تخمین عملکرد گندم آبی با استفاده از روش الگوریتم ترکیبی شبکه های عصبی مصنوعی و الگوریتم ژنتیک

Abstract

The study area is located in northwest of east Azarbaijan. In this study, 80 soil profiles were surveyed in irrigated wheat farms and soil samples were taken from each genetic horizon for physical and chemical analyses. In this region soil moisture and temperature regimes are Aridic border to Xeric and Mesic, respectively. The soils were classified as Entisols and Aridisols. We used 1×1 m wooden square plots in each profile to determine the amounts of yield. Because of nonlinear trend of yield, a nonlinear algorithm hybrid technique (neural-genetics) was used for modeling. At first step, the average weight of soil characteristics (from depth of 100 cm) and landscape parameters of selected profiles were measured for modeling according to the annual growing period of wheat. Then, land components and wheat yield considered inputs and output of model, respectively. For this reason, genetic algorithm was investigated to train neural network. Finally, estimated wheat yield was obtained using input data. Root mean square error (RMSE) and Coefficient of determination (r2) indexes was used for assessing the capability of this method. The sensitivity analysis of model showed that soil and land parameters such as total nitrogen, available phosphorus, slope percentage, content of gravel, soil reaction and organic matter percentage act as important characteristics in amount of wheat yield in study area. The most effective soil index in quality of produced wheat and quantity of its yield is total nitrogen based on the correlation between surveyed features and yield, and soil organic matter has at least correlation. We found that network learning process based on genetic algorithms in the learning process has less error. Findings showed that beside of confirming the desired results in the case of using sigmoid activation function in the hidden layer and linear activation function in the output layer of all neural networks demonstrated the proposed hybrid technique had much better results. These findings also confirm better prediction ability of neural network based on error back propagation algorithm or Levenberg-Marquardt training algorithm compared to other types of neural network confirms. Using nonlinear techniques in modeling and forecasting wheat yield due to its nonlinear trend and influencing variables is inevitable. Recently, genetic algorithms and neural network techniques is considered as the most important in modeling nonlinear and complex processes. Despite the advantages of these techniques there are a lot of weaknesses. Impose specific conditioned form by researchers in the techniques of genetic algorithms and stopping neural network learning at the optimal points are the main weaknesses of these techniques, instead the search for global optimal point and not impose a specific functional forms are strengths of genetic algorithm techniques and neural networks, respectively. Results of this study indicated that the proposed hybrid technique has much better results. So that, correlation coefficient (0.92) and average deviation square error (357.9) were high and low, respectively. It is obvious the surveyed soil indexes have very strong relationship with the yield. So, implementation of land management practices is inventible for improving soil and land characteristics to maintain high yield, Preventing land degradation and preserve it for future generations to create sustainable development