Post on 03-Aug-2018
Seite 1
Predictive Emission Monitoring (PEM):Suitability and Application in View of U.S. EPAand European Regulatory Frameworks
Thomas EisenmannDr. Roland Bianchin
David Triebel
CEM 2014 - Istanbul, 14th May 2014
PEMS provide a reasonable alternative to CEMS (AMS) where there is
a reliable and predictable correlation between plant operating
conditions and emissions.
Predictive Emission Monitoring Systems (PEMS) are used for
continuous monitoring of emissions at stationary sources as an
alternative and backup for ‘Continuous Emission Monitoring
Systems’ (CEMS, in Europe usually AMS - Automated Monitoring
System).
PEMS define the relationship between a number of characteristic
process parameters of an emission source and the corresponding
emission concentration.
Introduction
This is generally applicable to gas and oil
fired plants, e.g. Boilers
Turbines
Reciprocal Internal Combustion Engines
(RICE)
Heaters
Furnaces
Ships
Just to name a few
Introduction
Introduction
Control Building
Fuels: Gas, Oil etc.
Fuel FlowEmission
Source
NOx, CO, O2
etc.
DCS
DAHS plus Emissions Model
Inputs (read only)
Data Interface by bus
systems (Modbus,
Profibus, OPC etc.)
Control
D-DAS 2010 D-PEMS
DAHS: Data Acquisition &
Handling System
DCS: Distributed Control System
CEMS as well as PEMS in the USA and countries
following U.S. Environmental Protection Agency (EPA)
regulations are mainly governed by
40 C(ode)F(ederal)R(egulation) Part 60 (Part 63) and
40 CFR Part 75.
PEMS as Alternative Monitoring Method (AMS) is
specifically addressed in
Part 60 Performance Specification PS-16 and
Part 75 Subpart E Alternative Monitoring Methods.
Regulatory Framework – U.S. EPA
New Source Performance Standard - NSPS,
promulgated first 1971
US EPA Part 60
Subparts for each type of source with e.g. subpart D covering boilers, GG
covering stationary gas turbines and J Petroleum Refineries
Requires Continuous Monitoring of Primary Pollutants (NOx, SO2, CO,
Opacity and VOC)
Part 60 is flanked by a series of test methods for emission measurements
and 13 (17) Performance Specifications
Industrial Units >100 mmBTU (about 29 MW), in some case also to smaller
sources (e.g. Subpart Dc covering small industrial boilers)
Performance Specifications
www.epa.gov/ttn/emc
• Performance specifications are used for evaluating the
acceptability of CEMS (and PEMS) at the time of or soon after
installation and whenever specified in the regulations;
• Quality assurance procedures in 40 CFR Part 60 Appendix F
are used to evaluate the effectiveness of quality control (QC)
and quality assurance (QA) procedures and the quality of data
produced by any CEMS / PEMS.
Continuous Emission Monitoring, originally published in 1993
US EPA Part 75
Continuous Emission Monitoring (CEM) and reporting requirements under EPA’s
Acid Rain Program (ARP), instituted in 1990 under Title IV of the Clean Air Act
ARP regulates electricity generating units (EGUs) that burn fossil fuels and that
serve a generator > 25 megawatts
It requires continuous monitoring and reporting of SO2 mass emissions, CO2 mass
emissions, NOx emission rate, and heat input
Included is a cap and trade program as market-based approach to reducing
emissions
Emissions data must be reported for each operating hour, without exception
Each CEMS or PEMS must be equipped with an automated DAHS
Regulatory Framework – Europe
PEMS in compliance with
prEN 264153 (WG 37) “PEMS Standard”
IED 2010/75/EU
The Large Combustion Plant Directive LCPD (2001/80/EC)
EN 15267. Applicability of minimum requirements as per Part 3 needs to
be reviewed.
EN 14181. Applicability of QAL1?
The Waste Incineration Directive WID (2000/76/EC)
CEN / TC 264 WG 37 “PEMS Standard”
• CEN (European Committee for Standardization) established
within the Technical Committee (TC) 264 „Air Quality“ the
working group WG37
• Main stimulus came from the Netherlands. They use PEMS in
particular for refinery / petrochemical applications
• The normative work started of from a Dutch draft and is
entitled
prEN 264153 Predictive Emission Monitoring
Systems (PEMS) – Applicability Execution and
Quality Assurance• Legal basis is the EU directive 2010-75-EU for industrial
emissions
• Four Task Groups:
Building a PEMS
QA of PEMS
Validation of a PEMS
Generic example of execution and QA of a PEMS.
• WG37 standard must accommodate the one-of-a-kind
uniqueness of PEMS for a specific plant;
• This is a particular challenge for dealing with EN15267 Part 3
and the array of questions concerning type tests;
• Detailed suitability of QAL2 test schemes of EN14181 will be
investigated
CEN / TC 264 WG 37 “PEMS Standard”
U.S. EPA
Part 60 (PS-16) as well as
Part 75 (Subpart E)
focus primarily on a
certification for a specific plant (initial RATA).
VGB PEMS Project
R&D project funded by VGB Power Tech, the European technical association for
power and heat generation:
“Evaluation of the application of Predictive Emissions
Monitoring Systems (PEMS) in Europe taking into account the
boundary conditions of the Standards EN 14181 and EN 15267“
Conducted at two combined cycle power plants in Germany
Major objective was to compare PEMS with plant CEMS and the results of a
QAL2 conducted by an independent, accredited stack tester
Lessons learned gave good insight, where improvements can be achieved and
what needs to be observed to successfully conduct the PEMS implementation
process
A detailed assessment of performance criteria in EN15267 Part 3 with respect to
their applicability for PEMS can be found in the final report of TUEV Rheinland
Some Other PEMS Relevant Activities
Another working group, WG 9 “Quality assurance of automated
measuring systems (Amendment to EN14181:2004)” within TC 264
is related to WG 37. It is working on a draft standard for data
acquisition and handling systems (DAHS).
To address communication and digital interfaces, a quite significant
issue also for PEMS and their integration in plant-wide IT networks,
the Association of German Engineers VDI has published a series of
guideline documents as VDI 4201, Part1-4
Practical Minimum Performance Requirements I
Units: Any unit wanted by the operator or stipulated by environmental
regulation (ppm, mg/Nm3, g/GJ, etc.).
Best model selectable and application specific to provide the best fit
to a plant process and the most economical way.
Components: NOx / O2 / CO / SO2 / CO2 / NH3 / HC / VOC / HCHO /
others as needed. Model parameters individual per component
Ranges only limited by used analyzer for the training data set. Model
building incorporating data from multiple analyzers and multiple
ranges. Ranges fully adjustable.
Sensor Validation System: Determination of defective and erroneous
process sensors and inputs. PS-16 requires it daily, recommended is
once per minute. Redundant sensors and analysis of reliable inputs.
Accuracy and precision as per PS-16 /Subpart E / CEMS standard
Response time: 99,9 % less than 1 sec.
Executable on standard server and workstation hardware
Visualization (graphic user interface - GUI) By means of a data
acquisition systems, but also own (simple graphic representation)
Prevention of data manipulation and unauthorized reproduction of
licenses, e.g. log file recording immutably all modification of software
or model, model encryption, new revision number, license key etc.
All standard interfaces, e.g. OPC, ModBus, Profibus etc.
Data availability 99 % +
Measurement frequency 1 sec.
Practical Minimum Performance Requirements II
Quality Assurance
PEMS is an Analyzer!
Measure Frequency
Sensor Validation Once per minute, minimum daily
Relative Accuracy Audit RAA Quarterly / Monthly
Relative Accuracy Test Audit RATA Annual
EN 14181 QAL1 Uncertainty: PEMS Model Building
EN 14181 QAL 2 Initial calibration
EN 14181 QAL 3 Periodic checks at regular intervals
EN 14181AST Annual
Initial
certification
Sensor
Validation
Quarterly
RAAs
Subsequent
RAAsYearly RATA
Yearly Relative Accuracy
Test Audit (RATA):
Minimum 9-runs at the
normal operating level.
According PS-16
(continual
compliance):
Minimum 27-run
(9 runs at three
operating levels
of “the key
parameter that
most affects
emissions”)
relative accuracy
test (RAA).
Minimum daily
check,
recommended
once per minute
or once per
monitoring
cycle.
RAAs in the first
year after initial
certification:
At least three 30-
minute portable
analyzer or
reference method
(RM)
determinations*.
All quarterly RAAs
passed in the first
year and the RATA
in the second year:
One may elect to
perform a single
mid-year RAA in
the second year.
Quality Assurance EPA Part 60
*The average of the portable analyzer or Reference Method
determinations must not differ from the PEMS average value
by more than 10 percent or the test is failed
Step 3
Step 1
Step 2
Step 4
Step 5
Quality Assurance EPA Part 75
Subpart E
Approval
720 operating hour
demonstration that
the AMS has same
or better precision,
reliability,
accessibility, and
timeliness (PRAT) as
a CEMS
Alternative Monitoring System (AMS)
Compliance
Data Availability
Valid paired data for
at least 90 % of the
minimum 720
operating hours
Reliability
Valid one-hour
averages for
minimum 95.0 % of
operating hours for
one year
Sensor Validation
System
shall include an alarm
to inform the operator
when sensors need
repair and when the
PEMS is out-of-
control. Sensor failure
test.
RAA & RATA
Semi-annual or annual
RATA
3-run (minimum) RAAs
in every month in
which the unit operates
for at least 56 hours
(may be only for ozone
season May-
September or waived
for a three-month RAA)
Normal operating
levels
Summary and Comparison PEMS – CEMS
Common
FeaturesCEMS / PEMS
ContinuousBoth methods can be used for continuous emissions
monitoring.
Plant
TypesFor all oil- and gas-fired sources.
Accuracy /
Precision
Accuracy and precision are comparable provided that the
same quality assurance is applied.
Quality
assurance
Securing data quality with procedures of EN14181 /
requirements of EN 15267 (EU) as well as Part 60 / Part 75
RATA / RAA (USA)
Data
Acquisition
For data representation and reporting of monitoring results,
use of data acquisition and handling systems.
Summary and Comparison PEMS – CEMS
Differences CEMS PEMS
Hardware
• Gas Analyzers
• Accessories like probes, heated
lines, racks, shelters etc. needed
• Standard server hardware with means
for data back-up and securing data
integrity
Application
CEMS more universally applicable:
• Plants fired with variable solid
fuels
• Components like particulate
matter and Hg
• Basically not suitable for solid,
moisture absorbing fuels
• Not applicable at e.g. waste
incinerators
• Restricted for coal-fired plants
Cost
• Capital cost: Approximately 50 % of a comparable CEMS. In case of model
transferability or for ex-proof areas, cost difference may even be much higher
• Operations and maintenance: Approximately 10-20 % of CEMS cost
• Quality assurance: No cost difference
Quality
assurance
• EU: Type approval
• U.S. Daily Zero and Span Check
(Part 60)
• EU: EN15267-3 may not be reasonably
applicable due to the plant specific, one-
of-a-kind characteristic of PEMS
• U.S. Sensor Validation System (PS-16,
Subpart E). Resilience to input failures.
Seite 22
Predictive Emission Monitoring (PEM):Suitability and Application in View ofU.S. EPA and European Regulatory Frameworks
CEM 2014 - Istanbul, 14th May 2014