摘要
山区地质复杂、环境敏感、生态脆弱,致使铁路隧道建设难度大,且在资源节约、环境保护等方面面临较高要求。绿色设计作为绿色建设的先行环节,对减少隧道建设对资源环境的扰动作用,达到资源节约、环境友好的目的起关键引领作用。为定量评估山区铁路隧道工程的资源环境影响效应、衡量隧道设计的绿色程度,提出一种山区铁路隧道工程资源环境影响效应分析方法。首先,基于“驱动力-状态-响应”模型(Driving force-State-Response,DSR),以隧道设计参数为驱动力指标、资源环境状况为状态指标、隧道设计措施为响应指标,构建山区铁路隧道工程资源环境影响效应评估指标体系,并建立各指标分级标准;其次,运用一种具有自学习自调整能力的支持向量回归模型(Support Vector Regression,SVR)对隧道工程的资源环境影响效应进行评估;再次,以驱动力指标和响应指标为分析对象,运用考虑指标间关联关系及叠加效应的敏感性分析法,甄别对隧道工程资源环境影响效应优化具有重要影响的隧道设计因素;最后,以某山区铁路隧道工程为例,得到该隧道工程的资源环境影响效应值为4.7157,对应等级为较好,表明该隧道工程绿色设计水平较好,资源集约节约利用较合理、环境保护力度较大,可为其他类似工况隧道工程的绿色设计提供借鉴。此外,分析结果显示,注浆加固效果是导致该隧道工程资源环境影响效应变化最敏感的因素,其次为清污分流比例,可着重从这2个方面进行优化设计,以实现隧道工程资源环境影响效应的进一步优化。研究结果验证了本文研究方法的适用性和可操作性,可为山区铁路隧道工程资源环境影响效应评估及明确隧道工程优化设计方向提供科学依据。
The complex geology,sensitive environment,and fragile ecology in mountainous areas make railway tunnel construction difficult,and face great requirements in resource conservation and environmental protection.Green design,as a leading link in green construction,plays a key leading role in reducing the disturbance of tunnel construction on resources and environment,and achieving the goal of resource conservation and environmental friendliness.To quantitatively evaluate the level of resource and environmental impact effects of railway tunnel engineering in mountainous areas and measure the green degree of tunnel design,a method for analyzing the resource and environmental impact effects of railway tunnel engineering in mountainous areas was proposed.First,based on the“Driving force-State-Response”model,with tunnel design parameters as the driving force indices,resource and environmental conditions as the status indices,and tunnel design measures as the response indices,a resource and environmental effect evaluation index system for railway tunnel engineering in mountainous areas was constructed,and grading standards for each index were established.Second,a support vector regression model with self-learning and self-adjusting ability was used to evaluate the level of resource and environmental effects of tunnel engineering.Once again,by taking driving force indices and response indices as the analysis objects,the sensitivity analysis method considering the correlation relationship and superposition effects among indices was used to identify tunnel design factors that have important impacts on the optimization of resource and environmental impact effects in tunnel engineering.Finally,by taking a railway tunnel engineering in a mountainous area as an example,it is obtained that the level of resources and environment of the tunnel engineering is 4.7157,and the corresponding grade is good,indicating that the green design level of the tunnel engineering is good,the intensive and economical utilization of resources is reasonable,and the environmental protection is strong,which can provide reference for the green design of tunnel engineering under other similar conditions.In addition,the analysis results show that the grouting reinforcement effect is the most sensitive factor leading to changes in the resource and environmental effects of the tunnel engineering,followed by the proportion of clean and sewage diversion.Optimization design can be focused on these two aspects to further optimize the resource and environmental effects of the tunnel engineering.The research results validate the applicability and operability of the research method in this article,which can provide scientific basis for evaluating the resource and environmental impact effects of railway tunnel engineering in mountainous areas and clarify the direction of tunnel engineering optimization design.
作者
闫林君
陈慧鑫
鲍学英
王起才
李亚娟
YAN Linjun;CHEN Huixin;BAO Xueying;WANG Qicai;LI Yajuan(School of Civil Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)
出处
《铁道科学与工程学报》
EI
CAS
CSCD
北大核心
2024年第4期1613-1623,共11页
Journal of Railway Science and Engineering
基金
中央引导地方科技发展资金资助项目(22ZY1QA005)
国家自然科学基金资助项目(51768034)。
关键词
山区铁路隧道工程
资源环境影响效应
“驱动力-状态-响应”模型
支持向量回归模型
敏感性分析
railway tunnel engineering in mountainous areas
resource and environmental impact effects
“Driving force-State-Response”model
support vector regression model
sensitivity analysis