Research On Flood Control Management Optimization Of Sanyuan River Basin In Guizhou Province Based On K-means And ARMA Water Situation Analysis

As a result of global warming,extreme weather events are on the rise,making it difficult to follow regular patterns of events such as rainfall and floods.Good river basin flood control management plays a huge role in ensuring the safety of people’s lives and property,and also reflects the ability of government departments to deal with public crises.River basin flood control should be the result of comprehensive control of engineering measures and management measures,and should not only rely on the construction of flood control projects,but neglect the weakening of management.Big data technology is used to optimize the management of small watershed strategy,is a new attempt,scientific management,digital study not only can enrich the theoretical basis of the flood control and management decision-making and management of flood control and the actual decision optimization has good auxiliary function,to explore new type of small watershed flood control and management has a positive practical significance.Through cleaning the watershed data of Sanyuan River basin in Guizhou for several years,the K-means clustering algorithm of Jupyter Notebook analysis tool was used to analyze the extract of flood elements in the basin,and the relevant hydrological characteristics were obtained.ARMA time series algorithm is used to predict the flow of the basin,and the results are basically consistent with the actual situation of the basin.Combined with the current situation of flood control management in Sanyuan River basin of Guizhou province,the optimization of flood control management based on data analysis is put forward,which shows that the management can be optimized.It is feasible to optimize flood control management by using big data technology to analyze the characteristics and changes of river basin water regime.Although it is not difficult to obtain watershed data at present,there is still a long way to go to better use these data to improve management,which is also the future development direction of management science.