2 二段階ニューラルネットワークモデルによる損傷同定手法と剛性低下率に基づく地震被害予測

概要

論文の詳細を見る
Recent interest in inspecting the condition of a structure for quality assurance has been strong. Until now, however, inspection of the condition of a structure has been done manually, and inspections are limited due to labor costs. Therefore, monitoring systems that require only the information from sensors to estimate the structural integrity are desired. Outline of the structural integrity assessment system was proposed, and then we also indicated the damage identification by using improved MDLAC (Multiple Damage Location Assurance Criterion) method, neural networks, and so on. In this paper, we describe the damage identification by using improved neural networks, and then the forecast of seismic damage based on stiffness degrading. As a result, it is pointed out that there are some possibilities to confirm the structural integrity assessment system.
一般社団法人日本建築学会の論文
2005-03-20