Increasing the Accuracy of Soil Nutrient Prediction by Improving Genetic Algorithm Backpropagation Neural Networks

Soil nutrient prediction has been eliciting increasing attention in agricultural production. Backpropagation (BP) neural networks have demonstrated remarkable ability many scenarios. However, directly utilizing BP soil may not yield promising results due to the random assignment of initial weights and thresholds tendency fall into local extreme points. In this study, a network model optimized by an improved genetic algorithm (IGA) was proposed predict time series with high accuracy. First, crossover mutation operations (GA) were improved. Next, IGA used optimize model. The symmetric nature lies its feedforward feedback connections, i.e., same must be for forward backward passes. An empirical evaluation performed using annual data from China. pH, total nitrogen, organic matter, fast-acting potassium, effective phosphorus selected as indicators. IGA–BP, GA–BP, models compared analyzed. For IGA–BP model, coefficient determination pH 0.8, while those all greater than 0.98, exhibiting strong generalization ability. root-mean-square errors reduced 50% models. indicated that method can accurately content future series.