U.S. patent application number 15/152039 was filed with the patent office on 2016-09-01 for power point tracking.
This patent application is currently assigned to SunPower Corporation. The applicant listed for this patent is SunPower Corporation. Invention is credited to Gary B. Baker, Robert Batten, Vincenzo DiTommaso, Edward Gee, Adam Heiberg, Triet Tu Le, Henry F. Pruett.
Application Number | 20160254673 15/152039 |
Document ID | / |
Family ID | 44188497 |
Filed Date | 2016-09-01 |
United States Patent
Application |
20160254673 |
Kind Code |
A1 |
Batten; Robert ; et
al. |
September 1, 2016 |
POWER POINT TRACKING
Abstract
A local power converter may include a controller to manipulate
the operating point of a local power converter to cause the power
point tracking feature of a central power converter to operate at a
point determined by the controller. In some embodiments, the
controller can manipulate the operating point of the local power
converter by alternating between at least two modes. In some other
embodiments, the controller can manipulate the operating point of
the local power converter to provide a substantially constant
slope. In some other embodiments, the controller can maintain a
substantially constant impedance ratio. In some other embodiments,
the controller enables perturbations from the power point tracking
feature of the central power converter to reach the power
source.
Inventors: |
Batten; Robert; (Tualatin,
OR) ; DiTommaso; Vincenzo; (Beaverton, OR) ;
Baker; Gary B.; (Sherwood, OR) ; Pruett; Henry
F.; (North Bend, OR) ; Heiberg; Adam;
(Tualatin, OR) ; Le; Triet Tu; (Portland, OR)
; Gee; Edward; (Camas, WA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
SunPower Corporation |
San Jose |
CA |
US |
|
|
Assignee: |
SunPower Corporation
San Jose
CA
|
Family ID: |
44188497 |
Appl. No.: |
15/152039 |
Filed: |
May 11, 2016 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
12972398 |
Dec 17, 2010 |
9342088 |
|
|
15152039 |
|
|
|
|
61291613 |
Dec 31, 2009 |
|
|
|
61316750 |
Mar 23, 2010 |
|
|
|
61291613 |
Dec 31, 2009 |
|
|
|
Current U.S.
Class: |
307/112 |
Current CPC
Class: |
H02J 3/387 20130101;
H02J 3/385 20130101; H02J 3/386 20130101; Y02E 10/76 20130101; Y02E
10/56 20130101; G05F 1/67 20130101 |
International
Class: |
H02J 3/38 20060101
H02J003/38 |
Claims
1. A method of operating a local power converter comprising:
operating a controller of a local power converter to periodically
alternate operation of the local power converter between two or
more operating modes, the local power converter receiving voltage
from a voltage generating source and the local power converter
outputting to a central power converter having a power point
tracking feature; wherein one of the operating modes is a constant
power mode and one of the operating modes is a constant voltage
ratio mode, the constant power mode outputs power to the central
power converter within a predetermined range and conducts Maximum
Power Point Tracking for the local power converter to identify a
Maximum Power Point, the constant voltage ratio mode including
maintaining a constant ratio between input and output voltage of
the local power converter, the ratio causing an MPPT algorithm in
the central power converter to lock onto an operating point
selected by the local power converter.
2. The method of operating a local power converter of claim 1
wherein the voltage generating source comprises one or more of a
solar panel, a fuel cell, a wind turbine or a battery.
3. The method of operating a local power converter of claim 1
wherein a Maximum Power Point is identified before maintaining a
constant ratio between input and output voltage for the constant
voltage ratio mode and wherein the controller moves marginally off
the Maximum Power Point and uses this marginal move to maintain the
constant ratio between input and output voltage for the constant
voltage ratio mode.
4. The method of operating a local power converter of claim 3
wherein marginal movement off of the Maximum Power Point provides
an amount of slope that causes an MPPT function in the central
power converter to lock onto an operating point selected by the
local power converter rather than an operating point the central
power converter would have selected.
5. The method of operating a local power converter of claim 1
wherein the local power converter continuously cycles between the
constant power mode and the constant voltage ratio mode and wherein
the relative amount of time spent in each mode may be adjusted.
6. The method of operating a local power converter of claim 1
wherein an operating point may be preselected and resident in the
controller as well as be variable and configurable by an outside
source.
7. A method of operating a local power converter comprising: at a
local power converter, receiving voltage from a voltage generating
source, the local power converter outputting to a central power
converter having power point tracking functionality, determining a
ratio of local converter impedance to photovoltaic panel impedance;
using the determined ratio in a controller of the local converter
as a control variable, and sending the determined ratio to the
central power converter to maintain a constant impedance ratio and
sustain an operating point at the central power converter at a
constant distance from a maximum power point.
8. The method of operating a local power converter of claim 7
wherein an impedance ratio of one indicates that the local power
converter is operating at its maximum power point.
9. The method of operating a local power converter of claim 7
wherein operating the local converter with an impedance ratio of
less than one causes the central power converter to slope down in
operating voltage.
10. The method of operating a local power converter of claim 7
wherein operating the local power converter with an impedance ratio
of greater than one causes the central power converter to slope up
in operating voltage.
11. The method of operating a local power converter of claim 7
further comprising: repeatedly perturbing the voltage generating
source using a frequency outside of the frequency band that can be
monitored by the central power converter; and monitoring the
results of the perturbations by the local power converter
independent of conventional perturbations generated by the central
power converter.
12. A method of operating a local power converter comprising:
initializing a local power converter at a desired operational
voltage ratio, the local power converter receiving voltage from a
voltage generating source and the local power converter outputting
to a central power converter having a power point tracking feature;
observing at the local power converter perturbations caused by the
central power converter and determining an AC portion of the
pertubations, and associated direction and slope amount, for the
central converter.
13. The method of operating a local power converter of claim 12
further comprising: operating a control loop in the local power
converter to move a DC level up or down to maintain a voltage ratio
and wherein the central power converter raises or lowers its
operating point using this voltage level maintained by the local
central converter.
14. The method of operating a local power converter of claim 12
wherein when the voltage ratio is above the desired ratio value, a
DC level is adjusted such that the central converter pertubations
are centered at a power point above a Maximum Power Point voltage
level.
15. The method of operating a local power converter of claim 14
wherein if the voltage ratio is observed to be within bounds of a
desired level then the DC level of the local power converter is
maintained such that inverter pertubations are centered at a power
point below a Maximum Power Point voltage level.
16. The method of operating a local power converter of claim 15
wherein the Maximum Power Point (MPP) is associated with the
central power converter and wherein the MPP always remains within a
ripple range of pertubations of the central power converter.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application is a continuation of U.S. patent
application Ser. No. 12/972,398 titled Power Point Tracking, which
was filed on Dec. 17, 2010, and which claims priority from U.S.
Provisional Patent Application Ser. No. 61/291,613 titled Power
Point Tracking filed Dec. 31, 2009 and U.S. Provisional Patent
Application Ser. No. 61/316,750 titled Inverter Interaction filed
Mar. 23, 2010. The '398 application is incorporated herein by
reference in its entirety.
BACKGROUND
[0002] FIG. 1 illustrates a prior art energy conversion system in
which an array of photovoltaic (PV) panels 10 are connected in
series string to generate a DC bus voltage V.sub.A+, V.sub.A- from
the array. The DC bus voltage is converted to AC power in the form
of an AC voltage V.sub.AC by an inverter 12.
[0003] Referring to FIG. 2, curve 16 illustrates the
voltage-current characteristic (V-I curve) of a typical PV panel
under certain operating conditions, while curve 18 illustrates the
corresponding power characteristic (power curve) for the same panel
under the same conditions. The V-I curve is zero volts with a value
of I.sub.SC which is the short-circuit current generated by the
panel when the output terminals are shorted together. As the output
voltage increases, the V-I curve remains at a fairly constant level
of current until it reaches a knee at which it descends rapidly
toward zero current at V.sub.OC, which is the open-circuit output
voltage of the panel.
[0004] The power curve is simply the current times the voltage at
each point along the V-I curve. The power curve has a maximum value
corresponding to a certain voltage level and a certain current
level. This is known as the maximum power point or MPP. The maximum
power point tends to change based on changes in operating
conditions such as illumination level, temperature, age of the
panel, etc.
[0005] Inverters for PV systems typically include maximum power
point tracking (MPPT) functionality that attempts to operate the
system at the maximum power point. Since the maximum power point
changes based on changes in operating conditions, algorithms have
been devised for tracking the MPP as it changes over time.
Conventional MPPT algorithms often employ a perturb and observe
technique in which the operating voltage or current is varied,
thereby causing the operating point to vary around the MPP as shown
in FIG. 2. By observing the change in power as the voltage or
current changes, the MPP can be determined.
[0006] In systems that only include a single PV panel, existing
MPPT algorithms enable the system to operate at the true maximum
power point for the single panel. In systems having multiple PV
panels as shown in FIG. 1, existing MPPT algorithms 14 attempt to
find the average maximum power point for the overall system.
However, finding the average maximum power point for the system
does not necessarily enable each individual panel to operate at its
maximum power point. By not operating each panel at its maximum
power point, the energy produced by the system may be reduced
resulting in lost income and a longer return on investment. This
loss of efficiency may be especially problematic with PV arrays
that are subjected to shading in part of the array because the
reduction in efficiency is often disproportionately larger than the
area that is shaded.
BRIEF DESCRIPTION OF THE DRAWINGS
[0007] FIG. 1 illustrates a prior art photovoltaic energy
conversion system.
[0008] FIG. 2 illustrates the voltage-current curve and power curve
of a typical PV panel.
[0009] FIG. 3 illustrates an embodiment of an energy conversion
system according to some inventive principles of this patent
disclosure.
[0010] FIG. 4 illustrates the output power characteristic of a
local power converter having an MPPT feature.
[0011] FIG. 5 illustrates an example embodiment of a local power
converter 22 according to some inventive principles of this patent
application.
[0012] FIGS. 6 and 7 illustrate an example embodiment of a
technique for switching between fixed duty cycle and MPPT modes
according to some inventive principles of this patent
disclosure.
[0013] FIGS. 8 and 9 graphically illustrate the interaction between
duty cycle and current for increasing and decreasing duty cycles,
respectively, for an embodiment of a technique for switching
between fixed duty cycle and MPPT modes according to some inventive
principles of this patent disclosure.
[0014] FIG. 10 illustrates an example embodiment of a local/central
converter interaction technique utilizing alternating modes
according to some inventive principles of this patent
disclosure.
[0015] FIG. 11 illustrates an aspect of the operation of the
embodiment of FIG. 10.
[0016] FIG. 12 illustrates an example embodiment of a local/central
converter interaction technique utilizing constant slope according
to some inventive principles of this patent disclosure.
[0017] FIG. 13 illustrates the manner in which an MPPT algorithm in
a central power converter may generate perturbations about a
maximum power point.
[0018] FIG. 14 illustrates an example embodiment of a local/central
converter interaction technique utilizing a central converter
perturb method according to some inventive principles of this
patent disclosure.
[0019] FIG. 15 illustrates an aspect of the operation of the
embodiment of FIG. 14.
[0020] FIGS. 16-18 illustrate efficiency curves for a local power
converter, a central power converter, and a composite curve,
respectively, according to some inventive principles of this patent
disclosure.
DETAILED DESCRIPTION
[0021] Some of the inventive principles of this patent disclosure
relate to the use of local power converters to optimize the
operating points of individual power sources that are connected to
a centralized power converter. FIG. 3 illustrates an embodiment of
an energy conversion system having local power converters according
to some inventive principles of this patent disclosure. In the
system of FIG. 3, each power source 20 has a corresponding local
power converter 22 that includes a power conversion stage to
transfer power from the power source to a central power converter
24 that includes a power point tracking (PPT) feature 26. The local
power converters 22 are connected in a series string to generate a
power bus V.sub.B+, V.sub.B-, but the inventive principles are not
limited to any specific topology, power sources, power converters,
etc.
[0022] For example, if the power sources are photovoltaic (PV)
panels, the local power converters 22 may be implemented as DC/DC
converters, and the central power converter 24 may be implemented
as a DC/AC inverter having conventional MPPT functionality that
attempts to operate the system at the maximum power point. However,
to achieve the most energy from a PV system, especially under
shading and inter-panel mismatch conditions, each panel may be
individually optimized for a maximum power point. Thus, the local
power converters 22 may be implemented as power optimizers (PO)
that include local MPPT functionality to individually optimize
their corresponding panels.
[0023] FIG. 5 illustrates an example embodiment of a local power
converter 22 according to some inventive principles of this patent
application. A power conversion stage (PCS) 30 may be implemented
with any suitable power conversion topology. For example, in a PV
system, the PCS may be implemented as a buck converter, a boost
converter, a buck-boost converter, a push-pull converter followed
by a rectifier, or any other DC/DC converter topology. A controller
32 may be implemented with analog and/or digital hardware,
software, firmware or any suitable combination thereof. Local power
point tracking functionality 34 may likewise be implemented with
analog and/or digital hardware, software, firmware or any suitable
combination thereof. In a system with a controller based on a
digital signal processor (DSP), the PPT functionality may be
implemented as a pure software or firmware solution. The system of
FIG. 5 may be used, for example, to implement any of the methods
according to the inventive principles of this patent disclosure as
described herein.
[0024] One potential problem with using a conventional inverter
having MPPT functionality is that the addition of MPP tracking
hardware and/or software at the panel level means that the system
may have two or more independent MPP tracking systems. Since the
panel level MPP tracking functionality may attempt to output
constant power as shown in FIG. 4, regardless of the load
impedance, the MPP tracking of the inverter may not see the panel
behavior that it expects to see. In general, this may result in the
inverter MPP system not being able to find an optimal operating
point and constantly varying the system string voltage in a
sometimes random manner as the operating point moves as shown by
the arrows in FIG. 4.
[0025] Therefore, each local power converter 22 may include
functionality to manipulate or optimize the operating point of the
power source and/or the corresponding local power converter in a
manner that may influence the PPT functionality 26 of the central
power converter 24. For example, the PPT functionality in a local
power converter 22 may cause the power point tracking feature of
the central power converter to operate at a point determined by the
local power converter. In some embodiments, control over the power
point tracking feature of the central power converter may be
achieved by manipulating the operation of the local power converter
to cause the central power converter to see a slope in the power
characteristic of the local power converter. Thus, the PPT
functionality of the central power converter may be manipulated by
giving it a direction in which to move in search of a maximum power
point that it otherwise would not see.
[0026] Therefore, to provide a complete system solution, the
maximum power point tracking hardware that is installed at each
panel may interact with the central inverter. A potential benefit
of the inventive principles is that hardware and/or software at the
local level may work with most conventional central inverters and
may not require special wiring, setup, configuration, software
changes, etc., at either the system level or the inverter level. In
other words, the inventive principles may enable dynamic control of
the central power converter (e.g., inverter) operating point
autonomously without any communication between the local power
converters (e.g., power optimizers) and the central inverter to
operate in an optimized region without having to change hardware
and/or software in existing inverters. Moreover, the inventive
principles may enable the implementation of a system that is
suitable for use with a wide range of inverters from different
manufacturers.
[0027] In a system according to some inventive principles of this
patent disclosure, there may be specific output voltage and/or
voltage ratio (duty cycle) ranges where efficiency is highest. It
may be advantageous for the inverter to cause the system to operate
in this region. When the system is in this region it may be
advantageous for the inverter to believe that it is operating at
the peak of the power curve for the system. As described below,
there are numerous ways to accomplish this according to the
inventive principles of this patent disclosure.
Fixed Duty Cycle Method
[0028] Some inventive principles of this patent disclosure relate
to operating a local power converter in a fixed duty cycle mode
that enables a central power converter to see the power curve of
the local power source through the local power converter, instead
of the flat power curve that would normally be presented by a local
power converter implementing an MPPT algorithm. Operating at a
fixed duty cycle essentially produces a fixed voltage ratio V.sub.R
such that V.sub.R=V.sub.O/V.sub.I where V.sub.I and V.sub.O are the
input and output voltages, respectively.
[0029] In one example embodiment, the local power converter may be
implemented as a switching power converter such as a buck
converter. In fixed duty cycle mode, the converter is operated at a
fixed duty cycle which results in a fixed voltage ratio between the
input and output voltages. The converter may initially begin
operation in a normal mode using a conventional MPPT algorithm.
Once the local power converter has found the MPP, the controller
enters fixed duty cycle mode by maintaining the converter at a
fixed duty cycle (or voltage ratio) corresponding to the MPP. This
mode may essentially cause the local power converter to be
transparent to the MPPT algorithm in a central power converter,
thereby preventing the MPPT algorithm from wandering.
[0030] As a further refinement, limits may be placed on the amount
of time and/or duty cycle (voltage ratio) in fixed duty cycle mode.
For example, allowing the system to stay in a transparent, fixed
duty cycle mode may negate the optimizing effect of local and/or
system optimizers. To avoid this, the value of the current through
the local power converter at the maximum power point just before
entering fixed duty cycle mode may be saved. If the system current
deviates too far from the saved current, for example, as a percent
of the saved value, the local power converter (e.g., power
optimizer) may return to normal MPPT mode since it apparently is
not at an ideal operation point. Various thresholds may be used to
determine when the system current or other parameter has deviated
too far from the save value. For example, modes may be switched
when the difference between the stored value and the presently
measured instantaneous value of the parameter exceeds a
predetermined limit. As another example, the threshold may be based
on the difference between the stored value and the measured value
exceeding a specific value for a predetermined length of time.
[0031] FIGS. 6 and 7 illustrate an example embodiment of a
technique for switching between fixed duty cycle and MPPT modes
according to some inventive principles of this patent disclosure.
The fixed duty cycle check routine illustrated in FIG. 6 may be run
at regular intervals during operation of the local power converter.
The fixed duty cycle check routine is entered at 600 and proceeds
to determine if the system is in fixed duty cycle mode by checking
a fixed duty cycle flag at 602. If the controller is already in
fixed duty cycle mode, the fixed duty cycle routine at 604 (which
is illustrated in FIG. 7) is run, then the check routine terminates
at 606. If the controller is not already in fixed duty cycle mode,
a delay loop is checked at 608. If the delay is not over, the check
routine terminates at 606. Once the delay is finished, the present
duty cycle is compared to a threshold (trip point) at 610. If the
trip point has not been reached, the check routine terminates at
606. Once the threshold is reached, the value of the input current
is saved at 612, the fixed duty cycle flag is set at 614, and the
duty cycle of the converter is frozen at the present value at
616.
[0032] Once the system is in fixed duty cycle mode, the open
routine shown in FIG. 7 monitors the current in the system at 702
and if it becomes larger than the LargeTriggerPoint (LTP) threshold
sets the system back into normal (MPPT) mode at 704-708. If the
system spends more time outside of the SmallTriggerPoint (STP) than
inside at 712-718 as determined by the length of time set by TTP,
the system will also return to normal mode. This second STP window
is designed to allow the system to become even more optimized once
all the system transients have settled out by providing a method of
removing units that are operating at a non-optimal point. This
relaxes the requirements on the LTP window, making it easier for
the central inverter to find the optimal operating point in the
presence of noise and other changing variables.
[0033] FIGS. 8 and 9 graphically illustrate the interaction between
duty cycle and current for increasing and decreasing duty cycles,
respectively. As the duty cycle crosses the trip point, it becomes
fixed over a range of operating currents defined by the value of
LTP. The STP window is not shown on these figures but would be
similar, albeit with the locations of the STP edges positioned
closer to the value of Saved_input_current.
[0034] Thus, some of the inventive principles may be implemented to
cause the central inverter voltage to operate in a desired region
without needing to make any modifications to the existing inverter
hardware or software.
Alternate Modes Method
[0035] Some additional techniques for optimizing the operation of a
local power converter while interacting with a central power
converter according to some inventive principles of this patent
disclosure relate to methods in which a controller for the local
power converter periodically alternates between two or more
modes.
[0036] For example, the local power converter, or power optimizer
(PO), may alternate (or "ping-pong") between a first mode in which
it operates at constant power (CPM or CP mode), and a second mode
in which it operates at a constant ratio (CR mode). In constant
power mode, constant power is maintained regardless of the output
operating point (within limits), and the controller implements an
MPPT algorithm to determine a maximum power point. In constant
voltage ratio mode, the controller steps (leans) slightly off the
maximum power point and maintains a constant ratio between the
input and output voltage of the local power converter, thereby
providing a certain amount of slope that causes the MPPT algorithm
in the central inverter to lock onto the operating point selected
by the local power converter rather than the operating point it
would have selected on its own. Although stepping off the precise
maximum power point may sacrifice a small amount of efficiency at
the local power converter, this amount may typically be very small
compared to the gain in overall system efficiency.
[0037] The controller in the local power converter continuously
cycles between these two modes, updating the MPP in CP mode, then
presenting a synthesized power slope to the central power converter
in CR mode until the voltage ratio is within an ideal range. The
relative amount of time spent in each mode may be adjusted, either
analytically or empirically, to work with a suitable range of
available central power converters. For example, in some
embodiments, the controller may operate in CP mode for about 1
second to determine the MPP, then operate in CR mode for about 15
seconds to present a slope to the central power converter.
[0038] In some embodiments, an ideal operating point may be
implemented as a fixed value that is programmed into the local
power converter. In other embodiments, the operating point may be
variable and configurable, for example, by an outside source.
[0039] FIG. 10 illustrates an example embodiment of a local/central
converter interaction technique utilizing alternating modes
according to some inventive principles of this patent disclosure.
The example of FIG. 10 is described in the context of a PV system,
but the inventive principles may be applied to other types of
systems including those based on fuel cells, wind turbines,
batteries, etc.
[0040] The algorithm is initialized at 1000. Constant power mode
begins at 1002 where the controller finds the maximum power point
for the PV panel. The controller remains in CP mode for Y seconds
at 1004, and at 1006, the measured (e.g., filtered) voltage ratio
is determined. Calculation of the operating point begins at 1008
where a determination is made was to whether the voltage ratio is
below, above or at (within bounds) the ideal operating point
voltage ratio.
[0041] If the voltage ratio Vr is below the ideal value, then at
1010 the target voltage ratio (TVR) is set to the measured voltage
ratio plus N percent. If the voltage ratio Vr is below the ideal
value, then at 1014 the TVR is set to the measured voltage ratio
minus N percent. If the voltage ratio Vr is at the ideal value
(within an acceptable band), then the algorithm returns to the
beginning.
[0042] Bounds on the value of N used in 1010 and 1014 above may be
determined analytically or empirically by testing with a range of
commercially available inverters or other central power converters
having built-in power point tracking. Moreover, the value of N may
be fixed, or a function of any suitable parameter such as Vr.
[0043] Once the target value of the TVR is determined, the
controller enters constant voltage ratio mode (CR mode) at 1016
where the controller attempts to set the overall system to the
desired TVR by stepping slightly off the maximum power point by a
certain percentage. The controller remains in CR mode for Z seconds
at 1018, then returns to CP mode at 1020.
[0044] The operation of the method of FIG. 10 is illustrated
graphically in FIG. 11 where the amount that the algorithm steps
off the MPP is shown at points A and B corresponding to the value
of N used in the corresponding steps in FIG. 10. The operating
point of the system constantly alternates between the MPP and point
A or point B depending on the direction in which the slope
leans.
Constant Impedance Ratio Method
[0045] Some additional techniques for optimizing the operation of a
local power converter while interacting with a central power
converter according to some inventive principles of this patent
disclosure relate to methods in which the local power converter
uses a constant impedance ratio, or constant slope, to maintain an
operating point at an approximately constant distance from the
maximum power point.
[0046] This may be accomplished, for example, by using the
incremental impedance to panel impedance ratio as a control
variable for constant slope. An impedance to impedance ratio of one
(unity) means that power converter is at its maximum power point.
An impedance to impedance ratio of less than one means the PV
operating voltage is higher than the maximum power point voltage
and vice versa. An impedance to impedance ratio of less than one
may cause the central inverter to constantly see higher power at
voltages lower than its current operating point and may cause the
central inverter to constantly slope down in voltage. An impedance
to impedance ratio of greater than one may cause the central
inverter to constantly see higher power at voltages higher than its
current operating point and may cause the central inverter to
constantly slope up in voltage.
[0047] The incremental impedance may be expressed as an incremental
change in voltage divided by the corresponding incremental change
in current. I.e., the incremental impedance may be expressed as
.DELTA.V/.DELTA.I where .DELTA.V is an incremental change in
voltage, and .DELTA.I is the corresponding incremental change in
current.
[0048] The impedance ratio may be determined, for example, by
enabling the local power converter to continually perturb the power
source at a frequency that is outside of the band that can be
monitored by the central power converter. Filters and correlations
may then be used to monitor the results of the perturbations by the
local power converter independently of the conventional
perturbations generated by the central power converter and other
system changes. Simultaneously, the perturbations generated by the
central power converter are allowed to reach the power source so
that the central power converter can determine a direction to move.
Because both perturbation loops can operate at the same time, this
method may eliminate the need to switch back and forth between
modes. The faster loop is used to determine the operating point of
the power optimizer. The operating point can then be adjusted to
present the desired slope to the central power converter.
[0049] FIG. 12 illustrates an example embodiment of a local/central
converter interaction technique utilizing constant impedance
according to some inventive principles of this patent disclosure.
The example of FIG. 12 is described in the context of a PV system,
but the inventive principles may be applied to other types of
systems including those based on fuel cells, wind turbines,
batteries, etc.
[0050] The algorithm is initialized at 1200. Constant power (CP)
mode begins at 1202, and the controller measures the voltages for
the local power converter at 1204. Then constant impedance (CI)
mode is entered at 1206.
[0051] Simultaneously, the faster loop begins running at 1210 where
a perturb signal is used to determine the PV panel slope. At 1211,
the impedance is calculated, and at 1212, the slope amount and
direction are calculated based on the perturbations and
corresponding observed voltages, and the desired system operating
point. The slope amount and direction output from 1212 are used at
1208 where the system is maintained at the desired impedance by
controlling the local power converter duty cycle. The loop around
1208 is substantially slower than the faster loop including
1210-1212, so the perturbations generated by the central power
converter make their way back to the PV panel and enable the MPPT
functionality in the central power converter to see the shape of
the PV panel power curve.
[0052] An advantage of a constant impedance ratio method according
to the inventive principles of this patent disclosure is that it
may eliminate or reduce cycling between constant power mode and
constant impedance mode because the calculation of the slope amount
and direction can be performed in a faster background loop
simultaneously with the slower loop that maintains the system at
the desired slope.
Central Converter Perturb Method
[0053] Some additional techniques for optimizing the operation of a
local power converter while interacting with a central power
converter according to some inventive principles of this patent
disclosure relate to methods in which the perturbations generated
by the central power converter are observed and used to determine a
slope direction and amount.
[0054] For example, FIG. 13 illustrates the manner in which an MPPT
algorithm in a central power converter may generate perturbations
about a maximum power point V.sub.MPP to enable it to maintain its
operation about the maximum power point. The controller in the
local power converter may allow these perturbations to reach the
power source so the controller can observe the AC portion of the
perturbations to determine the direction and slope amount. The
control loop in the local power converter may then move the DC
level up or down as shown by the arrow in FIG. 13 to maintain the
voltage ratio such that the inverter perturbations see the desired
power curve shape for the power source. Thus, the control loop may
control the direction that the PPT feature in the central power
converter thinks it should raise or lower its operating point.
[0055] FIG. 14 illustrates an example embodiment of a local/central
converter interaction technique utilizing a central converter
perturb method according to some inventive principles of this
patent disclosure. The example of FIG. 14 is described in the
context of a PV system, but the inventive principles may be applied
to other types of systems including those based on fuel cells, wind
turbines, batteries, etc.
[0056] The algorithm is initialized at 1400, and at 1402, the local
power converter is started at the desired voltage ratio. At 1404,
the perturbations caused by the central inverter are observed. At
1406, the controller determines whether the voltage ratio is at,
above or below the desired level. If the voltage ratio is above the
desired ratio, then at 1408 the DC level is adjusted so the
inverter perturbations are centered at a power point above the MPP
voltage level such that the lowest voltage excursion reaches the PV
panel maximum power point. If the voltage ratio as at the desired
level (within bounds), then at 1410 the DC level is maintained so
the inverter perturbations are centered at the MPP voltage level.
If the voltage ratio is below the desired ratio, then at 1412 the
DC level is adjusted so the inverter perturbations are centered at
a power point below the MPP voltage level such that the highest
voltage excursion reaches the PV panel maximum power point. The
control loop then returns to 1404 and continues to maintain the
local power converter operating point at a DC level that causes the
central inverter to settle in to the operating point selected by
the local power converter.
[0057] The operation of the method of FIG. 14 is illustrated
graphically in FIG. 15 where the variation in the operating point
is shown by arrow A when the voltage ratio as at the desired level,
arrow B when the voltage ratio is below the desired amount, and
arrow C when voltage ratio is above the desired amount.
[0058] An advantage of the embodiment illustrated in FIG. 15 is
that the maximum power point MPP may always remain within the
ripple range of the perturbations of the central inverter.
Coordinated Operation of Power Converters
[0059] Some additional inventive principles of this patent
disclosure relate to systems in which there is communication
between one or more local power converters, and optionally a
central power converter. In such a system, one or more additional
levels of optimization can take place according to the inventive
principles.
[0060] For example, in a central inverter system with multiple PV
panels and multiple local power converters (power optimizers),
optimization may be performed at a system level. Either on a
central or on a distributed basis, the local units can monitor the
maximum and minimum duty cycles of each local panel power optimizer
and make system adjustments to the duty cycle trip points to
maximize efficiency. By knowing the efficiency curves and power
generation of each device in the system, including the central
inverter, the best operating point can be chosen.
[0061] An example embodiment is illustrated with respect to FIGS.
16-18. FIG. 16 illustrates the efficiency curve .eta..sub.local for
a local power optimizer which has a peak efficiency at Vout/Vin=1.0
volts. FIG. 17 illustrates the efficiency curve .eta..sub.central
for the central inverter which has a peak efficiency at
Vout/Vin=0.9 for a combined string. FIG. 18 illustrates a composite
efficiency curve obtained by multiplying .eta..sub.local by
.eta..sub.central. Thus, the best overall efficiency may be
obtained at Vout/Vin=0.95 even though this operating point is not
necessarily the peak efficiency point for any specific
component.
[0062] Another more specific example is based on the following
assumptions: a system has 200 power optimizers installed; each
power optimizer in the system handles 250 W of power; at a duty
cycle of 50%, a power optimizer is 92% efficient (produces 230 W);
at a duty cycle of 80%, a power optimizer is 98% efficient
(produces 245 W); in this system 10 units are running at 80%
(within the LTP window); and in this system 190 units are running
at 50%.
[0063] The total power out of the system is 46150 W (10*245
W+190*230 W=2450 W+43700 W).
[0064] If the 10 units at 80% duty cycle were removed from the
system (turned off for example, although that is not necessary it
just makes the calculation easier) and the rest of the system was
allowed to operate at 80% duty cycle the output power would then be
46550 W (190*245 W). The second system is producing more power and
would be the more efficient configuration. This can also be applied
to the inverter as well. If the efficiency over input voltage and
current for the inverter is known, the system can optimize both its
own operating point and the voltage at the input to the inverter to
achieve the best overall system operating efficiency.
[0065] In some embodiments, the optimization of the system, done by
an overall supervisory system or done ad-hoc, at both the local
level and the system level may be an important feature. This system
optimization information can be used at a high level to adjust
system voltages and currents. It can also be used in conjunction
with other local power optimizer functions such as phase shedding.
With the phase shedding example, if the supervisory system has
information about the efficiency curves for each local power
optimizer, including the efficiency effects of phase shedding,
system voltages and currents can be optimized to make a local power
optimizer run with reduced phases, resulting in increased
efficiency.
[0066] The inventive principles of this patent disclosure have been
described above with reference to some specific example
embodiments, but these embodiments can be modified in arrangement
and detail without departing from the inventive concepts. Such
changes and modifications are considered to fall within the scope
of the following claims.
* * * * *