列车振动环境影响预测的准确度与可靠性研究

With the large-scale construction and networked operation of urban rail transit, the environmental impact of train vibrations is becoming serious problem. Accurate and reliable prediction is the key procedure to solve this problem. A design stage-accompanied system of prediction methodology with three levels have been put forward by the applicant and the scholars at home and abroad, but it is still limited to the fixed value prediction rather than the stochastic processes prediction. Therefore, the accuracy and reliability of the prediction results can not be evaluated scientifically, and accordingly the design and evaluation of engineering vibration isolation is still blind, which is also an important factor that the investment risk is out of control. In order to solve this problem, the applicant will combine the rich cases and related research projects accumulated over the past ten years to explore the forecastology research on the environment impact of train vibrations from the perspective of stochastic system for the first time. By systematically analyzing the stochastic factors, especially paying attention to the research of vibration source and the influence to prediction results by the combination of stochastic factors of prediction approaches, accuracy and reliability calculation method for prediction result of each level can be put forward. In addition, the prediction principle to ensure the accuracy and reliability will also be built. The expected results of this project can provide a scientific and reasonable solution in avoiding the blindness of environmental impact assessment and vibration isolation design of railway vibration, reducing the investment risk of vibration abatement and noise reduction in rail transit engineering, and improving the level and quality of environmental impact assessment.

随着城市轨道交通大规模建设及成网运行，列车振动的环境影响问题日趋严重。对其进行准确而可靠的预测是解决这一问题的关键环节。申请人及国内外学者已提出了分阶段三等级的预测方法体系，但仍囿于定值预测而非随机过程。因此预测结果的准确度及可靠性无法科学评价，故而减隔振设计与评估仍然存在盲目性，这亦是投资风险失控的重要因素。为解决这一问题，申请人将结合近十多年来积累的丰富案例及相关课题，首次探索从随机系统角度开展列车振动环境影响的预测学研究。通过对各随机因素的系统分析，重点研究振源、预测方案组合随机因素变量对预测结果的影响，进而提出各等级预测结果的准确度及可靠性计算方法，同时提出保障相应准确度及可靠性的预测方法工作原则。本项目的预期研究成果，可为避免轨道交通振动的环境影响评估及减隔振设计工作的盲目性，降低轨道交通工程建设在减振降噪方面的投资风险，提高环境影响评价水平与质量，提供一个科学合理的解决方法。

轨道交通在给人们的出行带来便利的同时，其运行产生的振动也不可避免地会对周边的居民、古建筑和精密仪器产生消极的影响。对轨道交通振动环境影响进行精准预测是解决这一问题的关键环节。传统的预测方法局限于确定性框架内，且对预测结果的精准度和预测方法的可靠性鲜有研究。针对这一问题，本课题项目组在近二十年工作积累的基础上，针对有关的关键科学问题，开展了四项工作，尝试将确定性预测扩展到了概率预测，并且对精准度和可靠性进行了探索研究。1）课题组首先建立了实测数据集，为整个研究工作提供了数据基础；2）针对振源的不确定性，我们建立了列车动荷载识别模型，能够根据实测信号用动态反演分析方法得到列车动荷载，并对荷载的不确定性进行了量化；3）针对振动传播路径的不确定性，我们建立了考虑地层参数随机性的响应面预测模型。探索了多指标评价预测结果精准度的方法；4）最后，我们针对预测方法的可靠性评价问题，提出了对预测方法可靠性评价的多元指标法，并提出了提高预测方法可靠性的技术策略。作为应用案例，针对北京地区的地质特点，我们对《环评导则》推荐的链式公式预测方法进行了可靠性研究。提出了提高链式公式预测方法可靠性的策略。本项目的研究成果将有利于提高地铁振动环境影响预测的精准度，提高预测可靠性，有利于降低根据环评预测结果采取工程措施的风险，对轨道交通工程减振降噪措施的选择做出更为科学合理的决策。在本项目资助下，我们公开发表相关期刊论文共计11篇；培养博士研究生3人，硕士研究生3人；参加国内外学术会议交流共计5人次；申请专利1项。本项目结余经费83028元，将用于本课题的后续研究工作中。