本文選題：粒子群-支持向量機 + 支持向量機��； 參考：《震災(zāi)防御技術(shù)》2017年01期

【摘要】：本文根據(jù)城市橋梁群體的實際震害資料數(shù)據(jù),采用粒子群算法(PSO)來優(yōu)化支持向量機(SVM)參數(shù),選擇影響橋梁震害等級的8個因素作為特征輸入向量,充分用2種算法的優(yōu)點建立PSO-SVM的橋梁震害預(yù)測模型。通過比較PSO-SVM和SVM模型對橋梁震害的預(yù)測能力,發(fā)現(xiàn)PSO-SVM模型具有較高預(yù)測精度和較高的推廣價值。本文的研究成果對橋梁震害等級的預(yù)測具有一定的參考價值和指導(dǎo)意義。
[Abstract]:Based on the actual earthquake disaster data of urban bridge population, particle swarm optimization (PSO) algorithm is used to optimize the parameters of support vector machine (SVM), and eight factors affecting the earthquake damage grade of bridge are selected as the characteristic input vector. The bridge damage prediction model of PSO-SVM is established by fully using the advantages of the two algorithms. By comparing the prediction ability of PSO-SVM model and SVM model to bridge earthquake damage, it is found that PSO-SVM model has higher prediction accuracy and higher popularizing value. The research results of this paper have certain reference value and guiding significance for the prediction of bridge earthquake damage grade.
【作者單位】： 中國海洋大學(xué)工程學(xué)院土木工程系;
【分類號】：P315.9;U442.55

【相似文獻】

相關(guān)期刊論文 前1條

1 解麗;盧永坤;非明倫;;汶川8.0級地震橋梁震害[J];地震研究;2008年S1期

相關(guān)會議論文 前1條

1 Q笮，

本文編號：2063745