基于神经网络和证据理论的农田虫害预测算法

Farmland pest prediction algorithm based on neural network and evidence theory

农田虫害预测是促进农业发展和增加农民收入的关键部分。针对目前农田虫害预测算法准确性差和适应性不佳的问题,提出一种基于神经网络和证据理论的农田虫害预测算法。该方法首先分别采用BP神经网络、RBF神经网络和Elman神经网络进行虫害预测,然后利用证据理论中的组合决策思想,结合神经网络预测结果,进行权值提取和权值融合,最后通过融合后的权值实现农田虫害预测。试验结果表明,权值融合后具有更高的预测精度,相比神经网络传统预测方案,该方法的预测精度相比BP神经网络、RBF神经网络和Elman神经网络分别提升了约5倍、3倍和2倍,预测性能优于任何一种单一神经网络模型。

Farmland pest prediction is a key part of promoting agricultural development and increasing f*armers income.Aiming at the problem of poor accuracy and adaptability of farmland pest prediction algorithm.a farmland pest prediction algorithm based on neural network and evidence theory is proposed.Firstly,BP neural network,RBF neural network and Elman neural network were used to predict pests.And then,neural network prediction results were combined based on the fusion rule in evidence theory.*Finally the model of weighted fusion is used for farmland pest prediction.The experimental results showed that the weighted model has higher prediction accuracy,and compared with the traditional neural network prediction scheme,The prediction accuracy of this method is flbout 5 times,3 times and 2 times higher than BP neural network,RBF neural network and Elman neural network respectively.Predictive performance is superior to any single neural network model.