滑坡变形预测与失稳预报问题的几点讨论

Discusses on some problems of deformation and failure prediction for landslides

滑坡问题的研究关系到水利、交通、能源等基本工程建设,并与人们生命财产安全、生态环境保护及可持续发展问题相关,因此一直备受关注。滑坡是一个复杂的地质体,影响滑坡稳定的内外各种因素多种多样,影响程度各异,并且各因素又是动态变化的,决定了滑坡变形机制与变形特点十分复杂,因此滑坡变形预测与失稳预报十分困难。文章认为,变形突变预测、失稳时间预报及失稳判据是滑坡预测预报的3个核心问题,变形突变预测是为了预测出滑坡从匀速变形阶段进入加速变形阶段的特征点或时间,是回答滑坡能否失稳的关键,是提高预测结果可靠性及提高预测研究水平的基础。分析了目前各种预测模型的特点及可靠性情况,模型的优缺点。探讨了变形突变难以预测的实质,并指出采用多因素预测模型进行探索性研究。文章还对失稳时间预报的预报尺度、预测误差的计算、滑坡空间预测及多测点协同预测等问题进行分析,并对可能的解决办法提出了建议或展望。

Landslide research is related to basic construction of water conservancy engineering,transportation engineering, energy resource engineering, etc, as well as human safety, environment protection and sustainable development, to which a great deal of attention has been paid. There are various complex factors which affect landslide stability, and the affecting degrees of different factors are changeable dynamically. The mechanisms and features of landslide deformation are very complicated. So the prediction of deformation and failure of landslide is very difficult.Deformation mutation prediction, failure time prediction and failure judging criterion are three core problems in the researches of deformation prediction. The purpose of deformation mutation prediction is to analyze or predict the time which indicates the deformation from uniform speed stage to accelerated one. It is the key to failure prediction, and the base to promote reliability of prediction results and research level of landslide prediction. Then, various prediction models at present are sorted, models' reliabilities and their advantages and disadvantages are discussed. The paper analyzed the reasons of difficulty in deformation mutation prediction. Multi-factor prediction models are suggested in order to explore the puzzle. In addition, prediction dimension of failure time prediction, calculation of prediction error, spatial prediction and multi-monitoring-spot cooperating prediction are analyzed in the paper, some possible approaches in landslide prediction researches are suggested or prospected.