Prediction of tubular corrosion rate under multiphase dynamics condition based on wavelet neural network

doi:10.16085/j.issn.1000-6613.2017-1724

Abstract

Abstract: Wavelet neural network model was obtained to predict tubular corrosion rate for oil-gas gathering and transferring under multiphase dynamics condition. The laboratory multiphase dynamics corrosion experiment was adopted firstly in order to gain coupon corrosion rate in conditions of various operating situation for learning and testing of wavelet neural network prediction model. Then, a multi-factors analysis of variance was used to research the influence on corrosion rate of temperature, pressure, velocity, hydrogen sulfide content, carbon dioxide content, dissolved oxygen content, moisture content, salt content and pH to fulfill factor validity filtration. Finally, the suitable wavelet neural network prediction model was established by the way of learning and testing in the basis of recognition of neuron number of hidden layer. The model reliability was further verified. The result showed that above factors except pressure had great impact on corrosion rate and were considered as effective input signals. 8-17-1 type of wavelet neural network structure exhibited the favorable accuracy and stability when neuron number hidden layer was 17. Levenberg Marquardt optimization algorithm was chosen to train model repeatedly until its root mean square error less than convergence tolerance 0.001. The predicted value was approximately linear with the experimental value. The determination coefficient of learning stage and testing stage was 0.9992 and 0.9967, respectively, and demonstrated the superior correlation. There was also no significant difference between model predicted value and verified one. Therefore, the prediction model of wavelet neural network possessed well capacity of tubular corrosion rate for oil-gas gathering and transferring under multiphase dynamics condition.

Key words: neural network, factor validity, structure, corrosion, prediction

摘要： 利用小波神经网络模型预测多相动态环境下油气集输管道腐蚀速率。首先通过室内多相动态腐蚀实验，获得了不同工况条件下的挂片腐蚀速率，用于训练和检验小波神经网络预测模型，然后利用多因子方差分析研究了温度、压力、流率、硫化氢含量、二氧化碳含量、溶解氧含量、含水率、盐含量和pH对腐蚀速率的影响程度，实现了各因素的有效性筛选，最后在确定隐含层节点数基础上通过训练、测试建立起适宜的小波神经网络预测模型，并进一步验证了模型可靠性。结果表明：除了压力外，各因素对腐蚀速率均有十分显著的影响，属于有效输入信号。当隐含层节点数为17时，8-17-1型小波神经网络结构表现出良好的准确性和稳定性。利用Levenberg Marquardt优化算法对模型进行了反复训练，直至其均方根误差小于容许收敛误差限0.001，预测值与实际值近似呈线性关系，训练、测试阶段决定系数分别为0.9992、0.9967，相关性较高，模型预测值和验证值亦不存在显著差异。因此小波神经网络预测模型对多相动态环境下油气集输管道腐蚀速率具有良好的预测能力。

关键词: 神经网络, 因素有效性, 结构, 腐蚀, 预测

CLC Number:

TG172.8

FAN Zheng, FU Wenyao, ZHAO Xiaonan, LI Yazhou, LI Wenhong, MAO Zhenxing. Prediction of tubular corrosion rate under multiphase dynamics condition based on wavelet neural network[J]. Chemical Industry and Engineering Progress, 2018, 37(08): 2904-2911.

范峥, 付文耀, 赵笑男, 李亚洲, 李稳宏, 毛振兴. 基于小波神经网络预测多相动态管道腐蚀速率[J]. 化工进展, 2018, 37(08): 2904-2911.

References

[1] 王丹, 袁世娇, 吴小卫, 等.油气管道CO₂/H₂S腐蚀及防护技术研究进展[J].表面技术, 2016, 45(3):31-37. WANG Dan, YUAN Shijiao, WU Xiaowei, et al.Research progress of CO₂/H₂S corrosion in oil and gas pipelines and the protection techniques[J]. Surface Technology, 2016, 45(3):31-37.
[2] 连艺秀, 冯春艳, 刘文博, 等.油气管道的腐蚀与防护技术[J].油气田地面工程, 2014, 33(10):104-105. LIAN Yixiu, FENG Chunyan, LIU Wenbo, et al.Corrosion and protection techniques of oil and gas pipelines[J]. Oil-gas Field Surface Engineering, 2014, 33(10):104-105.
[3] 亓树成, 曾丽华, 石剑英, 等.腐蚀控制与防护技术在新疆油田的应用[J].化工进展, 2014, 33(5):1351-1355. QI Shucheng, ZENG Lihua, SHI Jianying, et al.Application of corrosion control and protection techniques in Xinjiang Oilfield[J]. Chemical Industry and Engineering Progress, 2014, 33(5):1351-1355.
[4] 付柯, 谢良才, 闫雨瑗, 等.改进BP神经网络预测Ni/Al₂O₃催化CH₄-CO₂重整反应[J]. 化工进展, 2017, 36(7):2393-2399. FU Ke, XIE Liangcai, YAN Yuyuan, et al.Predicting model of CH₄-CO₂ reforming on Ni/Al₂O₃ catalyst by improved back propagation(BP) neural network[J]. Chemical Industry and Engineering Progress, 2017, 36(7):2393-2399.
[5] 宋泓阳, 孙晓岩, 项曙光.人工神经网络在化工过程中的应用进展[J]. 化工进展, 2016, 35(12):3755-3762. SONG Hongyang, SUN Xiaoyan, XIANG Shuguang.Progress on the application of artificial neural network in chemical industry[J]. Chemical Industry and Engineering Progress, 2016, 35(12):3755-3762.
[6] 李君, 陈佳文, 廖伟丽, 等.基于小波神经网络的轴流泵性能预测[J].农业工程学报, 2016, 32(10):47-53. LI Jun, CHEN Jiawen, LIAO Weili, et al. Performance prediction of axial pump based on wavelet neural network[J]. Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(10):47-53.
[7] 江婷, 沈振中, 徐力群, 等.基于支持向量机-小波神经网络的边坡位移时序预测模型[J].武汉大学学报(工学版), 2017, 50(2):174-181. JIANG Ting, SHEN Zhenzhong, XU Liqun, et al.Time series prediction model of slope displacement based on support vector machines-wavelet neural network[J].Engineering Journal of Wuhan University, 2017, 50(2):174-181.
[8] NASSER S, GHOLAM H V, HAJIR K, et al.Prediction of the overall sieve tray efficiency for a group of hydrocarbons, an artificial neural network approach[J].Journal of Natural Gas Science and Engineering, 2011, 3(1):319-325.
[9] YANG H J, HU X.Wavelet neural network with improved genetic algorithm for traffic flow time series prediction[J].International Journal for Light and Electron Optics, 2016, 127(19):8103-8110.
[10] QI X N, LIU Y Q, GUO Q J, et al.Performance prediction of a shower cooling tower using wavelet neural network[J].Applied Thermal Engineering, 2016, 108(9):475-485.
[11] 王守琰, 宋诗哲. 小波分析和神经网络研究纯锌大气腐蚀[J]. 中国腐蚀与防护学报, 2005, 25(5):257-261. WANG Shouyan, SONG Shizhe.Predictive model based on wavelet transformation and artificial neural network for zinc atmospheric corrosion[J]. Journal of Chinese Society for Corrosion and Protection, 2005, 25(5):257-261.
[12] VASILIOS P A, ANTONIOS K A, IOANNIS F G, et al.Wavelet neural network methodology for ground resistance forecasting[J]. Electric Power Systems Research, 2016, 214(19):935-943.
[13] 周潇, 杨功流, 蔡庆中.基于小波神经网络的高精度惯导重力扰动补偿方法[J].中国惯性技术学报, 2016, 24(5):571-576. ZHOU Xiao, YANG Gongliu, CAI Qingzhong.Compensation on gravity disturbance for high-precision INS based on wavelet neural network[J]. Journal of Chinese Inertial Technology, 2016, 24(5):571-576.
[14] 成谢锋, 傅女婷, 陈胤, 等.一种心音小波神经网络识别系统[J].振动与冲击, 2017, 36(3):1-6. CHENG Xiefeng, FU Nüting, CHEN Yin, et al.A recognition system for a heart sound wavelet neural network[J].Journal of Vibration and Shock, 2017, 36(3):1-6.
[15] 钟诗胜, 李洋.基于小波过程神经网络的飞机发动机状态监视[J].航空学报, 2007, 28(1):68-71. ZHONG Shisheng, LI Yang.Condition monitoring of aeroengine based on wavelet process neural networks[J].Acta Aeronautica Et Astronautica Sinica, 2007, 28(1):68-71.
[16] 米洋, 陈家斌, 刘红光, 等.一种采用小波神经网络的GPS精密单点定位方法[J].中国惯性技术学报, 2016, 24(3):337-341. MI Yang, CHEN Jiabin, LIU Hongguang, et al.GPS precise point positioning method using wavelet neural network[J].Journal of Chinese Inertial Technology, 2016, 24(3):337-341.
[17] 刘阳, 赵晖.基于小波神经网络的城市交通出行方式交互预测[J].长安大学学报(自然科学版), 2015, 35(1):41-48. LIU Yang, ZHAO Hui.Interactive prediction for traffic modal split based on wavelet neural networks[J]. Journal of Chang' an University (Natural Science Edition), 2015, 35(1):41-48.
[18] 刘伟韬, 穆殿瑞, 杨利, 等.倾斜煤层底板破坏深度计算方法及主控因素敏感性分析[J].煤炭学报, 2017, 42(4):849-859. LIU Weitao, MU Dianrui, YANG Li, et al.Calculation method and main factor sensitivity analysis of inclined coal floor damage depth[J]. Journal of China Coal Society, 2017, 42(4):849-859.
[19] 陈美舟, 孙雪峰, 程修沛, 等.卧辊式摘穗机构摘穗辊高度差对玉米籽粒损失的影响[J].农业工程学报, 2017, 33(4):63-68. LIU Meizhou, SUN Xuefeng, CHENG Xiupei, et al.Effect of height difference of horizontal corn snapping rollers device on grain loss[J]. Transactions of the Chinese Society of Agricultural Engineering, 2017, 33(4):63-68.
[20] MOHAMMAD M G, ALIZERA B, SOHRAB Z.Estimation of triethylene glycol(TEG) purity in natural gas dehydration units using fuzzy neural network[J]. Journal of Natural Gas Science and Engineering, 2014, 17(3):26-32.
[21] 翁永基.腐蚀预测和计量学基础-从试验到数据分析, 建模与预测[M].北京:石油工业出版社, 2011. WENG Yongji.Corrosion prediction and basic chemometrics-from experiments to data analysis, modeling and prediction[M].Beijing:Petroleum Industry Press, 2011.