基于遗传小波神经网络的GPS可降水量预测  

Prediction of GPS precipitable water based on genetic wavelet neural network

作　　者：谢劭峰 苏永柠 王新桥 杨帆 梁春丽 邢尹 XIE Shaofeng;SU Yongning;WANG Xinqiao;YANG Fan;LIANG Chunli;XING Yin(College of Geomatics and Geoinformation,Guilin University of Technology,Guilin 541006,China;Guangxi Key Laboratory of Spatial Information and Geomatics,Guilin 541006,China)

机构地区：[1]桂林理工大学测绘地理信息学院,桂林541006 [2]广西空间信息与测绘重点实验室,桂林541006

出　　处：《中国科技论文》2018年第9期1041-1044,共4页China Sciencepaper

基　　金：国家自然科学基金资助项目(41704027);广西自然科学基金资助项目(2017GXNSFBA198139);广西中青年教师基础能力提升项目(2017KY0267;KY2016YB189);广西空间信息与测绘重点实验室资助课题(16-380-25-11;15-140-07-11)

摘　　要：针对GPS可降水量时间序列具有随机性和非线性的特点,利用遗传算法优化小波神经网络的输入参数,建立基于遗传小波神经网络的GPS可降水量预测模型。结果表明,遗传小波神经网络预测方法的均方根误差为0.124mm,平均绝对百分误差为0.167%,其精度相比BP神经网络和小波神经网络方法均有明显提高,能更好地反映可降水量的变化特征。Aiming at the random and nonlinear characteristic of the time series of GPS precipitable water vapor（PWV）,this paper used the genetic algorithm to optimize the input parameters of wavelet neural network and established a new prediction model of GPS PWV based on genetic wavelet neural network.The experimental results show that the root mean square error（RMSE）of genetic wavelet neural network prediction method is 0.124 mm,and its mean absolute percent error（MAPE）is 0.167%.Compared with the BP neural network and wavelet neural network methods,the accuracy of genetic wavelet neural network methods is obviously improved.This prediction model can better reflect the variation characteristics of PWV.

关 键 词：GPS可降水量 遗传算法 小波神经网络 BP神经网络 预测 

分 类 号：P457.6[天文地球—大气科学及气象学] P228.9