基于改进的Elman神经网络的病虫害预测模型研究

Research on Pest and Disease Prediction Model Based on Improved Elman Neural Network

农业物联网是物联网应用的重要发展方向之一,为农业的发展带来了很好的前景。在农业物联网中,数据处理的研究所面临的挑战是缺乏考虑数据动态性和历史数据影响的预测模型,为此,提出了一种基于输出—输入反馈机制的神经网络,进一步提高网络的预测精度,使得网络具有良好的泛化能力和较强的鲁棒性,具有非线性时间序列的特征,应用领域较广,有较好的实际应用价值。

Agricultural iot is one of the important development directions of iot application, which brings a good prospect for the development of agriculture.In the Internet of things of agriculture, the challenge for the data processing of the institute is the lack of prediction model considering the influences of dynamic data and historical data, to this end, this paper proposes a neural network based on output-input feedback mechanism, further improves the prediction precision of the network, which not only has good generalization ability and strong robustness, but the characteristics of nonlinear time series, a wide application field, and good application value as well.