sensors

Article
Effect of Sensors on the Reliability and Control
Performance of Power Circuits in the Web
of Things (WoT)
Sungwoo Bae 1 and Myungchin Kim 2, *
1 Department of Electrical Engineering, Yeungnam University, Gyeongsan 38541, Korea; sbae@yu.ac.kr
2 Agency for Defense Development, Daejeon 34186, Korea
* Correspondence: abekim81@gmail.com; Tel.: +82-42-821-0551

Academic Editors: Massimo Poncino and Ka Lok Man

Received: 25 June 2016; Accepted: 31 August 2016; Published: 6 September 2016

Abstract: In order to realize a true WoT environment, a reliable power circuit is required to ensure
interconnections among a range of WoT devices. This paper presents research on sensors and their
effects on the reliability and response characteristics of power circuits in WoT devices. The presented
research can be used in various power circuit applications, such as energy harvesting interfaces,
photovoltaic systems, and battery management systems for the WoT devices. As power circuits
rely on the feedback from voltage/current sensors, the system performance is likely to be affected
by the sensor failure rates, sensor dynamic characteristics, and their interface circuits. This study
investigated how the operational availability of the power circuits is affected by the sensor failure
rates by performing a quantitative reliability analysis. In the analysis process, this paper also includes
the effects of various reconstruction and estimation techniques used in power processing circuits
(e.g., energy harvesting circuits and photovoltaic systems). This paper also reports how the transient
control performance of power circuits is affected by sensor interface circuits. With the frequency
domain stability analysis and circuit simulation, it was verified that the interface circuit dynamics may
affect the transient response characteristics of power circuits. The verification results in this paper
showed that the reliability and control performance of the power circuits can be affected by the sensor
types, fault tolerant approaches against sensor failures, and the response characteristics of the sensor
interfaces. The analysis results were also verified by experiments using a power circuit prototype.

Keywords: power circuit; Web of Things (WoT); energy harvesting; reliability; filter; sensors

1. Introduction
The use of advanced power processing circuit technologies has increased continuously for
numerous applications such as wireless sensor networks [1,2], IoT devices [3,4], and home
appliances [5]. To realize a genuine Web of Things (WoT) environment, a reliable power circuit for
various connected devices is required [6]. As shown in Figure 1, the power circuits inside such devices
perform energy harvesting and power processing so that the devices are operational and connected
to the WoT. The processors and transmitters of such WoT devices can successfully function using
the energy that is harvested and processed by the power circuits. Since power circuits and different
types of power sources are commonly used in the WoT devices [7,8], their design engineers should not
overlook the effect of power circuits (e.g., energy harvesting interfaces) on system performance factors
such as reliability [9], cost [10], and efficiency [11]. Examples of power sources for WoT applications
that are considered in this paper include batteries [2] and solar cells [1,7]. To enable the effective and
efficient operation of different power sources, the system designers should consider the reliable circuit
control strategies and power management algorithms for power circuits used in WoT devices.

Sensors 2016, 16, 1430; doi:10.3390/s16091430 www.mdpi.com/journal/sensors

Sensors 2016, 16, 1430 2 of 18

reliable circuit control strategies and power management algorithms for power circuits used in WoT
Sensors 2016, 16, 1430 2 of 18
devices.

Figure 1. Role of power circuits in the WoT.

WoT.

Figure 22 shows
Figure shows aa block
block diagram
diagram of of aa typical
typical power
power circuit
circuit configuration
configuration for for aa WoT-connected
WoT-connected
device. While the actual power conversion in the power circuit is performed by the power
device. While the actual power conversion in the power circuit is performed by the power stage,
stage,
controllers are generally used to control the major variables of interest, such as
controllers are generally used to control the major variables of interest, such as voltage and current. voltage and current.
Since sensors
Since sensorsare are typically
typically usedused for feedback
for feedback in control
in control systems,systems,
satisfactorysatisfactory
operationoperation of the
of the controller
controller would not be possible when the feedback information is unavailable,
would not be possible when the feedback information is unavailable, which may result from the failure which may result
from
of the failure
sensors. of sensors.
As shown As shown
in Figure in Figurecircuit
2, an interface 2, an interface circuitthe
exists between exists between
sensor outputtheand
sensor output
the control
input. Hence, the characteristics of the actual feedback signal are expected to be determined notto
and the control input. Hence, the characteristics of the actual feedback signal are expected be
only
determined
by notbut
the sensors, only by by
also thetheir
sensors, but also
interface by their
circuits. interface
In other circuits.
words, In other
the failure words,
rate the failure
of sensors rate
can affect
of sensors can affect the reliability of power conversion circuits [12].
the reliability of power conversion circuits [12]. In the same manner, the dynamics of the sensorsIn the same manner, the
with their interface circuits can affect the control performance [13]. As a power circuit control methoda
dynamics of the sensors with their interface circuits can affect the control performance [13]. As
power circuit
generally control
requires method generally
the installation requires
of sensors, the installation
this paper of sensors,
reports the effects this and
of sensors paper reports
their the
interface
effects of sensors and their interface circuits on the performance of power circuits.
circuits on the performance of power circuits. In particular, this study focuses on how the reliability In particular, this
study
and focuses
control on how theare
performance reliability
affectedand control
by the sensorperformance
failure ratesareand affected
dynamics.by the sensor failure rates
and dynamics.
The analysis process, observation, and verification results of this study address hardware
The analysis
and system design process, observation,
challenges involving andtheverification
connectionresults of this
of various study
WoT addressSpecifically,
devices. hardware andthe
system design challenges involving the
contributions of this study can be summarized as follows:connection of various WoT devices. Specifically, the
contributions of this study can be summarized as follows:
Sensors 2016, 16, 1430 3 of 18
Sensors 2016, 16, 1430 3 of 18

FigureFigure 2. Block diagram of a typical power conversion approach for a WoT connected device.
2. Block diagram of a typical power conversion approach for a WoT connected device.
First, this research provides a comprehensive analysis result on the reliability and control
First, this research
performance of WoTprovides
devices by a comprehensive
considering the effects analysis result on
of sensors and the
theirreliability and control
interface circuits.
performance
Although oftheWoT devices
sensor byhas
reliability considering
been considered the effects
explicitlyofin sensors andstudies
several case their[12,14],
interface circuits.
the use
Althoughof sensor reliability
the sensor data was
reliability haslimited to only investigating
been considered explicitly theindetailed
severalfailure modes and
case studies failure
[12,14], the use
criticality. Compared to previous research, this study does not only focus
of sensor reliability data was limited to only investigating the detailed failure modes and failure on the reliability of the
sensor, but also examines how alternative sensing mechanisms other than the direct feedback from
criticality. Compared to previous research, this study does not only focus on the reliability of the
sensors can improve the system reliability. For example, the use of current information reconstruction
sensor, but also examines how alternative sensing mechanisms other than the direct feedback from
or estimation schemes that can increase the fault tolerance against current sensor faults is considered
sensorsascan improvetothe
a method system
increase thereliability. For example,
system reliability. Whilethe use of works
research currentoninformation reconstruction
the reconstruction or
or estimation schemes that can increase the fault tolerance against current
estimation of voltage/current feedback have been performed [15–19], this paper provides sensor faults is considered
the as
a method to increase
quantitative the system
analysis results reliability.
on the levelWhile research
of system works
reliability on the reconstruction
improvement or estimation of
that can be achieved.
Second,
voltage/current this paper
feedback considers
have the practical[15–19],
been performed hardware design
this paperaspects of power
provides circuits in sensor
the quantitative analysis
resultsinterfaces
on the levelthat ofhave not been
system explored
reliability thoroughly in that
improvement the past.
can Inbe particular,
achieved.this study is motivated
by the following observations: (1) measurement sensors used in the power circuits can also fail or
Second, this paper considers the practical hardware design aspects of power circuits in sensor
wear out [12]; and (2) interface circuits are generally used with sensors to address issues related to
interfaces that have not been explored thoroughly in the past. In particular, this study is motivated
electromagnetic interference (EMI), analog-to-digital conversion (ADC), and switching mode power
by the regulation
following [2]. observations:
Through both (1) measurement
theoretic analysis sensors used inoriented
and circuit the power circuits this
verification, can also
paperfail or
wear out [12]; andthat
demonstrates (2) such
interface circuits
practical are generally
constraints can affect theused with sensors
reliability to address
and regulation issues of
performance related
to electromagnetic
power circuits that interference
are commonly (EMI),
usedanalog-to-digital
for energy harvesting conversion (ADC), and switching mode
in WoT devices.
Third, a system
power regulation designerboth
[2]. Through can easily apply analysis
theoretic the analysis andapproach
circuitoforiented
this paperverification,
to study the effects
this paper
demonstrates that such practical constraints can affect the reliability and regulationfor
of different circuit configurations and feedback schemes that are used in power circuits different
performance of
power types
circuitsof embedded devices. When
that are commonly usedmultiple
for energy approaches
harvesting are used
in WoTto acquire
devices. feedback of the same
variable, the reliability analysis framework considered in this paper can be applied easily to reliability
Third, a system designer can easily apply the analysis approach of this paper to study the effects of
calculation models of such approaches. As the considered reliability analysis method provides
different circuit configurations and feedback schemes that are used in power circuits for different types
quantitative results, the system designer can perform a more comprehensive and practical trade-off
of embedded devices.
study among When multiple
the various approaches
design alternatives are used
during to acquire
the system feedback
planning of theresults
stage. These same can
variable,
also the
reliability analysis
be used framework
as a metric considered
to perform in this paper
design optimization can be
to meet applied
various easily (e.g.,
objectives to reliability calculation
cost, reliability,
modelsand of such approaches.
complexity). As the energy
For example, considered reliability
harvesting analysis
interfaces method
of sensors canprovides
be optimized quantitative
so that theresults,
design approach with the highest reliability level can be selected
the system designer can perform a more comprehensive and practical trade-off study among the among various design candidates.
variousWhile
designresearch works on
alternatives designing
during sustainable
the system energy stage.
planning harvesting
Theseapproaches
results can for also
sensors
be have
used as a
received continuous interest [1,7,8,20], most previous research works have focused only on the
metric to perform design optimization to meet various objectives (e.g., cost, reliability, and complexity).
For example, energy harvesting interfaces of sensors can be optimized so that the design approach
with the highest reliability level can be selected among various design candidates. While research
works on designing sustainable energy harvesting approaches for sensors have received continuous
interest [1,7,8,20], most previous research works have focused only on the development of circuit and
Sensors 2016, 16, 1430 4 of 18

power control approaches; and there seem to be limited studies on performing a system level analysis
considering the effect of power circuits on the reliability characteristics.
The remainder of this manuscript is organized as follows: the following section reviews the related
prior research works. Section 3 discusses the effects of sensors on the system reliability and introduces
the reliability analysis approach. The reliability analysis result is also provided. Section 4 shows how
the dynamics of the sensor interface circuit affect the system control performance. The verification
results of both simulation and experiments are provided in Section 5. Finally, Section 6 concludes the
paper by summarizing the findings and introducing directions for future work.

2. Review of Related Works

The design of power circuits in the WoT is critical to enabling sustainable operation of devices
that are powered by batteries [2] and PV cells [1,7]. Therefore, reliability of the power circuits should
be high enough to maximize the operational period and life cycles of WoT devices. Recently, research
works on a quantitative reliability analysis and the fault tolerant operation of power circuits have
been performed actively to establish a systematic approach for a reliable system design. For example,
the effects of different switching device configurations (i.e., discrete power switches and integrated
power modules) on the system reliability were studied [21]. Comparison results in [21] showed that
the overall reliability performance can differ depending on how power switches are configured even
in the same circuit topology. A design approach for power circuits in aerospace applications, which
requires a high level of reliability, was explored [9]. In particular, the effects of different capacitor types,
pulse-width-modulation (PWM) strategies, switching frequencies, and power switch configurations
on the system reliability were studied based on the reliability prediction results. The effects of
different circuit topologies on the reliability of PV system interfaces were also studied [22]. Studies on
reliability calculation frameworks have been proposed for actuation systems and renewable power
systems [12,23,24]. To design a system with improved reliability, a control approach for power circuits
with hardware redundancy was also introduced [25]. Previous studies [14,25,26] not only provided
reliability analysis results on the case studies of control approaches that have a redundancy, but also
discussed how practical fault scenarios can be handled by the proposed control approach. Although
reliability studies on power circuits have been investigated in previous studies, the majority of these
studies have focused only on power stage components, such as the power devices and reactive
components. In other words, the effects of sensors on the system reliability have not been fully
examined in power circuits used in WoT devices. In order to address such gap in the literature, this
paper performed a reliability analysis of power circuits with a focus on the effects of sensors and
alternative sensing approaches.
Studies on the approaches that can minimize the use of sensors have also been performed with
motivations to realize cost effective power management solutions and highly integrated power circuits
in a compact size. For example, a current estimation approach that uses an Extended Kalman Filter
(EKF) has been used in a boost converter [15]. Approaches that use either inductor voltage [16,17]
or inductor voltage-second [18] information to estimate the inductor current have been explored
to sense the inductor current without sensors. Using the approach of [19], it is possible to know
the current information of each individual stage in an interleaved dc converter by installing only a
single sensor at the common dc link instead of installing sensors at each stage. These approaches can
contribute to decreasing the overall cost and size of the power circuit by reducing the number of sensors
required. From a reliability perspective, such reconstruction or estimation techniques can be considered
for designing power circuit interfaces with fault tolerant capability [27]. Not only does the power
processing circuit become less constrained to the status of the sensor, but these estimation techniques
can also be used as a back-up sensing mechanism that can be activated as soon as a fault in the sensor
is detected. Although the approaches of previous studies can be applied to design a power processing
circuit that is tolerant to sensor faults, a quantitative analysis of reliability improvement should also be
required for optimized approaches that can satisfy both the cost and reliability requirements.
Sensors 2016, 16, 1430 5 of 18

Previous studies have also shown that the control performance can be affected by the interface
circuitsSensors
that2016,
are 16,
commonly
1430 used with sensors. For example, [13] demonstrated through a5 ofseries 18 of
experiments that subharmonic stability problems can arise in buck converters depending on the
parametersPrevious studies have also shown that the control performance can be affected by the interface
of the interface circuit. The effects of the current sensor or its interface circuit on the
circuits that are commonly used with sensors. For example, [13] demonstrated through a series of
stability performance of current mode controlled power converters have also been reported [28].
experiments that subharmonic stability problems can arise in buck converters depending on the
parameters of the interface circuit. The effects of the current sensor or its interface circuit on the
3. Reliability of Sensors and Systems
stability performance of current mode controlled power converters have also been reported [28].
This section introduces the power circuit configuration, control approach and reliability model
that are3.considered
Reliability of Sensors and Systems
in the following discussion of this paper.
This section introduces the power circuit configuration, control approach and reliability model
3.1. Control and
that are Reliabilityinofthe
considered Power Circuits
following discussion of this paper.

This
3.1.paper
Controlconsiders twoofrepresentative
and Reliability Power Circuits power processing circuit configurations. Figure 3 shows
the configuration and control approach of a dc-dc buck converter. Thanks to its ability to step down the
This paper considers two representative power processing circuit configurations. Figure 3 shows
output voltage, a buck converter is commonly used for energy harvesting in low power applications,
the configuration and control approach of a dc-dc buck converter. Thanks to its ability to step down
such asthewireless sensor networks [29] and small portable devices [30]. The buck converter can be
output voltage, a buck converter is commonly used for energy harvesting in low power
modeled as followssuch
applications, [31]:as wireless sensor networks [29] and small portable devices [30]. The buck
di
converter can be modeled as follows [31]:L L = q(t) E − vo (1)
dt
dvo = ( ) −vo (1)
C = iL − (2)
dt R
= − (2)
where i L is the inductor current, E is the input (source) voltage, vo is the capacitor (load) voltage, L is
the inductance,
where C the
is is the
inductor current, R is
capacitance, is the
theinput
load (source)
resistance, and q(t)isisthe
voltage, thecapacitor
switching function
(load) voltage,of the
power Lswitch Q. The switching
is the inductance, function q(t)
C is the capacitance, is the
R is defined as:
load resistance, and q(t) is the switching function of
the power switch Q. The switching function
( q(t) is defined as:
1,1,when Q isison
when Q on
q(t) =( ) = (3) (3)
0,0,when
when Q
Q isisoff
off

iO

iL
q (t ) iC vO

(a)

vO*
iL*

vO iL
(b)
Figure 3. (a) Buck converter configuration; (b) control block diagram.
Figure 3. (a) Buck converter configuration; (b) control block diagram.
A current-fed push-pull interface can be considered when a boost operation through isolation at
Ahigher
current-fed push-pull
power levels interface
is required can be
for safety. considered
Figure when
4 shows the a boost
circuit operation
configuration through
of this isolation at
boost power
higher processing unit,isand
power levels the circuit
required for can be modeled
safety. Figure 4asshows
[32]: the circuit configuration of this boost power
processing unit, and the circuit can be modeled as [32]:
Sensors 2016, 16, 1430 6 of 18

Sensors 2016, 16, 1430 6 of 18

dii
Li = v − vp (4)
dt = i −
(4)

dvo
= |is | − io
dt = | | −
Co (5)
(5)

dφ vs v p1 v p2
== = = ==−− (6)
(6)
dt Ns Np Np
Np
= [1
vp = − q1((t)) ∙·q2 (t)] ·∙
[1 − ·vo
Ns ∙
(7)
(7)

Lp
Ls
φ == iQ1 −
− −
iQ2 − is (8)
(8)
Np Ns
vs==[ [q(1 ()t)−− q(2 ()]t)]∙ ·v(9)
o (9)
(9)
where iii is
where is the
the current
current of inductor LLii;; vvi is
of inductor is the
the input voltage; vvpp is
input voltage; is the
the primary voltage; vvoo is
primary voltage; is the
the capacitor
capacitor
(output) voltage; C
(output) voltage; is the
Coo is the output capacitance; iiss is
output capacitance; is the
the secondary
secondary currentcurrent of
of the transformer; iioo is
the transformer; is the
the
output current;
output current; φ is flux; v
is the transformer magnetization flux; vs is the transformer secondary voltage;sNiss
the transformer magnetization s is the transformer secondary voltage; N
is the number of turnsininthe
the number of turns thetransformer
transformersecondary side;NNp pisisthe
secondaryside; thenumber
number of
of turns
turns in
in the transformer
the transformer
primary side;
primary and qqi(t)
side; and (t) is
is the
the switching
switching function
function ofof the
the power switch Q
power switch Qii (i(i == 1,
1, 2).
2). The
The switching
switching
function q (t) is defined in the same manner as
function qii(t) is defined in the same manner as Equation (3). Equation (3).

q1 (t ) q2 (t )

Figure 4.
Figure Configuration of
4. Configuration of aa current-fed
current-fed push-pull
push-pull converter.
converter.

To regulate
regulatethe output
the output voltage, vo , avcascaded
voltage, controlcontrol
o, a cascaded approach [33] (i.e.,[33]
approach Figure 3b)Figure
(i.e., was considered.
3b) was
As shown inAs
considered. Figure
shown 3b,inthe outer
Figure 3b,voltage
the outer controller
voltage generates
controller the reference
generates the command for the inner
reference command for
current
the innercontroller, and the current
current controller, and the controller generates generates
current controller the controlthe command for the power
control command switch,
for the power Q.
To regulate
switch, Q. Tothe load voltage,
regulate the load thevoltage,
voltagethe controller
voltagerequires
controller therequires
feedback theoffeedback
the load of voltage value,
the load and
voltage
the current
value, and thecontroller
currentuses the feedback
controller uses theoffeedback
the inductor
of thecurrent
inductorthrough
current sensors
through [34–36].
sensorsTherefore,
[34–36].
failure to provide
Therefore, failurethe
to feedback
provide the of accurate
feedback information
of accurate oninformation
the inductor on current to the current
the inductor controller
current to the
might cause
current an unsatisfactory
controller might cause performance
an unsatisfactory or failure in the voltage
performance regulation
or failure of the power
in the voltage circuit.
regulation of
Thispower
the paper circuit.
uses theThis
inductor
papercurrent
uses the forinductor
voltage current
regulation forpurposes, whereas the
voltage regulation current whereas
purposes, informationthe
has beeninformation
current used in otherhasapplications
been usedofinpower othercircuits for various
applications objectives.
of power circuitsA representative example of
for various objectives. A
energy harvesting
representative purposes
example is the harvesting
of energy maximum power purposes point tracking
is the maximum (MPPT) method
power pointoftracking
PV modules
(MPPT) [7].
methodWhile themodules
of PV control performance
[7]. is determined by the controller configuration, controller gains,
and the
Whilecircuit
the parameter values, theisreliability
control performance determined performance is determined
by the controller by the failure
configuration, rate gains,
controller of the
interface
and hardware.
the circuit In this
parameter paper,
values, theit reliability
is assumed that the reliability
performance of each
is determined byhardware
the failurecomponent
rate of the
can be modeled
interface hardware.to have
In this anpaper,
exponential distribution
it is assumed that thewith a constant
reliability failure
of each rate (λ),component
hardware as in previouscan
studies
be [9,10,12,25].
modeled That
to have an is, the reliability
exponential function
distribution withof aa constant
component failure R(t))(λ),
(i.e.,rate is as
expressed
in previousas [9,10,22]:
studies
[9,10,12,25]. That is, the reliability function of a component (i.e., R(t)) is expressed as [9,10,22]:
R (t) = e−λt (10)
( )= (10)
where λ is the failure rate (failures/hours), and t is time (hours).
Sensors 2016, 16, 1430 7 of 18

where λ is the failure rate (failures/hours), and t is time (hours).

3.2. Reliability Analysis Considering Sensors

Although each hardware device of the power circuit (e.g., capacitor, inductor, and power switch)
can fail during the system operation, this study simplifies the reliability analysis by assuming that
only the current sensors are subject to failures. Reliability studies that consider the failures of other
components can be found in other studies [22,24]. By assuming that only the inductor current sensor
can fail, the probability that the power circuit is operational (i.e., RPC (t)) can be calculated by using the
failure rate data of the current sensor as follows:

R PC (t) = RCS (t) = e−λCS t (11)

where RCS (t) is the reliability function of the current sensor, and λCS is the failure rate of the
current sensor. Examples of current sensor failure rate values can be found in previous reliability
analyses [14,25] or in datasheets provided by the manufacturers [37].
While the reliability can be calculated using Equation (11) based on the assumption that the
inductor current sensor is the only method for sensing the inductor current value, the inductor current
can still be known without requiring explicit feedback information of the sensor output. For example,
the inductor current can be estimated using the inductor voltage value as follows [16,17]:

1 1
Z Z
iL = v L dt = (q (t) E − vo ) dt (12)
L L

where vL is the inductor voltage. This estimation approach is possible based on the voltage-current
relationship of the inductors. When the inductor current feedback is performed based on an alternative
method instead of using the current sensor, the failure rate to be considered in Equation (11) should be
modified to represent the actual sensing configuration. For example, the failure rate in Equation (11)
should represent the properties of the voltage sensor instead of the current sensor if the inductor
voltage based estimation approach (i.e., Equation (12)) is used. A different approach for estimating the
inductor current is by reconstructing the inductor current value through other current information
of the interface. In particular, the capacitor current and load current can be used for such purposes.
Furthermore, such alternative sensing approaches can be used as a backup mechanism with an objective
to achieving fault tolerant capability against the faults of the inductor current sensor. The system can
be designed such that the controller receives feedback from the inductor current sensor as primary,
and the source of the inductor current value feedback can be switched from the current sensor to an
alternative approach (e.g., Equation (12)) when a fault in the inductor current sensor is detected.
When such alternative approaches are available, a more comprehensive analysis can be performed
to quantify the level of reliability improvement. Table 1 lists the possible operation conditions of the
three current sensors in the buck converter.

Table 1. Cases of the current sensor status.

Case Number Inductor Current Capacitor Current Load Current

1 O O O
2 O X O
3 O O X
4 O X X
5 X O O
6 X X O
7 X O X
8 X X X

Cases 1 to 4 represent the operation scenarios that the inductor current information is available
through the inductor current sensor itself. Since the explicit feedback information of an inductor
Sensors 2016, 16, 1430 8 of 18

current value from the sensor is available, the system is regarded as being operational regardless of the
operation status of the capacitor current sensor and the load current sensor. In other words, the system
is operational as long as the inductor current sensor is healthy. Hence, the probability (or the system
reliability function, RPC1 ) that the system is operational can be expressed as [10]:

R PC1 = p IL = e−λCS t (13)

where pIL is the probability that the inductor current sensor is healthy.
When the inductor current sensor is not operational, as in the other cases (i.e., Cases 5 to 8),
voltage regulation can still be performed using alternative approaches for current feedback information.
A detailed explanation of these cases is as follows.
In the case that the capacitor current sensor and the load current sensor are operational (i.e.,
Case 5), the instantaneous inductor current value can be reconstructed from the output of the capacitor
current sensor and the output of the load current sensor as follows:

i L_rec = iC + iO (14)

where iL_rec is the reconstructed inductor current, iC is the measured capacitor current, and iO is the
measured load current. Applying Kirchhoff’s current law at node A of Figure 3a enables such a current
reconstruction approach. Therefore, by installing a capacitor current sensor, a load current sensor,
and a mechanism (e.g., algorithm) that can provide the result of Equation (14) to the current controller,
the reliability of the system (i.e., RPC2 ) can be increased from Equations (13) to (15) as follows:

R PC2 = R PC1 + ∆R1 = R PC1 + q IL · p IC · p IO (15)

where qIL is the probability that the inductor current sensor fails (qIL = 1 − pIL ), pIC is the probability
that the capacitor current sensor is operational, and pIO is the probability that the load current sensor is
operational. An increase in reliability, ∆R1 , is achieved at the expense of installing additional current
sensors (i.e., capacitor current sensor and load current sensor) and implementing the reconstruction
approach, as expressed in Equation (14). Depending on the level of reliability required, some system
design cases will require such a backup scheme at the cost of installing additional sensors.
Although only the capacitor current sensor is available as shown in Case 7, it is still possible
to estimate the load current under the assumption that the load resistance is known. As the output
voltage sensor would have already been installed to perform voltage regulation in the buck converter,
the load current estimation can be performed using the sensed output voltage without installing extra
sensors. The inductor current value can be estimated using the capacitor current and the output
voltage value as follows:
v
i L_est = iC + iO_est = iC + O (16)
R
where iO_est is the estimated load current. Accordingly, the reliability of the system (i.e., RPC3 ) can be
increased further from Equations (15) to (17) as

R PC3 = R PC2 + ∆R2 = R PC2 + q IL · p IC ·q IO · pVO (17)

where qIO is the probability that the load current sensor fails (qIO = 1 − pIO ), and pVO is the reliability
function of the output voltage sensor. The primary objective of this study is not to develop fault tolerant
control or sensing approaches for power circuits but to introduce and perform a quantitative reliability
analysis considering the effects of sensors and various alternative feedback approaches. Based on this
objective, this paper considered rather straightforward sensing approaches for the reliability analysis.
Figure 5 shows the reliability computation results from Equations (13), (15) and (17) using the
sensor failure data [14]: λ = 2 failures/106 hours. The reliability value shows a remarkable increase with
increasing number of cases that make the system operational by applying additional mechanisms for
Sensors 2016, 16, 1430 9 of 18

current sensing. In other words, the system reliability increases with increasing number of alternative
approaches for acquiring the current value. For example, the reliability values at t = 10,000 h for
Equations (13), (15) and (17) are 0.9802, 0.9992 and 0.9996, respectively. The reliability analysis result
of this study, such
Sensors as1430
2016, 16, the plot shown in Figure 5, can be used to characterize the reliability 9 of 18 of power
circuits in various WoT devices. In addition, the analysis result can also provide estimation on system
reliability analysis result of this study, such as the plot shown in Figure 5, can be used to characterize
availability of networks that consist of such WoT devices. Using the quantitative reliability data
the reliability of power circuits in various WoT devices. In addition, the analysis result can also
(i.e., Figureprovide
5) from the proposed
estimation on systemreliability
availabilityanalysis process,
of networks a power
that consist circuit
of such designer
WoT devices. canthe
Using also choose
proper sensors or alternative
quantitative sensing
reliability data approaches
(i.e., Figure 5) from for
the aproposed
reliablereliability
power circuit
analysisin a more
process, systematic and
a power
circuit designer
comprehensive way. Ascan also choose
reliability proper sensorscan
performance or alternative sensingduring
be considered approaches
thefor a reliable
design power reliability
process,
circuit in a more systematic and comprehensive way. As reliability performance can be considered
issues of power circuits for energy harvesting and power processing in WoT devices can be addressed
during the design process, reliability issues of power circuits for energy harvesting and power
by applying the analysis
processing in WoTapproach
devices can of this study.
be addressed by applying the analysis approach of this study.

1
Equation (7)
Equation (9)
0.9 Equation (11)

0.8
Reliability

0.7

0.6

0.5

0.4

0 1 2 3 4 5
Time [Hrs] 5
x 10

Figure 5. Reliability calculation results.

Figure 5. Reliability calculation results.
3.3. Further Considerations
3.3. Further Considerations
As shown in Figure 5, the system reliability can be improved by considering alternative
approaches that perform identical tasks (e.g., control and sensing). In the case of buck converter
As shown in Figure 5, the system reliability can be improved by considering alternative
control, voltage regulation can also be performed using the cascaded control configuration that relies
approaches on that perform
the feedback of theidentical
capacitortasks
current(e.g.,
insteadcontrol and sensing).
of the inductor In The
current [38]. thecapacitor
case of current
buck converter
control, voltage
control regulation
approach cancan also be when
be activated performed
a fault inusing the cascaded
the original controller control
or in the configuration
inductor current that relies
sensor isof
on the feedback detected so that thecurrent
the capacitor regulationinstead
performance
of the is not interrupted
inductor by the[38].
current resulting
Thefault. With current
capacitor
such back-up capability, Case 7 in Table 1 can also be included as an operational case without
control approach can be activated when a fault in the original controller or in the inductor current
requiring a load current estimation and can contribute to improving the system reliability. In a similar
sensor is detected so that the
manner, the effects regulation
of alternative performance
approaches is not interrupted
for estimating the current onby thethe resulting
system fault.
reliability can With such
back-up capability, Case 7 in Table 1 can also be included as an operational case without requiring a
be studied further. Instead of relatively simple approaches that were considered in the previous
analysis,
load current it is possible
estimation and can to include the effects
contribute of different types
to improving the of advanced
system current reconstructions
reliability. In a similar or manner, the
estimation approaches that can be used when the inductor current sensor fails [39,40]. An estimation
effects of alternative approaches for estimating the current on the system reliability can be studied
of the capacitor current can be considered, and even all current values can be estimated when the
further. Instead of relatively
direct feedback from all simple approaches
current sensors that were
is unavailable, considered
as in Case in theisprevious
8. As the system equipped with analysis, it is
possible tosome
include
level ofthe effects
fault tolerantofcapability
different types of advanced
by introducing redundancycurrent reconstructions
in the control or sensing function,or estimation
approaches it isthat
possible
cantobe expect
used some improvement
when in reliability
the inductor currentsimilar to previous
sensor fails studies [14,25,26].
[39,40]. On the
An estimation of the
other hand, a reliability analysis that includes other alternative approaches was not performed in this
capacitor paper
current can be considered, and even all current values can be estimated
because the development of alternatives is not in the scope of this paper. In addition, the actual
when the direct
feedback from all current sensors is unavailable, as in Case 8. As the system
applicability of these alternative approaches is decided not only by the reliability requirements, but is equipped with some
also by the constraints of other factors, such as cost, complexity, and volume.
level of fault tolerant capability by introducing redundancy in the control or sensing function, it is
possible to expect some improvement in reliability similar to previous studies [14,25,26]. On the
other hand, a reliability analysis that includes other alternative approaches was not performed in this
paper because the development of alternatives is not in the scope of this paper. In addition, the actual
applicability of these alternative approaches is decided not only by the reliability requirements, but also
by the constraints of other factors, such as cost, complexity, and volume.
Sensors 2016, 16, 1430 10 of 18
Sensors 2016, 16, 1430 10 of 18

4. Dynamics of the Sensor Interface Circuit and System

4. Dynamics
While Sectionof the 3Sensor Interface
discussed howCircuit
sensorsandandSystem
alternative sensing approaches can affect the
reliability
Whileperformance of the power
Section 3 discussed circuits
how sensors in alternative
and WoT devices, this section
sensing approachesdiscusses howthe
can affect sensors and
reliability
their interface circuits could affect the control performance of power circuits.
performance of the power circuits in WoT devices, this section discusses how sensors and their interface Analogous to the
relationship between sensor failure rate and system reliability, the dynamic
circuits could affect the control performance of power circuits. Analogous to the relationship between characteristics of the
sensor failure
sensor itself andrate its
andinterface circuit can
system reliability, thealso affectcharacteristics
dynamic the control performance
of the sensorof power
itself circuits.
and its interfaceTo
illustrate such an effect, this paper considers the control block diagram of a buck
circuit can also affect the control performance of power circuits. To illustrate such an effect, this paper converter whose
output voltage
considers is regulated
the control throughoffeedback
block diagram control as
a buck converter shown
whose in Figure
output voltage6 [31]. Since the
is regulated actual
through
feedback signal that is fed to the voltage controller is not the immediate sensor
feedback control as shown in Figure 6 [31]. Since the actual feedback signal that is fed to the voltageoutput but the signal
that is processed by the sensor interface circuit, the signal characteristics of
controller is not the immediate sensor output but the signal that is processed by the sensor interfacethe voltage controller
output the
circuit, (i.e.,signal
control command) would
characteristics be affected
of the voltage by theoutput
controller properties
(i.e., of the interface
control command) circuit.
would Suchbe
interface circuits are commonly used to mitigate the effects of switching noises and
affected by the properties of the interface circuit. Such interface circuits are commonly used to mitigate to prevent aliasing
[31,41,42].
the effects of switching noises and to prevent aliasing [31,41,42].

vO* vO

Figure 6.
Figure Control block
6. Control block diagram
diagram of
of aa buck
buck converter.
converter.

study the
To study the effects
effects of
of interface
interface circuits
circuits on
on the
the control
control performance,
performance, this paper examines the
control loop
frequency response of the control loop gain
gain [43]
[43] shown
shown in in Figure
Figure 6.6. In particular, this study is
interested in exploring how the dynamics of the interface circuit affect the the regulation
regulation performance
performance
when external disturbances in the load or the the source
source are
are introduced.
introduced. The loop gain of the control
diagram shown in Figure 6 can can be
be expressed
expressed asas [43]:
[43]:
( )= ( )∙ ( )∙ ( ) (18)
Lc (s) = GC (s) · Gvd (s) · G IC (s) (18)
where Gc(s) is the transfer function of the voltage controller, Gvd(s) is the control input-to-output
voltageGctransfer
where (s) is the function
transfer function of the converter,
of the buck and GICG(s)
voltage controller, vd (s)
is is
thethetransfer
control input-to-output
function of the voltage
sensor
transfer
interfacefunction of feedback
circuit. In the buck converter, and GICthe
control systems, (s) is thegain
loop transfer
has function
been used of extensively
the sensor interface circuit.
for design and
In feedback control systems, the loop gain has been used extensively for design
characterization purposes of their systems because the reference-to-output transfer function of the and characterization
purposes
converterof their systems
voltage in Figurebecause theexpressed
6 can be reference-to-output
as [43]: transfer function of the converter voltage in
Figure 6 can be expressed as [43]:
( ) ( ) ( ) ( ) ( )
∗ ( )
= = (19)
v (s) ( ) (
GC (s) Gvd (s) ) ( ) ( )
GC (s) Gvd (s)
= = (19)
Assuming that the voltagev∗ (s) controller
1 + GC (sis )G PI(scontroller
a vd ) G IC (s) and the1 + dynamics
Lc (s) of the interface circuit
are approximated
Assuming that asthe
thevoltage
first order low pass
controller is afilter as:
PI controller and the dynamics of the interface circuit
are approximated as the first order low pass filter as: 1
( )= (20)
1+ 1
G (s) = (20)
where is the time constant of the interfaceICcircuit, the
τs +bode
1 plot of the loop gain (i.e., Equation (18))
can be analyzed as shown in Figure 7 using the system parameter values listed in Table 2.
where τ is the time constant of the interface circuit, the bode plot of the loop gain (i.e., Equation (18))
can be analyzed as shown in Figure 7 using the system parameter values listed in Table 2.
Sensors 2016, 16, 1430 11 of 18
Sensors 2016, 16, 1430 11 of 18

Table 2. System
Table parameter
2. System parametervalues usedfor
values used forinterface
interface circuit
circuit effects.
effects.

System
System Parameter
Parameter Value
Value
Inductance 25 μH
Inductance 25 µH
Capacitance 22 μF
Capacitance 22 µF
Source Voltage 16 V
Source Voltage 16 V
Voltage Controller Kp =K 0.2, Ki = 1000
Voltage Controller p = 0.2, Ki = 1000
Switching
Switching Frequency
Frequency 100100
kHzkHz

Figure 7. Bode
Figure plot
7. Bode plotofofthe
theloop
loop gain withdifferent
gain with differentsensor
sensor interface
interface circuits.
circuits.

Figure 7 shows the effects of different time constants on the frequency response of the loop gain.
Figure 7 shows the effects of different time constants on the frequency response of the loop gain.
Table 3 lists the values of the considered time constants. Although both the magnitude and the phase
Tableof3 the
listsloop
the gain
valuesareof the considered
affected timeinconstants.
by the change the dynamicsAlthough both thecircuit
of the interface magnitude and
(i.e., time the phase of
constant),
the loop
this gain
paper arefocuses
affectedonbythe
thephase
change in the(PM)
margin dynamics of theIninterface
variation. circuit
particular, the (i.e., time constant),
PM decreases with this
paperincreasing
focuses on theconstant
time phase margin
of the (PM) variation.
interface circuit.In particular, that
Considering the PM
the decreases
performance with of increasing
controlled time
constant of the
systems interface
is affected by circuit. Considering
the PM [42,43], it can bethat the performance
concluded of controlled
that the control performance systems
will beisaffected
affected by
by the dynamics of the interface circuit. Such a decrease in the PM value will result
the PM [42,43], it can be concluded that the control performance will be affected by the dynamics of the in a greater
oscillatory
interface circuit.system
Such response
a decrease to disturbances.
in the PM value It is worth noting
will result inthat a similar
a greater observation
oscillatory can beresponse
system made to
for digital filters.
disturbances. It is worth noting that a similar observation can be made for digital filters.
Table 3. Time constant and PM of considered cases.
Table 3. Time constant and PM of considered cases.
Case Number Interface Circuit Time Constant PM
Case 1
Case Number 1 μsTime Constant
Interface Circuit 63.1°PM
Case 2 5 μs 51.7° ◦
Case 1 1 µs 63.1
Case
Case 2 3 10 μs
5 µs 42.7°
51.7◦
Case 3 10 µs 42.7◦
Although the dynamics of the interface circuit can be considered during the controller (e.g.,
control law and controller gains) design process, the characteristics of the interface circuits can be
Although the dynamics
changed during the overalloflife
thecycle.
interface circuit can
For example, be considered
the capacitance during thevalues
or inductance controller
of the (e.g.,
passivecontrol
law and controller gains) design process, the characteristics of the interface circuits
components in analog circuits can be affected by the operation conditions (e.g., temperature, can be changed
during the overall
humidity) or thelife cycle.
aging For example,
of materials, the capacitance
and component or [34,35].
tolerances inductance
Whenvalues
multipleof sensing
the passive
approaches
components are implemented,
in analog circuits canitbeis affected
necessarybytothe
perform studies
operation for all of(e.g.,
conditions the feedback options
temperature, to
humidity)
or theensure
agingthat the performance,
of materials, particularly
and component stability, is
tolerances unaffected
[34,35]. When bymultiple
the sensing mechanism.
sensing For are
approaches
implemented, it is necessary to perform studies for all of the feedback options to ensure that the
performance, particularly stability, is unaffected by the sensing mechanism. For digital controllers,
in addition to the dynamics of the sensor interface circuit, the effects of analog-to-digital conversion
Sensors 2016, 16, 1430 12 of 18
Sensors 2016, 16, 1430 12 of 18

(ADC), sampling,
digital and in
controllers, computation candynamics
addition to the also be non-negligible [42,44].
of the sensor interface The the
circuit, delay timeofcaused
effects by such
analog-to-
digital
digital conversion (ADC), sampling, and computation can also be non-negligible [42,44]. The delay the
tasks is dependent on the sampling frequency, complexity of the sensing approach, and
time time
required
causedforbyfault
suchdetection or is
digital tasks controller
dependentreconfiguration.
on the sampling frequency, complexity of the sensing
approach, and the time required for fault detection or controller reconfiguration.
5. Results Analysis
5. Results Analysis
To demonstrate that the faults in the current sensor can be a direct reason for the failures of power
circuits, a To demonstrate
fault thatsimulated
scenario was the faults for
in the
thecurrent sensor can
buck converter be a the
using direct reason forlisted
parameters the failures
in Tableof4 [33].
power circuits, a fault scenario was simulated for the buck converter using the parameters listed in
Figure 8 presents the effects of this sensor fault scenario on the control performance. The load voltage
Table 4 [33]. Figure 8 presents the effects of this sensor fault scenario on the control performance. The
was regulated to 10 V using a cascaded control configuration as shown in Figure 3b. Initially, the buck
load voltage was regulated to 10 V using a cascaded control configuration as shown in Figure 3b.
converter shows
Initially, an acceptable
the buck regulation
converter shows performance
an acceptable owing
regulation to the controller,
performance owing toas
theshown in Figure
controller, as 8.
At t = 0.15 s, an inductor current sensor fault was introduced to the system by disconnecting
shown in Figure 8. At t = 0.15 s, an inductor current sensor fault was introduced to the system by the
inductor current feedback
disconnecting loop
the inductor as depicted
current feedbackinloop
Figure 8.
as depicted in Figure 8.

60
Output Voltage (V)

40

20

0
0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55

60
Inductor Current (A)

40

20

0
0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55

60
Feedback Current (A)

40

20

0
0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55
Time (s)

Figure Effect
8. 8.
Figure Effectofofsensor
sensor faults onthe
faults on thecontrol
controlperformance.
performance.

Table
Table 4. Systemparameters
4. System parameters used
used for
forsensor
sensorfailure (Figure
failure 8). 8).
(Figure
System Parameter Value
System Parameter
Inductanace 1 mHValue
Inductanace
Capacitance 125 1μFmH
Capacitance
Source Voltage 50125
V µF
Source Voltage
Voltage Controller Kp = 0.1, K50 V
i = 83.33
Voltage
CurrentController
Controller Kp =K0.666, KiK=i 5555
p = 0.1, = 83.33
Current Controller Kp = 0.666, Ki = 5555
Switching Frequency 100 kHz
Switching Frequency 100 kHz

A discrepancy between the actual inductor current value and the feedback current value exists
in discrepancy
A the simulationbetween
waveformthe between
actualt inductor
= 0.15 s and t = 0.4 svalue
current because
andof the
the sensor fault.current
feedback As soonvalue
as theexists
in thefault is introduced
simulation to the current
waveform between sensor, the sload
t = 0.15 andvoltage
t = 0.4becomes
s becauseuncontrollable andfault.
of the sensor the voltage
As soon as
fails is
the fault to introduced
be regulatedto
to the
the current
referencesensor,
value, 10
theV.load
Oncevoltage
the sensor fault isuncontrollable
becomes removed at t = 0.4 s and
and thethe
voltage
current feedback is resumed, it can be seen that the load voltage is regulated to the reference value
fails to be regulated to the reference value, 10 V. Once the sensor fault is removed at t = 0.4 s and the
current feedback is resumed, it can be seen that the load voltage is regulated to the reference value
and the controllability is recovered. This result shows that a fault in the sensor can result in a failure
of the power processing circuit interface. The cascaded control configuration (e.g., Figure 3b), which
 Sensors
Sensors 2016,2016, 16, 1430
16, 1430 1313ofof1818

and the controllability is recovered. This result shows that a fault in the sensor can result in a failure
of thethe
requires power processing
feedback circuit
of both the interface.
voltage and The current
cascadedvalues,
controlhas configuration
been reported (e.g.,to
Figure 3b), which
improve control
requires the
performance feedback
[31,33]. of boththe
However, the results
voltageof and current
Figure values,
8 show thathasthebeen reportedinner
additional to improve
currentcontrol
control
performance [31,33]. However, the results of Figure 8 show that the additional
loop could cause system failure when a fault in the current sensor occurs. Such reliability characteristic inner current control
loop on
is based could causethat
the fact system failure when
the cascaded controla configuration
fault in the current
generally sensor occurs.
requires Such sensors
feedback reliabilityfor
characteristic is based on the fact that the cascaded control configuration
two control loops (i.e., the inner control loop and the outer control loop) to be available. In generally requires feedback
realistic
sensors for two control loops (i.e., the inner control loop and the outer control loop) to be available.
settings, the protection logic of the converter would have been activated because of the excessive
In realistic settings, the protection logic of the converter would have been activated because of the
current level. Such activation would cause an interruption to the operation of the power circuit. Hence,
excessive current level. Such activation would cause an interruption to the operation of the power
sensor reliability performance should also be reviewed to ensure that the system reliability is not
circuit. Hence, sensor reliability performance should also be reviewed to ensure that the system
affected by the sensor failure rates. When a WoT device is required to have a relatively high reliability
reliability is not affected by the sensor failure rates. When a WoT device is required to have a
level, the alternative
relatively feedback
high reliability level,approaches
the alternativethat feedback
were discussed
approaches in Section 3 can
that were be used
discussed in to achieve
Section 3
reliable
can bepower
usedcircuit operation
to achieve reliable that is more
power tolerant
circuit against
operation thatcurrent
is more sensor faults.
tolerant Among
against the sensor
current various
available
faults. approaches
Among the that could
various be considered
available approaches as anthat
alternative
could be forconsidered
acquiring feedback information,
as an alternative for
the acquiring
reliabilityfeedback
analysisinformation,
results of Figure 5 can be used to select the most suitable
the reliability analysis results of Figure 5 can be used to select approach from
the a
system
mostreliability perspective.
suitable approach from a system reliability perspective.
In order to support
In order to support the the analysis
analysisofofSection
Section4,4,a asimulation
simulationwas was performed
performed to to examine
examinethe theeffects
effects
of sensor
of sensorinterface
interface circuit dynamics
circuit dynamicson onthethe performance
performance of ofaapower
powercircuit.
circuit.
The The parameters
parameters listed
listed in
in Tables
Tables 22 andand 3 werewere used
usedfor forthethesimulation.
simulation. Figure
Figure 9 shows
9 shows the load
the load regulation
regulation performance.
performance. The
Thevoltage
voltagereference
reference was set set
was to 3.3
to V.
3.3AsV.shown
As shown in Figure 9, the load
in Figure 9, thevoltage
load experiences a short transient
voltage experiences a short
period that is cleared within 0.5 ms. The response of Case 3 shows a larger
transient period that is cleared within 0.5 ms. The response of Case 3 shows a larger overshoot value overshoot value compared
to that of
compared to Case
that of1. The
Caseovershoot of Case 3of
1. The overshoot (0.3/3.3
Case =3 9.09%)
(0.3/3.3is =approximately three times larger
9.09%) is approximately three than
times
larger than Case 1 (0.1/3.3 = 3.03%). Such an increase in the overshoot value can be explained by in
Case 1 (0.1/3.3 = 3.03%). Such an increase in the overshoot value can be explained by the decrease the
the PM
decrease invalue,
the PM asvalue,
discussed in the previous
as discussed in thesection.
previous section.

3.8
Case 1
Case 3
3.6

3.4
Voltage [V]

3.2

2.8

2.6

49 49.5 50 50.5 51
Time [ms]

Figure Load
9. 9.
Figure regulation
Load regulationperformance
performancecomparison.
comparison.

An An additional
additional simulation
simulation was
was performedtotodemonstrate
performed demonstrate the
the effects
effects of
of interface
interfacecircuits
circuitson
onthe
the
regulation performance of power circuits during the transient period, as shown in Figure 10. The low
regulation performance of power circuits during the transient period, as shown in Figure 10. The low
pass filter cut-off frequency was set to 20 kHz, which is five times slower than in Case 3, as shown in
pass filter cut-off frequency was set to 20 kHz, which is five times slower than in Case 3, as shown in
Figure 9. Compared to the waveforms of Figure 9, it can be seen that the transient response shown in
Figure 9. Compared to the waveforms of Figure 9, it can be seen that the transient response shown
Figure 10 is more oscillatory and requires more time for the transient responses to be cleared in this
in Figure 10 is more oscillatory and requires more time for the transient responses to be cleared in
this case. Such a simulation result shows that the interface circuit should not be ignored during the
controller design process.
Sensors 2016, 16, 1430 14 of 18

case.Sensors
Such2016, 16, 1430
a simulation 14 of the
result shows that the interface circuit should not be ignored during 18

controller design process.

case. Such a simulation result shows that the interface circuit should not be ignored during the
Sensors 2016, 16, 1430 14 of 18
controller design process.
3.8
3.8
3.6
3.6
3.4
3.4
3.2
Voltage [V]
3.2
Voltage [V] 3
3
2.8
2.8
2.6
2.6
2.4
2.4
2.2
492.2 49.5 50 50.5 51
49 49.5 Time 50[ms] 50.5 51
Time [ms]
Figure 10. Load regulation performance (20 kHz).
Figure
Figure 10.10. Load
Load regulationperformance
regulation performance(20
(20kHz).
kHz).

An experiment was also performed to demonstrate that the dynamics of the interface circuits
An experiment was also performed to demonstrate that the dynamics of the interface circuits
An
affect theexperiment
control was also performed
performance of of
power to demonstrate that
circuits. the dynamics of the interface
thecircuits affect
affect the control performance power circuits.Figure
Figure11
11shows
showsthe
theconfiguration
configuration ofof the experiment
experiment
the control performance of power circuits. Figure 11 shows the configuration of the experiment setup.
setup.
setup.

(a)
(a)

(b)
(b)
Figure 11. (a) Configuration of the experiment setup; (b) prototype of a power circuit.
Figure
Figure 11.
11. (a)
(a) Configuration
Configuration of
of the
the experiment
experiment setup;
setup; (b)
(b) prototype of aa power
prototype of power circuit.
circuit.

The step-up operation (i.e., boost mode) of the current-fed push-pull converter was performed.
Figure 12 shows that the output voltage waveforms using two different digital low pass filters.
Sensors 2016, 16, 1430 15 of 18

SensorsThe step-up
2016, 16, 1430 operation
(i.e., boost mode) of the current-fed push-pull converter was performed.
15 of 18
Figure 12 shows that the output voltage waveforms using two different digital low pass filters.

(a)

(b)
Figure
Figure 12.
12. Voltage
Voltage waveforms
waveforms with
with different
different filter time constants
filter time constants (a)
(a) 50
50 ms;
ms; (b)
(b) 0.1
0.1 ms.
ms.

In the experiment, two different filter time constants (i.e., 50 ms and 0.1 ms) were considered,
In the experiment, two different filter time constants (i.e., 50 ms and 0.1 ms) were considered, and
and the input voltage of the experimented power supply interface was 14 V. Throughout the
the input voltage of the experimented power supply interface was 14 V. Throughout the experiment,
experiment, the output voltage was commanded to be regulated to 48 V. While the voltage showed
the output voltage was commanded to be regulated to 48 V. While the voltage showed a satisfactory
a satisfactory performance until the load changed, the output voltage showed a transient response
performance until the load changed, the output voltage showed a transient response period as soon
period as soon as the load power was changed from 50 W to 150 W. A comparison of Figure 12a,b
as the load power was changed from 50 W to 150 W. A comparison of Figure 12a,b shows that the
shows that the dynamics of the interface circuits can affect the transient response characteristics of
dynamics of the interface circuits can affect the transient response characteristics of the power circuits.
the power circuits.
6. Conclusions
6. Conclusions
This study examined how the reliability and transient response characteristics of power circuits
usedThis
in thestudy
WoTexamined
devices arehow the reliability
affected and and
by sensors transient response
interface characteristics
circuits. As sensors of power
play circuits
an essential
used in the WoT devices are affected by sensors and interface circuits. As sensors play
role in a power circuit control unit, it is necessary to study their potential effects on the overall system an essential
role in a power
reliability circuit control
and control unit, it isAnecessary
performance. to study
quantitative their potential
reliability analysiseffects on the overall
was performed system
to provide
reliability and control performance. A quantitative reliability analysis was
detailed information of the feedback mechanism used in power circuits. The sensor failure rate dataperformed to provide
detailed
were used information of thehow
to characterize feedback mechanism
the sensor useditself
hardware in power
couldcircuits. Thepower
affect the sensorcircuit
failurereliability.
rate data
were
In addition, this paper studied how the system reliability is affected by introducing different types In
used to characterize how the sensor hardware itself could affect the power circuit reliability. of
addition, this paper studied
sensing mechanisms to acquire how the system
feedback reliability
information suchis as
affected by introducing
reconstruction different
or estimation types of
approaches.
sensing mechanisms
The reliability to acquire
analysis result showedfeedback information
that the overall such as depends
system reliability reconstruction
not onlyoronestimation
the failure
approaches. The reliability analysis result showed that the overall system
rate of the sensor, but also on how the system is designed to operate with sensor faults. reliability depends not only
In particular,
on the failure rate of the sensor, but also on how the system is designed to operate with
it was shown that a higher level of system reliability can be achieved at the cost of installing additional sensor faults.
sensors. Furthermore, the analysis result can characterize the reliability performance in a quantitative
Sensors 2016, 16, 1430 16 of 18

manner so that further optimization can be conducted during the power circuit design stage to satisfy
multiple objectives using a systematic approach. The verification results demonstrated the importance
of the sensor being operational by showing the case of power circuits losing regulation capability.
In addition, the effects of sensor interface circuits on the stability characteristics were explored by
studying the dynamic frequency response of the control loop gain. The analysis results in this paper
showed that the control loop phase margin is affected by the dynamics of the sensor interface circuits.
The effects of sensor faults and its interface circuit dynamics on the system performance were also
verified by both simulations and experiments.
As the proposed analysis approach can be easily applied to other types of power circuits in WoT
devices, the analysis approach on reliability and dynamic properties of the sensors, their interface
circuits, and alternative sensing approaches is under investigation for future work. Examples of such
work include the comprehensive reliability analysis of power circuits for various WoT devices used at
different operation conditions (e.g., environment, and operation periods), the development of design
standards on sensor integration of power circuits used in WoT devices, and the effects of different
reliability models on the estimation accuracy of reliability performance. Exemplar case studies on
reliability of actual WoT devices are also works of interest to be performed.

Acknowledgments: This work was supported by the 2013 Yeungnam University Research Grant.
Author Contributions: Sungwoo Bae and Myungchin Kim performed the analysis; Sungwoo Bae performed the
experiments; Myungchin Kim analyzed the data.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Zou, T.; Lin, S.; Feng, Q.; Chen, Y. Energy-Efficient Control with Harvesting Predictions for Solar-Powered
Wireless Sensor Networks. Sensors 2016, 16, 53. [CrossRef] [PubMed]
2. Hwang, Y.-S.; Chen, J.-J.; Lai, B.-H.; Ku, Y.-T.; Yu, C.-C. A Fast-Transient Boost Converter with
Noise-Reduction Techniques for Wireless Sensor Networks. IEEE Sens. J. 2016, 16, 3188–3197. [CrossRef]
3. Martinez, B.; Montón, M.; Vilajosana, I.; Prades, J.D. The Power of Models: Modeling Power Consumption
for IoT Devices. IEEE Sens. J. 2015, 15, 5777–5789. [CrossRef]
4. Gubbi, J.; Buyya, R.; Marusic, S.; Palaniswami, M. Internet of Things (IoT): A vision, architectural elements,
and future directions. Future Gener. Comput. Syst. 2013, 29, 1645–1660. [CrossRef]
5. Ahn, J.-H.; Kim, D.-H.; Lee, B.-K.; Jin, H.-C.; Shim, J.-S. DC Appliance Safety Standards Guideline through
Comparative Analysis of AC and DC Supplied Home Appliances. J. Electr. Eng. Technol. 2012, 7, 51–57.
[CrossRef]
6. Ahmad, M. Reliability Models for the Internet of Things: A Paradigm Shift. In Proceedings of the 2014
IEEE International Symposium on Software Reliability Engineering Workshops (ISSREW), Naples, Italy,
3–6 November 2014; pp. 52–59.
7. Jafer, I.; Stack, P.; MacNamee, K. Design of New Power Management Circuit for Light Energy Harvesting
System. Sensors 2016, 16, 270. [CrossRef] [PubMed]
8. Brunelli, D. A High-Efficiency Wind Energy Harvester for Autonomous Embedded Systems. Sensors 2016,
16, 327. [CrossRef] [PubMed]
9. Burgos, R.; Chen, G.; Wang, F.; Boroyevich, D.; Odendaal, W.G.; Van Wyk, J.D. Reliability-Oriented Design
of Three-Phase Power Converters for Aircraft Applications. IEEE Trans. Aerosp. Electron. Syst. 2012, 48,
1249–1263. [CrossRef]
10. Yu, X.; Khambadkone, A.M. Reliability Analysis and Cost Optimization of Parallel-Inverter System.
IEEE Trans. Ind. Electron. 2012, 59, 3881–3889. [CrossRef]
11. Kolar, J.; Krismer, F.; Lobsiger, Y.; Muhlethaler, J.; Nussbaumer, T.; Minibock, J. Extreme Efficiency Power
Electronics. In Proceedings of the 2012 International Conference on Integrated Power Electronics Systems
(CIPS), Nuremberg, Germany, 6–8 March 2012; pp. 1–22.
12. Bazzi, A.M.; Dominguez-Garcia, A.; Krein, P.T. Markov Reliability Modeling for Induction Motor Drives
under Field-Oriented Control. IEEE Trans. Power Electron. 2012, 27, 534–546. [CrossRef]
Sensors 2016, 16, 1430 17 of 18

13. Fang, C.-C.; Redl, R. Subharmonic Stability Limits for the Buck Converter with Ripple-Based Constant
on-Time Control and Feedback Filter. IEEE Trans. Power Electron. 2014, 29, 2135–2142. [CrossRef]
14. Julian, A.L.; Oriti, G. A Comparison of Redundant Inverter Topologies to Improve Voltage Source Inverter
Reliability. IEEE Trans. Ind. Appl. 2007, 43, 1371–1378. [CrossRef]
15. Tong, Q.; Chen, C.; Zhang, Q.; Zou, X. A Sensorless Predictive Current Controlled Boost Converter by using
and EKF with Load Variation Effect Elimination Function. Sensors 2015, 15, 9986–10003. [CrossRef] [PubMed]
16. Midya, P.; Krein, P.T.; Greuel, M.F. Sensorless Current Mode Control-An Observer-Based Technique for
DC-DC Converters. IEEE Trans. Power Electron. 2001, 16, 522–526. [CrossRef]
17. Kimball, J.W.; Krein, P.T.; Chen, Y. Hysteresis and Delta Modulation Control of Converters using Sensorless
Current Mode. IEEE Trans. Power Electron. 2006, 21, 1154–1158. [CrossRef]
18. Lopez, V.M.; Azcondo, F.J.; de Castro, A.; Zane, R. Universal Digital Controller for Boost CCM Power Factor
Correction Stages Based on Current Rebuilding Concept. IEEE Trans. Power Electron. 2014, 29, 3818–3829.
[CrossRef]
19. Kim, H.; Falahi, M.; Jahns, T.M.; Degner, M.W. Inductor Current Measurement and Regulation Using a Single
DC Link Current Sensor for Interleaved DC-DC Converters. IEEE Trans. Power Electron. 2011, 26, 1503–1510.
[CrossRef]
20. Li, Y.; Yu, H.; Su, B.; Shang, Y. Hybrid Micropower Source for Wireless Sensor Network. IEEE Sens. J. 2008, 8,
678–681. [CrossRef]
21. Abdi, B.; Ranjbar, A.H.; Gharehpetian, G.B.; Milimonfared, J. Reliability Considerations for Parallel
Performance of Semiconductor Switches in High-Power Switching Power Supplies. IEEE Trans. Ind. Electron.
2009, 56, 2133–2139. [CrossRef]
22. Chan, F.; Calleja, H. Reliability Estimation of Three Single-Phase Topologies in Grid-Connected PV Systems.
IEEE Trans. Ind. Electron. 2011, 58, 2683–2689. [CrossRef]
23. Smater, S.S.; Dominguez-Garcia, A.D. A Framework for Reliability and Performance Assessment of Wind
Energy Conversion Systems. IEEE Trans. Power Syst. 2011, 26, 2235–2245. [CrossRef]
24. Dhople, S.V.; Davoudi, A.; Domínguez-García, A.D.; Chapman, P.L. A Unified Approach to Reliability
Assessment of Multiphase DC–DC Converters in Photovoltaic Energy Conversion Systems. IEEE Trans.
Power Electron. 2012, 27, 739–751. [CrossRef]
25. Julian, A.L.; Oriti, G.; Blevins, S.T. Operating Standby Redundant Controller to Improve Voltage-Source
Inverter Reliability. IEEE Trans. Ind. Appl. 2010, 46, 2008–2014. [CrossRef]
26. Geyer, T.; Schroder, S. Reliability Considerations and Fault-Handling Strategies for Multi-MW Modular
Drive systems. IEEE Trans. Ind. Appl. 2010, 46, 2442–2451. [CrossRef]
27. Jiang, W.; Fahimi, B. Current Reconstruction Techniques for Survivable Three-Phase PWM Converters.
IEEE Trans. Power Electron. 2010, 25, 188–192. [CrossRef]
28. Aroudi, A.E.; Calvente, J.; Giral, R.; Al-Numay, M.; Martinez-Salamero, L. Boundaries of Subharmonic
Oscillations Associated to Filtering Effects of Controllers and Current Sensors in Switched Converters Under
CMC. IEEE Trans. Ind. Electron. 2016, 63, 4826–4837. [CrossRef]
29. Chen, J.-J.; Hwang, Y.-S.; Yu, J.-H.; Ku, Y.-T.; Yu, C.-C. A Low-EMI Buck Converter Suitable for Wireless
Sensor Networks with Spur-Reduction Techniques. IEEE Sens. J. 2016, 16, 2588–2597. [CrossRef]
30. Krug, M.; Hartmann, J.; Gröll, L.; Gengenbach, U.; Nagel, J.; Bretthauer, G. Model-Based Dual-Mode
Controller for Low-Power Buck Converters. In Proceedings of the 2014 International Conference on
Optimization of Electrical and Electronic Equipment (OPTIM), Bran, Romania, 22–24 May 2014; pp. 489–497.
31. Krein, P.T. Elements of Power Electronics; Oxford University Press: New York, NY, USA, 1998.
32. Hartmann, L.V.; Corrêa, M.B.; Lima, A.M. A Simple and Effective Control Strategy for Improved Operation
of a Current-Fed Push-Pull Converter. In Proceedings of the 2010 IEEE Energy Conversion Congress and
Exposition, Atlanta, GA, USA, 12–16 September 2010; pp. 1098–1103.
33. Tsang, K.; Chan, W. Cascade Controller for DC/DC Buck Convertor. IEE Proc. Electr. Power Appl. 2005, 152,
827–831. [CrossRef]
34. Dallago, E.; Passoni, M.; Sassone, G. Lossless Current Sensing in Low-Voltage High-Current DC/DC Modular
Supplies. IEEE Trans. Ind. Electron. 2000, 47, 1249–1252.
35. Pajer, R.; Milanoviĉ, M.; Premzel, B.; Rodiĉ, M. MOS-FET as a Current Sensor in Power Electronics Converters.
Sensors 2015, 15, 18061–18079. [CrossRef] [PubMed]
Sensors 2016, 16, 1430 18 of 18

36. Patel, A.; Ferdowsi, M. Current Sensing for Automotive Electronics—A Survey. IEEE Trans. Veh. Technol.
2009, 58, 4108–4119. [CrossRef]
37. LF 2005-S/SP13 User Guide, LEM SA. Available online: http://www.lem.com/docs/manuals/LF%202005-
S%20SP13_en.pdf (accessed on 10 August 2016).
38. Kapat, S.; Krein, P.T. Formulation of PID control for DC–DC converters based on capacitor current:
A geometric context. IEEE Trans. Power Electron. 2012, 27, 1424–1432. [CrossRef]
39. Lukic, Z.; Zhao, Z.; Ahsanuzzaman, S.; Prodic, A. Self-Tuning Digital Current Estimator for Low-Power
Switching Converters. In Proceedings of the 2008 IEEE Applied Power Electronics Conference and Exposition
(APEC), Austin, TX, USA, 24–28 February 2008; pp. 529–534.
40. Mezger, F.; Killat, D. A Digital Observer based Current Loop Control for Buck Converters. In Proceedings of
the 2013 Conference of the IEEE Industrial Electronics Society (IECON), Vienna, Austria, 10–13 November
2013; pp. 181–186.
41. Castelló, J.; Espí, J.M.; García-Gil, R. A New Generalized Robust Predictive Current Control for
Grid-Connected Inverters Compensates Anti-Aliasing Filters Delay. IEEE Trans. Ind. Electron. 2016, 63,
4485–4494. [CrossRef]
42. Ellis, G. Control System Design Guide: A Practical Guide, 3rd ed.; Elsevier Academic Press: San Diego, CA,
USA, 2004.
43. Erickson, R.W.; Maksimovic, D. Fundamentals of Power Electronics, 2nd ed.; Kluwer Academic Publishers:
Norwell, MA, USA, 2001.
44. Chang, Y.-T.; Lai, Y.-S. Effect of Sampling Frequency of A/D Converter on Controller Stability and Bandwidth
of Digital-Controlled Power Converter. In Proceedings of the 2007 International Conference on Power
Electronics (ICPE), Daegu, Korea, 22–26 October 2007; pp. 625–629.

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).