中国农业科学院农业信息研究所 中国农学会
Agricultural Information Institute of CAAS China Association of Agricultural Science Societies
En
Ch
中国农业科学院农业信息研究所 中国农学会 主办

AI检索

订阅

返回顶部
当前位置: 首页 > 文章 > 基于PCA-NARX神经网络的溶解氧预测 山东农业大学学报(自然科学版) 2019,50 (5) 902-907
Position: Home > Articles > Prediction of Dissolved Oxygen Based on PCA-NARX Neural Network Journal of Shandong Agricultural University(Natural Science Edition) 2019,50 (5) 902-907

基于PCA-NARX神经网络的溶解氧预测

作  者：
袁红春;黄俊豪;赵彦涛
单  位：
上海海洋大学信息学院
关键词：
溶解氧预测;NARX神经网络;主成分分析
摘  要：
溶解氧是水产养殖中的一项重要水质参数,为了准确掌握溶解氧的变化趋势,提出了基于PCA-NARX神经网络的溶解氧预测模型.通过主成分分析法提取的主成分变量作为网络输入,优化了网络结构,并根据渔业养殖用水溶解氧标准,进行了NAR、NARX模型对溶解氧的短期(64 h)预测实验对比,仿真结果表明:PCA-NARX模型在16 h内均方根误差(RMSE)最小;32、48 h内,NAR模型与PCA-NARX模型预测精度基本一致;总体64 h之内,PCA-NARX模型相对于NAR、NARX模型具有更好的泛化能力,对溶解氧的预测性能较好.
译  名：
Prediction of Dissolved Oxygen Based on PCA-NARX Neural Network
作  者：
YUAN Hong-chun;HUANG Jun-hao;ZHAO Yan-tao;College of Information/Shanghai Ocean University;
关键词：
Prediction of dissolved oxygen;;NARX neural network;;principal component analysis
摘  要：
Dissolved oxygen is an important aquatic parameter. In order to accurately grasp the trend of the dissolved oxygen accurately, the dissolved oxygen prediction model based on the PCA-NARX neural network is developed in this paper. The principal component variables extracted by principal components analysis(PCA) are used as exogenous inputs and the network structure was optimized,and short-term(64 h) prediction experiments of dissolved oxygen by NAR and NARX models were compared according to the standards of dissolved oxygen in fishery and aquaculture water. Simulation results show that the PCA-NARX model has a minimum root mean square error(RMSE) within 16 h,the prediction accuracy of the NAR model and the PCA-NARX model is basically the same within 32, 48 h. In addition, the comparisons with other models show that PCA-NARX neural network has better nonlinear fitting ability and superior in dissolved oxygen prediction based on the RMSE in short term(64 h). In total, within 64 h, PCA-NARX model has better generalization ability than the NAR and NARX model and better prediction performance for dissolved oxygen.

相似文章

导出引用
计量
文章访问数: 13
HTML全文浏览量: 0
PDF下载量: 0

所属期刊

推荐期刊

友情链接Friendly link
农业信息研究所Agricultural information institute 农业智能知识服务平台Agricultural intelligent knowledge service platform 农业学术服务平台Agricultural Knowledge Service System 中国知网CNKI 维普网VIP 万方WANFANG DATA
联系我们Contact us

电话Tell：010-82109652-808

邮箱E-mail：agrijournal@caas.cn

服务Service
关于我们About us 意见建议Opinions and suggest
版权说明Copyright
单位：中国农业科学院农业信息研究所 中国农学会 © 2017中国农业科学院农业信息研究所 ICP备案号：京ICP备09089781号-46