三峡库区EFDC模型集成与应用

为满足三峡库区大尺度的水质预测需求，降低专业水质模型使用门槛，实现模型业务化运行，采用涵盖"模型封装-接口服务-系统集成"全过程的模型封装集成技术方法，开展了EFDC模型集成研究，并在三峡库区水质安全评估与预警系统中进行验证和应用.结果表明:①基于实例化模型建立的模型预置参数库，是模型业务化运行中参数简化的前提，可有效解决模型本地驯化及模型实例的基础数据复用问题；②依据参数识别结果，将模型参数识别为"必调参数、可调参数、默认参数"三类，并采用B/S架构对输入文件、输出文件和主控文件中的不同参数进行分类封装，达到参数简化目的；③采用半紧密型方式建立的基于Web services的EFDC模型集成接口服务及标准，可为信息系统提供通用的EFDC模型计算服务；④该方法在EFDC模型与三峡库区水质安全评估与预警系统集成中得到验证，实现了全库区干流及主要支流COD_Cr、总磷、总氮等污染因子的长时间序列水动力水质预测联机在线运算，满足了用户在水质影响预测预警中低修改量、高运算效率和良好用户体验等需求，实现了模型的业务化运行. 

The EFDC Model Integration and Application in the Three Gorges Reservoir

In order to meet the needs of large-scale water quality prediction in the Three Gorges Reservoir (TGR) area, it is important to lower the threshold of professional water quality model and realize the business operation of the model. A model encapsulation integration technique, which covers the whole process of''model encapsulation-interface service-system integration'' was used to study the integration of EFDC model, and has been tested and applied in Water Quality Safety Assessment and Early Warning System for the TGR Area. The results showed that:1) The preset parameter database based on the instantiation model were the precondition for reducing parameters to facilitate the business operation of the mode. Tt could solve the calibration problem for a certain area, and made it easy to reuse some essential data when necessary. 2) According to parameter identification results, the model parameters were firstly divided into three classes:namely the imperatively-adjusted-parameter, the adjustable-parameter and the default-parameter. Using B/S architecture, the input files, output files and master files with classified parameters were also encapsulated to achieve parameter simplification. 3) The interface services and interface standard of model integration of Environmental Fluid Dynamics (EFDC), set up by Web services with half-compact type, provided the united service of EFDC to information systems. 4) The method has been verified by Water Quality Safety Assessment and Early Warning System for the TGR Area. The on-line calculation and water quality prediction with long time series for the pollutants of COD_Cr, total phosphorus, total nitrogen was realized in the main stream and some tributaries of the TGR. It meets the demand of low modification, high computing efficiency and good user experience in the water quality forecast and warning, and t realized the business operation of the model.