摘要
利用兰州大学半干旱气候与环境观测站(SACOL站)2009-2010年的地基微波辐射计亮温资料和榆中站探空资料,基于伪逆学习算法建立了应用于地基微波辐射计温度、相对湿度和水汽密度反演的神经网络(PIFN),并将反演结果与地基微波辐射计自带反演产品进行了对比,研究了伪逆学习算法在地基微波辐射计气象要素反演算法本地化的应用效果.结果表明:PIFN反演的温度、相对湿度和水汽密度的均方根误差的最大值分别为6.41K,31.21%和1.5g/m3,地基微波辐射计温度、相对湿度和水汽密度产品的均方根误差最大值分别为11.93K,53.18%和3.06g/m3,与微波辐射计自带神经网络反演结果在不同高度层进行比较可以看出PIFN对2~10km、1~7km和0~3km的大气温度、相对湿度和水汽密度廓线的反演均有明显改善,伪逆学习算法能够应用于地基微波辐射计气象要素的反演算法的本地化.
In order to estimate the application of pseudoinverse learning algorithm in the localization of ground-based microwave radiometer meteorological elements inversion algorithms,the observed data of the ground-based microwave radiometer from the Semi-Arid Climate and Environment Observatory(SACOL)of Lanzhou University and the radiosonde data from the Yuzhong Station during 2009 to 2010are used to establish a neural network(PIFN)for temperature,humidity and water vapor density inversion based on the pseudoinverse learning algorithm(PLA),and the inversion results are compared with the products of the ground-based microwave radiometers.The result show that the maximum mean square root error of temperature,relative humidity and water vapor density inversed by PIFN are 6.41 K,31.21% and 1.5g/m3,respectively,and the maximum root mean squared root error of temperature,relative humidity,and water vapor density products recorded by the ground-based microwave radiometers are11.93 K,53.18%and 3.06g/m3,respectively.PIFN significantly improves the inversion performance of temperature,relative humidity and water vapor density profiles between 2~10km,0~3km,1~7km,respectively.It is concluded that the inversion result of PLFN has a better performance than the microwave radiometer's own products and is more close to the radiosonde data and that PLA can be introduced to ground-based microwave radiometer inversion algorithm localization field.
引文
[1]郭丽君,郭学良.利用地基多通道微波辐射计遥感反演华北持续性大雾天气温、湿度廓线的检验研究[J].气象学报,2015,73(2):368-381.
[2]赵金霞,范苏丹,朱晓晶.微波辐射计资料在大雾预报预警中的应用[J].沙漠与绿洲气象,2015,9(3):31-35.
[3]唐仁茂,李德俊,向玉春,等.地基微波辐射计对咸宁一次冰雹天气过程的监测分析[J].气象学报,2012,70(4):806-813.
[4]赵兵科,邵德民,鲁小琴,等.多通道地基微波辐射计在0713号“韦帕”台风登陆前后探测性能及特征分析[J].地球科学进展,2009,24(9):1001-1008.
[5]王叶红,赖安伟,赵玉春.地基微波辐射计资料同化对一次特大暴雨过程影响的数值试验研究[J].暴雨灾害,2010,29(3):201-207.
[6]鲍艳松,钱程,闵锦忠,等.利用地基微波辐射计资料反演0~10km大气温湿廓线试验研究[J].热带气象学报,2016,32(2):163-171.
[7] CIMINI D,HEWISON T J,MARTIN L,et al.Temperature and Humidity Profile Retrievals from Ground-Based Microwave Radiometers During TUC[J].Meteorologische Zeitschrift,2006,15(5):45-56.
[8]刘亚亚,毛节泰,刘钧,等.地基微波辐射计遥感大气廓线的BP神经网络反演方法研究[J].高原气象,2010,29(6):1514-1523.
[9]黄兴友,张曦,冷亮,等.基于MonoRTM模型的微波辐射计反演方法研究[J].气象科学,2013,33(2):138-145.
[10]GUO P,WANG K,XIN X.Autoencoder,Low Rank Approximation and Pseudoinverse Learning Algorithm[C]//2017IEEE International Conference on Systems.Banff,BB,Canada:Man and Cykemetics,2017:948-953.
[11]GUO P,LYU M R.A Pseudoinverse Learning Algorithm for Feedforward Neural Networks with Stacked Generalization Applications to Software Reliability Growth Data[J].Neurocomputing,2004,56(1):101-121.
[12]WANG K,GUO P,YIN Q,et al.A Pseudoinverse Incremental Algorithm for Fast Training Deep Neural Networks with Application to Spectra Pattern Recognition[C]//2016International Joint Conference on Neural Networks.BC,Canada:Vancouver,2016:3453-3460.
[13]GUO P,LYU M R.A Pseudoinverse Learning Algorithm for Feedforward Neural Networks with Stacked Generalization Applications to Software Reliability Growth Data[J].Neurocomputing,2004,56(1):101-121.
[14]赵玲,马玉芬,张广兴,等.MP-3000A微波辐射计的探测原理及误差分析[J].沙漠与绿洲气象,2009,3(5):54-57.
[15]王振会.大气温度分布的地面微波遥感数值实验[J].南京气象学院学报,1988,11(3):346-355.