[1]王乙竹,陶伟,陆思宇.基于神经网络的“回南天”观测数据质量控制方法初探[J].气象研究与应用,2024,45(02):37-44.[doi:10.19849/j.cnki.CN45-1356/P.2024.2.06]
 WANG Yizhu,TAO Wei,LU Siyu.Preliminary study on the quality control method for observation data of“Continuous Wet Weather”based on neural network[J].Journal of Meteorological Research and Application,2024,45(02):37-44.[doi:10.19849/j.cnki.CN45-1356/P.2024.2.06]
点击复制

基于神经网络的“回南天”观测数据质量控制方法初探()
分享到:

气象研究与应用[ISSN:1673-8411/CN:45-1356/P]

卷:
第45卷
期数:
2024年02期
页码:
37-44
栏目:
研究论文
出版日期:
2024-06-15

文章信息/Info

Title:
Preliminary study on the quality control method for observation data of“Continuous Wet Weather”based on neural network
作者:
王乙竹1 陶伟1 陆思宇2
1. 广西壮族自治区气象技术装备中心, 南宁 530022;
2. 广西壮族自治区气象科学研究所, 南宁 530022
Author(s):
WANG Yizhu1 TAO Wei1 LU Siyu2
1. Guangxi Meteorological Technical Equipment Center, Nanning 530022, China;
2. Guangxi Institute of Meteorological Sciences, Nanning 530022, China
关键词:
质量控制反向传播神经网络粒子群优化“回南天”
Keywords:
quality controlback propagation neural networkparticle swarm optimizationContinuous Wet Weather
分类号:
P412
DOI:
10.19849/j.cnki.CN45-1356/P.2024.2.06
摘要:
为判别“回南天”观测设备数据可靠性,基于传统反向传播神经网络(BPNN),结合粒子群优化算法(PSO-BPNN),对广西“回南天”观测数据进行质量控制研究。结果表明:(1)在模型估算温度与实测温度对比验证中,与BPNN 模型相比,PSO-BPNN 模型精度更高,PSO-BPNN 模型没有明显高估或低估,而 BPNN 模型在 10 ℃附近出现较大偏差。(2)在使用测试集数据对模型进行测试中,瓷砖地面和墙面温度在 10~30℃范围,模型的适用性更强,PSO-BPNN模型稳定性优于BPNN模型。(3)在随机添加人工误差进行的模型检验中,PSO-BPNN模型瓷砖地面、墙面、水泥地面温度的最佳质量控制参数分别为1.73、1.64、1.68,BPNN模型分别为1.82、1.83、1.78。
Abstract:
In order to determine the reliability of the observation data of“Continuous Wet Weather”,a quality control study on the observation data of“Continuous Wet Weather”in Guangxi was carried out based on the traditional back-propagation neural network (BPNN),combined with particle swarm optimization (PSO) algorithm (i. e. PSO-BPNN). The results show that: (1) compared with the traditional BPNN model,the accuracy of the PSO-BPNN model is higher in comparing the model-estimated temperature with the measured temperature,without any significant overestimation or underestimation in the PSO-BPNN model,while the BPNN model shows a large deviation around 10 °C. (2) In the tests of PSO-BPNN and BPNN model,tile floor and wall temperatures in the range of 10~30 °C show greater applicability of the models,and the PSO-BPNN model is more stable than the BPNN model. (3) Randomly adding artificial errors for model validation,the optimal quality control parameters for the temperatures of tile ground,wall,and cement ground in the PSOBPNN model are 1.73,1.64,and 1.68,respectively,and 1.82,1.83,and 1.78 for the BPNN model, respectively.

参考文献/References:

[1] 罗小莉,古明悦,钟利华,等.近年广西典型"回南天"现象成因分析[J].气象科技,2015,43(4):659-665, 674.
[2] 陈见,李佳颖,高安宁,等.广西"回南天"发生特征及预报着眼点[J].气象,2015,41(3):372-379.
[3] 黄剑钊,李艳萍,陶伟,等.广西回南天观测系统装备与数据分析[J].气象研究与应用,2020,41(2):89-92.
[4] 张新甲,甘晓英,刘艳辉,等."回南天"自动监测仪的研发与应用前景[C]//中国气象学会.第28届中国气象学会年会——S1第四届气象综合探测技术研讨会.台山市气象局,2011:5.
[5] JIMENEZ P A,GONZALEZ R J F,NAVARRO J,et al. Quality assurance of surface wind observations from automated weather stations[J]. Journal of Atmospheric and Oceanic Technology,2010,27(7):1101-1122.
[6] QI Y C,MARTINAITIS S,ZHANG J,et al. A real-time automated quality control of hourly rain gauge data based on multiple sensors in MRMS system[J]. Journal of Hydrometeorology,2016,17(6):1675-1691.
[7] BEGES G,DRNOVSEK J,BOJKOVSKI J,et al. Automatic weather stations and the quality function deployment method[J]. Meteorological Applications,2015(22):861-866.
[8] RIENZNER M,GANDOLFI C. A procedure for the detection of undocumented multiple abrupt changes in the mean value of daily temperature time series of a regional network[J]. International journal of climatology,2013,33(5):1107-1120.
[9] 闵锦忠,王晨珏,贾瑞怡.苏皖地面自动站资料的质量控制及结果分析[J].大气科学学报,2018,41(5):637-646.
[10] HUDDARD K G,YOU J. Sensitivity analysis of quality assurance using the spatial regression approach-A case study of the maximum/minimum air temperature[J]. Journal of Atmospheric and Oceanic Technology,2005, 22(10):1520-1530.
[11] 殷利平,刘宵瑜,盛绍学,等.基于粒子群优化-BP神经网络-马尔科夫链的地面能见度观测资料质量控制[J].科学技术与工程,2022(13):5125-5133.
[12] SHA Y K,GAGNE L,DAVID J,et al. Deep-learningbased precipitation observation quality control[J]. Journal of Atmospheric and Oceanic Technology,2021,38(5):1075-1091.
[13] YE X L,YANG X,XIONG X,et al. A quality control method based on an improved random forest algorithm for surface air temperature observations[J]. Advances in Meteorology,2017:1-15.
[14] 闵晶晶,孙景荣,刘还珠,等.一种改进的BP算法及在降水预报中的应用[J].应用气象学报,2010,21(1):55-62.
[15] 张志富,任芝花,张强,等.自动站小时气温数据质量控制系统研究[J].气象与环境学报,2013,29(4):64-70.
[16] 杨娜,刘良明,向大享,等.利用BP神经网络由特征气象要素预测土壤湿度[J].土壤通报,2011,42(6):1324-1329.
[17] 张天虎,鲍艳松,钱芝颖,等.基于BP神经网络与遗传算法反演大气温湿廓线[J].热带气象学报,2020, 36(1):97-107.
[18] COUCEIRO M,SIVASUNDARAM S. Novel fractional order particle swarm optimization[J]. Applied Mathematics&Computation,2016(283):36-54.
[19] 韩力群.人工神经网络教程[M].北京:北京邮电大学出版社,2006:1-330.
[20] BAYRAM S,GEZICI S. Stochastic resonance in binary composite hypothesis-testing problems in the NeymanPearson framework[J]. Digital Signal Processing,2012, 22(3):391-406.
[21] 叶小岭,沈云培,熊雄.一种基于改进克里金法的地面气温质量控制算法[J].气候与环境研究,2016,21(5):614-620.

相似文献/References:

[1].自动站温度、雨量数据的质量控制方法和应用研究[J].气象研究与应用,2014,35(01):99.
[2]黄淑娟,陈健.自动站实时数据质量控制失败的案例分析[J].气象研究与应用,2014,35(02):72.
[3]杜丽英,王楚钦,余珂,等.地面气象观测资料一体化质控中出现的问题及处理[J].气象研究与应用,2016,37(02):75.
 DU Liying,WANG Chuqin,YU Ke,et al.The Problems and Treatment of the Integrated Quality Control of Surface Meteorological Observation Data[J].Journal of Meteorological Research and Application,2016,37(02):75.
[4]赵华睿,宋煜,李昱茜,等.暴雨强度公式编制之基础数据质量控制[J].气象研究与应用,2018,39(04):58.
 Zhao Huarui,Song Yu,Li Yuqian,et al.Basic data quality control for rainstorm intensity formula[J].Journal of Meteorological Research and Application,2018,39(02):58.
[5]刘昭武,刘非,张其忠.基于滨州市台站降水资料的空间内插方法分析[J].气象研究与应用,2018,39(04):63.
 Liu Zhaowu,Liu Fei,Zhang Qizhong.Analysis of Spatial Interpolation Method Based on Precipitation Data of Binzhou Station[J].Journal of Meteorological Research and Application,2018,39(02):63.
[6]侯江生,邹哲馨.贺州市区域自动气象站监控平台设计与实现[J].气象研究与应用,2018,39(04):74.
 Hou Jiangsheng,Zou Zhexing.Design and implementation of monitoring platform for Hezhou regional automatic weather station[J].Journal of Meteorological Research and Application,2018,39(02):74.
[7]黄剑钊,李艳萍,陶伟,等.广西回南天观测系统装备与数据分析[J].气象研究与应用,2020,41(02):89.[doi:10.19849/j.cnki.CN45-1356/P.2020.2.18]
 Huang Jianzhao,Li Yanping,Tao Wei,et al.Equipment and data analysis of Guangxi continuous wet weather observation system[J].Journal of Meteorological Research and Application,2020,41(02):89.[doi:10.19849/j.cnki.CN45-1356/P.2020.2.18]
[8]江益,王立俊,郑虹晖.海南省实时能见度质量控制方法及应用[J].气象研究与应用,2021,42(02):53.[doi:10.19849/j.cnki.CN45-1356/P.2021.2.10]
 Jiang Yi,Wang Lijun,Zhen Honghui.Research on Quality Control Method of Real-time Visibility of Hainan[J].Journal of Meteorological Research and Application,2021,42(02):53.[doi:10.19849/j.cnki.CN45-1356/P.2021.2.10]

备注/Memo

备注/Memo:
收稿日期:2023-11-2。
基金项目:广西气象科研计划项目(桂气科2021Z05)、广西壮族自治区气象技术装备中心自立项目(ZBKY202304)
作者简介:王乙竹(1996-),硕士,助理工程师,主要从事气象装备运行保障、质量控制研究。E-mail:617575956@qq.com
通讯作者:陶伟(1983-),高级工程师,主要从事综合气象观测研究。E-mail:249525294@qq.com
更新日期/Last Update: 1900-01-01