基于BP人工神经网络的土壤含水量预测模型的研究

Study on Forecasting Model of Soil Water Content based on BP Artificial Neural Network

土壤含水量是制约植物生长的主要因子之一,因此科学地预测土壤含水量对充分利用土壤水资源具有十分重要的意义。提出了基于BP人工神经网络的土壤含水量预测模型,BP人工神经网络采用收敛速度较快和误差精度较高的动量-自适应学习速率调整算法。并通过基于这种模型的土壤含水量预测实验,结果表明BP人工神经网络预测模型提高了收敛速度和减少陷入局部最小的可能,并且提高了预测精度。

Soil moisture is one of the factors restricting plant growth, so it has great significance to scien- tifically forecast soil moisture for making full use of soil water. In this paper, the soil moisture predicting mod- el was put forward based on the BP neural network. The BP neural network using adaptive learning rate momentum algorithm has fast convergence rate and high error precision. According to the soil moisture forecast experiment, the BP neural network predicting model increased the convergence rate, reduced the possibility of getting into local minimum, and improved the prediction accuracy.