摘要
以云南省程海为例,基于CE-QUAL-W2计算平台开发了水动力-水质模型.由于程海现有的数据不足以支撑开发可靠的流域水文与污染物负荷模型,为弥补由此引起的水质模型边界条件缺失的限制,本研究提出了基于逆向模拟与遗传算法的多模式负荷-参数识别方法来实现对水质模型的校正和鲁棒校验.应用校验后的模型对程海不同水位调控情景的水质响应进行鲁棒预测分析,为决策者提供了基于不确定性的量化水质响应,从而有助于制定可靠的管理策略.研究结果表明,程海水位调控对TN、TP及COD等水质指标的改善作用非常有限.
A water quality model was developed through incorporating the water surface elevation and water quality data of Lake Chenghai into the CE-QUAL-W2 computational platform.The model integrates the water surface elevation and water quality into a holistic dynamic system based on the data of Lake Chenghai,and was calibrated against observed data using a multiple pattern inverse water quality modeling technology,which was driven by a robust genetic algorithm（GA）.After the model was calibrated,it was used to produce robust predictions of the lake water quality in response to various water elevation controlled scenarios.The model established a basis for quantifying the water quality responses under uncertainty,and is valuable for supporting effective and reliable management decision making.The results of this research suggest that various water elevation control scenarios only result in insignificant water quality improvement in terms of TN,TP,and COD concentrations,therefore,it does not recommend to consider water elevation control to be the major water quality management option for Lake Chenghai.
出处
《环境科学》
EI
CAS
CSCD
北大核心
2011年第11期3193-3199,共7页
Environmental Science
基金
云南省科技厅社会事业发展专项(2009CA002)
国家水体污染控制与治理科技重大专项(2008ZX07102-001)
关键词
水质水动力模型
逆向模拟
遗传算法
多模式参数识别
鲁棒预测
hydrodynamic and water quality model
inverse modeling
genetic algorithms（GA）
multiple pattern parameter identification
robust prediction
