The Research About Intelligent Recognition And Predictive Method Of Algal Bloom In Lakes And Reservoirs

Nowadays, the phenomenon of water eutrophication is very outstanding in lakes and rivers in our country. Due to the accumulation of abundant nitrogen, phosphorus and other eutrophic materials in water, some dominant algae has an abnormal reproduction which leads to algal bloom in varying degrees. How to effectively identify the algal bloom and carries on the effective forecast has become one of the key research fields of water environment.In this article, the current research status about identification and prediction methods towards blooms have been analyzed comprehensively and some further researches have been done as well. Firstly, on the basis of deep research about remote sensing inversion method, information fusion method is put forward based on the information of site monitoring combined with remote sensing ones by means of D-S evidence theory so that the recognition of blooms in concerned region is achieved. Secondly, with the comprehensive analysis towards the eutrophication evaluation index, KPCA is adopted to confirm key factors affecting the formation of blooms thereby the comprehensive time series predictive model of blooms based on error compensation is established. Thirdly,considered the fact that the weather factors in natural lakes do have the influence on blooms, ANFIS is adopted to make prediction towards chlorophyll which is the characteristic factor of blooms so that the problem of low precision under the condition of abrupt change of weather is solved. Finally, the research studied above is embedded into the lake water quality monitoring and algae bloom forecast early warning system which is an effective manage tool for environmental protection administration.