Kuparuk Gas Lift Optimizer February 2006 Kenneth Lloyd Martin - - PowerPoint PPT Presentation

Kuparuk Gas Lift Optimizer February 2006 Kenneth Lloyd Martin

• Kuparuk Production/ Gas Lift System
• Development of Model
• Operation
• Results

Kuparuk Kuparuk

Kuparuk Kuparuk Gas System Gas System

Drill Site Manifold

Gas Lift Compressors Gas Injection Compressors (MI)

1400 psi Separation

MI to injection wells at 3900 psi

Lift Gas to Wells

NGL’s

Central Production Facility

CPF3 Production Lines CPF3 Production Lines

Drill Sites and Wells Drill Sites and Wells

• 6 Drill Sites with automated gas lift chokes
• 9 Drill Sites with manual gas lift chokes
• About 160 gas lifted wells
• 1 ESP
• 4 Reverse-Flow Jet Pumps with water as

power fluid

• 2 Naturally Flowing

Existing Method for Gas Lift Existing Method for Gas Lift Optimization Optimization

• Equal-Slope (IGOR) Method based on KWPS

“Rate Tables” (performance curves)

• Single target IGOR for all wells at each facility

(or by DS if automated)

• Board operator reviews “space”, changes

target IGOR on DS’s

• Trial and Error on rate using SCADA executed

calculation

See 2003 ASME GL Kuparuk Presentation (Martin, Nations)

Rate Table Curves (PC Rate Table Curves (PC’ ’s) s)

Equal Equal-

• Slope Gas Lift Optimization

Slope Gas Lift Optimization (and allocation) (and allocation)

Project Scope Project Scope

• Build production gathering network model
• Include common lines & history match ∆P
• Incorporate well performance curves from KWPS
• Implement optimization to maximize oil rate
• Integrate network models with SCADA (SetCim) for
• n-line use
• Add temporary pressure instrumentation for tuning

Project Goal Project Goal

• To develop a near-real time system which

will increase Kuparuk production by modeling the surface hydraulics, which will in turn allow improved optimization of the current facilities.

• Provide a planning tool for maintenance,

debottlenecking, and expansion.

Project Plan Project Plan

• Develop single Drill Site model as test of

hydraulics and for vendor/company understanding of model

• Develop single Facility (CPF3) model and

test for results

• Develop CPF1 and CPF2 models and

planning tools

Project Execution Project Execution

• RFP

– APA Petroleum Consultants

• Petroleum Experts Software

– GAP for Optimizer – OpenServer for data transfer – In-house KWPS for performance curves

– Glacier Services

• In-house developments on SCADA

Operator Interface Operator Interface

Temperature Impacts Temperature Impacts

20F 20F 15 MMSCF 15 MMSCF

Optimization Run Optimization Run

• Constraints

– Water, Oil, Formation Gas, Total Gas – Individual well minimum/maximums

• Objective

– Maximum Oil Rate

Setcim Setcim Engineer Engineer’ ’s COE s COE Machine Machine KIOS. KIOS. Office Office /GAP /GAP Principal Server Principal Server FTP Server FTP Server CPF3 Model Server CPF3 Model Server KIOS.MS KIOS.MS / GAP / GAP KIOS. KIOS. PS PS UDP UDP Control Room Control Room

Comparison of Upstream Pressures Comparison of Upstream Pressures

Planning Interface Planning Interface

Model Model Maint

• Maint. / Engineering Interface

. / Engineering Interface

Mapping Equipment Mapping Equipment

Well Data Well Data

Results Results

• Expectations
• Benefits
• Challenges

Kuparuk INM Benefit Predictions Kuparuk INM Benefit Predictions

• Vendor Predictions 8% to 15%
• Gas lift production increases achieved in other

automated fields using integrated models 1% to 3%

• Potential GKA Target:
• Internal predictions: 500-4000 BOPD increase

(0.25%-2.0%)

• CPF3 detailed project estimate 200 BOPD (0.4%)

Initial Install Benefit Initial Install Benefit

205 BOPD Predicted Benefit 205 BOPD Predicted Benefit

Oil Rate

30000 32000 34000 36000 38000 40000 42000 44000 46000 48000 50000 20-May 3-Jun 17-Jun 1-Jul 15-Jul 10 20 30 40 50 60 70 80 GASOPT KINM Temperature 24 per. Mov. Avg. (Temperature) Linear (KINM) Linear (GASOPT) SD-TAPS Proration-WW

First Install First Install

Going Going “ “to model to model” ” after 10 days off after 10 days off

165 BOPD 165 BOPD

Compression Utilization Compression Utilization

Other Benefits Other Benefits

• Identified well with high velocity through flowline,

likely erosion concerns

• Identified at least 10 wells and headers with

inaccurate instrumentation

• Provided estimates of impacts for new well drilling
• Used with Prosper for downhole choke designs
• Provided estimates of impacts of 3rd party facility

use

• Developed better visualization tool for operators.
• Identified loss in compressor rates due to forest

fire smoke

• Difficult mix: troubleshooting vs. confidence
• 90/10 rule is no joke

– 6 months from approval for CPF3 to initial install – Still debugging 8 months later!

Lessons Learned Lessons Learned

Project Difficulties Project Difficulties

• Evaluating Rate Change

– Change is smaller than hourly variation in field rate. – Step changes only work when going “to model”

• Unable to keep model tuned with existing

instrumentation – Drifts of up to 20 psi during periods of low gas rates – Requiring large investment in additional instrumentation

• Rapid optimization software version changes

– Significant time required to review version for “bugs” – Revisions are not all “backward-compatible”

Existing Challenges Existing Challenges

• Computing time is issue during rapid

temperature changes

• Sensitive to small errors in PC’s
• At low total gas rates, ignores minimum LG

requirements

• In new version, either minimum lift rate

constraint or allow SI not both

• Errors are difficult to trace because code is

proprietary so don’t know what’s “under the hood”

Major Expenditures Major Expenditures

• Optimization Developers
• Automation Developers
• Project Engineering
• Software
• Computers
• Pressure Gauges

Annual O&M Costs Annual O&M Costs

• Software Licenses
• Automation Maintenance

1 month/yr

• Optimizer Maintenance

1 month/yr

• Engineering Model Tuning

1 day/mo/CPF

• Engineering Well Models *

1 hr/day/CPF

• Instrumentation

4 mh/tr/3yr * Existing Cost

Successful? Successful?

• Met Benefit Estimate
• I/O with SCADA working well
• Better Operator Visualization

– Operator focus on compressor rates, in addition to “gas handling available”

• Long development-Ongoing troubleshooting
• Operator acceptance

Acknowledgements Acknowledgements

• CPF3 Production Staff
• Network Model User’s Group
• COP Knowledge Sharing Network Members
• APA Petroleum Engineering, Calgary and

Dallas

• Glacier Services, Joe Griffo/Alex Charyna
• Others at CPAI and BPXA