Research On Water Level Prediction Model Based On Linear Regression

River water level is an important basis for flood control and drought resistance,is to master the hydrological situation and projections of other hydrological elements of valuable information.Accurate observation of water level in flood control is an important noninvasive measure that can reduce losses and is also a non-engineering measure for rational use of water resources.So far,there are many problems in the measurement of river water level,such as measuring the river water level with water ruler,water jelly perennial soaking in the river is easy to be eroded,the instrument damage will produce a great error on the measurement results,the use of water meters to measure the water level work The amount of large,and manual measurement is also likely to cause errors;through the use of precision equipment to measure the river water level,the latter part of the equipment needs maintenance and repair,high cost and so on.It is important to find a way to accurately measure the water level and to reduce the cost of the inputs.In view of the above methods of measuring the water level above the above shortcomings,while the impact of water level in many factors,the rainfall changes is an important factor affecting the water level movements.It is important to find a way to accurately measure the water level and to reduce the cost of the inputs.Therefore,this paper is based on the linear regression model to study the relationship between rainfall and water level.The rainfall is used as the independent variable and the water level is the dependent variable.The linear regression model is used to predict the water level by rainfall.The work of the first from the professional website,such as the Yellow River network,the Central Meteorological Observatory and Henan rainfall summary query system collected from 2011 to 2016 historical rainfall and water level data.And data mining,data mining by day by day,will be repeated and incomplete data removed for the establishment of the model to provide clean,accurate and concise data to improve the accuracy of rainfall and water level forecast.The rainfall and water level data from 2011 to 2016 are used as training sets,and the rainfall and water level data from 2014 to 2016 are used as test sets.The regression parameters in the linear regression equation are calculated by using the data in the training set,and the regression test is carried out to analyze the predicted results.The predicted results are compared with the actual observations.And then use the test data to test the accuracy of the regression equation.The results show that the linear regression model can accurately predict the water level by rainfall.