基于樽海鞘算法优化支持向量机的连续梁桥损伤识别

Damage Identification of Continuous girder Bridges Based on Support Vector Machine Optimized by the Cunningham Sheath Algorithm

为了能够更加准确、高效地判断桥梁结构损伤位置和程度,本文提出了基于樽海鞘群优化支持向量机(SSA-SVM)方法进行连续梁桥损伤识别的方法。该方法以敏感性较高的曲率模态差作为损伤识别指标,利用樽海鞘群(SSA)算法寻找支持向量机(SVM)最优参数,建立SVM预测模型,通过建立一座三跨连续梁桥有限元模型,以桥梁易损区域作为损伤识别对象进行数值模拟。结果表明:以曲率模态差作为损伤识别指标,能够有效地识别并定位单点或多点的损伤状况,同时准确评估损伤的严重程度。与传统SVM模型比较,SSA-SVM模型实现了参数的自动优化,同时也拥有了更为精准的预测能力。

In order to be able to determine the location and degree of bridge structural damage more accurately and efficiently,this paper proposes a new method for damage identification of continuous girder bridges based on salp swarm algorithm optimization support vector machine method.This study proposes a novel approach for damage identification,utilizing the curvature mode difference as a highly sensitive index.The salp swarm algorithm(SSA)is employed to optimize the parameters of the Support Vector Machine(SVM)and establish the SVM prediction model.Numerical simulations are conducted using a finite element model of a three-span continuous girder bridge,with the vulnerable area of the bridge as the target for damage identification.The study demonstrates that employing curvature modal difference as a damage identification index effectively localizes and assesses the damage degree of bridge unit placement and multi-location damage.Additionally,the SSA-SVM model,with automatic parameter optimization,exhibits superior prediction accuracy compared to the conventional SVM model.