【摘要】：本文在系統(tǒng)查閱、歸納和分析國內(nèi)外文獻(xiàn)資料基礎(chǔ)上,結(jié)合現(xiàn)場監(jiān)測結(jié)果,對公路邊坡穩(wěn)定性影響因素進(jìn)行分析,選取影響邊坡穩(wěn)定的主要因素作為評價指標(biāo)。結(jié)合模糊聚類理論、模糊模式識別和模糊優(yōu)選理論,建立公路邊坡穩(wěn)定性模糊相似聚類模型。在此基礎(chǔ)上建立公路邊坡穩(wěn)定性模糊相似聚類RBF神經(jīng)網(wǎng)絡(luò)模型。通過研究,得出了如下主要研究成果：1.通過查閱文獻(xiàn)資料和現(xiàn)場調(diào)研,結(jié)合湖南公路邊坡滑坡災(zāi)害現(xiàn)狀、地形地貌特征以及工程地質(zhì)環(huán)境,對影響公路邊坡穩(wěn)定性的主要因素進(jìn)行分析、歸納和分類,將巖土體容重、粘聚力、內(nèi)摩擦角、坡高、坡角、孔隙水壓力比作為邊坡穩(wěn)定性分析的模糊評價指標(biāo)。2.采用二元對比排序法計算影響公路邊坡穩(wěn)定性評價的六個主要指標(biāo)所占權(quán)重,根據(jù)加權(quán)模糊聚類算法,建立了公路邊坡穩(wěn)定性模糊聚類預(yù)測模型,對現(xiàn)有模糊聚類迭代模型進(jìn)行了相應(yīng)的改進(jìn)。3.結(jié)合模糊聚類理論、模糊模式識別以及模糊優(yōu)選理論,建立了公路邊坡穩(wěn)定性模糊相似聚類模型,進(jìn)一步優(yōu)化模糊聚類算法,提高了計算效率和評價的準(zhǔn)確度。針對不同的相似聚類水平β進(jìn)行對比研究,得出了最佳相似聚類水平夕為0.8。4.將模糊相似聚類模型引入到RBF神經(jīng)網(wǎng)絡(luò)中,建立模糊相似聚類神經(jīng)網(wǎng)絡(luò)模型。結(jié)合工程實(shí)例分析,該模型能可靠地應(yīng)用于公路邊坡穩(wěn)定性評價與預(yù)測,進(jìn)一步優(yōu)化了邊坡穩(wěn)定性的評價方法。
[Abstract]:On the basis of systematic reference, induction and analysis of domestic and foreign literature and data, combined with site monitoring results, this paper analyzes the influencing factors of highway slope stability, and selects the main factors affecting slope stability as the evaluation index. Combined with fuzzy clustering theory, fuzzy pattern recognition and fuzzy optimal selection theory, fuzzy similar clustering model of highway slope stability is established. On this basis, the fuzzy similar clustering RBF neural network model of highway slope stability is established. Through research, the following main research results: 1. 1. By referring to the literature and field investigation, combining with the present situation of landslide disaster of highway slope in Hunan Province, the characteristics of landform and geomorphology and the engineering geological environment, the main factors affecting the stability of highway slope are analyzed, summarized and classified, and the bulk density of rock and soil is classified. Cohesion, internal friction angle, slope height, slope angle and pore-water pressure ratio are the fuzzy evaluation indexes of slope stability analysis. The weight of six main indexes affecting the evaluation of highway slope stability is calculated by using the binary contrast ranking method. According to the weighted fuzzy clustering algorithm, the fuzzy cluster prediction model of highway slope stability is established. The existing fuzzy clustering iterative model is improved. 3. Combined with fuzzy clustering theory, fuzzy pattern recognition and fuzzy optimal selection theory, the fuzzy similar clustering model of highway slope stability is established, and the fuzzy clustering algorithm is further optimized to improve the calculation efficiency and evaluation accuracy. According to the comparative study of different similarity clustering levels 尾, the best similarity clustering level is 0.8.4. The fuzzy similar clustering model is introduced into RBF neural network and the fuzzy similar clustering neural network model is established. The model can be reliably applied to the evaluation and prediction of highway slope stability, and the evaluation method of slope stability is further optimized.
【學(xué)位授予單位】：長沙理工大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號】：U416.14

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 羅國煜,劉松玉,楊衛(wèi)東;區(qū)域穩(wěn)定性優(yōu)勢面分析理論與方法[J];巖土工程學(xué)報;1992年06期

，

本文編號：2259723