dim División de Ingeniería de Máquinas

!"#$%&'$(

!"#$%&'()*&)*+&($,-+./($%&0#""12#$&-+/)-(,2#$/&$+'&3+*(,"+&+4,(+$,5&-+61(-+7+$)/&.-(/+8&9-.$/0#-)&/+,)#-&(/&#$+&#:&)*+&7#/)&.;+,)+<&=5&)*+&-+%1".2#$/>&

'*(,*&*./&)#&7.$1:.,)1-+&0#'+-:1""&+$#1%*&3+*(,"+/&-+<1,($%&)*+(-&+7(//(#$/&.)&)*+&/.7+&27+8&9*(/&(/&)*+&-+./#$&'*5&"(%*)'+(%*)&)+,*$#"#%5&(/&%-#'($%&

($&(70#-).$,+>&/($,+&($)-#<1,($%&$+'&"(%*)&7.)+-(."/&"(?+&,.-=#$&@=+-&-+($:#-,+<&0"./2,/&ABCDEF&($&7.//&0-#<1,2#$&#:&3+*(,"+/&'#1"<&.""#'&)*+7&)#&-+<1,+&

+$+-%5&,#$/1702#$&.$<&&)*+-+:#-+>&)#&,#70"5&'()*&)*+&+7(//(#$&-+/)-(,2#$/8&

B.-=#$& @=+-& 0.-)/& 7.$1:.,)1-($%& 0-#,+//+/& -+61(-+& ($)+$/(3+& *.$<'#-?& .$<& .1)#7.2$%& )*+7& (/& .& *1%+& )+,*$(,."& ,*.""+$%+8& G$& .<<(2#$>& 7.)+-(."&

.$(/#)-#05&=-($%/&.=#1)&.&*(%*&<+0+$<.$,+&=+)'++$&)*+&/)-1,)1-."&,#$,+0)&#:&)*+&0.-)&.$<&)*+&7.$1:.,)1-($%&0-#,+//8&9*+&0-+/+$)&-+/+.-,*&(/&:#,1/+<&($&

)*+&-+/#"12#$&#:&)*(/&0-#="+7>&=5&@$<($%&$+'&'.5/&)#&#027(H+&)*+&<-.0($%&#:&)*+&@=+-/&($)#&)*+&7#"<>&($&#-<+-&)#&0-#<1,+&0.-)/&'()*&*(%*+-&7+,*.$(,."&

0-#0+-2+/&."#$%&'()*&($,-+./+<&0-#<1,23()5&#:&)*+&7.$1:.,)1-($%&0-#,+//+/8&

)*+(,$%-'$-%&.(/01'*2$#(30%($4*(5&##(

6%07-'801(03(/0920#:$*(6&%$#(

&

!.*;&17%0(!"0-<!##&.:(=07%>?-*@A(

B1%:C-*(/4&'D1(E&1&%%0A((

'

F-&1(5&1-*.(5-G0@(H-:;0#&(

(

&&.*;&17%0I&%07%:?-*@J-29I*#A

(

K92%0L*($4*(C-&.:$M(03(9&##(2%07-'801(/N6=(2&%$#O(!L0:7(2%*30%9:1?(7*3*'$#(&

( ( ( ( ( ( ( ( ( ( ( ( ( ( (

)*+(&.?0%:$49(30%(P:1*9&8P(7%&2:1?(#:9-.&801(+:$4('092.*Q(7%&2:1?(.:1*#(((

&

&

&

&

&

&

&

&

&

&

&

I(;+-+$)&)+J2"+&.0"(,.2#$&0.)*/&A<-.0($%&"($+/>&-+<&,#"#1-+<F&"+.<&)#&<(;+-+$)&@=+-&<(-+,2#$/&.$<&)+J2"+&+$<&/*.0+8&G)&(/&+//+$2."&)#&#027(H+&)*(/&<-.0($%&

"($+/&($&#-<+-&)#&.3#(<&@=+-&<+:+,)/&.$<&)#&#=).($&.&0-#0+-&@=+-&0#/(2#$8&

/&%"01(R"*%($*Q8.*(S/N6=T( E*Q8.*(&7&2$&801($0($4*(90.7(

) . :$4 3 P: 8P 7 : : . 8

B.-=#$

(

)*+(U5V(W(,*%:*#I(,0-%'*O(U5V(

60$*18&.(7*3*'$#(7-*($0('-%%*1$(2%*30%9:1?(2%0'*##*#((

C(=+-&7(/."(%$7+$)K&-+<1,+<&7+,*.$(,."&0-#0+-2+/&

=E5A($4*(90#$(#2%*&7(9&1-3&'$-%:1?(2%0'*##(:1(&-$0908L*(:17-#$%M((

6%02*%(R"*%(&.:?19*1$(

dim

División de Ingeniería de Máquinas

164

Abstract

The research on refractory materials has been increased in the last years due to the fact that the metallurgical industry imposes more and more exacting specifications. On the other hand, an interesting way to work with composite materials is the powder technology. This technique consists in producing solid and resistant pieces from powder material. There are two methods to apply this technique: the first method consists in producing the piece by hot pressing and the second method consists in a sintering process after the compaction of the powder. The aim of this work is the generation of a discrete element model that allows to simulate the sintering process of a refractory material. After the model has been adjusted, this is used to look for other mixtures that improve the properties of the sintered pieces. The analyzed process consists in compacting a powder to obtain a green body. After that, the body is subjected to high temperatures to increase the mechanical strength of the piece. Therefore, the compaction and sintering processes are separately studied.

Numerical discrete simulation improving new refractory designs for ultraclean steels

a

Cristina Ramírez-Aragón,

Joaquín Ordieres-Meré,

Fernando Alba-Elías

aETSII, Universidad Politécnica de Madrid ([email protected])

bETSII, Universidad de La Rioja

Method

The EDEM® software is used to simulate the process through the discrete element method.

The simulations are carried out in two stages: simulation of compaction process and simulation of sintering process.

Powder

To simulate the sintering process a contact model is implemented. This consists in increasing the sintering force each timestep. Thereby, the overlap between particles augments as the sintering force is increased. This overlap simulates the neck growth between particles during the sintering.

Green body Sintered piece

Powder compaction Green body sintering

Finally, the mechanical properties of the sintered piece are analyzed and solutions that improve those properties are sought.

Powder compaction process

Green body sintering process

Improvement of the sintered pieces

To simulate the compaction process a contact model with bonding is used. The created bonds make the particles remain united forming a solid body.

At this model, a set of particles is created inside a die and then an upper puncheon goes down until the green body has the desired height. At the end, the puncheon goes up and the piece is pulled out of the die.

Stage I

Stage II

Stage III

In this simulation, the green body that was generated during the compaction process is placed over a horizontal surface and remains at rest while the sintering force is acting.

Simulated processes.

Simulated green body.

Powder compaction process.

Overlap between particles produced in simulations.

Refractory bricks. Ladle in a casting process.

5. (s_max,i ;P_f,i,DA) ! Three-parameter Weibull cdf (m s_u s₀) referred to DA

6.

Abstract

The large scatter in the strength values of brittle materials requires the use of cumulative distribution functions (cdf) to adequately describe the fracture stress. The three-parameter Weibull distribution function is widely used for this purpose. The failure probability of a loaded specimen expressed by this function depends on the size of the specimen (the so-called size-effect) and is valid for a uni-axial and uniform stress state. For the estimation of the parameter values of the cdf the process is just reversed i.e. the three parameters are estimated by using the calculated failure stresses of a series of destructive laboratory tests. The size-effect must be considered and, depending on the test employed, multi axial stress states can be found in the specimens before failure. An iterative algorithm to determine the three parameters is proposed. The size effect as well as the multi axial stress case is considered. This procedure is applied to determine the strength of crystalline silicon wafers. Due to the particularities of the specimens (i.e. very low thickness), non-linear behaviour during the tests must be taken into account. This leads to a wider field of applications of the proposed method.

Obtaining the cumulative distribution function of fracture stress for brittle materials. Application to silicon wafers

a

Josu Barredo,

Alberto Fraile,

Covadonga Alarcón,

Lutz Hermanns

a,b,d

Departamento de Ingeniería Mecánica, ETSII-UPM ([email protected])

c

Departamento de Ingeniería de Materiales, ETSIM-UPM

Crystalline silicon wafers

Samples characteristics

Anisotropy of mono-crystalline silicon:

c₁₁= 165.6 GPa; c₁₂= 63.9 GPa; c₄₄= 79.5 GPa Sample and test dimensions

Test

Finite Element Model (shell elements) Anisotropy; Large displacements; Contact

Fitting of test results-FE Model

Stress state of all samples before failure is reached

Obtaining the three parameters of the Weibull cdf

Weibull Model

Non-uniform stress Equivalent area of test i:

Multi-axial stress state

Principle of Independent Actions (PIA):

Iterative procedure to obtain the three parameters of the Weibull cdf

s

s q

+ s s

1. Assignation of experimental failure probability (N tests):

ss

2. (s_max,i ;P_f,i) ! Fitting to a three-parameter Weibull cdf using the least square method

(m s_u s₀) “experimental”

3. Calculation of the equivalent area of each test:

4. Calculation of the failure probability of test i referred to a differential area DA

NO YES END

Project sponsored by:

(Ref: ENE2014-56069-C4-2-R)

166

" # '

!(8 -5 = J 2 2

. 2 I 3B % ! B 9 ( % 1 , %

I # 2 . I

!(8 0 # #

! * (V # 2 E

!(8 0 " #

; # ! =; % U &; 2 E

!(8 0 # 3 ' #

/ " # ; # ! : %

!(8 ' 0 9

# & 9 E

!(8 - 0 '# ' 92 2

#

168

!"#$%&'$(

!"#$%&'#()*+,#-.-/.*'#)0#$%.#123#4565#7.'.8*9%#:*);<#<*.'."$#$%.&*#-)'$#*.9."$#*.';=$'>#4%.#7.'.8*9%#:*);<#?4%.)*@#8"A#5<<=&98B)"'#)0#

6)"'$*;9BC.#5<<*)D&-8B)"E#F4565G#(8'#9*.8$.A#&"#HIIJ#/@#-8$%.-8B9&8"',#<%@'&9&'$'#8"A#."K&"..*'>#4%.#-.-/.*'#)0#$%&'#-;=BA&'9&<=&"8*@#K*);<#

9)==8/)*8$.#&"#'.C.*8=#*.'.8*9%#=&".',#/@#8<<=@&"K#-8$%.-8B98=#8<<*)D&-8B)"#-.$%)A'#$)#89$;8=#<*)/=.-'#&"#L"K&"..*&"K,#8'#&$#&'#'%)("#&"#$%&'#()*+>###

)*+,-./)-012(!33.*405!-0*+(0+(,*52(!3360)!-0*+,(

&

!789&:;%<(=&%><?(@8%:&%;<(;8(7&()&778?(5&%A&(278:&(B<CA:DE8>?(F&"%G87&(

,&:#GD%8?(

B&HG;(6E8:D<?(

B<7<%8#(@&%%G<#(

&2-,(;8(0:D8:G8%<#(0:;E#$%G&78#([email protected](

"2-,(;8(0:D8:G8%A&(K(,G#$8C&#(;8(-878'<CE:G'&'GL:(

'2-,(;8(0:D8:G8%<#(0:M<%CNO'<#(

#

#

#

#

PE&;%&$E%8(M<%CE7&##

5#".(#9=8''#)0#";-.*&98=#M;8A*8$;*.'#&'#A.N".A#&"$."A.A#$)#&"$.K*8$.#<.*&)A&9#0;"9B)"'#OHP>#4%.&*#-8&"#0.8$;*.'#8*.Q#

R  5"#8<<*)D&-8B)"#)0#$%.#C8=;.#)0#$%.#&"$.K*8=#8'#(.==#8'#8"#.'B-8$.#)0#$%.#.**)*#9)--&S.A#8*.#)/$8&".A>#

R  4%.#M;8A*8$;*.'#')#A.N".A#8*.#)<B-8=#&"#$%.#'."'.#$%8$#&"$.K*8$.#.D89$=@#<)=@")-&8='#)0#%&K%#A.K*..#8"A#$%.*.#A)#")#.D&'$#)$%.*#M;8A*8$;*.'#

K&C&"K#.'B-8$.'#)0#$%.#C8=;.#)0#$%.#&"$.K*8=#8"A#$%.#.**)*#9)--&S.A#(&$%#%&K%.*#A.K*..#)0#<)=@")-&8=#.D89$".''>#

R  4%.#M;8A*8$;*.#(.&K%$'#8*.#<)'&BC.#8"A#$%.#")A.'#)0#$%.#M;8A*8$;*.#8*.#'&-<=.#8"A#/.=)"K#$)#$%.#&"$.*C8=#)0#&"$.K*8B)">#4%.'.#<*)<.*B.'#8*.#

<*)C."#0)*#';T9&."$=@#=8*K.#";-/.*#)0#")A.'#8"A#8#(&A.#9=8''#)0#(.&K%$#0;"9B)"'>#

R  4%.#")A8=#<)=@")-&8='#8*.#8=K./*8&98==@#9%8*89$.*&U.A#/@#-.8"'#)0#$%.#-)-."$'#)0#$%.#(.&K%$#0;"9B)">#

OHP#V>#A.#=8#68==.#W'.*",#?X<B-8=#LD$."'&)"#)0#$%.#YU.KZ#M;8A*8$;*.E,#!35#[>#\;-.*>#5"8=>#]^#FHI_^G,#`HHa`bc>###

)<CQ%8##8;(#8:#G:D(M<%('R&::87(8#OC&O<:(G:(SG%878##('<CCE:G'&O<:##

!"#(&*.=.''#9)--;"&98B)"',#$%.#9%8"".=#N=$.*#&'#;';8==@#B-.aC8*@&"Kd#&$#&'#".9.''8*@#$)#.'B-8$.#$%.#9%8"".=#N=$.*#R#8$#$%.#*.9.&C.*Q#

#

#

#

#

#

#

#

!"#);*#()*+#O_P,#(.#9)"'&A.*#-#8'#$%.#e&'9*.$.#6)'&".#4*8"'0)*-#4@<.a!#Fe64_G#8$#/)$%#$%.#$*8"'-&S.*#8"A#$%.#*.9.&C.*>#X;*#-8&"#9)"$*&/;B)"'#8*.Q#

R  f.#<*.'."$#8"#8=K)*&$%-#(%&9%#89%&.C.'#8"#899;*8$.#.'B-8B)"#)0#'@--.$*&9#9%8"".=#N=$.*'#R#;'&"K#)"=@#8#'-8==#";-/.*#)0#$*8&"&"K#'@-/)='#g₊>##

R  4%.#')=;B)"#&'#)/$8&".A#/@#;'&"K#.&$%.*#-8$*&D#&"C.*'&)"#)*#9)-<*.''.A#'."'&"K#8=K)*&$%-'>##

R  f.#<*)C&A.#$%.#$%.)*.B98=#*.';=$'#(%&9%#K;8*8"$..#$%.#C8=&A&$@#)0#$%.#<*)<)'.A#$.9%"&M;.#0)*#$%.#e64_>#

#O_P#3>#L>#e)-h"K;.U,#e>#i;."K),#:>#Y8"'&K*.Q#?L'B-8B)"#)0#Y@--.$*&9#6%8"".='#0)*#e&'9*.$.#6)'&".#4*8"'0)*-#4@<.a!#3;=Ba98**&.*#Y@'$.-'Q#5#

6)-<*.''.A#Y."'&"K#5<<*)89%E,#4%.#Y9&."BN9#f)*=A#[);*"8=,#j)=>#HI_^#F\)C>HI_^G,#5*B9=.#!e#_^_]`I,#http://dx.doi.org/10.1155/2015/151370##

B&%"<ET($%&:#M<%C&O<:#(

• 4%.#e8*/);D#$*8"'0)*-8B)"'#<*)C&A.,#k)&"$#(&$%#)$%.*#8<<=&98B)"',#8#-.$%)A#0)*#)/$8&"&"K#')=;B)"'#)0#')-.#&"$.K*8/=.#'@'$.-'>#!"#O]P,#$%.#9)"9.<$'#)0#

e8*/);D#089$)*&U8B)"#8"A#e8*/);D#$*8"'0)*-8B)"'#0)*#8*/&$*8*@#l.''."/.*K#/8"A.A#-8$*&9.'#8*.#8"[email protected]>#Y<.9&N98==@,#$%.#.D&'$."9.#)0#$%&'#+&"A#)0#

089$)*&U8B)"#&'#'$;A&.A,#8"A#')-.#';T9&."$#9)"A&B)"'#0)*#$%.#;"&M;.".''#8*.#A.$.*-&".A

>#

• !"#ObP#$%.#*.=8B)"#/.$(.."#$%.#e8*/);D#$*8"'0)*-8B)"#8"A#$%.#')=;B)"'#)0#$%.#0;==#m)'$8"$#4)A8#=8n9.#&'#8"[email protected]>#4%.#A&'9*.$.#m)*$.(.K#A.#j*&.'#

.M;8B)"#&'#;'.A#$)#)/$8&"#';9%#')=;B)"'>#

O]P#e>#V8**&)'#7)=8"h8,#e>#38"*&M;.,#X"#$%.#.D&'$."9.#)0#e8*/);D#$*8"'0)*-8B)"'#0)*#/8"A.A#-8$*&9.',#5<<=>#38$%>#8"A#6)-<;$>#H^]#FHI_^G,#<<>#__Ja_H^>#

ObP#e>#V8**&)'#7)=8"h8,#e8*/);D#$*8"0)*-8B)"#8"A#')=;B)"'#)0#')-.#&"$.K*8/=.#'@'$.-',#';/-&S.A#$)#5<<=>#38$%>#8"A#6)-<;$>##

U<E%G8%('<8V'G8:$#(&:;(<%$R<D<:&7(Q<7K:<CG&7#(

4%.#.T9&."$#9)-<;$8B)"#)0#o);*&.*#9).T9&."$'#)0#)*$%)K)"8=#<)=@")-&8='#.D<8"'&)"'#&'#8#C.*@#&-<)*$8"$#<*)/=.-#&"#";-.*&98=#8"8=@'&'#8"A#8<<=&.A##

-8$%.-8B9'#(&$%#8#(&A.#*8"K.#)0#8<<=&98B)"'#&"9=;A&"K,#k;'$#$)#-."B)"#8#0.(,#p;8"$;-#-.9%8"&9',#')=;B)"#)0#<8*B8=#A&q.*."B8=#.M;8B)"'#)*##

M;8A*8$;*.'>#X;*#$.8-,#$)K.$%.*#(&$%#')-.#9)==8/)*8$)*',#A.C.=)<.A#')-.#@.8*'#8K)#8"#8=K)*&$%-#F98==.A#\5j!35#8=K)*&$%-#O^PG##$)#9)-<;$.#$%&'##

9).T9&."$'#*.9;**."$=@>#!"#OJP#$%&'#8=K)*&$%-#&'#9)"'&A.*.A#&"#$%.#9)"$.D$#)0##)*$%)K)"8=#<)=@")-&8='#)0#8#A&'9*.$.#C8*&8/=.#8"A#&$#&'#8<<=&.A#$)#')=C.##

')-.#9)"".9B)"'#<*)/=.-'#0)*#-)"&9#/&C8*&8$.#m*8C9%;+#<)=@")-&8='>#

#

O^P#L>#:)A)@,#7>#7)"C.8;D,#5>#r8*U),#8"A#!>#5*.8>#3&"&-8=#*.9;**."9.#*.=8B)"'#0)*#9)"".9B)"#9).T."$'#/.$(.."#9=8''&98=#)*$%)K)"8=#<)=@")-&8='Q##

9)"B";);'#98'.>#[>#6)-<;$>#5<<=>#38$%>#cb#F_ss`G#H^`aH`^>#

OJP#!>#5*.8,#L>#:)A)@,#5>#7)"C.8;D,#[>#7)A8=#8"A#5>#r8*U),#V&C8*&8$.#m*8($9%);+#<)=@")-&8='Q#!"C.*'&)"#8"A#9)"".9B)"#<*)/=.-'#(&$%#$%.#\5j!35##

8=K)*&$%-,#[>#6)-<;$#5<<=>#38$%#Hcb#FHI_^G#^Ia^`>

(

(

169

Abstract This thesis attempts to explore the possibility to model human behavior and how it guides financial markets. According to Behavioral Finance theory the stock market ecosystem is influenced by the decision making of the individuals trading in it. The traders are heterogeneous in nature, with each group having their own belief and expectation. This thesis tries to answer the question Can human behavior and its responses to macroeconomic events be modelled and used as an indicator to predict price directions? To answer the former question, the research has delved deep into exploring human behavior guiding financial markets. Different exogenous variables representing stock broker behavior has been explored. These variables are derived from market data of local markets like the Madrid Stock Exchange, and also from micro blogging sites and website visit statistics. A local market microstructure is guided mostly by its local players and macroeconomic events. Where as more global stock markets are more guided by global macroeconomic events. This research constructs exogenous variables which effect the small stock exchanges and bigger stock exchanges alike. In this research different data set are constructed from web search volumes, sentiment scores of Twitter posts to page visit statistics of Wikipedia articles. The exogenous time series constructed is then used as a predictor variable for different supervised and unsupervised machine learning algorithms for future price predictions. In this research different categories of machine learning algorithm were used from simple tree based ensemble learning models to SVM (support vector machine) and kernel based models to more complex Deep Learning algorithms. The implication of the research is that it will help financial managers and traders use these correlations with social sentiment indexes to predict financial markets with certain accuracies. It will also provide them with early warnings of market downturns risk and indication of crisis.

Keywords: Social Media Analytics, Stock Market Prediction, Machine Learning Incorporation of human knowledge to the

stock markets for improving forecasts

a

Swarnava Mitra ,

Joaquin Ordieres Mere

a,b

Universidad Politécnica de Madrid, Department of Industrial Engineering, Business Administration and Statistics ([email protected])

170

Abstract

With the rapid development of Internet of Things (IoT) technology, billions of smart devices are being connected into a whole network and streaming out a huge amount of data every moment. Unimaginable potential value can be mined from these data with the help of “Cloud Computing” and “Machine Learning” techniques. The target of our research is to address the benefits of IoT in social applications, especially in healthcare area, by developing a multilayer framework. Low cost data collection, efficient data transfer, flexible data management and accurate data analysis mechanisms will be included in the framework. A Smart Decision Support System is supposed to be developed on the basis of this framework.

Relevant framework for social applications of IoT by means of Machine Learning techniques

a

Xiaochen Zheng (PhD researcher),

Joaquín Ordieres (Professor)

a

Technical University of Madrid ([email protected])

Figure 2 An example of the data collection and management system Figure 1 Overal framework of social applications on the basis of Internet of Things

Introduction

From its earliest days, the insurance industry has been data-centric. In the past, insurance companies relied on historical data from policy administration solutions, claims management applications and billing systems. Today, the explosion of new data is turning the insurance business model on its head. The growth in Quantified-Self products has had an especially large influence. Insurance has become a data industry .The factor of accelerated aging, deterioration of the living environment, Improvement of living standards are affecting people's lives. More and more people attach importance to their own health and they will pay attention health insurance.

The main method of Quantified-Self are data collection , data visualization, make a Business decisions . Wearable devices can collect the body data, analysis fitness of each policyholder. And to grasp the data of the human data from social network. The Data collection can do so by facilitating the discovery of risk factors for disease at population and individual levels, and by improving the effectiveness of interventions to help people achieve healthier behaviors in healthier environments. The wearable devices will provide the basic layer of continuous data stream to integrate with those coming from the smartphone in order to derive specific features along the time. Thus, trends can be estimated for different type of factors and human related factors can be mixed to establish online clustering to be processed against the accidents by a survival analysis. The concept behind this proposal is to be behind this first layer of application which consists in gather continuous data, but designing learning processes capable from knowing data sets and their evolution, in such a way that forecasting can benefit users as they will be informed about potential risks of specific actions, including causal patterns for biometric behavior.

Many opponents to the introduction of in Quantified-Self products have expressed concerns in regarding of privacy and insurance companies’ abilities to track their insured customers. Through the sensor, insured customers are more aware of how their premiums are calculated, and they’ll know that their health status and attitudes are recognized by The idea of using data to generate personalized pricing and provide meaningful risk information appears simple. This research will necessitate discovery it helps people to successfully modify their risk behaviors and reduces healthcare costs.

Relevant contributions of Internet of Things and Data Science to the Insurance Business by means of Machine Learning techniques

LIU YANG Professor ˖Joaquin, Ordieres [email protected] [email protected]

Industrial Organization, Business Administration and Statistics Business Intelligence

The figure1 ,-The structure map

172

+g]WIEjhd][h]gIGDs

¥.INÃÁÂşÆÊÄß ߟ.¦

Dhjg<Ej

.IEI[j <Gp<[EIh Q[ hshjIZ QGI[jQNQE<jQ][ N]g Z]G<Y jIhjQ[O Q[ EQpQY I[OQ[IIgQ[O Q[EYkGI jPI d]hhQDQYQjs j] <ddYs <gjQNQEQ<Y ¥ZI<hkgIG¦ N]gEIh j] <

hjgkEjkgI Q[ <GGQjQ][ j] jPI k[ZI<hkgIG <ZDQI[j IrEQj<jQ][ <[G j] QGI[jQNs < Z]GIY jP<j <EE]k[jh N]g D]jP IrEQj<jQ][ h]kgEIh [ dg<EjQEI jPI

<ZdYQjkGI]NjPI<gjQNQEQ<YN]gEIhE<[DIhZ<YYE]Zd<gIGj]jPI<ZdYQjkGI]NjPI<ZDQI[jN]gEIhh]hZ<YY<[Gdg<EjQE<Y<Ejk<j]ghE<[DIkhIG][

gIY<jQpIYsY<gOIhjgkEjkgIh

0PIdg]EIGkgIQhG<j<E]YYIEjQ][¥D]jPhshjIZQ[dkj<[G]kjdkj¦hshjIZQGI[jQNQE<jQ][<[GZ]G<Yd<g<ZIjIgIhjQZ<jQ][0PIhshjIZQGI[jQNQE<jQ][

hjIddY<sh<EgkEQ<Yg]YIQ[jPIfk<YQjs]NjPIZ]G<Yd<g<ZIjIghjP<j<gIGIgQpIGNg]ZjPIQGI[jQNQIGhshjIZZ]GIY<hqIYY<hQ[jPI[kZDIg]NZ]G<Y

d<g<ZIjIghjP<jE<[DIGIjIgZQ[IG0PQhIrdY<Q[hjPIQ[EgI<hQ[OQ[jIgIhjQ[h]dPQhjQE<jIGhshjIZQGI[jQNQE<jQ][<YO]gQjPZhN]gZ]G<Y<[<YshQh

0PI hj<jIŸhd<EI Z]GIY E<[ DI khIG <h jPI hshjIZ Z]GIY DIE<khI Qj E<[ j<XI Q[j] <EE]k[j D]jP ZI<hkgIG N]gEIh <[G k[ZI<hkgIG N]gEIh 7I

dg]d]hI j] IhjQZ<jI jPI hj<jIŸhd<EI Z]GIY khQ[O Z<rQZkZ YQXIYQP]]G ZIjP]G <[G jPI rdIEj<jQ][Ÿ!<rQZQv<jQ][ <YO]gQjPZ DIE<khI Z<rQZkZ

YQXIYQP]]G P<h ]djQZ<Y hj<jQhjQE<Y dg]dIgjQIh hkEP <h E][hQhjI[Es <[G INNQEQI[Es 0PI dIgN]gZ<[EI ]N jPI dg]d]hIG ZIjP]G Qh <[<YsvIG khQ[O <

hQZkY<jIGhjgkEjkgI

[dkj›]kjdkjpIghkh]kjdkjŸ][YsG<j<<[<YshQhN]g

Z]G<Yd<g<ZIjIghIhjQZ<jQ][

 <j<

QO/QZkY<jIGhjgkEjkgI

/shjIZ ]kjdkj Qh E]ZdkjIG khQ[O /IE][G "Iqj][ <q NjIg jP<j

ZI<hkgIZI[j[]QhIP<pIDII[<GGIG

QOÐ 0hshjIZQ[dkjŽ.0hshjIZ]kjdkj<j[]GI

Đ !]G<Yd<g<ZIjIghIhjQZ<jQ][

!]G<Yd<g<ZIjIgh

qPIgI<gIjPIIQOI[p<YkIh]NjPIIhjQZ<jIGZ<jgQr

QOĐ!]G<Yd<g<ZIjIghIhjQZ<jIGkhQ[OZ]GIYÂ¥OgII[G]jh¦<[GZ]GIYÃ¥DYkIG]jh¦Q[ÂÁÁhQZkY<jIGE<hIh

<

<pQIg <g<

Ihmhk<[

[gQfkIY<gE_[

<0IEP[QE<Y1[QpIghQjs]N!<GgQG

¥W<pQIgE<g<³kdZIh¦

Ð /shjIZQGI[jQNQE<jQ][

p<QY<DYIG<j<

 $kjdkjG<j<

Ð [dkjG<j<EPQgdhQO[<Y¥hIIQOæ

+g]d]hIGZ]GIYhN]gjPIG<j<

Ÿ !]GIYhj<jIŸhd<EIZ]GIYN]g]kjdkjŸ][YsG<j<



Ÿ !]GIYÏhj<jIŸhd<EIZ]GIYN]gQ[dkj›]kjdkjG<j<

]jP Z]GIYh P<pI DII[ IhjQZ<jIG Ng]Z G<j< khQ[O !<rQZkZ

QXIYQP]]G<[GjPIrdIEj<jQ][Ÿ!<rQZQv<jQ][<YO]gQjPZDsZI<[]N

E]ZdkjIgE]GIhGIpIY]dIGDsjPI<kjP]gh¥¦

  







Abstract

In this work, a short-term electric load forecasting method is developed, using as an explanatory variable the temperature forecasts from an external agent. A regression-spline model has been designed to implement the relationship between temperature and energy demand. The proposed method is general and can be implemented in any real-time electricity energy system.

To check the method's performance, a real data set from the Spanish electricity system has been used and the developed algorithm has been tested employing the MSE as the performance metric. Results denote that the forecasting method is both accurate and computationally efficient.

Short-term forecasting of electric load in Spain

a

Eduardo Caro,

Jesús Juan,

Javier Cara

a

Laboratory of Statistics - Technical University of Madrid

([email protected] – [email protected] – [email protected])

Motivation and aim

For any transmission system operator, it is important to have a proper prediction for the short-term future consumption. This is particularly valuable in the electric power framework: since the electricity cannot be efficiently stored, the electricity system operator must meet generation and demand on a real-time basis.

The aim of this work is to enhance the quality of the short-term load forecasts, based on a more accurate temperature model.

Fig. 1. Relationship between ambient temperature and the electric load (Blue: weekdays, Green: Saturdays, Red: Sundays, Black: public holidays)

4 a.m. 8 a.m.

4 p.m. 8 p.m.

0 10 20 30 40 + 10%

+5%

0 -5%

0 10 20 30 40 + 10%

+5%

0 -5%

+ 10%

0 10 20 30 40 + 10%

+5%

0 -5%

0 10 20 30 40 + 10%

+5%

0 -5%

+ 10%

Temperature Temperature

Influence over the loadInfluence over the load

Results

Fig. 4. Resulting model for several representative hours.

Spanish Electric Load

Temperature

Temperature influence

As it can be observed from Fig. 1, there is a non-linear relationship between temperature and electric load. Specifically, there is a negative correlation among these two variables for winter months (a lower temperature causes a higher demand), and a positive relationship for summer months (a higher temperature provokes a higher demand).

Fig. 2. Spline-regression additive model applied to temperature.

Temperature modelling

In order to model properly the relationship between ambient temperature and the Spanish electric load, a spline-regression additive model has been employed.

Figure 2 provides an illustrative example of an application of this model, using three nodes and T_min= 5ºC and T_max= 37 ºC, resulting the equation below:

The following geographical locations have been employed for computing the daily T_ivalue using AEMET weather forecasts:

!= "_#+ "_$· %_!+ "_&· ' %_!, (_$ + "₎· ' %_!, (_& + "_*· ' %_!, (₎ ···

10 20 30 40 0 %

-1 %

-2 %

-3 %

-4 %

-5 %

Influence over the load

Temperature ^ü_ü

ü

ü ü ü

ü ü ü

ü

The spline-based temperature model has proved to be accurate and reliable. The resulting relationship between load and temperature is provided in Fig. 4, denoting that this relation significantly varies among hours.

As an illustrative example, Fig. 3 provides the forecasted (green line) and the actual load demand (blue line), denoting accurate results.

Fig. 3. Real and forecasted electric load for a recent week using the developed model.174

" # % " '

!!( 00 # $ # / ) 2 #

/ , * 7 3 * & 9 E

!!( 04 # . " #

> # " 7 2 % " &

> # * 7 * !G / & &

!!( 05 ; # # " 7 2 %

() &; & , " & > # * 7 * !G

!!( 4 3 #

IB 2 2 . 2 )

!!( ' 4 #

# / J 2 ! & 9 * 2

6 ) *

!!( - 4 # #

@ # 9 * 2 6 )

*

!!( 0 4 " % () &;

J 2 9&;)# # ; & 9 E 2 1 92 2

. $ 8 # B 3

176

Abstract

The interest for modelling of human actions acting on structures has been recurrent since the first accidents on suspension bridges in the nineteenth century like Broughton (1831) in the U.K. or Angers (1850) in France. The use of new materials allowing the design of slender structures, the simultaneous interest in the structural serviceability performance and accidents such as during the opening ceremony of the London Millenium Footbridge made it mandatory to carry on and in-depth analysis of the equivalent actions to be used in the numerical analysis of structures.

In this work, some studies have been carried out in order to compare the existing load models used to simulate periodic jumping and develop a new load model. Tests have been performed on a structure designed to be a gymnasium, which has natural frequencies within that range of the excitation frequencies. In addition, a finite element model of the structure has been developed. The load models have been applied on this model to calculate predictions of the structural response. Test results have been compared with predictions based on the load modelling alternatives.

Characterization of human activities on structures

a

Javier Fernández,

Alberto Fraile,

Lutz Hermanns,

Enrique Alarcón

a

Departamento de Ingeniería Mecánica, ETSII-UPM ([email protected])

Jumping load models

Model 1. Steel Construction Institute

• Periodic loads

• Fourier coefficients (group effect !)

Model 2. Oxford University (Sim)

• Statistical model

• Randomness in the phase lag among individuals

Model 3. Proposed model

• Force plate tests

• Least squares fitting

• Statistical post-processing

Methodology

Structure monitoring

• ETSII gymnasium

Modal Identification and numerical model

• Operational Modal Analysis (OMA)

• Finite Element Model

Jumping tests

• Jumping area: 12 m²

• Up to 30 jumpers

• Occupyng densities: 0.5, 1, 1.5, 2, 2.5 p/m²

• Jumping frequencies: 1.5, 2, 2.5 Hz

Numerical-experimental comparison

Project sponsored by:

(Ref: BIA2014-59321-C2-1-R)

"(#) = $ · %& + ' *+·,-.(/ · + · 0 · 12·# + 3+)

4 +=&

5

6_7,8=#_7,8

9:·/0 6; =#̅7

9 ·/0

2 2

1 1 1 2 2 2

F(t) = A cos (w(t-t ))+A cos (w(t-t ))

Mode Experimental Freq. (Hz)

Numerical

Freq. (Hz) Error (%)

1^st Floor Bending 5.74 5.71 0.6

2^st Floor Bending 6.75 6.59 2.3

3^st Floor Bending 8.52 8.20 3.8

4^st Floor Bending 10.85 10.89 0.4

PSD diagram: Model 3 (proposed) fits better to the test results for frequencies

higher than 10 Hz Running RMS: Models 1 and 2

underestimate and model 3 (proposed) overestimate the experimental values

177

ďƐƚƌĂĐƚ

/DWZKs/E't/d,&Dd,'>^^&ZdhZ^dZ^^

Kd/E&ZKDKy/>Kh>Z/E'd^d

Ă

:ĞƐƷƐ ůŽŶƐŽͲůǀĂƌĞnj͕

:ŽƐĠ ͘WĂƌƌĂͲ,ŝĚĂůŐŽ͕

ŶƚŽŶŝĂ WĂĐŝŽƐͲ

ůǀĂƌĞnj͕

Dǐ ŽŶƐƵĞůŽ,ƵĞƌƚĂ͕

&ƌĂŶĐŝƐĐŽ ĂƉĞů

Ă

d^//ʹ hWD;ũĞƐƵƐĂůŽŶƐŽĂůǀΛŐŵĂŝů͘ĐŽŵͿ

ď

d^//ʹ hWD

/sͲ ^/

DĂŝŶ ŽďũĞĐƚŝǀĞ

&Žƌ ďŽƚŚ ƐĂŵƉůĞƐ͕ ĂdžŝĂů ůŽĂĚ ĂƉƌŽdž͘ ϭϱ ŬE͘

dŚĞƌĞ ĂƌĞ ŶŽƚ ďŝŐ ĚŝĨĨĞƌĞŶĐĞƐ ŽŶ ĨƌĂĐƚƵƌĞ ƉĂƚƚĞƌŶ ;ŵŽƌĞ ƚŚĂŶ ϭϱ ƐƉĞĐŝŵĞŶƐ ƉĞƌ ƐĞƌŝĞͿ ƚŚĞ Ĩŝůŵ ĂƉƉĂƌĞŶƚůLJ ĚŽĞƐ ŶŽƚ ŝŶĨůƵĞŶĐĞ ƚŚĞ ďĞŚĂǀŝŽƌ͘

WƌŽũĞĐƚƐƉŽŶƐŽƌĞĚ ďLJ

džƉĞƌŝŵĞŶƚĂůĐĂŵƉĂŝŐŶƐƵƐĞĚĨŽƌƚŚĞƐƚƵĚLJ

/ϮϬϭϭͲϮϴϵϱϵͲϬϮͲϬϮ

WƌŽƉŽƐĞĚĞƐƚŝŵĂƚŝŽŶƉƌŽĐĞĚƵƌĞĨŽƌƚŚĞĨƌĂĐƚƵƌĞƐƚƌĞƐƐĐƵŵƵůĂƚŝǀĞĚŝƐƚƌŝďƵƚŝŽŶĨƵŶĐƚŝŽŶ

&ĂŝůƵƌĞƉƌŽďĂďŝůŝƚLJĚŝƐƚƌŝďƵƚŝŽŶĨƵŶĐƚŝŽŶ;tĞŝďƵůůͿ

ŽŵƉĂƌŝƐŽŶ ŽĨĞdžƉĞƌŝŵĞŶƚĂů͕ƐƚĂŶĚĂƌĚ ĂŶĚ&DƐƚƌĞƐƐ;ĂůƵŵŝŶƵŵ ƉůĂƚĞͿ

ŽĂdžŝĂůŽƵďůĞ ZŝŶŐdĞƐƚ͘Z_ŝϵϬŵŵ͖Z_ĞϭϮϬŵŵ͖_ƌĞĨϮϱϰϳŵŵ^Ϯ

ŽŶĐůƵƐŝŽŶƐ

ϭϭϬƐƉĞĐŝŵĞŶƐ͘džĂŵƉůĞ͗ŶŶĞĂůĞĚŐůĂƐƐĂĨƚĞƌĨƌĂĐƚƵƌĞǁŝƚŚŽƵƚ;ͿĂŶĚǁŝƚŚĨŝůŵ;&Ϳ dŚĞ ƉŽƐƚͲĨƌĂĐƚƵƌĞ ďĞŚĂǀŝŽƌ ŽĨ ƚŚĞ ůĂŵŝŶĂƚĞĚ ŐůĂƐƐĞƐ ŝƐ ŽŶĞ ŽĨ ƚŚĞ ƌĞƐĞĂƌĐŚ ƚŽƉŝĐƐ ƚŚĂƚ ĂƌĞ ďĞŝŶŐ ƐƚƵĚŝĞĚ ƚŽ ĞdžƉůĂŝŶ ƚŚĞ ůŽĂĚ ĐĂƉĂĐŝƚLJ ĂĨƚĞƌ ƚŚĞ ďƌĞĂŬ ŽĨ ƚŚĞ ĨŝƌƐƚ ƐŚĞĞƚ͘ &Žƌ ŚƵŵĂŶ ŝŵƉĂĐƚ ŝŶ ĂŶŶĞĂůĞĚ ŐůĂƐƐĞƐ͕ ƚŚĞ ĐĂƉĂĐŝƚLJ ŽĨ ďĞĂƌŝŶŐ ůŽĂĚ ŝƚ ĐĂŶ ďĞ ƵƉ ƚŽ ϯ ƚŝŵĞƐ ŚŝŐŚĞƌ ǁŝƚŚŽƵƚ ĐůĞĂƌ ĞdžƉůĂŶĂƚŝŽŶ ŽĨ ƚŚĞ ƐƚƌƵĐƚƵƌĂů ďĞŚĂǀŝŽƌ ŽĨ ƚŚĞ ƉůĂƚĞ͘ dŽ ŵĂŬĞ ĂŶ ĂƉƉƌŽdžŝŵĂƚŝŽŶ ƚŽ ƚŚĞ ƉŽƐƚͲĨƌĂĐƚƵƌĞ ƌĞƐŝƐƚĂŶĐĞ͕ Ă ĞdžƉĞƌŝŵĞŶƚĂů ƉƌŽŐƌĂŵ ƚŽ ƚĞƐƚ ĂŶŶĞĂůĞĚ ;Ϳ͕ ŚĞĂƚͲƚĞŵƉĞƌĞĚ ;,dͿ ĂŶĚ ƚŽƵŐŚĞŶĞĚ ;dͿ ŐůĂƐƐ ƉůĂƚĞƐ ŚĂƐ ďĞĞŶ ƉƌĞƉĂƌĞĚ͘ dǁŽ ĂĚĚŝƚŝŽŶĂů ƐĞƌŝĞƐ ŽĨ ĂŶŶĞĂůĞĚ ĂŶĚ ŚĞĂƚͲƚĞŵƉĞƌĞĚ͕ ǁŝƚŚ Ă ůĂLJĞƌ ŽĨ ƉŽůLJǀŝŶLJů ďƵƚLJƌĂů ĂĚŚĞƌĞĚ ũƵƐƚ ĂĨƚĞƌ ƚŚĞ ŵĂŶƵĨĂĐƚƵƌŝŶŐ ƉƌŽĐĞƐƐ͕ ŚĂǀĞ ĂůƐŽ ďĞĞŶ ŝŶĐŽƌƉŽƌĂƚĞĚ͘ ŽĂdžŝĂů ŽƵďůĞ ZŝŶŐ ǁŝƚŚ ůĂƌŐĞ ƚĞƐƚ ƐƵƌĨĂĐĞ ĂƌĞĂƐ ŝƐ ƚŚĞ ƐƚĂŶĚĂƌĚ ƚŚĂƚ ŚĂƐ ďĞĞŶ ĨŽůůŽǁĞĚ͘ dŽ ŵĂŬĞ ƚŚĞ ĐŽŵƉĂƌŝƐŽŶ ŽĨ tĞŝďƵůůΖƐ ĚŝƐƚƌŝďƵƚŝŽŶƐ ŽĨ ƚŚĞ ĚŝĨĨĞƌĞŶƚ ĨƌĂĐƚƵƌĞ ƐƚƌĞƐƐ͕ ĂŶ ŝƚĞƌĂƚŝǀĞ ƉƌŽĐĞƐƐ ďĂƐĞĚ ŽŶ ƚŚĞ ĂĐƚƵĂů ƐƚƌĞƐƐ ĚŝƐƚƌŝďƵƚŝŽŶ ŽďƚĂŝŶĞĚ ǁŝƚŚ Ă ĨŝŶŝƚĞ ĞůĞŵĞŶƚƐ ŵŽĚĞů ƵƉĚĂƚĞĚ ǁŝƚŚ ĞdžƉĞƌŝŵĞŶƚĂů ƌĞƐƵůƚƐ ŚĂƐ ďĞĞŶ ƵƐĞĚ͘ &ŝŶĂů ĐŽŵƉĂƌŝƐŽŶƐ ƐŚŽǁ Ă ŐƌĞĂƚ ƐƚƌĞƐƐ ŝŵƉƌŽǀĞŵĞŶƚ ĨŽƌ ƚŚĞ ĂŶŶĞĂůĞĚ ŐůĂƐƐ ƉůĂƚĞƐ ;ϰϱ йͿ͕ ĂŶĚ Ă ŵŝŶŽƌ ŝŶĐƌĞŵĞŶƚ ĨŽƌ ƚŚĞ ŚĞĂƚͲƚĞŵƉĞƌĞĚ ;Ϯϱ йͿ͘

Ϭ͘Ϭ Ϭ͘Ϯ Ϭ͘ϰ Ϭ͘ϲ Ϭ͘ϴ ϭ͘Ϭ

ϱ ϭϱ Ϯϱ ϯϱ

&ƌĂĐƚƵƌĞWƌŽďĂďŝůŝƚLJ

džŝĂů>ŽĂĚ΀ŬE΁

ŶŶĞĂůĞĚ

WĨŽďƐĞƌǀĂĚŽ WĨ/ƚĞƌĂĐŝſŶ/ŶŝĐŝĂů WĨ/ƚĞƌĂĐŝſŶ&ŝŶĂů ,ĞĂƚ dĞŵƉĞƌĞĚ

ǁŝƚŚ &ŝůŵ

dĞŵƉĞƌĞĚ

ŶŶĞĂůĞĚ ǁŝƚŚ &ŝůŵ

,ĞĂƚdĞŵƉĞƌĞĚ

&ƉŽďƐĞƌǀĞĚ

&ƉŝŶŝƚŝĂů ŝƚĞƌĂƚŝŽŶ

&ƉĨŝŶĂůŝƚĞƌĂƚŝŽŶ

$N1

36  36 

$)N1 Ϭ

ϭϬϬ ϮϬϬ ϯϬϬ

Ϭ ϭϬ ϮϬ ϯϬ ϰϬ

ZĂĚŝĂů^ƚƌĞƐƐ΀DWĂ΁

džŝĂů>ŽĂĚ΀ŬE΁

ʍƌĂĚŝĂůĞŶƐĂLJŽĐĞŶƚƌŽ

ʍEŽƌŵĂ

ʍƌĂĚŝĂůĐĞŶƚƌŽ

ʍƌĂĚŝĂůĂŶŝůůŽ

dĞƐƚ Ͳ ʍ_ƌĐĞŶƚĞƌ

^ƚĂŶĚĂƌĚ Ͳ ʍ

&DͲ ʍ_ƌĐĞŶƚĞƌ

&DͲ ʍ_ƌƌŝŶŐ

Ϭ ϭϬϬϬ ϮϬϬϬ ϯϬϬϬ

Ϭ ϭϬ ϮϬ ϯϬ ϰϬ

ɀŝĐƌŽƐƚƌĂŝŶ

džŝĂů>ŽĂĚ ΀ŬE΁

İUB&HQWURB0()>ȝİ@

İUB&HQWUR>ȝİ@

İUB$QLOORB0()>ȝİ@

İUB$QLOOR>ȝİ@

İșB$QLOORB0()>ȝİ@

İșB$QLOOR>ȝİ@

&D Ͳ ɸƌĐĞŶƚĞƌ dĞƐƚ Ͳ ɸ_ƌĐĞŶƚĞƌ

&D Ͳ ɸ_ƌƌŝŶŐ dĞƐƚ Ͳ ɸƌƌŝŶŐ

&D Ͳ ɸɽ ƌŝŶŐ dĞƐƚ Ͳ ɸɽ ƌŝŶŐ

Ϭ ϭ Ϯ ϯ ϰ

ŝƐƉůĂĐĞŵĞŶƚ;ŵŵͿ

ŝƐƉůĂĐĞŵĞŶƚĂƚƚŚĞĐŽƌŶĞƌ

;ŵĞĂƐƵƌĞĚǁŝƚŚ>sdͿ

dŚĞ ĐŽŶƚŽƵƌƐ ŽĨ ƐƵƌǀŝǀĂů ƉƌŽďĂďŝůŝƚLJ ŝŶ ƚŚĞ &D ŽĨ Ь ŽĨ ƉůĂƚĞ͕ ǁŝƚŚ ƚŚĞ tĞŝďƵůů͛Ɛ ƉĂƌĂŵĞƚĞƌƐ ĨŽƌ ĂŶŶĞĂůĞĚ ŐůĂƐƐ

;Ϳ͕ ƚŚĞ ƚŽƚĂů ƐƵƌǀŝǀĂů ƉƌŽďĂďŝůŝƚLJ ŝƐ Ϭ͘Ϭϱ ĨŽƌ ƚŚĞ ĂŶŶĞĂůĞĚ ŐůĂƐƐ ǁŝƚŚ Ĩŝůŵ ;&Ϳ͕ ƐƵƌǀŝǀĂů ƉƌŽďĂďŝůŝƚLJ ŽďƚĂŝŶĞĚ ŝƐ ŽĨ Ϭ͘ϯϱ͘ ^ŝŶĐĞ ƐƚƌĞƐƐ ĚŝƐƚƌŝďƵƚŝŽŶ ŝƐ ƉƌĂĐƚŝĐĂůůLJ ƚŚĞ ƐĂŵĞ͕ ĚŝĨĨĞƌĞŶĐĞ ŝƐ ŝŶ ƚŚĞ ĚŝƐƚƌŝďƵƚŝŽŶ ĨƵŶĐƚŝŽŶ ĨŽƌ ƚŚĞ ĨƌĂĐƚƵƌĞ ƐƚƌĞƐƐ͘

$)

$

dŚŝƐ ĨŝŐƵƌĞƐ ƐŚŽǁ ƚŚĞ ĐŽŵƉĂƌŝƐŽŶ ďĞƚǁĞĞŶ ĞdžƉĞƌŝŵĞŶƚĂů ǀĂůƵĞƐ͕ ƚŚĞ ŽďƚĂŝŶĞĚ ǁŝƚŚ ƐƚĂŶĚĂƌĚ ĂŶĚ ƚŚĞ ŵŽĚĞů ƉƌŽƉŽƐĞĚ ƚŚĞ ƐƚƌĂŝŶ ĚŝƐƚƌŝďƵƚŝŽŶ ĐĂŶ ďĞ ĐĂůĐƵůĂƚĞĚ͘

hŶĚĞƌƐƚĂŶĚŝŶŐ ƚŚĞ ŝŶĨůƵĞŶĐĞ ŽĨ ŝŶƚĞƌůĂLJĞƌ ƚŽ ƉƌĞǀĞŶƚ ĨƌĂĐƚƵƌĞ ŽĨ ƚŚĞ ŐůĂƐƐ ƉĂŶĞ ĚƵĞ ƚŽ ƚŚĞ ŵŽĚŝĨŝĐĂƚŝŽŶ ŽŶ ĚŝƐƚƌŝďƵƚŝŽŶ ĨƵŶĐƚŝŽŶ ŽĨ ĨƌĂĐƚƵƌĞ ƐƚƌĞƐƐ͘

ϱϬ ϭϬϬ ϭϱϬ ϮϬϬ ϮϱϬ ϯϬϬ ϯϱϬ

ůͲ&Ds

>ͲDs

ůͲ,Ds

>Ͳ,&Ds

>ͲdDs

ϭ Ϭ͘ϴ Ϭ͘ϲ Ϭ͘ϰ Ϭ͘Ϯ

ϬϱϬϭϬϬϭϱϬϮϬϬϮϱϬϯϬϬϯϱϬ

^ƚƌĞƐƐ΀DWĂ΁

&ƌĂĐƚƵƌĞWƌŽďĂďŝůŝƚLJ

&

,d ,d&

d

&ŝŐƵƌĞ ĨƌŽŵ ƚŚĞ ůĞĨƚ ƉƌĞƐĞŶƚƐ ƚŚĞ ƌĞƐƵůƚƐ ŽĨ ƚŚĞ ĨĂŝůƵƌĞ ůŽĂĚ ƉƌŽďĂďŝůŝƚLJ ŽďƚĂŝŶĞĚ ĨƌŽŵ͗

Ͳ ƚŚĞ ƚĞƐƚ͖

Ͳ &D ǁŝƚŚ ŝŶŝƚŝĂů ĐƵŵƵůĂƚŝǀĞ ĚŝƐƚƌŝďƵƚŝŽŶ ĨƵŶĐƚŝŽŶ ĨŽƌ ƚŚĞ ĨƌĂĐƚƵƌĞ ƐƚƌĞƐƐ͖ ĂŶĚ

Ͳ &D ǁŝƚŚ ĨŝŶĂů ĐƵŵƵůĂƚŝǀĞ ĚŝƐƚƌŝďƵƚŝŽŶ ĨƵŶĐƚŝŽŶ͘

/ƚ ĐĂŶ ďĞ ŽďƐĞƌǀĞĚ ƚŚĂƚ ƚŚĞ ǀĂůƵĞƐ ĐĂůĐƵůĂƚĞĚ ǁŝƚŚ ƚŚĞ ƉƌŽƉŽƐĞĚ ŵĞƚŚŽĚ ƉƌĞƐĞŶƚƐ Ă ďĞƚƚĞƌ ĂƉƉƌŽĂĐŚ͘

WƌŽƉŽƐĞĚ ŵĞƚŚŽĚ ŝƐ ďĂƐĞĚ ŝŶ ƚŚĞ ĞƐƚŝŵĂƚŝŽŶ ŽĨ ĞĨĨĞĐƚŝǀĞ ĂƌĞĂ ĨŽƌ ĞĂĐŚ ƐƉĞĐŝŵĞŶ͘

 ŵĞƚŚŽĚ ƚŚĂƚ ĂůůŽǁƐ ƚŽ ĐŽŵƉĂƌĞ ƚŚĞ ƌĞƐƵůƚƐ ŽĨ ĐƵŵƵůĂƚŝǀĞ ĚŝƐƚƌŝďƵƚŝŽŶ ĨƵŶĐƚŝŽŶ ŽĨ ŐůĂƐƐ ĨƌĂĐƚƵƌĞ ƐƚƌĞƐƐ ŽďƚĂŝŶĞĚ ŚĂƐ ďĞĞŶ ƉƌŽƉŽƐĞĚ͘ ĂƐĞ ŽŶ ƚŚĂƚ͕

ĚŝĨĨĞƌĞŶƚ ŚLJƉŽƚŚĞƐĞƐ ĐĂŶ ďĞ ƐƚƵĚŝĞĚ͖ ĨŽƌ ĞdžĂŵƉůĞ͕ ƚŚĞ ƉƌĞĐŝƐŝŽŶ ůĞǀĞů ŝŶ ƚŚĞ ĂƉƉůŝĐĂƚŝŽŶ ŽĨ ƚŚĞ ůŽĂĚ ĂŶĚ ƉƌĞƐƐƵƌĞ ĂŶĚ ƚŚĞ ƌĞůĂƚŝŽŶ ǁŝƚŚ ƚŚĞ ĂƐƐŝŐŶĞĚ ĨƌĂĐƚƵƌĞ ƐƚƌĞƐƐ͕ ƚŚĞ ŵĞƚŚŽĚ ŽĨ ƐƚƌĞƐƐ ĐŽŵďŝŶĂƚŝŽŶ Žƌ ƚŚĞ ĞĨĨĞĐƚ ŽĨ ƚŚĞ ŵĞŵďƌĂŶĞ ƐƚƌĞƐƐ ƚŚĂƚ ĂƉƉĞĂƌƐ ŝŶ ƚŚĞ ŶŽŶ ůŝŶĞĂƌ ĐĂƐĞƐ͘178

ďƐƚƌĂĐƚ

dŚŝƐ ƉĂƉĞƌ ƐŚŽǁƐ ƚŚĞ ƌĞƐƵůƚƐ ŽĨ ĂŶ ĞdžƉĞƌŝŵĞŶƚĂů ƚĞƐƚ ĐĂŵƉĂŝŐŶ ŽŶ ĚŝĨĨĞƌĞŶƚ ŐůĂƐƐ ƐůĂďƐ ŽĨ ƚŚĞ ƐĂŵĞ ĚŝŵĞŶƐŝŽŶƐ ;ϴϳϲ ŵŵ džϭϵϯϴ ŵŵͿ ƵŶĚĞƌ ŚƵŵĂŶ ŝŵƉĂĐƚ ůŽĂĚŝŶŐ ĂĐĐŽƌĚŝŶŐ ƚŽ hEͲE ϭϮϲϬϬ͗ ƚĞŵƉĞƌĞĚ ;dϭϬͿ ĂŶĚ ĂŶŶĞĂůĞĚ ůĂŵŝŶĂƚĞĚ ;>ϱϱͿ ƐĂĨĞƚLJ ŐůĂƐƐĞƐ͘ dŚƌŽƵŐŚ Ă ŵŽĚĂů ƚĞƐƚ͕ Ă ĐůĂƐƐŝĐĂů ĂƉƉƌŽĂĐŚ ĂŶĚ ĂŶ ŽƉĞƌĂƚŝŽŶĂů ŵŽĚĂů ƚĞƐƚ ƵƐŝŶŐ ŚƵŵĂŶ ŝŵƉĂĐƚ ƚĞƐƚ͕ ŝƚ ĐĂŶ ďĞ ƐŚŽǁŶ ŚŽǁ ƚŚĞ ďĞŚĂǀŝŽƌ ŽĨ ůĂŵŝŶĂƚĞĚ ŐůĂƐƐ ĐĂŶ ďĞ ŵŽĚĞůĞĚ ĂƐ ŵŽŶŽůŝƚŚŝĐ͕ ĨŽƌ ƚŚŝƐ ƉŚĞŶŽŵĞŶĂ͕

ǁŚĞŶ ƚŚĞ ŝŶĨůƵĞŶĐĞ ŽĨ ƚŚĞ ƚĞŵƉĞƌĂƚƵƌĞ ŝƐ ŶĞŐůĞĐƚĞĚ͘

>D/Ed'>^^,s/KhZ'/E^d

,hDE/DWd

Ă

:ŽƐĠ ͘WĂƌƌĂͲ,ŝĚĂůŐŽ͕

:ĞƐƷƐ ůŽŶƐŽ͕

:ĂĐŽďŽ ^ĄŶĐŚĞnj͕

ď

ŶƚŽŶŝĂ WĂĐŝŽƐͲůǀĂƌĞnj͕

DǐŽŶƐƵĞůŽ,ƵĞƌƚĂ

Ă

d^//Ͳ hWD;ũŽƐĞĂŶƚŽŶŝŽƉĂƌƌĂŚŝĚĂůŐŽΛŐŵĂŝůͿ

ď

d^//Ͳ hWD

DĂŝŶŽďũĞĐƚŝǀĞƐ

‡ dŽ ƉƌĞƐĞŶƚ ƚŚĞ ƐƚƌƵĐƚƵƌĂů ďĞŚĂǀŝŽƌ ŽĨ ŐůĂƐƐ ƉůĂƚĞƐ͕ ĂŶŶĞĂůĞĚ ůĂŵŝŶĂƚĞĚ ;ǁŝƚŚ Ϭ͘ϯϲ ŵŵ Ws ŝŶƚĞƌůĂLJĞƌͿ ĂŶĚ ƚĞŵƉĞƌĞĚ ŵŽŶŽůŝƚŚŝĐ͕ ƚŚƌŽƵŐŚ ƐĞǀĞƌĂů ĚLJŶĂŵŝĐ ƚĞƐƚƐ͕ ŵŽĚĂů ƚĞƐƚ͕ ĂŶĚ ŚƵŵĂŶ ŝŵƉĂĐƚ ƚĞƐƚ͘

‡ dŽ ƉƌŽǀĞ ƚŚĞ ƚLJƉŝĐĂů ĂƉƉƌŽĂĐŚ ŽĨ ŵŽŶŽůŝƚŚŝĐ ďĞŚĂǀŝŽƌ ĨŽƌ ůĂŵŝŶĂƚĞĚ ŐůĂƐƐ ĨŽƌ ĚLJŶĂŵŝĐ ƚĞƐƚ ;ĨƌĞƋƵĞŶĐLJ ŚŝŐŚĞƌ ƚŚĂŶ ϱ ,njͿ͕

ŝŶĐůƵĚŝŶŐ ŶŽŶůŝŶĞĂƌ ĞĨĨĞĐƚ ŽŶ ŚƵŵĂŶ ŝŵƉĂĐƚ͕ ĂƐ ǀĂůŝĚ͘

‡ dŽ ƵƐĞ ƌĞƐŽƵƌĐĞƐ ŽĨ KƉĞƌĂƚŝŽŶĂů DŽĚĂů ŶĂůLJƐŝƐ ĨŽƌ ďŽƚŚ͗ ƚŽ ĚĞĨŝŶĞ ƚŚĞ ĨƌĞƋƵĞŶĐLJ ƌĞƉƌĞƐĞŶƚĂƚŝŽŶ͕ ĂŶĚ ƚŽ ĞƐƚŝŵĂƚĞ ƚŚĞ ĨƌĞƋƵĞŶĐLJ ĂŶĚ ĚĂŵƉŝŶŐ͘

ϭ͘ͲϮ ϭ͘Ͳϭ ϭ͘нϬ ϭ͘нϭ ϭ͘нϮ

Ϭ ϭϬϬ ϮϬϬ ϯϬϬ

ĐĐĞůĞƌĂƚŝŽŶͬ&ŽƌĐĞ;ŵͬEƐϮͿ

&ƌĞƋƵĞŶĐLJ;,njͿ

dϭϬ>ϱϱ

ϭ͘Ͳϯ ϭ͘ͲϮ ϭ͘Ͳϭ ϭ͘нϬ ϭ͘нϭ ϭ͘нϮ

Ϭ ϮϬ ϰϬ ϲϬ ϴϬ ϭϬϬ

ĐĐĞůĞƌĂƚŝŽŶͬ&ŽƌĐĞ;ŵͬEƐϮͿ

&ƌĞƋƵĞŶĐLJ;,njͿ

dϭϬ

>ϱϱ

ͲϮϬϬ ϱϬϬ ϭ͕ϮϬϬ ϭ͕ϵϬϬ Ϯ͕ϲϬϬ ϯ͕ϯϬϬ

ͲϯϱϬ ͲϮϱϬ ͲϭϱϬ ͲϱϬ ϱϬ ϭϱϬ

ϯ ϯ͘ϰ ϯ͘ϴ ϰ͘Ϯ

ʅƐƚƌĂŝŶ

ĐĐĞůĞƌĂƚŝŽŶ;ŵͬƐϮͿ

dŝŵĞ;ƐͿ

dϭϬWůĂƚĞĐĐ dϭϬWĞŶĚĐĐ

>ϱϱWůĂƚĞĐĐ

>ϱϱWĞŶĚĐĐ dϭϬ,ʅƐƚƌĂŝŶ dϭϬsʅƐƚƌĂŝŶ

>ϱϱ,ʅƐƚƌĂŝŶ

>ϱϱsʅƐƚƌĂŝŶ W_W_

W_

W_ W_ W_

KŶ ĨƌĞĞͲĨƌĞĞ ĐŽŶĚŝƚŝŽŶƐ͕ dϭϬ ĂŶĚ >ϱϱ ƐƉĞĐŝŵĞŶƐ ƉƌĞƐĞŶƚ ĂůŵŽƐƚ ƚŚĞ ƐĂŵĞ ĨƌĞƋƵĞŶĐŝĞƐ ďƵƚ ĚŝĨĨĞƌĞŶƚ ĚĂŵƉŝŶŐ ĚƵĞ ƚŽ ƚŚĞ Ws͘ KŶ ŝŵƉĂĐƚ ƚĞƐƚ ďŽƵŶĚĂƌLJ ĐŽŶĚŝƚŝŽŶƐ͕ ƚŚĞ ĨŝƌƐƚ ŵŽĚĂů ĨƌĞƋƵĞŶĐLJ ĂŶĚ ƐƚŝĨĨŶĞƐƐ ĂƌĞ ƚŚĞ ƐĂŵĞ͕ ǁŚŝůĞ ƚŚĞ ŵŽĚĂů ĚĂŵƉŝŶŐ ŝƐ ƐŝŐŶŝĨŝĐĂŶƚůLJ ĚŝĨĨĞƌĞŶƚ͘ dŝŵĞ ŚŝƐƚŽƌŝĞƐ ŽĨ ƚŚĞ ŚŽƌŝnjŽŶƚĂů ĂŶĚ ǀĞƌƚŝĐĂů ŵŝĐƌŽƐƚƌĂŝŶƐ ĂŶĚ ƉĞŶĚƵůƵŵ ĂĐĐĞůĞƌĂƚŝŽŶ ĂƌĞ ƉƌĂĐƚŝĐĂůůLJ ƚŚĞ ƐĂŵĞ͘ Ŷ ŝŶ ĚĞƉƚŚ ƐƚƵĚLJ ŽĨ ƚŚĞ ĞǀŽůƵƚŝŽŶ ŽĨ WŽǁĞƌ ^ƉĞĐƚƌĂů ĞŶƐŝƚLJ ĚƵƌŝŶŐ ƚŚĞ ŝŵƉĂĐƚ ƐŚŽǁƐ Ă ŐƌĞĂƚ ĐŽƌƌĞƐƉŽŶĚĞŶĐĞ ŽĨ ƌĞƐƵůƚƐ͘ ůĂƐƐŝĐĂů ĂŶĚ ŽƉĞƌĂƚŝŽŶĂů ŵŽĚĂů ƉƌŽǀĞ ƚŚĂƚ ůĂŵŝŶĂƚĞĚ ŐůĂƐƐ ďĞŚĂǀĞ ůŝŬĞ ŵŽŶŽůŝƚŚŝĐ ŽŶĞ ƵŶĚĞƌ ŚƵŵĂŶ ŝŵƉĂĐƚ͘

WƌŽũĞĐƚƐƉŽŶƐŽƌĞĚďLJ

džƉĞƌŝŵĞŶƚĂůƌĞƐƵůƚƐ

/ϮϬϭϭͲϮϴϵϱϵͲϬϮͲϬϮ

/ƚ ŝƐ Ă ŐƌĞĂƚ ƐŝŵŝůĂƌŝƚLJ ďĞƚǁĞĞŶ ƚŚĞ ƚǁŽ ƉůĂƚĞƐ͘ dŚĞ ŵĂŝŶ ĚŝĨĨĞƌĞŶĐĞ ŝƐ ĨŽƵŶĚ ŝŶ ƚŚĞ ƉĞĂŬ ƐŚĂƉĞƐ͖ ĨŽƌ ůĂŵŝŶĂƚĞĚ ŐůĂƐƐ͕ ĂƌĞ ƐŽĨƚĞŶĞĚ ĚƵĞ ƚŽ ŝƚƐ

ŚŝŐŚĞƌ ĚĂŵƉŝŶŐ

ƉƌŽĚƵĐĞĚ ďLJ ƚŚĞ Ws͘

DŽĚĂů ƉĂƌĂŵĞƚĞƌƐ͗

ŶĂƚƵƌĂů ĨƌĞƋƵĞŶĐŝĞƐ ĂŶĚ ĚĂŵƉŝŶŐ͘

KƉĞƌĂƚŝŽŶĂůŵŽĚĂůĂŶĂůLJƐŝƐǁŝƚŚŝŵƉĂĐƚƚĞƐƚĚĂƚĂ

^ƚĞƉϭ͘&Z&ĨƌĞĞͲĨƌĞĞĐŽŶĚŝƚŝŽŶƐdϭϬǀƐ>ϱϱ

^ƚĞƉϮ͘&Z&ƐƵƉƉŽƌƚĐŽŶĚŝƚŝŽŶƐdϭϬǀƐ>ϱϱ DŽĚĂů ƉĂƌĂŵĞƚĞƌƐ͗

ƚŚĞ ƉƌĞǀŝŽƵƐ ĂŶĚ ƉŽŝŶƚ ŵŽĚĂů ƐƚŝĨĨŶĞƐƐ ŽŶ ƚŚĞ ŵŝĚĚůĞ ŽĨ ƚŚĞ ƉůĂƚĞ͘ dŚĞ ĚŝĨĨĞƌĞŶĐĞ ŝŶ ĨƌĞƋƵĞŶĐLJ ďĞƚǁĞĞŶ

>ϱϱ ĂŶĚ dϭϬ ŝƐ Ϭ͘ϴй ͕ ĂŶĚ ŝŶ ƐƚŝĨĨŶĞƐƐ ŝƐ ϴ͘ϳй͘ ĂŵƉŝŶŐ ŝƐ ĐůĞĂƌůLJ ĚŝĨĨĞƌĞŶƚ͗ ф Ϭ͘ϴй ĨŽƌ dϭϬ͕ ĂŶĚ Ϯ͘Ϭй ĨŽƌ >ϱϱ͘

^ƚĞƉϯ͘,ƵŵĂŶŝŵƉĂĐƚƚŝŵĞŚŝƐƚŽƌLJdϭϬǀƐ>ϱϱ

ĐĐĞůĞƌĂƚŝŽŶƐ ;ƵƉƉĞƌ ŐƌĂƉŚͿ ĂŶĚ ƐƚƌĂŝŶ ;ůŽǁĞƌ ŐƌĂƉŚͿ ƌĞŐŝƐƚĞƌƐ͘ dŚĞ ƉůĂƚĞ ďĞŚĂǀŝŽƌ ĐĂŶ ďĞ ƵŶĚĞƌƐƚŽŽĚ ďLJ ůŽŽŬŝŶŐ Ăƚ ƚŚĞ ƚŝŵĞ ŚŝƐƚŽƌŝĞƐ͘ dŚĞ ďĞŚĂǀŝŽƌ ŽĨ ďŽƚŚ ƉůĂƚĞƐ ŝƐ ƋƵŝƚĞ ƐŝŵŝůĂƌ͕ ĞƐƉĞĐŝĂůůLJ ĚƵƌŝŶŐ ƚŚĞ ŝŵƉĂĐƚ͘

sĞƌƚŝĐĂů ůŝŶĞƐ ĐŽƌƌĞƐƉŽŶĚƐ ǁŝƚŚ ŬĞLJ ƉŽŝŶƚƐ͗ ƚ_ϭ ŝƐ ƚŚĞ ďĞŐŝŶŶŝŶŐ ŽĨ ƚŚĞ ĐŽŶƚĂĐƚ͕ ƚ_Ϯƚ_Ϯ^ΎĂŶĚ ƚ_ϯĐŽƌƌĞƐƉŽŶĚƐ ǁŝƚŚ ĚŝĨĨĞƌĞŶƚ ĚĞĨŝŶŝƚŝŽŶƐ ŽĨ ƚŚĞ ŝŵƉĂĐƚ ĚƵƌĂƚŝŽŶ͕ ƚ_ϰŝƐ ǁŚĞŶ ƚŚĞ ƉĞŶĚƵůƵŵ ƌĞƐŝĚƵĂů ĂĐĐĞůĞƌĂƚŝŽŶ ŝƐ ŶĞĂƌ njĞƌŽ ĂŶĚ ƚ_ϱŝƐ ǁŚĞŶ ƚŚĞ ŐůĂƐƐ ǀŝďƌĂƚŝŽŶ ŝƐ ĨƌĞĞ͘ dŚĞ ĐŽŶƚĂĐƚ ďĞŐŝŶƐ͕ ĂŶĚ ĨƌŽŵ ϱϬŵƐ

;ƚ_ϭͲƚ_ϮͿ ĂƉƉƌŽdžŝŵĂƚĞůLJ͕ ƚŚĞ ƉĞŶĚƵůƵŵ ĂŶĚ ƉůĂƚĞƐ ŝŶƚĞƌĂĐƚ ƚŽŐĞƚŚĞƌ ǁŝƚŚ ŶŽŶůŝŶĞĂƌ ĞĨĨĞĐƚ ĚƵĞ ƚŽ ƚŚĞ ůĂƌŐĞ ĚŝƐƉůĂĐĞŵĞŶƚ͘ >ŽƐŝŶŐ ƚŚĞ ĐŽŶƚĂĐƚ ƉƌŽĚƵĐĞƐ Ă ŶĞǁ ŶŽŶůŝŶĞĂƌ ďĞŚĂǀŝŽƌ ƚŚĂƚ ŝƐ ƌĞĨůĞĐƚĞĚ ŽŶ ƚŚĞ ƉůĂƚĞ ĂĐĐĞůĞƌĂƚŝŽŶ͕ ƚŚĂƚ ƉƌĞƐĞŶƚƐ ƚŚĞ ĐŽŶƚƌŝďƵƚŝŽŶƐ ŽĨ ƐĞǀĞƌĂů ŵŽĚĞƐ ;ƚ_ϰͿ͘ ^ŽŵĞ ŵŝůůŝƐĞĐŽŶĚƐ ĂĨƚĞƌ ;ƚ_ϱͿ͕ ƚŚĞ ƌĞƐƉŽŶƐĞ ĐŽƌƌĞƐƉŽŶĚƐ ǁŝƚŚ Ă ĨƌĞĞ ǀŝďƌĂƚŝŽŶ ŽĨ ƚŚĞ ƉůĂƚĞ͘

,ƵŵĂŶŝŵƉĂĐƚŝƐĐŚĂƌĂĐƚĞƌŝƐĞĚďLJĨƌĞƋƵĞŶĐŝĞƐƐŵĂůůĞƌƚŚĂŶϭϬϬ,nj͘

ŽŶĐůƵƐŝŽŶƐ

ϭ͘Ͳϳ ϭ͘Ͳϱ

ϭ͘Ͳϯ ^dϭϬdϭdϭϬdϮ

>ϱϱdϭ

>ϱϱdϮ

ϭ͘Ͳϵ ϭ͘Ͳϳ ϭ͘Ͳϱ ϭ͘Ͳϯ

Ϭ Ϯϱ ϱϬ ϳϱ ϭϬϬ

dϭϬdϰ

>ϱϱdϰ

ƵƌŝŶŐ ƚŚĞ ŝŵƉĂĐƚ ŝƚ ŝƐ ŶŽƚ ƉŽƐƐŝďůĞ ƚŽ ŽďƚĂŝŶ ƚŚĞ &Z& ďĞĐĂƵƐĞ ŶŽ ĨŽƌĐĞ ŵĞĂƐƵƌĞŵĞŶƚ ŝƐ ĂǀĂŝůĂďůĞ͘ dŚĞ WŽǁĞƌ ^ƉĞĐƚƌĂů ĞŶƐŝƚLJ ;W^Ϳ ŝƐ ƐĞůĞĐƚĞĚ ĨŽƌ ƚŚĞ ĨƌĞƋƵĞŶĐLJ ƌĞƉƌĞƐĞŶƚĂƚŝŽŶ ŽĨ ƚŚĞ ƉŚĞŶŽŵĞŶĂ͘ dŚĞ W^ ĐŽƌƌĞƐƉŽŶĚŝŶŐ ƚŽ dϭ ;ƚ_ϱʹ ƚ_ϭͿ ĂŶĚ dϮ ;ƚ_ϱʹ ƚ_ϮΎͿ ĂŶĚ dϰ ;ƚ_ϱʹ ƚ_ϯͿ ĂƌĞ ƐĞůĞĐƚĞĚ ƚŽ ĐŽŵƉĂƌĞ ƚŚĞ ďĞŚĂǀŝŽƵƌ ŽĨ dϭϬ ĂŶĚ >ϱϱ ƉůĂƚĞƐ͘

ƵĞ ƚŽ ƚŚĞ ĚŝĨĨŝĐƵůƚŝĞƐ ŝŶŚĞƌĞŶƚ ƚŽ ƚŚĞ ŶŽŶůŝŶĞĂƌ ďĞŚĂǀŝŽƵƌ ƚŚĞ ĐŽŵƉƵƚĞĚ W^͕ ŝƚ ŝƐ ĚŝĨĨŝĐƵůƚ ƚŽ ŽďƚĂŝŶ ƚŚĞ ŵŽĚĂů ĐŚĂƌĂĐƚĞƌŝƐƚŝĐƐ ǁŝƚŚ ƚŚĞ ƵƐƵĂů ĂƉƉƌŽĂĐŚ͘  ƐŝŵƉůŝĨŝĞĚ ŵĞƚŚŽĚ ŝƐ ƵƐĞĚ ƚŽ ĂĚũƵƐƚ ƚŚĞ ƉĂƌĂŵĞƚĞƌƐ ŽĨ ƚŚĞ ĨŝƌƐƚ ŵŽĚĞ͘ dŚĞ ĨƌĞƋƵĞŶĐLJ͕ ŵĂƐƐ͕ ĚĂŵƉŝŶŐ ŽĨ ƚŚĞ ĨŝƌƐƚ ŵŽĚĞ ĂŶĚ ƚŚĞ ƌĞƐŝĚƵĂů ĐŽŶƚƌŝďƵƚŝŽŶ ŽĨ ƚŚĞ ŽƚŚĞƌ ŵŽĚĞƐ ĂƌĞ ƚŚĞ ĚĞƐŝŐŶ ǀĂƌŝĂďůĞƐ ŽďƚĂŝŶĞĚ ǁŝƚŚ ŽƉƚŝŵŝnjĂƚŝŽŶ ƉƌŽĐĞƐƐ͕ ŵŝŶŝŵŝnjŝŶŐ ƚŚĞ ŵĞĂŶ ƐƋƵĂƌĞ ĞƌƌŽƌ ŽĨ ƚŚĞ W^͘ ƋƵĂƚŝŽŶ ϭ ŝƐ ƵƐĞĚ ĨŽƌ ƚŚĞ W^ ĂŶĚ ƚŚĞ ŝŶŝƚŝĂů ǀĂůƵĞƐ ĨŽƌ ƚŚĞ ƉĂƌĂŵĞƚĞƌƐ ĂƌĞ ĞƐƚŝŵĂƚĞĚ ǁŝƚŚ ĐůĂƐƐŝĐĂů ĂƉƉƌŽĂĐŚ͘





k c i m k

A

H m ¸

¹

¨ ·

©

§

˜

˜



˜

 

 ˜

Z Z

Z Z

W^ŝŵƉĂĐƚϮϬϬŵŵʹ dϭϬǀƐ͘>ϱϱ

&ƌĞƋƵĞŶĐLJ;,njͿ

ĐĐĞůϮͬ&ƌĞƋ;ŵϮͬƐϰ,njͿ