IRJET- Different Control Strategies for Power Control of Voltage Source Converters in a Microgrid

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 05 Issue: 11 | Nov 2018 www.irjet.net p-ISSN: 2395-0072
© 2018, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 303
Different Control Strategies for Power Control of Voltage Source
Converters in a Microgrid
Er. Shashank S. Deshmukh1, Er. Shridhar S. Khule2
1PG student, Department of Electrical Engineering, Matoshri College of Engineering and Research Centre,
Eklahare, Nashik, Maharashtra, India.
2Head of Department, Department of Electrical Engineering, Matoshri College of Engineering and Research
Centre, Eklahare, Nashik, Maharashtra, India.
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract - Traditional proportional integral (PI) control has
been extensively used for power control of voltage source
converters in microgrid systems. A new Critic based PI
controller is discussed in this paper. In another method using
the PR controllers, the converter reference tracking
performance can be enhanced and previously known
shortcomings associated with conventional PI controllers can
be alleviated. Another control strategy includes Vector
controlled VSC's. In it the main improvement is tosuppressthe
possible dc-link voltage fluctuations under power line faults
and unbalanced conditions. Another control strategy includes
Hysteresis Current Controller employed for the inverter
control. The proposed inverter control technique interfaces
renewable energy sources and the AC bus of micro grid. It
offers the possibility to inject power from the renewable
sources and also improves the powerqualityinthesamemicro
grid. Also a vector control strategy for regulating the current
of grid-tied voltage source converters in a rotating reference
frame is suggested in this paper.
Key Words: fuzzy logic, proportional-resonant (PR)
controllers, High-voltage direct current (HVDC),
Lyapunov methods, pulse width modulation (PWM),
RTDS, vector-controlled VSC, Synchronous Reference
Frame, DQ-Current Control, Voltage Source Converters.
I. INTRODUCTION
A microgrid is an integration of multiple distributed
energy resources such as generation, storage, and loadunits
at the distribution level. This single united system is capable
of operating in both grid-connected and stand-alone or
islanded modes. The nature of most distributed energy
resources in the microgrid requires power electronic-based
converters to reshape their generated energy to a form
compatible with the utility grid’s voltage and frequency.The
majority of DGs are hence interfaced to the microgrid
through voltage source converters (VSCs). Furthermore,
compared to traditional electric machines, use of power
electronic converters yields to more operation and control
flexibility for the microgrid sources. Nevertheless, it should
be noted that the lower physical inertia of power electronic
converters leads to higher sensitivity to the network
disturbances and may in occasions result in power
oscillations and violation of overall stability. Power control
of VSCs has been traditionally implemented in synchronous
(dq), stationary (αβ), or natural (abc) reference frames and
depending on the reference frame used the adopted control
structures can be chosenfromlinearornonlinearalgorithms
such as proportional integral (PI), proportional resonant
(PR), vector control, hysteresis control, etc., [1]–[8].
The design simplicity and easier implementationoflinear
control algorithms used in synchronous or stationary
reference frames have resulted in their relative popularity
compared to nonlinear approaches. The synchronous
reference frame which has received more attention uses a
Park transformation to convert three-phase rotating
quantities to direct and quadrature dc variables. These dc
quantities can be regulated by using simple and linear PI
controllers. Despite good performance in controlling dc
variables, PI controllers have the well-known drawback of a
nonzero steady state error; furthermore they fail to operate
well in case of load changes and high penetration of DGs. in
microgrids. Poor compensation capability for low-order
harmonics has also been reported. Application of PR
controllers in the stationary reference frame eliminates the
steady state error, however, is accompanied with other
drawbacks such as sensitivity to grid frequency variations,
poor transient response during step changes, and low
stability margins [9]. The use of natural reference frames on
the other hand, requires more complex control structures
such as hysteresis control. Application of such nonlinear
structures will itself result in variableswitchingfrequencies,
which requires complex adaptive band hysteresis control
[10]. In addition to classic control approaches, intelligent
control algorithms such as fuzzy logic and artificial neural
networks have been also used in recent years with the
purpose of improving the transient performance and
adaptivity; however, in most casesthesealgorithmsincrease
the control system complexity [11]–[13].
The recently introduced proportional-resonant (PR)
controllers and filters, and their suitability for
current/voltage control of grid-connected converters, are
described. Using the PR controllers, the converter reference
tracking performance can be enhanced and previously
known shortcomings associated with conventional PI

controllers can be alleviated. These shortcomings include
steady-state errors in single-phase systems and the needfor
synchronous d–q transformation in three-phase systems.
Based on similar control theory, PR filters can also be used
for generating the harmonic commandreferencepreciselyin
an active power filter, especially for single-phase systems,
where d–q transformation theory is not directly applicable.
Another advantage associated with the PR controllers and
filters is the possibility of implementing selective harmonic
compensation without requiring excessive computational
resources.
Voltage-source converter (VSC)-based transmission
systems have attractive potential features in termsofpower
flow control and stability of the network.Althoughrelatively
low switching frequencyoperationofhigh-powerconverters
(9–15 times the line frequency) is desirable, it makes them
sensitive to power network imbalances when they may be
needed the most. Hence a control structure to improve the
performance of high-power vector-controlled back-to-back
VSC systems for conventional and emerging utility
applications is proposed. The main improvement is to
suppress the possible dc-link voltage fluctuations under
power line faults and unbalanced conditions. The proposed
controller structure is designed based on regulating the
converter system’s states locally in dq synchronous
reference frame withoutsequencecomponentsextractionor
resonant notch compensator.RTDSresultsverifythevalidity
of the proposed control architecture during normal and
unbalanced power system conditions.
In another method a vector control strategy for regulating
the current of grid-tied voltage source converters in a
rotating reference frame is discussed. The proposed
approach is based on shaping the open-loop andclosed-loop
transfer matrices ofthesystem.Solvinga constrainedconvex
optimization problem, the shaping is achieved, which
guarantees the stability of the closed loop system. The
designed controller results in the desired dynamic
performance and decouples the direct and quadrature (dq)
current axes. The structure of the proposed controller is
similar to that of its predecessors and consists of four
proportional-integral (PI) controllers.
Researches related to non-conventional energy sources
have grown significantly in the present scenario. The
electrical energy from the PV panel is considered as one of
the most useful natural resources. The last method in this
paper deals with the operation and control of a grid
connected PV system with a nonlinear load. A utility grid
connection is provided in order to replenish energy levelsin
case of power shortage from the renewable energy sources.
Hysteresis Current Controller is employed for the inverter
control. The proposed inverter control technique interfaces
renewable energy sources and the AC bus of micro grid. It
offers the possibility to inject power from the renewable
sources and also improves the power quality in the same
micro grid.
II. DISCUSSION OF DIFFERENT METHODS FOR POWER
CONTROL USING VSC'S IN MICROGRID
1. A Novel Self tuning PI Control system:
A novel self-tuning PI control system is proposed in this
method to cope with the aforementioned deficiencies of
traditional PI control by increasing the control system
adaptivity while avoiding unnecessary complexity. To cope
with the slow adaptation of traditional fixed-gain PI
regulators to power changes, disturbances, and parameter
variations a continuous scheme can be used to adjust the
controller gains over time. The traditional solutionistotune
the gains manually by observing the output of the system.
However, to avoid unknown number of trial and error tasks
in manual control and increase the reliability, an online and
automatic tuning approach must be adapted. The authors
have previously implemented a fuzzy system to
automatically tune the gains of PI controllers in microgrids
in [12]. In this method, no exact mathematical formula was
used for the learning process,insteadbasedonthereasoning
capability of the fuzzy system the change in integral and
proportional gains were estimated and used to tune the PI
controller. In this paper, a novel gain-tuning algorithm is
developed in which online tuning is achieved by combining
the reasoning capability of a so-called fuzzy critic and a
mathematical-learning algorithm. Critic based learning
algorithms, also known as neuro-dynamicprogramming, are
a wide group of learning methods that are usually used to
facilitate the online tuning of intelligentcontrollersincase of
highly nonlinear and noisy systems [13]. The proposed
control system contains a PI controller and a critic agent
whose task is to continuously assess the performance and
generate an evaluation signal. The PI controller gains are
updated using a learning algorithm with the objective of
optimizing critic’s output signal, which represents the
credibility of the system performance. The proposed
approach is a non model-basedadaptivestructurewhich will
be named the critic-based self-tuning PI (CSPI) controller,
hereafter in this paper. The CSPI controller isusedtocontrol
the active and reactive power of VSCs within a power
electronic-based microgrid.
2. Proportional Resonant Controllers:
Over the years, power converters of various topologies
have found wide application in numerous grid-interfaced
systems, including distributed power generation with

renewable energy sources (RES) like wind, hydro and solar
energy, microgrid power conditioners and active power
filters. Most of these systems include a grid-connected
voltage-source converter whose functionality is to
synchronise and transfer the variable produced power over
to the grid. Another feature of the adopted converter is that
it is usually pulse-width modulated (PWM) at a high
switching frequency and is either current- or voltage-
controlled using a selected linear or nonlinear control
algorithm. The deciding criterion when selecting the
appropriate control scheme usually involves an optimal
trade off between cost, complexity and waveform quality
needed for meeting (for example) new power quality
standards for distributed generation in low-voltage grids,
like IEEE-1547 in the USA and IEC61727 in Europe at a
commercially favourable cost. With the above-mentioned
objective in view while evaluating previously reported
control schemes, the general conclusion is that most
controllers with precise reference tracking are either
overburdened by complex computational requirements or
have high parametric sensitivity (sometimes both). On the
other hand, simple linear proportional– integral (PI)
controllers are prone to known drawbacks, including the
presence of steady-state error in the stationary frame and
the need to decouple phase dependency in three phase
systems although they are relatively easy to implement.
Exploring the simplicity of PI controllers and to improve
their overall performance, many variations have been
proposed in the literature including the addition of a grid
voltage feed forward path, multiple-state feedback and
increasing the proportional gain. Generally, these variations
can expand the PI controller bandwidth but, unfortunately,
they also push the systems towards their stability limits.
Another disadvantage associated with the modified PI
controllers is the possibility of distorting the line current
caused by background harmonics introduced along the feed
forward path if the grid voltage is distorted. This distortion
can in turn trigger LC resonance especially when a LCL filter
is used at the converter AC output for filtering switching
current ripple [11, 12]. Alternatively, for three-phase
systems, synchronous frame PI control with voltage feed
forward can be used, but it usually requires multiple frame
transformations, and can be difficult to implement using a
low-cost fixed-point digital signal processor (DSP).
Overcoming the computational burden and still achieving
virtually similar frequency response characteristics as a
synchronous frame PI controller, develops the P+ resonant
(PR) controller for reference tracking in the stationary
frame. Interestingly, the same control structure can also be
used for the precise control of a single-phase converter. In
brief, the basic functionality of the PR controller is to
introduce an infinite gain at a selected resonant frequency
for eliminating steady state error at that frequency, and is
therefore conceptually similar to an integrator whose
infinite DC gain forces the DC steady-state error to zero. The
resonant portion of the PR controller can therefore be
viewed as a generalised AC integrator (GI), as proven in[10].
With the introduced flexibility of tuning the resonant
frequency, attempts at using multiple PR controllers for
selectively compensating low-order harmonics have also
been reported in for three-phase active power filters, in for
three-phase uninterruptible power supplies(UPS)andin for
single phase photovoltaic (PV) inverters. From the view
point that electronic power converters will find increasing
grid-interfaced applications either as inverters processing
DC energy from RES for grid injection or as rectifiers
conditioning grid energy for different load usages, the
information aims to provide a comprehensive reference for
readers on the integration of PR controllers and filters to
grid-connected converters for enhancing their tracking
performances. To begin, the method reviews frequency
domain derivation of the ideal and non-ideal PR controllers
and filters, and discusses their similarities as compared to
classical PI control.
It is desirable to have high-power high-voltage converter
based systems available during power system faults when
they may be needed the most. If theprotectionmeasurestrip
the converter system, it can take several fractionsofanhour,
depending on the size of the converter, to discharge the dc
link and check the healthiness of the whole system. Hence,
several practical methods have been proposed and
implemented to keep a system operating under power
system faults and disturbances. A voltage-source converter
(VSC) is the main building block for flexible ac transmission
systems (FACTS) devices and, as of today HVDC technology
up to several hundred megawatts.
The increasing emergence of VSC-based transmission is
the result of development in semiconductor devices, power
electronic circuits, control, and executive engineering.
Previously, the lack of these developments had prohibited
the VSC-based technology from being the first choice. While
each development is moving forward individually,theresult
of each one influences the design criteria and application
requirements of the overall system. However,generally,less
dependence on power semiconductor characteristics
amounts to having more supplier possibilities for the VSC-
based transmission. The most important limiting factor of
power semiconductors is their switching properties since
they are usually optimized for the conduction intervals.
Hence, high-power electronic converters are desired to
operate with relatively low switching frequencies
(maximum9–15times the line frequency, and even lowerfor
multilevel converters). The low switching frequency

operation of VSC systems imposes control limitationsincase
of power system faults and disturbances when they may be
needed the most. To the best of the authors’ knowledge, in
the installed operating FACTS and HVDC systems, the ride-
through capability is obtained either by proper passive
element design [7], [8] or a change in the control mode [1].
On the other hand, with emerging high-power applications
such as 10-MW wind generation turbines [9] or
transportable recovery transformers [10], the dynamic
operation of the VSC under powersystemdisturbancesmust
be revisited. This paper discusses an alternative control
framework to obtain robust dc-link voltage with specific
attention to design the VSC controller in the back-to-back
(BTB) configuration,. The proposed controller is
implemented in the dq (rotating) synchronous reference
frame without sequence extraction.
3. Decoupled DQ-Current Control ofGrid-tiedVoltageSource
Converters Using Nonparametric Models:
Voltagesourceconverters(VSCs)areextensivelyutilized
in many grid-tied applications while either their control
system directly regulates their input current, or the adopted
control system comprises an internal current feedback loop
[7]. Therefore, the quality of the adopted current control
strategy plays an important role in the overall performance
of the system. Distributed power generationwith renewable
energy resources such as photovoltaic(PV)energyandwind
energy, HVDC systems, active power filters (APF) etc, are
some examples of such applications. Variouscurrentcontrol
approaches have been proposed for the grid-tied VSCs,
which can be categorized into two major classes [1]: (i)
linear and (ii) nonlinear controllers. Due to their structural
simplicity and fully digital implement ability, linear
approaches such as stationary reference frame (SRF) and
rotating reference frame (RRF) controllers, are more
commonly used. Among SRF controllers, proportional-
resonant (PR) regulators are very well-known and popular.
PR-controllers track AC references in the stationary frame
with zero steady state error. However, they suffer from
several drawbacks, e.g., sensitivity to grid frequency
variations, exponentially decaying transients during step
changes, and being pushed toward instability margins even
by a small phase shift introduced by the adopted current
sensors . Moreover, since PR-controllers have more poles
than PI-controllers, they introduce a greater phase lag inthe
Nyquist plot of the open loop transfer function,whichmakes
their tuning more complicated compared to PI-controllers.
Furthermore, the choice of the damping ratio of the non-
ideal PR-controllers, which are used in practice instead of
ideal PR-controllers, is itselfanotherdesignissue. Therefore,
RRF-based controllers are normally preferred in many
applications and show acceptable performance. AmongRRF
controllers, PI-based regulators are the most renowned and
easy-to-design ones, which provide fulfilling performance.
The control scheme proposed in [10] is probably the most
well-known and utilized approach in the literature. In this
method, two current axes, i.e., direct and quadrature (dq)-
axes, are defined, and a control strategy is proposed based
on the mathematical model of the system. Using feed
forward signals and PI regulators, the axes are
independently controlled. However, due to imperfect
disturbance rejectionofPI-controllers,theparametricerrors
and the mathematical modelling mismatch result in coupled
current axes. That is, upon a step change in each axis, the
other axis experiences a transient, which results in power
quality problems and performance degradation. Another
RRF-based approach is proposed in whose structure is
similar to that of the controller of , however, its axes
decoupling capability is more efficient. This current control
strategy is based on the pseudo-continuous model of the
converter, and hereinafter,it iscalledthePseudo-continuous
MultivariableCurrentController(PMCC).AlthoughthePMCC
results in very sound performance, however, even adopting
this approach, the current axes cannot be fully decoupled.
The reason is that the PMCC also relies on the mathematical
model of the system and the exact values of the system
parameters. In this paper, it is shown that the parametric
errors and modelling mismatch deteriorate the dynamic
performance of the PMCC and result in the coupling of the
dq-axes. The widespread applicability of the dq-current
controller provides substantial inducements to investigate
and propose alternative regulation schemes that can
overcome the axes coupling problem. Using nonparametric
models, this paper proposes a robust current regulation
scheme for VSCs, in which the d and q-axes are decoupled.
The design procedure consistsofanoptimization-basedloop
shaping, which guarantees the stability and the desired
performance of the closed-loop system [8,9]. The structure
of the achieved controller is similar to that of the PMCC [12],
however, instead of integrators in the transversal arms of
the controller, PI-controller are used. The proposed method
provides better dynamic performance compared to its
predecessors and is structurally simple.
III. RESULT ANALYSIS OF METHODS
1. A Novel Self tuning PI Control system
The CSPI controller is applied to the microgrid system
shown in Fig. 1. Simulations are carried out in
MATLAB/SIMULINK and to verify the feasibility of the
proposed controller different operational scenarios in the
grid connected mode and a case in islanded mode have been
studied. To show the superiority of the CSPI algorithm
compared to conventional PI control, the results from the

proposed controller are compared to those of traditional PI
control in the first four cases. The PI coefficients are the best
of the two sets of gains obtained from the Root Locus and
Ziegler Nichols algorithms. The gains obtained from Root
Locus algorithm resulted in a better response and then a
manual tuning was employed to achieve to the best
response.
Fig1:Single line diagram of microgrid system.
Case1: Grid connected microgrid:
Fig2: Real and Reactive power tracking under CSPI
control over transient period
Fig3: Three phase current of DG1 under CSPI control &
Three phase current of DG1 under PI control
As seen, in the CSPI control a smooth transient
response is achieved and the output reaches its steady state
within 0.02 s which is in the reasonable margin of 2–3cycles
for a power system and is almost one fourth of the transient
time of 0.075 s in PI control. Also the initial power
oscillations that are observed in PI control are completely
removed for real power and significantly reduced for
reactive power. It is worthnotingthatinpractical microgrids
large power oscillations can lead to the unwanted activation
of protection systems, inwhichcasethemicrogridDGsmight
fail to maintain the desired supply to loads. This can violate
the overall system reliability and stability. Comparing both
figures in fig. 2, it is evident that the proposed CSPI control
achieves to a high level of harmonics attenuation compared
to traditional PI control and the desired current shape is
maintained.
Case2: Islanded Microgrid:
Fig4: Real and reactive power under PI control & Real and
reactive power under CSPI control.
zFig5:Three phase current in DG1 under PI Control &
Three phase current in DG1 under CSPI control.
In the islandedmicrogrid,theCSPIcontrollerisused
in conjunction with frequency and voltage droops and an
outer voltage control loop. As in this mode, the utility grid is
not present, the major control tasks are voltage and
frequency control. In the droop control, the frequency and
voltage deviations are limited within a maximum allowed
range to achieve proper active and reactive power sharing
among multiple DGs. During the simulations switch S2, S1
and the switch at PCC are open and therefore an islanded 3-
DG microgrid with two constantimpedanceloadsrated at15
kW and 3 kVAR is studied. As seen, the proposed CSPI
controller results in significantly reduced initial power
overshoot and steady state error; hence a faster and more
accurate control is yielded.
2. PR Controllers and Filters.
Given the advantages of PR controllersandfilters,a number
of applications have since been proposed in the literature
with most focusing on the control of converters interfaced
directly to the utility grid. In this Section,oneexamplecaseis
presented for demonstrating the effectiveness of using PR
controllers in a single-phase PV converter.
Single-phase PV grid-connected inverter
Single-phase grid inverters are commonly used in
applications like residential RES (typically PV or fuel cell
systems) and UPS. Figure shows a typical RES where the
DC-link voltage, active P and reactive Q powerarecontrolled
in the outer control loops (labelled as voltage controller and

reference generator in the Figure). The reference current
outputs of the outer loops (idd
+ and i+) are next tracked by an
inner current loop whose output is eventually fed to a PWM
modulator for switching the inverter.
Typically, the inner current loop is implemented
using a stationary PI current controller with voltage feed
forward, as shown in Fig.6. Using PI control, however, leads
to steady-state current error (both in phase and magnitude)
when tracking sinusoidal input, and hence a poor harmonic
compensation performance is expected [9]. Synchronous PI
control can mitigate the tracking error, but is generally
difficult to apply. Instead, the equivalent stationary PR
controller can be used as the inner current controller, as
shown in Fig. 10b. Compared to a stationary PI controller,
the only computational requirement imposed by the PR
controller is an extra integrator for implementing a second-
order system, but with a modern low-cost 16-bit fixed-point
DSP, this increase in computation can generally be ignored
[9]. Besides that, using a PR controller would allow the
removal of the grid voltage feed forward path, as proven in
[9], and the simple cascading of a HC compensator for
eliminating selected low-order harmonics.
Fig6: Block diagram of typical single-phase RES system
Fig7: Single phase Grid Inverter control. a)Stationary PI
control b)Stationary PR inner current control
The designed control scheme in Fig.6 has been
tested using an experimental 3kW PV full-bridge inverter
with an output LCL filter, as shown in Fig 5. The inverter is
powered from a regulated DC power supply (set to
UD=350V) for simulating a PV string, and is interfaced to the
utility grid with a voltage of Ug=230VRMSanda background
THD of 1.46%. The resulting system is controlled digitally
using a 16-bit fixed-point TMS320F24xx DSP platform with
an execution time of 40 ms (includingHCcompensation) and
the controller gains set as Kp=2, Ki=300 and Kih=300 for h=3,
5 and 7. With these settings, the grid current and voltage at
50% load using PR and PR+HC controllers are shown in
Figures. As seen in Fig., there is no phase error noted
between the grid current and voltage, confirmingtheproper
functioning of the PR controller. The harmonic distortion in
Fig. 7 can be further reduced by cascading an HC
compensator, as demonstrated by the smoother current
waveform in Fig. 8.
Fig.8: Schematic Representation of experimental single
phase PV Inverter.
Fig.9: Waveforms captured using PR controller at 50%
load Grid voltage Ch2 [100V/div.], grid current Ch1
[5A/div.] and DC voltage Ch4 [250V/div.]
Fig.10: Waveforms captured using PR+HC controller at
50% load Grid voltage Ch2 [100V/div.], grid current Ch1
[5A/div.] and DC voltage Ch4 [250V/div.]
The improved performance achieved here with a single
phase inverter can obviously be extended to a three-phase
RES (e.g. small wind or water turbines and high-power PV
plant).

3. Vector-Controlled Voltage-Source-Converters
This section presents and evaluates the dynamic
performance of BTB VSC systems in different applications.
Applications are categorized as HVDC, drive (wind), and
hybrid power system applications. The proposed controller
has been implemented in RTDS and compiled on a GPC
processor card with a controller sampling time of 50μs and
key circuit parameters as tabulated in Table.
Fig.11: The unbalanced system is represented by a 50%
voltage drop in phase A & the unbalanced system is
represented by a 50% voltage drop in phase B
In Fig. 10, the unbalanced system is represented by a 50%
voltage drop in phase A of the PCC voltages Vabc1.The fault
remains for six cycles. A can be observed the dc-link voltage
remains practically stiff, and harmonic measurement of the
dc-link voltage for the second and fourth harmonicsshows a
satisfactory level of compensation. The inverter currents
change (increase in phase A) to compensate for the
unbalanced power flowing into the system; however, it
remains within the safe operating area of the switches. In
Fig. 11, the system performance under more severe
imbalances is shown represented by a 50%and90%voltage
drop in phases B and C of the power flow controller side,
respectively.
Fig.12: roller side and drive side operating at 30Hz.(RTDS
results.) & Dynamic performance of the BTB VSC system in
drive (wind) applications under an unbalanced condition
of 50% in phase B and 30% in phase C voltage sag in the
dc-link controller side (grid) and drive side operating at
30 Hz. (RTDS results.)
The dynamic performance of the VSC BTB system in the
drive application under unbalanced conditions in the
inverter side or grid side (dc-link voltage controller
converter) is presented in Fig.11. The unbalance system is
represented by 50% voltage drop in phase B of the inverter-
side PCC. And a more severe fault occurs which is
represented by a 50% voltage sag in phase B in addition to a
30% voltage drop in phase C of the grid. From the results
obtained, it can be concluded that the proposed control
scheme is effective in maintaining a robust dc-link voltage
even under unbalanced grid conditions. In other words, the
drive side in this case of high power wind generators with a
BTB VSC interface can operate independently from the grid
disturbances translating to higher availability of the
renewable resources.
Fig.13: Schematic of microgrid interfaced to utility grid
using power quality compensator
Fig.14: Voltage control scheme of shunt inverter A
4. Decoupled DQ-Current Control of Grid-tied Voltage
Source Converters Using Nonparametric Models
The Optimization-based Multivariable-PI Current
Controller (OMCC) is based on minimizing the difference
between the open-loop transfer matrix of the system and a
desired open loop transfer matrix. The minimization is
subject to some convex constraints and forms a convex
optimization problem.
Performance Evaluation:
The purpose of this section is to evaluate the performance
of the OMCC and also to compare it with that of the PMCC
both in simulation and experiment. For the simulation part,
the system of Fig.13, whose parameters are set according to
Table I, is simulated in MATLAB/PLECS environment.

Fig.15: A one-line diagram of the three-phase test system.
TABLE1: Parameters of the three-phase test system
To convert the control and signal processing commands in
discrete time, the bilinear method is used. The continuous
time transfer functionsarethentransformedtodiscretetime
and are developed into a S-function in MATLAB/Simulink.
Moreover, for the experimental part, adopting the test
system of Fig. 14 and the correspondingparametersofTable
I, a three-phase experimental setup is implemented. To
implement the control strategies and signal processing
algorithms, an FPGA-based controller, which provides a C
programming environment, is used. The adopted FPGA is
XC4010PC84FPGAmanufactured byXILINX.Thecontrol and
signal processing algorithms are first discretized using the
bilinear method and then developed into C codes. It should
be noted that in order to achieve perfect isolation between
the power and control circuits, optic-based gate drivers are
adopted for driving the IGBTs.
For the performance evaluation, a reference tracking test
is conducted for each control strategy. Figs. 16 (a) and (b)
depict the reference signals utilized in the conducted tests
for the d and q-axes, respectively. It,d,ref consists of three step
changes, i.e., 0 to 10 A, 10 to 6 A, and 6 to 0 A, at the time
instants t = 0.01s, t = 0.02s, and t = 0.045s, respectively.
Moreover, It,q,ref experiences two step changes from 0 to 8 A
and from 8 to -10 A, at the time instants t = 0.03s and t =
0.055s, respectively. To design the PMCC, for the simulation,
the parameters of Table I are adopted. Moreover, for the
experimental part, the measuredvaluesoftheresistance and
inductance of the filter along with those of the transformer
are used, which are measured by an RLC meter. The
experimentally measured values comply with thoseofTable
I. The difference between the measured value of Rt = 0.3 Ω
and the identified value of Rt = 1.1 Ω represents the
parametric errors and also the un modeled resistive
elements of the system. Therefore, to design the OMCC for
the simulation and the experimental case, the simulation-
based nonparametric model corresponding to Rt = 1.1 Ωand
the experiment-based nonparametric model of Fig. 13 are
adopted, respectively. Moreover, to mimic the experimental
case in the simulation part, although the PMCC is designed
for Rt = 0.3 Ω, Rt is set to 1.1 Ω in the simulated system. Of
course, it must be noted that using a try and error approach,
one can tune the PMCC to achieve better results and to
reduce the error between the real value of Rt and the value
used for the design. However, the purpose of this paper is to
demonstrate the deficiencies of the PMCC, while
systematically designed, compared to the OMCC. To design
the controllers in thefollowingsections, thesamebandwidth
is selected for all cases, which is ωc = 1000rad/ s . Moreover,
the weighting filter adopted for the OMCC designisW1(jω)=
0.5. This value guarantees a gain margin of at least 2 and a
phase margin of greater than 29 degrees.
Fig.16: Reference value for the inverter current: a) d-
component & b) q component
A. Simulation Results
In the simulation part, Rt is set to 1.1 Ω. However, in
order to mimic the parametric errors of a real scenario, for
the PMCC design, Rt is assumed to be 0.3 Ω. Then, adopting
the aforementioned design parameters, the PMCC is
designed, and it is represented by
Moreover, utilizing the simulation-based
nonparametric model corresponding to Rt =1.1 Ωandtaking
the aforementioned designprocedure,theOMCCisdesigned,
which is represented by

Fig.17: Simulation results of the transient response of the
response PMCC during step changes in d- and d-axis: a)
line currents, b) dq-component of the currents, and c) dq-
components of and the control signals. & Simulation
results of the transient of the OMCC during step changes a)
line currents, b) dq-component of the currents, c) dq-
components of the control signals.
In the following, the performance of the designed
controllers of are respectively evaluated in simulation.
1) Reference Tracking Test for PMCC: In this section,
adopting the aforementioned reference tracking test
scenario in MATLAB/PLECS environment, the performance
of the PMCC is evaluated. It is shown that due to the
parametric errors and modeling mismatch, the dynamic
performance of the PMCC deteriorates, and subsequent to
step changes in the d-axis (qaxis), a non-negligible transient
is experienced in the q-axis (d-axis). This verifies that the
PMCC may not fully decouple the current axes. It must be
noted that even if the parameters could be precisely
determined, the modeling mismatch of the system, e.g., the
modeling mismatch of the PWM converter, would result in
slightly coupled current axes.
The test signals of Figs. 15(a) and (b) are applied to
It,d,ref and It,q,ref, respectively. Upon each step change, the line
currents and their corresponding dq-components,asshown
in Figs. 16 (a) and (b), change to track the reference signals
in less than 2 ms with zero steady state errors. However, Fig.
16 (b) shows that upon each step change in each axis, the
other axis experiencesa transient whosemagnitudedepends
on the amount of the step change and verifies that the PMCC
suffers from axes coupling problem. Fig. 16 (c) depicts the
dq-components of the control signals, which subsequent to
each step change, are adaptedtothenewoperatingpointsby
the PMCC.
2) Reference Tracking Test for OMCC: Utilizingthesystemof
Fig. 1 equipped with the OMCC of and applying the test
sequences of Figs. 15 (a) and (b) to It,d,ref and It,q,ref,
respectively, the same test is conducted for the OMCC to
evaluate its performance. Subsequent to each change in the
d- or q-axis reference value, the OMCC regulates the line
currents at the desired level in less than 2 ms, as shown in
Figs. 17 (a) and (b). Fig. 17 (b) depicts the dq-components of
the line currents, which subsequent to each step change, are
regulated at the desired values as fast as the PMCC case and
with zero steady state errors. Moreover, Fig. 17 (b) verifies
that upon each step change in the d-axis(q-axis),the q-
component(d-component)of the currents does not
experiences any transient, which confirms the full axes
decoupling capability of the OMCC. Fig. 17 (c) depict the dq-
components of the control signals.
Tonumericallycomparethe decouplingcapabilityof
the controllers, the second norm of the errorind-axisduring
a step change in the q-axis is calculated. The utilized
criterion is
summarized as . Calculating this norm
for t1 = 0.055 s and t2 = 0.075 s, d = 3.82 for the PMCC, while
d = 0.001 for the OMCC, which confirms more effective
decoupling capability of the OMCC compared to that of the
PMCC. Moreover, although the rise time of both controllers
are the same, the OMCC settling time is much less than that
of the PMCC. For example, subsequenttothelaststepchange
in the q-axis reference value, the OMCC reaches the ±5%
steady state boundary within 2 ms while it takes 20 ms for
the OMCC to reach that.
B. Experimental Results
In this section, the performance of the OMCC is
experimentally evaluated and is compared with that of the
PMCC. To design the PMCC, the filter parameters are of most
importance and are measured using an RLC meter. The
measured inductance and resistance of the line reactorfilter
and the transformer are Lt = 5.0 mH and Rt = 0.3 Ω, whichare
used to design the PMCC. However, to design the OMCC, the
nonparametric model of the system is obtained through a
PRBS-based identification on the experimental setup. Note
that although the measured Rt is 0.3 Ω, however, the
experiment-based identified nonparametricmodel complies
with a theoretical model corresponding to Rt = 1.1 Ω.
The designed controllers in discrete time are
represented by

In thefollowing,usingtheaforementionedreference
tracking test, the performance of the designed controllers
are experimentally evaluated.
Fig.18: Experimental results of the transient response of the
PMCC during step changes in d- and d-axis: a) line currents,
b) dq-component of the currents, and c) dq-components of
the control signals & Experimental results of the transient
response of the OMCC during step changes in d- and d-
axis:a) line currents, b) dq-component of the currentsand c)
dq-components of the control signals.
1) Reference Tracking Test for PMCC: In this section, the
performance of the PMCC is experimentally evaluated in
terms of tracking reference signals. It is shown that in
accordance with the simulation results, utilizing the PMCC,
the dq-axes of the current are not fully decoupled, mainly
due to the parametric errors and modeling mismatch.
To experimentally evaluate the performance of the
PMCC, the reference tracking test is conducted. Figs. 17 (a)
and (b) show the inverter current and their dq-components,
respectively, which follow the step changes in the reference
values in less than 2 ms. However, subsequent to each step
change in the reference value of the d-axis (q-axis) current,
the q-axis (d-axis) current experiences a non-negligible
transient. Note that the dq components of the currents
contain 300 Hz ripples, which are causedbythedead-timeof
the PWM strategy. Fig. 17 (c) shows the dq-components of
the control signals that upon each step change, vary to
accommodate the requested change. This results confirms
that the PMCC may not fully decouple the axes due to the
parametric errors and modeling mismatch.
2) Reference Tracking Test for OMCC: Utilizing the
implemented experimental setup equipped with the OMCC
and applying the same reference tracking test, the
performance of the OMCC is experimentally evaluated. Fig.
17 (a) and (b) depict the line currents and their
corresponding dq-components, respectively. Fig. 17 (b)
shows that subsequent to each step change in the reference
value of the d- or q-axis, the OMCC regulates the line
currents at the desired level in less than 2 ms. Moreover, it
confirms that upon a step change in each axis, the current of
the other one remains unchanged. Fig. 17 (c) depicts the dq-
components of the control signals,whichsubsequenttoeach
step change, are adapted to the new operating points by the
OMCC.
The presented test results confirm the simulation
results and show that the proposedcontroller,i.e.,the OMCC,
(i) is capable of tracking reference signals with zero steady-
state errors, (ii) has fast dynamics, and (iii) contrary to the
PMCC, provides fully decoupled d- and q-axes of currents.
Thus we can say that this method relies on shaping
the nonparametric open-loop and closed-loop transfer
matrices of the system through constrained convex
optimization. The adopted constraintsensurethestability of
the closed-loop system and also result in the desired
performance.
5. Hysteresis Current Controller for a Microgrid
Application
a) System description.
Voltage source inverter based three phase grid
connected inverter with control circuit is utilized in this
paper. It acts as a shunt active power filter and is connected
in parallel with the harmonic producing loads at the Point of
Common Coupling (PCC). Configuration of the system is
shown in Fig 1. Shunt active power filter generates a current
equal and opposite to that of harmonic currentdrawn bythe
load and injects it at the point of common coupling making
the source current sinusoidal. Filtering algorithm used and
calculation of load current harmonics decides the
characteristicsofharmoniccompensation.Referencecurrent
extracted by Synchronous reference method is compared
with the filter current in the hysteresis loop and
corresponding pulses are given for inverter switching.
Inductor provides smoothing andisolationofhighfrequency
components. Desired current waveform is obtained by
controlling the switching of IGBT switches in the inverter. A
nonlinear load of 10 kW is connected to a threephasesource
of 415 V. PV system with a capacity of 3.3 kW is
interconnected to the grid through an inverter. PV system is
equipped with MPPT control, DC-DC converter and a
bidirectional inverter.
Fig.19: System Configuration.

b) Simulation Results:
The PV system is modeled usingMATLABSimulink platform.
Table2 shows the parameter details of a PV panel.
Table 2 Parameter details of a PV panel
Parameter Set Value
Rated output Power 220 W
Short circuit current 10 A
Open circuit voltage 22.2V
Current at maximum power 9.9 A
Voltage at maximum power 17.2 V
Fig.20: Waveform of source voltage, source current, load
current
The PV system with MPPT control algorithm is
integrated to grid with hysteresis current controller with
same nonlinear load of 10 kW.Real powersharing,harmonic
elimination and power factor improvement are analyzed.
The current waveformafterthecompensationisshownin fig
19.
Fig.21: current waveform after compensation
It is inferred that PV system andgridsharestheload
and also the harmonics in the source current are eliminated.
The source voltageandcurrentwaveformforpower
factor analysis is shown in fig 21. Power factor is improved
after a delay of 0.15 seconds.
Fig.22: Power factor improvement
When an additional load of 5 kW is added to the
system at 0.4 sec, grid provides the power required for the
additional load since maximum capacity of PV generator has
reached. Fig 22 shows the power sharing between PV and
grid.
Fig.23: Power sharing when additional load of 5 kW is
added
When the nonlinear load connected to the systemis
1200W, the PV itself can support the load. So no power is
drawn from the grid. The fig 23 shows the power sharing
waveform when the load is 1200 W.
Fig.24: power sharing waveform when load demand is
1200 W
If any unfavorable situation occurs,ie,whenthereis
no supply from PV generators, the grid supports the entire
load demand. The power sharing of the system in the
absence of PV is shown in fig 24.

Fig.25: Power sharing in the absence of PV system
From the simulation doneforthevariousconditions
of load, it is found that the THD of system source current is
improved from 26.62%to3.84%aftercompensation.ThePV
system was capable of injecting 3.2 kW power to the system.
IV. CONCLUSION
Thus we have discussed differenttypesofstrategies
and methods of controlling power in Voltage Source
Converter based Microgrid power system. In CSPI based
controller the results verifytheeffectivenessoftheproposed
algorithm and provethatthe CSPI controllerhassignificantly
improved the microgrid performance by reducing the
convergence time, power oscillations, and tracking error
especially in case of higherDGpenetrations.Implementation
wise, the PR technique requires lesser computational
overhead and does not require an explicit grid voltage feed
forward control path, while still achieving the same
performance as a synchronous PI controller. The vector
controlled VSC's obviates the need for the sequence
extraction blocks or the resonant compensators. Therefore,
there is no diminishing bandwidth factor. The RTDS
verification of the controller attained less than 1% dc-link
voltage deviation under most common faults and
disturbances. The proposed optimization-based method
tracks the reference signals with zero steady state errors,
has better dynamic performancecomparedtothepreviously
proposed approaches and results in decoupled currentaxes.
Lastly in Hysteresis current controller based system the PV
system is tested and results confirm better real power
sharing, power factor improvement and harmonic
elimination.
V. REFERENCES
[1] M. Prodanovic and T. C. Green, “Control and filter design
of three-phase inverters for high power quality grid
connection,” IEEE Trans. Power Electron., vol. 18, no. 1, pp.
373–380, Jan. 2003
[2] N. Kroutikova, C. A. Hernandez-Aramburo, and T. C.
Greean, “Statespace model of gridconnected invertersunder
current control mode,” IET Elect. Power Appl., vol. 1, no. 3,
pp. 329–338, 2007.
[3] K. Sao and P. W. Lehn, “Control and power management
of converter fed microgrids,” IEEE Trans. Power Syst., vol.
23, no. 3, pp. 1088–1098, Aug. 2008.
[4] C. Schauder and H. Mehta, “Vector analysis and control of
advanced static VAr compensators,” IEEE Proc. C Gen.
Transmiss. Distrib., vol. 140, no. 4, pp. 299–306, Jul. 1993.
[5] F. Katiraei and M. R. Iravani, “Power management
strategies for a microgrid with multiple distributed
generation units,” IEEE Trans. Power Syst., vol. 21, no. 4, pp.
1821–1831, Nov. 2006
[6] N. Pogaku, M. Prodanovic, and T. C. Green, “Modeling,
analysis and testing of autonomous operationofaninverter-
based microgrid,” IEEE Trans. Power Electron., vol.22,no.2,
pp. 613–625, Mar. 2007.
[7] D. N. Zmood and D. G. Holmes, “Stationary frame current
regulation of PWM inverters with zero steady-state error,”
IEEE Trans. Power Electron., vol. 18, no. 3, pp. 814–822,May
2003.
[8] R. Teodorescu, F. Blaabjerg, M. Liserre, and P. C. Loh,
“Proportionalresonant controllers and filters for grid-
connected voltage-sourceconverters,”IEEProc.Elect.Power
Appl., vol. 153, no. 5, pp. 750–762, Sep. 2006.
[9] G. Shen, X. Zhu, J. Zhang, and D. Xu, “A new feedback
method for PR current control of LCL-filter-based grid-
connected inverter,” IEEE Trans. Ind. Electron., vol. 57,no.6,
pp. 2033–2041, Jun. 2010.
[10] F. Blaabjerg, R. Teodorescu,M.Liserre,andA.V. Timbus,
“Overview of control and grid synchronization for
distributed power generation systems,” IEEE Trans. Ind.
Electron., vol. 53, no. 5, Oct. 2006, pp. 1398–1409
[11] G. K. Venayagamoorthy, “Potentials and promises of
computational intelligence for smart grids,” in Proc. IEEE
Power Energy Soc. Gen. Meeting, Calgary, AB, Canada, Jul.
2009, pp. 1–6.
[12] G. Rami, T. Tran-Uoc, and N. Hadjsaid, “Fuzzy logic
supervision and control of distributed generators,” in Proc.
18th Int. Conf. Exhibit. Elect. Distrib. (CIRED), Turin, Italy,
2005, pp. 1–5.
[13] N. L. Diaz, F. H. Barbosa, and C. L. Trujillo, “Analysis and
design of a nonlinear fuzzy controller applied to a VSC to
control the active and reactivepowerflow,”inProc.Electron.
Mech. Autom. Conf., 2007, pp. 417–422.

IRJET- Different Control Strategies for Power Control of Voltage Source Converters in a Microgrid

More Related Content

What's hot (20)

Similar to IRJET- Different Control Strategies for Power Control of Voltage Source Converters in a Microgrid (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Different Control Strategies for Power Control of Voltage Source Converters in a Microgrid