Template for an Acta IMEKO event paper

ACTA IMEKO
ISSN: 2221-870X
December 2017, Volume 6, Number 4, 80-88

ACTA IMEKO | www.imeko.org December 2017 | Volume 6 | Number 4 | 80

Simple methods of voltage dip tracking – case study
Tomasz Tarasiuk

Gdynia Maritime University, Morska 81-87, 81-225 Gdynia, Poland

Section: RESEARCH PAPER

Keywords: voltage short and long term variations, low-cost measurement, low-pass filtering

Citation: Tomasz Tarasiuk, Simple methods of voltage dip tracking–case study, Acta IMEKO, vol. 6, no. 4, article 13, December 2017, identifier: IMEKO-ACTA-
06 (2017)-04-13

Section Editor: Konrad Jedrzejewski, Warsaw University of Technology, Poland

Received January 26, 2016; In final form November 15, 2017; Published December 2017

Copyright: © 2017 IMEKO. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited

Funding: Ministry of Science and Higher Education under Grant DS/430/2017

Corresponding author: Tomasz Tarasiuk, e-mail: t.tarasiuk@we.am.gdynia.pl

1. INTRODUCTION
The problem of power quality in electric power networks is

one of the hottest topics in electrical power engineering over
recent years. Proliferation of non-linear loads and renewable
energy resources have led to notorious voltage disturbances,
like voltage and current waveform distortions, voltage dips etc.
The intelligent metering and monitoring systems are necessary
[1] to monitor power flow and various voltage and current
parameters in numerous locations thorough the grid. However,
the proper solutions have to be implemented, in order not to
increase the cost of the whole infrastructure. In author’s
opinion, apart from high grade power quality analysers, a
number of low-cost devices will be used, like instruments based
on dedicated integrated circuits with fixed signal processing
algorithms, Application Specific Integrated Circuits (ASICs).
Fortunately, such low-cost ASICs are already available on “the
shelf”, e.g. Analog Devices Single Phase, Multifunction Metering IC
with Neutral Current Measurement ADE7953 [2] or Cirrus Logic
Single phase bi-directional Power/Energy IC CS5461A [3]. They
enable the measurement of voltage and current r.m.s. values,
power and energy as well as dips or swells assessment.
Although they are based on various processing principles, the

common feature of the devices is signal processing executed by
fixed function digital signal processor (DSP) [2]. Among the
applied algorithms, the input signal filtrations are used for
different aims like: elimination of input channels offset, zero-
crossing detection or r.m.s. values of the signals and active
power measurements [2].

This paper is focused solely on the singular feature of these
ICs, namely capacity of voltage dip or swell detection. There are
several methods of dip detection and characterization, starting
from simple solutions as: voltage peak value monitoring [4]-[8]
or the traditional voltage r.m.s. value calculation [4]-[8], up to
more complex solutions based on d-q transformation [4], [9],
short time Fourier transform [5], wavelet transform [4], [5], [8],
[10], [11], Kalman filtering [4], [5], [12] or a combination of
both wavelet and Kalman filtering [5], to name the most
popular. The comparative study on chosen methods
performance can be found e.g. in [6] and [8]. Unfortunately, the
last group of methods lacks simplicity required for
implementation in simple, low cost and low energy consuming
ASICs due to complex models and/or the necessity of
implementation of a number of conditional and branching
operations. Therefore, the simpler solutions are used in ASICs.

ABSTRACT
The paper presents results of an experimental study of two methods of voltage dip tracking. The first is based on half cycle absolute
peak value monitoring, whereas the second is based on low-pass filtration of squares of voltage samples. Both methods are devised
for application in low-cost integrated circuits, dedicated to power quality monitoring. The two real voltage dips have been considered
for the aim. The results are compared with the reference method recommended in IEC Std. 61000-4-30, based on calculation of the
r.m.s. voltage refreshed each half cycle. Further, the application of the low-pass method for assessment small voltage variations is
considered, both short term (r.m.s. voltage refreshed each half cycle) and long term (r.m.s. voltage calculated over 10 cycles of voltage
fundamental component). The research confirmed sufficient accuracy of the method based on low-pass filtering for the class A of
measurement.

ACTA IMEKO | www.imeko.org December 2017 | Volume 6 | Number 4 | 81

For instance, the manufactures of above mentioned ASICs [2],
[3] use the method based on peak value monitoring. Simply put,
if the magnitude of the instantaneous voltage (its absolute
value) is below the pre-defined threshold value, a dip is
detected. Similarly, the swell is to be identified if the absolute
value of the magnitude of the instantaneous voltage is above
the pre-defined threshold. The method is simple and easy to
implement in ASICs. But the downside of the solution is a
notorious impact of noise and distortions of the voltage on the
outcome of the dip or swell identification [4], [8]. This can be
avoided by superseding this method by an equally simple one
based on low-pass filtering of the squares of the input voltage
samples, like the method already used by some manufacturers
for the r.m.s. value of voltage and current measurements [2],
although not used for dip and swell identification. Therefore,
the paper aim is to explore the possibility of using this solution
in its simplest form, which can be easily applied in ASICs, for
dip and swell identification. It is compared with the above
mentioned method based on the absolute value of
instantaneous voltage monitoring. Both solutions are evaluated
against the reference measurement procedure laid in IEC
Standard 61000-4-30 for Class A measurements, based on the
measurement of the r.m.s. voltage over 1 cycle, commencing at
a fundamental zero crossing, and refreshed each half-cycle
Urms(1/2) [13]. This is recommended for voltage dips, swell and
interruption detection and evaluation [14]. It has to be added
that some documents use the term sag as synonym to the term
dip [2], [3], [14]. But since the term dip is used by IEC, it will be
used consequently thorough this paper.

Finally, the paper presents a comparative study of the
performance of these two methods. The first method is based
on monitoring the voltage peak value. The second method is
based on the r.m.s. voltage value determined by low pass
filtering. Because the idea behind the paper is to use real cases
for the method’s performance assessment, results obtained by
both solutions are compared against quantity values provided
by the reference measurement procedure to assess the
measurement trueness of the measured quantity values obtained
from analysed procedures [15]. The reference procedure is
described in detail in Section 2. This section includes also basic
information about limitations of the standard procedure. All
subsequent analyses of experimental data are based on two
voltage dips registered in real microgrids, namely the network
of a sensitive data centre during its island operation mode as
well as a marine microgrid, a network of a ferry, during switch-
on of the high power electric motor (for driving the bow
thruster). Further, another marine network with high distortion
(voltage THD=11.8 %) and voltage and frequency modulation
was considered for determining the method’s performance for
tracking short-term voltage variations. The paper is the
extended and updated version of the work presented during the
XXI IMEKO World Congress, 2015, Prague, Czech Republic
[16].

2. STANDARD FRAMEWORK – REFERENCE MEASUREMENT
PROCEDURE

The voltage dip is defined as the sudden decrease in r.m.s.
value of the supply voltage to a value between 10 % and 90 %
of the declared voltage (in another standard it is 1 % to 90 %
[17]) for durations from 0.5 cycles to 1 min [14]. A voltage dip
is to be described by a pair of data: residual voltage (sometimes
dip’s depth) and its duration [13]. The residual voltage is the

lowest r.m.s. value of the voltage during the considered event,
whereas duration is the time difference between the start time
(falling of r.m.s. voltage below the dip threshold) and end the
time (increase of r.m.s. voltage above the dip threshold plus
hysteresis typically equal to 2 % of declared voltage) [13]. In
some cases, the voltage dip is followed by a voltage peak (small
swell), e.g. during a asynchronous motor start up in the
microgrid [18]. Typically, a voltage swell is monitored by the
same methods like a voltage dip, e.g. voltage instantaneous
values [2] or r.m.s. values [13].

It was mentioned above that for Class A measurements the
considered r.m.s. voltage should be calculated over 1 cycle,
commencing at a fundamental zero crossing, and refreshed
each half-cycle. It is designated as Urms(1/2) [13] and it includes all
components like: harmonics, interharmonics, etc. [13]. It has to
be added that Class S has been defined at the IEC Standard
61000-4-30 as well. For this class the dip assessment is to be
carried out in a similar way like the above described but the
voltage r.m.s. value can be refreshed each cycle. Manufacturers
of measuring instrument should specify which method is used
[13].

Finally, the r.m.s. value is to be calculated as square root of
the mean value of squares of the voltage samples registered
over the considered time interval, like required in IEC Standard
61557-12 [19]. The evaluation of a real dip carried out by the
method devised for Class A is used in this paper as a reference
measurement procedure for evaluations of the other methods,
including whole voltage shape assessment during the
considered processes, namely bulk load startup in two
investigated microgrids and resulted dips. Next, it is compared
with the same shape determined by the other simple method
under investigation based on monitoring the voltage peak value.

It must be firmly stressed that the standard approach is
contested by some authors. For instance, in [6], the detailed
analysis of the standard method’s performance for short dips
(duration below two cycles) was presented. Authors of [6]
pointed out the ambiguity in determining the dip parameters,
both duration and residual voltage, by the standard solution for
such a short duration phenomenon. This is chiefly related to
synchronization based on fundamental zero crossing [6], [8].
The proposed solution can be based on determining the rms
value of the voltage over a one-cycle sliding window and
computing the rms value at every sampling point [6], [7].
However, the solution entails higher computational burden for
ensuring that the window length fits the actual fundamental
period [6]. Another frequently investigated solution is tracking
the energy of the band 0-100 Hz after wavelet transformation
for detection of short duration dips [10] but this also entails an
increase in computational complexity.

Next, the standard method has another limitation, namely it
“does not provide information about the phase angle of voltage
supply during the event, which could be of interest for some
applications” [8]. Reference [20] lists seven dip characterization
methods, including four which require phase angle information.

Despite the above mentioned drawbacks of the standard
method, it arguably remains one of the most popular due to its
simplicity and sufficient performance for a number of most
common applications.

3. THE METHODS UNDER INVESTIGATION
Notwithstanding its simplicity, the application of the above

described standard procedure for dip detection and evaluation

ACTA IMEKO | www.imeko.org December 2017 | Volume 6 | Number 4 | 82

can be inconvenient in simple low-cost devices, due to some
requirements for hardware resources of the measuring
instrument. In shorthand, it requires storing of voltage samples
for at least one cycle, conditional operation, some data address
generating, etc. Obviously, it can be easily implemented in
digital signal processors (DSPs) but not necessary in low-cost
dedicated ASICs. Therefore, manufacturers of low-cost
measuring devices implement other principles of dip detection
and evaluation [2], [3]. It is permissible for Class B
measurements if the manufacturer specifies the method used
for the aim [13]. So, the paper explores the performance of two
arguably simplest methods of voltage dip detection and
characterization, which can be used for less demanding
applications and both easy to apply in ASICs.

3.1. Dip measurement based on voltage peak value tracking
Arguably, the simplest solution of the considered problem is

detection of time instants when the absolute value of the
voltage falls below the programmable threshold [2], [3]. This
feature is easy to implement in ASICs. Simply, the dedicated
peak register is updated with a new instantaneous voltage value
every time that the absolute value of the voltage sample exceeds
the sample value already stored in the register. The register can
be cleared after reading [2], which can be synchronized with the
voltage fundamental component zero-crossing in order to
determine the end of each half-cycle. This feature enables
continuous recording the maximum value of the voltage
waveforms for each half cycle. Apart of its simplicity, the
obvious disadvantage of the solution is the above mentioned
possible impact of voltage distortions or noise [4], [8], since for
distorted signals one cannot conclude on its r.m.s. value basing
only on the maximum value of the voltage waveforms. It will be
proved below that even in the case of low distortion, the results
of dip analysis by the method can differ noticeably from the
reference method.

3.2. Dip measurement based on low-pass filtering of squares of
voltage samples

The standard dip assessment method is based on the
determination of the voltage r.m.s. value over one cycle. It can
be easily noted that

)( 2
1

21
k

k
Nrms uLPFuU k ≈∑⋅=

−

= (1)
where N is the number of voltage samples recorded over an
integer number of cycles (in fact one cycle for dip assessment),
uk is the voltage sample, LPF is a low-pass filter

The solution consists in superseding the mean filter with N
coefficients (each with value of 1/N) by another LPF filter.
Since N varies depending on the instantaneous frequency it
seems much easier to implement in ASICs a low-pass filter with
a constant number of coefficients. Because of the ripples of the
LPF output, the reading should be synchronised with the
voltage fundamental component zero-crossing. It is determined
after low-pass filtering of the voltage samples by another LPF.
The algorithm [2] after some modifications is shown in Figure
1. It should be added that the solution is applied using an
ADE7953 [2] for the r.m.s. measurement but not for dip
detection and evaluation.

The signal processing path shown in Figure 1 can be used
for both: measuring the r.m.s. value of the voltage during
steady-state as well as dip monitoring. The reading is to be
carried out after each zero-crossing of the voltage fundamental

component, similar as in the case of the Urms(1/2) measurement.
This value is used directly to the dip or swell detection and
evaluation. Namely, it is to be compared with the assumed
threshold level, once again similar as in the case of the standard
method. In order to obtain a 10-cycles rms voltage,
accumulation of Urms(1/2) and averaging is necessary. The block
“delay” shown in Figure 1 is to account for the LPF1 and LPF2
characteristics (group delays of both filters). It means shifting of
the samples of the LPF2 output. It depends on the actual
power frequency. So, adaptation of the shift is to be performed.
A simpler solution is to use constant sample shifts related to the
rated power frequency but it can impact the accuracy of the
proposed solution if used in islanded microgrids due to possible
power frequency changes.

4. RESULTS OF EXPERIMENTAL RESEARCH
The real examples are used for this paper’s purpose. So, the

research consisted in voltage samples registration in real
networks and subsequent processing by various signal
processing methods. A National Instruments controller PXIe-
8106 equipped with two data acquisition boards PXIe-6124 was
used for voltage samples recording in an office building. The
analog input channel consisted of CV3-1500 LEM voltage
transducers and LTC 1564 anti-aliasing filters. The cut-off
frequency of the anti-aliasing filters was equal to 10 kHz. In the
case of marine systems, data acquisition board PCI703-16/A
Eagle Technology, a resistive voltage divider and isolation
amplifiers ISO 124 of Burr Brown were used. The cut- off
frequency of the anti-aliasing filter was equal to 3.5 kHz.

Figure 1. Block diagram of the signal processing path for the measurement
of r.m.s. voltage, including dip detection and evaluation.

ACTA IMEKO | www.imeko.org December 2017 | Volume 6 | Number 4 | 83

Finally, the momentary voltages for dip detection are
calculated by both investigated methods. The results obtained
by analysis of the voltage local peak values (like used in [2] and
[3]) are designated as UABS. The recorded absolute values of the
voltage waveform are divided by √2 in order to obtain the
voltage r.m.s. value and subsequently to compare with the
reference method. The results calculated by voltage samples
squaring and low-pass filtering of the result are designated as
ULPF. It represents the r.m.s. voltage momentary values. For the
paper purpose, a third order Butterworth filter is used with
various cut-off frequencies. It has been mentioned (see Figure
1) that, in order to diminish the effect of ripples of the filter
output, its reading is to be synchronised with the fundamental
component zero-crossing. Zero-crossing of the fundamental
component of the voltage is determined after low-pass filtering
by another third order Butterworth filter, but with higher cut-
off frequency equal to 80 Hz. Such a solution is recommended
in the IEC 61000-4-30 standard for diminishing the impact of
higher frequency components. The cut-off frequency is chosen
after [2]. To compare the differences between the considered
methods and the reference method, the square root of the
mean value of squares of the differences is calculated by:

∑ −=
−

2
21

1 N

k
rmsmeth kk

UU
N

diffSQR )()( )/(
(2)

where methU is the method under investigation (UABS or ULPF),
N is the number of considered half cycles for the dip
assessment. It is assumed that N=146 starting at the dip
beginning.

4.1. Dip in emergency power system of office building
The voltage recording took place in the network of an office

building that contained a very important data centre and
sensitive for power quality disturbances. Particularly,
interruptions can lead to severe consequences. Therefore, it was
equipped with two UPSs and two generators driven by diesel
engines for power backup. The network rated voltage was equal
to 230 V and the rated frequency 50 Hz. The whole research
was carried out during the object island operation mode, due to
suspicion of power quality problems during the mode. Various
parameters of voltage and currents were analysed in various
points of the system. During the investigation the process of
switching-on the bulk load was recorded. The waveform of the
recorded voltage is presented in Figure 2.

It is easily discernible that the process of switching bulk load
on in the power network of the data centre causes a voltage dip
up to 85 % of the rated voltage followed by a small voltage
swell. It is only 101.6 % of the rated voltage but approximately
above 5 % of the registered mean steady-state voltage, which
was approximately equal to 96.8 % of the rated voltage (230 V).
Nevertheless, this phenomenon is considered as well, in order
to properly asses the methods under investigation. The details
of Urms(1/2) calculated according to the IEC 61000-4-30 standard
and the resulting voltage shape are shown in Figure 3.

Obviously, the processes of switching bulk load on in
microgrids cause voltage changes and concurrent momentary
frequency changes. For the considered example, the lowest
momentary frequency understood as reciprocal of fundamental
cycle is equal to 48.61 Hz followed by a frequency increase up
51.51 Hz. Typically, standards related to marine microgrids, e.g.
[21], [22] deal with the phenomenon, but it has not been
analysed for the paper’s aim. However, both investigated

methods, which depend to some respect on the fundamental
component zero-crossing, enable concurrent assessment of
momentary frequency changes, including their value and
duration. In fact, assessment of rapid power frequency changes
in marine systems is similar as in the case of voltage dip
assessment [21], [22].

Finally, the results of the calculation of parameter pairs for
the dip assessment (residue voltage Umin and duration),
completed by the r.m.s. value of the mini-swell Umax voltage that
followed the considered dip are shown in Table 1. Moreover,
results of determining SQR(diff) for each of the methods are
presented in the table as well.

Analysis of results in Table 1 leads to the conclusion that
results of dip shape assessment for filters with cut-off
frequencies from 15 Hz up to 19 Hz lead to comparable results.
Observed minor differences can be neglected. The lowest
SQR(diff) value is obtained for a filter with cut-off frequency 17
Hz. The analysis confirms that the UABS method leads to worse
results than the method based on low-pass filtering of the
squares of voltage samples for most of the used filters.
Next, the capability of tracking the voltage shape for each of
the methods is assessed. It is carried out by determining the
differences between ULPF and Urms(1/2) (they are graphically
presented in Figure 4 for LPF2 with cut-off frequency equal to
17 Hz) and the differences between UABS and Urms(1/2). The
results of comparison are shown in Figure 5. SQR(diff) values
for these examples are given in Table 1 as well.

Figure 2. Voltage waveform during switching bulk load on in a power
network of data centre during island operation.

Figure 3. Variations of Urms(1/2) value (reference method) during switching
bulk load on in a power network of a data centre during island operation;
residue voltage equals 195.86 V (85.16 % of Urated), dip duration 18 half
cycles, swell equals 233.58 V.

-400

-300

-200

-100

100

200

300

400

time [ms]

u(
t)

[V
]

800 0

190

200

210

220

230

240

time [s]

U
rm

s(
1/

2)
[V

]

4 0

ACTA IMEKO | www.imeko.org December 2017 | Volume 6 | Number 4 | 84

Comparison of Figures 4 and 5 once again reveals clear
superiority of the first solution, i.e. ULPF. The difference
between the considered method and the reference method does
not exceed 1.2 V. Maximal values are observed for the dip
beginning. Similar differences for the UABS method reach values
above 4 V. The reason is distortion of the investigated voltage,
although with a relatively low level. The value of the voltage
THD was equal to 3.13 % prior to the considered dip and 3.26
% after the dip. It can be noted that even this small increase
leads to a significant increase in differences between UABS and
Urms(1/2) (see Figure 5). The differences increase nearly two times

under steady-state conditions: before and after dip. The main
reason is the increase of the 5th harmonic content by 0.19 % of
the fundamental component, completed by a phase shift of
some harmonics in relation to the fundamental component
after the dip and a resulting increase in the voltage maximal
instantaneous value.

4.2. Dip in power system of a ferry
The second example concerns the dip caused by switching

on the high power electric motor with rated power 1.72 MW on
board a ferry with a 6.6 kV - 60 Hz system. The motor start
prior to the ship manoeuvring has led to a severe voltage dip,
namely the voltage dropped below the permissible limit -20 %
of the rated voltage [21], [22].

Similar as in the previous example, the dip parameters are
determined by the two investigated methods. The waveform of
the recorded voltage is presented in Figure 6 and the details of
Urms(1/2) calculated according to the IEC 61000-4-30 standard
and the resulting voltage shapes are shown in Figure 7. The
calculation results of the parameter pairs for the dip assessment
(residue voltage Umin and duration), completed by the r.m.s.
voltage of the mini-swell Umax that followed the considered dip
are shown in Table 2 and the differences between ULPF and
Urms(1/2) and the differences between UABS and Urms(1/2) are given
in Figures 8 and 9, respectively

Once again, analysis of the above results leads to the
conclusion that application of LPF filters with cut-off
frequencies 12-20 Hz gives satisfactory results. The solution
superiority over UABS method is clearly visible, despite the fact
that for the ferry voltage THD hardly exceeded 1.2 %.

Table 1. Calculation results of dip (duration and Umin) and swell (Umax)
parameters completed by SQR(diff).

fcutoff half
cycl.

Umin Umax SQR(diff)

[V] [%] [V] [V]

Urms(1/2) 18 195.86 85.16 233.58 ---

UABS 17 197.6 85.91 236.47 3.26
ULPF 1 Hz --- 212.87 92.55 226.99 6.23

2 Hz 15 203.64 88.52 228.20 3.51

3 Hz 17 198.28 86.20 232.53 1.72

4 Hz 17 196.26 85.33 233.46 1.13

5 Hz 17 195.20 84.87 233.68 0.88

6 Hz 17 194.99 84.77 233.69 0.85

7 Hz 18 194.96 84.76 233.72 0.68

8 Hz 17 195.16 84.85 233.71 0.55

9 Hz 18 195.43 84.97 233.66 0.61

10 Hz 17 195.69 85.08 233.69 0.53

11 Hz 18 195.87 85.16 233.67 0.36

12 Hz 18 195.98 85.20 233.66 0.45

13 Hz 18 196.00 85.20 233.67 0.64

14 Hz 18 195.93 85.17 233.67 0.51

15 Hz 18 195.87 85.16 233.65 0.35

16 Hz 18 195.83 85.14 233.62 0.23

17 Hz 18 195.82 85.14 233.59 0.20

18 Hz 18 195.83 85.14 233.58 0.25

19 Hz 18 195.87 85.16 233.59 0.34

20 Hz 18 195.87 85.16 233.59 0.88

Figure 4. The differences between r.m.s. value of voltage calculated by
filtering of square samples ULPF (LPF2 cur-off frequency=17 HZ) and
reference method Urms(1/2).

Figure 5. The differences between rms value of the voltage calculated on
the basis of the absolute peak value of voltage UABS and reference Urms(1/2).

Figure 6. Voltage waveform during switch-on of the thruster in power
network of ferry.

ACTA IMEKO | www.imeko.org December 2017 | Volume 6 | Number 4 | 85

4.3. Voltage variations tracking in the system with high
level of voltage distortion

The voltage registered on board of a chemical tanker was
chosen for the assessment of the performance under
significantly distorted conditions. The vessel is equipped with a
shaft generator working on main bus bars via a power converter
to obtain a constant frequency. The system rated voltage was
440 V and rated frequency 60 Hz. The voltage waveform
registered in the system of the chemical tanker is shown in
Figure 10. For the assessment, the measurement results of the

voltage modulation by low-pass filtering ULPF and by the Urms(1/2)
method were compared. It is to exemplify the capability of
tracking of the voltage variations by the ULPF method. It must
be stressed that the considered case is very hard to analyze
because of concurrent presence of significant waveform
distortions as well as fundamental voltage and frequency
modulations shown in Figure 11. The instantaneous frequency
and voltage for Figure 11 are calculated by a zoom-DFT with
Kaiser window, refreshed every 0.5 ms and with a frequency
resolution 0.001 Hz. For this case, a similar modulation was
observed for harmonics, e.g. the r.m.s. of the 5th harmonic
varied between 15 and 25 V.

Figure 8. The differences between the r.m.s. value of voltage calculated by
filtering of square samples ULPF (LPF2 cur-off frequency=17 Hz) and
reference method Urms(1/2).

Figure 9. The differences between the rms value of the voltage calculated
on the basis of the absolute peak value of voltage UABS and reference
Urms(1/2).

Figure 10. Voltage waveform in a marine system with a shat generator
working via a power converter, THD=11.8 %.

Figure 7. Variations of Urms(1/2) (reference method) during switching of the
thruster onboard a ferry; residue voltage equals 5015 V (75.98 % of Urated),
dip duration equals to 51 half cycles, swell equals to 7083.1 V.

Table 2. Results of calculation of dip (duration and Umin) and swell (Umax)
parameters completed by SQR(diff).
fcutoff half

cycl.
Umin Umax SQR(diff

)

[V] [%] [V] [V]

Urms(1/2) 51 5015.0 75.98 7083.1 ---

UABS 49 5211.6 78.96 7299.6 170.03

ULPF 1 Hz 52 5554.2 84.15 6671.4 326.89

2 Hz 48 4990.2 75.61 6958.7 146.99

3 Hz 49 4884.4 74.01 7083 108.95

4 Hz 49 4891.1 74.11 7105.6 87.96

5 Hz 50 4918.0 74.52 7111.5 76.13

6 Hz 50 4926.7 74.65 7109.9 64.24

7 Hz 50 4929.8 74.69 7105.7 55.97

8 Hz 50 4938.7 74.83 7100.5 51.63

9 Hz 51 4953.1 75.05 7093.8 46.56

10 Hz 52 4974.3 75.38 7086.9 42.23

11 Hz 51 4993.7 75.66 7080.4 38.26

12 Hz 51 5016.9 76.01 7074.7 42.65

13 Hz 51 5015.5 75.99 7083.1 33.88

14 Hz 51 5014.9 75.98 7083.1 28.86

15 Hz 51 5013 75.95 7079.6 28.70

16 Hz 52 5012.8 75.95 7080.8 32.15

17 Hz 51 5014.3 75.97 7082.0 38.19

18 Hz 51 5013.6 75.96 7083.3 28.18

19 Hz 51 5013.6 75.96 7084.2 22.58

20 Hz 51 5014.8 75.98 7085.9 18,43

ACTA IMEKO | www.imeko.org December 2017 | Volume 6 | Number 4 | 86

The two voltage tracking methods are graphically compared
in Figure 12, for a steady-state condition. Assumed cut-off
frequency of LPF2 filter is equal to 17 Hz.

Analysis of Figure 12 once again indirectly proves that the
ULPF method would be appropriate for the considered aim,
namely tracking of voltage short-term variations, including
voltage dip or swell detection and evaluation, even in the case
of heavily distorted signals.

Finally, it has to be added that the r.m.s. value of voltage
over 10 or 12 cycles time interval (depending on the power
system’s rated frequency) should be calculated for the
assessment of magnitude of the supply voltage (apart from dips
and swells detection) [13]. Because it would be unwise to design
a separate signal processing path for the 10 or 12 cycles r.m.s.
value determination, it seems that the simplest solution would

be determining the average value of readings of the LPF2 filter
over a specified number of cycles. It can easily be implemented
in ICs. To assess the accuracy of the solution, the average
values of twenty or twenty four consecutive readings of the
LPF2 filter with cut-off frequency equal to 17 Hz is determined
for the office building (50 Hz system) and the chemical tanker
(60 Hz system), respectively. They are graphically (Figure 13)
compared with respective r.m.s. values determined for the very
same time interval (exactly 10 cycles) by the reference method.

The analysis of the results presented in Figures 4, 8, 12 and
13 clearly indicates that the same signal processing path can be
easily used for assessing voltage dips and swells as well as
determining 10/12 cycles voltage magnitude after simple
averaging. For example, the maximum difference between
reference 10/12 cycle r.m.s. value and the respective value
determined by averaging the LPF2 readings (cut-off frequency
17 Hz) is below 0.02 V for the case presented in Figure 13(a)
and it is below 0.5 V for the case presented in Figure 13(b). The
value of SQR(diff) is below 0.01 V for the former case and 0.3 V
for the latter case.

5. CONCLUSIONS
The paper aim was the investigation of some simple

algorithms for dip or swell detection and evaluation, easily
applicable in low-cost ICs dedicated to multifunction electricity

Figure 11. Fluctuations of voltage (a) and its frequency (b) on board of
chemical tanker determined by zoom-DFT.

Figure 12. Comparison of tracking capabilities of short-term small voltage
variations by ULPF method with Urms(1/2) method.

Figure 13. Comparison of reference r.m.s. value calculated over 10 or 12
cycles of input voltage and respective value of averaged LPF2 readings over
the same time interval, (a) emergency supply of the office building, (b) bow
thruster subsystem with non-linear load.

ACTA IMEKO | www.imeko.org December 2017 | Volume 6 | Number 4 | 87

parameters measurements. It was proved that the method of
signal processing based on low-pass filtering of the input
voltage samples is more suitable for the aim than the commonly
used method based on absolute values of the momentary
voltage peak. In the latter case (UABS method) the significant
impact of voltage distortions on the measurement accuracy can
be observed. It is true even for slightly distorted signals, which
are rather norm than exception in nowadays power systems.

Therefore, using the method based on low-pass filtering is
arguably a better solution. Since it is already used for r.m.s.
value estimation by the same ICs, it requires only some design
modification , but special attention has to be paid to the cut-off
frequency and group delay of implemented LPFs. The used cut-
off frequency has to be increased in comparison with that
currently used in ICs [2] but it inevitably leads to an increase in
ripples of the filter output. Fortunately, the ripples impact is
limited, if synchronisation of reading with voltage zero
crossings is implemented. The resulting overall accuracy is
better than in the case of the hitherto used solution, based on
tracking of the absolute instantaneous value of the input
voltage. Moreover, the reading of the used LPF2 filters can be
used for determining the 10-cycles voltage magnitude with
good accuracy after simple averaging. It simplifies the IC
design, since the same signal processing path can be used for
both aims, dip detection and the 10-cycles voltage magnitude
measurement. Currently it is determined by two separate signal
processing paths, in dedicated ICs with fixed DSP.

It has to be added that the solution based on low-pass
filtering was devised for low-cost and low power consumption
applications, with sufficient accuracy for trouble-shooting
applications, like defined in the IEC Standard 61000-4-30 [13].
Its application for contractual purposes will require additional
research, particularly for other voltage shapes during dips.

Next, the cut-off frequency of the applied LPF2 filter has to
be carefully chosen, since it can affect the uncertainty of the
method. If it is too low, it results in an increase in response
time and the assessed residue voltage would be higher than the
real one. This can affect the accuracy of determining the residue
voltage during very short dips. However, a too high cut-off
frequency would lead to an increase of ripples of the filter
output and increase in the measurement uncertainty since
reading exactly at the moments of zero-crossing is hardly
possible. In fact, the results are affected by both LPF1 and
LPF2 characteristics. Their correction can assume discrete
values with a resolution equal to the sampling period, since
delay means shifting samples of the LPF2 output. So, the
accuracy of correction of the group delay for the respective
filters is in the range of ±0.5 Ts (Ts – sampling period, which for
the considered research equals 40 µs for the office building and
95 µs for the two later cases, namely the marine microgrids).
The remaining factors influencing measurement uncertainty are
the same as considered for typical performance monitoring and
measuring devices and discussed in standard [19]. Particularly,
the limitation of the method based on low-pass filtering is the
necessity of synchronisation of the reading with the
fundamental zero crossing. However, this is also a limitation of
the solution based on the peak value, which can be determined
once every half cycle.

Nevertheless, the research carried out for two various dip
cases and completed by research of modulated, significantly
distorted voltages (network of chemical tanker) proved to have
good accuracy of the proposed solution, even under such
unfortunate circumstances. The maximum difference between

the reference method (for 12 cycles time interval) and the
described proposal hardly exceeded 0.11 % of the voltage rated
value.

Summing up, in author’s opinion, the proposed solution
based on a simple low-pass filtering gives acceptable accuracy
for less demanding applications and can be used for everyday
monitoring of power systems performance. However, for more
demanding or legal purposes other, more complex solutions,
for instance mentioned in the introduction, should be
considered.

REFERENCES

[1] EUROPEAN COMMISSION, “Communication From the
Commission to the European Parliament, the Council, the
European Economic and Social Committee and the Committee
of the Regions, Smart Grids: from innovation to deployment.”
COM(2011) 202 final, Brussels, 12.4.2011.

[2] Analog Devices, “Single Phase, Multifunction Metering IC with
Neutral Current Measurement ADE7953”, Data Sheet, 2011-
2013.

[3] Cirrus Logic, “Single phase bi-directional Power/Energy IC
CS5461A”, Data Sheet, 2011.

[4] A. Khoshkbar Sadigh, K.M. Smedley, “Fast and precise voltage
sag detection method for dynamic voltage restorer (DVR)
application”, Electric Power Systems Research, vol. 130, 2016,
pp. 192–207.

[5] E. Perez, J. Barros, “A Proposal for On-Line Detection and
Classification of Voltage Events in Power Systems”, IEEE
Trans. on Power Delivery vol. 23, no. 4, 2008, pp. 2132-2138.

[6] D. Gallo, C. Landi, M. Luiso, E. Fiorucci, “Survey on voltage dip
measurements in standard framework”, IEEE Trans. on
Instrumentation and Measurement, vol. 63, no. 2, 2014, pp. 374-
387.

[7] M. Bollen, I. Gu, “Signal processing of power quality
disturbances”, New York, USA, Wiley, 2006.

[8] A. Moschitta, P. Carbone, C. Muscas, “Performance Comparison
of Advanced Techniques for Voltage Dip Detection”, IEEE
Trans. on Instrumentation and Measurement, vol. 61, no. 5,
2012, pp. 1494-1502.

[9] O. Hua, B. Le-ping, Y. Zhong-lin, ”Voltage Sag Detection Based
on dq Transform and Mathematical Morphology Filter”,
Procedia Engineering, vol. 23, 2011, pp. 775 – 779.

[10] M. Apraiz, J. Barros, R. Diego, “A real-time method for time–
frequency detection of transient disturbances in voltage supply
systems”, Electric Power Systems Research, vol. 108, 2014, pp.
103– 112.

[11] J. Decanini, M. Tonelli-Neto, F. Malange, C. Minussi, “Detection
and classification of voltage disturbances using a Fuzzy-
ARTMAP-wavelet network”, Electric Power Systems Research,
vol. 81, 2011, pp. 2057– 2065.

[12] E. Pérez, J. Barros, “An extended Kalman filtering approach for
detection and analysis of voltage dips in power systems”, Electric
Power Systems Research, vol. 78, 2008, pp. 618–625.

[13] IEC Standard 61000-4-30, “Testing and Measurement
Techniques – Power Quality Measurement Methods”, 2015.

[14] IEEE Standard 1159-2009, “IEEE Recommended Practice for
Monitoring Electric Power Quality”, 2009.

[15] Joint Committee for Guides in Metrology “International
vocabulary of metrology – Basic and general concepts and
associated terms,” 3rd edition, , JCGM 200:2012.

[16] T. Tarasiuk, “Comparative study on chosen methods of voltage
dip tracking based on real example”, XXI IMEKO World
Congress “Measurement in Research and Industry” August 30 −
September 4, 2015, Prague, Czech Republic.

[17] EN Standard 50160, “Voltage characteristics of electricity
supplied by public distribution networks”, 2007.

[18] A. Vicenzutti, D. Bosich, G. Giadrossii, G. Sulligoi, “The role of
voltage controls in modern all-electric ships”, IEEE

ACTA IMEKO | www.imeko.org December 2017 | Volume 6 | Number 4 | 88

Electrification Magazine, vol. 3, no. 2, 2015, pp. 49-65.
[19] IEC Standard 61557-12, “Electrical safety in low voltage

distribution systems up tp 1000 V a.c. and 1500 V d.c. –
Equipment for testing, measuring or monitoring of protective
measures – Part 12: Performance measuring and monitoring
devices (PMD)”, 2007.

[20] Y. Wang, M. Bollen, A. Bagheri, X. Xiao, M. Olofsson, “A
quantitative comparison approach for different voltage dip

characterization methods”, Electric Power Systems Research,
vol. 133, 2016, pp. 182–190.

[21] DET NORSKE VERITAS, “Rules for classification of ships /
high speed, light craft and naval surface craft. Electrical
installations”, Part 4, Chapter 8, 2011.

[22] IEEE Std. 45-2002, “IEEE Recommended Practice for
Electrical Installations on Shipboard”, 2002.

Simple methods of voltage dip tracking – case study

<<
/ASCII85EncodePages false
/AllowTransparency false
/AutoPositionEPSFiles true
/AutoRotatePages /None
/Binding /Left
/CalGrayProfile (Dot Gain 20%)
/CalRGBProfile (sRGB IEC61966-2.1)
/CalCMYKProfile (U.S. Web Coated \050SWOP\051 v2)
/sRGBProfile (sRGB IEC61966-2.1)
/CannotEmbedFontPolicy /Error
/CompatibilityLevel 1.4
/CompressObjects /Tags
/CompressPages true
/ConvertImagesToIndexed true
/PassThroughJPEGImages true
/CreateJobTicket false
/DefaultRenderingIntent /Default
/DetectBlends true
/DetectCurves 0.0000
/ColorConversionStrategy /CMYK
/DoThumbnails false
/EmbedAllFonts true
/EmbedOpenType false
/ParseICCProfilesInComments true
/EmbedJobOptions true
/DSCReportingLevel 0
/EmitDSCWarnings false
/EndPage -1
/ImageMemory 1048576
/LockDistillerParams false
/MaxSubsetPct 100
/Optimize true
/OPM 1
/ParseDSCComments true
/ParseDSCCommentsForDocInfo true
/PreserveCopyPage true
/PreserveDICMYKValues true
/PreserveEPSInfo true
/PreserveFlatness true
/PreserveHalftoneInfo false
/PreserveOPIComments true
/PreserveOverprintSettings true
/StartPage 1
/SubsetFonts true
/TransferFunctionInfo /Apply
/UCRandBGInfo /Preserve
/UsePrologue false
/ColorSettingsFile ()
/AlwaysEmbed [ true
]
/NeverEmbed [ true
]
/AntiAliasColorImages false
/CropColorImages true
/ColorImageMinResolution 300
/ColorImageMinResolutionPolicy /OK
/DownsampleColorImages true
/ColorImageDownsampleType /Bicubic
/ColorImageResolution 300
/ColorImageDepth -1
/ColorImageMinDownsampleDepth 1
/ColorImageDownsampleThreshold 1.50000
/EncodeColorImages true
/ColorImageFilter /DCTEncode
/AutoFilterColorImages true
/ColorImageAutoFilterStrategy /JPEG
/ColorACSImageDict <<
/QFactor 0.15
/HSamples [1 1 1 1] /VSamples [1 1 1 1]
>>
/ColorImageDict <<
/QFactor 0.15
/HSamples [1 1 1 1] /VSamples [1 1 1 1]
>>
/JPEG2000ColorACSImageDict <<
/TileWidth 256
/TileHeight 256
/Quality 30
>>
/JPEG2000ColorImageDict <<
/TileWidth 256
/TileHeight 256
/Quality 30
>>
/AntiAliasGrayImages false
/CropGrayImages true
/GrayImageMinResolution 300
/GrayImageMinResolutionPolicy /OK
/DownsampleGrayImages true
/GrayImageDownsampleType /Bicubic
/GrayImageResolution 300
/GrayImageDepth -1
/GrayImageMinDownsampleDepth 2
/GrayImageDownsampleThreshold 1.50000
/EncodeGrayImages true
/GrayImageFilter /DCTEncode
/AutoFilterGrayImages true
/GrayImageAutoFilterStrategy /JPEG
/GrayACSImageDict <<
/QFactor 0.15
/HSamples [1 1 1 1] /VSamples [1 1 1 1]
>>
/GrayImageDict <<
/QFactor 0.15
/HSamples [1 1 1 1] /VSamples [1 1 1 1]
>>
/JPEG2000GrayACSImageDict <<
/TileWidth 256
/TileHeight 256
/Quality 30
>>
/JPEG2000GrayImageDict <<
/TileWidth 256
/TileHeight 256
/Quality 30
>>
/AntiAliasMonoImages false
/CropMonoImages true
/MonoImageMinResolution 1200
/MonoImageMinResolutionPolicy /OK
/DownsampleMonoImages true
/MonoImageDownsampleType /Bicubic
/MonoImageResolution 1200
/MonoImageDepth -1
/MonoImageDownsampleThreshold 1.50000
/EncodeMonoImages true
/MonoImageFilter /CCITTFaxEncode
/MonoImageDict <<
/K -1
>>
/AllowPSXObjects false
/CheckCompliance [
/None
]
/PDFX1aCheck false
/PDFX3Check false
/PDFXCompliantPDFOnly false
/PDFXNoTrimBoxError true
/PDFXTrimBoxToMediaBoxOffset [
0.00000
0.00000
0.00000
0.00000
]
/PDFXSetBleedBoxToMediaBox true
/PDFXBleedBoxToTrimBoxOffset [
0.00000
0.00000
0.00000
0.00000
]
/PDFXOutputIntentProfile ()
/PDFXOutputConditionIdentifier ()
/PDFXOutputCondition ()
/PDFXRegistryName ()
/PDFXTrapped /False

/CreateJDFFile false
/Description <<
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
/BGR <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>
/CHS <FEFF4f7f75288fd94e9b8bbe5b9a521b5efa7684002000410064006f006200650020005000440046002065876863900275284e8e9ad88d2891cf76845370524d53705237300260a853ef4ee54f7f75280020004100630072006f0062006100740020548c002000410064006f00620065002000520065006100640065007200200035002e003000204ee553ca66f49ad87248672c676562535f00521b5efa768400200050004400460020658768633002>
/CHT <FEFF4f7f752890194e9b8a2d7f6e5efa7acb7684002000410064006f006200650020005000440046002065874ef69069752865bc9ad854c18cea76845370524d5370523786557406300260a853ef4ee54f7f75280020004100630072006f0062006100740020548c002000410064006f00620065002000520065006100640065007200200035002e003000204ee553ca66f49ad87248672c4f86958b555f5df25efa7acb76840020005000440046002065874ef63002>
/CZE <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>
/DAN <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>
/DEU <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>
/ESP <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>
/ETI <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>
/FRA <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>
/GRE <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>
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
/HRV (Za stvaranje Adobe PDF dokumenata najpogodnijih za visokokvalitetni ispis prije tiskanja koristite ove postavke. Stvoreni PDF dokumenti mogu se otvoriti Acrobat i Adobe Reader 5.0 i kasnijim verzijama.)
/HUN <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>
/ITA <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>
/JPN <FEFF9ad854c18cea306a30d730ea30d730ec30b951fa529b7528002000410064006f0062006500200050004400460020658766f8306e4f5c6210306b4f7f75283057307e305930023053306e8a2d5b9a30674f5c62103055308c305f0020005000440046002030d530a130a430eb306f3001004100630072006f0062006100740020304a30883073002000410064006f00620065002000520065006100640065007200200035002e003000204ee5964d3067958b304f30533068304c3067304d307e305930023053306e8a2d5b9a306b306f30d530a930f330c8306e57cb30818fbc307f304c5fc59808306730593002>
/KOR <FEFFc7740020c124c815c7440020c0acc6a9d558c5ec0020ace0d488c9c80020c2dcd5d80020c778c1c4c5d00020ac00c7a50020c801d569d55c002000410064006f0062006500200050004400460020bb38c11cb97c0020c791c131d569b2c8b2e4002e0020c774b807ac8c0020c791c131b41c00200050004400460020bb38c11cb2940020004100630072006f0062006100740020bc0f002000410064006f00620065002000520065006100640065007200200035002e00300020c774c0c1c5d0c11c0020c5f40020c2180020c788c2b5b2c8b2e4002e>
/LTH <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>
/LVI <FEFF0049007a006d0061006e0074006f006a00690065007400200161006f00730020006900650073007400610074012b006a0075006d00750073002c0020006c0061006900200076006500690064006f00740075002000410064006f00620065002000500044004600200064006f006b0075006d0065006e007400750073002c0020006b006100730020006900720020012b00700061016100690020007000690065006d01130072006f00740069002000610075006700730074006100730020006b00760061006c0069007401010074006500730020007000690072006d007300690065007300700069006501610061006e006100730020006400720075006b00610069002e00200049007a0076006500690064006f006a006900650074002000500044004600200064006f006b0075006d0065006e007400750073002c0020006b006f002000760061007200200061007400760113007200740020006100720020004100630072006f00620061007400200075006e002000410064006f00620065002000520065006100640065007200200035002e0030002c0020006b0101002000610072012b00200074006f0020006a00610075006e0101006b0101006d002000760065007200730069006a0101006d002e>
/NLD (Gebruik deze instellingen om Adobe PDF-documenten te maken die zijn geoptimaliseerd voor prepress-afdrukken van hoge kwaliteit. De gemaakte PDF-documenten kunnen worden geopend met Acrobat en Adobe Reader 5.0 en hoger.)
/NOR <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>
/POL <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>
/PTB <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>
/RUM <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>
/RUS <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>
/SKY <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>
/SLV <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>
/SUO <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>
/SVE <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>
/TUR <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>
/UKR <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>
/ENU (Use these settings to create Adobe PDF documents best suited for high-quality prepress printing. Created PDF documents can be opened with Acrobat and Adobe Reader 5.0 and later.)
>>
/Namespace [
(Adobe)
(Common)
(1.0)
]
/OtherNamespaces [
<<
/AsReaderSpreads false
/CropImagesToFrames true
/ErrorControl /WarnAndContinue
/FlattenerIgnoreSpreadOverrides false
/IncludeGuidesGrids false
/IncludeNonPrinting false
/IncludeSlug false
/Namespace [
(Adobe)
(InDesign)
(4.0)
]
/OmitPlacedBitmaps false
/OmitPlacedEPS false
/OmitPlacedPDF false
/SimulateOverprint /Legacy
>>
<<
/AddBleedMarks false
/AddColorBars false
/AddCropMarks false
/AddPageInfo false
/AddRegMarks false
/ConvertColors /ConvertToCMYK
/DestinationProfileName ()
/DestinationProfileSelector /DocumentCMYK
/Downsample16BitImages true
/FlattenerPreset <<
/PresetSelector /MediumResolution
>>
/FormElements false
/GenerateStructure false
/IncludeBookmarks false
/IncludeHyperlinks false
/IncludeInteractive false
/IncludeLayers false
/IncludeProfiles false
/MultimediaHandling /UseObjectSettings
/Namespace [
(Adobe)
(CreativeSuite)
(2.0)
]
/PDFXOutputIntentProfileSelector /DocumentCMYK
/PreserveEditing true
/UntaggedCMYKHandling /LeaveUntagged
/UntaggedRGBHandling /UseDocumentProfile
/UseDocumentBleed false
>>
]
>> setdistillerparams
<<
/HWResolution [2400 2400]
/PageSize [612.000 792.000]
>> setpagedevice