Based On Set Pair Analysis Of Water Resources System Prediction Method And Its Application

Water is a n irreplaceable and importa nt resource for hum an survival and development.When entering the 21 st cen tury,many countries are facing the challenges of water crisis,such as water shortage,water pollution,floods and other serious obstacles which hinder China's economic and social development,resulting in a series of social problem s.It is of gr eat significance to study the water resources forecasting problem and to realize the sustai nable utilization of water resources and promote the sustainable developm ent of water resources and econom y,society and ecological environment and alleviate the water crisis in China.In the pr ocess of water resources prediction,the main question is how to deal with the uncertainty,and the set pair analysis method has obvious advantages in dealing with these problems.Based on this,the paper mainly studies the forecasting method of water resources system and its application problem on the basis of set pair analysis theory,and obtains the following research results:(1)In view of the uncertaintie s in the current runof f classification,it is suggested that the appropriate runoff prediction model should be considered and the runof f data should be exam ined in order to ensure the accuracy of the forecast.Select the distribution curve of the annual runof f data,according to the cum ulative frequency of annual runoff in the fr equency curve to de termine the cr itical value of the states “wet”,”beween wet and dry”,and”dry”.The set of historical r unoff and pairing of current runoff are established and coupled with BP neural network method to predict runoff.The results show that the predicted method is effective.(2)Considering that there is subjectivity in the classification of runoff gradation,the fuzzy runoff grading standard is put forw ard,and the weight fa ctors of the runof f distribution are introduced,then calculate the com prehensive membership of each level of the runof f,establish the set of membership degree and ideal classification membership degree,introduce the concept of close degree in the fuzzy m athematics,use the m ethod of the fuzzy identification to classify the elem ents in the runof f collection.The results show that the relative error of the predicted results is small,and the classification method is accurate and effective.(3)To analysis and id entify the relevant influencing factors of urban water consumption,calculate the gray relational degree of each influencing factor,the influence factors of urban water consumption are determined by comparing the size of gray correlation degree,and the weight of each influence factor is calcu lated by gray relational degree.In the establishment of the set pair analysis cluster prediction model,the weights of each influencing factor are calculated by grey correlation degree and analytic hierarchy process.Establish the connection degree between the predicted sample and the reference system,predicate the amount of water comsumption about the forecasted years.The calculation of the former has a higher accuracy,and it is also the ideal method of predicting urban water consumption.