CENORM_RickGould

7/29/2019 CENORM_RickGould

1/40

The Environment Agencys

Monitoring Certification Scheme(MCERTS)

- and recent CEN developments

Richard GouldMonitoring & Assessment Process

Environment Agency


2/40

Outline

z Introduction words and confusionz Backgroundz Legal driversz Monitoring and performance standardsz Alignment and future European schemez Existing model for verification and approval


3/40

Introduction

z Environment Agency for England and Walesz Europes biggest environmental regulatorz

Industrial air emissions, waste, waterz Flood defence, contaminated land, ecologyz Monitoring is essential for controlz Monitoring and assessment

Guidance and training on emissions monitoring Application of standards Compliance assessment and auditing

z

CEN and ISO standards for monitoring and qualityassurancez Monitoring Certification Scheme

Type testing and approval scheme for Automated Measuring

Systems (AMS)


4/40

Big words and confusion?

z Verificationz Certificationz Approvalz Type-testingz Validationz Evaluation

z All the words have the same practicalmeaning?


5/40

Proof of performance

It does what it says on the tin

Google 502,000 examples


6/40

Proof of performance Stage 1

Super battery charger Fully charged battery = very quick

Testing


7/40

Certification Stage 2

AssessManufacturingprocesses

Official

Approval


8/40

Background - Quality of Monitoring


9/40

Industrial emissions


10/40

Monitoring systems


11/40

A hazardous waste treatment site


12/40

Legal demands

z Laws to control pollutionz Industrial operators have permits

Permits = the law applied to the processz Processes have emission limit valuesz Processes must be controlledz Operators must monitor emissions

Independent checks too Compliance assessment

z So how do we know that the monitoring is any good?z Are the results real and valid?


13/40

Monitoring systems


14/40

Equipment


15/40

Monitoring systems


16/40

Monitoring systems


17/40

Provisions for monitoring


18/40

Reference tests


19/40

Legal drivers

z EU Directives - incineration & combustionz Automated measuring systems (AMS)z Uncertainty budgets (95% confidence

intervals) Accuracy and precision

z Use CEN, ISO, national/other internationalstandards and methods

Performance requirements within monitoringstandards


20/40

Accuracy and precision

AccuratePrecise

InaccuratePrecise(bias)

AccurateImprecise

(random errors)

Inaccurate

Imprecise


21/40

Margins of error

ELV

95% CI


22/40

Past problems

z AMS varied in quality Very good to very bad

z Variable accuracy and precisionz Poor reliabilityz Unreliable dataz AMS did not always do what they said on the

tin


23/40

Solution

z Apply performance requirements Reinforce with laws

z Different grades of performance requirement Higher standards for more demanding applications Divisions of requirements

z

AMS with proven performance Testing (Verification) Certification (Approval)

z

Requirements for all test laboratories Standards, accreditation

z Standards applied through MCERTS


24/40

Performance specifications

z Linearityz Repeatabilityz Cross-sensitivityz Response timez

Detection limitz Environmental conditions (controlled chamber)z Vibration (test rig)z Reproducibilityz Availability and maintenance intervalz

Zero and span driftz Comparison with a reference method in a field test


25/40

Zero and span

y

/ m g

/ m 3

0,0

10,0

20,0

30,0

40,0

50,0

0,0 10,0 20,0 30,0 40,0 50,0 x / mg/m 3

Drift ?

St d d t ti g d


26/40

Standards - testing andcertification

Issue Standard

Quality of testing ISO 17025

AMS performance testing ISO andCENstandards

Product certification EN 45011

Measurement uncertainty ISO 14956


27/40

Evaluation the MCERTS model

Laboratory tests 3 month field tests Certification

Surveillance

Manufacturer audit


28/40

Paired testing laboratory tests

z Artificial stackz Gas mixing systemsz Environmental controls

Field testing comparison with


29/40

Field testing comparison witha reference method

Check of emissions, andAMS-readings, using astandard referencemethod (SRM)

SRM yields a truereading (or as close aswe can get)

SRM -vs- AMS data


30/40

SRM vs AMS data- a good agreement

CEM and SRMdata

SRM -vs- AMS data - a good


31/40

gagreement

y = 0.9572x - 0.0598

R2 = 0.9989

-2

0

2

4

6

8

10

12

14

16

0 2 4 6 8 10 12 14 16 18

AMS

Plot of half-hourlyaverages for SRMversus AMS

SRM

SRM -vs- AMS data - a poor


32/40

SRM vs AMS data a poor agreement

AMSand SRMdata

Time


33/40

Surveillance - Design changes

z Stage 1 - type testing; verificationz Stage 2 - Certification; approvalz Stage 3 - Continuing surveillance

control of design changes

control of manufacturing reproducibility maintenance of performance in modified designs


34/40

MCERTS expansion


35/40

MCERTS expansion

z Model applied to: Continuous ambient monitoring systems (CAMs) Continuous water monitoring systems (CWMs)

z MCERTS testing schemes: Manual stack emissions monitoring Flow monitoring verification Chemical testing of soils


36/40

European developments

E


37/40

European context

z Germany type-approval scheme for over 25 years AMS for stack emissions and ambient air World leaders in testing and verification for AMS UK and Germany aligned their schemes in 2002 Mutual Recognition

z Italyz Francez CEN member states

Updating standard

Unified performance specifications Standards for testing AMS International framework for testing and certification CEN TC 264

CEN TC 264


38/40

CEN TC 264

Standard Scope

prEN 15267-1 Framework for testing(verification) and certification(approval)

prEN 15267-2 Quality management system forAMS design and manufacturingcontrol

prEN 15267-3 Stack-AMS performance

specifications and testproceduresprEN 15267-4 Ambient-AMS performance

specifications and testprocedures

S li t i t


39/40

Salient pointsz MCERTS established in 1998z Based on international standards for:

Performance specifications, testing, design control andquality assurance

z Triggers: Laws, need to improve quality of monitoring

z International developments: Aligned with the German scheme in 2002 European co-operation Basis of new CEN standards (2007-8)

z Model for verification and approvalz www.mcerts.net

L h l d


40/40

Lessons we have learned

z Use standards if available Benchmarks for performance Framework for flexibility

z Make standards performance based black-box approach Input-output model Do not stifle innovation

z Provide and apply performance standards for testlaboratories

z

Strengthen through legislation Voluntary systems are weaker z Practicality, not academic exercises

Keep costs down for manufacturers

CENORM_RickGould

Documents