摘要
目前,水体富营养化及蓝藻水华污染已成为最为突出的世界性水环境问题之一,做好水华的预测预警工作显得尤为重要。预警模型是水体富营养化及蓝藻水华预警工作的前提与基础。文章讨论了主要预警参数因子的选择,包括物理因子、化学因子和生物因子;介绍了常用的水体富营养化及蓝藻水华预警模型的研究进展,总结了各个常用模型的概念、优缺点、建模思路以及应用现状。神经网络模型已成为国内外预警模型应用的热点,其中遗传算法模型、支持向量机模型、决策树方法受到了广泛的重视,另外,一些其它的模型得到了一定程度的应用,如模糊系统预测模型、灰色系统预测模型。综合运用各种预警监测技术,获取实时监测数据,提高预警模型的准确性是今后水华预警研究的重点方向。
Since water eutrophication and cyanobacteria bloom are becoming one of the most serious worldwide water environment problems,it is very important to predict cyanobacteria bloom.Model is the precondition and foundation of the early warning for water eutrophication and cyanobacteria bloom.Choice of main parameters and factors for early warning of cyanobacteria bloom were discussed,including physical factors,chemical factors and biotic factors.Progress of early warning model for water eutrophication and cyanobacteria bloom was also introduced,with analysis of the concepts,advantages and disadvantages,model building idea and application of commonly used models.Artificial neural network model has become the application focus for early warning model,in which genetic algorithm model,support vector machine model and decision trees model aroused wide attentions.Some other models are used to some extent,such as fuzzy system prediction model and grey system prediction model.It was proposed that applying a variety of monitoring technologies,acquiring in-time monitoring data,and raising accuracy of early warning model is the trend for further studying of early warning of water eutrophication and cyanobacteria bloom.
出处
《环境科学与技术》
CAS
CSCD
北大核心
2011年第11期152-157,204,共7页
Environmental Science & Technology
基金
国家"863"计划重点课题(2009AA063005)
安徽省科技攻关计划(08010301107)
关键词
富营养化
蓝藻水华
预警参数因子
预警模型
eutrophication
cyanobacteria bloom
early warning parameters and factors
early warning model
