Gas Network Optimization by MINLP 1st Edition

Jesco Humpola download

https://ebookname.com/product/gas-network-optimization-by-
minlp-1st-edition-jesco-humpola/

Get Instant Ebook Downloads – Browse at https://ebookname.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Sniffer Pro Network Optimization and Troubleshooting

Handbook 1st Edition Robert J. Shimonski

https://ebookname.com/product/sniffer-pro-network-optimization-
and-troubleshooting-handbook-1st-edition-robert-j-shimonski/

Evolved Cellular Network Planning and Optimization for

UMTS and LTE 1st Edition Lingyang Song

https://ebookname.com/product/evolved-cellular-network-planning-
and-optimization-for-umts-and-lte-1st-edition-lingyang-song/

Identification of essential oil components by gas

chromatography mass spectrometry 4th Edition Robert P.
Adams

https://ebookname.com/product/identification-of-essential-oil-
components-by-gas-chromatography-mass-spectrometry-4th-edition-
robert-p-adams/

Learning JavaScript Design Patterns Addy Osmani

https://ebookname.com/product/learning-javascript-design-
patterns-addy-osmani/
Launching the Imagination 3D 3rd Edition Mary Stewart

https://ebookname.com/product/launching-the-imagination-3d-3rd-
edition-mary-stewart/

The Reign of Botnets Defending Against Abuses Bots and

Fraud on the Internet 1st Edition David Senecal

https://ebookname.com/product/the-reign-of-botnets-defending-
against-abuses-bots-and-fraud-on-the-internet-1st-edition-david-
senecal/

Leading the Congregation Caring for Yourself While

Serving the People 1st Edition Norman Shawchuck

https://ebookname.com/product/leading-the-congregation-caring-
for-yourself-while-serving-the-people-1st-edition-norman-
shawchuck/

The Professional Practice of Landscape Architecture A

Complete Guide to Starting and Running Your Own Firm
Second Edition Walter Rogers

https://ebookname.com/product/the-professional-practice-of-
landscape-architecture-a-complete-guide-to-starting-and-running-
your-own-firm-second-edition-walter-rogers/

From Exclusion to Inclusion in Old Age A Global

Challenge 1st Edition Thomas Scharf

https://ebookname.com/product/from-exclusion-to-inclusion-in-old-
age-a-global-challenge-1st-edition-thomas-scharf/
Real Legal Certainty and its Relevance Essays in honor
of Jan Michiel Otto 1st Edition Adriaan Bedner

https://ebookname.com/product/real-legal-certainty-and-its-
relevance-essays-in-honor-of-jan-michiel-otto-1st-edition-
adriaan-bedner/
 Jesco Humpola

Gas Network Optimization

by MINLP

λογος
 Gas Network Optimization
by MINLP

vorgelegt von
Dipl.-Math. Jesco Humpola
aus Melle

von der Fakultät II – Mathematik und Naturwissenschaften

der Technischen Universität Berlin
zur Erlangung des akademischen Grades

Doktor der Naturwissenschaften

– Dr. rer. nat. –

genehmigte Dissertation

Promotionsausschuss:

Vorsitzender: Prof. Dr. Peter Bank

Gutachter: Prof. Dr. Dr. h.c. mult. Martin Grötschel
Gutachter: Prof. Dr. Thorsten Koch
Gutachter: Prof. Andrea Lodi, Ph.D.

Tag der wissenschaftlichen Aussprache: 06. November 2014

Berlin 2014
D 83
Bildnachweis Einband Vorderseite:
Mit freundlicher Genehmigung von Open Grid Europe GmbH, Essen.

Bibliografische Information der Deutschen Nationalbibliothek:

Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der

Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im
Internet über http://dnb.d-nb.de abrufbar.

©Copyright Logos Verlag Berlin GmbH 2017

Alle Rechte Vorbehalten.

ISBN: 978-3-8325-4505-5

Logos Verlag Berlin GmbH

Comeniushof, Gubener Str. 47,
10243 Berlin
Tel.: +49 (0)30 42 85 10 90
Fax: +49 (0)30 42 85 10 92
INTERNET: http://www.logos-verlag.de
Abstract

One quarter of Europe’s energy demand is provided by natural gas distributed

through a vast pipeline network covering the whole of Europe. At a cost of 1 million
Euros per kilometer the extension of the European pipeline network is already a
multi billion Euro business. The challenging question is how to expand and operate
the network in order to facilitate the transportation of specified gas quantities at
minimum cost. This task can be formulated as a mathematical optimization problem
that reflects to real-world instances of enormous size and complexity. The aim of this
thesis is the development of novel theory and optimization algorithms which make it
possible to solve these problems.
Gas network topology optimization problems can be modeled as nonlinear mixed-
integer programs (MINLPs). Such an MINLP gives rise to a so-called active trans-
mission problem (ATP), a continuous nonlinear non-convex feasibility problem which
emerges from the MINLP model by fixing all integral variables. The key to solving
the ATP as well as the overall gas network topology optimization problem and the
main contribution of this thesis is a novel domain relaxation of the variable bounds
and constraints in combination with a penalization in the objective function. In case
the domain relaxation does not yield a primal feasible solution for the ATP we offer
novel sufficient conditions for proving the infeasibility of the ATP. These conditions
can be expressed in the form of an MILP, i.e., the infeasibility of a non-convex NLP
can be certified by solving an MILP. These results provide an efficient bounding
procedure in a branch-and-bound algorithm.
If the gas network consists only of pipes and valves, the ATP turns into a passive
transmission problem (PTP). Although its constraints are non-convex, its domain
relaxation can be proven to be convex. Consequently, the feasibility of the PTP can
be checked directly in an efficient way. Another advantage of the passive case is that
the solution of the domain relaxation gives rise to a cutting plane for the overall
topology optimization problem that expresses the infeasibility of the PTP. This cut is
obtained by a Benders argument from the Lagrange function of the domain relaxation
augmented by a specially tailored pc-regularization. These cuts provide tight lower
bounds for the passive gas network topology optimization problem.

iii
 The domain relaxation does not only provide certificates of infeasibility and
cutting planes, it can also be used to construct feasible primal solutions. We make
use of parametric sensitivity analysis in order to identify binary variables to be
switched based on dual information. This approach allows for the first time to
compute directly MINLP solutions for large-scale gas network topology optimization
problems.
All the research in this thesis has been realized within the collaborative research
project “Forschungskooperation Netzoptimierung (ForNe)”. The developed software
is in use by the cooperation partner Open Grid Europe GmbH.
Parts of this thesis have been published in book chapters, journal articles and
technical reports. An overview of the topics and solution approaches within the
research project is given by Martin et al. (2011) and Fügenschuh et al. (2013).
Gas network operation approaches and solution methods are described in detail by
Pfetsch et al. (2014) and with a special focus on topology optimization in Fügenschuh
et al. (2011). The primal heuristic presented in this thesis is published by Humpola
et al. (2015a). The method for pruning nodes of the branch-and-bound tree for an
approximation of the original problem is described in Humpola and Fügenschuh
(2015) and Humpola et al. (2015b) and Humpola and Serrano (2017). The Benders
like inequality is introduced by Humpola et al. (2016).

iv
Zusammenfassung

Ein Viertel des europäischen Energiebedarfs wird durch Gas gedeckt, das durch
ein europaweites Pipelinesystem verteilt wird. Aufgrund von Ausbaukosten von
1 Mio. Euro pro Kilometer ist der Netzausbau ein Milliardenunterfangen. Die größte
Herausforderung besteht darin zu entscheiden, wie das Netzwerk kostengünstig
ausgebaut und genutzt werden kann, um notwendige Gasmengen zu transportieren.
Diese Aufgabe kann mit Hilfe eines mathematischen Optimierungsproblems formuliert
werden, wobei anwendungsnahe Instanzen eine enorme Größe und Komplexität
aufweisen. Ziel der vorliegenden Arbeit ist die Entwicklung neuer mathematischer
Theorien und damit einhergehender Optimierungsalgorithmen, die es ermöglichen,
derartige Probleme zu lösen.
Die Optimierung der Topologie eines Gasnetzwerks kann mit Hilfe eines nichtlinea-
ren gemischt-ganzzahligen Programms (MINLP) modelliert werden. Durch Fixierung
aller ganzzahligen Variablen ergibt sich ein kontinuierliches Zulässigkeitsproblem, das
als aktives Transmissionsproblem (ATP) bezeichnet wird. Die zentrale Methode um
dieses ATP zu lösen, ist eine neuartige Relaxierung, welche Variablenschranken und
einige Nebenbedingungen relaxiert und in der Zielfunktion bestraft. Diese Relaxierung
bildet den Kern der in dieser Arbeit vorgestellten Theorie und ermöglicht so die
effiziente Lösung der Topologieoptimierung eines Gasnetzwerkes. Für den Fall, dass
die Relaxierung keine Primallösung für das ATP liefert, ist es uns gelungen, hinrei-
chende Bedingungen für die Unzulässigkeit des ATP zu formulieren, die durch ein
MILP dargestellt werden. Kurz gefasst kann die Unzulässigkeit eines nicht-konvexen
NLP durch Lösung eines MILP bewiesen werden. Beide Methoden liefern effiziente
Schranken in einem branch-and-bound Lösungsverfahren.
Besteht ein Gasnetzwerk nur aus Rohren und Schiebern, dann wird das ATP als
passives Transmissionsproblem (PTP) bezeichnet. Obwohl die Nebenbedingungen
des PTP nicht konvex sind, konnten wir zeigen, dass seine Relaxierung konvex ist.
Daher kann die Unzulässigkeit des PTP direkt auf effiziente Weise geprüft werden.
Außerdem können mit Hilfe der Relaxierung in diesem speziellen Fall Schnittebenen
für das Topologieoptimierungsproblem aufgestellt werden. Diese repräsentieren die

vii
Unzulässigkeit des PTP und folgen aus der Lagrange Funktion der Relaxierung
zusammen mit einer speziellen Erweiterung, der sogenannten pc-Regularisierung.
Abgesehen von den genannten Klassifizierungen kann die Relaxierung auch genutzt
werden, um primale Lösungen zu konstruieren. Hier nutzen wir die parametrische
Sensitivitätsanalyse, um mit Hilfe dualer Informationen Binärvariablen des ATP zu
identifizieren, deren Werte angepasst werden müssen. Dieser Ansatz erlaubt es zum
ersten Mal, direkt MINLP Lösungen für das Topologieoptimierungsproblem realer
Gasnetzwerke zu berechnen.
Die Resultate dieser Arbeit wurden im Rahmen des Forschungsprojekts “For-
schungskooperation Netzoptimierung (ForNe)” erarbeitet. Die entwickelte Software
wird vom Kooperationspartner Open Grid Europe GmbH aktiv genutzt.
Teile dieser Arbeit sind in Buchkapiteln, Journalen und technischen Berichten
publiziert. Eine Übersicht über die Themen und Lösungsansätze im ForNe-Projekt
veröffentlichten Martin u. a. (2011) und Fügenschuh u. a. (2013). Für Lösungsme-
thoden für die operative Nutzung von Gasnetzwerken verweisen wir auf Pfetsch u. a.
(2014). Ansätze für eine Topologieoptimierung wurden von Fügenschuh u. a. (2011)
beschrieben. Die in dieser Arbeit präsentierte primale Heuristik ist publiziert von
Humpola u. a. (2015a). Die genannte Methode, um Knoten innerhalb des branch-and-
bound Baums abzuschneiden, wurde für eine Approximation des Topologieproblems
von Humpola und Fügenschuh (2015) und Humpola u. a. (2015b) und Humpola und
Serrano (2017) beschrieben. Ein Bericht über die Ungleichungen nach Benders ist in
Humpola u. a. (2016) nachzulesen.

viii
Acknowledgements

First of all, I would like to thank my supervisor Prof. Dr. Dr. h.c. mult. Martin
Grötschel for enabling me to work in such a fascinating field of work, making it
possible for me to apply my theoretical advances to real life scenario.
In addition I wish to thank my second supervisor Prof. Dr. Thorsten Koch. He
encouraged me to attend diverse mathematical conferences, resulting in profitable
discussions, novel ideas and a meeting with Prof. Andrea Lodi, who later agreed to
be one of my supervisors. Moreover, from very early on, he put his trust in me and
encouraged me to follow my own ideas.
Furthermore I would also like to thank Prof. Andrea Lodi for inviting me to
Bologna, having long and fruitful discussions on further applications of the research
of this thesis and for including my defense into his busy schedule.
I have to thank Prof. Dr. Ralf Borndörfer for the constructive discussions we had
over the years and all valuable insights and ideas he shared with me. Thank you for
supporting and motivating me and for providing me with the last finishing touches
and suggestions for this thesis.
I am especially obliged to Prof. Dr. Armin Fügenschuh, Dr. Thomas Lehmann
and Dr. Nam Dung Hoang for careful reading this thesis and for their diverse criticism
and suggestions. I have to emphasize Prof. Dr. Armin Fügenschuh for his willingness
to support my publication ideas, enhancing them and to make our final results so
engaging to read.
I am very grateful to my colleague Felipe Serrano for his essential comments on
Chapter 6 which resulted in the rewriting and rethinking of the whole chapter.
Thanks must go to my colleagues of the working group “Energy” at ZIB, Dr.
Benjamin Hiller, Ralf Lenz, Jonas Schweiger, Robert Schwarz. I always appreciated
our vital discussions and the pleasant working atmosphere. I also like to thank them
for commenting on parts of this thesis.
I would also like to thank our industry partners, in particular Dr. Klaus Spreck-
elsen, Dr. Lars Huke and his team from Open Grid Europe GmbH. They made it
possible to test the algorithms developed in this thesis within real-life applications
by providing the data.

xi
 I wish to thank all my former and fellow colleagues of the ForNe project team
and the SCIP group, especially Prof. Dr. Marc E. Pfetsch for all the hours of fertile
discussions we had.
Thank you to Dr. Radoslava Mirkov for all the advice, friendship and inspring
conservations we had over the past years. I am happy I came to Berlin for my PhD.
Also I have to give many thanks to Zara, Daniel, Nadine and my mother for all
the proof reading and rephrasing of my thesis.
I have to express my biggest gratitude to my family, especially to my parents for
supporting me in my studies. Because of you I became the person I am today!
Last but not least, I want to thank Nadine for all her patience and her constant
trust in me.

xii
Contents

1 Introduction: Gas Network Optimization 1

1.1 Optimization Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Previous Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Solution Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.4 Computational Study . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.5 Outline of the Thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2 Solving Mixed-Integer Nonlinear Optimization Problems 13

2.1 Definition of MINLP . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.2 Details on SCIP for Solving MINLP . . . . . . . . . . . . . . . . . . 14
2.3 Convex and Non-Convex MINLP . . . . . . . . . . . . . . . . . . . . 17
2.4 Necessary Conditions for Local Optimality . . . . . . . . . . . . . . . 19
2.5 A Specially Tailored Adaptation of SCIP . . . . . . . . . . . . . . . . 20

3 An MINLP Model for Gas Network Topology Optimization 23

3.1 Technical Background . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.2 An MINLP Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.3 Complexity Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.4 The Passive and Active Transmission Problem . . . . . . . . . . . . 41
3.5 Computational Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

4 Efficiently Solving the Passive Transmission Problem 51

4.1 Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
4.2 Relaxation of Domains . . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.2.1 Existence of a Solution . . . . . . . . . . . . . . . . . . . . . . 54
4.2.2 Characterization of the Feasible Region . . . . . . . . . . . . 58
4.2.3 Interpretation of Lagrange Multipliers . . . . . . . . . . . . . 64
4.3 Relaxation of Flow Conservation Constraints . . . . . . . . . . . . . 71
4.3.1 Preprocessing . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
4.3.2 Existence of a Solution . . . . . . . . . . . . . . . . . . . . . . 73
4.3.3 Characterization of the Feasible Region . . . . . . . . . . . . 78

xv
 4.3.4 Interpretation of Lagrange Multipliers . . . . . . . . . . . . . 85
4.4 Relaxation of Potential-Flow-Coupling Constraints . . . . . . . . . . 88
4.4.1 Conditions of the KKT System . . . . . . . . . . . . . . . . . 88
4.4.2 Different KKT Points . . . . . . . . . . . . . . . . . . . . . . 90
4.5 Solving the Passive Transmission Problem . . . . . . . . . . . . . . . 91
4.6 Integration and Computational Results . . . . . . . . . . . . . . . . . 95

5 An Improved Benders Cut for the Topology Optimization Problem 103

5.1 Valid Inequalities for the Passive Transmission Problem . . . . . . . 104
5.1.1 A Nonlinear Inequality . . . . . . . . . . . . . . . . . . . . . . 106
5.1.2 A Linear Inequality . . . . . . . . . . . . . . . . . . . . . . . 111
5.1.3 Feasibility Characterization by a Linear Inequality . . . . . . 114
5.1.4 A Linear Inequality derived from the Lagrange Function of the
Domain Relaxation . . . . . . . . . . . . . . . . . . . . . . . . 124
5.2 A Valid Inequality for the Topology Optimization Problem . . . . . 126
5.3 Integration and Computational Results . . . . . . . . . . . . . . . . . 133

6 Sufficient Conditions for Infeasibility of the Active

Transmission Problem 141
6.1 Non-Convex Relaxations for the Active Transmission Problem . . . . 142
6.1.1 Relaxation of Domains . . . . . . . . . . . . . . . . . . . . . . 143
6.1.2 Relaxation of Flow Conservation Constraints . . . . . . . . . 150
6.2 Detecting Infeasibility of the Active Transmission Problem by MILP 155
6.3 Interpretation of the Infeasibility Detection MILP . . . . . . . . . . 170
6.4 Integration and Computational Results . . . . . . . . . . . . . . . . . 175

7 A Primal Heuristic based on Dual Information 181

7.1 A Relaxation of the MINLP . . . . . . . . . . . . . . . . . . . . . . . 182
7.2 A Primal Heuristic for MINLP with Indicator Constraints . . . . . . 184
7.2.1 Theoretical Motivation . . . . . . . . . . . . . . . . . . . . . . 184
7.2.2 The Basic Dual Value MINLP Heuristic . . . . . . . . . . . . 187
7.2.3 Embedding the Heuristic in a Branch-and-Prune Search . . . 188
7.3 A Specialization to the Topology Optimization Problem . . . . . . . 188
7.3.1 The Relaxation . . . . . . . . . . . . . . . . . . . . . . . . . . 189
7.3.2 Handling Different Modes of Active Devices . . . . . . . . . . 190
7.3.3 Handling Loop Extensions . . . . . . . . . . . . . . . . . . . . 194
7.4 Implementation Details . . . . . . . . . . . . . . . . . . . . . . . . . 196
7.5 Computational Results . . . . . . . . . . . . . . . . . . . . . . . . . . 198

xvi
8 Conclusions 203

A Tables 207

Bibliography 227

xvii
Chapter 1

Introduction: Gas Network

Optimization

Natural gas is a nontoxic, odorless, transparent, and flammable gas that originates
from underground deposits. Today natural gas is mainly used for heating private
houses and office buildings, for the generation of electrical power, as fuel for vehicles,
and for several reactions in chemical process engineering. Natural gas usage represents
one quarter of the world’s energy demand (BGR 2013). It must be transported from
the deposits to the customers, sometimes over distances of thousands of kilometers.
For very long distances it is more economic to cool the gas down to −160° so that
it becomes liquid and can be transported by ships (see Cerbe 2008). For shorter
distances or for the delivery to the end customers large pipeline systems are used.
Existing gas networks have usually grown over time. In Germany, the high-pressure
pipeline system was built by gas supply companies. It has a size of approximately
35 000 km (FNBGas 2013).
Historically these companies were both gas traders and gas network operators.
They purchased gas from other suppliers and operated the necessary infrastructure
to transport the gas from those suppliers to their own customers. During the
liberalization of the German gas market these business functions were separated by
regulatory authorities (GasNZV 2005). Nowadays, there are companies that trade gas
and others whose sole task is the operation of gas networks for the transportation of
gas. One of the requirements set by the regulatory authorities is that every trader can
use the network infrastructure to transport gas. Open access to these gas networks
has to be granted to all the trading companies free of any discrimination. This means
that the gas supplies and demands cannot be fully controlled by the network operator.
Therefore the network operator is required to have a high degree of operational
flexibility. The majority of network management is carried out manually with the
aid of simulation software. There is a need to develop a more automated process

1
2 Chapter 1 Introduction: Gas Network Optimization

in order to cope with anticipated challenges associated with the addition of more
traders accessing the network. Here we have developed mathematical optimization
methods to improve the network operation and to enhance the cost effectiveness of
investments in the infrastructure.

1.1 Optimization Tasks

The physics of a gas transport network, which is used for the transportation of gas,
can roughly be described as follows: Most of the network elements are pipelines. A
gas flow through a pipeline (pipe) is induced if the gas has different pressures at
the end nodes of the pipe. Usually gas pipelines can withstand nominal pressures of
16 bar up to 100 bar. Typically at long transport distances of about 100 km to 150 km
the gas pressure gets too low which is technically not feasible. In situations where
lack of pressure is an issue compressors are used to increase the pressure again. High
gas pressures can also be problematical in parts of the network. Therefore in order to
protect the network it can be necessary to reduce the pressure by using control valves.
This is particularly important when the network includes older pipelines which have
a lower pressure limit. If parts of the network need to be deactivated, then valves
allow to split the network into physically independent subnetworks.
A gas network is mathematically modeled by a directed graph, refer to Korte and
Vygen (2007) for the notations in graph theory. This graph consists of nodes together
with connecting arcs. The end points of an arc are nodes. Each node corresponds
to a geographical position. Each arc models a network element. We use a directed
graph in order to distinguish between gas flow in the direction of an arc and in the
opposite direction. A network element can either be a pipeline, a valve, a compressor
or a control valve. All network elements determine a specific relation between the
flow through the element and the pressure at the end nodes. Pipelines are called
passive network elements, while valves, compressors and control valves are called
active network elements. Pipelines have a unique relation between the flow through
a pipeline and the pressures at its end nodes. This is different for active elements.
Their physical behavior can be influenced by the network operator, for instance, a
valve can be open or closed. An open valve means that the pressures at the end
nodes are equal, while closed means that there is no gas flow through the valve.
Compressors and control valves can be open or closed. This means that each active
network element has different operation modes or rather configurations available. An
accurate description of every network element is given in Section 3.1.
1.1 Optimization Tasks 3

∃ q, π

a4 s. t. q1 = 1,
−2.1 +1
q2 − q1 = 0,
v4 v5
q2 + q3 = 2.1,
a3 q3 + q4 = 2.1,
q4 = 1,
+2.1 v3
44 q1 |q1 | − (π2 − π1 ) = 0,
44 q2 |q2 | − (π3 − π2 ) = 0,
a2
44 q3 |q3 | − (π3 − π4 ) = 0,
−1
44 q4 |q4 | − (π5 − π4 ) = 0,
v1 a1 v2 π1 = 0,
(a) Test network. 0 ≤ π2 , π3 , π4 , π5 ≤ 92,
π1 , π2 , π3 , π4 , π5 , q1 , q2 , q3 , q4 ∈ R.

(b) Nomination validation problem for the net-

work shown in Figure 1.1a (left picture).

Figure 1.1: An example of a test network and a model of the corresponding nomination validation problem
as used in this thesis. The instance has two entries at node v3 with flow amount +2.1 and v5 with flow
amount +1. Two exits are at node v1 with flow amount −1 and v4 with flow amount −2.1. Node v2 is
a transmission node. These node flows imply the arc flows 1 for a1 and a2 and a4 and 1.1 for a3 . Every
arc is a pipeline which means a unique relation between the arc flow and the pressures at the end nodes.
The pressure at node v1 is fixed to zero which implies that any other node pressure is fixed. Hence the
pressure at node v5 is also fixed. The pressure bounds of the pipes imply that the pressure at node v5
violates its upper bound. So the nomination validation problem is infeasible for this instance.

We distinguish between two different optimization tasks. Given is a nomination

that specifies for each single node the amount of gas that enters or leaves the network
there. The network in combination with the nomination defines an instance. In
order to operate a gas transport network the task is to compute a configuration
of valves, control valves and compressors, a pressure for each node, and a flow for
each element. This computation requires that the flow specified by the nomination
is transported through the network and all technical and physical as well as legal
constraints are fulfilled. The arising problem is a feasibility problem which is called
nomination validation problem of the specified instance. Figure 1.1 shows an example.
A simple tree network together with a nomination is shown in 1.1a and the associated
nomination validation problem is infeasible. This means that it is not possible to
transport gas through the network according to the specified nomination. A model
of this problem as used in this thesis is shown in 1.1b and will be explained in detail
in Chapter 3. Due to the infeasibility of the nomination the network needs to be
extended. In principal all network elements, i.e., pipelines, valves, control valves and
compressors, can be added to the gas network. The additional elements are called
extensions. The task is to compute cost-optimal extensions in order to transport the
gas defined by the nomination through the network. The arising problem is called
topology expansion problem. Its objective is to minimize the building costs of the
4 Chapter 1 Introduction: Gas Network Optimization

(a) Extended by a control valve. (b) Extended by a pipe in combina-

tion with a compressor.

Figure 1.2: Extended versions of the network shown in Figure 1.1a. The discussion in Figure 1.1 explains
that the nomination validation problem 1.1b is infeasible because the pressure at node v5 violates its
upper pressure bound. Decreasing the pressure in node v4 is a possible adaptation in order to reduce the
pressure in node v5 . This adjustment is achieved by the proposed extensions in 1.2a and 1.2b.

additional network elements. Note that operation costs of the network elements are
assumed to be a constant term in this objective which is not taken into account
for the optimization. They are caused by compressors which consume energy for
the compression of the gas or pipelines which have to be maintained regularly, for
example. Figure 1.2 shows two different suitable extensions for the network in
Figure 1.1a. We refer to Figure 1.3 for a visualization of the effect of extending a gas
network. More precisely the node pressures are shown before and after extending
the network. The networks shown so far are only small examples. A real-world gas
network with approximately 4000 arcs for which we want to solve the nomination
validation problem is shown Figure 1.4.
The building costs of a new network element depend on its type. For building a
pipeline the costs are mainly made up of the material costs, the construction costs
and costs of getting permission rights for the use of land. Construction costs for
compressors are different as a compressor is typically built at a single place. We
assume that the construction costs also contain a term representing the operation
costs of the specific network element.

1.2 Previous Work

A general survey over the application of optimization methods in the natural gas
industry is given by Zheng et al. (2010). The surveys by Shaw (1994) and Ríos-
Mercado and Borraz-Sánchez (2012) are closer to the problems we are studying. A
monograph outlining the earlier state-of-the-art is described in the book of Osiadacz
(1987).
1.2 Previous Work 5

max min
Figure 1.3: Element flow and node pressure corresponding to a realization of a nomination in a test
network consisting of pipelines only. The line width represents the flow value (the thicker the more flow),
while its color depicts the mean value of the node pressures at both end nodes. The nodes are depicted
by squares and the node colors represent the node pressures. In our test case we added two pipelines to
the network which results in a different flow and pressure distribution (see the lower picture). In both
pictures the maximum of the node pressures corresponds to the color red and the minimum to the color
blue.
Exploring the Variety of Random
Documents with Different Content
 Films. 11 min., sd.,
b&w, 16 mm. © Coronet Instructional
Films, a division of Esquire, Inc.;
1May62; MP12603.

WHY FATHERS WORK. Encyclopaedia

Britannica Educational Corp.
14 min., sd., color, 16 mm. © Encyclopaedia
Britannica Educational
Corp.; 3Jul69; MP19611.

WHY GIRLS LEAVE HOME. See

PETTICOAT JUNCTION.

WHY IT WORKS. See

THE SLIDE RULE.

WHY MAN IN SPACE. See

CBS REPORTS.

WHY PLANTS GROW WHERE THEY DO. Coronet

Instructional Films. 11 min., sd.,
b&w, 16 mm. © Coronet Instructional
Films, a division of Esquire, Inc.;
1Jun66; MP16174.

WHY POLITICS. See

BUILDING POLITICAL LEADERSHIP.

WHY SEASONS CHANGE. Encyclopaedia

Britannica Films. 11 min., sd.,
b&w, 16 mm. © Encyclopaedia Britannica
 Films, Inc.; 25Oct60; MP10887.

WHY THEY BUY. Aluminum Co. of America.

Made by On Film. 10 min., sd., color,
16 mm. © Aluminum Co. of America;
17Jan60; LP16427.

WHY WAIT 'TILL NOVEMBER? See

THE FARMER'S DAUGHTER.

WHY WE LEARN. See

THE STUDY SERIES.

WHY WE LISTEN. See

THE STUDY SERIES.

WHY WE READ. See

THE STUDY SERIES.

WHY WON'T ANYBODY LISTEN? See

DR. KILDARE.

THE WICHITA ARROWS. See

RIVERBOAT.

WICHITA TOWN. Mirisch-McCrea Productions.

1 reel each, sd., b&w, 16 mm.
© Mirisch-McCrea Productions.

Afternoon in town. © 17Feb60; LP16613.

The avengers. © 3Feb60; LP16611.

The biggest man in town. © 30Dec59;

LP16606.

Brothers of the knife. © 10Feb60;

LP16612.

Bullet for a friend. © 14Oct59;

LP16598.

Compadre. © 18Nov59; LP16601.

Day of battle. © 18Nov59; LP16602.

Death watch. © 16Dec59; LP16818.

The devil's choice. © 23Dec59;

LP16605.

Drifting. © 28Oct59; LP16600.

The fight for Thursday. © 13Jan60

(in notice: 1959); LP16607.

The frontiersman. (The strange task

of Otis Stockett) © 2Mar60; LP16614.

The hanging judge. © 9Mar60; LP16615.

The legend of Tom Horn. © 30Mar60;

LP16618.

The long night. © 20Jan60; LP16609.

 Man on the hill. © 4Nov59; LP16595.

The night the cowboys roared.

© 30Sep59; LP16596.

Out of the past. © 9Dec59; LP16604.

Paid in full. © 23Mar60; LP16617.

Passage to the enemy. © 2Dec59;

LP16603.

Ruby Dawes. © 6Jan60 (in notice:

1959); LP16608.

Second chance. © 16Mar60; LP16616.

Seed of hate. © 27Jan60; LP16610.

Side-kicks. © 6Apr60 (in notice:

1959); LP16619.

They won't hang Jimmy Relson.

© 21Oct59; LP16599.

Wyndham's way. © 7Oct59; LP16597.

THE WICKED DREAMS OF PAULA SCHULTZ.

Theme Pictures. Released by United
Artists Corp. 113 min., sd., color,
35 mm. © Theme Pictures, Inc.;
4Jan68 (in notice: 1967); LP35531.

WIDE COUNTRY. Bar-Bar-A Productions.

Approx. 53 min. each, sd., b&w, 16 mm.
© Bar-Bar-A Productions.
The bravest man in the world.
© 6Dec62; LP26013.

The care and handling of tigers.

© 25Apr63; LP26030.

A cry from the mountain. © 17Jan63;

LP26018.

A devil in the chute. © 8Nov62;

LP26010.

Don't cry for Johnny Devlin.

© 24Jan63; LP26019.

Farewell to Margarita. © 21Mar63;

LP26026.

The girl from Nob Hill. © 28Mar63;

LP26027.

The girl in the sunshine smile.

© 15Nov62; LP26011.

Good old Uncle Walt. © 13Dec62;

LP26014.

A guy for Clementine. © 27Sep62;

LP26004.

Journey down a dusty road. © 4Oct62;

LP26005.

The Judas goat. © 21Feb63; LP26023.

The lucky punch. © 18Apr63; LP26029.

The man who ran away. © 7Feb63;

LP26021.

Memory of a filly. © 3Jan63 (in

notice: 1962); LP26016.

My candle burns at both ends.

© 20Dec62; LP26015.

Our Ernie kills people. © 1Nov62;

LP26009.

The quest of Jacob Blaufus. © 7Mar63;

LP26025.

The Royce Bennett story. © 20Sep62;

LP26003.

Speckle bird. © 31Jan63; LP26020.

Step over the sky. © 10Jan63 (in

notice: 1962); LP26017.

Straitjacket for an Indian. © 25Oct62;

LP26008.

Tears on a painted face. © 29Nov62;

LP26012.

To Cindy, with love. © 28Feb63;

LP26024.

What are friends for? © 18Oct62;

LP26007.
 Who killed Eddie Gannon? © 11Oct62;
LP26006.

Whose hand at my throat? © 14Feb63;

LP26022.

Yanqui, go home. © 4Apr63; LP26028.

THE WIDE OPEN SPACES. See

THE DONNA REED SHOW.

A WIDE PLACE IN THE ROAD. See

THE OUTSIDER.

THE WIDENING WORLD OF BOOKS. Ealing

Corp. Made by Milner-Fenwick.
1 reel, sd., b&w, 16 mm. (Starting
tomorrow) © Ealing Corp.; 15Nov68;
MP18942.

THE WIDESCREEN CAPER. See

77 SUNSET STRIP.

THE WIDOW. See

THE ANN SOTHERN SHOW.

GUNSMOKE.

THE WIDOW AND THE WEB. See

77 SUNSET STRIP.
THE WIDOW FROM RED ROCK. See

THE HIGH CHAPARRAL.

WIDOW IN WHITE. See

LARAMIE.

THE WIDOW-MAKERS. See

SEE HOW THEY RUN.

WIDOW MULDANE. See

WILD BILL HICKOK.

WIDOW OF EL DORADO. See

SHOTGUN SLADE.

THE WIDOW OF KILL COVE. See

CORONADO 9.

THE WIDOW OF WOODCOTE. See

IVANHOE.

THE WIDOW O'ROURKE STORY. See

WAGON TRAIN.

THE WIDOW POKE ARRIVES. See

THE BEVERLY HILLBILLIES.

THE WIDOW WOULDN'T WEEP. See

77 SUNSET STRIP.

THE WIDOWS. See

M-SQUAD.

A WIFE FOR ANDY. See

THE ANDY GRIFFITH SHOW.

WIFE VERSUS SECRETARY. See

THE JOEY BISHOP SHOW.

THE WIG. See

KAREN.

A WIG FOR MISS DEVORE. See

THRILLER.

THE WIG STORY. See

THE MOTHERS-IN-LAW. No. 24.

THE WILBERFORCE PENDULUM. Ohio State

University. 5 min., si., color,
16 mm. Eastman color. Produced
under a grant from The National
Science Foundation. © Franklin
Miller, Jr.; 25Oct63; MP13718.
WILBUR PEACOCK. See

CISCO KID.

WILBUR THE WANTED, No. 1. See

VAROOM SERVICE.

WILBUR/WILBUR'S.
For other titles beginning with Wilbur
or Wilbur's See MISTER ED.

WILD ABOUT HURRY. Warner Bros. Cartoons.

7 min., sd., Technicolor,
35 mm. (Merrie melodies cartoon;
Road Runner & Wile E. Coyote) © Vitaphone
Corp.; 10Oct59; MP10575.

THE WILD AND WONDERFUL THIRTIES. See

HOLLYWOOD AND THE STARS.

WILD AND WOOLLY HARE. Warner Bros. Pictures.

7 min., sd., Technicolor,
35 mm. (Looney tune cartoon; Bugs
Bunny special with Yosemite Sam)
© Vitaphone Corp.; 1Aug59; MP10334.

THE WILD ANGELS. American International

Pictures. 85 min., sd., color, 35 mm.
Color by Pathé. Panavision. © American
International Productions;
20Jul66; LP32873.

WILD ANIMAL FAMILIES. Film Associates

of California. 11 min., sd., color,
 16 mm. Eastman color. © Film
Associates of California; 23Jan57;
MP14017.

WILD BILL HICKOK. Screen Gems. Approx.

30 min. each, sd., b&w, 16 mm.
© Screen Gems, Inc.

The Angel of Cedar Mountain.

© 20Apr58; LP26457.

Clem's reformation. © 9Apr58; LP26454.

The counterfeit ghost. © 14May58;

LP26459.

The daughter of Casey O'Grady.

© 23Apr58; LP26456.

Good Indian. © 7May58; LP26458.

Jingles on the Jailroad. © 16Apr58;

LP26455.

Marvin's mixup. © 19Feb58; LP26447.

Meteor Mesa. © 19Mar58; LP26451.

Monkeyshines. © 5Mar58; LP26449.

The runaway wizard. © 12Mar58;

LP26450.

Sheriff of Buckeye. © 2Apr58; LP26453.

Spurs for Johnny. © 26Feb58; LP26448.

 Town without law. © 26Mar58; LP26452.

WILD BILL HICKOK. William F. Broidy

Productions. Approx. 30 min. each,
sd., b&w, 16 mm. © William F. Broidy
Productions, Inc.

Ambush. © 30Oct55; LP28061.

The avenging gunman. © 22Nov53;

LP28037.

Battle line. © 23Oct55; LP28060.

Behind southern lines. © 15Apr51;

LP26953.

Blacksmith story. © 10Feb62; LP27760.

Blake's kid. © 9Oct55; LP28058.

Blind alley. © 9Sep56; LP28235.

Bold Raven Rodeo. © 21Oct56; LP28241.

Border City. © 9Dec51; LP27753.

The Border City election. © 14Oct51;

LP26966.

The boy and the bandit. © 5May52;

LP28019.

The boy and the hound dog.

© 16Jan55 (in notice: 1954);
 LP28048.

Buckshot comes home. © 11Sep55;

LP28054.

Chain of events. © 18Oct53; LP28032.

Civilian clothes story. © 27Jan52;

LP27758.

A close shave for the marshal.

© 16Jun62; LP28024.

Cry wolf. © 9Jan55 (in notice: 1954);

LP28047.

The doctor story. © 25Oct53; LP28033.

The dog collar. © 6May51; LP26956.

Ex-convict story. © 16Dec51; LP27754.

The fortune telling story. © 9Jun52;

LP28023.

The Gatling gun. © 18Nov56; LP28245.

Ghost rider. © 7Apr52; LP27768.

Ghost town lady. © 13Nov55; LP28063.

Ghost town story. © 17Jun51; LP26962.

Golden rainbow. © 25Sep55; LP28056.

Gorilla at Owl Hoot Mesa. © 26Dec54;

 LP28045.

Grandpa and genie. © 13Sep53; LP28027.

The great obstacle race. © 16Sep56;

LP28236.

Halley's Comet. © 2Sep56; LP28234.

Hands across the border. © 15Nov53;

LP28036.

Heading for trouble. © 4Oct53;

LP28030.

Hepsibah. © 2Dec51; LP27752.

The hideout. © 6Nov55; LP28062.

Homer Atchison. © 8Jul51; LP26965.

Indian bureau story. © 20May51;

LP26958.

Indian pony express. © 27May51;

LP26959.

Indians and the delegates.

© 1Nov53; LP28034.

The iron major. © 4Nov56; LP28243.

Jingles becomes a baby sitter.

© 26May52; LP28022.

Jingles gets the bird. © 7Oct56;

 LP28239.

Jingles wins a friend. © 11Nov56;

LP28244.

Johnny Deuce. © 1Jul51; LP26964.

A joke on Sir Anthony. © 12May52;

LP28020.

Kangaroo kaper. © 23Sep56; LP28237.

The kid from Red Butte. © 28Nov54;

LP28041.

The lady mayor. © 29Apr51; LP26955.

The lady school teacher. © 28Oct51;

LP26968.

Lost Indian mine. © 6Jan52 (in

notice: 1951); LP27757.

Lumber camp story. © 21Apr52; LP27770.

Marriage feud at Ponca City.

© 30Jun52; LP28026.

Masked riders. © 25Nov51; LP27751.

Masquerade at Moccasin Flats.

© 5Dec54; LP28042.

The maverick. © 21Nov54; LP28040.

Medicine show. © 3Feb52; LP27759.

Mexican gun running story. © 17Feb52;
LP27761.

Mexican rustlers story. © 18Nov51;

LP27750.

Missing diamonds. © 30Sep56; LP28238.

Monster in the lake. © 6Dec53;

LP28039.

The mountain men. © 20Nov55; LP28232.

Music teacher. © 18Sep55; LP28055.

Nephew from back East. © 20Sep53;

LP28028.

Ol' pardner rides again. © 19Dec54;

LP28044.

Old cowboys never die. © 2Oct55;

LP28057.

Outlaw flats. © 4Nov51; LP26969.

Outlaw's portrait. © 4Sep55; LP28053.

The outlaw's son. © 24Mar52; LP27766.

Papa Antinelli. © 23Dec51; LP27755.

Photographer's story. © 17Mar52;

LP27765.
Pony Express vs. telegraph.
© 21Oct51; LP26967.

Prairie flats land swindle.

© 23Jun52; LP28025.

The professor's daughter.

© 10Mar52; LP27764.

The rainmaker. © 28Oct56; LP28242.

The return of Red Hawk. © 27Nov55;

LP28233.

The right of way. © 29Nov53; LP28038.

The Rock Springs rustlers. © 22Apr51;

LP26954.

Sagebrush manhunt. © 13Feb55 (in

notice: 1954); LP28052.

Savvy, the smart little dog.

© 31Mar52; LP27767.

School teacher story. © 24Feb52;

LP27762.

The sheriff was a redhead. © 8Nov53;

LP28035.

The sheriff's secret. © 23Jan55

(in notice: 1954); LP28049.

Silver mine, protection story.

© 13May51; LP26957.
Silver stage holdup. © 11Nov51;
LP26970.

Slocum family. © 30Dec51; LP27756.

The steam wagon. © 25Nov56; LP28246.

Stolen church funds. © 12Dec54;

LP28043.

Sundown Valley. © 6Feb55 (in notice:

1954); LP28051.

Superstition stage. © 22Jan55 (in

notice: 1954); LP28046.

Tax collecting story. © 3Jun51;

LP26960.

To the highest bidder. © 30Jan55

(in notice: 1954); LP28050.

Trapper's story. © 28Apr52; LP27771.

Treasure trail. © 16Oct55; LP28059.

Vigilante story. © 3Mar52; LP27763.

Wagon wheel trail. © 27Sep53; LP28029.

Widow Muldane. © 10Jun51; LP26961.

Wild Bill's odyssey. © 14Oct56;

LP28240.
 Wild white horse. © 14Apr52; LP27769.

Wrestling story. © 19May52; LP28021.

Yellow haired kid. © 24Jun51; LP26963.

Young witness. © 11Oct53; LP28031.

WILD BILL HICKOK, THE LEGEND AND THE

MAN. See

THE GREAT ADVENTURE.

WILD BILL'S ODYSSEY. See

WILD BILL HICKOK.

THE WILD BUNCH. See

SUGARFOOT.

WILD BURRO OF THE WEST. See

WALT DISNEY PRESENTS. Show no. 18.

WILD BUT TRUE; or, Salmon River king.

Del Roby Film Productions. 18 min.,
sd., color, 16 mm. © Del Roby Film
Productions; 3Jan66; MP15909.

WILD CARGO. See

TALES OF WELLS FARGO.

WILD CAT SOUP. See

 KENTUCKY JONES.

THE WILD CHASE. Warner Bros. Pictures.

1 reel, sd., color, 35 mm. (Merrie
melodie) Technicolor. © Warner Bros.
Pictures, Inc.; 27Feb65; LP30650.

THE WILD ELEPHANT. See

MAYA.

THE WILD EYE. American International

Pictures. 91 min., sd., color,
35 mm. © American International Pictures;
14Aug68; LP36163.

WILD GEESE. See

SPORTSMAN'S HOLIDAY.

THE WILD GOAT. See

DAVEY AND GOLIATH.

THE WILD GOOSE RAID. See

RAT PATROL.

WILD HEART. See

WALT DISNEY'S WONDERFUL WORLD OF COLOR.

THE WILD HORSE. See

LASSIE.
WILD HORSE REVENGE. See

WALT DISNEY PRESENTS. Show no. 11.

WILD IN THE COUNTRY. Company of

Artists. Released by Twentieth
Century-Fox Film Corp. 115 min.,
sd., Color by DeLuxe, 35 mm. Jerry
Wald's Production. CinemaScope.
Based on a novel by J. R. Salamanca.
© Company of Artists, Inc. & Twentieth
Century-Fox Film Corp.; 8Jun61;
LP19481.

WILD IN THE STREETS. American International

Pictures. 86 min., sd.,
color, 35 mm. © American International
Productions; 29May68; LP35944.

WILD MAN OF THE JUNGLE. See

JUNGLE JIM. 16.

THE WILD NORTHWEST. See

MISCHIEF MAKERS. 1069.

WILD ON THE BEACH. Lippert. Released

by Twentieth Century-Fox Film Corp.
77 min., sd., b&w, 35 mm. From an
original story by Hank Tani. © Twentieth
Century-Fox Film Corp.; 11Aug65;
LP31750.

THE WILD ONE. See

 BONANZA.

THE WILD RACERS. Alta Vista Film Productions.

Released by American International.
80 min., sd., color, 35 mm.
© Alta Vista Film Productions, Ltd.;
20Mar68; LP35524.

THE WILD REED. See

JOHNNY STACCATO.

WILD RIVER. Twentieth Century-Fox Film

Corp. 109 min., sd., Color by
DeLuxe, 35 mm. CinemaScope.
Based on novela by William Bradford
Huie & Borden Deal. © Twentieth
Century-Fox Film Corp.; 22May60;
LP16409.

WILD SEED. Pennebaker, Inc. Released

by Universal Pictures Co. 99 min.,
sd., b&w, 35 mm. © Pennebaker, Inc.;
31Jul65; LP35002.

THE WILD SWANS. See

SHIRLEY TEMPLE'S STORY BOOK.

THE WILD WATER CHAMPIONSHIP. See

SPORTFOLIO.

THE WILD WEST SHOW. See

GUESTWARD HO.
WILD WEST STORY. See

OUT OF THE INKWELL.

THE WILD WESTERNERS. Four Leaf Productions.

Released by Columbia Pictures
Corp. 70 min., sd., Eastman Color,
35 mm. © Four Leaf Productions, Inc.;
1Jun62; LP22250.

THE WILD WEST'S BIGGEST TRAIN

HOLDUP. See

DEATH VALLEY DAYS.

WILD WHITE HORSE. See

WILD BILL HICKOK.

THE WILD, WILD PLANET. Southern

Cross Films, Ltd. Released by Metro-Goldwyn-Mayer.
95 min., sd., color,
35 mm. © Southern Cross Films, Ltd.;
31Dec65; LP34711.

WILD, WILD WAKE. See

RUN, BUDDY, RUN.

WILD, WILD WEST. Columbia Broadcasting

System. Approx. 60 min. each, sd.,
b&w, 16 mm. Produced in association
with CBS Television Network. © Columbia
Broadcasting System, Inc.
The night of a thousand eyes.
© 20Oct65; LP32109.

The night of the bars of hell.

© 1Mar66; LP33003.

The night of the big blast.

© 4Oct66; LP33818.

The night of the bottomless pit.

© 1Nov66; LP33822.

The night of the burning diamond.

© 5Apr66; LP33267.

The night of the casual killer.

© 12Oct65; LP31942.

The night of the dancing death.

© 2Nov65; LP32112.

The night of the deadly bed.

© 21Sep65; LP31945.

The night of the double-edged knife.

© 8Nov65; LP32111.

The night of the druid's blood.

© 22Mar66; LP33268.

The night of the eccentrics.

© 13Sep66; LP33815.

The night of the fatal trap.

© 17Dec65; LP32496.
The night of the flaming ghost.
© 18Jan66 (in notice: 1965);
LP33004.

The night of the flying pie plate.

© 18Oct66; LP33820.

The night of the freebooters.

© 28Mar66; LP32880.

The night of the glowing corpse.

© 26Oct65; LP32113.

The night of the golden cobra.

© 20Sep66; LP33816.

The night of the Grand Emir.

© 25Jan66; LP32877.

The night of the green terror.

© 14Nov66; LP34016.

The night of the howling light.

© 10Dec65; LP32497.

The night of the human trigger.

© 30Nov65; LP32499.

The night of the inferno. © 14Sep65;

LP31941.

The night of the murderous spring.

© 12Apr66; LP32881.

The night of the poisonous posey.

© 25Oct66; LP33821.
The night of the puppeteer.
© 23Feb66; LP32878.

The night of the raven. © 27Sep66;

LP33817.

The night of the red-eyed madman.

© 23Nov65; LP32495.

The night of the returning dead.

© 11Oct66; LP33819.

The night of the steel assassin.

© 3Jan66 (in notice: 1965); LP32875.

The night of the sudden death.

© 5Oct65; LP31943.

The night of the sudden plague.

© 19Apr66; LP32882.

The night of the torture chamber.

© 7Dec65; LP32498.

The night of the two-legged buffalo.

© 7Mar66; LP32879.

The night of the whirring death.

© 15Feb66; LP33002.

The night that terror stalked the town.

© 15Nov65; LP32110.

The night the dragon screamed.

© 11Jan66 (in notice: 1965); LP32876.
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookname.com