GA-BP神经网络的GPS可降水量预测Prediction of GPS perceptible water vapor based on GA-BP neural network
谢劭峰;赵云;李国弘;周志浩;
摘要(Abstract)：
针对传统BP神经网络模型存在的学习速度慢、易陷入局部极值以及网络结构参数取值的不确定性等问题,该文研究了一种基于遗传算法与BP神经网络相结合的GPS可降水量预测的新方法。该方法利用遗传算法对BP神经网络的初始权值和阈值进行优化,并对该模型进行训练,以提高预测模型的性能。实验结果证明了遗传BP神经网络模型用于GPS可降水量预测的可行性,其预测结果的均方根误差为0.16 mm、平均绝对百分误差为0.23%。相对于BP神经网络和小波神经网络模型,均方根误差分别降低了0.37和0.19 mm,平均绝对百分误差分别降低了0.62%和0.33%。同时遗传BP神经网络模型亦显示了很好的非线性拟合能力,能更好地预测GPS可降水量,对实际工作具有较强的参考价值。
关键词(KeyWords)： BP神经网络;遗传算法;GPS可降水量;预测
基金项目(Foundation): 国家自然科学基金项目（41864002,41704027）;; 广西自然科学基金项目（2018GXNSFAA281182）;; 广西中青年教师基础能力提升项目（2017KY0267）
作者(Authors): 谢劭峰;赵云;李国弘;周志浩;
DOI: 10.16251/j.cnki.1009-2307.2020.03.006
参考文献(References)：
- [1] 薛丰昌,高晓东,张亚琳.地基GPS反演大气水汽含量实践[J].全球定位系统,2012,37(3):33-36.(XUE Fengchang,GAO Xiaodong,ZHANG Yalin.Inversion of water vapor content in the atmosphere using foundation GPS practice[J].GNSS World of China,2012,37(3):33-36.)
- [2] 叶朗明,侯喜良,邵洋.GPS技术在水汽观测中的应用与前景[J].气象水文海洋仪器,2009(3):62-65.(YE Langming,HOU Xiliang,SHAO Yang.Application and prospects of ground-based GPS in water vapor observation[J].Meteorological,Hydrological and Marine Instrument,2009(3):62-65.)
- [3] 葛玉辉,熊永良,陈志胜,等.基于小波神经网络的GPS可降水量预测研究[J].测绘科学,2015,40(9):28-32.(GE Yuhui,XIONG Yongliang,CHEN Zhisheng,et al.Prediction method of GPS precipition based on wavelet neural network[J].Science of Surveying and Mapping,2015,40(9):28-32.)
- [4] 董淑华,邢贞相,娄丹,等.基于改进径向基函数的降雨量短期预测研究[J].沈阳农业大学学报,2017,48(3):367-372.(DONG Shuhua,XING Zhenxiang,LOU Dan,et al.Short-term rainfall forecasting based on a modified RBF function[J].Journal of Shenyang Agricultural University,2017,48(3):367-372.)
- [5] 刘严萍,杨晶,王勇.经验模态分解与神经网络的GPS可降水量预测[J].测绘科学,2013,38(5):91-93.(LIU Yanping,YANG Jing,WANG Yong.Prediction of GPS water vapor based on EMD and ANN[J].Science of Surveying and Mapping,2013,38(5):91-93.)
- [6] 王建敏,黄佳鹏,席克伟,等.大气可降水量预测模型的自适应Kalmam滤波改进[J].测绘科学,2017,42(12):127-133.(WANG Jianmin,HUANG Jiapeng,XI Kewei,et al.Improved prediction model precipitable water vapor using adaptive Kalmam filter [J].Science of Surveying and Mapping,2017,42(12):127-133.)
- [7] 张朝菡.基于EMD与RBF的GPS可降水量的降水预报研究[D].唐山:河北联合大学,2015.(ZHANG Chaohan.Precipitation.Research on precipitation forecast of GPS precioitable water vapor based on EMD and RBF[D].Tangshan:Hebei United University,2015.)
- [8] 文靳.神经网络理论与应用研究[M].成都:西南交通大学出版社,1996.(WEN Jin.Research of neural network theory and application[M].Chengdu:Southwest Jiaotong University Press,1996.)
- [9] 张金梦,刘慧君.遗传算法优化BP神经网络的泊车位数量预测[J].重庆大学学报,2018,41(3):76-81.(ZHANG Jinmeng,LIU Huijun.Prediction of spare parking spaces based on BP neural network optimized by genetic algorithm[J].Journal of Chongqing University,2018,41(3):76-81.)
- [10] 仲云飞,梅一韬,吴邦彬,等.遗传算法优化BP神经网络在大坝扬压力预测中的应用[J].水电能源科学,2012,30(6):98-101.(ZHONG Yunfei,MEI Yitao,WU Bangbin,et al.Parameters inversion analysis of facing rock fill dam based on BP neural network[J].Water Resources and Power,2012,30(6):98-101.)
- [11] 吴有训,王周青,汪文烈,等.遗传算法优化BP网络的汛期降水预测模型[J].安徽农业大学学报,2013,40(2):299-303.(WU Youxun,WANG Zhouqing,WANG Wenlie,et al.Optimization of BP neural network for short-term climate prediction of rainfall amount in flood season based on genetic algorithm[J].Journal of Anhui Agricultural University,2013,40(2):299-303.)