陈亮等:低合金钢海水腐蚀监测中的双率数据处理与建模 ·103· 建立的CIV-ANN和CIV-SVR进行对比.实验表 Sanya seawater environment.J Mater Sci Technol,2021,64:222 明,CIV-IRVR在MAE和RMSE两项误差指标上 [15]Liu x Q,Tang X,Wang J.Correlation between seawater environmental factors and marine corrosion rate using artificial 达到了最低,在CD指标上达到了最高,获得了最 neural network analysis.JChin Soc Corros Prot,2005,25(1):11 佳的预测效果 (刘学庆,唐晓,王佳.金属腐蚀速度与海水环境参数相关模型 (3)CIV-IRVR在保留更多数据信息的同时还 的人工神经网络分析.中国腐蚀与防护学报,2005,25(1):11) 具备较高的预测精度,能够很好地解决海水腐蚀 [16]Shirazi A Z,Mohammadi Z.A hybrid intelligent model combining 中双率数据的处理和建模问题,对于材料腐蚀状 ANN and imperialist competitive algorithm for prediction of 态的预测及进一步指导腐蚀防护工作有一定的参 corrosion rate in 3C steel under seawater environment.Neural Comput4ppl,2017,28(11):3455 考价值和现实意义 [17]Wen Y F,Cai C Z,Liu X H,et al.Corrosion rate prediction of 3C 参考文献 steel under different seawater environment by using support vector regression.Corros Sci,2009,51(2):349 [1]Li X G,Zhang D W,Liu Z Y,et al.Materials science:Share [18]Bi A R,Luo Z S,Qiao W,et al.Prediction of pipeline inner- corrosion data.Nature,2015,527(7579):441 corrosion based on principal component analysis and particle [2]LiZL,Fu D M,Li Y,et al.Application of an electrical resistance swarm optimization-support vector machine.Surf Technol,2018, sensor-based automated corrosion monitor in the study of 47(9):133 atmospheric corrosion.Materials,2019,12(7):1065 (毕做容,骆正山,乔伟,等.基于主成分和粒子群优化支持向量 [3]Chen M D.Zhang H J.Chen L,et al.An electrochemical method 机的管道内腐蚀预测.表面技术,2018,47(9):133) based on OCP fluctuations for anti-corrosion alloy composition [19]Li M X,Zhang H C.Qiu P Y,et al.Predicting future locations optimization.Anti-Corros Methods Mater,2018,65(3):325 with deep fuzzy-LSTM network.Acta Geodaet Cartograph Sin, [4] Pei Z B,Cheng X Q,Yang X J,et al.Understanding 2018.47(12):1660 environmental impacts on initial atmospheric corrosion based on (李明晓,张恒才,仇培元,等.一种基于模糊长短期神经网络的 corrosion monitoring sensors.J Mater Sci Technol,2021,64:214 移动对象轨迹预测算法.测绘学报,2018,47(12):1660) [5]Ni B Y,Xiao D Y.A survey on identification of multirate sampled [20]Chen M D,Zhang F,Liu Z Y,et al.Galvanic series of metals and systems.Control Theory Appl,2009,26(1):62 effect of alloy compositions on corrosion resistance in Sanya (倪博溢,萧德云,多采样率系统的辨识问题综述.控制理论与 seawater.Acta Metall Sin,2018,54(9):1311 应用,2009,26(1):62) (陈闽东,张帆,刘智勇,等.金属材料在三亚海水中的腐蚀电位 [6]Kranc G.Input-output analysis of multirate feedback systems./RE 序及合金成分对耐蚀性的影响.金属学报,2018,54(9):1311) Trans Autom Control,1957,3(1):21 [21]Jiang X Y,Li S.BAS:beetle antennae search algorithm for [7]Friedland B.Sampled-data control systems containing periodically optimization problems.Int J Rob Control,2018,1(1):1 varying members.IFAC Proc Vol,1960,1(1):371 [22]Zhao Y Q,Qian Q,Zhou T J,et al.Hybrid optimization algorithm [8]Khargonekar P,Poolla K,Tannenbaum A.Robust control of linear based on beetle antennae search and genetic evolution.J Chin time-invariant plants using periodic compensation.IEEE Trans Comput Sy3L,2020,41(7:1438 Autom Control,.1985,30(11):1088 (赵玉强,钱谦,周田江,等.天牛须搜索与遗传的混合算法.小 [9] Ni B Y,Xiao D Y.A recursive identification method for non- 型微型计算机系统,2020,41(7):1438) uniformly sampled systems.Acta Autom Sin,2009,35(12):1520 [23]Tipping M E.Sparse Bayesian leaming and the relevance vector (倪博溢,萧德云.非均匀采样系统的一种递推辨识方法.自动 machine.J Mach Learn Res,2001,1:211 化学报,2009,35(12):1520) [24]Wu X T.Prediction Model of Solar Irradiance Based on [10]Li D G,Shah S L,Chen T W.Analysis of dual-rate inferential Variational Mode Decomposition and Relevance Vector Machine control systems.Automatica,2002,38(6):1053 Dissertation].Wuhan:Huazhong University of Science and [11]Ding F,Chen T W.Modeling and identification for multirate Technology,2018 systems.Acta Autom Sin,2005,31(1):105 (吴小涛.基于变分模态分解和相关向量机的太阳辐照度预测 [12]Zhi Y J.Fu D M,Zhang D W,et al.Prediction and knowledge 模型研究[学位论文].武汉:华中科技大学,2018) mining of outdoor atmospheric corrosion rates of low alloy steels [25]Duan Q,Zhao J G,Ma Y.Relevance vector machine based on based on the random forests approach.Metals,2019,9(3)383 particle swarm optimization of compounding kemels in elextricity [13]Shi Y N,Fu D M,Zhi Y J,et al.Improved ANFIS-based load forecasting.Electr Mach Control,2010,14(6):33 imputation method for missing data on atmospheric corrosion (段青,赵建国,马艳.优化组合核函数相关向量机电力负荷预 environment.Equip Environ Eng,2016,13(6):78 测模型.电机与控制学报,2010,14(6):33) (石雅楠,付冬梅,支元杰,等.基于ANFIS改进的大气腐蚀环境 [26]Wu J,Cheng HC,Liu Y Q,et al.Leaming soft sensors using time 缺失数据填补方法.装备环境工程,2016,13(6):78) difference-based multi-kemnel relevance vector machine with [14]Wei X,Fu D M,Chen M D,et al.Data mining to effect of key applications for quality-relevant monitoring in wastewater alloying elements on corrosion resistance of low alloy steels in treatment.Environ Sci Pollut Res.2020.27(23):28986建立的 CIV−ANN 和 CIV−SVR 进行对比. 实验表 明 ,CIV−IRVR 在 MAE 和 RMSE 两项误差指标上 达到了最低,在 CD 指标上达到了最高,获得了最 佳的预测效果. (3)CIV−IRVR 在保留更多数据信息的同时还 具备较高的预测精度,能够很好地解决海水腐蚀 中双率数据的处理和建模问题,对于材料腐蚀状 态的预测及进一步指导腐蚀防护工作有一定的参 考价值和现实意义. 参 考 文 献 Li X G, Zhang D W, Liu Z Y, et al. Materials science: Share corrosion data. Nature, 2015, 527(7579): 441 [1] Li Z L, Fu D M, Li Y, et al. Application of an electrical resistance sensor-based automated corrosion monitor in the study of atmospheric corrosion. Materials, 2019, 12(7): 1065 [2] Chen M D, Zhang H J, Chen L, et al. An electrochemical method based on OCP fluctuations for anti-corrosion alloy composition optimization. Anti-Corros Methods Mater, 2018, 65(3): 325 [3] Pei Z B, Cheng X Q, Yang X J, et al. Understanding environmental impacts on initial atmospheric corrosion based on corrosion monitoring sensors. J Mater Sci Technol, 2021, 64: 214 [4] Ni B Y, Xiao D Y. A survey on identification of multirate sampled systems. Control Theory Appl, 2009, 26(1): 62 (倪博溢, 萧德云. 多采样率系统的辨识问题综述. 控制理论与 应用, 2009, 26(1):62) [5] Kranc G. Input-output analysis of multirate feedback systems. IRE Trans Autom Control, 1957, 3(1): 21 [6] Friedland B. Sampled-data control systems containing periodically varying members. IFAC Proc Vol, 1960, 1(1): 371 [7] Khargonekar P, Poolla K, Tannenbaum A. Robust control of linear time-invariant plants using periodic compensation. IEEE Trans Autom Control, 1985, 30(11): 1088 [8] Ni B Y, Xiao D Y. A recursive identification method for nonuniformly sampled systems. Acta Autom Sin, 2009, 35(12): 1520 (倪博溢, 萧德云. 非均匀采样系统的一种递推辨识方法. 自动 化学报, 2009, 35(12):1520) [9] Li D G, Shah S L, Chen T W. Analysis of dual-rate inferential control systems. Automatica, 2002, 38(6): 1053 [10] Ding F, Chen T W. Modeling and identification for multirate systems. Acta Autom Sin, 2005, 31(1): 105 [11] Zhi Y J, Fu D M, Zhang D W, et al. Prediction and knowledge mining of outdoor atmospheric corrosion rates of low alloy steels based on the random forests approach. Metals, 2019, 9(3): 383 [12] Shi Y N, Fu D M, Zhi Y J, et al. Improved ANFIS-based imputation method for missing data on atmospheric corrosion environment. Equip Environ Eng, 2016, 13(6): 78 (石雅楠, 付冬梅, 支元杰, 等. 基于 ANFIS 改进的大气腐蚀环境 缺失数据填补方法. 装备环境工程, 2016, 13(6):78) [13] Wei X, Fu D M, Chen M D, et al. Data mining to effect of key alloying elements on corrosion resistance of low alloy steels in [14] Sanya seawater environment. J Mater Sci Technol, 2021, 64: 222 Liu X Q, Tang X, Wang J. Correlation between seawater environmental factors and marine corrosion rate using artificial neural network analysis. J Chin Soc Corros Prot, 2005, 25(1): 11 (刘学庆, 唐晓, 王佳. 金属腐蚀速度与海水环境参数相关模型 的人工神经网络分析. 中国腐蚀与防护学报, 2005, 25(1):11) [15] Shirazi A Z, Mohammadi Z. A hybrid intelligent model combining ANN and imperialist competitive algorithm for prediction of corrosion rate in 3C steel under seawater environment. Neural Comput Appl, 2017, 28(11): 3455 [16] Wen Y F, Cai C Z, Liu X H, et al. Corrosion rate prediction of 3C steel under different seawater environment by using support vector regression. Corros Sci, 2009, 51(2): 349 [17] Bi A R, Luo Z S, Qiao W, et al. Prediction of pipeline innercorrosion based on principal component analysis and particle swarm optimization-support vector machine. Surf Technol, 2018, 47(9): 133 (毕傲睿, 骆正山, 乔伟, 等. 基于主成分和粒子群优化支持向量 机的管道内腐蚀预测. 表面技术, 2018, 47(9):133) [18] Li M X, Zhang H C, Qiu P Y, et al. Predicting future locations with deep fuzzy-LSTM network. Acta Geodaet Cartograph Sin, 2018, 47(12): 1660 (李明晓, 张恒才, 仇培元, 等. 一种基于模糊长短期神经网络的 移动对象轨迹预测算法. 测绘学报, 2018, 47(12):1660) [19] Chen M D, Zhang F, Liu Z Y, et al. Galvanic series of metals and effect of alloy compositions on corrosion resistance in Sanya seawater. Acta Metall Sin, 2018, 54(9): 1311 (陈闽东, 张帆, 刘智勇, 等. 金属材料在三亚海水中的腐蚀电位 序及合金成分对耐蚀性的影响. 金属学报, 2018, 54(9):1311) [20] Jiang X Y, Li S. BAS: beetle antennae search algorithm for optimization problems. Int J Rob Control, 2018, 1(1): 1 [21] Zhao Y Q, Qian Q, Zhou T J, et al. Hybrid optimization algorithm based on beetle antennae search and genetic evolution. J Chin Comput Syst, 2020, 41(7): 1438 (赵玉强, 钱谦, 周田江, 等. 天牛须搜索与遗传的混合算法. 小 型微型计算机系统, 2020, 41(7):1438) [22] Tipping M E. Sparse Bayesian learning and the relevance vector machine. J Mach Learn Res, 2001, 1: 211 [23] Wu X T. Prediction Model of Solar Irradiance Based on Variational Mode Decomposition and Relevance Vector Machine [Dissertation]. Wuhan: Huazhong University of Science and Technology, 2018 ( 吴小涛. 基于变分模态分解和相关向量机的太阳辐照度预测 模型研究[学位论文]. 武汉: 华中科技大学, 2018) [24] Duan Q, Zhao J G, Ma Y. Relevance vector machine based on particle swarm optimization of compounding kernels in elextricity load forecasting. Electr Mach Control, 2010, 14(6): 33 (段青, 赵建国, 马艳. 优化组合核函数相关向量机电力负荷预 测模型. 电机与控制学报, 2010, 14(6):33) [25] Wu J, Cheng H C, Liu Y Q, et al. Learning soft sensors using time difference-based multi-kernel relevance vector machine with applications for quality-relevant monitoring in wastewater treatment. Environ Sci Pollut Res, 2020, 27(23): 28986 [26] 陈 亮等: 低合金钢海水腐蚀监测中的双率数据处理与建模 · 103 ·