Method and device for operating a PV generator at the operating point of maximum power

Abstract

In a method for operating a PV generator (11) at the operating point of maximum power (MPP, 22) by means of an inverter (12) for converting electrical DC power (P_PV, P_DC) of the PV generator (11) into electrical AC power (P_AC), the PV generator (11) has a PV characteristic curve (21) and the inverter (12) is operated using a droop control curve (31). In this case, the droop control curve (31) specifies target values for the power (P_DC, PAC) of the inverter (12) on the basis of a DC voltage (U_DC) of the inverter (12), wherein the inverter (12) uses a power-electronic bridge circuit to set an actual value of the power (P_DC, P_AC) to the target value specified by the droop control curve (31) for the present DC voltage (U_DC), such that, during the operation of the PV generator (11), a working point (41) is adopted which corresponds to the intersection point between the present droop control curve (31) and the present PV characteristic curve (21). By iterative MPP tracking, the droop control curve (31) is changed step by step in order to shift the operating point (41) along the DC voltage (U_DC). The resulting actual value of the power (P_DC, P_AC) is detected, and the droop control curve (31) is changed in each tracking step such that the operating point (41) is shifted in a direction which depends on a direction of a change in the actual value of the power (P_DC, P_AC) at the operating point (41) owing to the change in the droop control curve (31) in a preceding tracking step. A further method relates to the operation of a PV generator (11) by means of at least two inverters (12) connected in parallel on the DC side and on the AC side. The invention also relates to an inverter (12) and to a device for carrying out the aforementioned method.

Description

[0001] 22-368-P-WO - 1 - submitted version

[0002] Method and device for operating a PV generator at the operating point of maximum power

[0003] TECHNICAL AREA

[0004] The application concerns methods and devices for operating a photovoltaic generator at the operating point of maximum power.

[0005] STATE OF THE ART

[0006] Common energy generation systems utilize photovoltaics (PV) and comprise a number of PV modules connected in series to form PV strings. One or more PV strings can also be connected in parallel to form a PV generator. A PV generator, comprising a single PV string or a parallel connection of several PV strings, can be connected to an input of an inverter. The inverter uses power electronic circuits to extract electrical power in the form of direct current (DC) from the PV generator, convert it into alternating current (AC), and feed it into an AC grid or utilize it for other purposes, such as powering a load. Depending on the size of the PV generators, the number of DC inputs, and the rated power of the inverter, a high electrical power output can be present at a single DC input of the inverter.For example, a so-called central inverter can have several dozen DC inputs and convert a nominal power of several megawatts, so that PV generators with several dozen to several hundred kilowatts of electrical nominal power can be connected to each DC input.

[0007] A photovoltaic (PV) generator has a characteristic curve that describes the current or power it generates as a function of the voltage applied to the generator under a given irradiance. This characteristic curve, also called the P(U) curve, typically has a maximum that corresponds to the maximum power point (MPP) of the PV generator. The peak power of a PV module is a given value determined using standardized methods and indicates the MPP power under optimal irradiance conditions. The sum of the peak powers of the PV modules in a PV generator is therefore a crucial design criterion for determining the maximum expected MPP power of a specific PV generator.

[0008] An inverter connected to a PV generator typically features MPP tracking, where a DC voltage at the inverter's DC-side connection, and thus the voltage at the PV generator, is adjusted and modified so that the PV module or generator delivers its maximum possible power output under a given solar irradiance. Various methods are known in the art to find and track this operating point of maximum power, the so-called Maximum Power Point (MPP), for example, when the irradiance on the PV generator changes.For example, EP 1 995 656 A1 discloses a method in which an inverter specifies a setpoint for a DC voltage to be applied to a PV generator and modifies it depending on the resulting PV power, wherein the direction of the modification of the setpoint depends on the change in power during previous modifications of the DC voltage applied to the PV voltage.

[0009] Instead of a central inverter with a rated power output that essentially corresponds to the sum of the peak power outputs of the connected PV generators, so-called string inverters with a fraction of the rated power of a central inverter can also be used. Provided that the rated power output of a string inverter is on the order of the peak power output of a given PV generator, this PV generator can be connected to this single string inverter.

[0010] However, if a given PV system with one or more high-peak-power PV generators is to be used with inverters whose rated power is only a fraction of the peak power of one of the PV generators, it can be advantageous to connect several inverters in parallel on both the DC and AC sides and to connect a PV generator to this parallel connection of multiple inverters on the DC side. This allows the power of the PV generator to be distributed across the inverters. In such a configuration, conventional MPP tracking methods, in which the inverters modify the DC voltage at their respective DC terminals, cannot be used because the same DC voltage is present at all inverters due to the parallel DC connection, and this voltage cannot be adjusted independently of any single inverter.Conventional MPP tracking methods in parallel-connected inverters would interfere with each other, and a controlled adjustment of the PV voltage would not be possible. Consequently, no correlation between PV voltage and PV power could be established. Furthermore, an imbalance in the power distribution between the individual inverters could occur, leading to lower efficiency or, in the event of an escalation, to overloading of individual inverters.

[0011] TASK OF INVENTION

[0012] The application is based on the task of demonstrating a method for operating a PV generator in which the PV generator is electrically connected to an inverter or to several inverters connected in parallel on the DC and AC sides, wherein, as submitted in 22-368-P-WO - 3 - the method is suitable for operating the PV generator stably and reliably at the operating point of maximum power.

[0013] SOLUTION

[0014] The problem is solved by a method having the features of claim 1 or claim 5, an inverter having the features of claim 12, and a device having the features of claim 13. Preferred embodiments are claimed by the respective dependent claims.

[0015] DESCRIPTION OF THE INVENTION

[0016] In a method for operating a PV generator at its maximum power point using an inverter to convert the PV generator's DC power into AC power, the PV generator has a characteristic curve, and the inverter is operated using a droop control curve. The droop curve specifies setpoints for the inverter's power output, i.e., the DC or AC power converted by the inverter, as a function of the inverter's DC voltage, i.e., the DC voltage applied to the inverter's DC side.Depending on the DC voltage and the droop control curve, AC setpoints for the AC power or AC current of the inverter can be determined. This can be achieved, for example, by using a DC droop control curve to specify DC setpoints as a function of the DC voltage, which are then compared with actual DC values ​​and converted into AC setpoints by a controller. Alternatively, a DC-AC droop control curve can directly specify the AC setpoints as a function of the DC voltage. The inverter uses a power electronic bridge circuit to adjust the actual power value to the setpoint specified by the droop control curve for the current DC voltage. This results in an operating point for the PV generator that corresponds to the intersection of the current droop control curve and the current PV characteristic curve.

[0017] A patented method is characterized by iterative MPP tracking, performed by stepwise modifying the droop control curve to shift the operating point along the DC voltage. The resulting actual power value, determined by the modified droop control curve, is acquired on the AC or DC side and compared with the actual value obtained in the previous step. In each step, the droop control curve is modified such that the operating point is shifted in a direction determined by the direction of any change in the actual power value at the operating point resulting from the modification of the droop control curve in a previous step.For example, if the operating point in a previous step (22-368-P-WO - 4 - submitted version) was shifted to a higher DC voltage by a corresponding change in the droop control curve, and the actual power value increased as a result, then the droop control curve is changed in the same direction as in the previous step to shift the operating point in the current step to an even higher DC voltage with (presumably) even higher power. Conversely, if the AC actual value decreased in the previous step, the droop control curve is changed in the opposite direction to that in the previous step. This ensures that the PV generator operates at its maximum power point after just a few steps.

[0018] The patented method has the advantage, among others, of being very simple and robust to implement, particularly in an inverter control unit. Furthermore, the method is very robust and predictable due to the use of a droop control curve, as the operating point of the PV generator quickly and stably adjusts to the intersection of the PV characteristic curve with the patented droop control curve.

[0019] The droop control curve can be defined by a base point and a slope, where the target power is zero at the DC voltage of the base point and increases essentially monotonically, e.g., linearly, beyond the base point voltage. The slope of the droop control curve can be constant and / or vary depending on the base point; that is, the droop control curve can be steeper at base points with relatively low DC voltages than at base points with relatively high DC voltages.

[0020] The droop control curve can be incrementally modified during iterative MPP tracking by shifting the base point of the droop control curve along the DC voltage. This shifts the intersection of the droop control curve with the current PV characteristic curve, and consequently, the operating point of the PV generator, also along the DC voltage. However, the change in DC power due to the shift in the operating point depends on the sign of the slope of the PV characteristic curve, which differs to the left and right of the MPP. This is taken into account in iterative MPP tracking by shifting the operating point in each subsequent step in a direction determined by the change in the actual power value resulting from the change in the droop control curve in the previous step.This allows the operating point of the PV generator to gradually approach the maximum power point (MPP) and to be adjusted accordingly in the event of changes in irradiance that lead to a change in the MPP's position. In one embodiment, the iterative MPP tracking can be carried out using a three-point method, where each step of the MPP tracking comprises several measurements of the actual power value at at least two different droop control curves. 22-368-P-WO - 5 - submitted version.

[0021] A registered inverter for operating a PV generator is designed to operate the PV generator using the described method in such a way that the operating point of maximum power is targeted and, if necessary, pursued.

[0022] In an alternative method for operating a PV generator at its maximum power point (MPP), at least two inverters are used, connected in parallel on the DC and AC sides and configured to convert the PV generator's DC power into AC power. The PV generator itself has a characteristic curve and is connected to all inverters via the DC-side parallel connection at the same DC voltage. The inverters are operated using individual droop control curves, which specify setpoints for the power of each inverter as a function of its DC voltage, in particular setpoints for the AC power or the equivalent AC current.The inverters adjust their respective actual power value to the setpoint specified by their individual droop control curve for the current DC voltage using a power electronic bridge circuit.

[0023] A key concept of this patented method is that the sum of the individual droop control curves forms an overall droop control curve, which determines the behavior of the parallel-connected inverters in conjunction with the PV generator. Specifically, during operation of the PV generator, an operating point is established that corresponds to the intersection of the overall droop control curve with the current PV characteristic curve. On the DC side, such an operating point is characterized by a DC voltage and a DC current from the PV generator, whereby the resulting DC power is essentially completely converted by the inverters into AC currents with a corresponding total AC power output.

[0024] In the patented method, iterative MPP tracking is performed by a control unit determining the total power of the inverters, in particular by direct measurement at a common grid connection point or as the sum of the current actual values ​​of the AC or DC power of all inverters, and incrementally modifying the overall droop control curve to shift the operating point along the DC voltage. The total power resulting from the modified overall droop control curve is recorded and compared with the total power established in the previous step. In each step, the overall droop control curve is modified such that the operating point is shifted in a direction that depends on the direction of any change in total power due to the modification of the overall droop control curve in a previous step.

[0025] Analogous to the method described above with only one inverter, the PV generator can thus be operated at its maximum power point after just a few steps (22-368-P-WO - 6 - submitted version). The method with at least two inverters also offers the same advantages as the method described above with one inverter. In addition, the usable power can be multiplied compared to the nominal power of a single inverter by connecting a large number of inverters in parallel on both the DC and AC sides, so that even very large PV generators can be operated at their maximum power point with comparatively compact inverters. Conversely, a large PV generator that was previously operated with a central inverter, for example, can also be operated with a large number of significantly smaller inverters, as per the application.

[0026] The overall droop control curve can be changed stepwise within the iterative MPP tracking by changing the individual droop control curve of at least one of the inverters or of exactly one of the inverters.

[0027] Droop control curves can be defined, in particular, by a base point and a slope, where the target power is zero at the DC voltage of the respective base point and increases linearly, for example, beyond the base point voltage. The slope of the droop control curves can be constant and / or vary depending on the base point; that is, the droop control curve can be steeper at base points with relatively low DC voltages than at base points with relatively high DC voltages.

[0028] To change the overall droop control curve, the individual droop control curve of an inverter can be modified by shifting the base point of this individual droop control curve relative to a base point reference voltage. This also changes the overall droop control curve, and the intersection point of the overall droop control curve with the current PV characteristic curve, and thus the operating point of the PV generator, is shifted along the DC voltage.

[0029] In one embodiment of the method, the iterative shift of the feedpoints can have a substantially fixed step size for successive MPP tracking steps. The individual droop control curves can have the same slope, and their feedpoints can be separated by a maximum of one step size. This embodiment is particularly advantageous when the at least two inverters are substantially identical in design. Due to the droop control curves being shifted relative to each other by a maximum of one step size, the DC power of the PV generator is distributed largely symmetrically between the at least two inverters.A completely symmetrical distribution of the DC power of the PV generator across at least two identical inverters is regularly achieved if the droop control curves are identical for all inverters in one step of the MPP tracking; preferably, however, the feedpoints differ from each other in at least one of two consecutive MPP tracking steps (22-368-P-WO - 7 - filed version), which regularly results in a slight asymmetry of the power distribution. This inherently undesirable asymmetry is, however, accepted according to the application and outweighed by the advantages resulting from the finer adjustment of the overall droop control curve with a fraction of the step size of the shift of the feedpoints of the individual droop control curves. Furthermore, the method is insensitive to communication latencies that can occur, for example, when...The control unit can only communicate with individual inverters and not all of them at exactly the same time, and even if communication fails, the inverters can initially continue to run stably with a given droop control curve.

[0030] In one embodiment of the method, the step size depends on the change in total power due to the change in the overall droop control curve in a previous step. The step size can be larger, in particular, the greater the change in total power in the previous steps. This allows the iterative MPP tracking to reach the maximum power point (MPP) more quickly when the current operating point is relatively far from the MPP and is changed towards it in correspondingly large steps. Furthermore, it allows the MPP to be set more accurately and monitored more effectively when the PV generator is operating at the maximum power point. Specifically, in one embodiment, the iterative MPP tracking can be performed using a three-point method, where each step of the MPP tracking comprises multiple measurements of the total power at at least two different overall droop control curves.

[0031] The methods described in the application can include droop control curves that link the DC power or the AC power (or the corresponding currents at given voltages) with the DC voltage. In particular, the droop control curves can specify DC setpoints, especially a DC power or a DC current as a function of the DC voltage, which are compared with corresponding DC input values ​​and converted into suitable AC setpoints by means of a controller. Alternatively, the droop control curves can directly specify AC setpoints, especially an AC power or an AC current as a function of the DC voltage. For controlling the power electronic bridge circuit, the resulting AC setpoints are then generally used and compared with corresponding AC input values, so that ultimately, in every case, the intersection of the PV characteristic curve with the droop control curve, or, in the case of multiple inverters, with the overall droop control curve, is set as the operating point.

[0032] A device for operating a PV generator, as described in the application, comprises at least two substantially identical inverters connected in parallel on the DC and AC sides, and includes a control unit. The inverters and the control unit are configured to operate the PV generator at its maximum power point using the method described above. In particular, the control unit may be configured to record the current DC or AC load values ​​of all inverters or the total AC power and, in each step of the iterative MPP tracking, to modify the droop control curve of at least one of the inverters depending on the recorded total power.

[0033] The device's control unit can be a separate unit or integrated into one of the inverters. In particular, one of the control units already present in the inverters can be designated as the device's control unit. The control unit is preferably connected to all or the other essentially identical inverters via a communication interface, for example, a LAN bus, a CAN bus, or another suitable communication interface.

[0034] In a preferred embodiment, the total rated power of the device, which corresponds to the sum of the rated powers of the substantially identical inverters, is at least 50 percent of the peak power of the PV generator, and particularly preferably, the total rated power of the device is greater than 70 percent of the peak power of the PV generator. With such a design of a power generation system, the MPP power can temporarily exceed the rated power of the inverters, so that the inverters must limit the power to a value lower than the MPP power. In such a case, the MPP cannot be achieved by the method according to the application due to its inherent limitations, but it is ensured that one of the usually two intersection points of the overall droop control curve with the PV characteristic curve is stably established.This does mean that a small portion of the maximum possible PV power has to be sacrificed; however, this disadvantage is offset by the possibility of using significantly more compact inverters.

[0035] BRIEF DESCRIPTION OF THE FIGURES

[0036] The invention will now be further explained and described with reference to exemplary embodiments shown in the figures.

[0037] Fig. 1 shows an energy generation plant with a PV generator and an inverter;

[0038] Fig. 2 shows an exemplary PV characteristic curve and a droop control curve;

[0039] Fig. 3 shows a section of the PV characteristic curve from Fig. 2 and various droop control curves in an embodiment of the method according to the application; 22-368-P-WO - 9 - filed version

[0040] Fig. 4 shows the PV characteristic curve from Fig. 2 and another droop control curve;

[0041] Fig. 5 shows a further section of the PV characteristic curve from Fig. 2 and various droop control curves in a further embodiment of the method according to the application;

[0042] Fig. 6 shows an energy generation plant with a PV generator and a device according to the application with several inverters;

[0043] Fig. 7 shows an exemplary PV characteristic curve, identical individual droop control curves and a resulting overall droop control curve;

[0044] Fig. 8 shows a section of the PV characteristic curve from Fig. 7, different individual droop control curves and a resulting overall droop control curve;

[0045] Fig. 9 shows the section of the PV characteristic curve from Fig. 8 and various overall droop control curves in an embodiment of the method according to the application;

[0046] Fig. 10 shows the energy generation plant from Fig. 6 with an embodiment of the device according to the application with several inverters;

[0047] Fig. 11 shows exemplary time courses of electrical parameters and control variables when carrying out the method according to the application using the device according to the application with several inverters;

[0048] Fig. 12 shows a section of the time series from Fig. 11.

[0049] FIGURE DESCRIPTION

[0050] Fig. 1 shows a power generation system 10 with a PV generator 11 and an inverter 12. Given a certain amount of solar irradiance, the PV generator 11 produces DC power P_PV in the form of a current l_PV at a voltage U_PV. The functional relationship between the generated power P_PV and the applied voltage U_PV is defined by a PV characteristic curve, see Figs. 2ff. The PV generator 11 is connected to the inverter 12 on the DC side, so that the DC voltage U_DC at the inverter 12 essentially corresponds to the PV voltage U_PV. The inverter 12 is operated by a control unit 13 and converts the PV power P_PV into an AC current l_AC at an AC voltage U_AC, so that the AC power P_AC is generated by the inverter 12.In this embodiment, the AC power P_AC is fed into an AC network 15 via a transformer 14, which is generally optional; this is a common configuration for high-power power generation plants 10 in the range of a few hundred kilowatts (22-368-P-WO - 10 - submitted version) up to a few megawatts of electrical power. Other uses of the AC power P_AC, e.g., for operating an island grid or a dedicated load, are also possible.

[0051] In an embodiment according to the application, the control unit 13 is configured to detect the DC voltage U_DC and to control the inverter 12 so that it generates a power in the form of a DC power P_DC or an AC current l_AC or an AC power P_AC, which is specified by a droop control curve 31 as a function of the DC voltage U_DC.

[0052] Fig. 2 shows an exemplary PV characteristic curve 21 and a droop control curve 31. The PV characteristic curve 21 is a property of the PV generator 11 and is therefore considered given in a specific implementation and irradiance situation. At a DC voltage of zero, the PV generator 11 is considered short-circuited and consequently delivers no DC power P_DC. The PV characteristic curve 21 rises with increasing voltage and reaches an operating point with maximum power P_MPP at the MPP voltage U_MPP. The operating point of maximum power is referred to below as MPP 22. Beyond MPP 22, the PV characteristic curve 21 drops sharply again and reaches an open-circuit state at an open-circuit voltage U_0, in which the power P_DC is again zero.

[0053] The droop control curve 31 specifies setpoint values ​​for the power P or the equivalent AC current l_AC, which the inverter 12 is to set depending on the current DC voltage U_DC at the inverter 12, e.g., by means of the control unit 13. The droop control curve, as shown in Fig. 2, is characterized in particular by a base point at a base point voltage U_P0 and a slope. The slope of the droop control curve 31 can be linear, as in the example shown in Fig. 2; other slopes are also possible. The slope ends at a maximum value P_max, which can correspond to a rated power of the inverter 12 and cannot or must not be exceeded, so that the droop control curve 31 exhibits the maximum value P_max constantly as it progresses to higher voltages.

[0054] In a specific implementation, the open-circuit voltage U_0 of the PV generator 11 can range between approximately 500 volts and 2000 V, depending on the number of PV modules used. The MPP 22 is typically located at an MPP voltage U_MPP, which is approximately 80 percent of the open-circuit voltage and is also influenced by environmental conditions such as irradiance and temperature. For a specific operation of the PV generator 11, the base point of the droop control curve 31 lies below the open-circuit voltage U_0 and ideally below the MPP voltage U_MPP. The slope of the droop control curve 31 can be selected such that the difference between the base point voltage U_P0 and the voltage at which the maximum power P_max is reached is a fraction of the open-circuit voltage U_0. The maximum power P_max, as submitted in the 22-368-P-WO - 11 version, should generally be greater than the MPP power P_MPP. Specifically, the slope of the droop control curve 31 can, for example,The P_max value is 1-2% per 1 volt U_DC, so that the droop control curve 31 covers a DC voltage range of approximately 50-100 volts U_DC between 0 and P_max, thus encompassing approximately 1 / 5 to 1 / 40 of the PV characteristic curve 21, depending on the specific position of the open-circuit voltage U_0. An optimum slope for the droop control curve 31 can, for example, result from the specific behavior of the PV characteristic curve 11 during fluctuations in irradiance. The steeper the droop control curve 31, the greater the change in the setpoint for the power when the MPP shifts due to irradiance, and the higher the requirements for measuring the DC voltage U_DC and the stability limits of the control system. A lower slope of the droop control curve 31 can therefore be advantageous for the stability of the control, but on the other hand leads to a smaller change in performance when the droop control curve 31 is shifted as per the application within the framework of MPP tracking (see below).

[0055] During the operation of the energy generation plant 10 according to Fig. 1 with a PV generator 11 with PV characteristic curve 21 and an inverter 12 with droop control curve 31, an operating point 41 is automatically established, which corresponds to the intersection of the droop control curve 31 with the PV characteristic curve 21. At this operating point 41, the PV power P_PV generated by the PV generator 11 is essentially identical to the AC power P_AC generated by the inverter 12 from the DC power P_DC. The system consisting of the PV generator 11 and the inverter 12 is therefore in energy equilibrium at operating point 41.Any deviations from this operating point 41, for example a higher DC voltage U_DC, lead to a counter-reaction of the inverter 12, which briefly converts a higher power P than the generated PV power P_PV and thus lowers the DC voltage U_DC, or conversely, at DC voltages U_DC below the operating point 41, briefly converts a lower power P than the generated PV power P_PV and thus increases the DC voltage U_DC.

[0056] In the patented method, iterative MPP tracking is performed to operate the PV generator 11 using an inverter 12 by stepwise modifying the droop control curve 31 to shift the operating point 41 along the DC voltage U_DC. In particular, the feedpoint voltage U_P0, and thus the entire droop control curve 41, can be shifted horizontally, as indicated in Fig. 2 by the double arrow on the droop characteristic curve 41.

[0057] Fig. 3 shows a section of the PV characteristic curve 21 from Fig. 2 and various droop control curves 31.1-31.3, which are successively set during the execution of the iterative MPP tracking according to the application. Initially, the droop control curve 31.1 is set, and the operating point 41.1 is established, which corresponds to the intersection of the current droop control curve 31.1 with the PV characteristic curve 21. In a subsequent step of the iterative MPP tracking version filed in 22-368-P-WO - 12 -, the modified droop control curve 31.2 is set, and the shifted operating point 41.2 is established at the correspondingly shifted intersection of the droop control curve 31.2 with the PV characteristic curve 21. The resulting power P at operating point 41.2 is now compared with the resulting power P at the previous operating point 41.1, for example by the control unit 13 recording and evaluating the corresponding actual power values ​​(see Fig. 1).Since the power P increased during the change from droop control curve 31.1 to droop control curve 31.2, the operating point is shifted further in the same direction in the next step of iterative MPP tracking by setting droop control curve 41.3. At the resulting shifted operating point 41.3, the power P has increased further. This increase in the actual power value due to the change in droop control curve 31 to the right along the DC voltage U_DC can be registered, for example, in the control unit 13 to determine the direction in which operating point 41 should be shifted in the next step—in this case, further to the right. This can be achieved by a corresponding change in droop control curve 31 to the right along the DC voltage U_DC (not shown).As a result, the droop control curve 31 and thus the operating point 41 are gradually shifted to the right along the PV characteristic curve 21, as long as this increases the electrical power at the operating point 41, which is registered in particular by recording the actual value of the AC power and / or the DC power in the equilibrium state after a tracking step by the control unit 13.

[0058] Figure 4 illustrates the desired target state of the patented method, in which the PV generator 11 is operated at the maximum power point (MPP) 22. This state is achieved by shifting the droop control curve 31 to the right (starting from a low DC voltage U_DC, see Figure 3) or to the left (starting, for example, from the open-circuit voltage, not shown) until the droop control curve 31 intersects the PV characteristic curve 21 at MPP 22. At this point, the iterative MPP tracking continues, and the droop control curve is further modified stepwise, in particular by shifting the feedpoint voltage U_P0 and thus the entire droop control curve 41 horizontally to the right or left, as indicated in Figure 4 by the double arrow on the droop characteristic curve 41.This allows changes in the position of the MPP 22 to be automatically followed, for example when the PV characteristic curve 21 changes due to changing irradiance on the PV generator 11.

[0059] Fig. 5 shows a section of the PV characteristic curve from Fig. 4 and various droop control curves 31.4-31.7, which are set successively when performing the iterative MPP tracking according to the application. Initially, for example as a result of a series of MPP tracking steps including droop control curves 31.1-31.3 (see Fig. 3), the version filed in 22-368-P-WO - 13 - is

[0060] Droop control curve 31.4 is set, and operating point 41.4 is established, corresponding to the intersection of the current droop control curve 31.4 with the PV characteristic curve 21. Assuming that the power P at operating point 41.4 is greater than the power P at a previous operating point 41.3 to its left (not shown), the operating point is shifted further to the right in a subsequent step of the iterative MPP tracking by setting the droop control curve 31.5 with a correspondingly higher DC voltage U_DC shifted to the feedpoint voltage, and the shifted operating point 41.4 is established at the correspondingly shifted intersection of the droop control curve 31.4 with the PV characteristic curve 21.

[0061] Operating point 41.5 is now located at MPP 22. Since the power P has increased slightly again as a result of this step, the operating point is shifted further to the right in the next step of the iterative MPP tracking by adjusting the droop control curve 31.6 with a correspondingly higher DC voltage U_DC and a correspondingly higher feedpoint voltage. The resulting operating point 41.6 is already beyond MPP 22, so the power P has decreased again; therefore, the subsequent step is performed in the opposite direction, i.e., to the left, by adjusting the droop control curve 31.7, which in this example is identical to the droop control curve 31.5 from a previous step. The power P briefly reaches the MPP power P_MPP again, and subsequently, the procedure oscillates left and right around MPP 22, tracking any shifts in the MPP 22.

[0062] By using a more complex implementation of iterative MPP tracking, in particular a 3-point method with a more precise evaluation of the effects of a tracking step on performance, any disturbances such as changes in irradiance can also be taken into account and mismatches avoided, see also Fig. 11f.

[0063] Fig. 6 shows a power generation system 100 with a PV generator 11 and three inverters 12. The inverters 12 are essentially identical in design and connected in parallel on both the DC and AC sides. The PV generator 11 is connected to all inverters 12 on the DC side, so that the current PV voltage U_PV is identically present as a DC voltage U_DC at all inverters 12. The inverters 12 are communicatively connected to a control unit 13 and convert the PV power P_PV into three AC currents l_AC at the common AC voltage U_AC, and thus into three AC powers P_AC, so that the system from the three inverters 12 generates a combined total AC power P_sys and feeds it into an AC grid 15 via an optional transformer 14.

[0064] The inverters 12 are each operated with a droop control curve, so that they each convert a power P and, in particular, generate an AC power P_AC, which is specified by a respective droop control curve 35 as a function of the DC voltage U_DC (22-368-P-WO - 14 - submitted version). The droop control curve 35 used here has essentially the same features as the droop control curve 31 disclosed in Fig. 2ff., except that the respective rated power of the inverters 12, and thus the maximum value P_max of each inverter 12 individually, is about three times smaller than the rated power of the PV generator 11 (see Fig. 7ff.). These reduced rated powers are sufficient because, due to the droop control, the PV power P_PV is distributed essentially evenly among the three inverters 12 with essentially identical droop control curves 35.

[0065] Fig. 7 shows an exemplary PV characteristic curve 21 and the identically implemented droop control curves 35.a-35.c of the inverters 12. The individual droop control curves 35.a-35.c each specify setpoint values ​​for the power P for the respective inverter 12 as a function of the current DC voltage U_DC. The sum of the individual droop control curves 35.a1-35.c1 forms an overall droop control curve 36.

[0066] In the operation of the energy generation plant 100 according to Fig. 6 with a PV generator 11 with PV characteristic curve 21 and three inverters 12 with identical droop control curves 35. a1- 35. c1, an operating point 41.1 is automatically established, which corresponds to the intersection of the overall droop control curve 36.1 with the PV characteristic curve 21. At this operating point 41.1, the PV power P_PV generated by the PV generator 11 is essentially identical to the total power P_sys of the AC power P_AC generated by the inverters 12. The system consisting of PV generator 11 and inverters 12 is therefore in energy equilibrium at operating point 41.1, and any deviations from operating point 41.1 are automatically counteracted. The total power P_sys is distributed essentially equally among the inverters 12.

[0067] In the patented method for operating the PV generator 11 using at least two, here three, inverters 12, iterative MPP tracking is performed by the control unit 13 determining the total power P_sys directly or as the sum of the current AC powers P_AC of the three inverters 12 and incrementally modifying the overall droop control curve 36 to shift the operating point along the DC voltage. Analogous to the method shown in Fig. 2ff., all individual droop control curves 35.a-35.c can be modified identically so that the overall droop control curve 36 is shifted horizontally, and thus the operating point 41 is shifted along the DC voltage U_DC (see Figures 3 and 5).

[0068] Alternatively or additionally, the overall droop control curve 36 can be modified in each step of the iterative MPP tracking by modifying at least one of the individual droop control curves 35.a-35.c of the inverters 12. 22-368-P-WO - 15 - submitted version

[0069] Fig. 8 shows a section of the PV characteristic curve 21 from Fig. 7 and a modified overall droop control curve 36.2 compared to the overall droop control curve 36.1 from Fig. 7. The modified droop control curve 36.2 is formed from the individual droop control curves 35.a1, 35.b1, and 35.c2. The droop control curves 35.a1 and 35.a2 remain unchanged compared to the overall droop control curve 36.1 from Fig. 7; however, instead of the droop control curve 35.c1 in Fig. 7, the modified droop control curve 35.c2 is now used in Fig. 8. Analogous to Fig. 2, the modified droop control curve 35.c2 can, in particular, exhibit a shifted feedpoint voltage U_P0 and thus be horizontally shifted overall compared to the droop control curve 35.c1.

[0070] The resulting overall droop control curve 36.2 exhibits a non-linear behavior. The feedpoint voltage U_P0 corresponds to the minimum feedpoint voltage U_P0 of the individual droop control curves 35.a1-35.c2. A first section of the overall droop control curve 36.2 has a first slope that corresponds to the sum of the slopes of the individual droop control curves 35.a1 and 35.b1. A second section of the overall droop control curve 36.2, between the feedpoint voltage U_P0 of the droop control curve 35.c2 and the attainment of the maximum power P_max of the inverters 12 by the droop control curves 35.a1 and 35.b1, exhibits a second slope that corresponds to the sum of the slopes of all three droop control curves 35.a1-35.c2 and is therefore identical to the continuous slope of the overall droop control curve 36.1. A third section of the overall droop control curve 36.2 extends from the attainment of the maximum power P_max by the droop control curves 35.a1 and 35.b1.b1 until the maximum power P_max is reached by the droop control curve 35. c2 and has a third slope that corresponds to the slope of the droop control curve 35. c2.

[0071] The modified overall droop control curve 36.2 resulting in the example according to Fig. 8 intersects the PV characteristic curve 21 at the operating point 41.2, which is shifted along the DC voltage compared to the operating point 41.1 reached with the overall droop characteristic curve 36.1.

[0072] Within the framework of iterative MPP tracking, the total power P_sys achieved at operating point 41.2 with the modified overall droop control curve 36.2 can be compared with the total power P_sys at operating point 41.1.

[0073] Fig. 9 shows again the section of the PV characteristic curve from Fig. 8 and various overall droop control curves 36.1-36.4 in an embodiment of the patent application method for operating the PV generator 11 at the maximum power point (MPP) using three inverters 12. 22-368-P-WO - 16 - filed version

[0074] The overall droop control curves 36.1-36.4 are set sequentially during iterative MPP tracking. First, the overall droop control curve 36.1 is set, which is formed from the sum of the identical individual droop control curves 35.a1-35.c1 (see Fig. 7), and the operating point 41.1 is established at the intersection of the overall droop control curve 36.1 with the PV characteristic curve 21. In a subsequent step of the iterative MPP tracking, the modified overall droop control curve 36.2 is set by setting the modified, in particular shifted, droop characteristic curve 35. c2 for one of the inverters 12 instead of the individual droop characteristic curve 35. c1 (compare Fig. 8), and the shifted operating point 41.2 is set at the correspondingly shifted intersection point of the overall droop control curve 36.2 with the PV characteristic curve 21.

[0075] The resulting total power P_sys at operating point 41.2 is now compared with the resulting total power P_sys at the previous operating point 41.1, for example by the control unit 13 determining and evaluating the total power P_sys directly or as the sum of the respective AC powers P_AC or DC powers of all inverters 12 in equilibrium.

[0076] Since the total power P_sys increased in the step of changing from total droop control curve 36.1 to total droop control curve 36.2, the operating point is shifted further in the same direction in the following step of iterative MPP tracking by setting the droop control curve 36.3. For this purpose, one of the droop control curves 35.a1 or 35.b1 is modified and, in particular, shifted so that it corresponds to the droop control curve 35.c2 modified in the previous step; this shifts the second section of total droop control curve 36.3 largely parallel to the corresponding second section of the preceding total droop control curve 36.2. The resulting operating point 41.3 is shifted further along the DC voltage, and the total power P_sys is further increased compared to the total power P_sys at the preceding operating point 41.2.

[0077] This increase in the total AC power P_sys can be evaluated, for example, in the control unit 13 to determine the direction in which the operating point 41 should be shifted in the next step, in this case further to the right towards higher DC voltages U_DC. For this purpose, the overall droop control curve 36.4 is set in the next step, which is again formed by identical individual droop control curves 35.a2-35.c2 (not shown, but see droop control curve 35.c2 in Fig. 8). As a result, the overall droop control curve 36, and thus the operating point 41, is shifted stepwise to the right along the PV characteristic curve 21, in particular by changing one of the droop control curves 35.a-35.c per step. Only the inverter 12 needs to readjust to the changed droop control curve 35. the other inverters 12 with unchanged Droop- 22-368-P-WO - 17 - submitted version

[0078] Control curve 35 only needs to follow the comparatively small change in the overall droop control curve 36 of the overall system.

[0079] The modification of a droop control curve 35.a-35.c can in particular include a shift of its individual footpoint voltage U_P0 with a predetermined step size, wherein the individual droop control curves 35.a-35.c have the same slope and their footpoints are preferably at most one step size apart during the iterative MPP tracking.

[0080] In the further course of the iterative MPP tracking, the criteria explained in connection with Figures 3 and 5 are applied analogously, in particular with regard to the selection of the direction of the shift of the operating point 41 in a step depending on the direction of the shift of the operating point 41 in a previous step and the resulting change in the total power P_sys. As a result, after a few steps, the method achieves operation of the PV generator at the operating point of maximum power MPP, which is maintained and, if necessary, tracked in particular by oscillating the operating point around the MPP.

[0081] Fig. 10 shows the power generation plant from Fig. 6 in an embodiment of the device according to the application with several inverters 12 and an exemplary schematic representation of each control system 16. The AC power P_AC of the inverters 12 is set by means of respective current controllers 17, by adjusting the respective AC current l_1, l_2, l_3 at a given AC voltage U_AC to a respective setpoint I_setpoint I_setpoint1, I_setpoint2, I_setpoint3. The respective AC setpoint l_setpoint is determined by means of a respective droop control curve 35.a, 35.b, 35.c as a function of the DC voltage U_DC, which is essentially identical for all inverters 12. The AC setpoints l_setpoint can be derived directly from an AC setpoint power, for example, taking into account the AC voltage U_AC, if the droop control curves 35 specify the AC power P_AC as a function of the DC voltage U_DC.Alternatively, the AC setpoints l_setpoint can also be determined by means of an additional controller (not shown), in particular a P controller or a PI controller, from the deviation of a DC power P_DC from a DC setpoint of the respective inverter 12, if the droop control curves 35 specify the DC power P_DC as a function of the DC voltage U_DC.

[0082] The droop control curves 35 used specifically in Regulations 16 have an identical slope. However, the respective base point U_P0 of the droop control curves 35 has an individual offset dU1, dU2, dU3 relative to a reference base point (not shown). The individual offsets dU1, dU2, dU3 are specified by the control unit 13, whereby the reference base point can be chosen practically arbitrarily (22-368-P-WO - 18 - submitted version) and corresponds, for example, to an open-circuit voltage of the PV generator 11 or a minimum DC voltage U_DC of the inverters 12.

[0083] The control unit 13 performs the MPP tracking as per the application by incrementally modifying the overall droop control curve 36. In each step of the iterative MPP tracking, the individual droop control curve 35 of at least one or exactly one of the inverters 12 is preferentially modified by increasing or decreasing the respective individual shift dll by one step (see Figures 8 and 9). The direction of the change in an individual shift dll, and thus the direction of the shift in the operating point of the overall system by the control unit 13, depends on how the total power P_sys changed in the previous step of the MPP tracking (see Figures 11 and 12).

[0084] Fig. 11 shows exemplary time courses of electrical parameters and control variables when carrying out the method according to the application using the device according to the application with several inverters 12.

[0085] The first diagram in Fig. 11 shows the MPP power P_MPP of a simulated PV generator 11 and the total power P_sys resulting from the execution of a patent-compliant method. It can be seen that the total power P_sys follows a change in the MPP power, here at approximately t=20s, practically immediately, and thus the PV generator 11 is operated at its maximum power point both before and after the change.

[0086] The second diagram in Fig. 11 shows the individual AC powers P_AC of the inverters 12, which together make up the total power P_sys. The individual iterative MPP tracking steps are already evident, as the individual AC powers deviate stepwise from one another, while the mean value changes only very slightly for a given MPP power (before and after t=20s).

[0087] The third diagram in Fig. 11 shows the MPP voltage P_MPP of the simulated PV generator 11 and the resulting DC voltage U_DC when a patented method is executed. The simulated change in MPP power P_MPP at t=20s is correlated with a simulated change in MPP voltage U_MPP. Before the change in MPP power, the simulated PV generator 11 is operated at the MPP, and the DC voltage U_DC essentially corresponds to the MPP voltage U_MPP. After the simulated change in MPP power and MPP voltage at t=20s, a stepwise shift of the operating point within the framework of MPP tracking and the resulting approximation of the DC voltage U_DC to the MPP voltage are clearly visible. From about t=140s 22-368-P-WO - 19 - submitted version, the DC voltage U_DC essentially corresponds to the MPP voltage and oscillates around the MPP voltage due to the continued MPP tracking steps.

[0088] The fourth diagram shows the individual shifts dU1, dU2, dU3 of the feedpoints U_PO of the droop control curves 35a, 35b, 35c (see Fig. 8) relative to a reference feedpoint, which here corresponds approximately to the open-circuit voltage U_0 of the PV generator 11. The individual MPP tracking steps are again visible, each comprising a change in the shift dll for exactly one of the three inverters 12. The shifts dll are negative because the MPP voltage U_MPP, and thus also feedpoints U_P0, are (significantly) below the open-circuit voltage U_0 of the PV generator 11, which serves as the reference feedpoint. After the simulated change of the MPP at t=20s, the shifts dU approach an asymptote, which is reached at approximately t=140s. After that, only one of the shifts dU, here dU1 or dU2, alternates between higher and lower values ​​for testing purposes in order to track the MPP within the framework of MPP tracking.

[0089] Fig. 12 shows an enlarged section of the time series from Fig. 11, starting with the change in the MPP power P_MPP and the MPP voltage U_MPP of the simulated PV generator 11 at t=20s.

[0090] In the first diagram of Fig. 12, it can be seen that the total power P_sys follows the change in MPP power at t=20s practically immediately, initially being slightly below the changed MPP power P_MPP and then asymptotically approaching the MPP power P_MPP through MPP tracking; in the third diagram, an analogous behavior of the DC voltage U_DC in relation to the MPP voltage U_MPP can be seen.

[0091] In the fourth diagram of Fig. 12, the individual shifts dU1 , dU2, dU3 of the foot points U_P0 of the droop control curves 35a, 35b, 35c are shown, which ultimately lead to the individual AC powers P_AC1 , P_AC2, P_AC3.

[0092] The time courses in Fig. 11 and Fig. 12 are based on a patented method that uses a 3-point method for MPP tracking. The individual steps of the 3-point method are delineated from each other in Fig. 12 by the plotted time points t0-t4. At time t0, the 3-point MPP tracking is performed in a situation with three identical displacements dU1, dU2, dU3 and correspondingly identical individual AC powers P_AC1, P_AC2, P_AC3 in the second diagram (compare the overall droop control curve 36.1 in Fig. 9).

[0093] Between tO and t1, an MPP tracking step is performed, comprising three sub-steps in which the individual shift dU1 is increased by one step, decreased again, and then increased once more. Accordingly, the DC voltage U_DC increases in the first sub-step (22-368-P-WO - 20 - submitted version), decreases again in the second sub-step, and increases once more in the third sub-step. In all three sub-steps, the resulting total power P_sys at equilibrium is determined, and then it is essentially checked whether the total power P_sys has increased or decreased due to the shift of dU1 and the resulting shift of the operating point along the DC voltage U_DC in the first and third sub-steps.

[0094] Since the total power P_MPP in the first and third substeps is consistently higher at the operating point with the higher DC voltage U_DC than in the second substep with the lower DC voltage U_DC of the initial state at tO, the operating point is shifted further to higher DC voltages U_DC in the next MPP tracking step between times t1 and t2 for testing purposes. To achieve this, the individual shift dU1 is left at the higher level of the previous step, and the individual shift dU2 is increased, decreased, and then increased again for testing purposes.

[0095] Since the total power P_sys is consistently higher in the MPP tracking step between t1 and t2 in the substeps with higher DC voltage U_DC, the operating point in the next MPP tracking step between times t2 and t3 is shifted further to higher DC voltages U_DC for testing purposes. This is done by increasing the individual shift dU3 in addition to the individual shifts dU1 and dU2, so that the three individual shifts dU1, dU2, dU3 are identical again (compare the overall droop control curve 36.4 in Fig. 9). Since the total power P_sys is also higher at the operating point with higher DC voltage U_DC in this MPP tracking step, the operating point in the next MPP tracking step between times t3 and t4 is shifted further to higher DC voltages U_DC for testing purposes.As in the MPP tracking step between tO and t1, the individual shift dU1 is first increased by one step, then decreased again and increased by another step.

[0096] The operating point is also shifted to higher DC voltages U_DC in subsequent MPP tracking steps after time t4 by cyclically changing the individual shifts dU1, dU2, dU3 in the described sequence, as long as this increases the total power P_sys. The individual base points U_P0 of the inverters 12 are preferably at most one step size apart and are different from each other in at least one of two consecutive MPP tracking steps.

[0097] If the MPP is indeed reached at approximately t=140s, the operating point is experimentally oscillated around the MPP by alternately increasing the displacement dU1 and decreasing the displacement dU2. However, since the simulated PV generator 11 is already at the MPP, these displacements invariably result in a reduction of the total power P_sys, so the direction of the operating point displacement is reversed in each MPP tracking step. To adjust the MPP more precisely in this situation, the step size of the individual displacement dU can be reduced, as can be seen from the smaller amplitude of the changes in the displacements dU and the smaller difference between the output powers P_AC1, P_AC2, P_AC3 in the MPP steps from approximately t=175s onwards.

[0098] Overall, the methods and devices described in the application thus offer a rapid and robust method for locating and tracking the MPP of a PV generator 11, in particular by connecting several inverters 12 in parallel on both the DC and AC sides, and processing the MPP power P_MPP and P_sys in essentially equal proportions by the inverters 12. Control commands are exchanged relatively infrequently between the control unit 13 and the controllers 16 in the inverters 12, namely only to modify an individual droop control curve 35. This places no special demands on the specific communication links used, and even in the event of a communication failure, the power generation plant can continue to operate stably with the last transmitted settings.

[0099] 2-368-P-WO - 22 - submitted version

[0100] REFERENCE MARK LIST

[0101] 10 Energy generation plant

[0102] 11 Photovoltaic generator (PV generator)

[0103] 12 inverters

[0104] 13 Control unit

[0105] 14 T transformer

[0106] 15 AC power grid

[0107] 16 Regulation

[0108] 17 current regulators

[0109] 21 PV characteristic curve

[0110] 22 Maximum Power Point (MPP)

[0111] 31, 31.1 - 31.7 Droop control curve 35a, 35b, 35c Individual droop control curve 36.1-36.4 Total droop control curve

[0112] 41, 41.1 - 41.7 Operating point tO - 14 Time

[0113] U_PV, U_DC, U_AC voltage

[0114] U_0 Open circuit voltage

[0115] U_MPP MPP voltage

[0116] U_P0 Base point voltage dU1 , dU2, dU3 Displacement l_PV, l_DC, l_AC Current

[0117] I_set1 - I_set3 Current setpoint

[0118] P_PV, P_DC, P_AC Power

[0119] P_MPP MPP power

[0120] P_ys Total Power

[0121] P_max Maximum power

Claims

22-368-P-WO - 23 - submitted version PATENT CLAIMS 1. Method for operating a PV generator (11) at the maximum power point (MPP, 22) by means of an inverter (12) for converting electrical DC power (P_PV, P_DC) of the PV generator (11) into electrical AC power (P_AC), wherein the PV generator (11) has a PV characteristic curve (21) and the inverter (12) is operated by means of a droop control curve (31), wherein the droop control curve (31) specifies setpoint values ​​for the power (P_DC, P_AC) of the inverter (12) as a function of a DC voltage (U_DC) of the inverter (12), wherein the inverter (12) adjusts an actual value of the power (P_DC, P_AC) to the setpoint specified by the droop control curve (31) for the current DC voltage (U_DC) by means of a power electronic bridge circuit, so that during operation of the PV generator (11) sets an operating point (41) which corresponds to the intersection of the current droop control curve (31) with the current PV characteristic curve (21),characterized in that iterative MPP tracking is performed by stepwise changing the droop control curve (31) to shift the operating point (41) along the DC voltage (U_DC), and the resulting actual value of the power (P_DC, P_AC) is recorded, wherein the droop control curve (31) is changed in each tracking step such that the operating point (41) is shifted in a direction that depends on the direction of a change in the actual value of the power (P_DC, P_AC) at the operating point (41) due to the change in the droop control curve (31) in a previous tracking step.

2. Method according to claim 1, wherein the droop control curve (31) has a foot point with a foot point voltage (U_P0) at which the setpoint for the power (P_DC, P_AC) is zero, and a slope, wherein the change of the control curve (31) is effected by a shift of the foot point voltage (U_P0).

3. Method according to claim 2, wherein the slope of the droop control curve (31) is constant or varies depending on the foot point voltage (U_P0).

4. Method according to any of the preceding claims, wherein the iterative MPP tracking is performed according to a 3-point method, wherein each tracking step comprises multiple measurements at at least two different droop control curves (31).

5. Method for operating a PV generator (11) at the maximum power point (MPP, 22) by means of at least two inverters (12) connected in parallel on the DC and AC sides and configured to convert electrical DC power (P_PV) of the PV generator into electrical AC power (P_AC), wherein the PV generator (11) has a PV- 22-368-P-WO - 24 - submitted version The characteristic curve (21) and the inverters (12) are operated by means of individual droop control curves (35), wherein the individual droop control curves (35) specify setpoints for the power (P_DC, P_AC) of the respective inverter (12) as a function of a DC voltage (U_DC) of the inverter (12), wherein the inverters (12) adjust their respective actual power value (P_DC, P_AC) to the setpoint specified by the individual droop control curve (35) for the current DC voltage (U_DC) by means of a power electronic bridge circuit, wherein a sum of the individual droop control curves (35) forms an overall droop control curve (36), so that an operating point (41) is established during operation of the PV generator (11) which corresponds to the intersection of the overall droop control curve (36) with the current PV characteristic curve (21), thereby characterized by the fact that iterative MPP tracking is performed,by a control unit (13) determining a total power (P_sys) of the inverters (12) and incrementally changing the total droop control curve (36) to shift the operating point (41) along the DC voltage (U_DC), wherein the total droop control curve (36) is modified in each tracking step such that the operating point (41) is shifted in a direction that depends on the direction of a change in the total power (P_sys) at the operating point due to the change in the total droop control curve (36) in a previous tracking step.

6. The method of claim 5, wherein the overall droop control curve (36) is modified in each step of the iterative MPP tracking by modifying the individual droop control curve (35) of at least one of the inverters (12) or of exactly one of the inverters (12).

7. Method according to claim 5 or 6, wherein the individual droop control curves (35) have a respective foot point at which the setpoint for the power (P_DC, P_AC) is zero, a foot point voltage (U_P0) and a slope.

8. Method according to claim 7, wherein the individual droop control curve (35) of an inverter (12) is modified by shifting the feedpoint voltage (U_P0) of this individual droop control curve (35) relative to a feedpoint reference voltage.

9. Method according to claim 8, wherein the displacement of the foot point voltage (U_P0) has a step size, wherein the individual droop control curves (35) have the same slope and their foot point voltages (U_P0) are at most one step size apart. 22-368-P-WO - 25 - submitted version 10. Method according to one of claims 7 to 9, wherein the footpoint voltages (U_PO) of the individual droop control curves (35) are different from each other in at least one of two successive MPP tracking steps.

11. Method according to claim 9, wherein the step size depends on the change in total power (P_sys) due to the change in the total droop control curve (36) in a previous step, wherein the step size is particularly larger the greater the change in total power (P_sys) was in the previous steps.

12. Method according to any one of claims 5 to 11, wherein the iterative MPP tracking is performed according to a 3-point method, wherein each step of the MPP tracking comprises multiple measurements of the total power (P_sys) at at least two different total droop control curves (36).

13. Method according to one of the preceding claims, wherein the individual droop control curve (31, 35) specifies DC setpoints, in particular a DC power (P_DC) or a DC current (l_DC) as a function of the DC voltage, wherein the DC setpoints are compared with corresponding DC setpoints (P_DC, l_DC) and converted into AC setpoints (P_setpoint, l_setpoint) by means of a controller, or wherein the individual droop control curve (35) specifies AC setpoints, in particular an AC power (P_AC) or an AC current (l_AC) as a function of the DC voltage (U_DC).

14. Inverter (12) for operating a PV generator (11), wherein the inverter (12) is configured to operate the PV generator (11) at the maximum power point (MPP, 22) using a method according to one of claims 1 to 4 and claim 13, insofar as related to claims 1 to 4.

15. Device for operating a PV generator (11) with at least two substantially identical inverters (12) connected in parallel on the DC and AC sides, and with a control unit (13), wherein the inverters (12) and the control unit (13) are configured to operate the PV generator (11) using a method according to one of claims 5 to 13.

16. Device according to claim 15, wherein the control unit (13) is configured to acquire the current power values ​​(P_AC) of each inverter (12) or the total power (P_sys) and to modify the individual droop control curve (35) of at least one of the inverters (12) in each step of the iterative MPP tracking as a function of the total power (P_sys) in previous tracking steps. 22-368-P-WO - 26 - submitted version 17. Device according to claim 15 or 16, wherein the control unit (13) is designed separately or is arranged in one of the inverters (12) and / or is connected via communication with the other substantially identical inverters (12).

18. Device according to one of claims 15 to 17, wherein the sum of the maximum powers (P_max) of the substantially identical inverters (12) results in a total rated power of the device which is at least 50% of a peak power of the PV generator and preferably greater than 70% of the peak power of the PV generator.

Images