Lecture: Systems Analysis Methodologies: Dr. John C. Wright

Lecture: Systems Analysis Methodologies
Dr. John C. Wright
MIT - PSFC
28 SEP 2010
Introduction Outline
O UTLINE
Scoping study
Systems analysis - increasing
detail
Life cycle analysis
Simulation models
Risk analysis and uncertainty
How are all these connected?
2 SE T-4 Systems Analysis

Introduction Outline
I NTRODUCTION
Many issues for sustainability
requiring balance
We need to quantify to
proceed
Deal with complexity and
Economics Environment
uncertainty
This is the goal Systems
Society
Analysis
End result often involves very,
very large computer codes
How do we make such
computer models?

Scoping study
S COPING STUDY CHARACTERISTICS
Well see more of this in a fuel costs example next week.

Basic guidelines for a scoping study:
Highly simplied
Mostly linear analysis - add separate costs

Very few feedback effects
Advantages
Relatively simple to understand
Good overall picture
Identication of weaknesses

Systems Analysis
S YSTEMS A NALYSIS IS THE NEXT STEP IN EVALUATION
Assume a favorable scoping study

Next step is a detailed systems analysis
All elements are analyzed in much greater detail
For example in our nuclear plant scoping study we gave the fuel
price in $/kg
In a system analysis model these costs are further broken down
Fuel costs:
Mining costs
Conversion costs
Enrichment costs
Finance costs

Systems Analysis
M ODULARIZATION OF SYSTEMS ANALYSIS
Each of these may be further analyzed one or two levels deeper.

Input data will be based on experience and future projections.
The analysis will account for uncertainties.
All lower level contributions are combined to form one module of
the systems code
the fuel cost module.

Systems Analysis
SA INCLUDES NON - LINEAR EFFECTS
A critical feature in SA is the inclusion of interdependencies.

Systems analysis are not linear.
They include feedback effects.
For example consider mining costs:
Plenty of Reserves no problem, linear relation works.
Reserves dwindle other issues arise
Fuel costs will rise
Will new fuel be found, if so how much?
How will this affect the projected cost of fuel?

Systems Analysis
B EWARE OF COMPLEX CODES .
Systems analysis code contains a large number of complex
modules
Often hard to understand the whole picture, often expert in part of

the picture.
Should be more reliable than a scoping study thought.
Warning:
Be very careful using complex systems
analysis codes!!

Systems Analysis
SA IS NOT ONLY ABOUT MONEY
Investors are not the only people to carry out systems analysis
Investors focus on nancial returns
Architectural engineers focus on technical credibility, schedule,
and cost
Environmentalists focus on pollution, waste disposal, greenhouse
gasses, etc.
Government focuses on the public good

Systems Analysis
G OVERNMENT IMPACT ON SA IS THROUGH REGULATION
Desirability of a regulation is in the eye of the stakeholder.

Everyone is a lobbyist.
Financial institutions.
Engineering rms.
Environmental groups.
Industrial groups.
Consideration of impact of regulation is part of any SA.

There often is an uncertain political aspect to a regulation.

Systems Analysis
S TRUCTURE OF A S YSTEMS A NALYSIS
Number of approaches to systems analysis

Method below is fairly typical
Goal of the analysis answer the question
Does it make sense to build a new power
plant (of type X)?
The end product a large, complex, hopefully all inclusive,
simulation code.

Systems Analysis
T HE S IMULATION C ODE
Technical aspects from a life cycle analysis

Regulation aspects from a risk analysis
Include feedback effects
Combine to create a nancial analysis

Life Cycle Analysis
Systems Analysis Life Cycle Analysis
L IFE C YCLE A NALYSIS (LCA) ELEMENTS
Emissions
Energy
Production of Manufacturing Use of Disposal

Raw Materials Process Product
Wastes Recycle Wastes

Wastes Wastes
Comprehensive cradle-to-grave, wells-to-wheels, dust-to-dust

analysisCosts, Resource use, Emissions, Wastes, Costs, Performance, etc.
Attributes:
Includes
Sum cumulative attributes over total life cycle of product to compare net impacts
Raw materials
Materials processing
Manufacturing
Distribution
Repair and maintenance
Waste disposal
Decommissioning

MacDonalds
Styrofoam or paper?
Trees (natural?) Oil (bad?)
Chemicals (worse)
Paper (good ?)
Styrofoam (??)
Oil
Chlorine or
Peroxide
Benzene + C2H4 + etc.
Pulp Acid or Alkali
CFCs
CO2 Styrene
Paper Hard to recycle
PCBs + Pentane
Dioxins
Plastic coating Polystyrene foam
Water McD
Wastewater Recycle
Landfill Trash
Courtesy of Elisabeth M. Drake. Used with permission.

Hydrogen Production Example
Make from steam methane

reforming?
Make from water electrolysis
using wind power?

Steam Methane Reforming
System Boundary Definition
Upstream
processes
Plant Construction
Resources & Decommissioning Hydrogen
in
Natural gas
production & distribution
Fossil fuel
energy in SMR plant
Electricity
operation Emissions
generation
air, water,
wastes
System Boundary

SMR Results
H2 is a clean fuel, but its production from natural gas

has environmental consequences
H2 plant itself produces few emissions, except CO2

CO2 is the largest air emission (98 wt%) and accounts
for 77% of the GWP

0.64 MJ of H2 produced for every 1 MJ of fossil
energy consumed

Wind/Electrolysis Study
turbines electrolyzer H2 storage
Wind turbines:
Atlantic Orient Corporation (50kW x 3)
Class 5 wind data from upper Midwest site
(North Dakota)
Electrolyzer:
Stuart Energy (30 Nm3/hr nominal capacity)
Cars fueled: fleet of 46 at 3 kg/car/week

GWP and Energy Balance
Wind/Electrolysis
Preliminary results:
GWP = 650 g CO2-eq/kg H2
Only 5% of the greenhouse gas emissions from SMR
Energy balance = 20 MJ of H2 produced for every
1 MJ of fossil energy consumed
31 times more than the net energy balance from SMR
Emissions are from equipment manufacture

Majority from concrete bases for wind turbines
Water consumption in electrolysis accounts for nearly
all resources

Hydrogen Production Choice?
Wind power offers significant reduction in GHG

emissions
For transportation, there is a mismatch between
wind turbine energy availability and the large
concentrated populations of cars
Costs for hydrogen from wind power are MUCH
higher than those from SMR
For SMR, more fossil energy is consumed than H2
energy produced

Systems Analysis Life Cycle Analysis
ACCURACY REFLECTS UNCERTAINTIES
Technical accuracy is good

Based on established engineering principles
Amount of fuel per year
Amount of stainless steel pipe
Average lifetime of valves
Converting technical into $ more difcult
Interest rates
Ination rates
Cost of fuel

Systems Analysis An Example
A N LCA OF NUCLEAR FUEL COST INCLUDING SCARCITY
Cost of nuclear fuel including scarcity

Reference case: U=$2000/kg
Breakdown from the MIT study for cost per kg
Ore $437
Enrichment $117
Fabrication $825
Storage and Disposal $351
Total $2040

#$%&'(&)&*&+,&'-.%/&*0
12 3'2.&4',$05'''''''''''''''''''''678889:;<=
>5? 3'!'@>& A?&*%B'C+B''''''7B!D!8!8':>?*9E*
F'3'F.*+'*G5&''''''''''''''''''''''!B!D!8H':>?*9:;
I2'3'>5?9F'3'2.&49E*''''''''''''7B8D!8J :;<=9E*
K2 3'I2D12 3'!05 E*').&4''''''''6J7'I9E*
KLC2.&4 3'2.&4',$059:>?* 8BMH',&+509:>?*
!"
#$%&'($)(*+,(%$'*($-($.,
/"(01($.,(23,45'(/(01(.,652(*$(&',(7
8$'*($-($.,(3'(*+&'(9::;01<$.,
=.,(.,',.>,'(6*(*+3'(?.3%,(@(A3
A3 )$*(B,44(0)$B)
C''&D,(E"(2,6.'($-(.,',.>,'(-$.(!E"()&0,'
F+,)
#! ! #$ " %$ " &%$ " $ " ! #&'% "#$ ! $ " ! &'% "#$" ()*+,-.
G6%+()&0,(.,5&%,'(.,',.>,'(H2( #$$ " % ()*+,-.
!"
#$%&'(%&)*+,-,./,-
01,23*.,-,./,-*2-*2*45&67%8&*84*7%9,
$! !% " ! $" #$&' #%

08-7*84*8.,*$%((*%&6.,2-,*.23%'(:*2-*.,-,./,-*
'$%&'(,
;%93(,*98',(
$ #$&' %
+()* !% " ! + " %&' ,# # ! +" %&' ,#
" #

$! $" #$&' #%

<" =*!>?
!"
"#$%&''()($*+,%-../0)1
2345/6%$.%*30,/+6%7,+*)1%4+8%09+*:/%
#()9%)(4/
&1%0$1)%$.%$6/%:$/1%37%*/#%6/1/6;/1%
4+8%5/%.$3*'
<$)9%/../0)1%9+;/%+*%(47+0)%$*%)9/%
+4$3*)%$.%6/1/6;/1
!!
#$%&'()*%+&,-&'(.%/
#$+%%&%--%01/&2%1%+)34345&'
6%7890%)%41&,-&,82&4(.%/
6%7890%&,82&0,98&78941/&:31$&;<!&-+%%&4(.%/
=>&%8%01+3031?&2%)942&:388&340+%9/%&%90$&
?%9+&@ )%%1&4%:&2%)942&:31$&4(.%/
<82&4(.%/&+%7890%2&*?&4%:&4(.%/
<82&0,98&+%7890%)%41/&A&'0,98B#&A&CD!EB?+
'%:&2%)942&A&F&340+%9/%&G&1,198&78941/&A&E
!"
#$%&'()*%+&,-&'(.%/&01,234
5,)*62%&3%+)/
* !+" ! #$%&'%()*+,-.& " /012*3.4215.6.(' " +.7*51415%'8
*!"#$
! *********%&'( ***** " ******* + ********9***:;:<*%&'(+
,)
! =:: " ==;>+
7%+%8&'2(.% 9&:;;8&'1,<= 9&!>;8&#? 9&";&@+/8&3&6/&

)%</(+%A&62&@%<+/
!"
#$%&'$($)*$(
'$($)*$(&+,&-,%.&o /,(0&,1&,)$&+,$(&23
4,(0&,1&,)$&+,$(&23&o ($5)/6&1,)&.$%&,)$
#$%&)$($)*$(&/,(0&7,)$&065.&8.80859&)$($)*$(
:20&;,2&9,($&7,)$&<$/52($&,1&8./)$5(8.+&
-$75.-&065.&;,2&+58.&%806&.$%&)$($)*$(
!"
#$%&'$($)*$(&+,-./0
1&(2345$&3-6$5
% !"# $& % % $
! % !"# &&
'$ $& % ! '$ " " ( # '$ " ! '() # *" * " '
% !"# $& %
% !"#
7$)$&'28 9&!:;<=8> ?@A-)$B&C2 9&DEE&F&?@A-)$B&
G.6&?H&9&!
#-/$&/I$&J$$6KG,?&5--4
C-)$+/0&6$4$.6(&-.&'2+/0
'2+/0&6$4$.6(&-.&C-)$+/0&
!"
E QUATIONS FOR THE COST OF ORE ARE NON - LINEAR
Known reserves and cost of ore are inter-related

10N(t)MF t
Core (t) = Ci exp k1
Ri (t) 10N(t)MF t

Ncoal
N(t) = Ni0 + + k2 Ni0 t
Tp

Core (t) Ci
Ri (t) = Ri0 1 + k3
Core (t)
k1 = 2.3, k2 = 0.05, k3 = 2

C OST OF ORE FROM SYSTEMS ANALYSIS
1,000 C[$/kg]
N[#]
800 R[107 kg] Note the
singular
600 response
around 40
400 years. What
causes this?
200 What does a
plot of R vs C
0 look like?
0 10 20 30 40
Year

C ONCLUSION
Cost of ore increases by 30

Ore is 1/5 the cost of uranium
COE of uranium = 0.56 cent/kWhr
This yields 3.8 cents/kWhr
Not as bad when you calculate the present value
Still it could be a problem
Uncertainty: what is the sensitivity to k1 , k2 , k3

Systems Analysis Risk Analysis
R ISK A NALYSIS
Risk analysis involves accidents to people or mechanical failures

Too many injuries or failures lower the capacity factor and reduce
revenue
We want to minimize risk but it is not possible to achieve zero risk
Qualitatively risk can be written as
Risk = Frequency Consequence

T YPES OF R ISK
Risk can be continuous or discrete

Continuous: exposure to toxic fumes
Discrete: steam pipe explosion
Consequences could cause minor injuries
Consequences could cause death
Consequences could involve land or water contamination
Even if no human or ecological damage, mechanical failures
lower capacity factor

AVOIDING R ISK
Three basic approaches:

Ultra robust design to minimize failure
Redundancy one system fails, another takes over
Increased shut-downs for maintenance and repairs

D ETERMINING R ISK
How do we determine risk?

This is the realm of risk analysis
Single component failures relatively easy
Qualication data available
History of real world experience
Can predict the mean time between failure
Single small failures often harmless
Single gigantic failures very rare

C OMPLEX FAILURES
Largest danger: often a sequence of

minor failures leads to major
catastrophe
For example: TMI, Challenger
Analysis requires sophisticated tools
Fault tree analysis
Event tree analysis
Uncertainty analysis
Probability of a severe accident
Greater for a sequence of minor
failures
Fault Tree example
Smaller for a single major failure

W HAT TO DO ?
Recommendations vary by group

Builders tend to underestimate risks to keep the cost down
Example: Dont worry the Big Dig is safe
Others tend to overestimate the risks to avoid or delay
construction
Example: Nuclear is unsafe dont build it.
Example: Wind kill birds dont build it.
Often the arbiter of risks are government agencies the EPA,
NRC, FDA, etc.
Desire risk informed regulations
Regulations consistent with severity of the risk

#$%&''()*+,*-./%/0,12'/
345%64/)%+-)7%,88/9*%94$)*+(9*-4$%94)*%48%,%
245/+%2',$*:
#))(1/%-$-*-,'%9,2-*,'%94)*%-$9'(6/)%,''%
+/;(',*-4$)%,))49-,*/6%5-*<%+-)7
#)%94$)*+(9*-4$%2+49//6)=%$/5%+-)7)%1,>%?/%
-6/$*-8-/6
@</)/%9,$%'/,6%*4%$/5%+/;(',*-4$)
A/5%+/;(',*-4$)%'/,6%*4%94$)*+(9*-4$%9<,$;/)
B<,$;/%4+6/+)%-$9+/,)/%*-1/%,$6%94)*%48%
94$)*+(9*-4$
!"
#$%&'()*+,-.+/()&0(1%2
3**-4%&+$%&/)/+/52&.56/+52&.(*+&/*&'.56
3**-4%&+$%&/)/+/52&.()*+,-.+/()&+/4%&/*&#.
7%8&,%9-25+/()*&(:+%)&(..-,&,5)1(42;
<(,&*/462/./+;&5**-4%&+$5+&)%8&,%9-25+/()*&
(..-,&5+&+$%&,5+%&r 6%,&;%5,
=5.$&)%8&,%9-25+/()&/).,%5*%*&.(*+&>;&'
=5.$&)%8&,%9-25+/()&/).,%5*%*&+/4%&>;&+
?(52@&.52.-25+%&+$%&)%8&.(*+&(:&+$%&625)+&
/).2-1/)9&)%8&,%9-25+/()*
!"
#$%&'%(&)*+,-./0-1*+&2%.1*3
'%(&0*+,-./0-1*+&-14%&#05&6&*.171+89&-14%&#0
:9/,&%;-.8&.%7/98-1*+&-14%&'-
'&6&.#05&6&-*-89&+/4<%.&*=&0$8+7%,
>8-$%48-10899?
"!a ! "! " # %$ ! "! " %"!a%$
@*9A%&=*.&#05
"!
"!a !
! % %$
'*-%&-$8-&.-&B&C&=*.&0*4:9%-1*+
!"
#$%&'%(&)*+,
-&+./.012&1234/%5,&$*06+&7*2&,$%&5%(&8*+,
$%!
a ! & !"# " ' %& ! & !"# "
& !"# %&
! $ %(
'*,%&,$1,&,&156&)&12%&2%01,%6
91:%&,&0123%2&; 7%(%2&(*2:%2+&5%%6%6<&0%16+&
,*&0*(%2&8*+,+
=4,&,$%>&12%&?1.6&*@%2&1&0*53%2&,./%
A.B%6&8*+,+&C%D3D&.5+42158%<&E%5%7.,+<&%,8DF&0%16&
,*&1&5%,&.582%1+%&.5&)81? 64%&,*&6%01>+
!"
"#$%&'(#)'*
+,-)%.%,/'0,-&1$%2'34)$54)%'4,$'&46#)
%( ! %( !"#$ " %( %"&'#
746#)'-#2/2'0,-)%42%'42'/'0,-)%42%2
+,-)%.%,/4&'&46#)'-#2/2'4)%'4221.%$'/#'2-4&%'
&0,%4)&8'50/3'0,0/04&'&46#)'-#2/2'*&46#)
%*
%( %"&'# ! +!( %"&'#
,)
9: 02'4'-#,2/4,/'46#1/'%;14&'/#':
!!
#$%&'())(*&+,-%
#$%&-%.&/(-0)12/),(-&/(0)&,0&3,4%-&56
%*! %+
a ! , !"# "
, !"# %, $"%& " -!, '"()%

! % %+ *!
#$%&5%0)&0)17)%36
8,-,0$&/(-0)12/),(-&70&0((-&70&9(00,5:%&
%4%-&,;&,)&*%7-0&5(11(.,-3&*(1%&*(-%6&
29&;1(-)
!"
#$%&'$()#*+'),+-)''
8
8
Total Capital Cost
6.5
Ccapny
( ) 5
3.5
2 2
0 0.1 0.2 0.3 0.4 0.5
0 y 0.5
delta t (years)
!"
MIT OpenCourseWare
http://ocw.mit.edu
22.081J / 2.650J / 10.291J / 1.818J / 2.65J / 10.391J / 11.371J / 22.811J / ESD.166J

Introduction to Sustainable Energy
Fall 2010
For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.

Lecture: Systems Analysis Methodologies: Dr. John C. Wright

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lecture: Systems Analysis Methodologies: Dr. John C. Wright

Uploaded by

Copyright:

Available Formats

Lecture: Systems Analysis Methodologies

Dr. John C. Wright

2 SE T-4 Systems Analysis

Many issues for sustainability

Deal with complexity and

This is the goal Systems

End result often involves very,

very large computer codes

How do we make such

3 SE T-4 Systems Analysis

S COPING STUDY CHARACTERISTICS

Well see more of this in a fuel costs example next week.

Mostly linear analysis - add separate costs

Relatively simple to understand

Good overall picture

4 SE T-4 Systems Analysis

S YSTEMS A NALYSIS IS THE NEXT STEP IN EVALUATION

Assume a favorable scoping study

5 SE T-4 Systems Analysis

M ODULARIZATION OF SYSTEMS ANALYSIS

Each of these may be further analyzed one or two levels deeper.

the fuel cost module.

6 SE T-4 Systems Analysis

SA INCLUDES NON - LINEAR EFFECTS

A critical feature in SA is the inclusion of interdependencies.

Plenty of Reserves no problem, linear relation works.

Reserves dwindle other issues arise

Fuel costs will rise

Will new fuel be found, if so how much?

How will this affect the projected cost of fuel?

7 SE T-4 Systems Analysis

B EWARE OF COMPLEX CODES .

Systems analysis code contains a large number of complex

Often hard to understand the whole picture, often expert in part of

Be very careful using complex systems

8 SE T-4 Systems Analysis

SA IS NOT ONLY ABOUT MONEY

9 SE T-4 Systems Analysis

G OVERNMENT IMPACT ON SA IS THROUGH REGULATION

Desirability of a regulation is in the eye of the stakeholder.

Consideration of impact of regulation is part of any SA.

10 SE T-4 Systems Analysis

S TRUCTURE OF A S YSTEMS A NALYSIS

Number of approaches to systems analysis

plant (of type X)?

The end product a large, complex, hopefully all inclusive,

11 SE T-4 Systems Analysis

Technical aspects from a life cycle analysis

12 SE T-4 Systems Analysis

L IFE C YCLE A NALYSIS (LCA) ELEMENTS

Production of Manufacturing Use of Disposal

Wastes Recycle Wastes

Comprehensive cradle-to-grave, wells-to-wheels, dust-to-dust

Repair and maintenance

13 SE T-4 Systems Analysis

Trees (natural?) Oil (bad?)

Courtesy of Elisabeth M. Drake. Used with permission.

Make from steam methane

Courtesy of Elisabeth M. Drake. Used with permission.

System Boundary Definition

production & distribution

Courtesy of Elisabeth M. Drake. Used with permission.

H2 is a clean fuel, but its production from natural gas

$! !% " ! $" #$&' #%