WO2015065291A1

WO2015065291A1 - A micro inverter for a solar panel

Info

Publication number: WO2015065291A1
Application number: PCT/SG2014/000509
Authority: WO
Inventors: Mostmans Marcus Josephus CATHARINA
Original assignee: Maxsolar Pte Ltd
Priority date: 2013-10-30
Filing date: 2014-10-30
Publication date: 2015-05-07
Also published as: SG2013081187A

Abstract

A micro inverter (106) for a respective solar panel (102) comprising storage or memory, configured to store performance characteristics for the respective panel (102); a buck and/or boost converter and inverter configured to connect to a DC output of the respective panel (102) to invert the DC output into an AC output to connect to an AC bus (110) and/or an AC grid (114); and a controller or processor configured to optimise the effective DC load of the converter and inverter seen by the respective panel (102) according to the stored performance characteristics and one or more dynamic operational parameters.

Description

A MICRO INVERTER FOR A SOLAR PANEL

FIELD

The present invention relates to a micro inverter for a solar panel, particularly though not solely to a method of maximum power point tracking.

BACKGROUND

Building integrated solar panels are usually mounted on the roof of a building and may be connected to a battery bank or directly to the grid. Each panel produces a DC output at around 30-50V and a current which varies considerably depending on the level of radiation and/or shading. Such solar panels may also be used in a solar farm.

Connecting the DC output of the panels to the AC network within the installation in a cost effective manner, remains a challenge. One method is to connect a number of the panels in series to generate a high DC voltage 300-600 V. This can then be chopped directly into AC by a centralized inverter. A second option is a micro inverter which boosts the DC voltage of individual panels above the AC line top voltage and then converts it to AC.

Central inverters are cheaper in larger installations, but have lower output in some scenarios. If a single panel is shaded it affects the whole string, and the inverter is only able to optimise the load matching for the entire string in some scenarios. Prior art micro invertor's may be more expensive in some scenarios but may improve the overall efficiency, as they are able to load match for each individual panel.

SUMMARY

In general terms the invention proposes a micro inverter for a solar panel which is preprogrammed with the specific performance characteristics of the panel or panels it is connected to. This may have the advantage that the power output is more accurately optimised for each panel and the optimisation is much faster. This may result in a significantly higher output for the overall installation.

In a first specific expression of the invention there is provided a micro inverter for a respective solar panel comprising storage or memory, configured to store performance characteristics for the respective panel;

a buck and/or boost converter and inverter configured to connect to a DC output of the respective panel to invert the DC output into an AC output to connect to an AC bus and/or an AC grid;

a controller or processor configured to optimise the effective DC load of the converter and inverter seen by the respective panel according to the stored performance characteristics and one or more dynamic operational parameters.

In a second specific expression of the invention there is provided a central controller for a network of solar panel micro inverters comprising

central storage or memory, configured to store performance characteristics specifically determined for a plurality of panels according to location, model, date and/or installation characteristics, and a plurality of versions of firmware for a plurality of micro inverters;

a central communications module configured to interrogate respective micro inverters; and

a central controller or processor configured to update a local storage in respective micro inverter depending on whether the performance characteristics or firmware in the central storage or memory is more appropriate than that sorted locally in the respective micro inverter.

Embodiments may be implemented according to any of claims 2 to 6 or 8.

BRIEF DESCRIPTION OF THE DRAWINGS

One or more embodiments will be described with the reference of the below drawings, in which: Figure 1 is a single line diagram of the electrical connections in an example embodiment of a solar panel installation;

Figure 2 is a flow diagram of an example embodiment of a method of maximum power point tracking;

Figure 3 is a graph of the panel power vs. voltage for different light levels;

Figure 4 is a flow diagram of an iteration to find the final maximum power point; and

Figure 5 is a block diagram of an example embodiment of a micro inverter.

DETAILED DESCRIPTION Prior art micro inverters may incorporate maximum power point tracking (MPPT) which adjusts the DC load conditions (V and I) to attempt to maximise the power output of each panel.

However there are a wide variety of different optimisation techniques, none of which are ideal. They all suffer from slow initial solution, and slow reaction time to changed external conditions. Typically for a building integrated solar panel the sun intensity may significantly change several times each minute. Thus if the optimisation takes anything more than several seconds, a significant portion of the potential yield will be lost.

Given that in the current market, manufacturers are typically striving for even small fractions of a percent increase in efficiency, embodiments may include an optimisation technique that is more accurate at the initial solution, and/or which solves more quickly in response to dynamic external conditions.

According to the installation 100 shown in Figure 1, an array of panels 102 are each connected with DC cables 104 to a micro inverter 106. Each Micro inverter is coupled to an AC bus 110, which connects via several isolators 112 to the grid 114.

Each micro inverter 106 is programmed to optimise the power output of the connected solar panel. Rather than iteratively optimising from an arbitrary starting point, Figure 2 shows a method 200 of optimisation which provides a relatively accurate optimised power point based on the ambient conditions. Firstly the algorithm initialises 202, then the model of the solar panel is either entered or retrieved from memory 204. Based on a lookup table, the predetermined maximum power points are retrieved, and matched against ambient conditions to determine an initial maximum power point 206.

For example as shown in Figure 3 the power against voltage is shown for high radiation 302, medium radiation 304 and low radiation 306. The power output from each panel 102 is determined by the load factor presented by the corresponding micro inverter 106. For each different radiation level or light intensity there is a different optimum point or maximum in the power output. So by adjusting the load towards the maximum the power output from each panel 102 can be maximised.

By measuring the OC voltage 308 (and/or the short circuit current), the micro inverter 106 can determine the ambient conditions, for example an estimate of the radiation level or light intensity. Knowing in advance the model of panel 102 it is possible to predetermine the maximum power points 310 for each OC voltage, either empirically or using a formula provided by the manufacturer. Thus, the initial maximum power point can be determined based on the OC voltage and/or short circuit current. The load point of the inverter 106 is then set based on the maximum power point. The maximum power point may also be determined based on the temperature of the panel and/or other environmental variables.

As the individual panel may have slight manufacturing variances, and/or because the lookup table will be for discrete radiation levels, the method 200 may further iterate 208 towards a final maximum. The iteration 208 may occur as shown in Figure 4. The voltage and current are measured 402, and the current power output is determined 404. If the power has increased since the last iteration 406, the voltage will be incremented 410 in the same direction 408 as the last iteration, and vice versa in case of a decrease.

The iteration 208 may be repeated 412 until a certain number of iterations has occurred, or when an end condition occurs eg: the change is power is low when near the maximum. It is always a compromise in finding and tracking the MPPT. As shown in Figure 3 the top of the curve is flat, meaning to be sure of the MPPT you have to change the power at relatively large steps.

Depending on the light intensity the step size may be varied. For low light intensity the delta power change will be around 5% at high power it need only around 0.5%.

The optimisation 206,208 may start again when a change is determined, for example the level of radiation or output power changes significantly, or when a predetermined period of time has elapsed. In a cluster of inverters, once the MPPT is found, only one inverter needs to be used to monitor if the condition's have changed. It is also possible to each inverter to be directed to start at a range of different initial points and the one that finds the maximum point first informs the others. This installation coordination or control function may be done at a multiplexer at each installation 100, which connects the communications ports of the micro inverters to a central sever.

The micro inverter 106 is shown in more detail in Figure 5. The DC input from the panel 102 is buffered on a series of electrolytic capacitors 502. A buck boost converter 504 converts the DC voltage to a higher level, for example 350V. The high DC voltage is smoothened by a DC filter bank 506. The smooth DC voltage is then converted to a single phase AC voltage by inverter 508 and provided to the grid 114. A controller (not shown) may control the converter 504 and inverter 508 according to the algorithm 200 mentioned earlier.

For smoothing out the DC ripple from the converters, two converters are connected in parallel and interleaved. This significantly reduced the step height and the harmonic generation.

Thus the load point is primarily determined by the available DC current (controlled by the PWM value of the converter 504). The panel 102 will deliver that current at a voltage according to its characteristic for the current light level. The buck boost converter is then controlled accordingly to achieve the target higher DC voltage and the selected load point. The high DC voltage must be higher than the peak line voltage of the grid, plus any losses in the other components such as wiring resistance.

In some applications it may be desirable to connect two panels to a dual inverter. The dc switching parts 504 and the capacitors 502 are duplicated for each panel. A single controller will independently execute the optimisation 200 for each panel.

Remote updating

Using information obtained over time, it is possible to enhance the data on each model of panel provided from the manufacturer. Thus it is possible to provide a higher definition look-up table regarding the maximum power points. So, for example rather than haying say 5 solutions on the maximum power points 310, this might be enhanced to 20 points for much smaller increments of voltage. In this way the initial maximum power point 206 can be more accurate, and the iteration 208 shorter.

Thus the look-up table could be remotely updated or any other part of the software as required.

The term micro inverter used herein is intended to mean a buck and/or boost converter attached to and independently inverting the DC output of a single or side by side PV solar panels into an AC output, and configured in hardware and/or software to operate as per the present invention. The combination of the panel(s) and the micro inverter are referred to herein as an AC solar module. If and where such terms are used in the prior art they are not intended to mean the same thing. Whilst exemplary embodiments of the invention have been described in detail, many variations are possible within the scope of the invention as claimed as will be clear to a skilled reader.

Claims

CLAIMS:

1. A micro inverter for a respective solar panel comprising

storage or memory, configured to store performance characteristics for the respective panel;

a buck and/or boost converter and inverter configured to connect to a DC output of the respective panel to invert the DC output into an AC output to connect to an AC bus and/or an AC grid; and

2. The inverter in claim 1 wherein the stored performance characteristics include a lookup table referencing a measured operational parameter to a corresponding optimum DC load.

3. The inverter in claim 1 or 2 wherein the storage or memory is configured to store a plurality of profiles, each profile corresponds to the performance characteristics for a specific model and manufacturer of panel.

4. The inverter in any preceding claim wherein the operational parameters include open circuit voltage and/or short circuit current.

5. The inverter in any preceding claim wherein the optimisation includes determining an initial solution based on the operational parameters and the stored performance characteristics; and performing a minor iteration based on the measured output power.

6. The inverter in claim 5 wherein a step size of the minor iteration depends on the operational parameters.

7. A central controller for a network of solar panel micro inverters comprising

central storage or memory, configured to store performance characteristics specifically determined for a plurality of panels according to location, model, date and/or installation characteristics, and a plurality of versions of firmware for a plurality of micro inverters; a central communications module configured to interrogate respective micro inverters; and

8. The central controller in claim 7 wherein the performance characteristics or firmware is more appropriate if it has a later version number than that stored in the respective micro inverter.