色一情一区二,毛多色婷婷

基于Stacking機(jī)器學(xué)習(xí)模型的南海北部海溫預(yù)報(bào)

單位：1. 國(guó)家海洋環(huán)境預(yù)報(bào)中心自然資源部海洋災(zāi)害預(yù)報(bào)技術(shù)重點(diǎn)實(shí)驗(yàn)室, 北京 100081;
2. 廈門大學(xué)海洋與地球?qū)W院, 福建廈門 361102

關(guān)鍵詞：機(jī)器學(xué)習(xí) Stacking 南海北部海溫預(yù)報(bào)

分類號(hào)：P731.31

出版年·卷·期（頁(yè)碼）：2023·40·第一期（39-45）

摘要：

基于 Stacking（ET-ET）的機(jī)器學(xué)習(xí)算法，利用美國(guó)國(guó)家環(huán)境預(yù)報(bào)中心再分析數(shù)據(jù)和MGDSST海溫融合數(shù)據(jù)，建立了一套高效的海溫長(zhǎng)期預(yù)報(bào)方法，并在南海北部海域開展了1 a的表層海溫長(zhǎng)期預(yù)報(bào)實(shí)驗(yàn)。結(jié)果表明：基于Stacking（ET-ET）機(jī)器學(xué)習(xí)模型的表層海溫長(zhǎng)期預(yù)報(bào)的均方根誤差降至0.52 ℃，平均絕對(duì)百分比誤差降至1.58%，明顯優(yōu)于基于支持向量機(jī)、人工神經(jīng)網(wǎng)絡(luò)和長(zhǎng)短期記憶模型的預(yù)報(bào)結(jié)果。

In this paper, an efficient long-term SST forecast method is established based on Stacking (ET-ET) machine learning algorithm using reanalysis data of National Centers for Environmental Prediction and Mergid satellite and in situ data Global Daily sea surface temperature (SST) fusion data, and long-term SST forecast experiment is carried out in the northern South China Sea for one year. The results show that the root mean square error of long-term SST forecast based on Stacking (ET-ET) machine learning model is reduced to 0.52 ℃, and the mean absolute percentage error is reduced to 1.58%, which is significantly better than the forecast results based on the support vector machine, artificial neural network and long short-term memory model.

參考文獻(xiàn)：

[1] 張建華. 海溫預(yù)報(bào)知識(shí)講座第一講海水溫度預(yù)報(bào)概況[J]. 海洋預(yù)報(bào), 2003, 20(4):81-85. ZHANG H J. Lecture on knowledge of sea temperature prediction, Lecture 1, Overview of sea water temperature prediction[J]. Marine Forecasts, 2003, 20(4):81-85.
[2] 吳磊, 王彬, 潘錫山, 等. 融合海表溫度產(chǎn)品在渤黃東海的對(duì)比分析及初步驗(yàn)證[J]. 海洋通報(bào), 2020, 39(6):657-668. WU L, WANG B, PAN X S, et al. Intercomparison analysis of merged sea surface temperature products for the Bohai, Yellow and East China Seas[J]. Marine Science Bulletin, 2020, 39(6):657-668.
[3] 吳新榮, 王喜冬, 李威, 等. 海洋數(shù)據(jù)同化與數(shù)據(jù)融合技術(shù)應(yīng)用綜述[J]. 海洋技術(shù)學(xué)報(bào), 2015, 34(3):97-103. WU X R, WANG X D, LI W, et al. Review of the application of ocean data assimilation and data fusion techniques[J]. Journal of Ocean Technology, 2015, 34(3):97-103.
[4] 王輝, 萬(wàn)莉穎, 秦英豪, 等. 中國(guó)全球業(yè)務(wù)化海洋學(xué)預(yù)報(bào)系統(tǒng)的發(fā)展和應(yīng)用[J]. 地球科學(xué)進(jìn)展, 2016, 31(10):1090-1104. WANG H, WAN L Y, QIN Y H, et al. Review of the application of ocean data assimilation and data fusion techniques[J]. Advances in Earth Science, 2016, 31(10):1090-1104.
[5] 劉娜, 王輝, 凌鐵軍, 等. 全球業(yè)務(wù)化海洋預(yù)報(bào)進(jìn)展與展望[J]. 地球科學(xué)進(jìn)展, 2018, 33(2):131-140. LIU N, WANG H, LING T J, et al. Review and prospect of global operational ocean forecasting[J]. Advances in Earth Science, 2018, 33(2):131-140.
[6] 張培軍, 周水華, 梁昌霞. 基于衛(wèi)星遙感海溫?cái)?shù)據(jù)的南海SST預(yù)報(bào)誤差訂正[J]. 熱帶海洋學(xué)報(bào), 2020, 39(6):57-65. ZHANG P J, ZHOU S H, LIANG C X. Study on the correction of SST prediction in South China Sea using remotely sensed SST[J]. Journal of Tropical Oceanography, 2020, 39(6):57-65.
[7] ZHANG Q, WANG H, DONG J Y, et al. Prediction of sea surface temperature using long short-term memory[J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14(10):1745-1749.
[8] 郝日栩, 趙玉新, 何忠杰, 等. 基于EOF-NAR神經(jīng)網(wǎng)絡(luò)混合模型的海溫預(yù)報(bào)方法研究[C]//中國(guó)海洋學(xué)會(huì)2019海洋學(xué)術(shù)(國(guó)際)雙年會(huì). 三亞, 2019:31-45. HAO R X, ZHAO Y X, HE Z J, et al. Study on sea surface temperature prediction method based on EOF-NAR neural network hybrid model[C]//2019 Marine Academic (International) Biennial Meeting of China Oceanographic Society. Sanya, 2019:31-45.
[9] 王兆毅, 李云, 王旭. 中國(guó)近岸海域基礎(chǔ)預(yù)報(bào)單元海溫預(yù)報(bào)指導(dǎo)產(chǎn)品研制[J]. 海洋預(yù)報(bào), 2020, 37(4):59-65. WANG Z Y, LI Y, WANG X. Development of forecast guidance product for sea temperature of basic forecast units in the Chinese coastal waters[J]. Marine Forecasts, 2020, 37(4):59-65.
[10] 陳希, 沙文鈺, 李妍, 等. 人工神經(jīng)網(wǎng)絡(luò)技術(shù)在海浪預(yù)報(bào)中的應(yīng)用[J]. 海洋通報(bào), 2002, 21(2):11-15. CHEN X, SHA W Y, LI Y, et al. Application of the artificial neural network in the sea wave forecast[J]. Marine Science Bulletin, 2002, 21(2):11-15.
[11] 齊義泉, 張志旭, 李志偉, 等. 人工神經(jīng)網(wǎng)絡(luò)在海浪數(shù)值預(yù)報(bào)中的應(yīng)用[J]. 水科學(xué)進(jìn)展, 2005, 16(1):32-35. QI Y Q, ZHANG Z X, LI Z W, et al. Application of artificial neural network to numerical wave prediction[J]. Advances in Water Science, 2005, 16(1):32-35.
[12] 王建華, 于紅兵, 宋運(yùn)法. 人工神經(jīng)網(wǎng)絡(luò)在潮汐數(shù)值預(yù)報(bào)中的應(yīng)用[J]. 海洋預(yù)報(bào), 2007, 24(2):47-51. WANG J H, YU H B, SONG Y F. Application of artificial neural network to numerical tidal prediction[J]. Marine Forecasts, 2007, 24(2):47-51.
[13] PAVLYSHENKO B. Using stacking approaches for machine learning models[C]//2018 IEEE Second International Conference on Data Stream Mining & Processing (DSMP). Lviv, Ukraine:IEEE, 2018:255-258.
[14] KANAMITSU M, EBISUZAKI W, WOOLLEN J, et al. NCEPDOE AMIP-II reanalysis (R-2)[J]. Bulletin of the American Meteorological Society, 2002, 83(11):1631-1644.
[15] SAKURAI T, YUKIO K, KURAGANO T. Merged satellite and insitu data global daily SST[C]//Proceedings.2005 IEEE International Geoscience and Remote Sensing Symposium. Seoul, Korea (South):IEEE, 2005:2606-2608.
[16] GEURTS P, ERNST D, WEHENKEL L. Extremely randomized trees[J]. Machine Learning, 2006, 63(1):3-42.
[17] WOLPERT D H. Stacked generalization[J]. Neural Networks, 1992, 5(2):241-259.
[18] BREIMAN L. Stacked regressions[J]. Machine Learning, 1996, 24(1):49-64.

服務(wù)與反饋：

【文章下載】【發(fā)表評(píng)論】【查看評(píng)論】【加入收藏】