Soil corrosion depth prediction of buried pipelines based on KPCA-ICS-ELM algorithm

doi:10.16265/j.cnki.issn 1003-3033.2020.12.014

Abstract

Abstract: In order to study soil corrosion situation of oil and gas pipelines and to ensure safe operation of buried pipelines, best corrosion factors were selected and high contribution factors of soil corrosion were obtained by using KPCA. Then, ICS was adopted to optimize deviation and threshold of ELM hidden layer, and with high contribution factors as input and corrosion depth as output target, ICS-ELM corrosion depth prediction model was established. Finally, taking field test of a buried oil pipeline in Shaanxi Province as an example, 8 influencing factors were selected to establish a soil corrosion index system of buried oil pipelines, and prediction and verification analysis were carried out through simulation. The results show that improved CS algorithm has faster iteration rate, and the lowest relative error, root mean square error and hill inequality coefficient of prediction results of KPCA-ICS-ELM model are 0.209%, 0.228% and 0.441% respectively. It is more stable and accurate compared with other models.

Key words: kernel principal component analysis (KPCA), improved cuckoo search (ICS), extreme learning machine (ELM), buried oil pipeline, corrosion depth

CLC Number:

X935

LI Yian, LUO Zhengshan. Soil corrosion depth prediction of buried pipelines based on KPCA-ICS-ELM algorithm[J]. China Safety Science Journal, 2020, 30(12): 100-105.

References

[1] 张新生,吕品品.考虑随机效应的腐蚀管道贝叶斯退化分析[J].中国安全科学学报,2019,29(8): 73-80.
ZHANG Xinsheng, LYU Pinpin.Bayesian degradation analysis of corroded pipeline considering random effect[J].China SafetyScience Journal,2019,29(8): 73-80.
[2] 毕傲睿, 骆正山, 王小完, 等. 基于土壤腐蚀主成分的金属管道退化维纳过程研究[J]. 材料保护, 2018, 51(1): 37-42.
BI Aorui, LUO Zhengshan, WANG Xiaowan, et al.Wiener process of degradation of metal pipelines based on principalcomponents of soil corrosion[J]. Materials protection, 2018, 51(1): 37-42.
[3] SÉRGIO B D C. A review of quantitative risk assessment of onshore pipelines[J]. Journal of Loss Prevention in the Process Industries, 2016, 44: 282-298.
[4] 罗小虎.基于IAHP和模糊灰色理论的土壤腐蚀等级评价[J].材料保护,2019,52(7): 54-59.
LUO Xiaohu.Evaluation of soil corrosion grade based on IAHP and Fuzzy Grey Theory[J].Materials Protection,2019,50(7):54-59.
[5] CALEYO F, ALMA V, LESTER A, et al. Bayesian analysis of external corrosion data of non-piggable underground pipelines[J]. Corrosion Science, 2015, 90: 33-45.
[6] EHSAN A, ROUZBEH A, VIKRAM G, et al. Developing a dynamic model for pitting and corrosion-fatigue damage of subsea pipelines[J]. Ocean Engineering, 2018,150: 391-396.
[7] 王齐, 胡林林. 基于BP神经网络的油气长输管道土壤腐蚀性预测 [J].当代化工, 2016, 45(9): 2 198-2 200.
WANG Qi, HU Linlin. Soil corrosion prediction of oil and gas pipeline based on BP neural network[J]. Contemporary Chemical Industry, 2016, 45(9): 2 198-2 200.
[8] 凌晓,徐鲁帅,梁瑞,等.基于改进PSO-BPNN的输油管道内腐蚀速率研究[J].中国安全生产科学技术,2019,15(10): 63-68.
LING Xiao,XU Lushuai,LIANG Rui,et al.Study on internal corrosion rate of oil pipeline based on improved PSO-BPNN[J]. Journal of Safety Science and Technology,2019,15(10): 63-68.
[9] 邵良杉,徐波.岩溶塌陷倾向性等级的KPCA-SVM预测模型[J].中国安全科学学报,2015,25(3): 60-65.
SHAO Liangshan, XU Bo.KPCA-SVM model for predicting karst collapse tendency level[J].China Safety Science Journal,2015,25(3): 60-65.
[10] 孙艳丽,杨娜,张正涛,等.基于核主元分析和支持向量机的结构损伤识别研究[J].应用基础与工程科学学报,2018,26(4): 888-900.
SUN Yanli,YANG Na,ZHANG Zhengtao,et al.Structural damage identification study based on kernel component analysisand support vector machine[J].Journal of Basic Science and Engineering,2018,26(4): 888-900.
[11] HUANG Guangbin, ZHU Qinyu, SIEW C K S. Extreme learning machine: theory and applications [J]. Neurocomputing, 2006, 70: 489-501.
[12] YANG Xinshe, DEB S. Engineering optimisation by cuckoo search[J].International Journal of Mathematical Modelling & Numerical Optimisation, 2010, 1 (4): 330-343.
[13] 高策,沈晓卫,章彪,等.改进布谷鸟搜索算法优化支持向量机的MEMS陀螺温度零偏补偿[J].宇航学报,2019,40(7): 811-817.
GAO Ce,SHEN Xiaowei,ZHANG Biao,et al.Temperature compensation of MEMS-Gyro based on improving cuckoo searchand support vector machines[J].Journal of Astronautics,2019,40(7):811-817.
[14] 高阳, 张碧玲, 毛京丽,等. 基于机器学习的自适应光伏超短期出力预测模型[J]. 电网技术, 2015, 39(2): 307-311.
GAO Yang,ZHANG Biling,MAO Jingli,et al.Machine learning-based adaptive very-short-term forecast model forphotovoltaic powe[J].Power System Technolog,2015,39(2): 307-311.
[15] 王文辉,骆正山,张新生.基于PSO-GRNN模型的埋地管道腐蚀剩余寿命预测[J].表面技术,2019,48(10):267-275,284.
WANG Wenhui,LUO Zhengshan,ZHANG Xinsheng.Prediction on remaining service life of buried pipeline after corrosionbased on PSO-GRNN model[J].Surface Technology,2019,48(10):267-275,284.