大数据驱动的我国典型重点流域水污染防控决策研究

Water pollution control in major river basins and pollutant emissions reduction were used as important tasks in environmental protection work in the new period of China. Traditional environmental management and decision-making were transform the linear paradigm of management process-dominated to the flattening paradigm data-centered in the context of big data. Concerning the current situation of the lack of accurate technology method to manage in the major rivers of China, water pollution control and decision-making were investigated to our major rivers in the context of big data. Environmental big data for pollution control in major rivers would be constructed by combing analyzation of current situation, using Songhua River Basin as research subject, collecting big data of multi-source and heterogeneous in major rivers and water environment, and researching its value. The technical research of multi-source and heterogeneous data fusion would be investigated basing the grey system model and the method of computational intelligence. The negative list for pollution control in major rivers would be found by forecasting analysis and statistic inference the environment big data of multi-source and heterogeneous. Developing the big data association analysis for manage and control in major rivers, modeling the data-driven model universality to research, formulating the data-driven model under the global schema, an integrated management approach and a treatment scheme of the pollutant source will be formatted. Building an integrated virtual-visual management platform which facing the circumstances of the aquatic ecological environment panorama integrated management and decision, to provide science and technology support for the integrated status of environmental management.

重点流域水污染防治以及污染物排放量的大力削减是新时期环保工作的重要任务。在大数据背景下，环境管理与决策模式正在从传统的管理流程为主的线性范式向以数据为中心的扁平化范式转变。针对我国重点流域精准治理管理技术方法不足的现状，开展大数据背景下我国典型重点流域水污染防控管理与决策情景研究。结合现状分析，以松花江流域为研究对象，通过重点流域水环境多源异构大数据收集及数据价值发现研究，建立重点流域污染防控环境大数据集。开展多源异构数据融合技术研究，基于灰色系统模型和计算智能方法，对环境多源异构大数据进行统计推断与预测分析，建立重点流域水污染防控负面清单。开展重点流域水污染防控大数据关联分析和数据驱动模型普适性建模研究，制定全局模式下数据驱动模型，形成污染源综合管理方法、环境精准治理方案。面向流域水生态环境全景式综合管理与决策情景，构建虚拟可视化综合管理平台。为有效提高环境综合管理水平提供科技支撑。

针对我国重点流域精准治理管理技术方法不足的现状，开展了大数据背景下我国典型重点流域水污染防控管理与决策情景研究。结合现状分析，以松花江流域为研究对象，通过松花江流域水环境多源异构大数据收集及数据价值发现研究，统筹水资源、水生态与水环境问题，建立了松花江流域污染防控大数据集。开展多源异构数据融合技术研究，基于灰色系统模型和计算智能方法，对环境多源异构大数据进行了统计推断与预测分析，开展松花江流域水污染防控大数据关联分析，建立起流域水污染防控水环境质量底线及管控分区、水资源利用上线及生态环境准入清单，形成了污染源综合管理方法、环境精准治理方案。构建面向流域水生态环境管理重大需求的全景式管理、智能分析、群体决策的一体化大数据可视化决策平台，为有效应对黑龙江伊春“3·28”鹿鸣矿业尾矿库溢流事故提供科技支撑，并参与撰写2020年4月突发性水污染事件应急技术研发与应用进展及防范对策建议科技专报，通过技术增强引入生态环境治理体系新视角，提升政府决策效率，降低决策风险，推动我国流域精准治理工作向智能化、规范化方向发展。