基于神经网络和证据理论的农田虫害预测算法
|
摘要
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 farmers'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 about 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.
Farmland pest prediction is a key part of promoting agricultural development and increasing farmers'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 about 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.
引文
[1]许章华,黄旭影,林璐,等.基于Fisher判别分析与随机森林的马尾松毛虫害检测[J].光谱学与光谱分析,2018,38(9):2888-2896.Xu Zhanghua,Huang Xuying,Lin Lu,et al.Dendrolimus punctatus walker damage detection based on Fisher discriminant analysis and random forest[J].Spectroscopy and Spectral Analysis,2018,38(9):2888-2896.
[2]Karp D S,Judson S,Daily G C,et al.Molecular diagnosis of bird-mediated pest consumption in tropical farmland[J].Springerplus,2014,3(1):1-8.
[3]窦志国,崔丽娟,武高洁,等.芦苇粉大尾蚜虫害下芦苇叶绿素高光谱反演估算[J].生态学杂志,2018,37(10):3163-3170.Dou Zhiguo,Cui Lijuan,Wu Gaojie,et al.Estimation of the hyperspectral inversion of reed chlorophyll under Hyalopterus pruni attack[J].Chinese Journal of Ecology,2018,37(10):3163-3170.
[4]张浩,王国伟,苑超,等.基于AIGA-BP神经网络的粮食产量预测研究[J].中国农机化学报,2016,37(6):205-209.Zhang Hao,Wang Guowei,Yuan Chao,et al.Research on forecast of grain production based on AIGA-BP neural network[J].Journal of Chinese Agricultural Mechanization,2016,37(6):205-209.
[5]王笑岩,王石.基于BP神经网络的辽宁省农机总动力预测[J].中国农机化学报,2015,36(2):314-317.Wang Xiaoyan,Wang Shi.Prediction on total power of agricultural machinery in Liaoning Province based on BP neural network[J].Journal of Chinese Agricultural Mechanization,2015,36(2):314-317.
[6]尤文坚,叶雪英,唐仕云.基于径向基神经网络农机数量预测的研究[J].中国农机化学报,2013(2):38-41.Ju Wenjian,Ye Xueying,Tang Shiyun,Research on forecast of the number of agricultural machinery based on RBFneural network[J].Journal of Chinese Agricultural Mechanization,2013(2):38-41.
[7]鲁敏,岑红蕾,王洪坤.基于LM-BP的新疆玛纳斯灌区用水量预测[J].中国农机化学报,2014,35(2):75-77.Lu Min,Yan Honglei,Wang Hongkun.Prediction of Xinjiang manas irrigation area water consumption based on LM-BP[J].Journal of Chinese Agricultural Mechanization,2014,35(2):75-77.
[8]王丽艳,郭树国.基于BP神经网络玉米蛋白粉吸水性的预测[J].中国农机化学报,2013,34(6):125-128.Wang Liyan,Guo Shuguo.Water absorption index prediction for corn gluten meal based on BP neural network[J].Journal of Chinese Agricultural Mechanization,2013,34(6):125-128.
[9]Monck-Whipp L,Martin A E,Francis C M,et al.Farmland heterogeneity benefits bats in agricultural landscapes[J].Agriculture Ecosystems&Environment,2017,253:131-139.
[10]韦艳玲.一种改进的RBF神经网络在预测虫害中的应用研究[J].科学技术与工程,2013,13(1):136-139.Wei Yanling.Application research on an improved RBFneural network in pests forecast[J].Science Technology and Engineering,2013,13(1):136-139.
[11]Rusch A,Valantin-Morison M,Roger-Estrade J,et al.U-sing landscape indicators to predict high pest infestations and successful natural pest control at the regional scale[J].Landscape and Urban Planning,2012,105(1-2):62-73.
[12]张子恺,济航,王上,等.基于Apriori算法的森林虫害预测方法[J].东北林业大学学报,2017,45(8):93-96.Zhang Zikai,Ji Hang,Wang Shang,et al.Forest pest prediction method with Apriori algorithm[J].Journal of Northeast Forestry University,2017,45(8):93-96.
[13]Xiao Z,Ye S J,Zhong B,et al.BP neural network with rough set for short term load forecasting[J].Expert Systems with Applications,2009,36(1):273-279.
[14]Yang R,Er P V,Wang Z,et al.An RBF neural network approach towards precision motion system with selective sensor fusion[J].Neurocomputing,2016,199(3):31-39.
[15]Jiang H,Wang R,Gao Z,et al.Classification of weld defects based on the analytical hierarchy process and Dempster-Shafer evidence theory[J].Journal of Intelligent Manufacturing,2017(6):1-12.
[2]Karp D S,Judson S,Daily G C,et al.Molecular diagnosis of bird-mediated pest consumption in tropical farmland[J].Springerplus,2014,3(1):1-8.
[3]窦志国,崔丽娟,武高洁,等.芦苇粉大尾蚜虫害下芦苇叶绿素高光谱反演估算[J].生态学杂志,2018,37(10):3163-3170.Dou Zhiguo,Cui Lijuan,Wu Gaojie,et al.Estimation of the hyperspectral inversion of reed chlorophyll under Hyalopterus pruni attack[J].Chinese Journal of Ecology,2018,37(10):3163-3170.
[4]张浩,王国伟,苑超,等.基于AIGA-BP神经网络的粮食产量预测研究[J].中国农机化学报,2016,37(6):205-209.Zhang Hao,Wang Guowei,Yuan Chao,et al.Research on forecast of grain production based on AIGA-BP neural network[J].Journal of Chinese Agricultural Mechanization,2016,37(6):205-209.
[5]王笑岩,王石.基于BP神经网络的辽宁省农机总动力预测[J].中国农机化学报,2015,36(2):314-317.Wang Xiaoyan,Wang Shi.Prediction on total power of agricultural machinery in Liaoning Province based on BP neural network[J].Journal of Chinese Agricultural Mechanization,2015,36(2):314-317.
[6]尤文坚,叶雪英,唐仕云.基于径向基神经网络农机数量预测的研究[J].中国农机化学报,2013(2):38-41.Ju Wenjian,Ye Xueying,Tang Shiyun,Research on forecast of the number of agricultural machinery based on RBFneural network[J].Journal of Chinese Agricultural Mechanization,2013(2):38-41.
[7]鲁敏,岑红蕾,王洪坤.基于LM-BP的新疆玛纳斯灌区用水量预测[J].中国农机化学报,2014,35(2):75-77.Lu Min,Yan Honglei,Wang Hongkun.Prediction of Xinjiang manas irrigation area water consumption based on LM-BP[J].Journal of Chinese Agricultural Mechanization,2014,35(2):75-77.
[8]王丽艳,郭树国.基于BP神经网络玉米蛋白粉吸水性的预测[J].中国农机化学报,2013,34(6):125-128.Wang Liyan,Guo Shuguo.Water absorption index prediction for corn gluten meal based on BP neural network[J].Journal of Chinese Agricultural Mechanization,2013,34(6):125-128.
[9]Monck-Whipp L,Martin A E,Francis C M,et al.Farmland heterogeneity benefits bats in agricultural landscapes[J].Agriculture Ecosystems&Environment,2017,253:131-139.
[10]韦艳玲.一种改进的RBF神经网络在预测虫害中的应用研究[J].科学技术与工程,2013,13(1):136-139.Wei Yanling.Application research on an improved RBFneural network in pests forecast[J].Science Technology and Engineering,2013,13(1):136-139.
[11]Rusch A,Valantin-Morison M,Roger-Estrade J,et al.U-sing landscape indicators to predict high pest infestations and successful natural pest control at the regional scale[J].Landscape and Urban Planning,2012,105(1-2):62-73.
[12]张子恺,济航,王上,等.基于Apriori算法的森林虫害预测方法[J].东北林业大学学报,2017,45(8):93-96.Zhang Zikai,Ji Hang,Wang Shang,et al.Forest pest prediction method with Apriori algorithm[J].Journal of Northeast Forestry University,2017,45(8):93-96.
[13]Xiao Z,Ye S J,Zhong B,et al.BP neural network with rough set for short term load forecasting[J].Expert Systems with Applications,2009,36(1):273-279.
[14]Yang R,Er P V,Wang Z,et al.An RBF neural network approach towards precision motion system with selective sensor fusion[J].Neurocomputing,2016,199(3):31-39.
[15]Jiang H,Wang R,Gao Z,et al.Classification of weld defects based on the analytical hierarchy process and Dempster-Shafer evidence theory[J].Journal of Intelligent Manufacturing,2017(6):1-12.