See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/224341475

Hierarchical control of Hybrid Power Systems

Conference Paper · September 2008

DOI: 10.1109/CIEP.2008.4653837 · Source: IEEE Xplore

CITATIONS READS
21 375

2 authors:

Manuel Eduardo Torres-Hernández Miguel Velez-Reyes

Mexican Institute of Social Security University of Texas at El Paso
5 PUBLICATIONS   74 CITATIONS    215 PUBLICATIONS   2,167 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Integrating Spatial and Spectral Information for Hyperspectral Unmixing View project

Advanced Unmixing of Hyperspectral Remote Sensing Data for Space Situational Awareness View project

All content following this page was uploaded by Miguel Velez-Reyes on 29 January 2016.

The user has requested enhancement of the downloaded file.

 Hierarchical Control of Hybrid Power Systems

María E. Torres-Hernández, Student Member, IEEE, and Miguel Vélez-Reyes, Senior Member, IEEE.
Center for Power Electronic Systems, University of Puerto Rico at Mayaguez
E-mail: maria.torres@ece.uprm.edu, m.velez@ieee.org

Abstract two-level hierarchical controller for a hybrid EPDS. At

This paper presents a Hierarchical controller for a
DC bus configuration Hybrid Power System consisting
of a Wind Turbine, Photovoltaic Panels, a Battery
Bank, capability for connection to grid, and load. The
proposed hierarchical control system consisted of lower
level controllers for each generation and storage
component designed using Sliding Mode Control and a
supervisory controller that had the objective of
maximizing use of the renewable sources while
minimizing connection to the grid. The supervisory
controller changed sliding surfaces and set points
according to the operation conditions and performed
load shedding based on load priorities when available
generation is insufficient. The functionality and
performance of the system is studied by means of
simulations. The paper will present HPS model
development, hierarchical controller design, and
simulation results under the different operating
scenarios. Simulation results show that the proposed
controller has the capability to meet the design
objectives with acceptable performance.
the lowest level, local nonlinear sliding mode control is
used to control the renewable sources and the battery
storage system while the second level is a supervisory
controller that changes the set points and sliding
surfaces according to the load demand and available
generation capability. The structure of the supervisory
system allows the potential implementation of different
modes of operation to meet different control objectives.
The work presented here builds on the work of [1].
In this paper, the proposed supervisory controller has
five modes of operation that maximizes use of
renewable generation capability while minimizing
connection to the grid with an improved battery
management scheme [2]. It also provides capability for
the system for load shedding when the system is
islanded and the generation capability of the renewable
resources is not enough to supply the demand.

1 Introduction
Hybrid Power Systems are power systems that
combine different electric power energy sources. An
important problem in hybrid power systems is the
control and supervision of the power distribution
system. The dynamic interaction between the Figure 1. Hybrid Generation System
renewable sources, the grid, the loads and the power
electronic components and the system low inertia can 2 Hybrid power system
lead, to new system, critical problems stability and
power quality different from those in conventional
The Hybrid Power System studied in this work
power systems.
combines wind energy, photovoltaic system, and is
Electronic Power Distribution Systems (EPDS) are
connected to the grid. A schematic of the overall
power distribution systems where the electric power
hybrid system studied here is shown in Figure 1. The
flow is controlled using power electronic converters.
HPS can be divided in six components: (i) generation
EPDS are present in many applications such as ship
(Wind turbine, PV Panels, and the Grid), (ii) energy
power systems, electric hybrid vehicles, and hybrid
storage (Battery Bank), (iii) power bus, (iv) electric
power systems among others. Advanced control
load, and (vi) power electronic converters. The dc bus
techniques can improve the performance of such
collects the energy generated by both units and delivers
systems by improving energy management capability
it to the load and, if necessary, to the battery bank. ; the
and its capability to adapt to faults and other significant
converters control the operation point of the wind
changes in operating conditions. This paper presents a
turbine and PV Panels. The loads served consist of

978-1-4244-2718-5/08/$25.00 ©2008 IEEE 169

three similar zones. Each zone is connected to the bus
by a DC/DC converter.
The wind generation unit comprises a windmill, a
multipolar permanent-magnet synchronous generator
(PMSG), a rectifier, and a dc/dc converter to interface
the generator with the dc bus. The solar unit comprises
several panels coupled to the dc bus through a DC/DC
converter. The dc bus collects the generated energy and
delivers it to the load and, if necessary, to the battery
bank.
The voltage of the dc bus is set by the battery bank,
which comprises lead-acid batteries coupled in a
serial/parallel array; the converters control the operation
point of the wind turbine and PV Panels.
The loads served consist of three similar zones.
Each zone is connected to the bus by a DC/DC
converter.
3 Energy sources models
3.1 Wind subsystem
The wind generation unit comprises a windmill, a
multipolar permanent-magnet synchronous generator
(PMSG), a rectifier, and a dc/dc converter to interface
the generator with the dc bus. The mathematical model
used in the simulation was obtained from [3].
The mechanical power and torque generated by a
where and are, respectively, the quadrature and
direct currents; is the electrical angular speed; L and
are the per-phase inductance and resistance of the
stator windings; J is the total inertia of the system; P is
the number of poles of the PMSG; and is the flux
linked by the stator windings.
The voltage is externally imposed by the DC/DC
converter as a function of the duty cycle, ux, and is
described by:
(6)
If we assume an ideal static conversion, the current
at the output of the DC/DC converter is given by:
(7)
3.2 Solar subsystem
The photovoltaic power model was proposed in [4].
This model depends on several cell parameters an on
variable environment conditions such as the
temperature over the solar panel T, the characteristic
constant for the I-V curves b, the percentage of
HIIHFWLYH LQWHQVLW\ RI WKH OLJKW RYHU WKH VRODU SDQHOV Į
the open circuit voltage Voc, the short circuit current Isc,
and a shading linear factor Ȗ7KLVPRGHOZDVVHOHFWHG
because it uses the electrical characteristics provided by
the manufacturers in the solar panel data sheet, and its
WXUELQHLVSURSRUWLRQDOWRWKHDLUGHQVLW\ȡWKHSRZHU electrical behavior could be modeled by a nonlinear
coefficient of the rotor (Cp), the cube of the wind speed current source connected with the intrinsic cell series
ȞDQGWKHVZHSWDUHD$DVVKRZEHORZ resistance. The I-V relation of the solar panel is:
(1)
The power coefficient Cp( ), depends upon the (8)
aerodynamics of the rotor blades, the blades angle, and where
the wind velocity. Cp is a function of the tip-speed
ratio , where is the blade length, and is (9)
the angular shaft speed.
(10)
The torque produced by the wind turbine is given
by:
(11)
(2)
(12)
where, , is the torque coefficient of the
turbine. (13)
The PMSG dynamic model in a rotor reference The PV panel is connected to the DC bus through a
frame is given by: DC/DC buck converter [5] described by
(14)
(3)
(15)

(4) where is the battery voltage, is the current injected

on the DC bus, is the voltage level on the PV panel
(5) array terminals, and is the control signal.

170
3.3 Battery bank
The Li-Ion battery model that used here was
developed by [6]. This model is capable of modeling
the steady state and the dynamic behavior of the
battery.
Figure 2. Battery model adapted from [6].
The battery model is described by two circuits
coupled via a current controlled current source and a
nonlinear voltage controlled voltage source as shown in
. Insufficient Wind Regime.
The state of charge (SOC) of the battery is
represented by a large capacitor ; the model of the
transient behavior and voltage-current relationship of
the battery was represented by two RC circuits and
series resistance on the right hand side of the diagram.
The state space model of the circuit is given by [8]:
(16)
where and are the capacitance and resistance in
the long transient RC circuit, and are the
capacitance and resistance in the short transient RC
circuit, is the series resistance, and g is the nonlinear
SOC function. The input u is the current entering the
battery, and the output y is the voltage across the
battery terminals.
4 Operation modes
4.1 SMC of wind subsystem
The main control objective for the wind subsystem is
to control the power produced into the DC bus to satisfy
the total demand (the load plus the battery charge
requirements ).
For this subsystem, there are two modes of operation
as described in [3]. The first model occurs when the
wind turbine produces enough power to satisfy the total
171
power demand. We denote this mode as the Sufficient
Wind Regime.
(17)
The second mode of operation occurs when the
maximum energy captured from the wind is not enough
to satisfy the total demand. In this case, the output
power reference is the optimum power extracted by the
turbine minus losses.
(18)
where
(19)
is the tip speed ratio that maximizes the power
extracted from the wind. We denote this mode as the
The boundary between the operation modes for the
wind subsystem is given by the angular shaft speed at
the boundary point that is obtained by equating the
mechanical power references for both modes of
operation.
(20)
Figure 3 shows the operation points of both sliding
surfaces. The decision to work in one mode of
operation or the other can be taken by comparing the
measured speed with the boundary speed as
follow:
Figure 3 Operation Points of both sliding
surfaces [7].
4.1.1. Sufficient wind regime. In this mode, the main
objective of the wind subsystem is to regulate the
power produced by the wind turbine following .
To accomplish this objective, the sliding surface for the
SMC was written in terms of currents on the battery
bank.
(21)
The switched control signal produces a deviation in
Te, and a high ripple on the current of the DC/DC
converter. Following [3], we included an integrator
prior to the input ux to turn it into a new state variable
and set the signal w as the new input to the system. As
a consequence, the algebraic dependence between the
input of the system and the sliding surface is broken.
The dynamic model of the extended system is shown
below [3].
(22)
Maximum Power Operation Point (MPOP). In MPOP,
the instantaneous and incremental conductance of the
array has a different sign and the same absolute value.
This condition arises from [4]:
(27)
(28)
The decision to work in one mode of operation or
the other is taken by comparing the power that the array
would generate if operating at the MPOP and .
This is illustrated on Figure 4, and could be expressed
as a fictitious power in the following
way [5]:

where the electromagnetic torque produced by the 4.2.1. Sufficient power generation conditions. In this
PMSG is given by and the mode, the cell temperature and the isolation are
switched control law w is given by: sufficient to satisfy . To accomplish this objective,
the sliding surface was written in terms of currents on
the battery bank [5].
(29)

where , ,
, and

4.1.2. Insufficient wind regime. In this mode, the

system is not able to operate on the sliding surface h1,
defined previously, and it is necessary to incorporate a
secondary sliding surface h2 whose main objective is to
extract the maximum power available in the wind.
(24)
and the switching signal is given by
Figure 4 Zones of operation for control modes
(25) under sufficient power generation conditions
[5].

4.2 SMC of the solar subsystem and the switch control signal is:

There are two modes of operation following [5]. (30)

The first occur when the PV array is capable of
generating enough power to satisfy the total power
demand.
(26)
The second mode of operation occurs when the
Panel is unable to generate enough power to supply the
power reference. In this mode, the PV operates on the
4.2.2. Insufficient power generation conditions. The
control objective is to operate the PV cell at the MPOP
because the cell temperature and the isolation are
insufficient to satisfy . In order to satisfy the
equation 27 the sliding surface must be [5]:
(31)

172
and the switched control signal must be: Only through the Grid: if the battery bank is
completely discharged, i.e. , the Grid
(32) must supply the extra power needed. The strategy used
is called slope-on strategy (see ).
in this case, the battery bank recharging is made at
instead of .

4.3 Control strategy for the battery bank and

the grid

For the Battery Bank, we set two modes of

operation. The first occurs when the battery current, Ib,
is positive (charging mode), and the second occurs
when the battery current is negative (discharging
mode).

4.3.1. Charging mode. In this mode, the power

generated by the renewable energy system is higher
than the power required by the load, so the extra power
is handled as follow [2]:

The extra power will be stored in the Battery Bank
as long as SOC is less than and is a
smaller amount than .
If , then the Battery bank is charged
by , and the extra power is reduced from the
PV and Wind subsystems through the individual
control units.
If , means that the battery bank is
fully charged and the extra power must be reduced
from the PV and Wind subsystems, i.e. the
component of the reference signal related to the
battery bank input to the controllers is set to zero.
4.3.2. Discharging mode. In this case, the load demand
is higher than the power available from the Wind and
PV subsystems, so the battery bank and the Grid must
supply the extra power needed. Depending upon the
SOC of the Battery Bank and the magnitude of ,
supplying the extra demand is complete according to
the next hierarchical order [2]:
Only through the Battery Bank as long as these
conditions are satisfied:
(33)
Through the Battery Bank and the Grid: this occurs
when the load demand is higher than the battery
discharge limitations, so the Grid is able to provide the
extra current. The conditions that have to be satisfied
are:
(34)
Figure 5 Implementation of slope-on
strategy [2].
The grid start level occur when the SOC is equal to a
level higher than Qmin, defined by the battery; when that
level Qmin*X is reached, the grid starts taking the load
progressively from the battery. The grid current starts
increasing with the same slope so that the total current
contribution from in cooperation battery bank and the
grid making .
4.4 Load shedding
This control mode was added to connect or
disconnect a particular load or zone from the total
power demand. The main reason to do that is to protect
the generation sources when there is insufficient
generation capacity to meet the demand. This occur
when the system is isolated from the grid, the battery is
discharged, and the load demand in higher than the
maximum power that the wind turbine and photovoltaic
panels can produce.
If the total demand is higher than the total available
generated power, load shedding will be performed
based on a priority scheme. The criteria to disconnect
load three is described as follow:
(35)

173
where is the maximum current generated by the battery bank is discharged. The supervisory controller
Grid, and and are the load currents for shifts load gradually from the battery to the grid to meet
each zone. the power demand [2]. On this mode, the battery
current is always zero. The On-Off control for the load
may be activated if the total current demand is higher
5 Supervisory controller than the total current generated by the system.
The operation strategy for the supervisory controller
The supervisory controller determines the operation of the HPS is shown in Figure 6 as a state transition
mode of each generation subsystem and sets the diagram for each mode of the supervisory controller.
corresponding ser point and sliding surfaces for the Figure 7 shows the transition criteria to perform load
SMC at each source. The operation modes are shedding. Load shedding can occur only if the system
determined by the energy balance between the total is operating in mode 4 or mode 5.
demand (load and battery bank) and the total generation
(wind, solar, and grid).
To design the supervisory controller, we selected the
wind subsystem as the main generator role and use the
solar subsystem and the grid as secondary and tertiary
sources respectively.
The five possible modes of generation are as
follows.
Mode 1: Occur when the wind generation is enough Figure 6 Mode transition criteria for the
to satisfy the total demand. The solar subsystem and supervisor control.
the grid are inactive whilst the battery bank is in
recharge mode at maximum charging current, and the
wind subsystem is set to power regulation. This mode
is running until the maximum available wind power is
exceed by the total power demand [1].
Mode 2: the wind generation is not enough to
satisfy the total demand, and it is set by the supervisor
control to operate at the point of maximum energy
conversion. The solar subsystem is set to follow a Figure 7 Load shedding scheme.
power reference required to complement the wind
subsystem and together satisfy the total power demand.
The battery bank is part of the total power demand,
because is in recharge cycle, and its current is always
the battery reference current [1].
Mode 3: The wind and the solar subsystems are set
to operate at their maximum energy conversion points,
and the battery bank is set to supply power to the load
instead to receipt energy. This mode is maintained as
long as the state of charge of the battery is greater than
a minimum required [1] or the battery current is higher
than a minimum discharging current [2].
Mode 4: The load demand is higher than the power
Figure 8 Control signals of Wind Turbine and
available from renewable sources, and the battery bank
PV array (Battery Bank Discharged).
current is equal or higher than maximum discharging
current. The supervisor control connects to the grid to
obtain the extra energy [2]. In this mode, the battery
current is always equal to the maximum discharging
current. On this mode, the On-Off control for the load
may be activated if the total current demand is highest
than the total current generated by the renewable energy
and the grid.
Mode 5: The total demand is more than the power
available from wind and solar subsystems, and the

174
Figure 9 Control signal and sliding surfaces of
solar subsystem
Figure 11 DC bus currents and operation
modes (Battery Bank Discharged).

Figure 10 Control signal and sliding surfaces
of wind subsystem
6 Simulation results
The initial condition of the battery bank is totally
discharged (SOC = -0.75) which is the limit to allow
connection of the HPS to the grid. The battery reference
current for the renewable source to charge the battery
bank is set equal to the maximum charging current,
which in this case is 20A.
Figure 8 shows the wind turbine and the
photovoltaic panel control signals. The individual
control signals and sliding surfaces for the solar
subsystem and the wind subsystem are shown in Figure
9 and Figure 10 respectively.
Figure 11 shows the HPS DC bus current. The first
graph presents the wind current (Set-Point and
generated current). Between 2 to 6, 16 to 22, 24 to 31,
and 64 to 81 seconds, we can observe that the reference
signal is equal to generated signal. As shown before,
that means that system is operating at mode 1 as shown
in graph six. The second graph illustrates the solar
current with the same behavior as the previous
example. The third graph shows the battery bank
current. Because the initial condition was totally
discharged, when there are enough generation from the
renewable sources, the battery reference current is set to
maximum charge current (20A), and the system works
under the first three modes. If the battery current
exceeds the maximum discharging current (-20A), the
system works on mode 4. If the SOC of the battery is
under Qmin, means that the system works under mode 5
as shown in graph six. The fourth graph shows the grid

175
 current. This signal is activated only if the maximum
discharge current of the battery bank was exceeded or
the SOC of the battery bank is charge to Qmin. The
maximum current that the grid could produce is 25A, so
when the diesel reference is higher than this value, the
load shedding control is activated to turn off one or two
zones of the total load as shown in graph five. The
graph six presents the operation modes of supervisor
control. This scenario activates mode five because this
mode depends on the lower limit of SOC of the battery.
The simulation results show satisfactory
performance of the HPS hierarchical control system.
7 Conclusion
This paper presented a Hierarchical controller for a
DC bus configuration Hybrid Power System consisting
of a Wind Turbine, Photovoltaic Panels, a Battery
Bank, capability for connection to grid, and load. The
proposed hierarchical control system consisted of lower
level controllers for each generation and storage
component designed using Sliding Mode Control and a
supervisory controller that had the objective of
maximizing use of the renewable sources while
minimizing connection to the grid. The supervisory
controller had five modes of operation where in each
mode the sliding surfaces and the set points of the lower
level controllers were changed according to the
operation conditions and performed load shedding
based on load priorities when available generation is
insufficient. The functionality and performance of the
system was studied by means of simulations.
The simulation results presented, demonstrated that
the designed control is capable to determining the
operating mode of each subsystem on the HPS. The
criteria, to make the decisions, was based on
measurable system variables such shaft speed, currents,
SOC, and power.
In the supervisory control scheme design, we
assumed that the main generation role would be carried
out by the renewable energy, i.e., wind and solar
subsystem respectively, while the traditional generation
play a complementary role. The role of the solar
subsystem and the wind turbine can be easily changed
due to the modular design of the system.
8 Acknowledgements
This work was primarily supported by the ERC
Program of the National Science Foundation under
grant EEC-9731677.
176
View publication stats
9 References
[1] Valenciaga, F and Puleston, P F. "Supervisor control for
stand-alone hybrid power system using wind and
photovoltaic energy". IEEE Trans. on Energy
Conversion. Vol. 20. No2, June 2005, pp. 398-405.
[2] Omari, Osama, Ortjohann, Egon and Morton, Danny.
"An Online Control Strategy for DC Coupled Hybrid
Power Systems". IEEE Power Engineering Society
General Meeting, June 2007. pp. 1-8.
[3] Valenciaga, F, Puleston, P.F and Battaiotto, P.E.
"Passivity/sliding mode control of a stand-alone hybrid
generation system". IEE Proc. Control Theory Appl. Vol.
147 No 6, November 2000, pp. 680-686.
[4] Ortiz-Rivera, E and Peng, F.Z. "Analytical Model for a
Photovoltaic Module using the Electrical Characteristics
provided by the Manufacturer Data Sheet". IEEE 36th
Power Electronics Specialists Conference, 11-14 Sept
2005. pp. 2087-2091.
[5] Valenciaga, F, Puleston, P.F and Battaiotto, P.E. "Power
Control of a Photovoltaic array in a hybrid electric
generation system using sliding mode techniques". IEE
Proc. Control Theory Appl. Vol. 148 No 6, November
2001, pp. 448-455.
[6] Chen, Min. and Rincón-Mora, Gabriel A. "Accurate
Electrical Battery Model CApable of Predicting Runtime
and I-V Performance". IEEE Transaction on Energy
Conversion. Vol. 21 No 2., June 2006, pp. 504-511.
[7] Valenciaga, F., Puleston, P.F and Battaiotto, P.E.
"Variable Structure System Control Design Method
based on a Differential Geometric Approach: application
to a Wind Energy Conversion Subsystem". IEE
Procedings on Control Theory Applications. Vol. 15.
Nº1. January 2004, pp. 6-12.
[8] Knauff, M.C, et al. "Simulink Model for Hybrid Power
System Test-bed". Proc. of the IEEE Electric Ship
Technologies Symposium (ESTS). Arlington VA. : s.n,
May 21-23, 2007., pp. 421-427.