Well-site Analysis Why?
Need to make timely decisions
Many decisions are based on fluid data
When accurate direct measurements are required in the order of days
or even hours, well-site analysis is required.
Well-site Analysis When?
To test or not to test?
confirm live fluid analyzer (MDT) contamination
Re-entry for additional samples
if contamination is unacceptable
if a compositional gradient is suspected
if the samples are not valid
When any H2S/CO2 is anticipated
rapid adsorption of H2S on sample cylinders
corrosion and gas economics
Monitor well test cleanup
Limited sample volume available
PVT Express
New hardware/software platform for
measured and predictive PVT
Designed for fast measurements
Provide reliable volumetric fluid property
predictions
Immediate PVT report delivery
Unit is rugged, portable, modular
PVT Express Modules
WFPM - HPHT fluid property measurement module
WGOR - Fluid GOR module, API, atmospheric viscosity
WFAC - Gas and liquid chromatographic analysis module
WFPI - Data acquisition and processing interface
PVT Expert - Fluid property prediction module
- PVT ANN, PVT-OBM
WFCAB - Customized 10x14 ft lab cabin (mercury free)
PVT Express Hardware
Gas Chromatograph
Liquid trap
Acquisition Box
pVT CELL
Gasometer
HP Pump module STO Viscometer
* Laptop inc. PVT Expert software not pictured.
* Densitometer not displayed as well.
Measured Fluid Properties
Saturation Pressure bubble point/dew point
Density psat, stock tank
Fluid compressibility
GOR heavy oil to lean gas
Composition C12+ gas, C36+ liquid
Viscosity stock tank (live oil option)
OBM contamination concentration
Predicted Fluid Properties
Black Oil PVT Study
Constant Composition Expansion
Relative volume, oil compressibility
Differential Liberation
Bo, Rs, oil density , API, viscosity
Liberated gas specific gravity and Z-factor
Separator test
Bo @ any specified separator p&T
Separator oil shrinkage 2 stage @ separator p&T
GOR separator, tank and total
Oil density separator and tank
Predicted Fluid Properties
Gas Condensate Study
CCE
Relative volume
V/Vsat @ p>pd and V/Vtotal @ p<pd
Gas compressibility
1 phase @ p> pd and 2 phase @ p< pd
Retrograde liquid deposit @ p< pd
CVD
Retrograde liquid deposit @ p< pd
Cumulative produced fluid as mol % of original
Produced gas Z factor
Produced gas specific gravity @ p< pd
Produced gas mole weight @ p< pd
WFPM
15 Ksi 180 C
Sample volume 25 cc
Pressure (+/- 10 psi)
process fluid/flush mount
Volume (+/- 0.05 cc)
LVDT
Temperature (+/- 0.2 C)
Bubble & dew point
fiber optic detection
magnetic coupled mixer
WFPM Cell
Cell Body
Sample Ports
Control Valves
Heating Jacket
Stirrer
Rodded Piston
End Cap
Bubble Point determination
PPP=P=P=10,000
15,000
=4,000
5,000
=500
15
0 psia
psia
psia
psia
psia
Bubble Point
Determination Gas
Evacuate Cell Low
Charge Sample
Start Mixer
Density
Expansion 1
Expansion 2
Bubble Point High
Liquid
2 Expansion
Abandonment
Dew Point
determination PP=P=P12,000
=6,000
=15 psia
15,000
8,000
7,000
5,000
2,500
0 psia
psia
psia
Dew Point
Determination Gas
Evacuate Cell Low
Charge Sample
Start Mixer
Density
Expansion 1
Expansion 2
Dew Point
2 Expansion High
Liquid
2 Expansion
Abandonment
WGOR
5000 +/- 1 cc gasometer
Counter balanced piston
position encoder
3 ft x 1 ft footprint
Accessories
liquid trap 25 +/- 0.02 cc
vacuum pump
gas sample port
j-type thermocouple
digital pressure transducer
WFAC
C12+ composition of gas GC
C36+ composition of liquid
GC
HID detector
For gas & liquid
No hydrogen
Rugged design
Small footprint
Chromatography Principle
blotter
paper
solvent
t0 t1 t2
Gas Phase Chromatography
Principle
oven
data
Injector
detector
column
Gas Chromatography Basics Components
Chromatogram Example
WFPI
Magnetic stirrer controls
Digital output
Cell pressure & temp
Set point temperature
Pump volume
LVDT position
110 or 240 Volt input Optical RI intensity
10 channel universal input Motor RPM & current
Serial port output to computer
PVT Lab vs PVT Express Data Comparison
Black Oil
400 371368
315 328
WFAS Data
300
LAB DATA
202197 202 206
200
100
42 43 50 49
7.86.7
0
C1 C12+ OBM MW STO MW GOR Psat bar
Mole % Mole% scf/stbbl
PVT Lab vs PVT Express Data Comparison
Gas Condensate
450
374 373
350 WFAS Data
LAB DATA
250
170 153
150
83 83
45 56
50 25 25
1.251.2
C1 C12+ MW STO MW GOR Dew Pt
Mole% Mole% bbl/mmcf bar
PVT Expert
An artificial neural network (ANN), also called a simulated
neural network (SNN) or just a neural network (NN), is an
interconnected group of artificial neurons that uses a
mathematical or computational model for information
processing based on a connectionist approach to computation.
An ANN can be defined as a multi-dimensional function
including a large number of parameters which relates input and
output data.
PVT Expert
Artificial Neural Network model (ANN)
ANN utilises a large number of parameters that relate input and output
data
Tuning done during training over 1000 PVT studies
Accurately predicts complete set PVT properties
Operator independent
Quality assessment
Artificial Neural Network Principle
w1.1
x1 y1
w1.2
x2 y2
w1.n
Transfer function (w)
xn yn
Input layer Hidden layer Output Layer
ANN principle
ANN Features
Pros:
Give an advantage to the database owner.
Light weight software.
Operator independent.
Crash proof.
Quality assessment
Cons:
Not seen as true science.
Limited to the database range (no extrapolation).
Limited to the specific problem (input and output).
PVT Expert Prediction Accuracy
Black Oil
4.0
3.5
Predicted Bo @ pb (vol/vol)
3.0
+/- 5%
2.5
2.0
1.5
1.0
1.0 1.5 2.0 2.5 3.0 3.5 4.0
Reported Bo @ pb (vol/vol)
PVT Expert Prediction Accuracy
Gas Condensate
20
Predicted maximum condensation (% vol)
18
16
14
12
10
0
0 2 4 6 8 10 12 14 16 18 20
Reported maximum condensation (% vol)
PVT Express and Vx
ASD provides gas and oil samples.
Vx provides GVF
ASD provides means for physically recombining gas and oil sample.
PVT Express can provide:
The minimum on-site measurement for calibrating the BOM.
a full PVT and compositional analyses for Running PVT Pro and
generating the Vx Fluid ID.
