本文選題：高鐵建設(shè)　切入點(diǎn)：LM-BP人工神經(jīng)網(wǎng)絡(luò)　出處：《石家莊鐵道大學(xué)》2014年碩士論文　論文類型：學(xué)位論文

【摘要】：近年來，我國的高速鐵路建設(shè)迅猛發(fā)展，運(yùn)營里程成倍增長，高鐵的建設(shè)和開通運(yùn)營有力的帶動了沿線區(qū)域經(jīng)濟(jì)的發(fā)展，成為了區(qū)域發(fā)展的新動力，同時高鐵的發(fā)展也給沿線的環(huán)境保持帶來了巨大的壓力。高鐵項目環(huán)境影響評價作為線路選擇、控制環(huán)境負(fù)面影響的關(guān)鍵性決策因素日益成為研究的熱點(diǎn)。隨著研究的深入，高鐵環(huán)評在研究內(nèi)容和程序上都有了進(jìn)一步的完善與規(guī)范，然而對于使用復(fù)雜系統(tǒng)科學(xué)方法進(jìn)行評價研究的卻很少，而這種方法十分適合解決非線性的環(huán)境系統(tǒng)問題，其得出的評價結(jié)果在準(zhǔn)確性、運(yùn)算效率方面都非常有優(yōu)勢，高鐵環(huán)評就屬于這樣的問題。因此，研究高鐵建設(shè)環(huán)境復(fù)雜系統(tǒng)科學(xué)評價方法具有十分重要的意義。 本課題主要研究如何進(jìn)行高鐵建設(shè)環(huán)境影響綜合評價指標(biāo)體系的建立和指標(biāo)數(shù)量化計算以及改進(jìn)過的BP人工神經(jīng)網(wǎng)絡(luò)在高鐵建設(shè)環(huán)境影響綜合評價中的應(yīng)用等問題。文章闡述了本課題的國內(nèi)外的研究進(jìn)展，以及研究目的及方法。分析了高鐵建設(shè)對環(huán)境的影響，依據(jù)相關(guān)評價標(biāo)準(zhǔn)建立了高鐵建設(shè)環(huán)境影響綜合評價的指標(biāo)評價體系。介紹了人工神經(jīng)網(wǎng)絡(luò)的相關(guān)發(fā)展，重點(diǎn)闡述了BP人工神經(jīng)網(wǎng)絡(luò)的相關(guān)原理，探討了將人工神經(jīng)網(wǎng)絡(luò)應(yīng)用到高鐵建設(shè)環(huán)境影響綜合評價中的方法。 本研究根據(jù)對既往已經(jīng)開通運(yùn)營的鐵路項目對環(huán)沿線境影響的深入調(diào)查和研究，建立了高鐵建設(shè)環(huán)境影響綜合評價的指標(biāo)體系，確定了每個指標(biāo)的量化考核因素，并對指標(biāo)值的量化計算進(jìn)行了說明。根據(jù)改進(jìn)過的BP人工神經(jīng)網(wǎng)絡(luò)建立了高鐵建設(shè)環(huán)境影響的三層前饋神經(jīng)網(wǎng)絡(luò)綜合評價模型，改進(jìn)過的BP神經(jīng)網(wǎng)絡(luò)算法是在傳統(tǒng)的算法的基礎(chǔ)上使用LM算法，，即梯度法(最速下降法)和高斯—牛頓迭代法的結(jié)合，該種方法提高了收斂速度以及精確度。利用連鹽城際鐵路作為實例驗證，在建立好的指標(biāo)體系的基礎(chǔ)上，深入分析該工程對環(huán)境影響的情況，設(shè)計出了針對連鹽鐵路的LM-BP神經(jīng)網(wǎng)絡(luò)評價模型，在MATLAB軟件中通過BP網(wǎng)絡(luò)工具箱的調(diào)用進(jìn)行了仿真評價，評價結(jié)果真是有效，該模型的建立可提高評價結(jié)果的客觀性和準(zhǔn)確性。
[Abstract]:In recent years, the construction of high-speed railway in our country has developed rapidly and the mileage of operation has increased exponentially. The construction and opening of high-speed railway have powerfully driven the development of regional economy along the route and become the new driving force for regional development. At the same time, the development of high-speed rail also brings great pressure to the environmental maintenance along the railway line. As a route selection, environmental impact assessment (EIA) of high-speed rail project becomes a key decision-making factor to control the negative impact of the environment. With the development of the research, environmental impact assessment has become a hot topic. The research contents and procedures of high-speed rail environmental assessment have been further improved and standardized. However, the use of complex system scientific methods for evaluation research is rare, and this method is very suitable for solving nonlinear environmental system problems. The evaluation results obtained are very advantageous in terms of accuracy and operational efficiency, which is the problem of high-speed rail EIA. Therefore, it is of great significance to study the scientific evaluation method of complex system in the environment of high-speed rail construction. This paper mainly studies how to establish the index system of environmental impact comprehensive assessment of high-speed railway construction and how to calculate the index quantificationally, and how to apply the improved BP artificial neural network in the comprehensive assessment of environmental impact of high-speed railway construction. The article expounds the research progress of this subject at home and abroad, The paper analyzes the environmental impact of high-speed railway construction, establishes an index evaluation system for environmental impact assessment of high-speed rail construction, and introduces the related development of artificial neural network. The related principle of BP artificial neural network is expounded, and the method of applying artificial neural network to the comprehensive assessment of the environmental impact of high-speed railway construction is discussed. Based on the in-depth investigation and study of the impact of the railway projects that have been opened on the environment along the ring, the paper establishes an index system for the comprehensive assessment of the environmental impact of the high-speed rail construction, and determines the quantitative assessment factors for each index. Based on the improved BP artificial neural network, a three-layer feedforward neural network comprehensive evaluation model for the environmental impact of high-speed iron construction is established. The improved BP neural network algorithm is based on the traditional algorithm using LM algorithm, that is, the combination of gradient method (the steepest descent method) and Gao Si Newton iterative method. This method improves the convergence speed and accuracy. By using the inter-city railway with salt as an example, on the basis of establishing a good index system, the paper deeply analyzes the influence of the project on the environment. The evaluation model of LM-BP neural network for Lianyan railway is designed and simulated by calling BP neural network toolbox in MATLAB software. The evaluation result is really effective. The establishment of the model can improve the objectivity and accuracy of the evaluation results.
【學(xué)位授予單位】：石家莊鐵道大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號】：X82

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 陳輝,李雙成,鄭度;基于人工神經(jīng)網(wǎng)絡(luò)的青藏公路鐵路沿線生態(tài)系統(tǒng)風(fēng)險研究[J];北京大學(xué)學(xué)報(自然科學(xué)版);2005年04期

2 閻平凡;人工神經(jīng)網(wǎng)絡(luò)的容量、學(xué)習(xí)與計算復(fù)雜性[J];電子學(xué)報;1995年05期

3 陳慶偉;梁鵬;;建設(shè)項目環(huán)評與“三同時”制度評析[J];環(huán)境保護(hù);2006年23期

4 李京榮,王家驥,婁安如,蘇德畢力格,何萍;淺析鐵路建設(shè)對生態(tài)環(huán)境的影響——以青藏鐵路格望段為例[J];環(huán)境科學(xué)研究;2002年05期

5 趙尊宇;陳文瑛;;模式預(yù)測法應(yīng)用于鐵路噪聲的預(yù)測研究[J];中國安全生產(chǎn)科學(xué)技術(shù);2011年05期

6 孫東東;張小平;;冀中南城市群城市生態(tài)系統(tǒng)安全評價研究[J];科技創(chuàng)業(yè)家;2013年02期

7 賀昌政,李曉峰,俞海;BP人工神經(jīng)網(wǎng)絡(luò)模型的新改進(jìn)及其應(yīng)用[J];數(shù)學(xué)的實踐與認(rèn)識;2002年04期

8 蘇衛(wèi)青;;高速鐵路噪聲影響評價研究[J];鐵道標(biāo)準(zhǔn)設(shè)計;2011年05期

9 焦大化;鐵路噪聲環(huán)境影響評價的方法[J];鐵道勞動安全衛(wèi)生與環(huán)保;2001年03期

10 鐵道部環(huán)境保護(hù)考察團(tuán);國外鐵路環(huán)境保護(hù)工作[J];鐵道勞動安全衛(wèi)生與環(huán)保;2003年01期

本文編號：1557353