优化的Gray Markov模型在埋地管道腐蚀速率预测中的应用
|
摘要
为提高埋地管道剩余寿命预测的精确度,对传统灰色马尔科夫预测模型进行优化。将管道腐蚀速率视为一个灰色系统,对灰色模型的原始数据光滑处理后建立等维新信息无偏灰色模型,预测腐蚀速率的宏观值。以优化的灰色模型预测值残差为基础结合马尔科夫链模型,进行二次平滑处理和白化系数寻优,得出残差修正值。最终,结合两种优化模型得出管道腐蚀速率的预测值。实例检验证明,该模型能有效克服系统长期动态预测上的不足,与传统灰色马尔科夫链预测模型相比,预测精度提高了40. 33%,预测结果与实测值有更高的拟合程度。
In order to improve the accuracy of the residual life prediction of buried pipelines, the traditional grey Markov prediction model was optimized. The corrosion rate of pipeline was regarded as a gray system, after smooth processing of the original data of the gray model, an unbiased gray model of iso-dimensional new information was established, and the macroscopic value of corrosion rate was predicted. Based on the residual of the predicted grey model predictive value, combined with the Markov chain model,the second smoothing process and whitening coefficient optimization were performed, and the residual correction value was obtained. Finally, a combination of two optimization models yielded a predicted value for the corrosion rate of the pipeline. The example test proved that the model could effectively overcome the shortcomings of long-term dynamic prediction of the system. Compared with the traditional grey Markov chain prediction model, the prediction accuracy was improved by 40. 33%, and the prediction result had a higher degree of fitting with the measured value.
In order to improve the accuracy of the residual life prediction of buried pipelines, the traditional grey Markov prediction model was optimized. The corrosion rate of pipeline was regarded as a gray system, after smooth processing of the original data of the gray model, an unbiased gray model of iso-dimensional new information was established, and the macroscopic value of corrosion rate was predicted. Based on the residual of the predicted grey model predictive value, combined with the Markov chain model,the second smoothing process and whitening coefficient optimization were performed, and the residual correction value was obtained. Finally, a combination of two optimization models yielded a predicted value for the corrosion rate of the pipeline. The example test proved that the model could effectively overcome the shortcomings of long-term dynamic prediction of the system. Compared with the traditional grey Markov chain prediction model, the prediction accuracy was improved by 40. 33%, and the prediction result had a higher degree of fitting with the measured value.
引文
[1]王如君,王天瑜.灰色-马尔科夫链模型在埋地油气管道腐蚀预测中的应用[J].中国安全生产科学技术,2015(4):102-106.
[2]张国帅.基于累积法的灰色马尔科夫预测模型及其应用[J].统计与决策,2015(24):157-158.
[3]文华.基于遗传算法的灰色马氏链模型预测油田产量[J].特种油气藏,2009,16(6):58-60.
[4] OSSAI C I, BOSWELL B, DAVIES I. Markov chain modelling for time evolution of internal pitting corrosion distribution of oil and gas pipelines[J]. Engineering Failure Analysis, 2016, 60:209-228.
[5] ZENG B. Forecasting the relation of supply and demand of natural gas in China during 2015-2020 using a novel grey model[J]. Journal of Intelligent&Fuzzy Systems,2017,32(1):141-155.
[6]张春晓,石晓磊.腐蚀条件下基于残差修正的飞机结构寿命灰色马氏链预测[J].腐蚀与防护,2016,37(3):230-235.
[7]周健,曹瑞霞,王兆卫.餐饮业短期客流量预测方法[J].同济大学学报(自然科学版),2014,42(3):493-498.
[8]张新生,赵梦旭.改进的Grey-Markov模型在油气管道腐蚀深度预测中的应用[J].地质科技情报,2017(6):286-291.
[9]汤国平,姜汉桥,丁帅伟,等.基于Gompertz的灰色模型预测油田产量[J].新疆石油地质,2013,34(4):462-464.
[10]童明余,周孝华,曾波.灰色NGM(1,1,k)模型背景值优化方法[J].控制与决策,2017,32(3):507-514.
[11] LIU S, HUANG S, BEIHAI X U, et al. Foundation pit settlement prediction based on unbiased greymarkov chain[J]. Journal of Geomatics,2015,40(4):10-13.
[12] LIN T,ZHANG N,UNIVERSITY F. Research on grey verhulst and equal dimension new information model in medium and long term load forecasting[J].Electrical Engineering,2017,118:969-984.
[13]俞树荣,韩竣羽,李淑欣.利用灰色马尔可夫模型预测腐蚀管道寿命[J].机械强度,2016,38,186(4):850-856.
[14]赵振武,麻建军.基于灰色马尔科夫模型的机场安检危险品数量预测[J].安全与环境学报,2017,17(1):51-53.
[15]黎锁平,刘坤会.平滑系数自适应的二次指数平滑模型及其应用[J].系统工程理论与实践,2004,24(2):95-99.
[16] WONG L Y,TOH M P,THAM L W. Projection ofpre-diabetes and diabetes population size in Singapore using a dynamic Markov model[J]. Journal of Diabetes.2016,9(1):65-75.
[17]高云龙,闫鹏.基于多种群粒子群算法和布谷鸟搜索的联合寻优算法[J].控制与决策,2016,31(4):601-608.
[18] CESAREND, CHAMORET D, DOMASZEWSKI M. A new hybrid PSO algorithm based on a stochastic Markov chain model[J]. Advances in Engineering Software,2015,90(C):127-137.
[19] DENG Z, KE Y, GONG H,et al. Land subsidence prediction in Beijing based on PS-InSAR technique and improved Grey-Markov model[J]. Giscience&Remote Sensing,2017(1):1-22.
[2]张国帅.基于累积法的灰色马尔科夫预测模型及其应用[J].统计与决策,2015(24):157-158.
[3]文华.基于遗传算法的灰色马氏链模型预测油田产量[J].特种油气藏,2009,16(6):58-60.
[4] OSSAI C I, BOSWELL B, DAVIES I. Markov chain modelling for time evolution of internal pitting corrosion distribution of oil and gas pipelines[J]. Engineering Failure Analysis, 2016, 60:209-228.
[5] ZENG B. Forecasting the relation of supply and demand of natural gas in China during 2015-2020 using a novel grey model[J]. Journal of Intelligent&Fuzzy Systems,2017,32(1):141-155.
[6]张春晓,石晓磊.腐蚀条件下基于残差修正的飞机结构寿命灰色马氏链预测[J].腐蚀与防护,2016,37(3):230-235.
[7]周健,曹瑞霞,王兆卫.餐饮业短期客流量预测方法[J].同济大学学报(自然科学版),2014,42(3):493-498.
[8]张新生,赵梦旭.改进的Grey-Markov模型在油气管道腐蚀深度预测中的应用[J].地质科技情报,2017(6):286-291.
[9]汤国平,姜汉桥,丁帅伟,等.基于Gompertz的灰色模型预测油田产量[J].新疆石油地质,2013,34(4):462-464.
[10]童明余,周孝华,曾波.灰色NGM(1,1,k)模型背景值优化方法[J].控制与决策,2017,32(3):507-514.
[11] LIU S, HUANG S, BEIHAI X U, et al. Foundation pit settlement prediction based on unbiased greymarkov chain[J]. Journal of Geomatics,2015,40(4):10-13.
[12] LIN T,ZHANG N,UNIVERSITY F. Research on grey verhulst and equal dimension new information model in medium and long term load forecasting[J].Electrical Engineering,2017,118:969-984.
[13]俞树荣,韩竣羽,李淑欣.利用灰色马尔可夫模型预测腐蚀管道寿命[J].机械强度,2016,38,186(4):850-856.
[14]赵振武,麻建军.基于灰色马尔科夫模型的机场安检危险品数量预测[J].安全与环境学报,2017,17(1):51-53.
[15]黎锁平,刘坤会.平滑系数自适应的二次指数平滑模型及其应用[J].系统工程理论与实践,2004,24(2):95-99.
[16] WONG L Y,TOH M P,THAM L W. Projection ofpre-diabetes and diabetes population size in Singapore using a dynamic Markov model[J]. Journal of Diabetes.2016,9(1):65-75.
[17]高云龙,闫鹏.基于多种群粒子群算法和布谷鸟搜索的联合寻优算法[J].控制与决策,2016,31(4):601-608.
[18] CESAREND, CHAMORET D, DOMASZEWSKI M. A new hybrid PSO algorithm based on a stochastic Markov chain model[J]. Advances in Engineering Software,2015,90(C):127-137.
[19] DENG Z, KE Y, GONG H,et al. Land subsidence prediction in Beijing based on PS-InSAR technique and improved Grey-Markov model[J]. Giscience&Remote Sensing,2017(1):1-22.