数据挖掘驱动的水沙模型时序耦合关键理论与技术研究

With the increasing demand for river flow-sediment simulation and the continuous improvement of numerical simulation theory and hydroinformatic technology, new requirements have been imposed for flow-sediment simulation, such as simulation spatiotemporal range expansion, strengthening of space-time mutual feedback, and simulation efficiency and accuracy enhancement. Based on the research of multi-model temporal data interaction method and spatial boundary hydraulic element connection model, this study explores the inherent law of model state spatiotemporal data, innovates in the research on data mining based multi-model temporal coupling evaluation and dynamic modification method, construct a cloud computing environment that adapts to heterogeneous model temporal coupling, achieves the simulation of the transportation of river water, sediment and their associated elements and the change trend of river itself, to provide technical support for the research on flow-sediment regulation and multi-material flux changes of the whole river. The Big Data and cloud computing technology introduced by multidisciplinary approach will also inject new impetus for the development of the river flow-sediment numerical simulation.

随着河流水沙模拟需求的不断提升，数值模拟理论与水信息学技术的不断进步，对水沙模拟提出了扩展模拟时空范围、加强模拟时空互馈，提升模拟效率和精度的新需求。本研究通过研究多模型时序间数据交互方法与空间边界水力要素连接模式等理论，探索模型状态时空属性数据的内在规律，创新基于数据挖掘多模型时序耦合评价与动态修正方法研究，构建适应异构模型时序耦合的云服务计算环境，实现河流水沙及其伴生要素输移及河流自身演变趋势模拟，为全河水沙调控和多物质通量变化研究提供技术支撑，多学科交叉引入的大数据与云计算技术也会为河流水沙数值模拟学科发展注入新的动力。

随着河流水沙模拟需求的不断提升，数值模拟理论与水信息学技术的不断进步，对水沙模拟提出了扩展模拟时空范围、加强模拟时空互馈，提升模拟效率和精度的新需求。本研究改进了模型耦合的数据交互方法，突破模型云化部署的关键技术，实现河流水沙及其伴生要素输移及河流自身演变趋势模拟，为全河水沙调控和多物质通量变化研究提供技术支撑，具有重要的理论价值和生产实践意义。成果不仅丰富了水沙数值模拟学科内容，而且可以广泛应用于水旱灾害防御、水资源管理与调配等生产实践中，具有良好的应用前景和推广价值。