Bayes理论在河流水质模型参数识别中的应用

参数识别是水环境数学模型建模的重要步骤.在实际模拟过程中,往往难以获得理想的数据进行模型参数识别.充分利用研究者已有的经验,可在一定程度上减少模拟过程的风险.Bayes理论为把研究者的经验或前验信息纳入到水质模拟提供了一个定量手段.采用离散Bayes理论的基本方法,以国内某河段实际监测数据为基础,完成了模型的参数识别过程,并对识别结果进行了分析.验证结果表明,采用Bayes理论获得的参数识别结果能够达到模型验证的要求.

Application of Bayes theorem in parameter identification for river water quality modeling

Parameter identification is one of the most important steps in water quality modelling. But it often happens that there is not enough data to support parameter identification, and thus it becomes dependent on the experience of the researchers to reduce the modelling risk to some extent. Bayes theorem makes it possible to quantify the experience of the researcher into the modelling. On the basis of the monitoring data of a river section, this paper identifies the parameters of a CSTR model, and the identification results are also analyzed. The verification shows that the identification results are reasonably acceptable.