Based on Corrected Model of General Regression Neural Network for Forecasting Water Consumption of Populus (P.×euramericana cv.“74/76<i>”</i>) Seedlings

Abstract

Water consumption of plants is a key parameter for formulating irrigation system, and the precise prediction play a important role in improving the use efficiency of limited water resources. In this experiment, by using the method of General Regression Neural Network (GRNN) and MATLAB DATA PROCESSING SYSTEM combined with the meteorological data of air temperature, relative air humidity, solar radiation, wind speed, soil water content and dew point temperature as the input variable, the author established the artificial neural network system to forecast the seedling water consumption of P.×euramericana cv.“74/76”, and made the amendment to the Neural Network Model, by comparison between original model and corrected model, we draw the conclusion that all parameters, except solar radiation, had better correlation between corrected values and measured values: relative air humidity (0.99>0.98), air temperature (0.62>0.61), dew point temperature (0.96>0.82), soil water content (0.56>0.54), wind speed (0.89>0.86), and the corrected model had higher forecasting accuracy. In addition, the maximum relative error of GRNN corrected model was 0.104, the minimum relative error was 0.010, the average relative error was 0.04. The corrected model is superior to the original model that the former performs a higher forecasting accuracy with relatively shorter time consumption and faster speed in training. Therefore, we propose that GRNN model and its corrected model can be used in prediction of seedling water consumption.