Content uploaded by Lefeng Cheng
Author content
All content in this area was uploaded by Lefeng Cheng on Aug 08, 2017
Content may be subject to copyright.
2017 3rd Asian Pacific Conference on Energy, Environment and Sustainable Development (APEESD 2017)
ISBN: 978-1-60595-435-6
1 INTRODUCTION
Energy is an important material basis for sustainable
development of national economy, improvements of
people’s living standard and promotions of social
progress (C. Y. Xia 2007, H. R. Xue et al. 2015).
Electricity occupies an important position in national
total energy amount, and its energy loss in total
energy losses shouldn’t be neglected. As we know,
transformers are main power equipment in a power
system, which play a significant role in generation,
transportation and scheduling process of electrical
energy, and its operation efficiencies directly affect
costs and benefits of the whole power system (K. M.
Wang et al. 2015, K. Alimjan et al. 2014).
In distribution network, requirements of
increasing distribution transformer layouts produce a
large number of distribution transformers, whose
capacities are far beyond that of generators. In
addition, power loss generated by transformers can
be considerable due to its large transportation energy
and long operating time. According to the statistics,
total power loss generated by the transformer can be
about 10% of total generation capacity, during 3~5
times voltage transformation experimenting from
electricity generation to electricity utilization (Z. B.
Li 2013). Therefore, studies of transformer energy
conservation are very urgent and necessary.
Currently, in aspect of transformer energy
conservation, domestic and foreign scholars have
carried out relevant researches. Among them,
optimization of transformer materials and
materials with low resistance and good conductivity
have been more and more widely used in transformer
manufacturing, which reduces transformer loss
greatly (N. F. Xu 1990, K. Zhao & L. Y. Zhang 2006,
Z. G. Zhang et al. 2008).
For domestic and foreign researches on
transformer energy conservation and reconstruction,
L. Yao et al. (2007) proposed a transformer structure
optimization method that the core structure was
changed from original direct seam to semi-straight
and semi-oblique and complete oblique joint, or
adjusted adequate the ratio of silicon steel sheet and
magnetic wire, which can increase the no-load loss
and load loss in large amplitude. In addition, Z. S.
Wan (2006) made a deep study on a three-phase roll-
core transformer with triangular structure of circle
section, as well as a further optimization of
transformer core structure, so that maximized the
excellent performance of materials. A large amount
of theoretical studies have overturned the
recognition that the conventional transformers
should have a higher load rate, based on which, J. S.
Hu et al. (1999) proposed a concept of optimal load
rate of transformer.
Transformer Energy Conservation and Reconstruction Methods and
Technologies in Low-voltage Area of Distribution Network
Yisheng Wu1, Lefeng Cheng2, Mingjie Cai1, Shunhao Li1 and Zhengjia Li2
1Panyu Power Supply Bureau of Guangzhou Power Supply Bureau Co., Ltd, Guangzhou 510000, China
2Suzhou Huatian Power Technology Co., Ltd, Suzhou 215000, China
ABSTRACT: New types of energy-saving transformers have been applied in many Chinese corporations,
under this background, it’s essential to study energy conservation and reconstruction methods and techniques
of distribution transformers for individual line enterprises. A brief review was made on domestic and abroad
energy conservation and reconstruction methods and techniques, including transformer material optimization,
life-cycle management, structure modification, condition-based maintenance, reasonable capacity selection,
economic operation and reactive power compensation devices application, etc. Moreover, the economic
switch method of transformer operation modes is given by a case study, and compared with conventional
means. Corresponding operation losses are made in contrast, which shows large energy conservation potential
exists in conventional operation modes. Another actual case is studied on transformer energy efficiency
diagnosis via reactive power compensation. Finally, weak links in current transformer energy conservation
and reconstruction methods are discussed. It’s concluded that new-type energy-saving transformers should be
promoted vigorously in Chinese enterprises. Hence, it can fully dig energy conservation potentials and
improve productive benefits for transformers, as well as give certain reference and guidance for the
management departments in aspect of energy conservation and reconstruction.
110
The sources of transformer power loss were
analyzed in the paper, and current transformer
energy conservation methods and technologies at
home and abroad were concluded and reviewed.
Then one of that was selected to make a case study,
in which the energy losses of a transformer in a
conventional operation mode and an economic
operation mode were compared, in addition, another
case study which adopted reactive power
compensation approach to realize transformer energy
conservation and reconstruction was given. The
results of two cases showed that large energy
conservation potential exists in the conventional
operation mode of a transformer. Finally, weak links
which exist in transformer energy conservation
research fields were discussed, prospects were made
and some suggestions were given.
2 TRANSFORMER LOSS ANALYSIS
The power loss of transformer can be divided into
active power loss and reactive power loss. The sum
of the active power loss of transformer and the added
grid active power due to its reactive power
consumption is called comprehensive power loss of
transformer, and its expression formula (C. Y. Xia
2007, H.R. Xue et al. 2015, K. M. Wang et al. 2015)
is shown as
z 0z KZ
2
P P P
(1)
Where P0Z is no-load comprehensive loss with
unit of kW, and P0Z=P0+kQ0, P0 is no-load active
power loss with unit of kW; Q0 is no-load reactive
power loss with unit of kW and Q0=I0%*SN×10-2;
PKZ is rated load comprehensive loss with unit of
kW and PKZ=PK+kQK; PK is load loss with unit of
kW; QK is short circle reactive power loss with unit
of kvar, and QK=UK%SN×10-2; β is transformer load
rate and β=S/SN; S is transformer load capacity with
unit of kVA; SN is transformer rated capacity with
unit of kVA; k is reactive power economic
equivalent with unit of kW/kvar.
2.1 Active power loss of transformer
Active power losses of a transformer are divided into
no-load loss and load one. The former is also called
iron loss, including eddy current loss and hysteresis
loss; the latter, also called copper loss, including
resistance loss and additional loss.
No-load loss (Z. B. Li 2013) is fixed loss which
has nothing to do with its load size, generally, larger
the transformer capacity is, larger its no-load loss
will be. Among them, hysteresis loss is electric
energy loss which is formed by heat release of
internal molecular mutual friction in silicon steel
sheet, due to its constant variation of magnetic line
direction and size, when AC current flows through
the transformer, the eddy current loss is generated by
iron core heating formed by the eddy current, and
which is produced by the induced electromotive
force in closed loop circuit.
In load loss, resistance loss is caused by
transformer coil resistance which is proportional to
the square of load current. Due to the short circuit
voltage UK is very small on primary side when the
transformer short circuit happens, the active power
loss that generated in transformer iron core can be
neglected, thus the transformer short circuit loss
△PK can be thought as copper loss; the additional
loss is mainly caused by transformer leakage flux,
which includes winding eddy current loss and
circulating current loss of parallel winding conductor
loss, and the structure loss, etc.
2.2 Reactive power loss of transformer
Transformation process of a transformer is finished
and based on its electromagnetic induction. During
power transmission of a transformer, reactive power
loss of the transformer itself is much larger than
active power loss. A part of transformer reactive
power loss is generated by excitation current of built
transformer main magnetic flux which has nothing
to do with load current and is a constant variation;
the other part is composed of transformer windings
reactance and the current flows through the windings
which is related with load current (Z. B. Li 2013).
The transformer reactive power loss can be
calculated as
2
0 30
K
TN N
%%
[ ( ) ]
100 100
IS
U
QS S
(2)
where I0% is percentage of transformer no-load
current on rated current; UK% is percentage of
transformer short circuit point voltage on rated
voltage; S30 is transformer calculation load; SN is
transformer rated capacity.
3 REVIEW OF TRANSFORMER ENERGY
CONSERVATION METHODS AND
TECHNOLOGIES
The significance of transformer energy saving is
research hotspot currently. Transformer energy
conservation methods can be divided into several
kinds, based on loss origin of the transformer, such
as a) Change loss characteristics of a transformer in
approach of optimizing the transformer magnetic
materials, which can originally reduce its no-load
loss and load loss, especially for transformers with
the amorphous alloy as iron materials can reduce its
no-load loss in large amplitude, thus they have a
very good energy conservation effect;
111
b) In order to maximize excellent function of high
quality materials, transformer structure design and
further reconstruction of transformer structure must
be achieved in approach, which can reduce losses
and save materials;
c) Utilize optimal load rate of a transformer or
adopt principle of minimum comprehensive energy
efficiency cost to select transformer capacity
reasonably, so as to avoid over large or too small
transformer capacity selection effectively, and
reduce the operation cost, and improve the energy
saving performance and economy;
d) After selection of a transformer, adopt a
transformer economic operation scheme to arrange
operations, such as make the transformer operate in
economic operation interval, economic switching of
transformer between each operation mode, and
economic dispatch between transformer loads, etc.
Adopting these schemes can improve transformer
operating efficiency and reduce its operation loss,
thus achieve transformer energy conservation;
e) Adopt a reactive power compensation method,
making the reactive power of transformer
compensated from the loads, thus can reduce the
reactive power loss and comprehensive power loss
of transformer. Reactive power compensation
methods include adoptions of conventional static
reactive power compensator SVC, static reactive
power generator SVG and active power filters (APF),
etc. Also we can adopt new-type distribution
transformer integration SVC technique to realize
comprehensive compensation and control reactive
power and power quality in junction positions in
high and low voltage levels. The transformer energy
conservation methods review is shown in the
flowing sections.
3.1 Optimization of transformer materials
A transformer utilizes its electromagnetic induction
to change network voltage, so that performance of
magnetic materials influences transformer loss
characteristics directly. Therefore, optimization of
transformer materials is an important transformer
energy conservation measure. In aspect of reducing
no-load loss, silicon steel sheets are constantly
developed and changed, and currently, which are
used in iron core magnetic materials are generally
with thickness of 0.23~0.30mm, among them, the
silicon steel sheet with thickness of 0.18mm has
been gradually applied, and the one with thinner
thickness has been a developing trend in future (N. F.
Xu 1990). In addition, application of amorphous
alloy materials also drives development of
transformers. A transformer based on the amorphous
alloy material was compared with one based on the
silicon steel sheet iron core, the no-load loss can be
reduced for 70%. K. Zhao and L. Y. Zhang (2006)
compared the losses of the transformer based on the
amorphous alloy iron core and an energy-saving type
of S11 transformer and concluded that the
transformer based on the amorphous alloy iron core
has a better superiority in aspect of no-load loss
reduction. In aspect of load loss reduction, a new-
type low resistance material has become a research
hotspot. Oxygen-free copper wire and electrician
aluminum wire which can make electrical
conductivity respectively are improved to 109% of
electrolytic copper and 104.2% of industrial
aluminum wire, thus they have been widely applied
in transformer energy conservation. Moreover,
utilize a feature that a superconducting material loses
resistance after it exceeds critical temperature, the
developed superconducting transformer not only can
reduce transformer loss but also improve its short
circuit performance (Z. G. Zhang et al. 2008).
3.2 Reconstruction of transformer structure
It’s not enough to reduce transformer loss by
adopting the materials with high magnetic properties;
the transformer structure must be improved to fully
utilize the better characteristic of materials. The
transformer structure construction is method to
optimize the transformer structure, thus achieve
materials saving, and losses reduction. The iron core
structure of transformer is changed from original
straight seam to semi-straight and semi-diagonal and
complete diagonal connection seam, thus can détente
the magnetic direction in iron core seam zone, and
reduce the no-load loss; adjust the ratio of silicon
steel sheet and magnetic line reasonably, and reduce
the current density, thus can reduce the load loss in
large amplitude (L. Yao and Z. S. Yao 2007); the
new-type roll-core transformer, due to its almost
none of accumulative joint, and the orientation of
silicon steel sheet is fully used in continuous roll-
winding, and is formed nature fastening state, which
avoids the loss increasing caused by clamping.
Compared with a traditional accumulative-type iron
core transformer, a transformer based on a roll-core
not only saves materials but also reduces its no-load
loss for 20%~35%, and no-load current is reduced
60~80% (N. F. Xu 1990). Based on this, Z. S. Wan
(2006) made a further optimization on the
transformer based on a roll-core, and proposed a
transformer based on a three-phase roll-core with
symmetrical triangle structure of circle section which
is totally symmetric, yoke is shortened in large
amplitude, magnetic resistance is reduced largely,
with no joint of an iron core, and high core column
filling coefficient, thus performance is improved
obviously and currently the transformer proposed is
the most ideal one with high efficiency, energy
conservation performance and environmental
protection.
3.3 Reasonable selection of transformer capacity
The transformer capacity is a most important
parameter when selecting transformer. If the
112
transformer capacity is selected over large, then the
investment of transformer is increased, as well as the
transformer no-load loss, which leads to operation
cost of transformer; otherwise, if the transformer
capacity is selected too small, the transformer loss is
increased, even causes overload of transformer in a
long period, and accelerates the isolation aging and
so as to shorten the transformer operation life.
Therefore, it’s of great energy conservation and
economic significance to select transformer capacity
reasonably. Aimed at this problem, there are mainly
two solutions, one is to select the transformer
capacity based on the transformer load rate (J. S. Hu
et al. 1999, Z. F. Wang et al. 2007, X. H. Shan et al.
2009) and the other is adopt the comprehensive
energy efficiency cost minimum principle to select
the transformer capacity (Z. L. Liu 2012, S. A. Weng
2007).
The load coefficient β of transformer is defined
Ncos
P
S
(3)
The (J. S. Hu et al. 1999) proposed that the
transformer has an optimal load coefficient, namely,
when
0 0 K
/PP
, the transformer efficiency is
largest, and meantime the copper loss of transformer
is equal to the iron loss. β0 is load coefficient in
smallest loss rate, and which is generally called
active power economic load coefficient. According
to β0, which is added into (3) above to select the
transformer capacity, thus can make the transformer
loss is reached to smallest. But in actual operation,
it’s unable to make selection according to the β0,
especially in current application of new-type
transformer series, the no-load loss is very low, if
it’s still according to the β0 to select the transformer
capacity, which will cause situation of “big horse
pulls small vehicle”, which is due to three aspects,
one is no taking consideration is took of the load
actual operation time, two is the reactive power
economic equivalent and three is the transformer
investment return.
Aimed at the circumstances, the (Z. F. Wang et al.
2007) proposed a more reasonable solution for
optimal load rate, that is, take the load actual
operation into consideration, the optimal load rate is
got as the following formula
0
max K
TP
P
(4)
where P0 is transformer no-load loss, the unit is kW;
PK is transformer short circuit loss, the unit is kW;
τmax is hours of maximum load loss, the unit is h; T
is total year utilization hours, the unit is hour.
When the reactive power loss is considered, the
optimal load rate is calculated as
00
zmax k max k
T P kT Q
P k Q
(5)
where Q0 is excitation power on power side when
the transformer is no-load (reactive power no-load
loss), the unit is kvar; QK is consumed leakage
power (reactive power load loss) when the
transformer is rated load, and the unit is kvar; k is
reactive power economic equivalent.
When the cased of transformer investment return
is considerate, the optimal load rate is calculated as
0 0 1 p e
opt max k max k
T P kT Q k k S
P k Q
(6)
where k1 is coefficient of converting the price into
power, when the price is took 0.5, k1 is took 2; kp is
converted present value coefficient, kp=[1-
1/(1+i)n]/i, i is annual interest rate; Se is rated
capacity of transformer, the unit is kVA.
Similarly, based on the above method, after
obtaining the more suitable d optimal load rate, then
which is took into Eq. (3) to select the transformer
capacity.
But due to constant variation of transformer load,
it’s hard to control the transformer always operates
in the optimal load rate, so it’s unable to adopt this
method to select in actual operation. Then X. H.
Shan, Y. M. Zhong and Y. Z. Wang (2009) proposed
a method to select the transformer economic
equivalent according to its economic interval, that is,
when the transformer economic operation interval is
given as [βL2, βL1], the transformer capacity is
selected according to the maximum load, and the
condition is
max L1
r
SS
(7)
The minimum load is used for calibration, and the
condition is
min L2
r
SS
(8)
Thus, it’s guaranteed that the transformer is not
over its economic operation interval when it is
operating, the solution of transformer economic
operation interval will be mentioned in next part.
The comprehensive energy efficiency cost
analysis method is not based on the transformer load
coefficient, and which is used to select the minimum
cost scheme as the optimal economic scheme via
calculation and analysis of transformer
comprehensive energy efficiency cost in each viable
technique scheme (Z. L. Liu 2012, S. A. Weng et al.
2007). First determine the transformer types, then
select multi-specification transformers as the
selective schemes, and finally according to the
known parameters. The optimal economic capacity
can be calculated as follows.
113
First, the transformer comprehensive energy
efficiency cost value is calculated as
EFC OEFC KEFC
TOC CI P P
(9)
Then, the transformer comprehensive equivalent
initial cost
ZEFC
P
is calculated as
ZEFC OEFC KEFC
P P P
(10)
where CI is distribution transformer initial cost;
POEFC is equivalent initial cost of no-load loss;
PKEFC is equivalent initial cost of load loss.
Finally, according to the economy and energy
conservation requirements of transformer capacity
selection, select the scheme that the value of TOC is
smallest and the scheme that the value of PZEFC is
smallest, then make a further comparative analysis of
the two schemes to select the optimal economic
capacity of transformer.
3.4 Transformer economic operation
After a transformer being designed according to
excellent materials and structure and selecting a
reasonable capacity, in actual operation, adopt
economic operation scheme can further reduce loss
and achieve transformer energy conservation.
Problems of transformer economic operation
contained three aspects of research, i.e. transformer
economic operation interval, transformer economic
switching between operation modes, and economic
dispatch between transformer loads. During study of
transformer economic operation, we can respectively
build an analysis model according to conditions of
the smallest active power, the smallest reactive
power loss and the smallest comprehensive power
loss. For example, if active power energy saving is
main, it should arrange economic operation
according to the minimum principle of active power
loss; or if power factor increasing is main, then it
should arrange economic operation based on the
minimum principle of reactive power loss; or if the
two aspects are considerate and system network loss
reduction is main, it should arrange economic
operation according to the minimum principle of
comprehensive power loss (H. T. Liu 2008).
1) Transformer economic operation interval
Transformer loss is changed with variation of
load rate, and when it operates in no-load or low
load, the main loss is iron loss; the loads of a
transformer increase, corresponding load loss is
gradually increased, and when load rate of a
transformer is larger than a certain value, the load
loss will again occupy the main part. It’s known
from section 3.3 that the transformer has an optimal
load rate, and in operation of this load rate,
comprehensive energy loss of the transformer is the
smallest, and operation efficiency is the highest.
Transformer load rate cannot be kept in the optimal
load rate for a long time, due to which a transformer
is always controlled to work in economic operation
interval and actual operation. Therefore, it both has
great significances to determine economic operation
interval of a transformer for ensuring transformer
economic operation and selecting reasonable
transformer capacity.
The Refs (H. T. Liu 2008, S. L. Liao & C. Yu
2011, Z. P. Cao 2005) both made a deep research on
determination methods of transformer economic
operation interval, which considered that upper limit
value of transformer economic operation interval
should be set as load rate β=1, and loss rate
corresponding to lower limit value should be equal
to the rated load loss rate. The transformer economic
operation interval determined by this method can
guarantee that, in transformer actual operation, the
loss rate is lower than the rated load loss rate, and
the efficiency is higher than the operation efficiency
in rated transformer load.
Except for the methods mentioned above, S. A.
Weng (1998), based on the constraint condition,
proposed that the transformer’s annual energy loss
△W% is not exceeded 1% of annual minimum
energy loss △Wmin% corresponding to consideration
of optimal load coefficient
z
β
of reactive power
loss, determined the transformer economic operation
interval. Thus, the load rate corresponds to the upper
and lower limit values of economic interval is
determined by the formula
c0
(11)
where
[0.718,0.905]
.
In this method, and compared with each grade
capacity of transformer, the transformer economic
operation interval suitable for transformer capacity
selection can be obtained with objective of lowest
energy loss, that is, if the graduation of two-grade
transformer capacity is 1.26 and the annual electrical
energy loss of transformer is compared through the
two grades, then solve the critical load coefficient
βr=0.905β0 when △W1%=△W2%, then combined
with (11), the load rate corresponds to the economic
operation interval that is suitable for transformer
capacity selection is
cm
, [0.718,0.905]
(12)
According to this solved interval, the annual
energy loss of the transformer that is selected will be
lower than the transformer with any other capacity,
so as to guarantee energy conservation performance.
2) Economic switching between transformer
operation modes
When a single transformer is operating
independently, it is generally in its economic
operation interval to realize energy conservation.
While in distribution network, there are other
transformer operation modes, i.e. one transformer
operates with a spare one, two or multi transformers
with same capacities in parallel operation, those with
114
different capacities in parallel operation, etc. In
actual operation, switching between different
operation modes is generally finished according to
variation of loads with a goal of the minimum
comprehensive loss, which is an economic switching
problem between the transformer operation modes.
The key to solve economic switching problems is
solving the turning point between two different
operation modes. T. T. Li (2010) thought that the
turning point is solved in approach of solving the
relationship between the comprehensive loss △P and
load S of two kinds of transformer operation mode
to be switched, the load that is corresponding to the
intersection point of two kinds of curve is the critical
load of switching between two operation modes. If
this critical load point is met with no exceeding of
its full load operation point, then the operation mode
switching can be finished at this point; otherwise,
when its operation mode reaches to the critical load
point, the operation mode switching should be
finished at the point.
It’s essential to stress that aimed at parallel
operation case of transformer with different
capacities, the load distribution coefficient C should
be took into consideration in solving the relationship
between the comprehensive loss △P and load S of
transformer operation mode.
J. Yang and X. Q. Ding (2006) proposed that, if
we are absolutely according to load variation and
frequently switching transformer operation situations
when it reaches the optimal economic operation
turning point, it will have negative effects on safe
and stable operation of substation, transformers and
switches. So in actual operation, we are considering
adopting a time-inverted control method, that is,
according to the solved theoretical turning point of
transformer operation mode switching, solves
transformer economic operation scheme that is
integrated economy and actual significance, and
distributes the transformer switching times in
advance.
3) Economic dispatch between the transformer
loads
When the total electricity consumption loads of
transformer are constant, as well as the operation
modes of transformer, the total active power loss and
reactive power consumption of transformer will be
changed with the variation of load distribution
between transformers. Therefore, the transformer
total active power loss and reactive power
consumption can be reduced minimum values
according to economic dispatch of loads between
transformers, so as to achieve transformer energy
conservation ( Z. P. Cao 2005). The problems of load
economic dispatch between transformers are mainly
solved in approach of mathematical methods.
Z. P. Cao (2005), J. H. Wang & N. Gu (2002), J.
S. Hu (1981) both built the mathematical
relationship model between the active power loss
△P and apparent power S of each load, utilize the
mathematical method to solve the load economic
dispatch coefficient when the active power loss is
lowest.
The load economic dispatch coefficient of two or
multi transformers with same capacities is
1
11
jn
jk iik
CPP
(13)
where Cj is load economic dispatch coefficient of
the transformer j; Pjk is the short circuit loss of
transformer j; Pik is short circuit loss of transformer
i. The load economic dispatch coefficient of two or
multi transformers with different capacities is
2
2
1
jN
jk
jniN
iik
S
P
CS
P
(14)
where SjN is rated capacity of transformer j; SiN is
rated capacity of transformer i.
Based on Eq. (13) and (14), the solved apparent
power of each transformer load is the optimal
economic scheme in capacity distribution.
Law of load economic dispatch got from Eq. (13)
~ (14) is that load economic dispatch coefficient
between two or multi transformers with same
capacities is inverse proportional to short circuit loss
of transformer itself; load economic dispatch
coefficient between two or multi transformers with
different capacities is inverse proportional to short
circuit loss of transformer itself, and proportional to
square of transformer capacity.
The above economic dispatch method can achieve
the goal of minimum power loss in transformer
operation, thus can save active power energy in large
amplitude. If the main goal is to increase the power
factor in actual operation, it should distribute loads
according to the principle of minimum reactive
power loss; if it is integrated with the two goals
above or main focuses on reduction of system
network loss, it should distribute loads according to
the principle of minimum comprehensive power loss,
aimed at which, J. S. Hu (1981) built the model, and
analyzed the load economic dispatch problems
respectively in principle of minimum reactive power
loss and comprehensive power loss, and met the
different operation requirements, and made the
model of economic dispatch between transformer
loads be more satisfactory.
4) Utilization of reactive power compensation
devices
It’s known from section 2.2 that the reactive
power loss is main transformer loss. If the reactive
power of transformer is compensated from the load,
then the transformer reactive power loss can be
115
reduced effectively. G. J. Song (2010), based on
experiment, compared the transformer losses in the
different load power and power factor and verified
the effective reduction of transformer reactive power
loss via improvement of the power factors on
transformer load side. Therefore, adopting reactive
power compensation devices on transformer load
side is an important measure, as well as a significant
method of transformer energy conservation.
Large amount of studies have been made on
reactive power compensation devices in domestic
and abroad, the conventional reactive power
compensation devices (Y. L. Dong et al. 2003) are
main static reactive power compensator SVC, static
reactive power compensator SVG, and active power
filter APF, etc. SVC can change the input equivalent
susceptances in grid by means of controlling its
thyristor trigger angle, so as to achieve the goal of
regulating the reactive power output (L. Su, S. Song
and J. Y. Chen 2004). Because there is no current
breaking ability of loop component, a larger amount
of harmonic current are generated within grid, aimed
at which, the reactive power compensation devices
are based on electric and electronical inverter
techniques, for example, static reactive power
compensator SVG and active power filter APF
which have good performance. SVG outputs voltage
phase and amplitude on AC side via adjustment of
bridge inverter circuit, or directly controls its AC
side current to generate or absorb reactive power that
meets requirements, so as to realize dynamic reactive
power compensation from inductive to capacitive
reactive power in total range (J. J. Liu 2012). The
main compensated goal of APF is the harmonic
current in system, among which, the parallel type of
APF can produce the harmonic current with equal
size and inverse direction, compared with the load
current, then the current is compensated as
sinusoidal wave (M. Hu and Y. Chen 2000), and
thus it’s able to implement simultaneous dynamic
compensations on the real-time changing harmonic
current and reactive power in grid.
Except for those above conventional distribution
network reactive power compensation devices which
are utilized for local compensation on user side,
aiming at problems of reactive power compensation
on transformer load side, Q. Xiong (2011) proposed
a new-type distribution transformer with integrated
static reactive power compensation technique (DT-
STATCOM). DT-STATCOM integrates the good
performance of conventional distribution
transformer and power electronic SVC devices, and
utilizes comprehensive the information on multi
transformer sides, so as to realize comprehensive
compensation controlling of reactive power and
power quality on high and low voltage grade
conjunction points. This kind of technique reduces
the withstand voltage requirements of power
electronic device effectively, and can fully utilize the
surplus capacity of transformer, and improve
efficiency, and which is a new method of utilizing
the reactive power compensation devices to achieve
transformer energy conservation.
3.5 Life-cycle management
Life cycle refers to each stage of a whole process in
which a product accesses resources and energy from
nature, through mining and smelting and other
production process, and then via storage, selling,
utilization and consumption until it is scrapped and
disposed, namely, a whole life cycle of a product in
material conversion from cradle to grave (R. R. Ning,
M. A. Tang and Z. C. Du 2010). Life cycle analysis
(LCA) is a technological approach used to evaluate
the impacts of a product on the environment during
its whole life cycle, from raw materials access,
production, utilization until to disposition after
utilization (C. Chen, H. Zhao and P. H. Wang 2009).
Y. Y. Zhang (2014) built a real-time status-based
transformer condition-based maintenance model,
which gave consideration to the economy and
reliability, and focused on some key problems for
transformer, such as insulation fault diagnosis,
insulation life assessment, transformer condition
evaluation, transformer risk evaluation, transformer
life-cycle cost evaluation, etc., to realize large power
transformer condition-based maintenance strategy
that comprehensively balanced the technicality and
economy. Thus utilize the life-cycle management to
urge the enterprises consider the maintenance cost
and system fault, after which, finally determine the
optimal maintenance scheme, as well as adopt
reasonable maintenance modes, to reduce equipment
detection and maintenance total cost, prolong
equipment utilization cycle, ensure equipment
reliability, and properly balance the factors of
reliability and economy for transformer safe
operation. Y. Y. Zhang (2014), based on an idea of
life cycle management, and started from transformer
real-time status, comprehensively considered the
reliability and economy for its operation and
maintenance, and then developed condition-based
maintenance strategies according to its status, to
completely improve the level of delicacy
management and scientific decision-making, and
narrow the gap with foreign advanced level.
Enterprises conduct energy conservation and
reconstruction of transformer need to focus real-time
operating status of transformer, based on reasonable
development of condition based maintenance
strategies, to improve the economy and reliability for
transformer safe operation to the greatest extent,
hence, realize transformer energy conservation and
reconstruction, avoid excessive maintenance to
transformer, and cause enormous waste of resources
and energy. The author has ever proposed a dynamic
correction based power transformer reliability
evaluation model (X. B. Guo et al. 2016), which
116
selected the transformer oil-paper insulation system
as evaluation object, began with the hot spot
temperature (HST) as the research problem, and
combined with Weibull distribution and Arrhenius
reaction law, the HST-based transformer aging
failure model was developed, which was applied to
describe the aging process of transformer, then the
failure rate of transformer oil-paper insulation
system was solved via calculation of winding HST,
next, utilized the dissolved gases analysis data of
transformer and combined with grey theory, the
HST-based aging fault model was modified
dynamically, to guarantee the evaluations can well
track and reflect the actual reliability level of
transformer. An example analysis based on the data
from Jiangmen power bureau was made (X. B. Guo
et al. 2016), by which, the validity of the model was
verified. The model proposed by the author can
assist power enterprises to track operating status of a
transformer in real time, develop proper condition-
based maintenance strategies, and reduce failure risk
via excessive maintenance, hence, the maintenance
efficiency of transformer was improved and
maintenance cost reduced through timely and
effective operating maintenance, which provided
implementation basis for further energy conservation
and reconstruction of transformer.
4 TRANSFORMER ENERGY CONSERVATION
ANALYSIS CASE 1: IN ECONOMIC
OPERATION MODE
4.1 Transformer energy conservation and diagnosis
flow chart design
In this section, aiming at two transformers in one
certain substation, the economic switching of
transformer operation modes discussed in section 3.4
is shown as an example, in which, the energy losses
of transformer in conventional operation mode and
economic operation mode are compared, then the
energy conservation potential was calculated and
analyzed. This process is based on the transformer
energy conservation analysis flow chart, which is
shown in Figure 1.
Transformer energy
conservation analysis
Transformer
parameters
initialization
Import data and
generate graphs
Energy conservation
diagnosis: whether the
operating mode is
economic?
No
Yes
Save data and keep
operating
Output results and
import the report
Energy conservation
scheme: economic
operating mode switching
Calculate the energy
conservation benefits
Figure 1. Transformer energy conservation
and diagnosis flow chart.
4.2 Basic parameters calculation
The basic parameters of the substation are shown in
Table 1 as follows.
Table 1. Basic parameters of substation.
Transformer
number
SN
/kVA
P0
/kW
PK
/kW
I0%
UK%
A
800
1.50
6.50
0.8
0
4.5
B
1600
2.70
15.20
0.6
0
4.5
Use the annual continuous load curve to represent
the substation load and which is shown in Figure 2.
220
635
3100 7100 8760
1810
Load Sc(kVA)
Time(Hour)
Figure 2. Annual continuous load curve of substation.
According to Table 1 and Eq.(1) (reactive power
economic equivalent k is took 0.1 kW/kvar), for
transformer A, QA0=IA0%*SAN*10-2=6.4kvar,
QAK=UAK%* SAN*10-2=72 kvar, PA0Z=PA0+kQA0=
2.14kW, PAKZ= PAK+kQAK =13.7kW, then the
comprehensive power loss of transformer A is
ΔPAZ=PA0Z+(SC/SAN)2*PAKZ; for transformer B,
QB0= IB0%*SBN*10-2=9.6kvar,
QBK=UBK%*SBN*10-2=72kvar, PB0Z=PB0+kQB0=
3.66 kW, and PBKZ=PBK+ kQBK=18.8kW, and the
comprehensive power loss of transformer B is
ΔPBZ= PB0Z+(SC/SBN)2*PBKZ.
When transformer A and B operate in parallel
mode, the parameters of them are added to Eq. (14)
and the load economic dispatch coefficient of the
two transformers are CA=0.369, CB=0.631.The
comprehensive power loss of transformer A and
transformer B in parallel operation mode is
calculated as ΔPABZ=PA0Z+PB0Z+
(SC*CA/SAN)2*PABZ+ (SC*CB/SBN)2*PBKZ.
4.3 Energy loss of transformer conventional
operation mode
Based on the conventional transformer operation
mode, the transformer is more economic in higher
load rate. Most transformers in domestic are adopted
the conventional operation mode, that is, when over
two transformers operate, only till one is in full load
situation, the other can be put into operation.
117
Therefore, the operation scheme of the substation in
conventional operation mode is shown as 1) when
0<h≤3100, arrange transformer A operates
independently; 2) when 3100<h≤7100, arrange
transformer B operates independently; 3) when
7100<h≤8760, the arrange transformer A and B
operate in parallel mode. In this designed operation
modes, the energy loss is calculated as
ΔAconventional= ΔA0<h≤3100 + ΔA3100<h≤7100 +
ΔA7100<h≤8760 = 9845.794 + 43086.141 + 41380.707
= 94312.642 kWh.
4.4 Energy loss analysis of transformer in
economic operation mode
The two transformers in this substation has three
operation modes, i.e. transformer A and B operates
independently and transformer A and B operate in
parallel mode, when two transformers operate in
parallel mode, the parameters of transformer A and
B are added into (14) and the load economic
dispatch coefficients of the two transformers are
AB
0.369, 0.631CC
.
Therefore, the relationship curve between the
comprehensive power loss P and load capacity SC
in the three operation modes can be shown in Figure
3 as follows.
Comprehensive power loss
ΔP (kW)
Load capacity SC (kVA)
ΔPA-S
ΔPB-S
ΔPAB-S
X:328.7
Y:4.454
X:484.9
Y:7.173
0 100 200 300 400 500 600 700 800
2
4
6
8
10
12
14
16
Figure 3. Relation curve between the comprehensive power loss
P and load capacity SC in three operation modes.
Three aspects can be concluded from Figure 3,
including a) when
C
0kVA 328.7kVAS
, arrange
transformer A in economic operation singly; b) when
C
328.7kVA 1600kVAS
, arrange transformer B in
economic operation singly; c) when
C1600kVAS
,
the load capacity has been exceeded the rated
capacity of transformer B, it should arrange
transformer A and B in parallel operation at this
point.
Therefore, the economic operation modes that are
adopted in this substation is designed as a) when
0 3100h
, arrange transformer A in operation
singly; b) when
3100 7100h
, arrange transformer
B in operation singly; c) when
7100 8760h
,
arrange transformer A and B in parallel operation. In
this designed economic operation scheme, the
electrical energy loss is
ΔAconventional=ΔA0<h≤3100+ΔA3100<h≤7100
+ΔA7100<h≤8760=9845.794+26484.734+41380.706
=77711.234 kWh.
The electrical energy loss calculation results in
conventional and economic approach are compared,
and when adopting the economic operation mode,
the electricity saving percent is calculated as
energy-saving
conventional economic
conventional
A%
A - A
=A
94312.64-77711.23
= 100%=17.6%
94312.64
(15)
Therefore, it can be concluded that adopt
economic switching for transformer operation modes
can reduce the electrical energy loss in large
amplitude and has a great economic significance.
5 TRANSFORMER ENERGY CONSERVATION
ANALYSIS CASE 2: REACTIVE POWER
COMPENSATION
5.1 A brief analysis for reactive power
consumption
In industrial and household electricity consumption,
resistance-inductance load occupies a large
proportion, such as the asynchronous motor,
transformer, etc. The reactive power consumption of
asynchronous motors and transformers occupies a
large proportion in the whole reactive power
provided by a power system, meanwhile, the electric
reactors and overhead lines in power system also
consume some reactive power, but the resistance-
inductance loads must have absorption of reactive
power, then it can work in normal conditions, which
are determined by its features. Some power
electronic apparatus especially non-linear devices
also consume reactive power, such as various phased
controlled ones, in addition, these devices also
produce a large amount of harmonic current, and the
harmonic sources also consume reactive power.
For reactive power, there are some impacts of
which on public grid, i.e. a) increasing of equipment
capacity. Increasing of reactive power causes
increasing of current and apparent power, then the
start of power users and the size and standard of
control equipment and measuring meters are both
improved; b) increasing of EQ and lines loss.
Increasing of reactive power causes increasing of
total current, thus the losses of equipment and lines
are increased; c) causes increasing of the voltage
drop of lines and transformers. If a reactive power
load is impacted, the voltage will generate strong
118
fluctuations, which causes severe reduction of power
supply quality.
Active power fluctuations generally have little
influence on grid voltage; the fluctuations of grid
voltage are mainly caused by that of reactive power.
During starting period of a motor, its power factor is
low, this impact reactive power will lead to severe
fluctuations of grid voltage, even causes the users
that connect to the same grid fail to work normally,
for example, the electric-arc furnaces can cause
frequent reactive power impacts, even influence the
power supply quality seriously (Z. A. Wang 2009).
5.2 A brief introduction of reactive power
compensation schemes
In a power system, voltage and frequency are the two
most important and fundamental indexes, which are
must be controlled in a certain range for normal
operation of power system. The frequency control is
related to active power controlling, while the voltage
control is one of most important approaches to
control the reactive power. There are many
approaches to control reactive power, we can adopt
a) synchronous generator; b) synchronous
condenser; c) parallel capacitor; d) static var
compensator (SVC). Among which, the parallel
capacitors have gradually replace the synchronous
condenser because of its simplicity, economy,
convenience and flexibility, while the SVC, as a new
kind of reactive power compensator, has constantly
developed and applied widely in recent years.
According to installation positions of capacitors,
there are generally three kinds of parallel capacitor
reactive power compensation modes, such as (L. F.
Cheng 2015, T. J. E. Miller 1990, Z. X. Han 1993, L.
F. Cheng, B.Zhou &T. Yu 2014, Z. L. Dou & F. Q.
Du 2015, Z. X. Han & G. Y. Wu 1994, F. Z. Peng, G.
W. Ott & D. J. Adams 1998, Z. Shu, S. Xie & Q. Li
2011, W. M. Wu, J. Lu & M. Tan 2015, B. Singh, K.
A. Haddad & A. Chandra 1998):
1) Concentrated compensation. The capacitor
banks are intensively installed on 6~10kV bus-bar of
an enterprise or local overall buck substation, to
improve the power factor of whole substation, and
make the reactive power in power supply area of the
subtraction basically balanced, and reduce the
reactive power loss of high voltage lines, and
improve the power supply quality of subtraction;
2) Component compensation. The capacitor banks
are respectively installed on the high or low voltage
bus-bars of workshop substation and distribution
station with low power factors, which also calls
disperse compensation, this compensation mode has
same advantages with concentrated compensation,
only the reactive power capacity and range are
relatively smaller, but the effects of component
compensation are more obvious and it’s also more
widely applied;
3) Local compensation. The capacities or
capacitor banks are installed on an asynchronous
motor or close to the inductive electric equipment;
compensate locally, which also calls single
compensation or individual compensation mode.
This mode not only improves the power supply loop
factor of electric equipment, but also the voltage
quality, thus which is very suitable for the middle
and small equipment. In recent years, with
improvement of ability, China is gradually equipped
with low voltage self-healing parallel capacities, and
the types and specifications are gradually complete,
some conditions are formed for promotion of local
compensation mode, and many successful cases have
been applied.
If we take balanced considerations and rational
distributions for the three compensation modes, they
will get much more technological and economic
benefits. So we can give the following suggestions
for energy conservation: a) preferentially replace the
old SJ-type transformers as new-S11-type energy-
saving ones; b) replace the S7-type transformers as
S11-type transformers according to the economic
conditions and synthesize each aspect of factor; c)
for two main transformers, adopt some controlling
devices to average balance the loads in operation, to
reduce transformer losses, the effect of energy
conservation is obvious, suggest adopt it.
5.3 Case study for energy conservation benefits
via reactive power compensation
An enterprise energy conservation and
reconstruction example of cement plant in
Guangzhou province are given, in which, the power
supply and distribution system and belonging
electric equipment are made energy conservation
analysis, six monitoring points are measured and the
site information of each measuring point are shown
in Table 2, and the position of each measuring point
and electric connection mode are shown in Figure 4.
Although the enterprise power factors are completely
kept between 0.92 and 0.95 in a long period, and
each month can be awarded by the power supply
corporation, but we can see from intraday actual
measuring values that the intraday average power
factors of some transformers are still lower than 0.9
(such as 3#3), and even some are lower than 0.8
(such as 2#2), thus it’s essential to improve the
power factors of transformers, next, we will compute
the energy conservation benefits the case brings
when the power factors of all transformers are
improved to above 0.95.
Table 2. The site information of
each measuring point.
Monitoring
point
Name
Type
Impedance
(%)
Connection
mode
No.1
1#
S7-8000/35
7.80
Y-yn0
No.2
2#
S7-8000/35
7.80
Y-yn0
119
No.3
3#3
S7-1250/6
4.50
Y-yn0
No.4
3#2
S7-630/6
4.50
Y-yn0
No.5
4#1
S7-1250/6
4.50
Y-yn0
No.6
2#2
S9-M-
1600/6
4.65
D,yn11
6kV
Refers to
measuring
point
Interconnection
switch
Closed
Zhongyin Line
(LGJ-150-35KV)
2# step-down
transformer
(SJ-50)
2# main
transformer
No.2
1#
capacitor
2#
capacit
or
Liangyin Line
(LGJ-150-35KV)
No.1
3#3
distribution
room
2#2
distribution
room
3#2
distribution
room
No.6
No.3
Interconnection
switch
Closed
1# main
transformer
No.4 No.5
4#1
distribution
room Load
1# step-down
transformer
(SJ-50)
Load
Figure 4. The position of each measuring point and electric
connection mode.
The reactive power compensation amount is
computed as (Q. Liu 1989, L. F. Cheng, B. Zhou &
T. Yu 2014, L. L. Zhang et al. 2010):
2
2k
22
N 1 2
11
cos cos
P
PP
S
(16)
where P2 is active power of transformer, cosφ1 is
measuring value of original power factor, cosφ2 is
measuring value after compensation, and it is took
0.95.
According to Eq. (15) and the actual measuring
values, we can calculate the electricity saving
amount in each 30s, finally, we can get the intraday
actual energy conservation amount via compensating
the power factors to 0.95 in a comprehensive
calculation, and it is calculated as the following
formula:
2
2k
22
N 1 2
11
Acos cos
P
P dT P dT
S
(17)
All variables in (15)~(16) are both selected the
measuring values, then the computational results are
very accurate, certainly, the conditions are supposed
that the transformer loads have no impacts in
measuring interval of 30s, which is absolutely
agreeable in engineering computation. Hence,
according to Eq. (15), we can get the energy
conservation amount of transformer in 30s, and that
is
2
2
30 k
22
N 1 2
11
A 30
cos cos
sP
P dT P s
S
(18)
While the P2 and cosφ1 are constantly changed in
a day, thus, it’s need to divide 24h of a day into
sections in each 30s, that is number of 2880, then the
approach to calculate energy conservation amount of
a day is just to add that of 2880 sections, and that is
2
2880 2880 2
24 30 k
22
11
N 1 2
11
30
cos cos
n
hs
nn n
P
A A P s
S
(19)
where P2n is active power measuring value in time
of the section n, cosφ1n is power factor measuring
value in time of the section n. All of these values are
recorded and saved by the measuring device, and can
be directly used for calculation.
We can use the Matlab to write calculation
programs, then each transformer data and measuring
data are imported Matlab and the intraday energy
conservation benefit via reactive power
compensation can be obtained, and we assume that
the intraday measuring values are whole year
average values, then which are multiplied by 365,
and we will get the whole year energy conservation
amount via reactive power compensation. We can
use the software that is developed to match the
Fluke435 to calculate the results, which are shown in
Table 3. “Para1” means “Year electricity
consumption amount before compensation (kWh)”;
“Para2”means “Year electricity consumption amount
after compensation (kWh)”; “Para3” means “Year
saved electricity amount after reactive power
compensation (kWh)”; “Para4” means “The rate of
electricity saving (%)”.
Table 3. The whole year energy conservation amount
of each measuring point after implementing reactive
power compensation.
Name of
transformer
Type
Para1
Para2
Para3
Para4
2# main
S7-
8000/3
5
159408.8
144752.9
14655.9
9.19
3#3
S7-
1250/1
0
41120.2
31832.6
9287.6
22.59
2#2
S9-M-
1600\6
27591.3
22620.6
4970.7
18.02
3#2
S7-
630/6
3259.3
1997.5
1261.8
38.71
4#1
S7-
1250/1
0
4274.7
3184.3
1090.4
25.51
Summation
---
235654.3
204387.9
31266.4
13.27
According to Table 3, we can calculate the total
year electricity saving benefits of the five
transformer measuring points via reactive power
compensation, namely, the year electricity saving
amount is 31266.4kWh, the reduced standard coal is
11.80tce(=31266.4/10000×3.6), the reduced CO2
120
emission is 30.92 tons, and the energy conservation
benefit is 15.1 thousand Yuan
(=31266.4/10000×0.46).
Although the energy conservation benefit is not
too high, which can be seen from the calculation
results, we can conclude that it’s very essential to
implement reactive power compensation for each
transformer in aspect of energy saving, certainly, for
a concrete enterprise, we need to consider its own
conditions when implement reactive power
compensation. We suggest the enterprises firstly
according to their development status, and
preferentially implement reactive power
compensation for those measuring points with lower
power factors and reactive power compensation
electricity saving benefits, to reduce electric energy
loss, and improve energy utilization rate. This
cement plant in Guizhou province has no higher
power factors for those distribution transformers, so
that the electricity saving rate of reactive power
compensation is considerably high, and the average
is reached to about 20%. Although current load rate
is lower, and absolute loss is not higher, while the
former is higher, the latter will increase sharply, and
then the effect of total energy conservation is very
outstanding.
6 PROSPECTS OF TRANSFORMER ENERGY
CONSERVATION METHODS
Except for the introduced transformer energy
conservation methods in the second part of this
paper, according to impact factors on transformer
economic operation, the prospects of deep
transformer energy conservation research directions
are made as follows.
1) Stablity of the transformer operation voltage.
Transformer active power loss is in proportion to
square of voltage, generally, when the transformer is
operating with 5% of over voltage, the iron loss will
be increased 15%; when with 15%, the iron loss will
be increased over 50% and the no-load loss
increased in a large amplitude as well, that is, the
reactive power loss of grid is increased. Therefore,
we should adopt an automatic voltage regulator and
other devices to control power quality and avoid
over-voltage operation for a transformer.
2) Maintain three-phase load balance of a
transformer. Load loss of a transformer is changed
with variation of a transformer operating load, and is
in proportion to load current. When a three-phase
load is balanced, the loss of transformer is the lowest;
while when is unbalanced, the load loss of
transformer is equal to the sum of three single
transformers’ load losses, and even the loss, in
condition of maximum unbalance, is three times of
in balanced condition. Moreover, line loss on a high
voltage side is generated due to three-phase load
unbalance on a low voltage side. Thus, taking
measures or adopting devices to maintain the
transformer three-phase load balance are important
approaches to reduce transformer loss.
3) Decrease transformer operation temperature.
Winding resistance of a transformer increases as
temperature rises, and the temperature of each liter
decreases 1℃,the load loss can be reduced 0.32%.
Therefore, it’s of great significance to reduce
transformer loss by optimizing temperature
controlling and reduction measures.
7 CONCLUSIONS
For the loss in power system transmission and
distribution networks, the loss of a transformer
occupies a large percent, which makes the
transformer energy conservation become one
imminent problem to be solved currently. On this
background, some aspects are studied as follows.
1) The source of transformer power loss is
analyzed, and the current transformer energy
conservation methods in domestic and abroad were
classified and summarized, which concretely
contained the optimization of transformer materials,
reconstruction of transformer structure, reasonable
selection of transformer capacity, transformer life-
cycle management, economic operation control of
transformer and adoption of reactive power
compensation devices, etc.
2) Transformer energy conservation methods are
reviewed in this paper, such as economic operation
controlling of a transformer, which included
determination of economic operation interval,
economic switching between transformer operation
modes, and economic dispatch between transformer
loads, etc. then, the calculation analysis of
transformer electrical energy loss in operation mode
economic switching approach was given as an
example, the calculation results show that adoption
of this approach, compared with the conventional
operation mode, can save 17.6% of energy, and
adoption of reactive power compensation is very
essential for energy conservation improvement.
Examples analysis showed that an obvious economic
significance exists in transformer energy
conservation and reconstruction work, and also
verifies that a greater energy conservation potential
in current conventional operation modes.
3) The transformer energy conservation research
was prospected in aspects of stabilizing transformer
operation voltage, maintaining transformer three-
phase load balance and decreasing transformer
operation temperature, etc. the research of their
aspects above are weak currently. Finally some
suggestions are given for those enterprises,
especially high energy consumption ones, should
vigorously promote the use of new-type energy-
121
saving power transformers, meanwhile, a deep
research on transformer energy saving and diagnosis
should be conducted in future.
8 ACKNOWLEDGEMENTS
The authors gratefully acknowledge the support of
the Science and Technology Projects of China
Southern Power Grid-Advanced research and
application of smart operation management and
optimization technologies in distribution area(K-
GZM2014-140).
REFERENCES
[1] C. Y. Xia. 2007. Analysis and application of transformer
economic operation. Transformer 44 (12): 24-28.
[2] H. R. Xue, K. H. Zhang, B. Li, et al. 2015. Fault diagnosis
of transformer based on the cuckoo search and support
vector machine. Power System Protection and Control 43
(8): 8-13.
[3] K. M. Wang, H. C. Shu, L. P. Cao, et al. 2015. Study of
OLTC running state evaluation method based on correlation
analysis. Power System Protection and Control 43 (19) :
54-59.
[4] K. Alimjan, J. L. Suonan, A. N. Amingul, et al. 2014. A
novel scheme for UHV auto-transformer protection based
on zero-sequence current longitudinal comparison principle.
Power System Protection and Control 42 (22): 24-31.
[5] Z. B. Li. 2013. Categorizing and analysis of energy saving
measures in voltage transformer. Applied energy technology
30 (5): 29-32.
[6] N. F. Xu. 1990. Summary of energy-saving measures of
transformers. Water Conservancy & Electric Power
Machinery 2 (2): 32-34.
[7] K. Zhao and L. Y. Zhang. 2006. International experience on
transformer energy conservation and consumption
reduction. Power Demand Side Management 8 (5): 63-64.
[8] Z. G. Zhang, L. X. Liu, Z. Y. Li, et al. 2008. Analysis of the
economic benefits to new energy-saving distribution
transformer, in 2008 IEEE China International Conference
on.Electricity Distribution, pp. 1-3.
[9] L. Yao and Z. S. Yao. 2007. Structure and characteristics of
energy saving transformer. Electric power 52 (9): 49-53.
[10] Z. S. Wan. 2006. Research of key technology of the new
type economy transformer with triangle configuration round
section roll-core. Masteral Dissertation: Donghua
University, Shanghai, China.
[11] J. S. Hu, et al. 1999. Transformer economic operation,
China Electric Power Press, Beijing, China.
[12] Y. Y. Zhang. 2014. Study on life cycle cost based
maintenance decision making for power transformers
considering condition assessment and insulation life
assessment. Masteral Dissertation: Chongqing University,
China.
[13] X. B. Guo, L. F. Cheng, G. P. Wang, et al. 2016.
Research on reliability model of power transformer based
on dynamic correction technique. Electric Power
Automation Equipment 36 (6) : 148-155.
[14] Z. F. Wang, D. Q. Zhao, J. Bai, et al. 2007. Selection of
reactive compensation and transformer capacity. Power
capacitor 28 (3): 16-19.
[15] X. H. Shan, Y. M. Zhong, and Y. Z. Wang. 2009. Choice
of distribution transformer’s economic operation range and
capacity. Guangxi Electric Power 32 (2) : 38-39.
[16] Z. L. Liu. 2012. Analysis on distribution transformer
capacity selection and economic running. Shanxi Electronic
Technology 40 (4): 42-44.
[17] S. A. Weng, J. L. Liu, and H. Wang. 2007. Economic
evaluation of distribution transformer capacity selection.
Electrotechnical Application 26 (1): 140-143.
[18] H. T. Liu. 2008. Energy saving study on the distributing
transformer. Masteral Dissertation: North China Electric
Power University, Beijing, China.
[19] S. L. Liao and C. Yu. 2011. The energy-saving methods
of distribution transformer. Modern Architecture Electric 2
(8) : 13-17.
[20] Z. P. Cao. 2005. Economic operation study on the
distribution transformer. Masteral Dissertation: Taiyuan
University of Technology, Taiyuan, China.
[21] S. A. Weng. 1998. The energy saving and economic
performance of distribution transformer capacity choice.
Low Voltage Apparatus 40 (2) : 55-57.
[22] T. T. Li. 2010. Research on energy-saving and loss-
decreasing of 10kv distribution network. Masteral
Dissertation: South China University of Technology,
Guangzhou, China.
[23] J. Yang and X. Q. Ding. 2006. Practical economic
operation method of two-winding transformer. Electric
Power Automation Equipment 26 (2) : 40-43.
[24] J. H. Wang and N. Gu. 2002. Study on economic
assignment loads of transformers. Energy engineering 22
(3): 15-16.
[25] J. S. Hu. 1981. Economic assignment loads of
transformers. Industry and mine automation 9 (4) : 46-50.
[26] G. J. Song. 2010. Distribution transformer reactive power
analysis and energy saving measures. China Plant
Engineering 36 (4) : 59-60.
[27] Y. L. Dong, J. Wu, N. C. Wang, et al. 2003. Summary of
reactive power compensation technique. Energy
Conservation 23 (9): 13-15+19.
[28] L. Su, S. Song and J. Y. Chen. 2004. The newest projects
of SVC implemented worldwide. International Electric
Power for China 8 (1) : 44-49.
[29] J. J. Liu. 2012. The research and design of static var
generator (SVG). Masteral Dissertation: Tianjin University
of Technology, Tianjin, China.
[30] M. Hu, and Y. Chen. 2000. Active power filter
technology and its application. Automation of Electric
Power Systems 24 (3): 66-70.
[31] Q. Xiong. 2011. Study on distribution transformer and
static reactive power compensator integration technology.
Masteral Dissertation: Huazhong University of Science and
Technology, Wuhan, China.
[32] R. R. Ning, M. A. Tang, and Z. C. Du. 2010. Fuzzy
evaluation on the universities’ building energy-saving
design. Mathematical Practice and Thinking 40 (17) : 66-
71.
[33] C. Chen, H. Zhao, and P. H. Wang. 2009. Multi-factors
evaluation on building energy consumption of public
institution. Shanghai Energy Conservation 16 (2): 20-24.
[34] Canadian Association of Electrical Engineering and
Operation Branch Static Compensator Committee, wrote, Q.
Liu, translated. 1989. Static compensators applied for
power system reactive power control, Water Resources and
Electric Power Press, Beijing, China.
[35] Z. A. Wang. 2009. Harmonic suppression and reactive
power compensation. China Machine Press, Beijing, China.
[36] L. F. Cheng. 2015. Design and research on key
technology of intelligent energy-saving and diagnosis
122
device based on dedicated users. Masteral Dissertation:
South China University of Technology, Guangzhou, China.
[37] T. J. E. Miller. 1990. Power system reactive power
control. Water Resources and Electric Power Press, Beijing,
China.
[38] Z. X. Han. 1993. Power system automatic control.
Zhejiang University Press, Hangzhou, China.
[39] L. F. Cheng, B. Zhou, and T. Yu. 2014. Design and
implementation of energy-saving potential automatically
detecting online and rapid energy audit intelligent system.
Power System Protection and Control 42 (14) : 105-111.
[40] Z. L. Dou and F. Q. Du. 2015. Research on the energy-
saving strategies and the methods for power system. The
Journal of New Industrialization 5 (10): 1-6.
[41] Z. X. Han, and G. Y. Wu. 1994. Power system analysis.
Water Resources and Electric Power Press, Beijing, China.
[42] F. Z. Peng, G. W. Ott, and D. J. Adams. 1998. Harmonic
and reactive power compensation based on the generalized
instantaneous reactive power theory for three-phase four-
wire systems. IEEE Transactions on Power Electronics 13
(6): 1174-1181.
[43] Z. Shu, S. Xie, and Q. Li. 2011. Single-phase back-to-
back converter for active power balancing, reactive power
compensation, and harmonic filtering in traction power
system. IEEE Transactions on Power Electronics 26 (2) :
334-343.
[44] W. M. Wu, J. Lu, and M. Tan. 2015. Multi-regional
reactive power optimization based on correlated equilibrium
Q-learning collaborative algorithm. The Journal of New
Industrialization 5 (6): 33-40.
[45] B. Singh, K. A. Haddad, and A. Chandra. 1998. A new
control approach to three-phase active filter for harmonics
and reactive power compensation. IEEE Transactions on
Power Systems 13 (1) : 133-138.
[46] L. L. Zhang, T. Yu, J. Zeng, et al. 2010. Transformer
energy-saving benefits accurate measurement method and
example analysis based on actual measured data.
Electrotechnical Application 29 (22): 44-47.
123