国产日韩欧美一区二区三区三州_亚洲少妇熟女av_久久久久亚洲av国产精品_波多野结衣网站一区二区_亚洲欧美色片在线91_国产亚洲精品精品国产优播av_日本一区二区三区波多野结衣 _久久国产av不卡

?

融合機制與高斯混合回歸算法的成品油管道順序輸送混油長度預(yù)測模型

2023-09-04 21:48:42袁子云劉剛陳雷邵偉明張鈺晗
關(guān)鍵詞:數(shù)據(jù)機制

袁子云 劉剛 陳雷 邵偉明 張鈺晗

摘要:成品油管道順序輸送過程中會出現(xiàn)混油現(xiàn)象,精確預(yù)測混油長度對油品批次切割具有重要意義,混油長度機制模型存在精度不高,數(shù)值計算量龐雜等問題。當(dāng)前基于機器學(xué)習(xí)算法構(gòu)建的全局預(yù)測模型未考慮實際工況多模態(tài)特性,預(yù)測精度受限;直接引入高斯混合回歸算法辨識數(shù)據(jù)模態(tài)難以準(zhǔn)確表征變量間復(fù)雜非線性關(guān)系。采用現(xiàn)有機制計算公式與高斯混合回歸算法構(gòu)建融合機制認(rèn)知的局部建模算法,基于真實成品油管道順序輸送混油長度數(shù)據(jù)集進(jìn)行不同模型預(yù)測結(jié)果對比試驗。結(jié)果表明,融合機制認(rèn)知與局部建模算法能有效表征變量間函數(shù)關(guān)系,新模型預(yù)測精度有明顯優(yōu)勢。

關(guān)鍵詞:成品油管道; 混油長度; 局部建模; 高斯混合回歸; 機制-數(shù)據(jù)

中圖分類號:TP 181 文獻(xiàn)標(biāo)志碼:A

引用格式:袁子云,劉剛,陳雷,等.融合機制與高斯混合回歸算法的成品油管道順序輸送混油長度預(yù)測模型[J].中國石油大學(xué)學(xué)報(自然科學(xué)版),2023,47(2):123-128.

YUAN Ziyun, LIU Gang, CHEN Lei, et al. Predictive model of mixed oil length for sequential transportation of multi-product pipeline by combining mechanism and Gaussian mixture regression algorithm [J]. Journal of China University of Petroleum (Edition of Natural Science), 2023,47(2):123-128.

Predictive model of mixed oil length for sequential transportation of multi-product pipeline by combining mechanism and

Gaussian mixture regression algorithm

YUAN Ziyun1,2 , LIU Gang1,2, CHEN Lei1,2, SHAO Weiming2,? ZHANG Yuhan3

(1.College of Pipeline and Civil Engineering in China University of Petroleum(East China),? Qingdao 266580, China;2.Shandong Provincial Key Laboratory of Oil & Gas Storage and Transportation Safety, Qingdao 266580, China; 3.Qingdao Operation Area, Shandong Branch, PipeChina,? Qingdao 266400, China)

Abstract: The oil mixing phenomenon? occurs during the sequential transportation of the multi-product pipeline, and the accurate prediction of the length of the mixed oil is of great significance for the cutting batch segment. The mechanism model is faced with problems such as low accuracy and complex numerical simulation. In the current global predictive models derived from machine learning algorithms, the multi-mode characteristics of actual operating conditions are ignored, thus the predictive accuracy is limited. The Gaussian mixture regression algorithm cannot accurately characterize the complex nonlinear relationship among variables if it is directly introduced to identify the data mode. Based on the existing mechanism equation and the Gaussian mixture regression algorithm, we develop a local modeling algorithm that integrates the mechanism knowledge. Based on the real product oil pipeline sequential transportation mixed oil length data set, a comparison experiment? among different models was carried out, and the results show that the mechanism and local modeling algorithm can effectively characterize the functional relationship of variables, and the predictive accuracy of the new model has obvious advantages.

Keywords: multi-product pipeline; mixed oil length; local modeling; Gaussian mixture regression; mechanism-data

成品油管道通常采取順序輸送,相鄰兩批次油品間不可避免產(chǎn)生混油[1],混油長度是順序輸送過程中油品批次切割的重要數(shù)據(jù)依據(jù)[2]。準(zhǔn)確預(yù)測成品油管道混油長度,對順序輸送過程的實時監(jiān)控、油品批次切割意義重大[3-4]。目前混油長度計算機制模型可分為一維模型與二維模型。一維模型如Austin-Palfrey混油計算公式[5]應(yīng)用簡便,但計算精度有待提升[6]。二維模型[7-8]能更準(zhǔn)確地刻畫混油形成與發(fā)展過程,但模型復(fù)雜度高,求解復(fù)雜,難以應(yīng)用于長距離成品油管道[6,9]。數(shù)據(jù)驅(qū)動建模方法具備良好非線性擬合能力[10-11],但該方法旨在盡可能擬合已有樣本,難以保證其泛化性[12-13]。因此Chen等[14-15]傾向于將Austin-Palfrey公式與數(shù)據(jù)驅(qū)動建模算法相結(jié)合,然而現(xiàn)有混油長度預(yù)測模型均依賴一個單獨的全局預(yù)測模型完成回歸任務(wù)?,F(xiàn)實場景中不同管道內(nèi)部的物理流動空間與流體流動機制存在差異,導(dǎo)致數(shù)據(jù)集呈現(xiàn)明顯多模態(tài)特性[16-17]。針對多模態(tài)問題通常采用“分而治之”理念,即為每個待預(yù)測樣本構(gòu)建局部預(yù)測模型以精準(zhǔn)挖掘數(shù)據(jù)關(guān)系[18]。采用高斯混合回歸算法(Gaussian mixture regression, GMR)辨識數(shù)據(jù)多模態(tài)特性是當(dāng)今主流方法[19],但其假定變量間服從簡單線性關(guān)系,模型預(yù)測精度存疑。針對混油段長度預(yù)測問題將GMR算法融合Austin-Palfrey混油公式,借助真實成品油管道混油長度數(shù)據(jù)集開展模型性能分析,融合機制的GMR算法具備明顯預(yù)測精度優(yōu)勢,驗證新算法對求解成品油管道混油長度預(yù)測問題的適用性。

1 原理與方法

1.3 融合機制公式與GMR算法的GMR-M模型

在GMR模型中假定第k個模態(tài)內(nèi)輸入輸出變量間的函數(shù)關(guān)系為簡單的線性關(guān)系。因成品油管道順序輸送過程受多因素耦合影響,且流體流動狀態(tài)復(fù)雜多變,輸入輸出變量間應(yīng)服從復(fù)雜非線性關(guān)系,完全基于GMR算法構(gòu)建的預(yù)測模型將難以準(zhǔn)確描述混油長度發(fā)展規(guī)律。因此考慮結(jié)合GMR算法與已有機制公式,將多維輸入變量與對應(yīng)輸出變量間的復(fù)雜非線性關(guān)系簡化為線性關(guān)系,再利用GMR算法辨識數(shù)據(jù)間隱含的多模態(tài)關(guān)系,在同一模態(tài)下構(gòu)建局部預(yù)測模型,實現(xiàn)預(yù)測精度的有效提升。GMR-M建模具體流程如下,相應(yīng)示意圖見圖2。

(1)基于式(5),將管道內(nèi)徑d、輸送距離L以及運行雷諾數(shù)Re整合成變量CAP。

(2)結(jié)合GMR算法探尋不同模態(tài)條件下輸入變量CAP與輸出變量即混油長度C間的函數(shù)關(guān)系。

(3)輸入待預(yù)測樣本xq可預(yù)測相應(yīng)混油長度預(yù)測值。

考慮到人工神經(jīng)網(wǎng)絡(luò)作為數(shù)據(jù)分析領(lǐng)域主流算法之一[21-22]及其對復(fù)雜非線性函數(shù)的優(yōu)秀擬合能力,選擇其為對照方法以驗證GMR-M模型在混油長度預(yù)測問題上的優(yōu)越性。此外為說明融合機制公式與考慮數(shù)據(jù)多模態(tài)特點的重要性,基于變量重組方式構(gòu)建了ANN-M模型。相較于GMR-M模型,基于人工神經(jīng)網(wǎng)絡(luò)算法的ANN-M模型并未辨識數(shù)據(jù)多模態(tài)信息。單純基于GMR,ANN構(gòu)建的預(yù)測模型的輸入變量信息為L、d、Re;GMR-M和ANN-M的輸入變量信息為CAP。

2 實 例

采用真實混油長度數(shù)據(jù)集以驗證GMR-M模型的適用性,以Austin-Palfrey公式,現(xiàn)有兩種預(yù)測模型以及單純基于GMR,ANN算法構(gòu)建模型與ANN-M模型的預(yù)測結(jié)果作為基準(zhǔn),對比分析GMR-M模型在預(yù)測精度方面的表現(xiàn)。其中GMR-M與GMR模型模態(tài)數(shù)均設(shè)置為3,ANN與ANN-M隱藏層神經(jīng)元個數(shù)設(shè)置為10。主要采用均方根誤差(root mean square error, RMSE)、最大絕對誤差(max absolute error, MAE)與決定系數(shù)R2作為評價模型的預(yù)測性能指標(biāo),評價指標(biāo)分別為

式中,ERMS和EMA分別為均方根誤差和最大絕對誤差;yq、q與q分別為樣本實際值、預(yù)測值與樣本均值;Q為測試樣本數(shù)量。

R2指標(biāo)越大,表明預(yù)測值與實際值吻合程度更高。而RMSE與MAE指標(biāo)越大,代表預(yù)測結(jié)果越偏離實際值。利用SCADA(supervisory control and data acquisition)系統(tǒng)采集的中國南方三條成品油管道生產(chǎn)運行數(shù)據(jù)作為樣本來源,部分樣本基本信息如表1所示。

前兩條管道共計1 948個樣本用于訓(xùn)練模型,第三條管道中528個樣本用于構(gòu)建測試數(shù)據(jù)集以評估不同預(yù)測模型的泛化性能。各模型相應(yīng)預(yù)測指標(biāo)列于表2。

由表2可知,對于現(xiàn)有預(yù)測模型,Chen模型的RMSE指標(biāo)已超過Austin-Palfrey公式的預(yù)測結(jié)果,表明該模型預(yù)測值擬合樣本實際值效果不佳。雖然Yuan模型表現(xiàn)出相對較優(yōu)的預(yù)測性能,但未考慮數(shù)據(jù)多模態(tài)特性仍導(dǎo)致其出現(xiàn)預(yù)測失真,相比于現(xiàn)有機制計算公式難以顯現(xiàn)出明顯預(yù)測優(yōu)勢。由于神經(jīng)網(wǎng)絡(luò)具備復(fù)雜非線性擬合能力,ANN與ANN-M模型在混油長度預(yù)測問題上表現(xiàn)出較好的預(yù)測能力。從整體來看,二者的決定系數(shù)R2超過0.94,模型預(yù)測結(jié)果與實際值較為接近。相較于現(xiàn)有兩種預(yù)測模型,基于ANN算法構(gòu)建的預(yù)測模型具備一定優(yōu)勢。此外相較于ANN模型,基于已有計算公式重組輸入變量得到的ANN-M能更精確捕捉變量間的函數(shù)映射關(guān)系,預(yù)測性能有一定的提升。但其MAE指標(biāo)均約為900且超過了現(xiàn)有兩種預(yù)測模型,說明基于神經(jīng)網(wǎng)絡(luò)構(gòu)建的預(yù)測模型對個別樣本出現(xiàn)了較嚴(yán)重的預(yù)測偏差,模型預(yù)測能力仍有待提升??芍蚝雎詳?shù)據(jù)集內(nèi)樣本可能來源于不同模態(tài)導(dǎo)致變量間函數(shù)關(guān)系存在的差異,即使融合已有機制公式并采用具備擬合非線性能力的數(shù)據(jù)分析算法,全局建模方法預(yù)測分屬不同模態(tài)樣本時適用性仍欠佳,導(dǎo)致預(yù)測結(jié)果偏離實際情況。

對比GMR模型,由于未有機結(jié)合已有機制公式,模型無法有效表征輸入變量與輸出變量間的復(fù)雜非線性關(guān)系,預(yù)測結(jié)果不理想。反映在GMR表現(xiàn)出最高的RMSE和MAE指標(biāo),說明直接引入GMR算法難以解決混油長度預(yù)測問題。與之相對的,融合了機制表達(dá)形式且采用局部建模方法的GMR-M模型,預(yù)測結(jié)果更貼近真實情況。GMR-M模型的RMSE與R2預(yù)測指標(biāo)均明顯優(yōu)于其他模型。具體而言,GMR-M是RMSE指標(biāo)唯一低于200的模型,且表現(xiàn)出最低的MAE指標(biāo),充分表明GMR-M具備良好的預(yù)測精度。上述結(jié)果有效驗證了融合機制公式與局部建模方法在準(zhǔn)確預(yù)測成品油管道順序輸送混油長度預(yù)測問題中的重要性。

圖3為各模型估計值與測試集樣本實際值的擬合情況。由圖3可以看到,相較于混油長度實際值,基于Austin-Palfrey公式得到的預(yù)測值偏低;因缺少模態(tài)識別步驟,Chen-和Yuan模型表現(xiàn)出明顯的預(yù)測偏差;單純基于GMR算法構(gòu)建的預(yù)測模型,由于未有機融合機制公式,導(dǎo)致模型陷入過擬合,預(yù)測精度不理想;ANN和ANN-M預(yù)測值較接近實際值,然而融合了機制公式與多模態(tài)識別功能的GMR-M模型表現(xiàn)出最高的預(yù)測精度,預(yù)測值擬合實際值效果最好,表明模型精準(zhǔn)捕捉到了輸入輸出變量間的函數(shù)映射關(guān)系。

3 結(jié)束語

為克服現(xiàn)有成品油管道順序輸送混油長度預(yù)測方法中存在的不足,提出了一種融合機制與GMR算法的成品油管道混油長度預(yù)測模型GMR-M。與現(xiàn)有預(yù)測模型以及ANN模型相比,因考慮了數(shù)據(jù)內(nèi)部多模態(tài)特性并針對性地構(gòu)建了多個局部預(yù)測模型完成回歸任務(wù),GMR-M模型能有效提高成品油管道混油長度預(yù)測精度;對比已有機制公式與單純采用GMR算法構(gòu)建的預(yù)測模型,通過耦合現(xiàn)場數(shù)據(jù)攜帶的關(guān)鍵信息與已有機制認(rèn)知,GMR-M模型能更有效表征輸入輸出變量間的復(fù)雜函數(shù)關(guān)系,預(yù)測結(jié)果更接近于真實情況。

參考文獻(xiàn):

[1]HE Guoxi, LIN Mohan, WANG Baoying, et al. Experimental and numerical research on the axial and radial concentration distribution feature of miscible fluid interfacial mixing process in products pipeline for industrial applications[J]. International Journal of Heat & Mass Transfer, 2018,127:728-745.

[2]SHAHANDEH H, LI Z. Modeling and optimization of the upgrading and blending operations of oil sands Bitumen[J]. Energy & Fuels, 2016,30(JUL.SPEC.):5202-5213.

[3]MORADI S, MIRHASSANI S A. Robust scheduling for multi-product pipelines under demand uncertainty[J]. The International Journal of Advanced Manufacturing Technology, 2016,87(9):2541-2549.

[4]CAFARO V G, CAFARO D C, MNDEZ CA. Optimization model for the detailed scheduling of multi-source pipelines[J]. Computers & Industrial Engineering, 2015,88:395-409.

[5]AUSTIN J E,PALFREY J R. Mixing of miscible but dissimilar liquids in serial flow in a pipeline[J]. P Mech Eng B-J Eng, 1963,178:377-389.

[6]孫健飛,梁永圖.成品油管道順序輸送混油模型研究進(jìn)展[J].油氣儲運,2019,38(5):496-502.

SUN Jianfei, LIANG Yongtu. Research progress on the mixed oil models for the batch transportation in products pipeline [J]. Oil & Gas Storage and Transportation, 2019,38(5):496-502.

[7]夏增艷,劉青泉.順序輸送混油過程的二維數(shù)值分析[J].力學(xué)與實踐,2010,32(6):13-17.

XIA Zengyan, LIU Qingquan. Numerical simulation of the contamination between batches in multi-product pipeline transport [J]. Mechanics in Engineering, 2010,32(6):13-17.

[8]馬鋼,白瑞.成品油管道二維混油理論數(shù)值分析研究[J].油氣田地面工程,2018,37(7):54-59.

MA Gang, BAI Rui. Study on the two-dimensional mixed oil theory numerical analysis for product oil pipelines [J]. Oil-Gas Field Surface Engineering, 2018,37(7):54-59.

[9]吳玉國.冷熱原油順序輸送技術(shù)研究[D].青島:中國石油大學(xué)(華東),2010.

WU Yuguo. Research on technology of the batch transportation of cold and hot crude oils [D]. Qingdao:China University of Petroleum(East China) ,2010.

[10]何玉榮,宋志超,張燕明,等.機器學(xué)習(xí)在水力壓裂作業(yè)中的應(yīng)用綜述[J].中國石油大學(xué)學(xué)報(自然科學(xué)版),2021,45(6):127-135.

HE Yurong, SONG Zhichao, ZHANG Yanming, et al. Review on application of machine learning in hydraulic fracturing [J]. Journal of China University of Petroleum(Edition of Natural Science), 2021,45(6):127-135.

[11]王艷松,趙惺,李強,等.基于油氣開采的海上油田中長期電力負(fù)荷預(yù)測[J].中國石油大學(xué)學(xué)報(自然科學(xué)版),2021,45(2):127-133.

WANG Yansong, ZHAO Xing, LI Qiang, et al. Medium and long term power load prediction of offshore oil field based on oil and gas exploitation [J]. Journal of China University of Petroleum (Edition of Natural Science), 2021,45(2):127-133.

[12]張黎明,陳昕晟,李國欣,等.基于集合和神經(jīng)網(wǎng)絡(luò)架構(gòu)搜索的自動歷史擬合方法[J].中國石油大學(xué)學(xué)報(自然科學(xué)版),2022,46(2):127-136.

ZHANG Liming, CHEN Xinsheng, LI Guoxin, et al. An automatic history matching method based on ensemble and neural architecture search [J]. Journal of China University of Petroleum (Edition of Natural Science), 2022,46(2):127-136.

[13]潘少偉,王朝陽,張允,等.基于長短期記憶神經(jīng)網(wǎng)絡(luò)補全測井曲線和混合優(yōu)化XGBoost的巖性識別[J].中國石油大學(xué)學(xué)報(自然科學(xué)版),2022,46(3):62-71.

PAN Shaowei, WANG Chaoyang, ZHANG Yun, et al. Lithology identification based on LSTM neural networks completing log and hybrid optimized XGBoost [J]. Journal of China University of Petroleum(Edition of Natural Science), 2022,46(3):62-71.

[14]CHEN L, YUAN Z Y, LIU G, et al. A novel predictive model of mixed oil length of products pipeline driven by traditional model and data[J]. Journal of Petroleum Science and Engineering,2021,205:108787.

[15]YUAN Z Y, CHEN L, SHAO W M, et al. A robust hybrid predictive model of mixed oil length with deep integration of mechanism and data[J]. Journal of Pipeline Science and Engineering, 2021,1(4):459-467.

[16]SOUZA F, RUI A. Mixture of partial least squares experts and application in prediction settings with multiple operating modes[J]. Chemometrics and Intelligent Laboratory Systems, 2013,130(2):192-202.

[17]SHAO W, GE Z, SONG Z. Soft-sensor development for processes with multiple operating modes based on semisupervised Gaussian mixture regression[J]. IEEE Transactions on Control Systems Technology, 2019,27(5):2169-2181.

[18]SHAO W, GE Z, SONG Z, et al. Data-driven predictive model based on locally weighted Bayesian Gaussian Regression:? 2019 IEEE 8th Data Driven Control and Learning Systems Conference (DDCLS)[C]. Dali: IEEE, 2019.

[19]WANG J, SHAO W, SONG Z. Bayesian regularized Gaussian mixture regression with application to soft sensor modeling for multi-mode industrial processes:2018 IEEE 7th Data Driven Control and Learning Systems Conference (DDCLS)[C]. Enshi: IEEE, 2018.

[20]SHAO W, XIAO C, WANG J, et al. Real-time estimation of quality-related variable for dynamic and non-Gaussian process based on semisupervised Bayesian HMM[J]. Journal of Process Control, 2022,111:59-74.

[21]宋先知,朱碩,李根生,等.基于BP-LSTM雙輸入網(wǎng)絡(luò)的大鉤載荷與轉(zhuǎn)盤扭矩預(yù)測[J].中國石油大學(xué)學(xué)報(自然科學(xué)版),2022,46(2):76-84.

SONG Xianzhi, ZHU Shuo, LI Gensheng, et al. Prediction of hook load and rotary drive torque during well-drilling using a BP-LSTM network[J]. Journal of China University of Petroleum(Edition of Natural Science),2022,46(2):76-84.

[22]鄭秋梅,商振浩,王風(fēng)華,等.基于深度神經(jīng)網(wǎng)絡(luò)和支持向量機的海底管線水合物生成預(yù)測模型[J].中國石油大學(xué)學(xué)報(自然科學(xué)版),2020,44(5):46-51.

ZHENG Qiumei, SHANG Zhenhao, WANG Fenghua, et al. Prediction model of submarine pipeline hydrate formation based on deep neural network and support vector machines[J]. Journal of China University of Petroleum(Edition of Natural Science), 2020,44(5):46-51.

(編輯 沈玉英)

收稿日期:2022-08-12

基金項目:國家重點研發(fā)計劃(2021YFA1000104);國家自然科學(xué)基金項目(52174068);中央高校自主創(chuàng)新基金項目(22CX01001A-5)

第一作者:袁子云(1998-),男,博士研究生,研究方向為油氣管網(wǎng)大數(shù)據(jù)分析。E-mail:yuanziyun@s.upc.edu.cn。

通信作者:劉剛(1975-),男,教授,博士,博士生導(dǎo)師,研究方向為油氣管道系統(tǒng)數(shù)據(jù)挖掘與智能決策的應(yīng)用。E-mail:liugang@upc.edu.cn。

猜你喜歡
數(shù)據(jù)機制
構(gòu)建“不敢腐、不能腐、不想腐”機制的思考
自制力是一種很好的篩選機制
文苑(2018年21期)2018-11-09 01:23:06
溫室控制系統(tǒng)及控制方法的研究
公路工程試驗檢測存在的問題及措施
價值工程(2016年30期)2016-11-24 16:39:03
焊接工藝仿真訓(xùn)練系統(tǒng)中焊點數(shù)據(jù)的建立方法
一種借助數(shù)據(jù)處理構(gòu)建的智能食堂管理系統(tǒng)
定向培養(yǎng) 還需完善安置機制
數(shù)據(jù)化藝術(shù)的生成探究
淺談計量自動化系統(tǒng)實現(xiàn)預(yù)購電管理應(yīng)用
破除舊機制要分步推進(jìn)
恩施市| 陆良县| 张家口市| 平远县| 铅山县| 宣武区| 黑水县| 河曲县| 洪湖市| 呈贡县| 广宗县| 莆田市| 大洼县| 宜兴市| 磴口县| 逊克县| 建始县| 龙川县| 电白县| 金乡县| 玉山县| 土默特右旗| 洛宁县| 鸡泽县| 年辖:市辖区| 镇巴县| 天长市| 普格县| 霍州市| 云浮市| 涿州市| 宜城市| 加查县| 康保县| 固始县| 宁明县| 黔西县| 弥勒县| 溆浦县| 如东县| 英超|