Agricultural Machine Well With Electric Water Folding Method And Its Application

There is a serious shortage of water resources in Hebei Province.With the development of the economy,the contradiction between supply and demand of water resources has become increasingly prominent.It is urgent to reform the water resources management method and to implement the reform of water resources fees.In March 2016,Hebei Province was identified as the only water resource tax reform pilot in the country,requiring to provide replicable,scalable experience and models to fully levy water tax to the country.There are a total of 1.04 million eye wells in Hebei Province,most of which don't have metering facilities installed,can't directly obtain water consumption data,so it is necessary to find a new method to determine the conversion coefficient between electricity consumed and water pumped.The influencing factor of the conversion coefficient between electricity consumed and water pumped was analyzed and the calculation method of its was proposed.The third method was predicted based on the measured data of typical wells in Hebei Province.The main research contents and results are as follows:(1)Analyze the influencing factors and distribution of the conversion coefficient between electricity consumed and water pumped,and find that it is affected by factors such as machine well efficiency,groundwater table burial depth,and machine well space layout,and the dispersion of the conversion coefficient between electricity consumed and water pumped in administrative regions and water resources in Hebei Province is more discrete.The average prediction error of the conversion coefficient between electricity consumed and water pumped of other wells is 22.53% or 27.74%,and the prediction error is larger.(2)Selected the groundwater level burial depth and the age of the well as the independent variables,the conversion coefficient between electricity consumed and water pumped as the dependent variable.Combined with the spatial distribution,the multivariate regression model is used to carry out regression analysis on the conversion coefficient between electricity consumed and water pumped of the wells in each administrative area.The prediction error less than 30% accounting for 63.32%;the prediction error of the conversion coefficient between electricity consumed and water pumped of each water resource division less than 30%accounting for 71.43%.(3)Taken the groundwater level burial depth,well age,pump age,position of the well(latitude and longitude coordinates)as input variables,the conversion coefficient between electricity consumed and water pumped as output variable,the support vector machine model is used to perform regression prediction and error analysis on the conversion coefficient between electricity consumed and water pumped of the wells in each administrative area.The prediction error less than 30% accounting for 81.16%.The average prediction accuracy is 17.71% higher than the average prediction and 24.45% higher than the multiple regression model.The hydropower conversion coefficients of each water resources division are predicted.The prediction error less than 30% accounting for 75.51%,the prediction accuracy is 23.95% higher than the average prediction,and 9.83% higher than the multiple regression model.It is a feasible method to predict the conversion coefficient between electricity consumed and water pumped of each well using the SVM model for the current situation in Hebei Province.