ES2415238B1 - SYSTEM AND METHOD OF LOSS REDUCTION IN PARALLEL TRANSFORMER SYSTEMS IN DISTRIBUTION CENTERS - Google Patents

Abstract

System and method of loss reduction in parallel transformer systems in distribution centers. The method comprises: # - measuring the total load power S {sub, L} of the parallel transformers (10, 10 ’) of the distribution center; # - comparing said load value S {sub, L} with a value of optimization power (5) obtained based on the total transformation power S {sub, N} of the transformers in parallel (10, 10 ’); # - select a specific transformer connection configuration (10, 10’) depending on this comparison; # - change to the selected connection configuration, in case it does not match the active connection configuration. # Applicable for centers where the use of electrical energy is highly variable (industrial estates, shopping centers , teaching centers, seasonal tourist centers, etc.).

Description

System and method of loss reduction in transformer systems in

parallel in distribution centers

Field of the Invention

5

The present invention is included within the field of centers of

transformation, and more specifically, in systems and methods of reducing

Losses in the voltage transformation carried out in these centers of

transformation.

1 o

Background of the invention

Transformation systems installed in connected buildings and industries

in distribution networks they must guarantee continuity of supply at the point of

Consumption in the most efficient way possible. The different international regulations

they demand this guarantee of supply by installing two or more

fifteen

parallel power transformers that allow operations of

maintenance in one of them without cutting off the supply, as well as guaranteeing the

continuity of operation in case of failure. Although the efficiency

power of distribution transformers has increased and the

total losses, sum of the losses in load and in empty, these still represent

twenty

an important value in electrical systems, between 16% and 40% of losses

energy companies associated with electrical distribution systems. It is therefore done

it is necessary to reduce the losses in the transformation systems and therefore the

emission of associated greenhouse gases.

Transformation systems account for a significant percentage of

25

losses of the entire electricity distribution in both Europe and States

United [1]. Although transformer manufacturers [2] have reduced form

Systematic losses by introducing new materials and manufacturing techniques (3-

6], there is a significant number of equipment already installed that are within

their useful life and have much lower energy efficiency values than

30

present the systems manufactured today. Modern manufacturing techniques

of transformers allow to obtain systems with practically null losses what

which implies the reduction of associated greenhouse gas emissions

[4,5]. The transformation systems with low losses allow to reduce the power

snapshot demanded by the installation and therefore reduce the power of

35

associated generation. In the case of fossil fuel based supply this

it implies a reduction of tons of equivalent C ~ and in the case of systems in distributed generation networks with renewable generation it implies a reduction of the installed power and facilitates a better forecast of the demand and control of the system.

The necessary investment in power distribution systems and the cost of electricity generation continue to increase, so that technologies that reduce energy consumption are in high demand today. Electricity supply companies and the final consumer in private facilities are directly benefited by the implementation of loss reduction systems in distribution transformers. The transformation systems are in mandatory operation whenever there is demand so that the energy savings associated with new technologies or transformation systems reach high values. In addition, the transformers installed operating in a vacuum have associated losses, so there is energy consumption without demand in the secondary. In a design phase of a new installation the decision between the use of a high efficiency system and one of greater losses is directly related to the expected economic savings during the equipment use phase. The TOC analysis (Total Owning Cost, total cost of ownership analysis) is used as a decision tool since it contemplates the sum of the cost of the transformer itself and the costs derived from losses in the equipment during its useful life [7). There are examples of the application of this method for the evaluation of facilities [8,9] depending on the type of use [10-12] and particularizing for industrial and commercial systems [13, 14]. The TOC calculation evaluates the losses of the transformer in load and in vacuum to determine which transformer is the most suitable depending on the cost of the losses during the useful life of the equipment. It is an ideal tool for decision-making during the design phase [14,15], manufacturing and purchase of equipment.

In the case of existing installations that have transformers in operation, it is necessary to propose techniques that allow reducing the energy losses associated with the equipment without having to replace them. The replacement of equipment that is not damaged is justifiable provided that when performing a TOC analysis the profitability of disposing of the existing system and the installation of a new transformer is verified. This substitution is not economically profitable in almost any case so it is necessary to propose savings alternatives. In the case of new installations the choice of transformers of higher efficiency class can be profitable when the reduction of energy losses justifies the

extra cost compared to a conventional transformer. Existing transformers

they must also be integrated into the new distributed generation systems so

that the proposal of loss reduction methods will facilitate this task. In the

industrial and commercial facilities transformation centers, usually

5

the transformers are kept connected whatever the power demand

snapshot. There are many facilities that have hourly, daily seasonality

and monthly usage, so the installed transformation power is not always

fits the optimum. In addition, for reasons of anticipating expansion of the installation,

It is usual to oversize the transformation systems by a minimum of 20%.

1O

Transformers are static electric machines with a high

performance that reaches values higher than 95%. It is based on the use of two or

more windings around a ferromagnetic core, and in a change of one level

from tension to another without a change in frequency. The transformers are

used massively in electrical distribution systems and perform

fifteen

basic functions as they act as isolation elements and transformers of

tension level The use of ferromagnetic materials and copper makes its

cost is high and they are also heavy and bulky machines. The means to

reduce the size of the machines is that the power density in the transformer

it is inversely proportional to the frequency so systems that work with

twenty

high frequencies allow more efficient use of the magnetic core and

therefore reduce the size of the equipment [16]. They are currently being developed

multiple investigations in electronic based transformation systems of

power. These systems, called OEPT (Distribution Electronic Power) systems

Transformef), allow to deal more effectively with the quality problems of the

25

energy in distribution systems and reduce losses [17-24]. The systems

PTSD are under development but have not been introduced into the systems

of electrical distribution.

The present invention focuses on the systems currently in existence and

that are installed in the distribution systems and that will be followed

30

installing in the coming years. According to studies carried out in the United States

Transformation system losses account for 40% of losses in

non-generating public service companies and 16% in service companies

private [1]. In the European Union, the European Copper Institute estimates that the

replacement of power transformers installed in Europe with another 40%

35

  more efficient would reduce energy consumption by more than 22 TWh which

would lead to a decrease in greenhouse gas emissions of more than 9

million tons of e ~ equivalents [25]. Due to the high cost of

equipment this solution cannot be carried out systematically so it

raises a method of reducing losses in transformation systems

5

regardless of its efficiency [26-29]. In June of the year 2011 the lntemational

Energy Agency (lEA) has made public that during 2010 and despite the crisis

economic world, the record of effect gas emissions has been reached

inventory [30]. This implies that the consequences of climate change are every

increasingly unavoidable and all possible strategies to reduce

1st

these emission levels. The present invention presents a low methodology

cost applicable to parallel transformation systems and study the potential

concrete in several real situations. Among these real situations, the

high potential for economic and energy savings in facilities that present

a temporary, hourly or seasonal operation. For example in polygons

fifteen

industrial, shopping centers, educational centers, etc., the operation occurs in

certain times of the day, in tourist centers the use is very seasonal and there are cases

combined, as teaching centers. In all these cases the power demanded in

Transformers vary substantially over time.

The present invention proposes a new method called PLO (Para / le / Losses

twenty

Optimization) and aimed at minimizing losses in the system during the

exploitation of it. The method allows adapting the operating system of the

transformation center to achieve the lowest possible losses for power

sued In addition, the proposed PLO method reduces losses throughout

type of parallel transformer installations being applicable for systems

25

existing or new allowing to increase energy savings whatever the

transformer efficiency class.

BIBLIOGRAPHIC REFERENCES

[1) Kennedy BW. Energy efficient transformers. New York: McGraw-Hill; 1998

30

[2] Hasegawa R, Azuma D. lmpacts of amorphous metal-based transformers on energy

efficiency and environment. J Magn Magn Mater2008; 320: 2451-6.

[3) Olivares JC, Yilu L, Canedo JM, Escarela-Perez R, Driesen J, Moreno P. Reducing

losses in distribution transformers. IEEE Trans Power Deliv 2003; 18: 821-6.

[4] Hasegawa R, Azuma D. lmpacts of amorphous metal-based transformers on energy

35

  efficiency and environment. J Magn Magn Mater 2008; 320: 2451-6.

[5] Georgilakis PS. Differential evolution solution to transform no-load loss reduction problem. IET Gener, Trans Distrib 2009; 3: 960-9. (6] Olivares-Galván JC, Georgilakis PS, Ocon-Valdez R. A review of transforming losses. Electr Power Compon Syst 2009; 37: 1046-62. (7] Nochumson CJ. Considerations in appllcation and selection of unit substation transformers. IEEE Trans lnd Appl2002; 38: 778-87.

(8) Rasmusson PR. Transform economic evaluation. IEEE Trans lnd Appl 1984; 20: 355-63.

[9] Bins DF, Crompton AB, Jaberansari A. Economic design of a 50 kV A distribution transform. Part 2: effect of different core steels and loss capitalisations. IEE Proc Gener, Trans, Distrib 1986; 133: 451--6.

[10] ANSI / IEEE. IEEE loss evaluation guide for power transformers and reactors. ANSI / IEEE Standard C57.120-1991; 1992.

[11] Nickel DL, Braunstein HR. Distribution transform loss evaluation: 1-proposed techniques. IEEE Trans Power Appar Syst 1981; 100: 788-97.

[12] Nickel DL, Braunstein HR. Distribution transform loss evatuation: 11 -load characteristics and system cost parameters. IEEE Trans Power Appar Syst 1981; 100: 798-811.

[13] Merritt S, Chaitkin S. No load versus load loss. IEEE lnd Appl Mag 2003; 9: 21--8.

[14] Georgilakis PS. Decision support system for evaluating transform investments in the industrial sector. J Mater Process Technol2007; 181: 307-12. [15) Baranowski JF, Hopkinson PJ. An altemative evatuation of distribution transformers to achieve the lowest TOC. IEEE Trans Power Deliv 1992; 7: 614-9.

[16] M. Kang, P.N. Enjeti, I.J. Pite !, Analysis and design of electronic transformers for electric power distribution system, IEEE Trans. Power Electr. 14 (1999) 1133-1141.

[17] W. McMurray, Power converter circuits having a high frequency link, US Patent 3,517,300, June 23, 1970.

[18] G. Venkataramanan, B.K. Johnson, A. Sundaram, An ac-ac power converter for custom power applications, IEEE Trans. Power Delivery 11 (1996) 1666-1671.

[19] J.L. Brooks, Solid State Transform Concept Development, Naval Material Command, Civil Engineering Laboratory, Naval Construction Battalion Ctr., Port Hueneme, CA, 1980.

[20] EPRI Report, Proof of the principie of the solid-state transform: the AC / AC switch mode regulator, EPRI TR-105067, Research Project 8001-13, Final Report, August 1995.

[twenty-one]

K. Harada, F. Anan, K. Yamasaki, M. Jinno, Y. Kawata, T. Nakashima, K. Murata,

H.

Sakamoto, lntelligent transform, in: Proceedings of the 1996 IEEE PESC Conference, 1996, pp. 1337-1341.

[22] M.O. Manjrekar, R. Kiefemdorf, G. Venkataramanan, Power electronic transformers for utility applications, in: Proceedings of the 2000 IEEEIAS Annual Meeting, 2000, pp. 2496-2502.

[23] E.R. Ronan, S.D. Sudhoff, S.F. Glover, D.L. Galloway, Apower electronicbased distribution transform, IEEE Trans. Power Delivery 17 (2002) 537-543.

[24] M. Marchesoni, R. Novaro, S. Savio, AC locomotive conversion systems without heavy transformers: is it a practicable solution, in: Proceedings of the 2002 IEEE International Symposium on Industrial Electronics, 2002, pp. 1172-1177.

[25] Targosz R. The potential for global energy savings from high energy efficiency distribution transformers. Leonardo Energy, 2005. <http://www.leonardoenergy.org/repository/Library/Reportsrrransformers-Giobal.pdf>. [26) Zhang Y-J, Wei Y-M. An overview of current research on EU ETS: evidence from its operating mechanism and economic effect. Appl Energy 2010; 87: 1804-14. [27) Chen WT, Li YP, Huang GH, Chen X, Li YF. A two-stage inexact-stochastic programming model for planning carbon dioxide emission trading under uncertainty. Appl Energy 2010; 87: 1033-47.

[28] Vad Mathiesen 8, Lund H, Karlsson K. 100% renewable energy systems, climate mitigation and economic growth. Appt Energy 2011; 88: 488-501.

[29] lrrek W, Topalis F, Targosz R, Rialhe A, Frau J. Policies and measures fostering energy-efficient distribution transformers. Report of European Commission Project No EIEI05 / 056 / SI2.419632; June 2008. [30) C02 emissions reach a record high in 2010; 80% of projected 2020 emissions from the power sector are already locked in. lntemational Energy Agency, 2011. <http://www.iea.org> [31) CENELEC. Three-phase oil-immersed distribution transformers 50 Hz, from 50 kVA to 2500 kVA with highest voltage for equipment not exceeding 36 kV -Part 1: General requirements. CENELEC Standard EN 50464-1; 2007

[32] FLUKE 430 Series Three-Phase Power Quality Analyzers, 2011. <http: //www.fluke.cornlfluke/usen/power-quality-toolslthree-phase/fluke-430-series.html>

Description of the invention

The present invention proposes a system and a method of optimizing transformer systems in parallel that solves the above problems, applicable to existing or future installations and adaptable to any type of transformer systems.

The optimization method is applied for transformation systems based on two transformers in parallel, a usual solution both industrially and commercially. The method calculates and characterizes the total losses in the transformation center for one, two or both connected transformers. For each transforming power one optimization power or point is established, here also called PLO (Paralell Losses Optimization) power, which is the breaking point from which it is more profitable to have one or the other of the transformers connected or both in parallel .

The method of reducing losses in parallel transformer systems in distribution centers comprises repeatedly performing the following steps:

-measure the total load power SL of the transformers in parallel of the distribution center;

- comparing said load value SL with at least one optimization power value obtained based on the total transformation power SN of the transformers in parallel;

-select a specific transformer connection configuration based on said comparison; -change to the selected connection settings! in case it does not match the active connection settings.

Preferably the transformation center has only two transformers connected in parallel. In that case the load value SL is compared with a single optimization power value.

In a preferred embodiment, the optimization power value depends linearly on the total transforming power SN of the transformers in parallel.

The transformer connection configuration can be determined as follows:

-if the load value SL is lower than the optimization power value, the connection configuration consists of activating a single transformer, which is the one that for that load value SL has lower losses;

-if the load value SL is greater than the optimization power value, the

Connection configuration consists of activating both transformers in parallel.

The object of the invention is also a loss reduction system in

parallel transformer systems in distribution centers, comprising:

5

-means of measurement of the total load power ~ of the transformers in

parallel distribution center;

-control means configured for:

• compare said load value SL with at least one power value

of optimization obtained based on the total transformation power SN

1 o

of transformers in parallel;

• select a transformer connection configuration

determined based on said comparison;

• switch to the selected connection settings, using the

activation of connection means (which can be part of the

fifteen

system), in case it does not match the connection settings in

active.

The control means may comprise a memory where the

At least one optimization power value.

The system may also comprise a remote monitoring unit with

twenty

Remote management functions of the control media configuration.

In the event that the transformation center has two

transformers connected in parallel, the control means may be

configured to compare the load value ~ with a single power value of

optimization The optimization power value can depend linearly on the

25

Total transforming power SN of the transformers in parallel.

The control means may be configured to determine the

transformer connection configuration as follows:

-if the load value SL is lower than the optimization power value, the

connection configuration consists of activating a single transformer, which is what

30

for this cargo value SL has lower losses;

-if the load value SL is greater than the optimization power value, the

Connection configuration consists of activating both transformers in parallel.

Brief description of the drawings

Then a series of drawings is described very briefly

that help to better understand the invention and that expressly relate to

an embodiment of said invention presented as a non-limiting example of

is.

Figure 1 shows, in a parallel transformer system, an example

5

of load / loss ratio curves of the different transformers when

They work separately and both in parallel.

Figure 2 shows in detail the optimum power point obtained for the

previous figure.

Figure 3 shows the correspondence between optimal power points and the

1 o

nominal transformation power used.

Figure 4 represents the system object of the invention.

Detailed description of the invention

Transformers have associated losses that are classified as:

fifteen

-Value losses: In a transformer connected to a system of

electrical distribution there are losses whenever the transformer is

under a voltage. These losses do not depend on whether or not the transformer is

subjected to a load so they are called empty losses. Its value is

constant and consists of five differentiated components:

twenty

• Hysteresis losses in central blades and losses due to currents

Foucault in the main lamination (99%)

• Dielectric losses (<0.50%)

• Losses due to free eddy currents and clamps

main, segments and components (<0.35%)

25

• 12R losses associated with idle current (<0.15%)

Because Foucalt's hysteresis and current losses represent

99% of total losses are usually assumed as the only ones

into account, while the others are assumed to be approximately equal to zero.

- Losses in load: As for the losses in load, as the

30

load in the transformer the losses in the equipment are modified since these are

due to heat losses in conductors caused by the load current

and by the currents of Foucault. These losses vary with temperature as the

conductivity characteristics of the materials and the most important value is the

  due to losses in copper characterized by 12R.

To ensure continuity of supply,

in extension forecast,

to facilitate maintenance and regulatory issues are very common

installations

formed by two transformers in parallel. Sayings systems

Multitransformer in parallel also have different losses. Can be presented

5

Three operation cases:

1. The demanded operating power is less than the nominal power

of the first transformer.

2. The demanded operating power is less than the nominal power

of the second transformer.

1 o

3. The operating power demand is greater than the nominal power of

Either of the two transformers.

These loading situations will be presented regularly in the same

installation depending on the load curve of the installation but the transformers

they remain connected identically regardless of the load.

fifteen

Taking into account the types of losses set forth above, in one case

from

a system formed by a transformer with a power nominal SN

characterized

by their losses in empty (the) Y its percentage impedance from

Short circuit (Use) total losses when subjected to a load power

Apparent SL will be calculated according to:

twenty

4 = 4 + (SL) 2 Use SN (1)

In

he case from two transformers connected in parallel with the

Test characteristics of percentage short circuit impedance, Use, the load is

will distribute between both transformers so that the lower the value of Use

The higher the short circuit currents.

25

The total load power, SL, will be distributed between transformers 1 and 2 of

according to Equations (2) and (3):

s

-sL.sN.I.usc.r

L, l - (sN, I + sN.J. USC, I (2)

s

_ sL.sN.2.usc.r

L, l- (sN, I + sN.2J · usc.2 <3>

The value Usc, T represents the average short circuit impedance determined according to Equation (4):

So each transformer can be characterized by its distribution

Load percentage PL1, PL2 as shown in Equations (5) and (6):

p ~ = SL, 1 SN, 1 (5)

PL2 = SL, 2

5

SN, 2 (6) The losses therefore in the system running in parallel will be calculated according to Equation (7):

SL, 1 2 (SL, 2) 2

Lr nnr = Lo 1 + -Use 1 + Lo 2 + --U se 2 (7)

• r- • (S),. S.

N, 1 N, 2

In the method proposed by the present invention, the loss characteristics of each of the transformers and of both of them operating in parallel for each point of the load curve, corresponding to the sum of the nominal powers, Equation (8) ):

SN = SN, 1 + SN, 2 (8) 15 Depending on the load and using the loss calculation formulas in Equations (1) and (7), determine the point or power of loss optimization in parallel (hereinafter optimization power or PLO power). The PLO power (Para / el / Losses Optimization) has been called the one at which from which the losses in the system make it advisable to connect both transformers even if the 100o / o load in the first one (SN, one). This strategy minimizes losses by establishing a connection strategy based on the minimum operating losses and not on the nominal powers of the equipment. 25 For the combination of transformers that make up the transformation center, losses at different load levels (St.) are simulated, operating with each of the transformers and with their operation in parallel

obtaining a load-loss curve like the one in Figure 1, where it is shown

as an example the result for a transformation power of 41 O kVA

(160 + 250 kVA EoDk efficiency class transformers).

Figure 1 therefore shows the loss curve of the first

5

transformer 1 running in isolation, the loss curve of the second

transformer 2 with the same operating strategy, the loss curve of the

parallel system 3 and the minimum loss curve 4 (i.e. the optimum). The

optimization point or power 5 (point or PLO power) as that power of

load from which the losses decrease when connecting both

1O

transformers, detail seen in Figure 2.

In zone A, to the left of optimization power 5, it will be optimal

work with a single transformer system connection (whichever is more

effective, the one with the least losses, in this case the first transformer), while

that in zone 8, to the right of optimization power 5, it will be optimal to work

fifteen

with a parallel connection of both systems. Thus, the curve of minimum losses 4

will coincide in zone A with the loss curve of the first transformer 1 and in the

zone B with the parallel loss curve of the system 3.

In short, the connection that is most energy efficient in choosing

SL load power function. In zone A the transformer is connected with

twenty

minor losses Depending on the characteristics of both transformers the

optimization point varies, so what you want is to connect the

lower loss transformer for that power SL, not being this one

necessarily the maximum rated power.

In a study conducted in different transformation centers, they have

25

combined transformers of the characteristics set forth in Table 1, to give

place to thirteen installation powers with three degrees of energy efficiency (EoDk,

DoCk, CoBk) so that a total of 39 case studies have been generated, according to the

Table 2. The analyzed powers correspond to typical cases of powers.

installed in a wide range of applications. All transformer data

30

They correspond to real teams.

~ (kVA)

ENSQ464..1 classification Use (%) PorN)

100

EoO ..- High 4 320

160

EoO ..- High 4 460

250

EoDk-High 4 650

400

EoOt-Hlgh 4 930

630

EoDk-High 4 1,200

1,000

EoDt-High 5 1,700

1,600

EoDt-High 6 2,600

100

DoCtc -Medium 4 260

160

DoCtc -Medium 4 375

250

DoCtc -Medium 4 530

400

CoS ..- Medium 4 750

630

DoCtc -Medium 4 940

1,000

DoCk-Medium 5 1,400

1,600

DoCtc -Medium 6 2,200

100

CoBk-Low 4 210

160

CoBk-LOW 4 300

250

CoS ..- Low 4 425

400

CoBk-Low 4 610

630

CoBk-Low 4 800

1,000

CoS ..- Low 5 1,100

1,600

CoS ..- Low 6 1,700

Table 1. Power transformers studied

~ (kVA)

Parallel transformers EN50464-1 classification

200

2x 100 EoDk, DoCtc. CoBt

260

100 + 160 EoO ... DoCtc, CoS ..

320

2x 160 EoDk DoCtc, CoS ..

410

160+ 250 EoDt. DoCk, CoS ..

500

2x250 EoDk DoCk, CoS ..

650

250+ 400 EoDt. DoCtc CoS ..

800

2x400 EoDt. DoCk, CoS ..

1,030

400 + 630 EoDt. DoCk, CoS ..

1,260

2x630 EoDt. DoCtc, CoS ..

1,630

630+ 1,000 EoDt. DoCtc, CoS ..

2,000

2x1,000 EoDk, DoCk, CoS ..

2,600

1000 + 1,600 EoDt. DoCtc, CoS ..

3,200

2x 1,600 EoDk, DoCtc, CoS ..

Table 2. Transformation facilities studied

For each case, the optimization point or power (PLO) has been calculated and the values indicated in Table 3 have been obtained. It is observed that as a general rule as the efficiency of the set of transformers increases the optimization power It increases although the values are very similar for all efficiency classes. The effect of increasing the efficiency class is a decrease in losses in any operating condition.

~ (kVA)

Half

200

56 55 54 55

260

86 82 79 82

320

89 92.8 89 90

410

128 134 129 130

500

142.5 145 140 143

650

217 220 212 216

800

228 232 229 230

1,030

332 340 340 337

1,260

333 340 340 338

1,630

491 535 522 516

2,000

520 530 500 517

2,600

820 860 800 833

Table 3. Optimization power results

10 Graphically representing the values obtained and performing a statistical treatment of them, a correlation is obtained that allows the calculation of the optimization power value 5 for a transforming power of the SN installation, Figure 3.

15 For each efficiency level (Eo {} ¡¡, DoCk. CoBk), the value that reaches the optimization power point 5 as well as the average value of the optimization power point for each transformation power SN are detailed. From these results a new method of optimization of transformers in parallel is deduced. Performing a statistical analysis to validate the correlation, it is obtained that

20 optimization power values (PLO) have an adequate relationship with the installed transformation power, according to Equation (9): PLO = f (SN) (9)

Specifically, it is observed that it presents an adequate linear correlation and a method of calculation of the optimization power (PLO) is obtained based simply on the installed transformation power SN, according to Equation (10}, with a e 9t, b e 9t:

PW = a · SN + b (10} For the specific case of study, an adjustment level R2 = 0.9771 is obtained and with a = 0.012 and b = -2.0583, according to the following

equation: PL0 = 0.012 · SN -2.0583

It is important to note that the a and b values would vary depending on the study performed, and any other values considered based on the results of other studies and / or tests may be used. The method proposed here allows, by applying a direct linear relationship with the installed transformation power, to implement the system in any existing or future installation regardless of its characteristics. The simplicity of the algorithm allows to program it in a simple programmable automaton that will govern the connection / disconnection systems of the transformers in parallel.

Due to the need to adapt the electrical measurement systems to the requirements of intelligent electricity distribution networks ("Smart Grid" systems), changes are being made to the electric meters of all power distribution systems in the US. and in Europe with a maximum time horizon of 2020. Intelligent measurement systems (smart meters, "Smart Metering") have the ability to program measurement outputs and in some cases can perform logical calculations with a loss reduction system. Like the one shown in Figure 4, it is possible to execute the proposed method with a practically zero cost by providing returns on the investment in a very low period.

The loss reduction system, applied to parallel transformers (10,10 '), comprises measuring means 12 (eg a smart meter or counter) of the total load power SL of the parallel transformers (10,10'} and control means11 (eg a control unit, a controller) responsible for configuring the connection and / or disconnection of the transformers in parallel (10.10 '), through connection means (13.13'}, depending on the total load power and optimization power points 5 previously calculated, it is therefore configured which transformer (s) are connected (a single transformer or the two transformers

in parallel). The optimization power points 5 could be obtained, as explained, simply in function -e.g. a linear relationship of the value of the transforming power of the SN installation, which corresponds to the sum of the nominal powers (SN, 1t SN.2) of the transformers (10, 10 '). Said optimization power points 5, which in the case of two transformers is a single point as shown in Figure 2, may be stored in a memory 16 (memory within the control means 11 or the control means measure 12, or an external memory accessible by any of said means). The optimization power points 5 will normally be previously calculated and stored in the memory 16, which could be updated remotely through a remote monitoring unit 14 of the smart grid 15.

The remote management capability of the smart meter also allows, through PLC ("Power Line Communication") protocol or any other communication system the supervision, control, update and monitoring of the system, being able to update the control form and the values of the points of optimization power 5 through a monitoring unit 14.

The proposed method is therefore validated for transformation facilities of three efficiency levels, which allows it to be implemented in any transformation installation. The new low loss transformer systems have lower losses but a higher economic cost, so their installation is profitable if the overall cost during the life cycle is lower. In existing transformers direct replacement is not profitable in almost any case. The proposed method reduces your losses throughout its useful life in new or existing installations. The proposed system does not require for its implementation of any additional maneuvering device and allows obtaining percentage savings of up to 41% compared to the initial losses.

The new energy measurement systems in smart electricity distribution networks facilitate installation and operation with this method. In installations with non-continuous hourly operation, the percentage of savings can be much higher since during the night hours and weekends the transformers are still connected in parallel, thereby increasing losses. Connecting and disconnecting them according to the optimization power point and current consumption reduces those losses. In installations with seasonal operation, annual losses can be reduced significantly by performing the proposed optimization.

Finally it is necessary to indicate that the present invention could work with three or more transformers. In the case of using three transformers there would be two optimization zones, although the commercial case of transformer centers with three transformers is extremely rare (the third one is very much in reserve). In

If there were three transformers, one would act in the following way: the most efficient of the three would always be connected as a base and the invention would be applied to select which of the remaining two would be optimal to connect for each power demand.

10

Claims (8)

one. Loss reduction method in parallel transformer systems in distribution centers, characterized in that it comprises repeatedly performing the following steps: CLAIMS 1. -measure the total load power SL of the parallel transformers (10,10 ') of the distribution center; -select a transformer connection configuration (10.10 ') determined according to said comparison; -change to the selected connection configuration, in case it does not match the active connection configuration. Method according to claim 1, wherein the transformer center has two transformers (10, 10 ') connected in parallel, characterized in that the load value SL is compared with a single optimization power value (5). Method according to claim 2, characterized in that the optimization power value (5) depends linearly on the total transforming power SN of the parallel transformers (10,10 '). Method according to claim 2 or 3, characterized in that the transformer connection configuration (10,10 ') is determined as follows: -if the load value SL is greater than the optimization power value (5), the connection configuration consists of activating both transformers in parallel (10, 10 '). 5.-compare said load value SL of the total transforming power SN the parallel transformers (10,10 '); (5). The load value SL is greater than the optimization power value (5), the connection configuration consists of activating both transformers in parallel (10, 10 '). Loss reduction system in transformer systems in 3. Method according to claim 2, characterized in that the optimization power value (5) depends linearly on the total transforming power SN of the parallel transformers (10.10 '). Method according to claim 2 or 3, characterized in that the transformer connection configuration (10,10 ') is determined as follows: Loss reduction system in transformer systems in -comparing said SL load value of of the parallel transformers (10.10 '); (5). The load value SL is greater than the optimization power value (5), the connection configuration consists of activating both transformers in parallel (10, 10 '). of transformers in one. Loss reduction method in parallel transformer systems in distribution centers, characterized in that it comprises repeatedly performing the following steps: -measure the total load power SL of the parallel transformers (10.10 ') of the distribution center; - comparing said load value SL with at least one optimization power value (5) obtained as a function of the total transformation power SN of the parallel transformers (10.10 '); -select a transformer connection configuration (10.10 ') determined according to said comparison; -change to the selected connection configuration, in case it does not match the active connection configuration. 2. 3. Method according to claim 2, characterized in that the optimization power value (5) depends linearly on the total transforming power SN of the parallel transformers (10,10 '). one. Loss reduction method in parallel transformer systems in distribution centers, characterized in that it comprises repeatedly performing the following steps: -measure the total load power SL of the parallel transformers (10.10 ') of the distribution center; -select a transformer connection configuration (10.10 ') determined according to said comparison; -change to the selected connection configuration, in case it does not match the active connection configuration. 2. 3. Method according to claim 2, characterized in that the optimization power value (5) depends linearly on the total transforming power SN of the parallel transformers (10.10 '). Four. Method according to claim 2 or 3, characterized in that the transformer connection configuration (10,10 ') is determined as follows: -if the load value SL is lower than the optimization power value (5), the connection configuration consists of activating a single transformer, which is the one that for that load value SL has lower losses; 1. -measure the total load power SL of the transformers in parallel (10.10 ') of the distribution center; -select a transformer connection configuration (10.10 ') determined according to said comparison; -change to the selected connection configuration, in case it does not match the active connection configuration. 2.