Expensive to drill wells when all goes to plan. Drilling problems add costs throughout the well s life.
|
|
- Julian Chase
- 6 years ago
- Views:
Transcription
1 Beau Rollins
2 Map View Cross Section View Expensive to drill wells when all goes to plan. Surface Location Drilling problems add costs throughout the well s life. The best scenario for minimizing costs and problems is to drill the plan as quickly as possible. Surface Location Best Planned vs. Actual
3 Drilling Subject Matter Experts Ali Charlton Tristan Arbus Garrett Robberson Real-Time Drilling Data Collection Don Morrison Amos Hall BHA Details and Directional Surveys Data Collection Beau Rollins
4 Position Measurement Measured at discrete points. Tools cannot take measurements while rotating. Measured Depth Azimuth Inclination Map View Cross Section View Continuous Drilling Data DP_1 DP_2 DP_3 DP_4 & DP_ Planned vs. Actual Directional Survey MD AZI INC BHA 1 Bottom Hole Assembly (BHA) Details θ 1 BHA 2 θ 2 New Components Stabilizer
5 Measured Direction Survey Data is acquired from measurement tools at regular survey stations of 30 to 90 a part. Interpolated Direction Survey Data is interpolated between measured survey stations using the minimum curvature method to populate data at 1 intervals. This interpolation is performed by a software called Compass. Sliding Slow course corrections. Train and Validate Score
6 The bit s measured position, at any point, is its most likely position in an ellipse of uncertainty. This uncertainty cumulatively grows as the bit moves laterally away from the point of origin. Even after advanced processing, position errors persist. Northing: 60 x 20 Easting 70 x 20 All the content on this slide is from this Halliburton presentation.
7 Can we predict our position before we take the next position measurement by predicting the change in azimuth and inclination of the bit based on drilling data and the details of the bottom hole assembly (BHA)? Can we create a general model to apply to all actively drilling wells Train and validate on one well. Score a second well. Drilling parameters can account for most of the variance in the targets Unmeasured rock properties could have impact The model focuses on the relationship between drilling parameters and position
8 What are the factors that can predict the changes in azimuth and inclination of the bit? Drilling Data Bottom Hole Assembly (BHA) Details Geologic Data Revolutions per Minute Rate of Penetration Toolface Data Differential Pressure Flow in Torque Weight on Bit Standpipe Pressure Hookload Block Position When are we Sliding? Is the Housing Stabilized? Distance to Stabilizer Bend Setting (Degrees) Bend Type Bit Box to Bend Length Gamma Ray
9 Targets The objective is to model the change in inclination and azimuth based on the values the drilling data for the current row. This project will have three types of error: 1) Predicted Difference 2) Predicted Orientation 3) Predicted Spatial Position
10 Target: Dif Azi Perfect correlation between these variables in this linear combination.
11 Target: Dif Inc
12 I changed my approach after analyzing the model diagnostics from the previous modeling work to a machine learning solution. Only horizontal portion of wellbore 70/30 Training/ Validation partitioning Model competition and scoring of dif_azi Model competition and scoring of dif_inc
13 The four models: Linear Regression Decision Tree Neural Network Random Forest Linear Regression Main Effects Only Model Selection: Stepwise Selection Criteria: Validation Error Model Settings Decision Tree Target Criteria: ProbF Maximum Branch: 3 Maximum Depth: 6 Assessment Measure: Validation Error Neural Network Network: Multi-Layer Perceptron with 1 Hidden Layer with 5 Hidden Units Optimization: Back Prop Input Standardization: Standard Deviation Max Iterations: 1000 Pre-Training: No Selection Criteria: Validation Error Random Forest Max Trees: 1000 Out of Bag Sampling: Proportional (40%) Exhaustive: 5000 Max Depth: 50 Variable Importance: Loss Reduction Selection Criteria: Validation Error
14 Model Competition Selection Statistic: Average Squared Error on Validation Set Dif Azi Competition Dif Inc Competition
15 n.. X1 X2 X3 X4 X5 T Training H Train 1 OOB 1 X2 X3 2 variables are randomly chosen 60% of training data randomly chosen to build model 40% of training data randomly chosen as OOB (self validating) Double validate with validation data Recursive partitioning to maximize difference in average between splits Ensemble scoring of new data Train 2 OOB 2 X1 X4 L Train n OOB n X2 X5 H LL H L Validation X1 X2 X3 X4 X5 T T Validation Find row in forest Find row in forest Find row in forest n Average of Trees Average of Trees Average of Trees
16 Residual by Observation Linear Regression Neural Network Decision Tree
17
18 Residual by Observation Linear Regression Neural Network Decision Tree
19
20 Model Competition Selection Statistic: Average Squared Error on Validation Set Dif Azi Competition Dif Inc Competition Each champion model is scored against a new well. The predictions are saved and sent to Enterprise Guide further analysis.
21 Model Predictions Applied Orientation Measured Orientation Measured Orientation Orientation Error
22 Minimum Curvature Method Do it in SAS
23 Predictions Position Calculation Apply to orientation measurement
24 Model performance on a scored well. The green line is the model prediction for bit position. The blue dots are the actual measured position. Map View Cross Section View
25 Big dots are slide flags The maximum error in almost 1 mile of drilling laterally is +/- 5 feet. The final East/West error at TD is 3 feet.
26 Machine learning can produce useful predictive models for bit position that can be trained on wells and deployed to other wells and still maintain predictive accuracy. The Base SAS code can implement engineering equations which translate machine learned predictions into orientation and position estimates. The model should help our engineers minimize drilling costs and reduce problems throughout the lifecycle of the well. If my company implemented this model, the reduction in lifecycle costs will directly impact the bottom line of each well we drill moving forward.
27
28 The Directional Survey table Variable 1: dif_azi (Outcome). Change in Azimuth, where azimuth is the position of the bit relative to North. Variable 2: dif_inc (Outcome). Change in Inclination, where inclination is the position of the bit relative to straight down. The BHA table Variable 3: Bend Setting (Predictor). The angle of the bend in the bottom hole assembly. Variable 4: Bit Box to Bed Length (Predictor). The distance between the bend to the bit. Variable 5: Bearing Housing Stabilized (Predictor). Flags whether the BHA is stabilized. Variable 6: Distance to Stabilizer (Predictor). Distance between the stabilizer and the bend. The Drilling Parameters table Variable 7: HDEPTH (Predictor). The current depth of the hole. Variable 8: TORQUE (Predictor). The current torque. Variable 9: RPM (Predictor). The current revolutions per minute. Variable 10: FLOW_IN (Predictor). The current flow in rate. Variable 11: SPPA (Predictor). The current standpipe pressure. Variable 12: DIFF_PRESS (Predictor). The current differential pressure. Variable 13: HKLD (Predictor). The current hook load. Variable 14: SWOB (Predictor). The current weight on bit. Variable 15: MUD_VOLUME (Predictor). The current mud tank volume. Variable 16: ROP (Predictor). The current rate of penetration. Variable 17: BLKHT (Predictor). The current block height. Variable 18: GTF (Predictor). The current reading from the gravity tool face. Variable 19: MTF (Predictor). The current reading from the magnetic tool face. Variable 20: PSUEDO_MSE (Predictor). The current value of the calculated mechanical specific energy. Variable 21: Slide_v_ Rotate (Predictor). Flags whether the drill bit is rotating or sliding.
29 The model will be deployed in near real-time on incoming drilling data and directional surveys. A time series transfer function or a linear regression model with serially correlated errors seemed relevant initially, but there are a few concerns I have with the data and the model. My concern with these approaches lies in the real-time deployment complications, and the inability of the predictive modeler to dynamically account for: Missing data values due to transmission errors; level shifts; and additive outliers. The data to be modeled is not as continuous as I had previously believed. The interpolation method of inclination and azimuth resulted in repetitive discrete numbers. I chose a machine learning approach to solve the problem for their robustness in handling missing values. I included a linear regression model in the modeling competitions out of curiosity, however I anticipated the data would violate the assumptions of the linear model. It is my opinion that predictive accuracy would trump parsimony in this business case. Akaike s Information Criteria should be the metric of choice when comparing models. However, tree based models, such as random forests and decision trees, do not have this metric calculated on them in Enterprise Miner. Minimizing average squared error on the validation set will be the deciding metric.
30 R_Dif_Azi R_Dif_Inc
31 Pred_azi_error Pred_inc_error
32 EW_Error NS_Error TVD_Error
High Accuracy Wellbore Surveys Multi-Station Analysis (MSA)
1 High Accuracy Wellbore Surveys Multi-Station Analysis (MSA) Magnetic MWD measurement Uncertainty Conventional MWD Tools report their own orientation: Inclination, Azimuth, and Toolface. These reported
More informationWell Advisor Project. Ken Gibson, Drilling Technology Manager
Well Advisor Project Ken Gibson, Drilling Technology Manager BP Well Advisor Project Build capability to integrate real-time data with predictive tools, processes and expertise to enable the most informed
More informationRandom Forest A. Fornaser
Random Forest A. Fornaser alberto.fornaser@unitn.it Sources Lecture 15: decision trees, information theory and random forests, Dr. Richard E. Turner Trees and Random Forests, Adele Cutler, Utah State University
More informationSeptember 23 rd 2010 ISCWSA Meeting Florence
Could the minimum curvature wellbore reconstruction lead to unrealistic wellbore positioning? A directional step by step approach enables to address this issue Introduction Problem Positioning In SPE paper
More informationPETROPHYSICAL DATA AND OPEN HOLE LOGGING BASICS COPYRIGHT. MWD and LWD Acquisition (Measurement and Logging While Drilling)
LEARNING OBJECTIVES PETROPHYSICAL DATA AND OPEN HOLE LOGGING BASICS MWD and LWD Acquisition By the end of this lesson, you will be able to: Understand the concept of Measurements While Drilling (MWD) and
More informationBig Data Methods. Chapter 5: Machine learning. Big Data Methods, Chapter 5, Slide 1
Big Data Methods Chapter 5: Machine learning Big Data Methods, Chapter 5, Slide 1 5.1 Introduction to machine learning What is machine learning? Concerned with the study and development of algorithms that
More informationBusiness Club. Decision Trees
Business Club Decision Trees Business Club Analytics Team December 2017 Index 1. Motivation- A Case Study 2. The Trees a. What is a decision tree b. Representation 3. Regression v/s Classification 4. Building
More informationPredicting Expenditure Per Person for Cities
Predicting Expenditure Per Person for Cities Group Gamma Daniel Eck, Lian Hortensius, Yuting Sun Bo Yang, Jingnan Zhang, Qian Zhao Background Client A government organization of State Planning Commission
More informationMaking the Most of Borehole Surveying
Making the Most of Borehole Surveying Prof Angus Jamieson University of the Highlands and Islands Video presentation available at www.uhi.ac.uk/surveying-summary This Presentation Covers... 1. Why survey
More informationPad Drilling Using Magnetic MWD
Pad Drilling Using Magnetic MWD Managing Magnetic Interference from nearby casing Neil Bergstrom, P.E. Wellbore Geodetic Specialist Devon Energy ISCWSA SPE Wellbore Positioning i Tech Section Denver 3
More informationPredictive Analytics: Demystifying Current and Emerging Methodologies. Tom Kolde, FCAS, MAAA Linda Brobeck, FCAS, MAAA
Predictive Analytics: Demystifying Current and Emerging Methodologies Tom Kolde, FCAS, MAAA Linda Brobeck, FCAS, MAAA May 18, 2017 About the Presenters Tom Kolde, FCAS, MAAA Consulting Actuary Chicago,
More informationThe New Generation of Rotary Systems May be Closer Than You Think Frank J. Schuh, Pat Herbert, John Harrell, The Validus International Company, LLC
1 AADE-03-NTCE-02 The New Generation of Rotary Systems May be Closer Than You Think Frank J. Schuh, Pat Herbert, John Harrell, The Validus International Company, LLC Copyright 2003 AADE Technical Conference
More informationWell Placement and Ranging Services. Charles Duck
Well Placement and Ranging Services Charles Duck The RADAR services encompass: SAGD (well twinning) applications Well Intercept (well kill) applications Well Avoidance applications Gravity MWD (provide
More informationRobotics. Lecture 5: Monte Carlo Localisation. See course website for up to date information.
Robotics Lecture 5: Monte Carlo Localisation See course website http://www.doc.ic.ac.uk/~ajd/robotics/ for up to date information. Andrew Davison Department of Computing Imperial College London Review:
More informationMIT Samberg Center Cambridge, MA, USA. May 30 th June 2 nd, by C. Rea, R.S. Granetz MIT Plasma Science and Fusion Center, Cambridge, MA, USA
Exploratory Machine Learning studies for disruption prediction on DIII-D by C. Rea, R.S. Granetz MIT Plasma Science and Fusion Center, Cambridge, MA, USA Presented at the 2 nd IAEA Technical Meeting on
More informationEnterprise Miner Tutorial Notes 2 1
Enterprise Miner Tutorial Notes 2 1 ECT7110 E-Commerce Data Mining Techniques Tutorial 2 How to Join Table in Enterprise Miner e.g. we need to join the following two tables: Join1 Join 2 ID Name Gender
More informationNetwork Traffic Measurements and Analysis
DEIB - Politecnico di Milano Fall, 2017 Sources Hastie, Tibshirani, Friedman: The Elements of Statistical Learning James, Witten, Hastie, Tibshirani: An Introduction to Statistical Learning Andrew Ng:
More informationMIT 801. Machine Learning I. [Presented by Anna Bosman] 16 February 2018
MIT 801 [Presented by Anna Bosman] 16 February 2018 Machine Learning What is machine learning? Artificial Intelligence? Yes as we know it. What is intelligence? The ability to acquire and apply knowledge
More informationKnowledge Discovery and Data Mining
Knowledge Discovery and Data Mining Lecture 10 - Classification trees Tom Kelsey School of Computer Science University of St Andrews http://tom.home.cs.st-andrews.ac.uk twk@st-andrews.ac.uk Tom Kelsey
More informationFathom Dynamic Data TM Version 2 Specifications
Data Sources Fathom Dynamic Data TM Version 2 Specifications Use data from one of the many sample documents that come with Fathom. Enter your own data by typing into a case table. Paste data from other
More informationENTERPRISE MINER: 1 DATA EXPLORATION AND VISUALISATION
ENTERPRISE MINER: 1 DATA EXPLORATION AND VISUALISATION JOZEF MOFFAT, ANALYTICS & INNOVATION PRACTICE, SAS UK 10, MAY 2016 DATA EXPLORATION AND VISUALISATION AGENDA SAS Webinar 10th May 2016 at 10:00 AM
More informationTHE NEW METHOD OF DIRECTIONAL DRILLING BY NON-ROTATING ADJUSTABLE STABILIZER
THE NEW METHOD OF DIRECTIONAL DRILLING BY NON-ROTATING ADJUSTABLE STABILIZER Eisa Novieri, E.novieri@siau.ac.ir, Department of Petroleum Engineering, Islamic Azad University Susangerd Branch, Susangerd,
More information7. Boosting and Bagging Bagging
Group Prof. Daniel Cremers 7. Boosting and Bagging Bagging Bagging So far: Boosting as an ensemble learning method, i.e.: a combination of (weak) learners A different way to combine classifiers is known
More informationADVANCED ANALYTICS USING SAS ENTERPRISE MINER RENS FEENSTRA
INSIGHTS@SAS: ADVANCED ANALYTICS USING SAS ENTERPRISE MINER RENS FEENSTRA AGENDA 09.00 09.15 Intro 09.15 10.30 Analytics using SAS Enterprise Guide Ellen Lokollo 10.45 12.00 Advanced Analytics using SAS
More informationTL LONGBOW v
http://www.trantlogistics.com eric@trantlogistics.com for licensing and technical support nate@trantlogistics.com for sales and pricing USER'S MANUAL Directional Driller, Proximity, Planner Edition January
More informationA Dynamic Support System For Wellbore Positioning Quality Control While Drilling
A Dynamic Support System For Wellbore Positioning Quality Control While Drilling Dr. Cyrille Mathis, Director & CSO, ThinkTank Maths (Edinburgh) IOGP Geomatics / Statoil Industry Day 2017 26 th April 2017
More informationAutomation of Static and Dynamic FEA Analysis of Bottomhole Assemblies
Automation of Static and Dynamic FEA Analysis of Bottomhole Assemblies Nader E. Abedrabbo, Lev Ring & Raju Gandikota 1 Weatherford International 11909 Spencer Road, Houston, TX nader.abedrabbo@weatherford.com
More informationOverview and Practical Application of Machine Learning in Pricing
Overview and Practical Application of Machine Learning in Pricing 2017 CAS Spring Meeting May 23, 2017 Duncan Anderson and Claudine Modlin (Willis Towers Watson) Mark Richards (Allstate Insurance Company)
More informationLecture 06 Decision Trees I
Lecture 06 Decision Trees I 08 February 2016 Taylor B. Arnold Yale Statistics STAT 365/665 1/33 Problem Set #2 Posted Due February 19th Piazza site https://piazza.com/ 2/33 Last time we starting fitting
More informationRecitation Supplement: Creating a Neural Network for Classification SAS EM December 2, 2002
Recitation Supplement: Creating a Neural Network for Classification SAS EM December 2, 2002 Introduction Neural networks are flexible nonlinear models that can be used for regression and classification
More informationDI TRANSFORM. The regressive analyses. identify relationships
July 2, 2015 DI TRANSFORM MVstats TM Algorithm Overview Summary The DI Transform Multivariate Statistics (MVstats TM ) package includes five algorithm options that operate on most types of geologic, geophysical,
More informationRandom Forests and Boosting
Random Forests and Boosting Tree-based methods are simple and useful for interpretation. However they typically are not competitive with the best supervised learning approaches in terms of prediction accuracy.
More informationDrilling Performance Services
Drilling Performance Services Helping Our Customers Achieve Superior Results REVit DAMPED Oscillation REVit PROVEN REAL-TIME STICK SLIP MITIGATION *Torque and RPM are out of phase and cancel one another
More informationComputer Vision Group Prof. Daniel Cremers. 6. Boosting
Prof. Daniel Cremers 6. Boosting Repetition: Regression We start with a set of basis functions (x) =( 0 (x), 1(x),..., M 1(x)) x 2 í d The goal is to fit a model into the data y(x, w) =w T (x) To do this,
More informationThe Basics of Decision Trees
Tree-based Methods Here we describe tree-based methods for regression and classification. These involve stratifying or segmenting the predictor space into a number of simple regions. Since the set of splitting
More informationElemental Set Methods. David Banks Duke University
Elemental Set Methods David Banks Duke University 1 1. Introduction Data mining deals with complex, high-dimensional data. This means that datasets often combine different kinds of structure. For example:
More informationRSM Split-Plot Designs & Diagnostics Solve Real-World Problems
RSM Split-Plot Designs & Diagnostics Solve Real-World Problems Shari Kraber Pat Whitcomb Martin Bezener Stat-Ease, Inc. Stat-Ease, Inc. Stat-Ease, Inc. 221 E. Hennepin Ave. 221 E. Hennepin Ave. 221 E.
More information5 Classifications of Accuracy and Standards
5 Classifications of Accuracy and Standards 5.1 Classifications of Accuracy All surveys performed by Caltrans or others on all Caltrans-involved transportation improvement projects shall be classified
More informationDirectional Drilling Optimization by Non-Rotating Stabilizer
Directional Drilling Optimization by Non-Rotating Stabilizer Eisa Noveiri, Adel Taheri Nia Abstract The Non-Rotating Adjustable Stabilizer / Directional Solution (NAS/DS) is the imitation of a mechanical
More informationData Preprocessing. Slides by: Shree Jaswal
Data Preprocessing Slides by: Shree Jaswal Topics to be covered Why Preprocessing? Data Cleaning; Data Integration; Data Reduction: Attribute subset selection, Histograms, Clustering and Sampling; Data
More informationFrom Building Better Models with JMP Pro. Full book available for purchase here.
From Building Better Models with JMP Pro. Full book available for purchase here. Contents Acknowledgments... ix About This Book... xi About These Authors... xiii Part 1 Introduction... 1 Chapter 1 Introduction...
More informationPredict Outcomes and Reveal Relationships in Categorical Data
PASW Categories 18 Specifications Predict Outcomes and Reveal Relationships in Categorical Data Unleash the full potential of your data through predictive analysis, statistical learning, perceptual mapping,
More informationData Rules. rules.ppdm.org. Dave Fisher & Madelyn Bell BUSINESS RULES WORKSHOP. March Business rules workshop Nov 2013
BUSINESS RULES WORKSHOP Data Rules Dave Fisher & Madelyn Bell rules.ppdm.org 1 March 2014 AGENDA Workshop objectives Definitions what is a data rule? It only grows with your help It takes longer then anyone
More informationPRESENTATION TITLE. Uncertainty - an introductory discussion Robert Wylie. Industry Steering Committee on Wellbore Survey Accuracy
1 PRESENTATION TITLE Tool Wellbore Face Positioning Efficiency, Technical Virtual SectionSurveys, and Positional Uncertainty - an introductory discussion Speaker Information President xndrilling, Inc.
More informationDecision Trees Dr. G. Bharadwaja Kumar VIT Chennai
Decision Trees Decision Tree Decision Trees (DTs) are a nonparametric supervised learning method used for classification and regression. The goal is to create a model that predicts the value of a target
More informationNominal Data. May not have a numerical representation Distance measures might not make sense. PR and ANN
NonMetric Data Nominal Data So far we consider patterns to be represented by feature vectors of real or integer values Easy to come up with a distance (similarity) measure by using a variety of mathematical
More information3D modeling of the Quest Projects Geophysical Datasets. Nigel Phillips
3D modeling of the Quest Projects Geophysical Datasets Nigel Phillips Advanced Geophysical Interpretation Centre Undercover Exploration workshop KEG-25 April 2012 Mineral Physical Properties: density sus.
More informationLecture 7: Decision Trees
Lecture 7: Decision Trees Instructor: Outline 1 Geometric Perspective of Classification 2 Decision Trees Geometric Perspective of Classification Perspective of Classification Algorithmic Geometric Probabilistic...
More informationText Categorization. Foundations of Statistic Natural Language Processing The MIT Press1999
Text Categorization Foundations of Statistic Natural Language Processing The MIT Press1999 Outline Introduction Decision Trees Maximum Entropy Modeling (optional) Perceptrons K Nearest Neighbor Classification
More informationSlides for Data Mining by I. H. Witten and E. Frank
Slides for Data Mining by I. H. Witten and E. Frank 7 Engineering the input and output Attribute selection Scheme-independent, scheme-specific Attribute discretization Unsupervised, supervised, error-
More information8. Tree-based approaches
Foundations of Machine Learning École Centrale Paris Fall 2015 8. Tree-based approaches Chloé-Agathe Azencott Centre for Computational Biology, Mines ParisTech chloe agathe.azencott@mines paristech.fr
More informationBIOL 458 BIOMETRY Lab 10 - Multiple Regression
BIOL 458 BIOMETRY Lab 0 - Multiple Regression Many problems in biology science involve the analysis of multivariate data sets. For data sets in which there is a single continuous dependent variable, but
More informationModeling Plant Succession with Markov Matrices
Modeling Plant Succession with Markov Matrices 1 Modeling Plant Succession with Markov Matrices Concluding Paper Undergraduate Biology and Math Training Program New Jersey Institute of Technology Catherine
More informationDecision Trees Oct
Decision Trees Oct - 7-2009 Previously We learned two different classifiers Perceptron: LTU KNN: complex decision boundary If you are a novice in this field, given a classification application, are these
More informationPattern Recognition. Kjell Elenius. Speech, Music and Hearing KTH. March 29, 2007 Speech recognition
Pattern Recognition Kjell Elenius Speech, Music and Hearing KTH March 29, 2007 Speech recognition 2007 1 Ch 4. Pattern Recognition 1(3) Bayes Decision Theory Minimum-Error-Rate Decision Rules Discriminant
More informationSLiP. Source Line Processing
SLiP Source Line Processing What is SLiP SLiP (Source Line Processing) is Windows based software for computing source positions. SLiP has been specifically written for OBC type seismic surveys. It deals
More informationOutrun Your Competition With SAS In-Memory Analytics Sascha Schubert Global Technology Practice, SAS
Outrun Your Competition With SAS In-Memory Analytics Sascha Schubert Global Technology Practice, SAS Topics AGENDA Challenges with Big Data Analytics How SAS can help you to minimize time to value with
More informationHighlight Mode and Text Mode In TVD Mode, the Mode menu at the top of the window contains three items: Text Mode, Highlight Mode, and Toggle TVD Mode.
Overview HMG Strata Log has a TVD Mode, which converts a regular measured depth log into a true vertical depth log using the available survey data. Moreover, it opens three separately scrollable graph
More informationDoc #: IDI06-11F Rev: 1.3 Issued: 22/02/18. Well Seeker PRO How To Guide Rev 1.3. Page 1 of 26
Well Seeker PRO How To Guide Rev 1.3 Page 1 of 26 Contents 1.0 - Getting Started... 4 1.1 - Display... 4 2.0 - Creating a new Well... 5 2.1 - Unit Selection... 5 2.2 - New Instant Plan / Survey... 6 2.3
More informationMinitab 17 commands Prepared by Jeffrey S. Simonoff
Minitab 17 commands Prepared by Jeffrey S. Simonoff Data entry and manipulation To enter data by hand, click on the Worksheet window, and enter the values in as you would in any spreadsheet. To then save
More informationAADE-05-NTCE-67. A Gravity-Based Measurement-While-Drilling Technique Determines Borehole Azimuth From Toolface and Inclination Measurements
AADE-05-NTCE-67 A Gravity-Based Measurement-While-Drilling Technique Determines Borehole Azimuth From Toolface and Inclination Measurements Herbert Illfelder, Ken Hamlin, Graham McElhinney - PathFinder
More informationDIRECTIONAL WELL TRAJECTORY DESIGN: THE THEORITICAL DEVELOPMENT OF AZIMUTH BENDS AND TURNS IN COMPLEX WELL TRAJECTORY DESIGNS
Nigerian Journal of Technology NIJOTECH) ol. 35, No. 4, October 2016, pp. 831 840 Copyright Faculty of Engineering, University of Nigeria, Nsukka, Print ISSN: 0331-8443, Electronic ISSN: 2467-8821 www.nijotech.com
More informationFaculty of Science and Technology MASTER S THESIS
Faculty of Science and Technology MASTER S THESIS Study program/ Specialization: Petroleum technology Drilling Technology Spring semester, 2011 Writer: Knut Tveitan Faculty supervisor: Eirik Kårstad Seyed
More informationDrillstring Magnetic Interference
Drillstring Magnetic Interference ISCWSA MWD Error Model Terms Harry Wilson ISCWSA 38, New Orleans, 3 October 2013 MWD Magnetic Azimuth Dominant error sources: Uncertainty associated with nominal declination
More informationApplying Supervised Learning
Applying Supervised Learning When to Consider Supervised Learning A supervised learning algorithm takes a known set of input data (the training set) and known responses to the data (output), and trains
More informationBuilding Better Parametric Cost Models
Building Better Parametric Cost Models Based on the PMI PMBOK Guide Fourth Edition 37 IPDI has been reviewed and approved as a provider of project management training by the Project Management Institute
More informationCS 229 Midterm Review
CS 229 Midterm Review Course Staff Fall 2018 11/2/2018 Outline Today: SVMs Kernels Tree Ensembles EM Algorithm / Mixture Models [ Focus on building intuition, less so on solving specific problems. Ask
More informationPractical Guidance for Machine Learning Applications
Practical Guidance for Machine Learning Applications Brett Wujek About the authors Material from SGF Paper SAS2360-2016 Brett Wujek Senior Data Scientist, Advanced Analytics R&D ~20 years developing engineering
More informationNotes based on: Data Mining for Business Intelligence
Chapter 9 Classification and Regression Trees Roger Bohn April 2017 Notes based on: Data Mining for Business Intelligence 1 Shmueli, Patel & Bruce 2 3 II. Results and Interpretation There are 1183 auction
More informationRegression on SAT Scores of 374 High Schools and K-means on Clustering Schools
Regression on SAT Scores of 374 High Schools and K-means on Clustering Schools Abstract In this project, we study 374 public high schools in New York City. The project seeks to use regression techniques
More informationMachine Learning Duncan Anderson Managing Director, Willis Towers Watson
Machine Learning Duncan Anderson Managing Director, Willis Towers Watson 21 March 2018 GIRO 2016, Dublin - Response to machine learning Don t panic! We re doomed! 2 This is not all new Actuaries adopt
More informationMachine Learning Techniques for Data Mining
Machine Learning Techniques for Data Mining Eibe Frank University of Waikato New Zealand 10/25/2000 1 PART VII Moving on: Engineering the input and output 10/25/2000 2 Applying a learner is not all Already
More informationLecture 20: Bagging, Random Forests, Boosting
Lecture 20: Bagging, Random Forests, Boosting Reading: Chapter 8 STATS 202: Data mining and analysis November 13, 2017 1 / 17 Classification and Regression trees, in a nut shell Grow the tree by recursively
More informationModel Based Impact Location Estimation Using Machine Learning Techniques
Model Based Impact Location Estimation Using Machine Learning Techniques 1. Introduction Impacts on composite structures result in invisible damages that need to be detected and corrected before they lead
More informationGetting more from your Engineering Data. John Chapman Regional Technical Manager
Getting more from your Engineering Data John Chapman Regional Technical Manager 2012 HALLIBURTON. ALL RIGHTS RESERVED. Getting more from your Engineering Data? extracting information from data to make
More informationMATH 1112 Trigonometry Final Exam Review
MATH 1112 Trigonometry Final Exam Review 1. Convert 105 to exact radian measure. 2. Convert 2 to radian measure to the nearest hundredth of a radian. 3. Find the length of the arc that subtends an central
More informationLouis Fourrier Fabien Gaie Thomas Rolf
CS 229 Stay Alert! The Ford Challenge Louis Fourrier Fabien Gaie Thomas Rolf Louis Fourrier Fabien Gaie Thomas Rolf 1. Problem description a. Goal Our final project is a recent Kaggle competition submitted
More informationChapter 2: Modeling Distributions of Data
Chapter 2: Modeling Distributions of Data Section 2.2 The Practice of Statistics, 4 th edition - For AP* STARNES, YATES, MOORE Chapter 2 Modeling Distributions of Data 2.1 Describing Location in a Distribution
More information3D S wave statics from direct VSP arrivals Michael J. O'Brien, Allied Geophysics, and Paritosh Singh, Colorado School of Mines
3D S wave statics from direct VSP arrivals Michael J. O'Brien, Allied Geophysics, and Paritosh Singh, Colorado School of Mines Summary We observe that direct S wave arrivals in 3D VSP data contain the
More informationPhysics 101, Lab 1: LINEAR KINEMATICS PREDICTION SHEET
Physics 101, Lab 1: LINEAR KINEMATICS PREDICTION SHEET After reading through the Introduction, Purpose and Principles sections of the lab manual (and skimming through the procedures), answer the following
More informationIntroduction to Classification & Regression Trees
Introduction to Classification & Regression Trees ISLR Chapter 8 vember 8, 2017 Classification and Regression Trees Carseat data from ISLR package Classification and Regression Trees Carseat data from
More informationCART. Classification and Regression Trees. Rebecka Jörnsten. Mathematical Sciences University of Gothenburg and Chalmers University of Technology
CART Classification and Regression Trees Rebecka Jörnsten Mathematical Sciences University of Gothenburg and Chalmers University of Technology CART CART stands for Classification And Regression Trees.
More informationPredictive Modeling with SAS Enterprise Miner
Predictive Modeling with SAS Enterprise Miner Practical Solutions for Business Applications Second Edition Kattamuri S. Sarma, PhD From Predictive Modeling with SAS Enterprise Miner TM, From Predictive
More informationUsing the DATAMINE Program
6 Using the DATAMINE Program 304 Using the DATAMINE Program This chapter serves as a user s manual for the DATAMINE program, which demonstrates the algorithms presented in this book. Each menu selection
More informationMASW Horizontal Resolution in 2D Shear-Velocity (Vs) Mapping
MASW Horizontal Resolution in 2D Shear-Velocity (Vs) Mapping by Choon B. Park Kansas Geological Survey University of Kansas 1930 Constant Avenue, Campus West Lawrence, Kansas 66047-3726 Tel: 785-864-2162
More informationMagnetic Referencing and Survey Accuracy for Horizontal Development in the Permian
Slide 1 of 35 Magnetic Referencing and Survey Accuracy for Horizontal Development in the Permian Wellbore position uncertainty Global and local geomagnetic models Quality control and advanced corrections
More informationAllstate Insurance Claims Severity: A Machine Learning Approach
Allstate Insurance Claims Severity: A Machine Learning Approach Rajeeva Gaur SUNet ID: rajeevag Jeff Pickelman SUNet ID: pattern Hongyi Wang SUNet ID: hongyiw I. INTRODUCTION The insurance industry has
More informationGLM II. Basic Modeling Strategy CAS Ratemaking and Product Management Seminar by Paul Bailey. March 10, 2015
GLM II Basic Modeling Strategy 2015 CAS Ratemaking and Product Management Seminar by Paul Bailey March 10, 2015 Building predictive models is a multi-step process Set project goals and review background
More informationCPSC 340: Machine Learning and Data Mining. Logistic Regression Fall 2016
CPSC 340: Machine Learning and Data Mining Logistic Regression Fall 2016 Admin Assignment 1: Marks visible on UBC Connect. Assignment 2: Solution posted after class. Assignment 3: Due Wednesday (at any
More informationWorkshop 8: Model selection
Workshop 8: Model selection Selecting among candidate models requires a criterion for evaluating and comparing models, and a strategy for searching the possibilities. In this workshop we will explore some
More informationMore on Neural Networks. Read Chapter 5 in the text by Bishop, except omit Sections 5.3.3, 5.3.4, 5.4, 5.5.4, 5.5.5, 5.5.6, 5.5.7, and 5.
More on Neural Networks Read Chapter 5 in the text by Bishop, except omit Sections 5.3.3, 5.3.4, 5.4, 5.5.4, 5.5.5, 5.5.6, 5.5.7, and 5.6 Recall the MLP Training Example From Last Lecture log likelihood
More informationModel Assessment and Selection. Reference: The Elements of Statistical Learning, by T. Hastie, R. Tibshirani, J. Friedman, Springer
Model Assessment and Selection Reference: The Elements of Statistical Learning, by T. Hastie, R. Tibshirani, J. Friedman, Springer 1 Model Training data Testing data Model Testing error rate Training error
More informationTHE L.L. THURSTONE PSYCHOMETRIC LABORATORY UNIVERSITY OF NORTH CAROLINA. Forrest W. Young & Carla M. Bann
Forrest W. Young & Carla M. Bann THE L.L. THURSTONE PSYCHOMETRIC LABORATORY UNIVERSITY OF NORTH CAROLINA CB 3270 DAVIE HALL, CHAPEL HILL N.C., USA 27599-3270 VISUAL STATISTICS PROJECT WWW.VISUALSTATS.ORG
More informationMultiple Regression White paper
+44 (0) 333 666 7366 Multiple Regression White paper A tool to determine the impact in analysing the effectiveness of advertising spend. Multiple Regression In order to establish if the advertising mechanisms
More informationChapter 2 - Getting Started
Chapter 2 - Getting Started Entering Survey Data The heart of this program is the survey data and the ability for this program to compute the derived survey parameters. Start the XSection Horizontal Log
More informationNonparametric Classification Methods
Nonparametric Classification Methods We now examine some modern, computationally intensive methods for regression and classification. Recall that the LDA approach constructs a line (or plane or hyperplane)
More informationEnsemble-based decision making for reservoir management present and future outlook. TPD R&T ST MSU DYN and FMU team
Ensemble-based decision making for reservoir management present and future outlook TPD R&T ST MSU DYN and FMU team 11-05-2017 The core Ensemble based Closed Loop Reservoir Management (CLOREM) New paradigm
More informationCurriculum Map for Accelerated Probability, Statistics, Trigonometry
Curriculum Map for Accelerated Probability, Statistics, Trigonometry Statistics Chapter Two September / October Targeted Standard(s): N-Q.1, N-Q.2, N-Q.3, S-ID.1, S-ID.2, S-ID.3, S-IC.1, S-IC.2, S-IC.3,
More informationAn Interactive GUI Front-End for a Credit Scoring Modeling System by Jeffrey Morrison, Futian Shi, and Timothy Lee
An Interactive GUI Front-End for a Credit Scoring Modeling System by Jeffrey Morrison, Futian Shi, and Timothy Lee Abstract The need for statistical modeling has been on the rise in recent years. Banks,
More informationGEOG 4110/5100 Advanced Remote Sensing Lecture 4
GEOG 4110/5100 Advanced Remote Sensing Lecture 4 Geometric Distortion Relevant Reading: Richards, Sections 2.11-2.17 Geometric Distortion Geometric Distortion: Errors in image geometry, (location, dimensions,
More information