, and a damage area estimating unit which estimates the damage area of the structure.;COPYRIGHT KIPO 2017
机译:提供一种用于检测异常行为并评估合并非周期性小规模数据的结构的安全性的系统及其方法。根据本发明的系统可以通过合并数据以使非周期性数据连续性来最小化可产生的噪声,可以通过使用非噪声数据生成与最小条件匹配的数据集来简单地执行结构的安全性评估和异常行为检测。周期性小规模数据,并且可以通过合并非周期性小规模数据来实现具有成本损失的硬件,从而降低用于结构的安全性评估和异常行为检测的运营成本。根据本发明的系统包括:非周期性小规模数据收集单元,其累积地收集结构的非周期性小规模数据;非周期性小规模数据合并单元,其输入非周期性小规模数据。通过合并和对齐非周期性小比例数据得到的比例数据,提取输入数据特征的输入数据特征提取单元,用于检测结构异常行为和结构损坏的单元,损坏指数比较单元,用于检查产生异常行为的传感器节点;损坏指数模式分析单元,用于分析损坏指数模式;以及损坏区域估计单元,用于估计结构的损坏区域。; COPYRIGHT KIPO 2017

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号