基于改进灰色模型的管壁蜡沉积厚度预测研究  

Research on prediction of wax deposition thickness on pipe wall based on improved grey model

作　　者：徐士祺[1,2] 殷彦龙 林清军 熊小伟 朱广虎 XU Shiqi;YIN Yanlong;LIN Qingjun;XIONG Xiaowei;ZHU Guanghu(College of Petroleum Engineering,Xi’an Shiyou University,Xi’an Shaanxi 710065,China;Shaanxi Key Laboratory of Advanced Stimulation Technology for Oil&Gas Reservoirs,Xi’an Shaanxi 710065,China;Oil Production Plant No.8 of Changqing Oilfield Company CNPC,Xi’an Shaanxi 710016,China;Oil Production Plant No.3 of Changqing Oilfield Company CNPC,Yinchuan Ningxia 750000,China;Shaanxi Xianyang Petroleum Branch of Sinopec Sales Co.,Ltd.,Xianyang Shaanxi 237000,China)

机构地区：[1]西安石油大学石油工程学院,陕西西安710065 [2]西安石油大学陕西省油气田特种增产技术重点实验室,陕西西安710065 [3]长庆油田分公司第八采油厂,陕西西安710016 [4]中国石油长庆油田分公司第三采油厂,宁夏银川750000 [5]中国石化销售股份有限公司陕西咸阳石油分公司,陕西咸阳237000

出　　处：《中国安全生产科学技术》2021年第10期32-38,共7页Journal of Safety Science and Technology

摘　　要：为对含蜡原油管道中的蜡沉积厚度进行准确预测,在函数cot(x^(2))变换的基础上,结合平移变换思想,利用cot(x^(2)+c)变换建立新的改进GM(1,1)模型。以现场管道结蜡数据和室内环道结蜡数据为例,对比改进GM(1,1)模型、基于函数cot(x^(2))变换建立的GM(1,1)模型及传统GM(1,1)模型之间的预测精度,并分析平移量c对改进GM(1,1)模型预测精度的影响。结果表明:改进GM(1,1)模型的预测精度最高,其次是基于函数cot(x^(2))变换建立的GM(1,1)模型,而传统GM(1,1)模型的预测精度最低;随着平移量的增大,改进GM(1,1)模型的平均相对预测误差呈现出先减小后增大的趋势,因此合理的平移量有助于模型精度的提高。应用改进GM(1,1)模型来预测管道蜡沉积厚度是可行的,该方法可为含蜡原油管道蜡沉积厚度的准确预测提供参考和借鉴。In order to accurately predict the wax deposition thickness in the waxy crude oil pipeline,a new improved GM(1,1)model was proposed by using the cot(x^(2)+c)transformation based on the function cot(x^(2))transformation and combined with the idea of translation transformation.Taking the field pipeline wax deposition data and indoor loop wax deposition data as examples,the prediction accuracies of the improved GM(1,1)model,the GM(1,1)model based on function cot(x^(2))transformation and the traditional GM(1,1)model were compared,and the influence of translation amount on the prediction accuracy of the improved GM(1,1)model was analyzed.The results showed that the improved GM(1,1)model had the highest prediction accuracy,followed by the GM(1,1)model based on function cot(x^(2))transformation,while the traditional GM(1,1)model had the lowest prediction accuracy.With the increase of translation amount,the average relative prediction error of the improved GM(1,1)model presented a trend of decreasing first and then increasing.Therefore,the reasonable translation amount was helpful to improve the model accuracy.It is feasible to apply the improved GM(1,1)model to predict the wax deposition thickness of pipelines,which can provide useful reference for the accurate prediction of wax deposition thickness of waxy crude oil pipelines.

关 键 词：蜡沉积厚度 函数变换 平移变换 改进GM(1 1)模型 精度对比 

分 类 号：X937[环境科学与工程—安全科学]