Professional Documents
Culture Documents
Download textbook Operations Research Proceedings 2013 Selected Papers Of The International Conference On Operations Research Or2013 Organized By The German Operations Research Society Gor The Dutch Society Of Operatio ebook all chapter pdf
Download textbook Operations Research Proceedings 2013 Selected Papers Of The International Conference On Operations Research Or2013 Organized By The German Operations Research Society Gor The Dutch Society Of Operatio ebook all chapter pdf
Download textbook Operations Research Proceedings 2013 Selected Papers Of The International Conference On Operations Research Or2013 Organized By The German Operations Research Society Gor The Dutch Society Of Operatio ebook all chapter pdf
Dennis Huisman
Ilse Louwerse
Albert P. M. Wagelmans Editors
Operations Research
Proceedings 2013
Selected Papers of the International
Conference on Operations Research,
OR2013, organized by the German
Operations Research Society (GOR),
the Dutch Society of Operations
Research (NGB) and Erasmus University
Rotterdam, September 3–6, 2013
Albert P. M. Wagelmans
Editors
Operations Research
Proceedings 2013
Selected Papers of the International
Conference on Operations Research,
OR2013, organized by the German
Operations Research Society (GOR), the
Dutch Society of Operations Research (NGB)
and Erasmus University Rotterdam,
September 3–6, 2013
123
Editors
Dennis Huisman
Ilse Louwerse
Albert P. M. Wagelmans
Econometric Institute, Erasmus School of
Economics
Erasmus University Rotterdam
Rotterdam
The Netherlands
ISSN 0721-5924
ISBN 978-3-319-07000-1 ISBN 978-3-319-07001-8 (eBook)
DOI 10.1007/978-3-319-07001-8
Springer Cham Heidelberg New York Dordrecht London
v
vi Preface
vii
viii Contents
Abstract A solution of the traveling salesman problem (TSP) with n nodes consists
of n edges which form a shortest tour. In our approach we compute an upper bound
u for the longest edge which could be in an optimal solution. This means that every
edge longer than this bound cannot be in an optimal solution. The quantity u can be
computed in polynomial time. We have applied our approach to different problems of
the TSPLIB (library of sample instances for the TSP). Our bound does not necessarily
improve the fastest TSP-algorithms. However, the reduction of the number of edges
might be useful for certain instances.
1 Introduction
The traveling salesman problem (TSP) is one of the most studied problems in com-
binatorial optimization and has got applications in many different areas. The TSP
consists of finding a shortest tour in a complete graph whose edges (i,j) have cost
(distance) ci j . A comprehensive treatment of the traveling salesman problem can be
found in [3].
In this paper we do not assume that the cost matrix is symmetric. However, our
figures will refer to symmetric instances. We consider a dual relaxation of the original
problem—the assignment problem A based on the same cost matrix. The result is a
dual relaxation, possibly with subtours, as shown in Fig. 1. This problem can also
be solved by any code for the assignment problem, e.g. [1].
n
n
A: min ci j xi j (1)
i=1 j=1
s.t.
n
xi j = 1 1≤ j ≤n (2)
i=1
n
xi j = 1 1≤i ≤n (3)
j=1
0 ≤ xi j ≤ 1 1 ≤ i, j ≤ n (4)
This optimal solution can be transformed into a tour as shown in the figure below.
The value of the objective function of the solution in this example is 2,744 and it is an
upper bound for the optimal solution (Fig. 2). By using the Lin-Kernighan heuristic
[4] we can obtain an even better upper bound 2,726. The optimal value of the dual
heuristic is 2,426. Hence, the length of an optimal tour is between these two values.
In this paper we introduce a new relaxation A∞ of the TSP. Due to inequality 6 and 7
every node must have at least one adjacent edge and at most two adjacent edges.
Equation 8 assures that there are exactly n − 1 edges.
Computing an Upper Bound for the Longest Edge in an Optimal TSP-Solution 3
n
n
A∞ : min ci j xi j (5)
i=1 j=1
s.t.
n
n
xi j + x ji ≤ 2 1≤ j ≤n (6)
i=1 i=1
n n
xi j + x ji ≥ 1 1≤ j ≤n (7)
i=1 i=1
n
n
xi j = n − 1 (8)
i=1 j=1
M, a set of valid TSP constraints (9)
0 ≤ xi j ≤ 1 1 ≤ i, j ≤ n (10)
The set M may consist of some valid TSP constraints which do not contradict
constraint 8. For example, M could be chosen as a set of subtour elimination con-
straints. We have tested our approach with M = {x : xi j + x ji ≤ 1, 1 ≤ i, j ≤ n}
to avoid 2-cycles. An optimal solution (objective value 2,624) for this problem is
shown in Fig. 3. If we delete the constraints of type 9 then the resulting problem A∗
is comparable to an assignment problem where only n − 1 nodes are assigned. In
[2] the first author analyzed the bipartite weighted matching problem with respect to
slightly changed problems of the original problem. In one type of problem two nodes
are deleted in the bipartite graph (one at each partition). The solution is of course a
complete matching (an assignment) with n − 1 edges and therefore also a solution
for A∗ which can be computed in O(n 3 ).
4 H. Achatz and P. Kleinschmidt
Let f (A∞ ) be the objective value of the above problem A∞ . OPT denotes the optimal
solution of the original TSP and P ∞ is any primal feasible solution. Then, of course
we have
f (A∞ ) ≤ OPT ≤ f (P ∞ ) (11)
Proof For any primal feasible solution P with objective value f (P) ≤ f (P ∞ ) we
claim:
If (i, j) is the longest edge in P then ci j ≤ f (P ∞ ) − f (A∞ ).
Suppose f (P ∞ ) − ci j < f (A∞ ) then P \ {(i, j)} is a feasible solution for problem
A∞ with objective value f (P) − ci j . Hence, f (P) − ci j ≤ f (P ∞ ) − ci j < f (A∞ ) by
our assumption. However, f (A∞ ) was optimal and therefore we have a contradiction.
This means that all edges longer than f (P ∞ ) − f (A∞ ) can not be in a better solution
than P ∞ , in particular all these edges can not be in an optimal solution. ∩
√
In our example our best primal solution was 2,726 and the objective value of A∞ is
2,624. Therefore the difference 102 of these values is an upper bound for the longest
edge in an optimal solution. This improves the value 2,352 computed via A∗ . In
Fig. 4 the edge (a, b) has length 104 and therefore this edge can not be in an optimal
solution. All in all 3,542 edges (or 83 %) are longer than the computed bound and
can be deleted.
Remark 1 There are TSP instances where the longest edge of the problem is in an
optimal solution.
If all cities are on a semicircular then the longest edge (the diameter of the circular)
is of course in the optimal solution. In this case our bound is useless.
Computing an Upper Bound for the Longest Edge in an Optimal TSP-Solution 5
3 Computational Results
We have analyzed our approach with several instances in TSPLIB [5] where our set
M was chosen to be M = {x : xi j + x ji ≤ 1, 1 ≤ i, j ≤ n}. The first two columns
denote the name and size of the problem.
The entries of the column “length” are the computed upper bounds for the respec-
tive instances. In the last two columns the number of edges longer than this bound
and their percentage is given. This means for example for the drilling problem in
instance a280 that 42 % of all edges are too long to be in an optimal solution. In all
instances, computing the euclidian distances from the problem data takes more time
than the computation of the LP-solution of A∞ . All primal feasible solutions were
produced by the Lin-Kernighan heuristic [4].
Our computed bounds may be helpful computationally as they lead to potentially
much sparser graphs to be considered in various algorithms.
6 H. Achatz and P. Kleinschmidt
References
1. Achatz, H., Kleinschmidt, P., & Paparrizos, K. (1991). A dual forest algorithm for the assign-
ment problem. In P. Gritzmann, B. Sturmfels & V. Klee (Eds.), Applied geometry and discrete
mathematics. The Victor Klee festschrift (pp. 1–10). Providence, R.I.: AMS (DIMACS’4).
2. Achatz, H. (1999). Sensitivity analysis of the bipartite weighted matching problem. In P. Kall
(Ed.), Operations Research Proceedings 1998. Selected Papers of the International Conference
on Operations Research, Zürich (pp. 135–141). Aug31–Sept 3, 1998. Berlin: Springer.
3. Applegate, D. L., Bixby, R. E., Chvátal, V., & Cook, W. J. (2007). The traveling salesman
problem. A computational story. Princeton, NJ: Princeton University Press (Princeton series in
applied mathematics).
4. Lin, S., & Kernighan, B. W. (1973). An effective heuristic algorithm for the traveling-salesman
problem. Operations Research, 21, 498–516.
5. Reinelt, G. (1991). TSPLIB—a traveling salesman problem library. ORSA Journal on Com-
puting, 3, 376–384.
Electricity Storage Systems and Their
Allocation in the German Power System
Abstract The need for daily electricity storage systems increases with the
growing share of volatile renewable energy in the generation mix. Since the loca-
tion of decentralized electricity generation (based on reneweable energy resource
potentials) and electricity demand (depending on industrial facilities and population
density) in Germany are geographically apart from each other, at the same time more
electricity has to be transported. At certain times, this might challenge the transmis-
sion grid. Storage systems can be used for storing the surplus production of renewable
energy and also help to prevent congestions in the grid. However, besides the tech-
nical feasibility there are economic criteria decisive for the installation of storage
systems. These depend firstly on potential alternative technologies as gas turbines or
the load shift potential of electric vehicles and secondly on the price development of
storage systems. In order to estimate the future demand and the strategic allocation
of daily storage systems in this context, expansion options for storage systems are
implemented in the optimizing energy system model PERSEUS-NET-TS. This is a
myopic material and energy flow model with an integrated nodal pricing approach. A
mixed-integer optimization calculates the expansion and use of power plants in Ger-
many until 2040 considering the DC restrictions of the transmission grid. Hence, the
commissioning and allocation of storage systems in the German transmission grid is
determined when the government target of 60 % renewable feed-in by 2040 is met.
For this paper about every forth car in Germany is considered to drive electrically by
2040. When they are charged uncontrolled, directly after arrival the results are that
by 2040 about 19 GW of storage systems are commissioned. Most are built closely
to generation centers, but some are allocated close to bottlenecks in the transmission
grid instead. When load shifting of the demand for electric mobility is allowed in
terms of a controlled charging the required daily storage capacity could be reduced
by more than half, so that only 8 GW are needed in 2040.
1 Introduction
personal vehicles in Germany are going to be electric by 2030. This share remains
stable until 2040 for the PERSEUS-NET-TS calculations. Depending on the settings
this extra demand can either be charged controlled or uncontrolled directly after
arriving at a charging opportunity at home or at the work place. When charged con-
trolled the only restriction is that each day the needed amount has to be charged, but
the model decides endogenously at which hour during the plug-in time. The driving
force of the optimization is the exogenously given hourly electricity demand for
a winter and a summer week. This hourly demand must be met at each grid node
considering restrictions of the transmission grid and techno-economic constraints of
the generation system. The electricity can be either generated at the grid node with
the assigned power plants or transmitted over the grid from one of the neighboring
nodes. The decision relevant expenditures are minimized for each of the considered
periods. These expenditures include fuel costs and CO2 certificate prices that are
based on the World Energy Outlook 2012 [7]. Furthermore, the variable costs of the
generation are considered, as well as costs for load changing of the thermal power
plants (coal, lignite and gas-combined cycle plants). The fixed costs of the resulting
power plant fleet in the current period are also added as well as investments for new
power plants.
The commissioning of storage systems is allowed from 2020 on. The capital
expenditures for storage systems are assessed to 1,000 EUR/kW in 2020 and are
gradually reduced to 700 EUR/kW in 2040. Additionally, for battery storage sys-
tems there is a fixed ratio of installed capacity to storage volume (kW to kWh) of 1
to 5 assumed. This ratio is chosen according to the characteristics of daily storage
systems in [1]. With the use of PERSEUS-NET-TS it is possible to determine the
technology and the allocation of new generation plants. In order to prevent storage
systems from storing and generating at the same time binary variables are needed
for each hour and each storage. Thus, the model has to be solved via a mixed-integer
optimization. With 336 considered hours (two weeks) for each period, the calcu-
lation time increases significant with every implemented storage option. Through
the storing and generating at the same time, the efficiency of the storage systems
can be used to “waste” electricity. This may make sense within the optimization to
avoid the shutdown of thermal power plants with load changing costs or to meet
the exogenously given targets for the renewable feed-in. To avoid this simultaneous
bidirectional use of storage systems, we chose a two-step approach for this analyses
(see Fig. 1). First, ideal storages with an efficiency of 100 % are implemented. In this
case PERSEUS-NET-TS can be solved linearly with a relatively short computation
time of about 9.5 h.1 Each of the calculated 7 periods consists of about 2.7 mil-
lion equations and 2.3 million variables. In this step, about 350 nodes distributed
across Germany are provided with extension options for battery storage systems. In
addition, a total of 30 pump storage power plants are considered, of which 10 are
integrated as expansion option because they are currently in the planning phase.
1With the use of 6 threats on Windows Server 2008 R2 Enterprise, Intel(R) Xeon(R) CPU E5-1650
@ 3.20 GHz 3.20 GHz; 96 GB RAM; 64 Bit.
10 S. Babrowski et al.
PERSEUS-NET-TS: PERSEUS-NET-TS:
- LP Allocation of storage - MIP
- ideal storages options - real storages
Input: Results:
- Generation system (2012) - Generation system
- Extension options - Electricity mix
- County specific demand - Curtailment of renewables
- Fuel and CO2 prices - Commissioning and dispatch
- Commissioning of Option for new
of storage systems
Existing Option for
renewables hydro storage battery storage hydro storage
3 Results
in the North Sea (see Fig. 3). However, looking at the bottleneck in the transmission
grid in the west of Germany (see Fig. 4) and the storage systems allocated on both
sides of that bottleneck in 2040 (see Fig. 3) it becomes clear that storage systems are
also used for grid management within the optimization.
In comparison to the results of an uncontrolled charging the need for storage
systems decreases by 55 % to only 8 GW when the charging process of the EVs is
controlled. In that case only about 2 GW battery storage systems are endogenously
commissioned in the northwest.
References
1. Association for Electrical, Electronic and Information Technologies (VDE). (2012). Energiespe-
icher für die Energiewende Speicherbedarf und Auswirkungen auf das Übertragungsnetz für
Szenarien bis 2050. Frankfurt.
2. Esser-Frey, A. (2012). Analysing the regional long-term development of the German power
system using a nodal pricing approach. Dissertation, Karlsruher Institute of Technology (KIT),
Karlsruhe.
3. German Federal Ministry for the Environment, Nature Conservation and Nuclear Safety (BMU).
(2010). Long-term scenarios and strategies for the deployment of renewable energies in Germany
in view of European and global developments—Pilot study. Berlin.
4. German Federal Ministry for the Environment, Nature Conservation and Nuclear Safety (BMU).
(2011). Long-term scenarios and strategies for the deployment of renewable energies in Germany
in view of European andglobal developments—Pilot study. Berlin.
Electricity Storage Systems and Their Allocation in the German Power System 13
5. German Federal Ministry of Economics and Technology (BMWi) and German Federal Ministry
for the Environment, Nature Conservation and Nuclear Safety (BMU). (2011). The Federal
Government’s energy concept of 2010 and the transformation of the energy system of 2011.
Berlin.
6. Heinrichs, H. (2013). Analyse der langfristigen Auswirkungen von Elektromobilität auf das
deutsche Energiesystem im europäischen Energieverbund unter Berücksichtigung von Netzre-
striktionen. Dissertation, Karlsruher Institute of Technology (KIT), Karlsruhe.
7. International Energy Agency, (IEA). (2012). World energy, outlook.
Misspecified Dependency Modelling: What Does
It Mean for Risk Measurement?
Theo Berger
1 Introduction
T. Berger (B)
University of Bremen, Wilhelm-Herbst-Str. 5, 28359 Bremen, Germany
e-mail: theoberger@uni-bremen.de
2 Methodology
2.1 Copulas
The Gaussian and t copula belong to the family of elliptical copulas and are derived
from the multivariate normal and t distribution respectively.
The setup of the Gaussian copula is given by:
whereas Φρ stands for the multivariate normal distribution with correlation matrix
ρ and Φ −1 symbolizes the inverse of univariate normal distribution.
Along the lines of the Gaussian copula, the t copula is given by:
in this setup tρ,v stands for the multivariate t distribution with correlation matrix ρ and
v degrees of freedom (d.o.f.). tv−1 stands for the inverse of the univariate t distribution
and v influences tail dependency. For v ≤ ∞ the t distribution approximates a
Gaussian.
In contradiction to the elliptical copulas, the Clayton copula belongs to the group
of Archimedean copulas and is given by:
1
C Clayton (x1 , x2 ) = max{x1θ + x2θ − 1, 0} θ , (6)
with θ ≥ [−1, ∞)\{0}. Note that the Clayton copula describes stronger dependence
in the negative tail than in the positive, for θ ≤ ∞ the Clayton copula describes
comonoticity, and for θ ≤ 0 independence.
Another popular Archimedean copula is represented by the Gumbel copula which,
in contradiction to the Clayton copula, exhibits higher dependence in the positive
tail than in the negative. The copula is given by:
1
C Gumbel (x1 , x2 ) = exp −[(−ln x1 )δ + (−ln x2 )δ ] δ , (7)
with δ ≥ [1, ∞). Analogue to the Gumbel copula, we get comonoticity for θ ≤ ∞
and independence for θ ≤ 0.
As well we introduce the Frank copula as defined by Nelson (1999) which is given
by:
Another random document with
no related content on Scribd:
Giovane che fin da’ suoi tempi e da lui e da altri prestantissimi
uomini se ne gridasse all’abolizione.
L’universale delirio per questi giuochi giadiatorj, l’affluenza del
pubblico, l’intervento del principe e de’ magistrati, la descrizione di
queste pugne e l’interessamento dovunque ad esse per parte d’ogni
classe di persone, non escluso il sesso che suolsi appellare gentile,
io dirò meglio colla viva dipintura che ne fa l’illustre poeta tedesco
Federico Halm, ossia, per togliergli il velo della pseudonimia lacerato
non ha guari da morte, il barone Münch Bellinghausen, nella sua
tragedia Il Gladiatore di Ravenna, la quale abbiamo la fortuna
d’avere egregiamente recata in italiani versi da quel chiarissimo
letterato che è il prof. dottor Giuseppe Rota:
. . . . . . . pectusque jacentis
Virgo modesta jubet, converso pollice, rumpi [143]
Non io di sferza
Nè di buone parole a te mi parco:
Tu dunque bada a farmi onor: m’intendi?
Impassibile mostrati e sicuro:
La coscïenza di vittoria è mezza
Già la vittoria: tieni gli occhi agli occhi
Del tuo rivale e dove intenda avverti
Pria ch’ei muova la man.
TUMELICO.
Lo so, il so bene,
GLABRIONE.
TUMELICO.
E qual?
GLABRIONE.
Nel caso....
Intendi ben.... ciò non sarà, ma pure
Esser potria.... nel caso che abbattuto,
Gravemente ferito.... egli è un supposto...
Tu ti sentissi, allor fa di cadere
Sovra il manco ginocchio e fuor protesa
La destra gamba e del sinistro braccio
Fatto puntello, declinato indietro,
Grazïoso a vedersi e pittoresco,
Statti aspettando il colpo estremo [144].
Talvolta il popolo era tanto feroce che dava tumultuosi segni
d’impazienza quando il combattimento durava un po’ più dell’usato,
senza che alcuno dei due campioni fosse rimasto ucciso o ferito.
V’eran tuttavia degli intervalli di riposo in queste lotte di gladiatori e si
chiamavano deludia: Orazio usò nell’Epistola XIX la frase deludia
posco, per dire chieggo un armistizio, togliendola a prestanza dallo
stile gladiatorio e dall’anfiteatro.
La presenza dell’imperatore faceva d’ordinario accordare la vita al
vinto, e fu ricordato come un esempio di crudeltà il fatto di Caracalla,
che a Nicomedia, in uno spettacolo gladiatorio, avesse licenziato
coloro che eran venuti ad implorarne la vita, sotto pretesto
d’interrogarne il popolo; lo che si ritenne quanto l’ordine di trucidarli.
Byron, l’immortale poeta del Corsaro, di Lara e di Don Juan, nel
Pellegrinaggio di Childe-Harold, dinnanzi al capo d’opera di Ctesilao,
il Gladiatore morente, da lui veduto in Roma, e del quale Plinio il
Vecchio aveva detto che l’artefice vulneratum deficientem fecit in
quo possit intelligi quantum restat animæ [145], così lo descrisse e gli
prestò tal sentimento da sembrare che le barbare orde settentrionali
e le sventure tutte piombate poi sull’Italia e Roma, non altro fossero
che la giusta espiazione del sangue sparso da’ poveri e innocenti
prigionieri di guerra condannati in sollazzo pubblico a’ cruenti
spettacoli dell’Anfiteatro.
Così alla meglio tento di rendere in italiano i bellissimi versi inglesi: