Font Size: a A A

Study On Water Demand Prediction Of Walnut Fruit During Swelling Period Based On Elman Neural Network Of Mind Evolutionary Algorithm

Posted on:2021-04-23Degree:MasterType:Thesis
Country:ChinaCandidate:H DengFull Text:PDF
GTID:2393330647964121Subject:Computer technology
Abstract/Summary:PDF Full Text Request
The shortage of water resources and irrational utilization are becoming more and more serious,and the amount of irrigation water used in agriculture accounts for 70%of all water.during the growth and development of crops,sufficient water supply must be ensured at all stages.Especially in certain critical periods,excessive watering will cause water waste,and insufficient watering will lead to lack of production.Therefore,the amount of irrigation for crops needs to be controlled within a reasonable range,which requires accurate prediction of crop water demand and water shortage.The planting area and output of walnut in my country are in the leading position in the world,and the research on it is very representative.Based on this,this article has done the following research:(1)Walnut water demand has complex nonlinear characteristics such as uncertainty and multiple inputs,and the relationship between various influencing factors is complicated.Neural networks have advantages in solving these problems.However,the traditional neural network will have problems of premature maturity,falling into local minimums,and random selection of initial weights and thresholds.Therefore,a mental evolutionary Elman neural network walnut water demand prediction model is proposed.TThe Mind Evolutionary Algorithm(MEA)has excellent global search capabilities,it can effectively solve such problems The prediction result is corrected by the error correction model based on the similar day algorithm to obtain a more accurate value of walnut water demand.(2)The predicted results are unavoidable from errors.The essence of error correction is to find a suitable correction factor to process the predicted results.Based on the traditional methods,this paper establishes an error correction model that combines the systematic clustering method,polynomial fitting method,and median method with the selection of similar days as a prerequisite.The analysis of the results shows that the predicted value of walnut water demand after model correction is closer to the actual value.(3)The results obtained from the previous work are only used to meet the water volume required for normal development of walnuts in the stage.However,in actual situations,natural rainfall should be taken into account.In the study of walnut fruitexpansion in area,it is precisely A period of very abundant rainfall.Therefore,the concept of effective rainfall is introduced to make a reasonable judgment on the utilization of natural rainfall in walnuts during the fruit expansion period and the actual water shortage of walnuts.Taking representative walnut crops as the research object,Combined with the historical data of the walnut fruit in the precision test field during the expansion period,the water demand forecasting work was carried out.Then introduce effective rainfall to make a reasonable judgment on the actual water shortage of walnuts during the period.It can provide an effective reference for realizing effective management of farmland water,saving agricultural water resources,and alleviating water shortages.
Keywords/Search Tags:water demand prediction, thinking evolution algorithm, Elman neural network, walnut water demand, error correction
PDF Full Text Request
Related items