FORECASTS OF GLOBAL FPSO MARKET DEMANDS FOR 2025, 2030 AND 2035 BY A COMBINED METHOD BASED ON MULTI-ALGORITHM INTEGRATED ANALYSIS

doi:10.13776/j.cnki.resourcesindustries.20241029.001

Abstract

Abstract: This paper studies the market demands of Floating Production Storage and Offloading (FPSO), presents a way to forecast the global FPSO market demands. By means of a method integrated with entropy, artificial neural network, random forest regression, ADABOOST regression, and ARMA model, this paper forecasts the global FPSO orders with the results showing that FPSO needs 9 ships in 2025, 10 in 2030 and 12 in 2035, suggesting a stable rising global FPSO market demand in the future ten years. China's ship manufacturers should pay close attention on the impacts of crude oil price on FPSO market demands and use new technologies in reducing carbon emission and increasing market shares.

Key words: FPSO, market demands, multi-algorithm integration, combined forecast, crude oil price

摘要： 调研分析国际浮式生产储油卸油装置（Floating Production Storage and Offloading, FPSO）的市场需求现状，提出国际FPSO市场需求的预测方法。采用熵值法、人工神经网络法、随机森林回归、ADABOOST回归、ARMA模型等多算法融合的方法，对国际FPSO订单进行组合预测，结果表明国际FPSO市场需求在2025年、2030年、2035年的需求量分别为9、10与12艘，可见国际FPSO市场需求在未来十多年间稳步回升。我国船企应密切关注原油价格变化对FPSO市场需求的影响，积极采用新产品技术，同时减少碳排放，提高市场份额。

关键词: 浮式生产储油卸油装置, 市场需求, 多算法融合, 组合预测, 原油价格

CLC Number:

F551

GUO Jingyi, WANG Ling, FANG Yutong. FORECASTS OF GLOBAL FPSO MARKET DEMANDS FOR 2025, 2030 AND 2035 BY A COMBINED METHOD BASED ON MULTI-ALGORITHM INTEGRATED ANALYSIS[J]. Resources & Industries, 2024, 26(5): 198-206.

芦琳娜, 张瑞朋, 于强, 李松, 解永乐, 徐珊珊, 于银杰, 张秋阳. 2025、2030、2035年国际FPSO市场需求组合预测——基于多算法融合分析[J]. 资源与产业, 2024, 26(5): 198-206.

References

艾信, 田鹏, 吉效科, 等, 2021. 多算法融合管道泄漏检测预警系统试验研究[J]. 石油矿场机械, 50(6): 26-33.〔AI X, TIAN P, JI X K, et al.，2021. Experimental study on multi-algorithm fusion pipeline leakage detection and early warning system[J]. Oil Field Machinery, 50(6): 26-33.〕

白新建,梁姝婷,万莉, 2018. FPSO生活模块海工装备市场经济预测研究[J]. 中国修船, 31(6): 45-48.〔BAI X J, LIANG S T, WAN L, 2018. Research on market economy prediction of FPSO life module marine equipment[J]. China Shipbuilding, 31(6): 45-48.〕

班文静,姜强,赵蔚,2022. 基于多算法融合的在线学习成绩精准预测研究[J]. 现代远距离教育(03): 37-45.〔BAN W J, JIANG Q, ZHAO W, 2022. Accurate prediction of online academic performance based on multi-algorithm fusion[J]. Modern Distance Education(03): 37-45.〕

陈华友, 2008．组合预测方法有效性理论及其应用[M]. 北京: 科学出版社.〔CHEN H Y, 2008. Validity theory and application of combination forecasting methods[M]. Beijing: Science Press.〕

戴家刚, 2007. 住房公积金中沉淀资金预测模型的研究与实现[D]. 北京: 清华大学.〔DAI J G, 2007. Research and implementation of precipitation fund prediction model in housing provident fund[D]. Beijing: Tsinghua University.〕

杜庆贵,沈晓婵,檀国荣,等,2017. FPSO应用现状及发展趋势浅析[J]. 海洋工程装备与技术, 4(2): 63-68.〔DU Q G, SHEN X C, TAN G R, et al.，2017. Analysis of application status and development trend of FPSO[J]. Ocean engineering equipment and technology, 4(2): 63-68.〕

冯明刚, 严伟, 葛新民, 等, 2018. 利用随机森林回归算法预测总有机碳含量[J]. 矿物岩石地球化学通报, 37(03): 475-481.〔F ENG M G, YAN W, GE X M, et al.，2018. Total organic carbon content was predicted using random forest regression algorithm[J]. Geochemical Bulletin of Minerals and Rocks, 37(03): 475-481.〕

黄吉, 姜晓翔, 甘霏斐, 2021. FPSO国内外发展及市场展望[J]. 船舶工程, 43(12): 29-36.〔HUANG J, JIANG X X, GAN F F, 2021. FPSO development and market outlook at home and abroad[J]. Ship Engineering, 43(12): 29-36.〕

纪绘,李玥,王开翔,2023. 基于多算法融合与人文特征的兰州市PM_(2.5)等级预测方法[J]. 软件导刊, 22(8): 24-32.〔JI H, LI Y, WANG K X, 2023. PM _ (2.5) grade prediction method of Lanzhou city based on multi-algorithm fusion and humanistic characteristics[J]. Software Guide, 22(8): 24-32.〕

李红卫，1992. FPSO系统90年代的市场预测[J]. 中国海洋平台 (2): 90.〔LI H W, 1992. Market forecast of FPSO system in 1990s[J]. China Offshore Platform (2) : 90.〕

李宁, 2020. 基于BP神经网络和果蝇算法的FPSO船体梁结构参数优化研究[D]. 青岛: 中国石油大学(华东).〔LI N, 2020. Optimization of structural parameters of FPSO hull girder based on BP neural network and fruit fly algorithm[D]. Qingdao: China University of Petroleum (East China).〕

李伟, 严珂, 陆慧娟, 等, 2019. 基于Adaboost.RT算法的隧道沉降时间序列预测研究[J]. 中国计量大学学报, 30(3): 331-336.〔LI W, YAN K, LU H J, et al.，2019. Prediction of tunnel settlement time series based on Adaboost.RT algorithm[J]. Journal of China Jiliang University, 30(3): 331-336.〕

李文彬, 张春梅, 2017. 多算法融合的电网用电量预测系统研究和实现[J]. 现代计算机(专业版), 22: 75-78.〔LI W B, ZHANG C M, 2017. Research and implementation of power grid electricity consumption prediction system based on multi-algorithm fusion[J]. Modern Computer (Professional Edition), 22: 75-78.〕

芦琳娜, 胡平樱, 雷涯邻, 2016. 2020年、2025年、2030年我国镍资源供需量熵值组合预测[J]. 中国矿业, 25(6): 38-43.〔LU L N, HU P Y, LEI Y L, 2016. 2020,2025,2030 China’s nickel resources supply and demand entropy combination forecast[J]. China Mining, 25(6): 38-43.〕

罗宇卓, 马瑜, 王文娜, 等, 2018. 基于FPSO优化的BP神经网络算法及环境监测应用[J]. 国外电子测量技术, 37(3): 136-142.〔LUO Y Z, MA Y, WANG W N, et al.，2018. BP neural network algorithm based on FPSO optimization and environmental monitoring applications[J]. Foreign Electronic Measurement Technology, 37(3): 136-142.〕

吕佳朋, 史贤俊, 2020. 基于AdaBoost的GPR预测算法研究及应用[J]. 电光与控制, 27(6): 43-46, 62.〔LU J P, SHI X J, 2020. Research and application of GPR prediction algorithm based on AdaBoost[J]. Electro-optics and control, 27(6): 43-46, 62.〕

吕龙德, 2023. 需求增长FPSO掀制造小高潮[J]. 广东造船, 42(5): 15-16.〔LU L D, 2023. Demand growth FPSO tilting manufacturing small climax[J]. Guangdong Shipbuilding, 42 (5): 15-16.〕

孙仪阳, 2017. 基于熵值法和数学预测模型的土地生态安全动态评价[D]. 泰安: 山东农业大学.〔SUN Y Y, 2017. Dynamic evaluation of land ecological security based on entropy method and mathematical prediction model[D]. Taian: Shandong Agricultural University.〕

谭家翔, 2016. 深水FPSO发展现状与趋势[J]. 船海工程, 45(5): 65-69, 75.〔TAN J X, 2016. Development status and trends of deepwater FPSO[J]. Ship-Ocean Engineering, 45(5): 65-69,75.〕

王祺, 2022. 基于FPSO的深水油田开发模式最优化算法研究[D]. 北京: 中国石油大学(北京).〔WANG Q, 2022. Research on optimization algorithm of deepwater oilfield development mode based on FPSO[D]. Beijing: China University of Petroleum (Beijing).〕

王荣波, 王亚杰, 黄孝喜, 等, 2018. 基于多算法融合的移动通信客户流失预测模型[J]. 计算机技术与发展, 28(8): 152-155, 159.〔WANG R B, WANG Y J, HUANG X X, et al.，2018. Mobile communication customer churn prediction model based on multi-algorithm fusion[J]. Computer Technology and Development, 28(8): 152-155, 159.〕

王天佐, 王常明, 姚爱军, 等, 2016. 基于时间序列的地铁横通道拱顶沉降预测[J]. 现代隧道技术, 53(3): 74-81.〔WANG T Z, WANG C M, YAO A J, et al.，2016. Prediction of vault settlement of subway cross passage based on time series[J]. Modern Tunnel Technology, 53(3): 74-81.〕

吴潇雨, 和敬涵, 张沛, 等, 2015. 基于灰色投影改进随机森林算法的电力系统短期负荷预测[J]. 电力系统自动化, 39(12): 50-55.〔WU X Y, HE J H, ZHANG P, et al.，2015. Short-term load forecasting of power system based on grey projection improved random forest algorithm[J]. Power System Automation, 39(12): 50-55.〕

吴晓阳, 张森, 陈先中, 等, 2020. 高炉煤气流分布过程的多算法融合预测模型[J]. 控制理论与应用, 37(6): 1241-1252.〔WU X Y, ZHANG S, CHEN X Z, et al.，2020. Multi-algorithm fusion prediction model for blast furnace gas flow distribution process[J]. Control Theory and Application, 37(6): 1241-1252.〕

徐煜程, 2017. 基于ARMA模型对我国金融机构存款的分析[J]. 中国集体经济(8): 77-78.〔XU Y C, 2017. Analysis of deposits in China‘s financial institutions based on ARMA model[J]. China’s Collective Economy(8): 77-78.〕

薛瑾艳, 高薇, 王东岳, 2019. FPSO投资项目基准收益率的确定办法[J]. 工程建设与设计(22): 243-245.〔XUE J Y, GAO W, WANG D Y, 2019. Determination method of benchmark yield of FPSO investment project[J]. Engineering Construction and Design (22): 243-245.〕

姚立平,潘中良,2019. 一种多算法融合的人脸识别方法研究[J]. 光电子·激光, 30(9): 960-967.〔YAO L P, PAN Z L, 2019. Research on a multi-algorithm fusion face recognition method[J]. Optoelectronics·Laser, 30(9): 960-967.〕

殷欣, 高峰, 刘泉声, 等, 2022. 面向隧道掘进机可掘性评价的多算法融合优化模型及其工程应用[J]. 岩石力学与工程学报, 41(S1): 2757-2771.〔YIN X, GAO F, LIU Q S, et al.，2022. Multi-algorithm fusion optimization model and its engineering application for tunnel boring machine drivability evaluation[J]. Journal of Rock Mechanics and Engineering, 41(S1): 2757-2771.〕

袁捷, 唐龙, 杜浩, 2015. 机场道面使用性能的动态自回归预测模型[J]. 同济大学学报(自然科学版), 43(3): 399-404.〔YUAN J, TANG L, DU H, 2015. Dynamic autoregressive prediction model of airport pavement performance[J]. Journal of Tongji University (Natural Science Edition), 43(3): 399-404.〕

张晶, 2017. 基于AdaBoost回归树的多目标预测算法[J]. 计算机与现代化(9): 89-95, 105.〔ZHANG J, 2017. Multi-objective prediction algorithm based on AdaBoost regression tree[J]. Computer and Modernization (9): 89-95,105.〕

赵羿羽, 徐晓丽, 郎舒妍, 等, 2018. FPSO生活模块市场需求预测[J]. 中国海洋平台, 33(2): 1-4, 21.〔ZHAO Y Y, XU X L, LANG S Y, et al.，2018. FPSO life module market demand forecast[J]. China Offshore Platform, 33(2): 1-4, 21.〕

郑斌,孙洪霞,王维民, 2020. 基于随机森林回归的汽油研究法辛烷值预测[J]. 石油炼制与化工, 51(12): 69-75.〔ZHENG B, SUN H X, WANG W M, 2020. Random forest regression based octane number prediction of gasoline research method[J]. Petroleum Refining and Chemical Industry, 51(12): 69-75.〕

ANDERSON O D, 1977. The Box-Jenkins approach to time series analysis[J]. RAIRO-Operations Research, 11(1): 3-29.

BARTON C M, 2016. FPSO market shows signs of resiliency[EB/OL]. (2016-08-11)[2024-06-10].https://www.offshore.mag.com/production/article/16754991/fpso-market-shows-signs-of-resiliency.

BATES J M, GRANGER C W J, 1969. The combination of forecasts[J]. Oper Res Soc, 20: 451-468.

BOX G E P, PIERCE D A, 1970. Distribution of residual autocorrelations in autoregressive-integrated moving average time series models[J]. Journal of the American Statistical Association, 65(332): 1509-1526.

BREIMAN L, 2001. Random forests[J]. Machine Learning, 45: 5-32.

BUSARI G A, LIM D H, 2021. Crude oil price prediction: a comparison between AdaBoost-LSTM and AdaBoost-GRU for improving forecasting performance[J]. Computers & Chemical Engineering, 155: 107513.

CHOI B S, 2012. ARMA model identification[M]. Secaucus: Springer Science & Business Media.

DENTON J W, 1995. How good are neural networks for causal forecasting?[J]. The Journal of Business Forecasting, 14(2): 17.

DUGGAL A, MINNEBO J, 2020. The floating production, storage and offloading system-past, present and future[C]. Offshore Technology Conference. Houston: OTC-30514-MS.

FREUND Y, SCHAPIRE R E, 1997. A desicion-theoretic generalization of on-line learning and an application to boosting[J]. Journal of Computer and System Sciences, 55(1): 119-139

HENERY D, INGLIS R B, 1995.Prospects and challenges for the FPSO[C]. Offshore Technology Conference. Houston: OTC-7695-MS.

JIANG H, FAN Y, SUN H, et al.，2021. Multi-algorithm fusion pharmaceutical sales forecasting mode[C]. 2nd International Conference on Machine Learning and Computer Application. Shenyang: ICMLCA 2021: 1-5.

JIN Y, JANG B S, 2015. Probabilistic fire risk analysis and structural safety assessment of FPSO topside module[J]. Ocean Engineering, 104: 725-737.

KELKBOOM E J C, ZHOU X, BREEBAART J, et al.，2009. Multi-algorithm fusion with template protection[C]. 2009 IEEE 3rd International Conference on Biometrics: Theory, Applications, and Systems. Washington DC: IEEE: 1-8.

KULKARNI S S, Sanghai N P, Baishya C, et al.，2024. Random forest regression for radiation pattern prediction of planar metasurface reflector antenna[J]. AEU-International Journal of Electronics and Communications, 174: 155018.

LEE L, CHOWDHURY A, SHUBITA M, 2023. Impact of Paris Agreement on financing strategy: evidence from global FPSO industry[J]. Technological Forecasting Social Change,188: 122266.

LIONG C T, CHUA K H, KUMAR N, et al.，2024. Deterministic prediction of vessel motion in real-time using artificial neural network[J]. Ocean Engineering, 294: 116835.

MONTAGNON C E, 2021. Forecasting by splitting a time series using Singular Value Decomposition then using both ARMA and a Fokker Planck equation[J]. Physica A: Statistical Mechanics and its Applications, 567: 125708.

REDDY A P, VIJAYARAJAN V, 2020. Audio compression with multi-algorithm fusion and its impact in speech emotion recognition[J]. International Journal of Speech Technology, 23(2): 277-285.

ROJAS I, VALENZUELA O, ROJAS F, et al.，2008. Soft-computing techniques and ARMA model for time series prediction[J]. Neurocomputing, 71(4/5/6): 519-537.

SIEBEL N T, MAYBANK S, 2002. Fusion of multiple tracking algorithms for robust people tracking[C] European Conference on Computer Vision, 2353: 373-387.

TANG A, GONG P, LI J, et al.，2022. A state-of-charge estimation method based on multi-algorithm fusion[J]. World Electric Vehicle Journal,13(4):70.

VINOD J, SARKAR K B, 2021. Francis turbine electrohydraulic inlet guide vane control by artificial neural network 2 degree-of-freedom PID controller with actuator fault[J]. Proceedings of the Institution of Mechanical Engineers, Part I: Journal of Systems and Control Engineering, 235(8):1494-1509.

[1]	XIAO Jian-zhong, WANG Xiu-juan, CHEN Dan-yang. THE VOLATILITY RELATIONSHIP BETWEEN INTERNATIONAL NATURAL GAS IMPORT PRICES AND CRUDE OIL PRICES [J]. Resources & Industries, 2016, 18(2): 88-95.
[2]	HOU Ming-yang, YANG Guo-feng. EVOLUTION, INFLUENCE AND OUTLOOK OF AMERICAS TIGHT OIL [J]. Resources & Industries, 2015, 17(1): 11-17.
[3]	HOU Ming-yang，ZHOU Qing-fan. DEVELOPMENT OF TIGHT OIL AND ITS INFLUENCE ON INTERNATIONAL OIL PRICE [J]. Resources & Industries, 2014, 16(1): 23-28.
[4]	WEI Jian-guo,LEI Ya-lin,AN Hai-zhong. DYNAMIC FEATURES OF CHINAS CRUDE OIL PRICE BASED ON MULTISCALE ENTROPY [J]. Resources & Industries, 2012, 14(6): 31-39.
[5]	LU Lin-na, LEI Ya-lin, CHE Gang. COINTEGRATION OF GLOBAL OIL PRICE, ECONOMIC GROWTH AND CRUDE OIL CONSUMPTION CHANGE TO DOMESTIC OIL PRICE [J]. Resources & Industries, 2012, 14(3): 16-23.