ABSTRACT

In this work, a novel approach in post-earthquake structural damage estimation is investigated. The approach is formulated as a problem of both damage approximation and localization. The inter-story drift ratio and the global damage index of Park/Ang (DI_G,PA) are the estimated damage indicators for each floor of the structure. Artificial neural networks (ANNs), random forests (RFs), support vector machines (SVMs) with linear and radial basis function (RBF) kernels and adaptive neuro-fuzzy inference systems (ANFISs) are tested to predict the seismic damage state of each floor of an 8-storey reinforced concrete (r/c) building subjected to 155 natural and artificially generated seismic accelerograms. The damage potential of the accelerograms is described by three seismic parameters extracted from the response of the structure. The set of seismic accelerograms is defined by combining two outlier detection techniques, isolation forests and Z-score, while the set of seismic parameters is confirmed by minimum redundancy maximum relevance (mRMR) feature selection algorithm. Optimization methods are used to fine-tune the performance of all networks. Results indicate RFs and ANNs among the models with optimal performances, reaching average correct classification rates of up to 96.87% and 91.87% with RFs, and 96.25% and 90.12% with ANNs, for DI_G,PA and ISDR, respectively.

摘要

在这项工作中，研究了一种震后结构损坏估计的新方法。该方法被表述为一个损伤近似和定位的问题。Park/Ang (DI _G,PA) 是结构每个楼层的估计损坏指标。测试人工神经网络 (ANN)、随机森林 (RF)、具有线性和径向基函数 (RBF) 内核的支持向量机 (SVM) 和自适应神经模糊推理系统 (ANFIS)，以预测每层楼的地震损坏状态8 层钢筋混凝土 (r/c) 建筑受到 155 个自然和人工生成的地震加速度图的影响。加速度图的潜在损坏由从结构响应中提取的三个地震参数描述。地震加速度图集是通过结合两种异常值检测技术隔离森林和 Z 分数来定义的，而地震参数集是通过最小冗余最大相关性 (mRMR) 特征选择算法来确认的。优化方法用于微调所有网络的性能。结果表明，在具有最佳性能的模型中，RFs 和 ANNs 的平均正确分类率高达 96.87% 和 91.87%，使用 RFs，使用 ANNs 达到 96.25% 和 90.12%，对于_G，PA和 ISDR，分别。