本文關(guān)鍵詞：隧道下穿建筑群爆破振動(dòng)控制置信度研究　出處：《北京交通大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

  更多相關(guān)文章： 隧道下穿建筑群 振動(dòng)監(jiān)測(cè) 回歸分析 置信預(yù)測(cè) BP神經(jīng)網(wǎng)絡(luò)分類

【摘要】：近年來,隨著國(guó)民經(jīng)濟(jì)持續(xù)增長(zhǎng),城市化進(jìn)程不斷深化,城市交通基礎(chǔ)設(shè)施建設(shè)事業(yè)也在快速發(fā)展,下穿城區(qū)建筑群的隧道工程也越來越多。由于受到工程地質(zhì)、經(jīng)濟(jì)等條件的制約,大多數(shù)的城市隧道工程仍然采用礦山法開挖施工。隧道爆破開挖引發(fā)隧道上部建筑群產(chǎn)生振動(dòng),并對(duì)居民日常生活產(chǎn)生影響。因此,研究下穿建筑群隧道爆破振動(dòng)控制技術(shù)十分重要。本文依托大連南部濱海大道隧道下穿同香山莊工程,對(duì)其爆破振動(dòng)現(xiàn)場(chǎng)監(jiān)測(cè)數(shù)據(jù)進(jìn)行回歸分析與置信研究,主要研究成果如下:(1)對(duì)于本研究所用的實(shí)測(cè)振速數(shù)據(jù),線性回歸得到的擬合直線只反應(yīng)數(shù)據(jù)點(diǎn)分布的平均程度,有50%的數(shù)據(jù)點(diǎn)分布于該擬合直線之上,其余50%的數(shù)據(jù)點(diǎn)分布于該擬合直線之下。在實(shí)際工程中,如果選擇該擬合直線作為振速控制的標(biāo)準(zhǔn),爆破振速必然有很大的可能(近50%)超出振速控制標(biāo)準(zhǔn),從而引發(fā)諸多問題。此時(shí),引入95%置信限十分有必要,利用95%置信限計(jì)算所需的藥量指導(dǎo)施工,可以將95%的爆破振動(dòng)控制在振速控制標(biāo)準(zhǔn)之下,既保證了上部建筑物安全又不過于保守以致影響施工,使下穿建筑群隧道的爆破開挖施工得以順利進(jìn)行。(2)相比常規(guī)分析預(yù)測(cè)模型,基于貝葉斯理論的分析預(yù)測(cè)模型雖然從原理上以及模型實(shí)現(xiàn)上較復(fù)雜,但其對(duì)振速置信限的預(yù)測(cè)精確度很高,尤其在樣本相關(guān)性不高的條件下。(3)樣本數(shù)據(jù)對(duì)回歸分析及置信限預(yù)測(cè)結(jié)果的精確度具有很大影響,一方面,樣本的數(shù)量對(duì)回歸分析及置信限預(yù)測(cè)精確度產(chǎn)生影響,一般來說,樣本數(shù)量越多,對(duì)置信限的預(yù)測(cè)精確度越高。另一方面,樣本數(shù)據(jù)自身的特征,如相關(guān)性、離散程度等也對(duì)置信限預(yù)測(cè)精確度產(chǎn)生影響。(4)在以往的研究中,BP神經(jīng)網(wǎng)絡(luò)方法在非線性回歸分析中展現(xiàn)出較高的精確度,但由于其預(yù)測(cè)能力受用于訓(xùn)練的樣本特征影響,對(duì)于本文中所用樣本數(shù)據(jù),BP神經(jīng)網(wǎng)絡(luò)方法給出了精確度并不高的預(yù)測(cè),不過其預(yù)測(cè)精度仍然高于擬合直線預(yù)測(cè)精度,其振速預(yù)測(cè)結(jié)果依然滿足置信限控制要求。
[Abstract]:In recent years, with the sustained growth of the national economy and the deepening of the urbanization process, the construction of urban transport infrastructure is also developing rapidly. There are more and more tunnel projects in urban buildings, which are restricted by engineering geology, economy and so on. Most of the urban tunnel projects are still excavated by mining method. The excavation of tunnel blasting causes the vibration of the upper tunnel building group and has an impact on the daily life of the residents. It is very important to study the blasting vibration control technology of the tunnel in the underpass building group. Based on the project of Tongxiang Mountain Villa under the Binhai Avenue Tunnel in the south of Dalian, the field monitoring data of blasting vibration are analyzed by regression analysis and confidence study. The main research results are as follows: (1) for the measured vibration velocity data used in this study, the fitting line obtained by linear regression only reflects the average degree of data point distribution. 50% of the data points are distributed above the fitting line, while the remaining 50% data points are distributed under the fitting line. In practical engineering, if the fitting line is chosen as the criterion of vibration velocity control. The blasting vibration velocity is likely to exceed the control standard of vibration velocity (nearly 50%), thus causing many problems. At this time, it is necessary to introduce 95% confidence limit. The blasting vibration of 95% can be controlled under the control standard of vibration velocity by using the quantity of charge needed to calculate the confidence limit of 95%, which not only ensures the safety of the upper building, but also does not affect the construction. Compared with the conventional analysis and prediction model, the Bayesian theory based analysis and prediction model is more complex in principle and model realization. However, the prediction accuracy of the confidence limit of vibration velocity is very high, especially under the condition that the sample correlation is not high.) the sample data have a great influence on the accuracy of regression analysis and confidence limit prediction results, on the one hand. The number of samples has an impact on regression analysis and confidence limit prediction accuracy. In general, the more samples, the higher the prediction accuracy of confidence limit. On the other hand, the characteristics of sample data, such as correlation. The degree of dispersion also has an effect on the accuracy of confidence limit prediction. (4) in previous studies, BP neural network method has shown a high accuracy in nonlinear regression analysis. But because its prediction ability is affected by the sample characteristics used for training, the BP neural network method of sample data used in this paper is predicted with low accuracy. However, the prediction accuracy is still higher than that of fitting straight line, and the prediction result of vibration velocity still meets the requirement of confidence limit control.
【學(xué)位授予單位】：北京交通大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：U455.6

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 鄒寶平;楊建輝;王建秀;胡力繩;;基于灰色理論與BP神經(jīng)網(wǎng)絡(luò)的特長(zhǎng)瓦斯隧道爆破參數(shù)優(yōu)選[J];鐵道標(biāo)準(zhǔn)設(shè)計(jì);2015年08期

2 周宜;王海亮;張祖遠(yuǎn);孫松;;城市淺埋隧道下穿建筑物的微振動(dòng)爆破技術(shù)研究[J];隧道建設(shè);2015年01期

3 趙華兵;龍?jiān)?劉好全;李興華;鐘明壽;;城市隧道爆破振動(dòng)速度峰值預(yù)測(cè)神經(jīng)網(wǎng)絡(luò)法應(yīng)用[J];工程爆破;2014年05期

4 管曉明;傅洪賢;王夢(mèng)恕;崔X鵬;林凡濤;;基于OMA試驗(yàn)?zāi)B(tài)參數(shù)的砌體結(jié)構(gòu)有限元建模及修正[J];振動(dòng)與沖擊;2014年14期

5 楊慶;王海亮;王軍濤;;城市硬巖隧道下穿磚木結(jié)構(gòu)建筑爆破控制技術(shù)[J];現(xiàn)代隧道技術(shù);2014年01期

6 史秀志;武永猛;唐禮忠;黃宣東;;偏最小二乘回歸神經(jīng)網(wǎng)絡(luò)模型在爆破振動(dòng)峰值速度預(yù)測(cè)中的應(yīng)用[J];振動(dòng)與沖擊;2013年12期

7 盧文波;張樂;周俊汝;金旭浩;陳明;嚴(yán)鵬;;爆破振動(dòng)頻率衰減機(jī)制和衰減規(guī)律的理論分析[J];爆破;2013年02期

8 王云龍;譚忠盛;;淺埋大斷面隧道下穿樓群施工爆破控制[J];北京交通大學(xué)學(xué)報(bào);2012年04期

9 胡玉璽;李夕兵;;復(fù)雜采空區(qū)危險(xiǎn)辨識(shí)的貝葉斯判別方法及應(yīng)用(英文)[J];Transactions of Nonferrous Metals Society of China;2012年02期

10 楊珊;陳建宏;郭宏斌;羅正;;基于回歸分析的隧道爆破振動(dòng)傳播規(guī)律研究[J];中國(guó)安全科學(xué)學(xué)報(bào);2011年10期

相關(guān)博士學(xué)位論文 前1條

1 尹光明;城市隧道臨近建筑物超深基坑支護(hù)理論與安全控制技術(shù)研究[D];中南大學(xué);2012年

相關(guān)碩士學(xué)位論文 前8條

1 周梓彬;基于波形疊加的合理微差時(shí)間選取[D];東北大學(xué);2014年

2 張建新;基于貝葉斯方法的有限元模型修正研究[D];重慶大學(xué);2014年

3 張振;大跨度淺埋隧道下穿文保建筑群爆破減震技術(shù)研究[D];合肥工業(yè)大學(xué);2014年

4 汪亞飛;復(fù)雜環(huán)境下隧道鉆爆施工降振試驗(yàn)研究及控制[D];北京交通大學(xué);2013年

5 景海濤;基于貝葉斯網(wǎng)絡(luò)的巖石隧道風(fēng)險(xiǎn)分析[D];南京理工大學(xué);2012年

6 饒?jiān)?隧道鉆爆施工對(duì)初期支護(hù)的影響及控制標(biāo)準(zhǔn)的研究[D];北京交通大學(xué);2010年

7 趙常偉;微振動(dòng)爆破有害效應(yīng)控制與評(píng)估方法研究[D];山東科技大學(xué);2009年

8 蔡志偉;考慮松動(dòng)圈的卡爾曼濾波與有限元耦合反分析理論及其在圍巖穩(wěn)定分析中的應(yīng)用[D];重慶交通大學(xué);2008年

，

本文編號(hào)：1407254