:::

詳目顯示

回上一頁
題名:以委員會機器為基底的滲透率預測模式
作者:陳昶旭
作者(外文):Chang-Hsu Chen
校院名稱:國立成功大學
系所名稱:資源工程學系碩博士班
指導教授:林再興
學位類別:博士
出版日期:2006
主題關鍵詞:保留法交叉驗證袋裝法滲透率類神經網路經驗公式井測預測分而治之委員會機器divide and conquerbaggingcross-validationholdoutpredictionempirical formulawell logpermeabilityneural networkcommittee machine
原始連結:連回原系統網址new window
相關次數:
  • 被引用次數被引用次數:期刊(0) 博士論文(0) 專書(0) 專書論文(0)
  • 排除自我引用排除自我引用:0
  • 共同引用共同引用:0
  • 點閱點閱:1
本研究目的為應用委員會機器(Committee Machine, CM)原理來發展新的滲透率預測模式,其中傳統的滲透率經驗公式及倒傳遞類神經網路(BPNN)皆被用來扮演委員會機器中的專家成員角色,發展新的滲透率預測模式;應用委員會機器原理及資料處理的統計技術所發展的滲透率預測模式,包括有CMEF模式(committee machine with empirical formulas, CM結合經驗式)、CNNH模式(committee neural network with holdout method, CNN結合保留法)、CNNKCV模式(CNN with K-fold cross-validation, CNN結合交叉驗證)、CNNB模式(CNN with bagging, CNN結合袋裝法)及CNNDC模式(CNN with divide and conquer, CNN結合分而治之法)。
在本研究中所有以CM為基底發展的模式,在利用井測資料及岩心資料,進行滲透率預測時,並不需要像傳統的滲透率經驗公式一樣,需要知道很明確的地層流體性質及岩石性質的詳細資料;此外,這些CM基底的滲透率模式,在進行預測時,都比傳統的經驗公式或單一的BPNN模式具有較佳的強健性(robust)及正確性(accurate)。
本研究應用所提出的CM滲透率模式成功地分析了一組現場資料,這些被用來當作模式輸入輸出的資料包括八種井測資料及相對應深度的岩心量測滲透率,八種井測包括井徑電測(CALI)、自然珈瑪電測(GR)、深感應式電阻電測(LLD)、微球聚式焦點電阻電測(MSFL)、淺感應式電阻電測(LLS)、聲波電測(DT)、密度電測(RHOB)、中子電測(NPHI)等。
本研究的分析結果中,CMEF模式結合CM的概念,將三個常用的滲透率經驗公式當作是預測專家,此模式展現的預測結果優於三種經驗公式的個別預測結果;CNNH模式利用不同的起始神經元連結值來訓練各個預測專家(BPNN)成員,而CNNKCV模式則利用不同組合的訓練資料集合來訓練各個預測專家(BPNN)成員,由最後預測的結果得知,CNNH模式及CNNKCV模式皆可獲得比單一BPNN模式更高的預測效能。
至於經過袋裝法(bagging)處理的CNNB模式,其判定係數遠比沒有進行bagging的模式為高,此結果充分展現了bagging技術的效果,此模式藉由bagging來擴大利用有限獲得的資料點,並提高了滲透率預測的效能。而CNNDC模式則把大範圍的滲透率分佈資料分割成低滲透率及高滲透率等兩個子範圍,然後對各自子範圍內的訓練集合資料點做訓練學習,並進行各自的解析預測,此模式不僅改善了滲透率預測的精確度,而且其概化(generalization)也獲得了較佳的結果。
This study aims to employ committee machines (CMs) to propose novel models for permeability estimation. In this study, conventional empirical formulas and backpropagation neural networks (BPNNs) were used as members of the CMs. These CM-based models were developed to improve prediction for formation permeability.
The proposed CM-based models included the CMEF model (CM with empirical formulas), CNNH model (committee neural network [CNN] with holdout method), CNNKCV model (CNN with K-fold cross-validation), CNNB model (CNN with bagging), and CNNDC model (CNN with divide and conquer).
In this study, all the CM-based models were developed to predict the permeability directly from well logs without explicit knowledge of the fluid and rock properties. This study demonstrated that the CM-based models are more robust and accurate than the conventional empirical formulas and a single BPNN.
The CM-based models were successfully applied to analyze field data consisting of eight well logs and core-measured permeability. The eight logs used included Caliper (CALI), Gamma Ray (GR), Laterolog Deep Resistivity (LLD), Micro Spherically Focused Resistivity (MSFL), Laterolog Shallow Resistivity (LLS), Interval Transit Time (DT), Bulk Density (RHOB), and Thermal Neutron Porosity (NPHI) logs.
The CMEF model, combining outputs from three empirical formulas, was superior to the three individual empirical formulas acting alone. As for the CNNH and CNNKCV models, BPNNs in the CNNH model were trained by varying the initial weights, whereas BPNNs in the CNNKCV model were trained by varying the combined training data. The results from the CNNH and CNNKCV models demonstrated the power of CM by obtaining larger R2 value compared with a single BPNN.
The CNNB model was developed to leverage the limited core data pairs by applying the bagging (bootstrap aggregating) technique. This model demonstrated its effect with a small data set by obtaining a considerably larger R2 value compared with a model without bagging. The CNNDC model was provided to demonstrate the benefits of modularity by decomposing the permeability range into two sub-ranges to increase the resolution. This implemented model improved the accuracy for permeability prediction and also earned its expected result by achieving the best generalization.
Ahmed, U., Crary, S.F., Coats, G.R., 1991. Permeability estimation: the various sources and their interrelationships. Journal of Petroleum Technology, 578-587.
Aminian, K., Ameri, S., Oyerokun, A., Thomas, B., 2003. Prediction of flow units and permeability using artificial neural networks. Paper SPE 83586.
Arbogast, J.S., Franklin, M.H., 1999. Artificial neural networks and high speed resistivity modeling software speed reservoir characterization. Petroleum Engineer International 72 (5), Houston, 57-61.
Asquith, G., Gibson, G., 1985. Basic Well Log Analysis for Geologists: Methods in Exploration Series. AAPG, Tulsa, Oklahoma 216pp.
Balan, B., Mohaghegh, S., Ameri, S., 1995. State-of-the-art in permeability determination from well log data: part I --a comparative study, model development. SPE Eastern Regional Conference and Exhibition, West Virginia, 17-21.
Bates, J.M., Granger, C.W.J., 1969. The combination of forecast. Operations Research Quarterly 20, 451-468.
Bhatt, A., Helle, H.B., 2002. Committee neural networks for porosity and permeability prediction from well logs. Geophysical Prospecting 50, 645-660.
Bhatt, A., Helle, H.B., Ursin, B., 2001. Application of committee machines in reservoir characterisation while drilling: a novel neural network approach in log analysis. Proceedings of the 6th Nordic Symposium on Peteophysics, Trondheim, Norway.
Breiman, L., 1996. Bagging predictors. Machine Learning 24 (2), 123-140.
Breiman, L., Spector, P., 1992. Submodel selection and evaluation in regression: the x-random case. International Statistical Review 60 (3), 291-319.
Carman, P.C., 1937. Fluid flow through granular beds. Transactions of the Institution of Chemical Engineers 15, London, 150-166.
Chen, Chang-Hsu, Lin, Zsay-Shing, 2006. A committee machine with empirical formulas for permeability prediction. Computers & Geosciences 32, 485-496.
Coates, G.R., Dumanoir, J.L., 1973. A new approach to improve log-derived permeability. Proceedings of SPWLA 14th Annual Logging Symposium, Paper R.
Debra, A., Osborne, D.A., 1992. Neural networks provide more accurate reservoir permeability. Oil & Gas Journal, 125-132.
Dewan, J.T., 1983. Essentials of Modern Open-Hole Log Interpretation. PennWell Publishing Company, Tulsa, Oklahoma 361pp.
Doveton, J.H., Prensky, S.E., 1992. Geological applications of wireline logs: a synopsis of developments and trends. The Log Analyst 33 (3), 286-303.
Efron, B., Gong, G., 1983. A leisurely look at the bootstrap, the jackknife, and cross-validation. The American Statistician 37 (1), 36-48.
Efron, B., Tibshirani, R., 1993. An Introduction to the Bootstrap. Chapman and Hall, New York 436pp.
Fernandes, A.M., Utkin, A.B., Lavrov, A.V., Vilar, R.M., 2005. Design of committee machines for classification of single-wavelengh lidar signals applied to early forest fire detection. Pattern Recognition Letters 26, 625-632.
Geman, S., Bienenstock, E., Doursat, R., 1992. Neural networks and the bias/variance dilemma. Neural Computation 4, 1-58.
Goldberg, D.E., 1989. Genetic Algorithms in Search, Optimization, and Machine Learning. Addison-Wesley, Reading, MA 412 pp.
GS Software, 1989. GS Version 4.0 User’s Manual. GeoScience Software, Inc, Florida 160 pp.
Hagan, M.T., Menhaj, M., 1994. Training feedforward networks with the marquardt algorithm. IEEE Transactions on Neural Networks 5 (6), 989-993.
Hansen, L.K., Salamon, P.R., 1990. Neural network ensembles. IEEE Transactions on Pattern Analysis and Machine Intelligence 12 (10), 993-1001.
Haykin, Simon, 1999. Neural Networks – A Comprehensive Foundation. Prentice Hall, New Jersey 842 pp.
Hornik, K., Stinchcombe, M., White, H., 1989. Multilayer feedforward networks are universal spproximators. Neural Networks 2, 359-366.
Huang, Z., Shimeld, J., Williamson, M., Katsube, J., 1996. Permeability prediction with artificial neural network modeling in the venture gas field, offshore eastern Canada. Geophysics 61, 422-436.
Jacobs, R.A., Jorsan, M.I., Barto, A.G., 1991. Task decomposition through competition in a modular connectionist architecture: the what and where vision tasks. Cognitie Science 15, 219-250.
Jamialahmadi, M., javadpour, F.G., 2000. Relationship of permeability, porosity and depth using an artificial neural network. Journal of Petroleum Science and Engineering 26, 235-239.
Johnson, W.W., 1994. Permeability determination form well logs and core data. The SPE Permian Basin Oil and Gas Recovery Conference in Midland, TX, SPE 27647.
Kapadia, S.P., Menzie, D.E., 1985. Determination of permeability variation factor V from log analysis. SPE Annual Technical Conference and Exhibition, 22-26 September, Las Vegas, Nevada, 1-12.
Kohavi, R., 1995. A study of cross-validation and bootstrap for accuracy estimation and model selection. In C. S. Mellish, ed., ‘Proceedings of the 14th International Joint Conference on Artificial Intelligence', Morgan Kaufmann Publishers, Inc., 1137-1143.
Matlab, 2000. Neural Network Toolbox User's Guide, The MathWorks, Inc., Natick, MA.
Mohaghegh, S., Arefi, R., Ameri, S., 1996. Petroleum reservoir characterization with the aid of artificial neural networks. Journal of Petroleum Science and Engineering 16, 263-274.
Naftaly, U., Intrator, N., Horn, D., 1997. Optimal ensemble averaging of neural networks. Network: Computation in Neural Systems 8, 283-296.
Nelson, P.H., 1994. Permeability-porosity relationship in sedimentary rocks. The Log Analyst 35 (3), 38-62.
Olson, T.M., 1998. Porosity and permeability prediction in low-permeability gas reservoirs from well logs using neural networks. Paper SPE 39964, 563-572.
Reddy, N.P., Buch, O.A., 2003. Speaker verification using committee neural networks. Computer Methods and Programs in Biomedicine 72, 109-115.
Schlumberger Limited, 1972. Log Interpretation, Volume I -- Principles. New York 113pp.
Schlumberger Limited, 1991. Schlumberger Log Interpretation: Principles/Applications. Schlumberger Educational Services, Houston, TX 198 pp.
Sharkey, A.J.C., 1996. On combining artificial neural nets. Connection Science 8, 299-314.
Sharkey, A.J.C., 1999. Multi-net systems. In (Ed) A.J.C. Sharkey Combining artificial neural nets: ensemble and modular multi-net systems , Springer-Verlag, 1-30.
Sharkey, A.J.C., Sharkey, N.E., Gerecke, U., Chandroth, G.O., 2000. The “test and select” approach to ensemble combination. In: Kittler, J. and Roli, F. (eds) Multiple Classifier Systems, Lecture Notes in Computer Science, Springer, 30-44.
Silpngarmlers, N., Guler, B., Ertekin, T., Grader, A.S., 2002. Development and testing of two-phase permeability predictirs using artificial neural networks. Paper 79547, 299-308.
Tang, Yu-Te, 1997. To estimate the formation permeability by using welllog data and permeability calculation models in offshore Taiwan. M. Sc. Thesis, National Cheng Kung University, Tainan, Taiwan 171pp. (in Chinese)
Timur, A., 1968. An investigation of permeability, porosity, and residual water saturation relationships for sandstone reservoirs. The Log Analysts 9 (4), 8-17.
Tixier, M.P., 1949. Evaluation of permeability from electric-log resistivity gradients. Oil and Gas Journal 48 (6), 113-122.
Twomey, J.M., Smith, A.E., 1998. Bias and variance of validation methods for function approximation neural networks under conditions of sparse data. IEEE Transactions on systems, man, and cybernetics 28 (3), part C, 417-430.
Weiss, S.M., Indurkhya, N., 1994. Decision tree pruning: biased or optimal. in Proceedings of the twelfth national conference on artificial intelligence. AAAI Press and MIT Press, 626-632.
Wong, P.M., Henderson, D.J., Brooks, L.J., 1997. Reservoir permeability determination from well log data using artificial neural network: an example from the ravva field, offshore India, SPE Asia Pacific Oil & Gas Conference and Exhibition, Kuala Lumpur, SPE 38034, 149-155.
Wong, P.M., Jang, M., Cho, S., Gedeon, T.D., 2000. Multiple permeability predictions using an observational learning algorithm. Computers & Geosciences 26, 907-913.
Wong, P.M., Taggart, I.J., Gedeon, T.D., 1995. Use of neural network methods to predict porosity and permeability of a petroleum reservoir. AI Applications 9 (2), 27-37.
Wyllie, M.R.J., Rose, W.D., 1950. Some theoretical considerations related to the quantitative evaluations of the physical characteristics of reservoir rock from electric log data. Transactions, the American Institute of Mining Engineers (AIME) 189, 105-118.
Yao, C.Y., Holditch, S.A., 1993. Estimating permeability profiles using core and log data. SPE Eastern Regional Meeting, 2-4 November, Pittsburgh, PA, 317-322.
Zhao, Z.-Q., Huang, D.-S., Sun, B.-Y., 2004. Human face recognition based on multi-features using neural networks committee. Pattern Recognition Letters 25, 1351-1358.
 
 
 
 
第一頁 上一頁 下一頁 最後一頁 top
QR Code
QRCODE