Air Emissions Inventory

Report
A Better Way to Estimate Emissions
from Oil and Gas Operations
Dr. Susan Stuver
Research Scientist
10/10/2013
Energy, Air & Water
A Comprehensive Study of the Issues of Today
UNIVERSITIES
HEALTH SCIENCES
AGENCIES
Problem Statement
Solution
Policy without Data
Conservative data assumptions leads to over
estimation – assume 100% engine load and
24/7 operation times
Measure Equipment Emissions
Engine load data and activity times
would make estimates more
realistic
Calculations Based on Outdated Standards
US Regulators use the USEPA AP-42 Standards
that assume worst possible emissions
Use Engine Emission Factors
Use EPA certified tiered engine
emission factor data from engine
manufacturers
3
Max Regional 8-hr Ozone
Impact derived from worst
case estimates
Wind Direction from Eagle Ford
into San Antonio Texas
4
CALCULATION – the old fashioned way…
So,
2,250 hp Pump truck deck engine and 12
pump trucks engines are used for the
hydraulic fracturing process
2,250 hp/pump truck engines x 12 pump
truck engines = 27,000 hp
ENOx = 2.4 x 10-02 lb NOx/hp-hr x 27,000 hp x 1
= 648 lb NOx/hr
Hey – who messed up the math?
ENOx = EF x HP x LF
Where,
ENOx = NOx Emissions (lb/hr)
EFNOx = NOx Emission Factor (lb NOx/hp-hr)No Emission Controls are assumed
HP = Total power output (hp)
LF = Load factor-100% assumed
5
Vehicles
Mobile Command Center
Crane (large) - Operating Engine
Pickup Trucks
Equipment
Light Tower (mobile)
Bulldozer
Backhoe
Generator - Cooling Room
Make/Model
SMT 503 Mobile Monitoring Unit
ATC3200
GMC 2500 HD
Make/Model
RL4000
TR95 w/Engine
420D
Model QAS25
Fuel
Diesel
Diesel
Gasoline
Rating
210 hp
517 hp
360 hp
Fuel
Diesel
Diesel
Diesel
Diesel
Number
1
1
2
Rating
13.6 hp
99 hp
88 hp
29.6 hp
Number
6
1
1
1
6
EMISSION GROUP RANKING SYSTEM
RANK 1 Largest Sources
- EX: Pump Trucks (Fracturing/Perfing)
RANK 2 Medium Sources
- EX: Fracturing Water Pumps
RANK 3 Small Sources
- EX: Light Carts
7
There are 7 of
them - Using
worst case can
cause huge
variation
8
GPSA Engineering Data Book (2004)6 and Carl Branan’s Process Engineer’s Pocket Handbook
(1984)7 provides the equation
BHp = Q x ∆P/1714(e)
Where,
BHp
Q
∆P
1714
(e)
= Required brake horsepower
= Fracturing fluid flowrate (gal/min)
= Pump discharge – Pump suction (inlet pressure) (psi)
= Conversion factor (convert to BHp)
= Pump efficiency = 90% (see Note)
Note 1 – Inlet pressure was not recorded so it is conservatively assumed to be zero (psi)
Note 2 - The combination of mechanical and volumetric efficiency is normally 90% or higher for
non-compressible fluids (GPSA Engineering Data Book, 2004)8
9
BHp = Q x ∆P/1714(e)
BHp = ((65 bbl/min x 42 gal/bbl x (6000 psi – 0
psi))/1714)/(0.90)=10,618 Bhp
LF= BHp and divide it by full horsepower. Full horsepower is
obtained by the engine HP times the number of engines.
LF = 10,618 Bhp/(2250 hp x 12 engines = 27,000 hp) = 39.2% while
the engine was under load.
10
2,250 hp Pump truck deck engine and 12
trucks
Epoll = EF x P x N x LFA
Where,
Epoll = Emissions for pollutant
EF = Emission factor in lb/hp-hr
P = Brake horsepower in hp
N = Number of units
LFA = Load factor Average (39% for 7.5hr
and 15% for 4.5hr)
So,
Nox = 108.5 lb Nox/hr
AP-42 factors with 100%
estimated load = 648 lb Nox/hr
that’s 539 pounds per hour
difference.
539 lb/hr x 12 hour day = 6,468
pounds (3.2 tons) per day for 1
frac job
ENOx = (1.34 x 10-02lb/hp-hr x 2250hp/eng x 12 eng) x 39% x 7.5 hr/day + 15% x 4.5 hr/day
12 hr/day
11
Lets compare the results
• Nox AP42 100% LF = 648 lb/hr
• Nox Tier Eng LF Avg = 108.5 lb/hr
• VOC AP42 100% LF = 19 lb/hr
• VOC Tier Eng LF Avg = 5.7 lb/hr
• CO AP42 100% LF = 149 lb/hr
• CO Tier Eng LF Avg = 20 lb/hr
• PM/PM10 100% LF = 19 lb/hr
• PM/PM10 LF Avg = 1.6 lb/hr
• 539 lb/hr
• 12 hour day = 6,468
pounds (3.2 tons) per
day for 1 frac job
• 2000 frac jobs = 6,400
tons of emissions
that are not really
there.
12
Tier IV engines would be much lower
Are actually measured in grams (not pounds) per hour
13
14
ENOx = EF x HP x LF
Using engine load is not realistic
Too much variance – need an
entirely new way to estimate
Conceivable to use
•
•
•
•
total fuel consumption,
brake specific fuel consumption
diesel density
tier rated emissions
to estimate emissions from a job
15
Assume a drilling operation uses 5,000 gallons of diesel per day for 5 days
And Assume we know the BSFC is 0.35 lb/bhp-hr
And Assume we know that diesel density is 7 lb/gallon
And Assume we know the engines are Tier 1 – 6.9 g/hp-hr for Nox
And we want to estimate NOx Emissions
Note:
numbers
aren't
actual
Calculating Hp-hr/job = (fuel for job, gallons) X 7 lb/gal = 500,000 Hp-hr/job
0.35 lb/bhp-hr
Tons = Hp-hr/job x tier rated emissions g/hp-hr x 1/454 g/lb x 1/2000 lb/ton
Nox Tons = 500,000 hp-hr/job x 6.9 g/hp-hr x 1/454 g/lb x 1/2000 lb/ton = 3.8 tons
This will be compared to the traditional method:
ENOx = EF from AP42 x total HP x 100% LF X 100% activity
And backed up with measurements to validate
16
No engine certification for
emission factors
Emission Reduction Potential
Not Recognized
engines are currently being
loaded into the regional air shed
model as diesel
Measurement of emission reduction is needed. We are securing
study sites to quantify the emissions reduced by switching to bifuels, dual fuels and /or NG
17
Remote Monitoring with Low Cost Sensors
18
Wireless Data Networking
19
Advanced Analytics for Fugitive Sampling
There's no protocol for the
use of relatively new
technology in oil and gas
operations
20
Water Screening Kits
Education and Outreach
Temporary Roads
Fugitive Emissions
Invasive Species
Adv. Analytics
Produced Water Trmt
Stray Gas
Livestock Impacts
Soil Impacts
Managing Misinformation
Wireless Data Loggers
Aggie
Pigeon
22

similar documents