湖库藻类水华智能识别与预测研究
发布时间：2019-07-05 20:56
【摘要】：当前,我国大多数湖库水体富营养化现象较为突出。由于水体中积聚了大量的氮、磷等营养物质,导致一些藻类异常繁殖,不断积聚而形成不同程度的蓝藻水华,如何对蓝藻水华这一水环境污染进行识别与预测预警,已经成为当今水环境领域研究的重点之一。本文综合分析了国内外湖库藻类水华识别与预测的研究现状,对湖库藻类水华的智能识别与预测方法进行了深入研究。首先,在对湖库水体遥感反演方法深入研究的基础上,提出了基于D-S证据理论的湖库站点监测与遥感监测的信息融合方法,实现了对关注区域内的蓝藻水华的有效识别;其次,通过对湖库水体富营养化评价指标的综合分析,采用核主成分分析法确定了蓝藻水华形成与暴发的关键影响因素,构建了基于误差补偿的蓝藻水华时序综合预测模型;在此基础上,考虑到自然湖库中环境因素对蓝藻水华形成的影响特征,采用自适应模糊推理专家系统对影响蓝藻水华暴发的表征因素叶绿素a进行预测,一定程度上解决了在环境突变情况下蓝藻水华预测精度不高的问题;最后,将研究成果嵌入到湖库水质监测与蓝藻水华预测预警系统中,并将其应用到实际湖库中,为环保部门进行湖库水环境监测和信息管理提供了辅助决策平台。
[Abstract]:At present, the eutrophication of most lakes and reservoirs in China is more prominent. Due to the accumulation of a large number of nitrogen, phosphorus and other nutrients in the water body, some algae reproduce abnormally and accumulate constantly to form different degrees of cyanobacteria blooms. How to identify, predict and warn the water environment pollution of cyanobacteria blooms has become one of the key points in the field of water environment. In this paper, the research status of algae bloom identification and prediction in lake and reservoir at home and abroad is comprehensively analyzed, and the intelligent recognition and prediction method of algae bloom in lake and reservoir is deeply studied. Firstly, based on the in-depth study of remote sensing inversion method of lake and reservoir water body, the information fusion method of lake reservoir site monitoring and remote sensing monitoring based on D / S evidence theory is proposed, and the effective identification of blue algae blooms in the area of concern is realized. Secondly, through the comprehensive analysis of the evaluation index of eutrophication in lake and reservoir, the key influencing factors of the formation and outbreak of cyanobacteria bloom were determined by nuclear principal component analysis, and the comprehensive prediction model of cyanobacteria bloom time series based on error compensation was constructed. On this basis, considering the influence of environmental factors on the formation of cyanobacteria blooms in natural lakes, the adaptive fuzzy reasoning expert system is used to predict the characterization factor chlorophyll a, which can solve the problem that the prediction accuracy of cyanobacteria blooms is not high in the case of environmental mutation. Finally, the research results are embedded in the lake reservoir water quality monitoring and blue algae bloom prediction and early warning system, and applied to the actual lake reservoir, which provides an auxiliary decision-making platform for environmental protection departments to carry out lake and reservoir water environment monitoring and information management.
【学位授予单位】：北京工商大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：X524
本文编号：2510806
[Abstract]:At present, the eutrophication of most lakes and reservoirs in China is more prominent. Due to the accumulation of a large number of nitrogen, phosphorus and other nutrients in the water body, some algae reproduce abnormally and accumulate constantly to form different degrees of cyanobacteria blooms. How to identify, predict and warn the water environment pollution of cyanobacteria blooms has become one of the key points in the field of water environment. In this paper, the research status of algae bloom identification and prediction in lake and reservoir at home and abroad is comprehensively analyzed, and the intelligent recognition and prediction method of algae bloom in lake and reservoir is deeply studied. Firstly, based on the in-depth study of remote sensing inversion method of lake and reservoir water body, the information fusion method of lake reservoir site monitoring and remote sensing monitoring based on D / S evidence theory is proposed, and the effective identification of blue algae blooms in the area of concern is realized. Secondly, through the comprehensive analysis of the evaluation index of eutrophication in lake and reservoir, the key influencing factors of the formation and outbreak of cyanobacteria bloom were determined by nuclear principal component analysis, and the comprehensive prediction model of cyanobacteria bloom time series based on error compensation was constructed. On this basis, considering the influence of environmental factors on the formation of cyanobacteria blooms in natural lakes, the adaptive fuzzy reasoning expert system is used to predict the characterization factor chlorophyll a, which can solve the problem that the prediction accuracy of cyanobacteria blooms is not high in the case of environmental mutation. Finally, the research results are embedded in the lake reservoir water quality monitoring and blue algae bloom prediction and early warning system, and applied to the actual lake reservoir, which provides an auxiliary decision-making platform for environmental protection departments to carry out lake and reservoir water environment monitoring and information management.
【学位授予单位】：北京工商大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：X524
【参考文献】
相关期刊论文 前1条
1 杨宁;冀德刚;李双金;;Pearson相关分析法在京津冀空气质量分析中的应用（英文）[J];Agricultural Science & Technology;2015年03期
，本文编号：2510806
本文链接：https://www.wllwen.com/kejilunwen/huanjinggongchenglunwen/2510806.html
