• No results found

Synchronous Railway Traction Systems with Railway Power Conditioners

N/A
N/A
Protected

Academic year: 2021

Share "Synchronous Railway Traction Systems with Railway Power Conditioners "

Copied!
51
0
0

Loading.... (view fulltext now)

Full text

(1)

UPTEC ES 16002

Examensarbete 30 hp Januari 2016

Compensating Unbalances in

Synchronous Railway Traction Systems with Railway Power Conditioners

Matilda Örnkloo

(2)

Teknisk- naturvetenskaplig fakultet UTH-enheten

Besöksadress:

Ångströmlaboratoriet Lägerhyddsvägen 1 Hus 4, Plan 0

Postadress:

Box 536 751 21 Uppsala

Telefon:

018 – 471 30 03

Telefax:

018 – 471 30 00

Hemsida:

http://www.teknat.uu.se/student

Abstract

Compensating Unbalances in Synchronous Railway Traction Systems with Railway Power Conditioners

Matilda Örnkloo

The electrified railway presents significant challenges for the electrical grid. This is due to the characteristics of the constructed railway system. Trains are single-phase loads, fed by two adjacent phases from the grid. Feeding phases will change continuously at every substation. This load characteristic will lead to unbalances and poor power quality in the grid. The poor power quality is caused by the unbalance in currents, voltage drops along the line, and induced harmonics from power electronic devices used in traction.

To decrease the impact of the railway traction system in the public grid, Static Var Compensators (SVCs) and Static Synchronous Compensators (STATCOMs) have been implemented. These installations offer voltage control, maintain balance and mitigate harmonics. This thesis investigates other power electronic technologies to improve the power quality in the grid for the 50 Hz railway traction system.

Handledare: Antonios Antonopoulos

(3)

Popul¨arvetenskaplig Sammanfattning

I dagens elektrifierade j¨arnv¨ag ger t˚agdriften upphov till obalanser i eln¨atet. T˚agen matas fr˚an en kontaktlina ovanf¨or r¨alsen. Kontak- tlinan matas i sin tur med str¨om fr˚an transformatorer kopplade till ett ¨overliggande eln¨at. Konstruktionen med en matande kontaktlina leder till en lastkarakt¨ar som medf¨or obalanser i eln¨atet.

N¨ar intensiteten ¨okar p˚a j¨arnv¨agen och allt tyngre laster transporteras blir dessa obalanser st¨orre. Det kan medf¨ora problem f¨or andra kunder i transmissionsn¨atet som efterfr˚agar el av bra kvalit´e. En bra elkvalit´e inneb¨ar bland annat att den effekt som skickas ut p˚a n¨atet best˚ar av r¨att sp¨anningsniv˚a, har lite ¨overtoner och h˚aller r¨att frekvens. Det konventionella j¨arnv¨agssystemet har enfas-transformatorer inkopplade l¨angs med banstr¨ackningen. Dessa transformatorer tar tv˚a faser fr˚an n¨atet och matar sin sekund¨ara sp¨anning till t˚agens kontaktlina samt en till jord. Fasmatningen ¨andras kontinuerligt f¨or att i st¨orre m˚an dra en j¨amn effekt fr˚an n¨atet. T˚agen ¨ar dock inte en konstant ef- fektf¨orbrukare, utan lasten varierar i och med t˚agens acceleration. Den h¨ar lastkarakt¨aren g¨or det mycket sv˚art att f˚a ett balanserat n¨at. Det finns ¨aven andra problem med dagens konventionella j¨arnv¨agssystem med enfas-transformatorer. N¨ar banan matas med sp¨anning fr˚an olika faser kan inte kontaktlinan ¨over t˚agen vara en kontinuerlig ledare. Vid varje transformatorstation m˚aste linan styckas upp. Detta ger upphov till str¨ackor d¨ar t˚agen k¨or utan effekttillf¨orsel. Str¨ackningen varierar i l¨angd men kan vara upp till 1000 meter l˚anga p˚a vissa str¨ackor. Vid installation av h¨oghastighetslinjer ¨ar detta inte en ¨onskv¨ard egenskap.

Studier har tidigare gjorts f¨or att studera system som kan f¨orb¨attra elkvalit´en i n¨atet. Tidiga l¨osningar har varit att installera reaktiva effektkompenserare i n¨atet. Senare har ¨aven aktiv-effektkompenserare studerats samt direktomvandling fr˚an h¨ogsp¨anningsn¨atet genom en tre-fas till en-fas konfiguration av tv˚a v¨axelriktare.

Det h¨ar examensarbetet kommer att studera aktiv effektkompenser- ing genom en s˚a kallad Railway Power Conditioner. Efter en litter- aturstudie valdes en topologi ut och dess analytiska uttryck togs fram f¨or att best¨amma dess funktion. Vidare gjordes simuleringar p˚a den valda topologin med aktiv effektkompensering samt simuleringar p˚a ett konventionellt system f¨or att kunna j¨amf¨ora prestation. Simu- leringar och ber¨akningar har utf¨orts i PSCAD respektive i MATLAB.

Resultatet fr˚an simuleringarna visar tydligt vad som ¨ar problem med dagens elektrifierade j¨arnv¨ag. Obalansen ¨ar stor och ger upphov till d˚alig elkvalit´e som kan st¨ora andra komponenter i elkraftssystemet.

(4)

Med den aktiva effektkompenseraren kunde d¨aremot elkvalit´en i n¨atet f¨orb¨attras avsev¨art. Det finns idag redan realiserade system installer- ade av denna typ i Shinkansen, Japan. F¨or att avg¨ora vilken metod som l¨ampar sig b¨ast f¨or j¨arnv¨agen kr¨avs dock ytterligare studier b˚ade inom systeml¨osningarnas tekniska prestation men ¨aven hur dessa kan m¨atas ur ett ekonomiskt perspektiv.

(5)

Executive summary

This thesis show how the use of a conventional railway traction system leads to unbalances in the electrical grid. Unbalances have a negative impact on apparatus in the network, leading to lower over-all performance of the power system. The power quality may even get worse with trains running at higher speeds or trains with a heavier load.

To mitigate this unbalance, power electronic devices can be installed.

This study show that railway power conditioners (RPCs) can mitigate the unbalance in the grid, as they reduce the degree of current unbal- ance significantly.

To determine if the RPC is a competitive solution for the railway traction system, further studies must be performed. The system must be investigated in terms of harmonics and in what way the harmonic distortion from both the converters and the traction equipment on the trains can be reduced. The RPC-system could be compared with other compensation systems, such as with a STATCOM. The compar- ison could include both cost and performance to determine the most suitable solution.

(6)

Acknowledgement

I appreciate all the guidance, advice and encouragement that I have got throughout this work, from people around me.

Especially I value the patience and assistance from my supervisor, Antonios Antonopoulos at ABB Corporate Research, V¨aster˚as. With his pedagogical explanations he could help me to overcome some mind barriers during the work and without his support I would not have been able to finish this work.

I also appreciate the time my subject reviewer, Urban Lundin at Uppsala University took to answering my questions and for reading my report. I also would like to thank my proof reader and opponent Siddy Persson for

additional comments.

At last I would like to thank the Swedish state for the opportunity of free education.

/Matilda ¨Ornkloo, Uppsala

(7)

Contents

1 Introduction 9

1.1 Scope of the Project . . . 9

1.2 Background . . . 9

1.3 Method . . . 11

1.4 Structure . . . 12

1.5 System boundaries . . . 12

2 Conventional System with Single-phase Transformers 13 2.1 Consequences with unbalances . . . 14

2.2 Symmetrical Components . . . 15

3 Solutions to Mitigate Unbalances with Power Electronic De- vices 16 3.1 Static Frequency Converter (SFC) . . . 16

3.2 Reactive power compensators, (SVC) and (STATCOM) . . . 17

3.3 Railway Power Conditioner (RPC) . . . 19

4 Railway Power Conditioner, Design and Function 21 4.1 Special Design Transformers . . . 21

4.2 Generation of Current and Voltage References for an RPC connected behind a V/V Transformer . . . 21

4.2.1 Control Schematic for the Railway Power Conditioner . 26 4.2.2 Parameter description and impact of parameter change 31 5 Simulations and Results for a Conventional system and a system with the RPC implemented 32 5.1 Conventional System . . . 32

5.1.1 Case 1 . . . 33

5.1.2 Case 2 . . . 35

5.1.3 Case 3 . . . 36

5.1.4 Case 4 . . . 38

5.2 Railway Power conditioner . . . 40

5.2.1 Case 1 . . . 40

5.2.2 Case 2 . . . 42

5.2.3 Case 3 . . . 43

5.2.4 Case 4 . . . 45

5.3 Unbalance factor . . . 47

6 Conclusion and Discussion 48

(8)

7 Future outlooks 49

8 References 50

(9)

1 Introduction

Trains appear as a single-phase loads in the public grid and create unbal- ances. This leads to poor power quality and may disturb other customers in the grid. When the traffic increases the issue becomes larger and solutions to improve the power quality are necessary. The poor power quality appears mainly from unbalances in the current on the grid side. The unbalances occur due to the train-load characteristic and it has a negative impact on trans- formers, generators and motors in the grid. It will also lower the capacity of the transmission lines.

If the railway is placed in remote areas, the grid might be weaker and more susceptible to voltage drops. Power-electronic device in the traction system also increase the harmonic content in the railway. To mitigate these issues SVCs and STATCOMs have been installed along the railway network. This thesis will study alternative power electronic applications to improve the power quality.

1.1 Scope of the Project

The purpose of this master thesis is to describe the unbalance problem in the synchronous railway system, study what consequences it has in the electricity network, and investigate systems that can mitigate these problems.

1.2 Background

The electrified railway system is divided into direct current (DC) and al- ternating current (AC) systems. Early electrification before 1950 was done using mainly DC systems or low-frequency AC systems. Later systems, de- signed after 1950 are often electrified by AC with an industrial frequency at 50 or 60 Hz. This transition of electrification systems is due to the following advancement in technology: Before the 50s railway-traction motors could not be fed by industrial frequencies, but at lower frequencies 16 2/3 Hz or 25 Hz [1]. To change the industry frequency from 50 Hz to a lower frequency that the railway can operate with, rotary converters were historically used [1].

When modernizing low-frequency railways, nowadays, static converters are installed that are based on semiconductor technology. Changing the railway electrification system in a region is often associated with large investment cost. Hence when a system is built and in operation, it will most likely be in operation during a long time. A system will rather be upgraded than being frequently redesigned from the beginning. Various system are, therefore, in

(10)

operation around the world.

DC systems are often fed by a low voltage, typically between 500 to 3500 Volts [1]. The comparative low voltage has to do with safety and constraints regarding technology and economics at the time the DC systems were intro- duced. The safety issue mainly refers to the fact that high-voltage breakers were difficult and expensive to make at that time. A reasonable way to trans- form the voltage easily between different DC voltage levels was not available.

However, with modern technology some of these constraints could be dealt with [2]. DC systems also require a large current for a given power and as result the resistive losses will increase. These systems are mainly used in urban areas for subways or trams.

AC systems span a larger voltage range, typically between 15 kV up to 50 kV. For low-frequency railways (16 2/3 Hz), 15 kV is used and for industrial frequency systems (50 or 60 Hz), 25 kV is used. Low-frequency railways can be found in northern and central Europe since the standard was set when traction-motor technology for higher frequencies was not mature. In America there are some grids operating at 25 Hz, and in Japan where the country has two frequency regions, 50 Hz and 60 Hz, most of the trains run at 60 Hz [3].

Modern electrification is generally done with industrial frequency at 50 Hz, but still there are several different systems in operation in the world [1]. The advantage with an AC system compared to a DC system is that a higher voltage can be used. For the same power the trains can thereby be fed by a lower current and the resistive losses are decreased. When higher power is needed for the trains, the more obvious the advantage with AC systems will be. With high speed trains, a DC system would require substations on fairly short distances. This will not be viable since trains are occasional consumers of power and do not require a constant power supply [1].

For modern railway electrification the most common system is a 25 kV AC system at 50 Hz. It does not require a frequency converter and the transform- ers can be lighter due to the higher frequency [1]. However, all AC systems are still being fed as single-phase loads. In theory it would be better to draw a three-phase current immediately from the grid and have more conductors on the railway. That would reduce the problem of unbalances, but this has not been a practical solution in reality. There are only a few three-phase railway traction systems. More conductors at each railway system would also cost too much since the loads are not a consumer of constant power.

(11)

1.3 Method

This thesis is divided into two main parts. The first part includes a literature study to introduce the reader to the content and provide an overview over the problem. The second part includes an attempt to find a solution for the problem.

First investigations have been done on power electronic devices that can mitigate unbalances in the electrical grid produced from the electrified rail- way traction system. Main focus in the thesis has been on the Railway Power Conditioner (RPC). The RPC is a power electronic device that has not been widely investigated before. The converter configuration enables active power to flow between two electrical subsystems, which can be compared to STAT- COMs that mainly work with reactive compensation.

Both a typical railway traction system, in this thesis named Conventional system and a conventional system with an RPC in operation are designed and simulated in this thesis. The purpose with this is to be able to show the performance of these systems when subjected to the same train-load. The two systems are built in Power Systems Computer Aided Design (PSCAD).

PSCAD is a simulation program to build, simulate and model a system of interest. PSCAD include both ready-made modules but also allow the user to construct and write code for costume-made modules.

(12)

1.4 Structure

The main structure of the thesis include following four chapters

• Chapter 2 - Conventional System

The typical railway traction system used in the railway industry today is described including its features

• Chapter 3 - Solutions to Mitigate Unbalances

This chapter describes power electronic solutions for the railway trac- tion system to mitigate the unbalance in the electrical grid

• Chapter 4 - Railway Power Conditioner

The Railway Power Conditioner is described and investigated in terms of an electrical analysis. Information from the investigation is further on used for construction and control of the converter configuration in PSCAD

• Chapter 5 - Simulations and Results

In this chapter results from the simulations are presented. Simulations have been done on the conventional system and on a conventional sys- tem with an RPC in operation. This to show their performance when subjected to a train-load

1.5 System boundaries

This thesis only considers railway systems from a three-phase AC system to the single-phase train load. The simulations are only done on systems operating with a frequency at fixed value of 50 Hz. The grid is assumed to be a three-phased sinusoidal system with no harmonic distortion and with a fixed voltage amplitude, unless noted otherwise. The only reference to harmonics will be in the rectified-traction-load characteristics and the railway power conditioner. Power losses and fault investigations are not carried out within the scope of this project.

(13)

2 Conventional System with Single-phase Trans- formers

In a conventional traction system trains are being fed by two adjacent phases through a single-phase transformer, as can be seen in figure 1. In order to equally load all three phases, the phase connection changes in every substa- tion. With this supply system the currents on the grid side will be unbalanced at every single instant in time, due to the single-phase load characteristic.

Furthermore, the railway is required to have neutral zones, marked as NZ in the figures, to divide the feeding sections that might vary in amplitude, frequency or phase shift. During these sections the trains will have to run unpowered [4].

The definition of a neutral zone is the distance where the train is not pow- ered. For high-speed and heavy-loaded trains this is a drawback, so the length and/or the amount of neutral zones should be reduced or avoided if possible. In Denmark there has been a detailed study about neutral sections since a high-speed railway is planned. For trains running at a speed below 200 km/h section insulators have been used and the neutral sections have been 8 meters long. For high-speed trains, with speeds above 200 km/h this construction does not fulfill the given standards and other types of neutral sections have to be used, where most of them are over 100 meters long or even up to 1000 meters long [4] [5].

The conventional system is simple, has a low investment cost, and the existing knowledge is large in this technology. However, when the expectations on the railway increases, regarding denser traffic and heavier loaded trains, certain improvements or even a complete redesign of the whole system is required [6].

It is no longer viable to have unpowered distances in the railway system.

Moreover, the conventional system provides poor power quality in the grid.

A modern railway system should preferably be able to maintain the voltage levels within given limits, mitigate harmonics from the traction equipment, and reduce the negative-sequence currents in the system. The amount of neutral zones should be reduced or avoided if possible.

(14)

Figure 1: Conventional railway feeding system with single-phase transformers and a conductor divided by neutral zones

2.1 Consequences with unbalances

Under unbalanced situations, negative-sequence current will occur in the grid. The magnitude of these negative sequence currents depends on the degree of unbalance. The negative-sequence currents have a negative impact on the grid and apparatuses that operate in the grid. It will decrease the transmission capability, lower the output power from the transformer, and disturb rotating machines like motors and generators [7].

Negative-sequence currents in electrical machines such as in a motor or a generator will decrease their efficiency. The negative-sequence currents will create a magnetic flux in the rotor that oppose the direction of rotation of the rotor. A relative motion will occur between the magnetic flux and the rotor.

This will cause rotor heating problems and may lead to insulation failure and mechanical problems [8]. Heating problems will demand extra cooling, and the efficiency of the machine will decrease accordingly. The positive sequence current will induce a magnetic flux in the same rotation as the rotor.

The negative-sequence current also affects the transmission lines. It flows in the transmission lines and does not perform any actual work. It cause additional power loss, and lower the transmission capability of the line.

(15)

Transformers will also be affected. The negative-sequence current in the grid results in asymmetrical three-phase currents on the primary side of trans- formers. The rated output of a transformer will be limited by the largest current on the primary side, and thereby the output power will be lower.

This leads to additional losses in a transformer and heating of the iron. The negative-sequence current may also disturb the relay protection in the grid [7].

The consequences in the grid from negative-sequence currents are severe and detrimental for most apparatus. The amount of negative-sequence current should, therefore, be reduced as much as possible.

2.2 Symmetrical Components

With symmetrical components it is possible to analyze unbalances in a multi- phase system in a simplified manner. The number of phases, n can be divided in n systems of balanced phasors, called the symmetrical components. Each set of symmetrical components is equal in length, and adjacent phasors will be separated by the same angle [9]. However, since most electrical systems operate with three phases the following description will be for a three-phase system. The main concept with symmetrical components is to transform the phase component to the new set of symmetrical components. The advantage of this method is that the sequence networks are easier to analyze compared to the three-phase network. Once the sequence networks are found, it is pos- sible to transpose it to the three-phase network again [10]. This method can be applied for both voltages and currents in the network.

The transformation from phase currents to the sequence currents can be seen further down.

 Ia Ib Ic

=

1 1 1

1 a2 a 1 a a2

 Ia0 Ia1 Ia2

 (1)

Ia0 = 1

3(Ia+ Ib + Ic) (2)

Ia1 = 1

3(Ia+ a2Ib+ aIc) (3) Ia2 = 1

3(Ia+ aIb+ a2Ic) (4)

(16)

3 Solutions to Mitigate Unbalances with Power Electronic Devices

To improve stability of the grid, primarily regarding voltage levels, it is possi- ble to have two feeding lines. Even if the grid dynamics may be improved with this solution it will not solve the problems with negative-sequence currents or harmonics. Double feeding lines will still require neutral zones to separate the feeding phases. There are also other solutions for the load-unbalance problem. Reactive-power compensators such as the SVC or STATCOMs can be used. The term static refers to the fact that no rotating part is involved, the devices use static semiconductor-based technology instead. Another in- teresting solution that is less investigated compared to the aforementioned ones is to use railway power conditioners (RPCs). In this project focus is to investigate the RPC and its features. Simulations are performed on the conventional system and the same system when the RPC is implemented.

Another available technology is static frequency converters (SFCs). SFCs operate with converters that immediately convert a three-phase input to a single-phase output and provide an in-line feeding system. This system does not require any neutral zones. However, it is considered to be quite expensive to install, as it requires full-power converters at every substation.

When a train brakes power is dissipated in the catenary. In a conventional system with single-phase transformers this power will flow back up through the transformer to the grid. This is often not desirable from the grid own- ers view, as the power fed back to the grid will be an uncontrolled power flow in time, magnitude and place. Therefore, there are often restrictions regarding power fed-back and some trains are installed with certain brakes to dissipate the excess power in terms of heat within the locomotive. With highly dense traffic railways, such as metro lines, the power can to a larger extent be used by adjacent trains in the system avoid feeding the power back to the network. With a system design with SFCs, it is possible to control the power flow feedback to the grid. All three phases can receive power in a symmetrical manner. This is not possible with single-phase transformers [1].

3.1 Static Frequency Converter (SFC)

Recently, full converters are proposed for the 50 Hz railway system, similarly to the full-converter design in the 3-phase AC 50 Hz to 1-phase AC 16.7 Hz systems, as shown in figure 2. The full-converter design consists of a three-phase converter connected to a single-phase converter in back-to-back

(17)

configuration with a common DC-link. Active power, reactive power, and harmonics can be controlled dynamically. This technology does not need neutral zones as the catenary is separated from the feeding grid. This makes the system flexible and suitable for fast and heavy-loaded trains. Step-up and step-down transformers can be used to obtain the desirable rating on the converters and for galvanic isolation against the grid [11]. The SFC is controlled in a way that the grid sees a symmetrical load, and therefore, the voltage requirements on the grid can be reduced. This technology can also provide the possibility for regenerating power flow, i.e., braking energy from the train can transferred to the grid. It can even improve the power quality in the grid, but has a higher capital cost compared to other systems [6].

Figure 2: Static Frequency Converter connecting directly to the high power grid through a coupling transformer

3.2 Reactive power compensators, (SVC) and (STATCOM)

In order to improve the power quality, without tremendously increasing the initial cost, FACTS devices can be installed in the conventional system. With these devices it is possible to use lower grid voltage for the railway than without them, since the unbalances will be reduced. Two common shunt- connected devices used for dynamical balancing is the Static Var Compen- sators (SVC) or the Static Synchronous Compensator (STATCOM). Both

(18)

technologies have been installed in several systems and have improved the power quality in the grid. In a railway grid, they can offer dynamic voltage control, mitigate harmonics and balance the load between phases. This tech- nology has been installed for this purpose in Japan, England and France [3]

[12].

These reactive power compensators still require neutral zones, since the feed- ing is done by single-phases transformers as in figure 3. SVCs and STAT- COMs are typically connected to the grid via a coupling transformer to step- down the voltage to a level that does not require a high rating of the com- pensator. The voltage regulation is done by the switching devices and the harmonics can be mitigated by additionally connected filters [12].

The STATCOM can also be connected in a system were the trains are fed by the same combination of phases all along the line. This requires higher ratings for the STATCOM, since the unbalances will be larger, as seen from the grid. However, one can theoretically avoid having neutral zones with this constellation. Section insulators might be needed between adjacent railway sections. As a result this is not widely used in practice as this technology can result in undesired power flows between two points in the grid. Through the railway and up through the feeding sections [11]. STATCOMs and SVCs can be suitable solutions for existing grids, since the infrastructure of the railway grid is already in place with single-phase transformers at the feeding points, and the STATCOM/SVC can be used as add-on equipment in the grid to increase the power quality.

(19)

Figure 3: Conventional system with single-phase transformers, grid con- nected to a STATCOM through a coupling transformer

3.3 Railway Power Conditioner (RPC)

Railway Power Conditioners (RPCs) were introduced by Japanese scholars in the 90s and has been in commercial operation in Shinkansen since the beginning of 2000 [13]. Similar devices have also been described in papers in other constellations or with added components that improve their perfor- mance [14]. RPCs do not have a unique definition, but they are often referred to a conditioner placed after a special design transformer, such as a Scott or V/V-transformer, as shown in figure 4. Furthermore the conditioners often consist of two single-phase AC/DC- converters connected back-to-back with a common DC-link. They are placed between the transformer secondary sides.

These devices enable active power flow between two sections of the cate- nary. During an upgrade of Shinkansen in 2002 several RPCs were installed in the railway system to obtain a balanced public grid. Later demonstration could show that balance were established between the phases but the RPC could also compensate for reactive power and keep the voltage levels within given requirements [13].

As modern trains also tend to a larger extent use pulse width modulation (PWM) control the need of compensating reactive power is decreased while

(20)

the need of active power transfer still exists [15].

The major advantage with this technology is that total balance can be achieved in the electrical grid. The major disadvantage is that this tech- nology still requires neutral zones.

Figure 4: Railway power conditioner connected to the grid via a step-down transformer

(21)

4 Railway Power Conditioner, Design and Func- tion

The main purpose with an RPC is to:

• Transfer active power between two electrical subsystems

• Compensate reactive power on each side of the converters

• Mitigate harmonics

In this thesis no detailed investigation is done to study harmonics. Focus has been on the first two items, transfer active power and reactive power compensation. The RPC is connected on the secondary side of a traction transformer. The traction transformer is often a special designed transformer since the primary side connects to the three-phase grid and the secondary side consists of two phases and one connection to ground.

4.1 Special Design Transformers

To obtain a more even distribution of loads when designing a railway traction system it is common to use special-design transformers. Special-design trans- formers include impedance-matching transformers or three-phase to two- phase transformers. Common traction transformers are the Scott, the Wood- bridge, the Le Blanc or the V/V transformer. Due to rapid changes in the train loads on each side of the transformer output, the system cannot main- tain its balance but still require an additional load compensator [16].

The choice of transformer will also affect the amount of negative-sequence currents injected in the grid. The V/V-transformer will inject more negative- sequence current than the Scott or the Woodbridge transformer [17]. How- ever, the V/V transformer has other desirable characteristics, such as a simple structure, and this is why it is widely used in railway traction systems [17].

4.2 Generation of Current and Voltage References for an RPC connected behind a V/V Transformer

To determine the required output from the converter to keep the system symmetrized, the electrical circuit must be analyzed. Figure 5 represents a schematic overview of the system.

(22)

From the three-phase grid a V/V-transformer is connected to the RPC. The V/V transformer is a transformer with three phases, two windings on the primary side connected to the grid. On the secondary side there are two phases connected to the load and one phase connected to the ground. Be- tween the phases on the secondary side two converters can be connected in back-to-back configuration. These two feeding phases are named according to the voltage across them, the ac-side (ac) and the bc-side (bc).

Figure 5: System overview, V/V transformer with connected RPC. Iac and Ibcrepresent the currents from the V/V transformer output-side and ILacand ILbc represent the load currents. Between the two feeding phases the railway power conditioner is placed. Two converters in a back-to-back configuration where Iacref and Ibcref determine the reference currents from the converters

In the initial system the three-phase voltages on the grid side are assumed to be purely sinusoidal with a phase shift of 120 degrees. The grid-side voltages are given in line-to-neutral values and the transformer turns ratio is a constant called K. The transformer is assumed to be ideal. If the input is symmetrical, then following equations can be written for the reference currents. Omega, ω represents the angular speed of the grid.

VA= V1cos(ωt) → IAref = I1cos(ωt) (5)

VB = V1cos(ωt − 2π/3) → IBref = I1cos(ωt − 2π/3) (6)

VC = V1cos(ωt + 2π/3) → ICref = I1cos(ωt + 2π/3) (7)

(23)

Assuming that the railway system appears only as an active load at the point of common coupling, the voltage and the current can be related by a pure resistance, I1 = VR1. The active power on the primary side of the grid can be described

P1 = 3

2V1I1 (8)

The train load on each of the secondary sides of the transformer can be represented by one resistive part and one inductive part,

Zac = Rac+ jωLac (9)

Zbc = Rbc+ jωLbc (10)

The voltages on the secondary side can be described as

Vac(t) = V2cos(ωt + ϕ) (11)

Vbc(t) = V2cos(ωt + γ) (12) The variables ϕ and γ represent the phase displacement from the initial angular speed for the system, and occur due to the load characteristic. As- suming that the train loads are PWM-controlled their reactive part can be neglected and the currents can be expressed as

ILac(t) = ILaccos(ωt + ϕ) (13) ILbc(t) = ILbccos(ωt + γ) (14) The active power on the secondary side

P2 = V2

2 (ILac+ ILbc) (15)

The active power drawn from the secondary side is assumed to be the same as on the primary side for an ideal transformer. Form (8) and (15) it is possible to find the current amplitude I1. The ratio between the line-to-neutral and line-to-line voltages and the transformer ratio must be considered at this step. Assuming P1 and P2 are equal this yields the current amplitude on primary side of the transformer, for a symmetrically shared load

I1 = k√ 3

3 (ILac+ ILbc) (16)

The reference input currents for a balanced system are expressed by

(24)

IAref = k√ 3

3 (ILac+ ILbc) cos(ωt) (17) IBref = k√

3

3 (ILac+ ILbc) cos(ωt − 2π/3) (18) ICref = k√

3

3 (ILac+ ILbc) cos(ωt + 2π/3) (19) The reference currents on the primary side must now be transferred to the secondary side. Figure 6 describes how the currents flow through the V/V transformer. Uppercase letters refer to the primary side of the transformer and lowercase letters refer to the secondary side.

Figure 6: Schematic over current division through the V/V transformer, primary side is noted by uppercase letters and the secondary side is noted by lowercase letters. The transformer has a three-phase input and two output- feeding phases and one connected to ground. Primary voltage is set to 100 kV RMS, L-N and the secondary voltage 25 kV RMS.

The current on the primary side can be written as

Ipr.AC = iA (20)

Ipr.BC = iB (21)

(25)

Ipr.AC+ic = −Ipr.BC → iC = −Ipr.BC−Ipr.AC → iC = −iB−iA (22) These currents must then be transferred to the secondary side of the transformer

Isec.ac= +iac (23)

Isec.bc= +ibc (24)

Isec.ac = −ic−Isec.bc → ic= −Isec.ac−Isec.bc → ic= 1

k(Ipr.AC+Ipr.BC) (25) isecC = 1

k(iA+ iB) (26)

The currents on the secondary side can now be expressed from the primary side currents

iac = Isec.ac = −1

kIprim.ac = −1

kiA (27)

ibc = Isec.bc = −1

kIprim.bc = −1

kiB (28)

In figure 5 the current schematic is shown and the reference currents from the converter can be described with (17), (18), (27) and (28)

irefac = −iac− iLac= 1

kiA− iLac (29)

irefbc = −ibc− iLbc= 1

kiB− iLbc (30)

Iacref =

√3

3 (ILac+ ILbc) cos(ωt) − ILaccos(ωt + ϕ) (31) Ibcref =

√3

3 (ILac+ ILbc) cos(ωt − 2π/3) − ILbccos(ωt + γ) (32) When the reference currents are found for the converters, the voltages can be calculated using Kirchhoffs voltage law. Reference voltage for the converter must be the voltage over the catenary minus the voltage over the inductor, as shown in figure 7.

(26)

Figure 7: Schematic over voltage division, Vac represent the voltage on the feeding phase, VL1 is the inductor voltage and Vacref refers to the voltage reference from the converter.

Vacref = Vac− VL1 → VL1 = jωL1Iacref (33)

Vbcref = Vbc− VL2 → VL2 = jωL2Ibcref (34) The inductance in the coupling reactor will cause a phase shift of 90 degrees and the reference voltages on each side will be

Vacref = Vac−ωL1(

√3

3 (ILac+ILbc) cos(ωt−π/2)−ILaccos(ωt+ϕ−π/2)) (35)

Vbcref = Vbc−ωL2(

√3

3 (ILac+ILbc) cos(ωt−2π/3−π/2)−ILbccos(ωt+γ −π/2)) (36) The reference currents and voltages for both converters will be variables for the control of the RPC.

4.2.1 Control Schematic for the Railway Power Conditioner The main function with this power conditioner is to allow active power to flow between two electrical subsystems. The power conditioner contains two converters connected in back-to-back configuration with a common DC-link, as in figure 8. These converters can operate in both rectification and inversion mode.

(27)

Figure 8: Two converters in back-to-back configuration, each converter con- sists of four switches two in each leg. The switching devices are Insulated-gate Bipolar Transistors (IGBTs). The DC-link capacitor is set to 5 µF.

The controller should use the reference currents and voltages to sym- metrize the currents on the grid side. This control system will use a sinu- soidal pulse width modulation (SPWM) to create gate signals for the IGBTs.

The SPWM compares a reference signal to a carrier signal to create the gate pulses. The carrier signal will be a triangular waveform with maximum out- put of 1 and a minimum output of -1. The frequency is initially set to 10 kHz for the triangular waveform.

From chapter 4.2, equations for the reference currents and the reference volt- ages can be found. The first attempt to build the controller only considered the voltage references. In figure 9 the voltage reference, Vacref is normalized over the DC-voltage. With a simple voltage control, the gate signals can be found immediately after the comparison of the signal and the carrier signal to determine if the switch should be in on-state or off-state. However, in this case a simple voltage control was not sufficient and an additional step was required.

(28)

Figure 9: Voltage control of the converter. The voltage reference is normal- ized over the DC-voltage and further on compared to a carrier signal

The controller based on only the reference voltages created currents that did not coincide with the reference currents. A control for the current was necessary. In figure 10 the measured current, IacM easured is allowed to deviate from the reference current, Iacref by a certain tolerance. In the simulations the tolerance is set to 100 A. A smaller tolerance will create a current that follows the reference current more accurately, but may cause unnecessarily high switching frequency. A larger tolerance might not be able to create currents and voltages that follow the references.

From the left side of the figure, Iacref is compared with an upper limit Iref +tolerance

ac .

The output from the comparator is 1 if signal A is larger than B, or 0 if sig- nal A is less than B. Further on in the input selector either A or B will be let through depending on the control signal, Ctrl. Signal A is set to 0 and signal B will either be 0 or 1 according to the voltage control output. When the control signal is 1, signal A will be let through. If the control signal is zero, signal B will be let through. The output from the first input selector in figure 10 will further on be used in the second input selector.

The second input selector uses a similar operation procedure. The output will determine if the device 1 and 2 will be on or off. This is the control func- tion for one leg in one converter. For the second leg in the same converter a similar procedure will be run as can be seen in figure 11.

(29)

Figure 10: Current control for gate 1 and gate 2. The current reference from the converter is let to vary within a band of tolerance. When the current appears outside this band the switches are forced to turn on or off depending on if more or less current is required

Figure 12 shows reference current, Iacref plotted along with the measured current, IacM easured and the band of limitation. The main purpose of the band is to maintain the current within given limits. For example, if IacM easured is under the band of limitation the current must be increased. This requires higher voltage and the switches must operate in a way that result in a higher voltage on the AC-side.

This is the control schematic for one of the converters, a similar concept was built for the second converter. The limitation band can be adjusted by changing the tolerance to obtain desirable accuracy.

(30)

Figure 11: Current control for gate 3 and gate 4. Similar function as the previous figure 10

Figure 12: Current limitation of the converter AC-side with a band of toler- ance 0.1 kA. Band of limitation is marked in red, the current reference in

(31)

With these control schematic it is possible to control the output from the converter. In a further step also the DC-link voltage should be controlled to maintain a constant voltage level.

4.2.2 Parameter description and impact of parameter change In this study values on the components of the RPC are chosen similarly to other simulated RPCs and some values are chosen after test simulations to find suitable ones. The main purpose with this thesis is to build a simple version of an RPC. In further studies it could be of interest to investigate the RPC and its components more detailed. This includes investigations regarding losses in the RPC, switching frequency in the IGBTs and DC-link capacitor size.

The switching frequency for the IGBTs in the simulated model is set to 10 kHz. High frequency lead to higher losses since each time the switch closes and opens losses occur. However too slow switching leads to slow respond and might give current and voltages that do not follow the reference values.

Since this control has an additional current control to maintain the current within the band of limitation the switching frequency will be slightly higher than 10 kHz.

The DC-link in the RPC includes a smoothening capacitor. This capac- itor shall reduce the ripple on the voltage waveform. The capacitor capaci- tance in this design was set to 5 µH after test simulations. A large capacitor can reduce the voltage ripple on the waveform but a too large capacitor will lead to high cost. A too small capacitor might lead to a converter design with poor performance.

In further studies the RPC could be optimized to find more suitable values on the components in the design.

(32)

5 Simulations and Results for a Conventional system and a system with the RPC imple- mented

The main goal with this chapter is to show how the conventional railway traction system and a conventional railway traction system with an RPC in operation perform under the same loading conditions. These loading condi- tions are divided into four cases.

All simulations are done with a three-phase grid voltage of 100 kV RMS and the voltages are assumed to be purely sinusoidal with a phase shift of 120 degrees between the phases. The grid is considered strong if nothing else is mentioned, and the frequency is set to 50 Hz.

The voltage after the traction transformer will be 25 kV RMS, either after the conventional system with single-phase transformers or after the system with an RPC and a V/V transformer.

The inductance of the grid represents the inductive behavior of a transmis- sion line. The transmission lines on the secondary side of the transformer are short lines (under 80 km), only a resistive and an inductive part is modeled in this simulation. The train load is simulated as a rectified load with one resistive part and one inductive part. The switching devices are Insulated- Gate Bipolar Transistors (IGBTs) connected with an anti-parallel diode.

Simulations are mainly done with a load of 10 MW. A typical train load can vary from a few MW up to 15 MW [1]. However, the train loads are assumed to increase in the future, and also several of them may be running within the same catenary section.

5.1 Conventional System

The first simulations were done on the conventional system with single-phase transformers. The worst-case scenario occurs when all trains are being fed by the same two phases all along the catenary. Here, the conventional system is simulated with a load connected between two phases. The train will draw current from these two phases, while no current will be drawn from the third phase. The single-phase transformers are assumed to be ideal. The conventional system was simulated in four cases.

(33)

5.1.1 Case 1

The first simulation was done with a strong grid i.e a low inductance of 3 mH, 0.01 p.u compared to the base-inductance.

Sb = 100 M V A Vb = 100 kV Zb = 100 Ω Zb = Rb = Xb Xb = jωL Lb = 0.318 H

Two trains are assumed to draw a constant power of 10 MW each, in each feeding section. The train is assumed to be a rectified load with one resistive part of 62.5 Ω and one inductive part of 10 mH. The load is assumed to be connected at the feeding point. As expected, the voltages are not affected by the load since the grid inductance is set to a low value. The voltage wave- forms can be seen in figure 13. However, the current waveforms are affected severely by the loading conditions. Current is drawn only from two phases through the transformer, and one phase-current will be zero, as shown in figure 14. The two phase-currents will be separated with 180 degrees, in a balanced situation it should be 120 degrees between all the three current waveforms.

(34)

Figure 13: Grid voltages for the conventional system. Each feeding section is loaded with 10 MW

Figure 14: Grid currents for the conventional system. Each feeding section is loaded with 10 MW

(35)

5.1.2 Case 2

In the next simulation the grid inductance remains the same, 3 mH. The load will remain the same as previously but the loads are assumed to be placed 50 km away from the feeding point. The transmission line is modelled as one resistor of 5 Ω and one inductance of 70 mH. The voltage and the current waveforms are not changing any significantly compared to the previously case. The voltage waveforms are still good as shown in figure 15 and the current waveforms in figure 16 are still as bad as before.

Figure 15: Grid voltages for the conventional system. Each feeding section is loaded with 10 MW, 50 km from feeding point

(36)

Figure 16: Grid currents for the conventional system. Each feeding section is loaded with 10 MW, 50 km from feeding point

5.1.3 Case 3

In case three one load of 50 MW is placed on the bc-side of the transformer output. The distance from the feeding point and the load will be 50 km.

The load contains a 12.5 Ω resistor and one inductor of 10 mH. The other section (the ac-side) is modeled as a section with no train connected to the catenary. This is done by connecting a resistor of 62500 Ω. Neither in this case did the voltage or current waveform change any significantly as can be seen in figure 17 and 18.

(37)

Figure 17: Grid voltages for the conventional system. One section loaded with 50 MW and the other section is modeled with no train connected

Figure 18: Grid currents for the conventional system. One section loaded with 50 MW and the other section is modeled with no train connected

(38)

5.1.4 Case 4

In the fourth case the bc-section draws a power of 10 MW and the ac-section initially draws a power of 1 MW. The 1 MW load contains of a load of 625 Ω and an inductor of 10 mH. After a certain time the train that draw a power of 1 MW will accelerate. The load will hence increase to 50 MW instantaneously, and to represent that, the resistor is changed to 12.5 Ω.

In this case, some disturbances can be observed in the voltage waveforms, as in figure 19 and the current waveforms in figure 20. The rectified load draws a high current and as can be seen in figure 20 the rectification will distort the current waveforms. This will also be transferred to the voltage waveforms, where some distortion is observable.

Figure 19: Grid voltages for the conventional system. One section is loaded with 10 MW and the other section is modeled as a train that accelerates at a certain time

(39)

Figure 20: Grid currents for the conventional system. One section is loaded with 10 MW and the other section is modeled as a train that accelerates at a certain time

(40)

5.2 Railway Power conditioner

The same system is now simulated with the addition of a railway power con- ditioner, under the same conditions as the conventional system. The railway power conditioner is placed between the two output phases at the secondary side of a V/V-transformer. Feeding can be done in both directions from the substation. The DC-link capacitor of the RPC has a capacitance of 5 mF.

Two single-phase transformers are connected between the converter and the load-sides. The converter coupling-inductance on each side of the converters is set to 4 mH, 10 percentage of the base-inductance. The RPC was also simulated in four cases, the same as for the conventional system.

Sb = 50 M V A Vb = 25 kV Zb = 12.5 Ω Zb = Rb = Xb Xb = jωL Lb = 0.039 H 5.2.1 Case 1

The first simulation was done with a strong grid, similar as case 1 for the conventional system. Two trains are assumed to draw a constant power of 10 MW each, on each feeding section. The train is assumed to be a rectified load with one resistive part of 62.5 Ω and one inductive part of 10 mH. As can be seen in figure 21, the voltage waveforms have been slightly affected by the switching actions of the RPC system. However, the current waveforms for the grid currents have become much better with this system compared to case 1 for the conventional system, as shown in figure 22.

(41)

Figure 21: Grid voltages for 10 MW load on both sections

Figure 22: Grid currents for 10 MW load on both sections

(42)

5.2.2 Case 2

In the next simulation the grid inductance remains the same at 3 mH. The load will remain the same as previously, but the loads are assumed to be placed 50 away km from the feeding point. The transmission line is modelled as one resistor of 5 Ω and one inductance of 70 mH. The voltage waveforms, shown in figure 23 will not be significantly affected compared to the previ- ous case for the RPC. The current waveforms in figure 24 will still be well balanced, even if the load characteristics is different (contains the 50-km line impedance).

Figure 23: Grid voltages for 10 MW load on both sections placed 50 km from feeding point

(43)

Figure 24: Grid currents for 10 MW load on both sections placed 50 km from feeding point

5.2.3 Case 3

In case 3, one load of 50 MW is placed in the bc-catenary section, close to the feeding point. The load on the ac-section is considered to be very small, modeled as a large resistor of 62500 Ω. The voltage waveforms in figure 25 are not changed significantly during this load condition compared to previous two cases for the RPC. The current waveform is, however, affected by the high load. The rectified load will draw a high current and significant harmonic distortion can be observed in the current waveforms in figure 26.

(44)

Figure 25: Grid voltages for the conventional system. One section loaded with 50 MW and the other section is modeled with no train connected

Figure 26: Grid currents for the conventional system. One section loaded with 50 MW and the other section is modeled with no train connected

(45)

5.2.4 Case 4

In the fourth case the bc-side draws a power of 10 MW the ac-side initially draws a power of 1 MW. After a certain time the train that drew a power of 1 MW will accelerate. The load will hence increase to 50 MW instantaneously.

The voltage waveforms in figure 27 still remained stable. The load step can be observed in the current waveforms in figure 28. The current wave- forms before the acceleration were quite good-quality sinusoidal waveforms.

After the acceleration the waveforms are also of adequate quality, and the balance among the three phases is still kept. However, some distortion can be observed from the high load with rectified characteristics. There is ap- proximately one period until the system reaches the steady state after the acceleration instant.

Figure 27: Grid voltage waveform for train acceleration

(46)

Figure 28: Grid current waveform for train acceleration

(47)

5.3 Unbalance factor

From equation (3) and (4) the unbalance factor can be found. Equation (3) will give the positive sequence current for a phase current and equation (4) will give the negative sequence current for a phase. The unbalance factor is determined as following:

Kunbalance= |Ia2

Ia1| (37)

For the first case, the conventional system, described in 5.1.1. Values for the phase currents are found during simulation.

Ia = 0 −175 A Ib = 3.75 ∗ 103 287 A Ic= 3.75 ∗ 103 108 A

The unbalance factor, Kunbalancefor the conventional system is approximately 98 %.

For the equivalent case in 5.2.1, simulated with the RPC in operation, the phase currents will be as following:

Ia = 171 −1.8 A Ib = 163 237.6 A Ic= 171 116.8 A

The unbalance factor, Kunbalance for the RPC-system is approximately 0.45

%.

(48)

6 Conclusion and Discussion

This study has shown that railway power conditioners (RPCs) can mitigate unbalances in the grid, as they reduce the degree of current unbalance sig- nificantly. In a conventional system, the unbalance factor can be nearly 100

%, when the train is the only load served by the grid. In the system with a V/V-transformer and a connected RPC, the unbalance factor for the same conditions decreased to about 0.45 %. The RPC could hence fulfill its pur- pose.

The conventional system is simple, cheap and there is a large knowledge base in this technology, but it creates poor power quality in the grid. In practice the feeding from the single-phase transformers will be alternating between the phases in the grid. This is done to equalize the average load between the phases. However, the stochastic behavior of the train load will make it impossible to have equal loading on all phases with single-phase transformers. This may disturb other customers in the grid and lower the over-all performance of the power system. The power quality may even get worse with trains running at higher speeds or trains with a heavier load. To obtain a better power quality, compensation must be implemented either on the grid-side or at the railway-side.

The RPC can serve as a power compensator on the railway-side to sym- metrize the grid currents. This device contains converters with switching devices that will create harmonics. To obtain a better performance more studies must be carried out to investigate its output harmonics, the losses and the ratings of an RPC.

(49)

7 Future outlooks

To determine if this is a competitive solution for the railway traction sys- tem, further studies must be performed. The system must be investigated in terms of harmonics and in what way the harmonic distortion from both the converters and the traction equipment on the trains can be reduced.

The RPC-system could also be compared with other compensation systems, such as the STATCOM. The comparison could include both cost and perfor- mance. In this study only the basic operation of the RPC has been inves- tigated. The behavior of this system could be further analyzed under fault conditions. It is also interesting to understand if the system can provide further support to the electrical grid, during the time that no train load is present.

(50)

8 References

[1 ] E.Andersson, M.Berg, Sp˚artrafiksystem och sp˚arfordon, Del 1 Sp˚artrafiksystem, Kapitel 5 Elektrisk t˚agdrift och dess anl¨aggningar, J¨arnv¨agsgruppen

KTH Centrum for forskning och utbildning i j¨arnv¨agsteknik, 4:de up- plagan, 2007

[2 ] L.Abrahamsson, T.Sch¨utte, S. ¨Ostlund, Use of converters for feeding of AC railways for all frequencies, Energy for Sustainable Development, vol 16, issue 3, page 368-378, 2012

[3 ] T.Uzuka, Faster than a speeding bullet - An overview of Japanese high- speed rail technology and electrification, IEEE Electrification Maga- zine,vol 1, issue 1, page 11-20, 2013

[4 ] N.Y.Da,i K.W.Lao, M.C.Wong, C.K. Wong, Hybrid power quality con- ditioner for co-phase power supply system in electrified railway, IET Power Electronics, Vol 5, Issue 7, page 1084-1094, 2012

[5 ] Copenhagen Ringsted Team, Preliminary design of neutral section - The New Line Copenhagen Ringsted Alignment and Railway Technology, page 1-19, 2012

[6 ] T.Bagnall, F.Silizar, Conference on railway excellence 2014 Adelaide, Power electronics based traction power supply for 50 Hz, 2014

[7 ] Q.Liu, Xr. Li, Xj. Li, B.Lei, D.Chen, Y.Pan, Impact of negative se- quence current of traction load on the grid running state, Advanced Power System Automation and Protection, International Conference, vol 2 , page 879-883, 2011

[8 ] J.Wang, R. Hamiltion, A review of negative sequence current, Protective Relay Engineers, 63rd Annual Conference, page 1-18, 2012

[9 ] J.J.Grainger, W.D Stevenson, Power System analysis, page 416-420, 1994

[10 ] J. Duncan Glover, M.S.Sarma, T.J.Overbye, Power system analysis and design, 5th edtion, page 428-431, 2012

[11 ] C.Zhao, Power supply for 50 Hz railway Systems: System simulation specification, unpublished ABB, 2014

(51)

[12 ] L.Xiaqing, Z.Li, Research on balance compensation of STATCOM, Beijing Institute, 2nd IEEE conference on industrial electronics and applications, page 563-568, 2007

[13 ] T.Uzuka, S.Ilkedo, Railway static conditioner field test, Quarterly re- port of RTRI, vol 45, no2, 2004

[14 ] N.Y.Dai, K.W. Lao, M.C.Wong, A hybrid railway power conditioner for traction power supply system, Applied Power Electronics Conference and Exposition, 28th Annual IEEE, page 1326-1331, 2013

[15 ] Z.Shu, S.Xie, K.Lu, Y.Zhao, X.Nan, D.Qiu, F.Zhou, S.Gao, Q.Li, Dig- ital detection, control, and distribution system for co-phase traction power supply application, IEEE Transactions on Power Electronics, vol 60, issue 5, page 1831-1839, 2011

[16 ] K.W.Lao, M.C.Wong, N.Y.Dai, C.K Wong, C.S Lam, A systematic ap- proach to hybrid railway power conditioner design with harmonic com- pensation for high-speed railway, IEEE Transactions on Power Elec- tronics, vol 62, issue 2, page 930-942, 2011

[17 ] A.Luo, C.Wu, J.Shen, Z.Shuai, F.Ma, Railway static power condition- ers for high-speed train traction power supply systems using three- phase V/V transformers, IEEE Transactions on Power Electronics, vol 26, issue 10, page 2844-2856, 2011

References

Related documents

What is more, the energy conversion efficiency by hydro turbine is exceedingly higher than the thermal turbine: the efficiency for hydro turbine is generally over 75%, and

In the specific case of shunt-connected FACTS controllers, such as a Static Synchronous Compensator (STATCOM) and Static Var Compensator (SVC), Transient Stabi- lity Enhancement

Our study has shown that optimizing the de-icing schedule at Stockholm Arlanda airport, while taking the total airport performance into consideration, enables the

Title: Environmental Product Declaration for power, signalling and telecom systems on the Bothnia Line3. Publication number: 2016:152 Date of publication:

As regards concrete and steel reinforcement quantities, fuel and electricity consumption, etc., specific data was collected from these contracts and used as a basis for

However, note that substructure (tunnels, bridges, track foundations) and power, signalling and telecom systems are not included.. The figure shows some of the most

The contribution from railway infrastructure to the total environmental impact for railway transport services is presented here in a transport scenario.. 10

The various uncertainties discussed in this paper makes that harmonic studies are needed before the connection of wind parks, and that these studies should be broader than