基于PSO-BP神经网络的草原放牧策略研究

Research on Grassland Grazing Strategy Based on PSO-BP Neural Network

摘要: 土壤有机碳是土壤化学物质的重要组成部分，建立其与其它化学物质的灰色关联模型，同时与放牧强度进行联合分析，得出适当放牧有助于提高土壤有机碳的循环。由于预测数据维度较多，选取BP神经网络模型能够有效预测，采用粒子群算法(Particle swarm optimization, PSO)对神经网络模型的参数进行优化，提高模型精度，最后求得土壤化学物质预测表。

Abstract: Soil organic carbon is an important component of soil chemicals. A grey correlation model was Established between organic carbon and chemicals, and conducting joint analysis with grazing intensity. It is concluded that appropriate grazing can help improve the cycling of soil organic carbon. Due to the large number of dimensions in the prediction data, A BP neural network model was selected to predict effectively. PSO is used to optimize the parameters of the neural network model and improve model accuracy. Finally a soil chemical substance prediction table was obtained.