A method and system for adaptive power decoupling droop control of multiple parallel inverters

Abstract

The application discloses a kind of adaptive power decoupling droop control method and system of multiple inverters parallel, including main controller, CAN bus, CAN-based multiple inverter module and adaptive power decoupling droop control module;Main controller is used to process global information;CAN bus realizes the data transmission between main controller and CAN-based multiple inverter module via CAN data line;CAN-based multiple inverter module is used to control power circuit;Adaptive power decoupling droop control strategy is by main controller, CAN-based multiple inverter module and CAN bus data interaction and through power decoupling PI regulator algorithm, generate the given value based on the frequency and amplitude of the ac output voltage of each inverter in CAN-based multiple inverter module, to form the new voltage reference quantity based on each inverter in CAN-based multiple inverter module.The application effectively suppresses the low-frequency circulation between parallel system through the pretreatment of virtual impedance and the deep processing of average power ring, and improves the stability of system operation.

Description

Technical Field

[0001] This invention relates to the field of power electronics and microgrid control technology, and in particular to an adaptive power decoupling droop control method and system for multiple inverters connected in parallel. Background Technology

[0002] With the rapid development of new energy power generation technology and new power system architecture, the demand for single-unit capacity of power electronic conversion devices is increasing day by day. However, due to the limitations of the voltage rating, heat dissipation threshold and physical structure size of power semiconductor devices, the power rating of a single inverter faces a bottleneck and it is difficult to independently meet the needs of high-power application scenarios. Therefore, using multiple inverters to operate in a modular parallel manner to expand the total system capacity has become the mainstream technical solution to improve power supply capacity, enhance system redundancy and reliability.

[0003] In inverter parallel control systems, droop control is widely used in microgrids and parallel inverter systems due to its ability to autonomously allocate power without the need for interconnection communication lines and relying on local information. Ideally, droop control is based on the assumption of inductive output impedance and utilizes the droop characteristics of active power-frequency and reactive power-voltage amplitude to theoretically achieve natural decoupling of active and reactive power control channels. That is, active power is controlled by adjusting the frequency, and reactive power is controlled by adjusting the voltage amplitude.

[0004] However, in practical engineering applications, the output impedance characteristics of multi-inverter parallel systems are often affected by multiple factors, including line impedance, filter parameters, and load characteristics. The line impedance angle is not an ideal 90-degree inductive angle but contains a significant resistive component. This leads to a strong cross-coupling effect between active power and voltage amplitude, and between reactive power and frequency, causing the droop control formula to be in an incompletely decoupled state during actual operation. This incomplete decoupling can cause power-coupled oscillations, leading to a decrease in the small-signal stability of the system. At the same time, due to incomplete decoupling, the voltage vectors between parallel units are difficult to synchronize precisely, generating significant fundamental circulating currents between units and causing uneven load distribution. Some units are prone to overload, reducing the system's lifespan.

[0005] With the increase in the number of parallel units and the expansion of system capacity, the control complexity and nonlinearity caused by incomplete decoupling become increasingly prominent. Existing conventional droop control strategies are no longer able to meet the requirements of high-precision and high-stability parallel operation. Therefore, how to improve the control strategy, overcome the coupling effects caused by line impedance and parameter differences, achieve complete decoupling of active and reactive power channels from the control principle, and ensure that the droop control formula achieves the ideal decoupling effect in physical implementation, thereby completely eliminating circulating current, achieving accurate current sharing, and ensuring strict synchronization of the amplitude and phase of the output voltage, is a key technical problem that urgently needs to be solved in the field of power electronics parallel technology. Summary of the Invention

[0006] To address the shortcomings of existing three-phase inverter parallel systems, such as severe droop control power coupling, uneven reactive power distribution, and large steady-state current sharing errors caused by line impedance mismatch, this invention provides an adaptive power decoupling droop control method and system for multiple inverters in parallel. It constructs an adaptive power decoupling droop control algorithm for multiple inverters in parallel, achieving deep decoupling of the active and reactive components of the inverters, thereby improving power sharing accuracy, effectively avoiding low-frequency oscillations caused by power coupling, and significantly enhancing the stability of the multi-inverter parallel system.

[0007] The objective of this invention is achieved in part by providing an adaptive power decoupling droop control method for multiple inverters in parallel, comprising the following steps:

[0008] 1) The main controller collects the active power and reactive power of each inverter in the parallel CAN-based multi-inverter module, calculates the average active power and average reactive power of all inverters in the CAN-based multi-inverter module in the main controller, and broadcasts them to each inverter in the CAN-based multi-inverter module via the CAN bus.

[0009] 2) The power error value obtained by subtracting the average active power and average reactive power received from the main controller by all inverters in the CAN multi-inverter module from the active power and reactive power of each inverter in the CAN multi-inverter module is used to generate active-frequency correction and reactive-voltage correction through the power decoupling PI regulator algorithm.

[0010] 3) The active-frequency correction and reactive-voltage correction are superimposed on the traditional droop control algorithm to obtain the AC output voltage amplitude and frequency setpoint of each inverter in the CAN multi-inverter module, thereby forming a new voltage reference.

[0011] Furthermore, step 2) specifically includes:

[0012] In each inverter based on the CAN multi-inverter module, the power decoupling PI regulator algorithm is used to obtain the active-frequency correction Δf by subtracting the received average active power from the active power of each inverter in the CAN multi-inverter module; and to obtain the reactive-voltage correction ΔE by subtracting the received average reactive power from the reactive power of each inverter in the CAN multi-inverter module, as shown in equation (1).

[0013]

[0014] In the formula, k p and k iThese are the proportional and integral parameters of the power decoupling PI regulator, respectively. P and Q represent the active and reactive power outputs of each inverter in the CAN multi-inverter module, respectively. and These are the average active power and average reactive power outputs of all inverters in the CAN multi-inverter module, respectively. Δf and ΔE are the active-frequency correction and reactive-voltage correction values ​​of each inverter in the CAN multi-inverter module, respectively.

[0015] Furthermore, step 3) specifically includes: algebraically superimposing the active-frequency correction and reactive-voltage correction with the active-frequency droop and reactive-voltage droop formulas in the traditional droop control, respectively, to form the algorithm of the adaptive power decoupling droop control strategy as shown in the following formula (2), so as to generate the frequency and amplitude given values ​​of the AC output voltage of each inverter in the CAN multi-inverter module, thereby forming a new voltage reference value of each inverter in the CAN multi-inverter module in the system;

[0016]

[0017] In the formula, f0 and E0 are the given values ​​of frequency and amplitude of AC output voltage of each inverter in the CAN multi-inverter module, respectively, and m and n are the active power-frequency droop coefficient and reactive power-voltage droop coefficient, respectively. and These are the reference power for the output active and reactive power of each inverter in the CAN multi-inverter module, respectively; f and E are the actual phase and actual voltage amplitude of the inverter output, respectively; P and Q are the active and reactive power output of each inverter in the CAN multi-inverter module, respectively; and Δf and ΔE are the active-frequency correction and reactive-voltage correction of each inverter in the CAN multi-inverter module, respectively.

[0018] Another aspect of the objective of this invention is achieved as follows: an adaptive power decoupling droop control system for multiple inverters in parallel, comprising a main controller, a CAN bus, a CAN-based multi-inverter module, and an adaptive power decoupling droop control module;

[0019] In a multi-inverter parallel system, the main controller is used to request the active and reactive power of all inverters in the CAN-based multi-inverter module, calculate the average active and average reactive power based on the active and reactive power uploaded by all inverters in the CAN-based multi-inverter module, and broadcast them to all inverters in the CAN-based multi-inverter module via the CAN bus.

[0020] The CAN bus is used to realize data transmission between the main controller and the CAN-based multi-inverter module via the CAN data line;

[0021] The CAN-based multi-inverter module is composed of n identical PWM inverter power supply modules connected in parallel, and is used to control the power circuit.

[0022] The adaptive power decoupling droop control module is used to generate given values ​​for the frequency and amplitude of the AC output voltage of each inverter in the CAN multi-inverter module through the power decoupling PI regulator algorithm, based on the data exchanged by the main controller, the CAN multi-inverter module and the CAN bus, thereby forming a new voltage reference value based on each inverter in the CAN multi-inverter module.

[0023] Furthermore, terminating resistors R are connected to both ends of the CAN bus. T It is used to eliminate signal reflection to ensure communication stability; the CAN bus is configured as a bidirectional interactive link, which not only transmits the control commands of the main controller, but also uploads the status data based on the CAN multi-inverter module, serving the adaptive power decoupling droop control module.

[0024] Furthermore, each of the CAN-based multi-inverter modules internally includes a DC bus, a PWM inverter circuit, a filter circuit, and an AC bus. The input terminal of the DC bus is connected to a DC source, and the output terminal is connected to the input terminal of the three-phase inverter circuit, providing DC energy to each inverter module. The PWM inverter circuit includes a power switch and a drive circuit, with its DC input terminal connected to the DC bus and its AC output terminal connected to the filter circuit. The input terminal of the filter circuit is connected to the output of the PWM inverter circuit, and its output terminal is connected to the AC bus. The AC bus is connected to the output terminal of the filter circuit of each PWM inverter circuit, collecting electrical energy and supplying power to the load.

[0025] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention proposes an adaptive power decoupling droop control strategy for multi-inverter parallel operation. Based on the traditional power droop control strategy, it constructs a parallel system of a main controller and CAN-based multi-inverter modules using an existing CAN bus. By introducing active-frequency correction and reactive-voltage correction quantities through average active control and average reactive control, respectively, it algebraically superimposes these values ​​with the active-frequency droop and reactive-voltage droop formulas in the traditional droop control, thereby constructing an adaptive power decoupling droop control algorithm for multi-inverter parallel operation. This achieves deep decoupling of the active and reactive components of the inverters, thereby improving power sharing accuracy, effectively avoiding low-frequency oscillations caused by power coupling, and significantly improving the stability of the multi-inverter parallel system.

[0026] This invention not only effectively suppresses circulating current in multi-inverter parallel systems, thus achieving current sharing from output to load, but also proposes a method for microgrid inverter parallel systems with existing CAN buses and a main controller handling global information such as data uploads to the cloud. This method incurs no additional cost; the control algorithm is implemented solely through program code, resulting in high cost-effectiveness. This invention facilitates the parallel operation of multiple inverters in photovoltaic energy storage microgrids within new power systems, contributing to the stable and reliable operation of novel energy storage inverter systems. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0028] Figure 1 This is the hardware architecture of the adaptive power decoupling droop control system for multiple inverters in parallel according to the present invention.

[0029] Figure 2 This is a schematic diagram of the adaptive power decoupling droop control strategy for multiple inverters in parallel according to the present invention.

[0030] Figure 3 This is the topology of the three-phase inverter of the present invention.

[0031] Figure 4 This is the architecture of an embodiment of the present invention.

[0032] Figure 5 The graph shows experimental data from a power analyzer using a combination of traditional droop control and virtual impedance control in an embodiment of the present invention.

[0033] Figure 6 The figure shows experimental data from a power analyzer employing adaptive power decoupling droop control and virtual impedance composite control in an embodiment of the present invention.

[0034] Figure 7 The waveform diagrams shown are from an oscilloscope experiment using a combination of traditional droop control and virtual impedance control in an embodiment of the present invention.

[0035] Figure 8 The waveform diagrams shown are from an oscilloscope experiment using adaptive power decoupling droop control and virtual impedance composite control in this embodiment of the invention.

[0036] Figure 1 Symbol names in:

[0037]

[0038] Figure 2 Symbol names in:

[0039]

[0040] Figure 3 Symbol names in:

[0041]

[0042] Figure 4 Symbol name:

[0043] Figure 5 and Figure 6 Symbol name:

[0044] Figure 7 and Figure 8 Symbol name:

[0045]

[0046] The system consists of 1 main controller, 2 CAN bus, 3 CAN-based multi-inverter module, 4 adaptive power decoupling droop control module, and 5 fuel tank. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] like Figure 1 The adaptive power decoupling droop control system for multiple inverters in parallel is shown, including a main controller 1, a CAN bus 2, a CAN-based multi-inverter module 3, and an adaptive power decoupling droop control module 4.

[0049] Main controller 1 is the only designated main controller 1 in a multi-inverter parallel system, used to process global information.

[0050] In a multi-inverter parallel system, the main controller 1 is used to request the active power and reactive power of all inverters in the CAN-based multi-inverter module 3, calculate the average active power and average reactive power based on the active power and reactive power uploaded by all inverters in the CAN-based multi-inverter module, and broadcast them to all inverters in the CAN multi-inverter module 3 using the CAN bus 2.

[0051] CAN bus 2 is used to realize data transmission between the main controller 1 and the CAN-based multi-inverter module 3 via the CAN data line; terminating resistors R are connected to both ends of CAN bus 2. T The CAN bus 2 is configured as a bidirectional interactive link to eliminate signal reflections and ensure communication stability. It transmits control commands from the main controller 1 and uploads status data based on the CAN multi-inverter module to serve the adaptive power decoupling droop control module.

[0052] The CAN-based multi-inverter module 3 consists of n identical PWM inverter power supply modules connected in parallel, used to control the power circuit. Each CAN-based multi-inverter module 3 internally includes a DC bus, a PWM inverter circuit, a filter circuit, and an AC bus. The input terminal of the DC bus is connected to a DC source, and the output terminal is connected to the input terminal of the three-phase inverter circuit, providing DC energy to each inverter module. The PWM inverter circuit includes power switching transistors and drive circuits, with its DC input terminal connected to the DC bus and its AC output terminal connected to the filter circuit. The input terminal of the filter circuit is connected to the output of the PWM inverter circuit, and its output terminal is connected to the AC bus. The AC bus is connected to the output terminal of the filter circuit of each PWM inverter circuit, collecting electrical energy and supplying power to the load.

[0053] The CAN multi-inverter module 2 in a multi-inverter parallel system mainly undertakes the following two tasks:

[0054] (1) Local voltage and current dual closed-loop control, traditional droop control and virtual impedance strategy are adopted;

[0055] (2) After receiving the active power and reactive power requests from the main controller 1, the active power and reactive power of each inverter in the CAN multi-inverter module 2 are uploaded to the main controller 1. After receiving the average active power and average reactive power of the main controller 1, the adaptive power decoupling droop control module 4 is performed.

[0056] The adaptive power decoupling droop control module 4 is used to generate the frequency and amplitude values ​​of the AC output voltage of each inverter in the CAN multi-inverter module through the power decoupling PI regulator algorithm, based on the data exchanged by the main controller 1, the CAN multi-inverter module 3 and the CAN bus 2, thereby forming a new voltage reference quantity for each inverter in the CAN multi-inverter module.

[0057] like Figure 2 As shown, an adaptive power decoupling droop control method for multiple inverters in parallel includes the following steps:

[0058] When the main controller 1 and the CAN-based multi-inverter module 3 meet the conditions for parallel connection of each inverter in the CAN-based multi-inverter module 3, the combined control of the adaptive power decoupling droop control module 4 and virtual impedance control is activated when the parallel relay is closed. At this time, the main controller 1 requests active and reactive power from all inverters in the CAN-based multi-inverter module 3 at the end of each grid cycle, i.e., every 20ms. After receiving the active and reactive power request signals from the main controller 1, each inverter in the CAN-based multi-inverter module 3 calculates the instantaneous active and reactive power according to its own output voltage and output current and uploads them to the main controller 1.

[0059] After receiving the active power and reactive power from all inverters in the CAN-based multi-inverter module 3, the main controller 1 calculates the average active power and average reactive power in the parallel system, and uses the broadcast mode of CAN bus 2 to send the average active power and average reactive power to each inverter in the CAN-based multi-inverter module 3.

[0060] After each inverter in the CAN multi-inverter module 3 receives the average active power and average reactive power calculated by the main controller 1, it compares the difference between the local active power and the average active power to form an active power error, which is then sent to the power decoupling PI regulator to output the active power-frequency correction amount as shown in equation (3). The active power-frequency correction amount is then superimposed on the output frequency of the traditional droop control to form a new frequency setting as shown in equation (4). The local reactive power is compared with the average reactive power to form a reactive power error, which is then sent to the power decoupling PI regulator to output the reactive power-voltage correction amount as shown in equation (3). The reactive power-voltage correction amount is then superimposed on the output voltage amplitude of the traditional droop control to form a new voltage amplitude setting as shown in equation (4), thus forming a new voltage reference amount. The new voltage reference amount is transformed by abc-dq to obtain the d-axis and q-axis voltage settings. The d-axis and q-axis voltage settings are then processed by the virtual impedance control algorithm to obtain the new d-axis and q-axis voltage settings, which are the voltage and current double closed-loop input settings. After the adaptive power decoupling droop control module 4, the active power and reactive power output by the parallel system are further decoupled, significantly improving the stability of the parallel system.

[0061] (3)

[0062] In the formula, k p and k i These are the proportional and integral parameters of the power decoupling PI regulator, respectively. P and Q represent the active and reactive power outputs of each inverter in the CAN multi-inverter module, respectively. and These are the average active power and average reactive power outputs of all inverters in the CAN multi-inverter module, respectively. Δf and ΔE are the active-frequency correction and reactive-voltage correction values ​​of each inverter in the CAN multi-inverter module, respectively.

[0063] (4)

[0064] In the formula, f0 and E0 are the given values ​​of frequency and amplitude of AC output voltage of each inverter in the CAN multi-inverter module, respectively, and m and n are the active power-frequency droop coefficient and reactive power-voltage droop coefficient, respectively. and These are the reference power for the output active and reactive power of each inverter in the CAN multi-inverter module, respectively; f and E are the actual phase and actual voltage amplitude of the inverter output, respectively; P and Q are the active and reactive power output of each inverter in the CAN multi-inverter module, respectively; and Δf and ΔE are the active-frequency correction and reactive-voltage correction of each inverter in the CAN multi-inverter module, respectively.

[0065] The core physical logic of traditional droop control lies in simulating the external characteristics of a synchronous generator, enabling each inverter unit in a parallel system to autonomously adjust its output power according to load fluctuations. Based on the CAN multi-inverter module 2, the current real-time active and reactive power are calculated using instantaneous power theory or coordinate transformation by sampling the voltage and current signals at the output terminals of each inverter. In an ideal inductive impedance environment, the power transmission characteristics show that active power is mainly controlled by voltage frequency, while reactive power is mainly affected by voltage amplitude. At the execution level, the traditional droop control strategy first relies on real-time high-frequency sampling of the voltage and current signals at the inverter output terminals, using instantaneous power theory or coordinate transformation to calculate the current real-time active and reactive power. Since, in an ideal inductive impedance environment, the power transmission characteristics show that active power is mainly controlled by voltage phase angle (i.e., frequency), while reactive power is mainly affected by voltage amplitude, this strategy establishes a linear negative feedback relationship between power and output frequency and voltage through a preset droop coefficient. This characteristic simulates the speed drop process of a synchronous generator, ensuring that multiple parallel inverters can automatically find a new power balance point as the frequency decreases, thereby achieving proportional distribution of active load. Simultaneously, the reactive-voltage characteristic exhibits a negatively correlated linear adjustment relationship between the output voltage amplitude and reactive power. When the system's reactive power demand increases, the inverters reduce the output voltage amplitude to share the load, maintaining relative stability of the parallel node voltage. In the control command generation stage, the angular frequency calculated from the droop equation needs to be converted into an angle reference value through an integrator. Combined with the calculated voltage amplitude, a three-phase reference voltage command is generated using sine wave synthesis technology, serving as the given signal for the subsequent virtual impedance control algorithm.

[0066] The virtual impedance technology introduced in this invention for the control system essentially simulates an equivalent complex impedance in the feedback loop of the inverter output voltage through a control algorithm. This alters the inverter's output characteristics without generating actual power loss or heat. In traditional droop control applications, power decoupling heavily relies on the system impedance exhibiting pure inductive characteristics. However, in low-voltage microgrid parallel systems, due to short cable lengths and high internal resistance, the physical line impedance often exhibits significant resistive-inductive coupling characteristics. To correct this physical defect, a virtual complex impedance Z=sL-R is introduced. This appropriately reduces the resistive component in the output impedance while retaining some resistivity to provide system damping capability, and increases the inductive component in the output impedance to ensure decoupling of active and reactive power. The specific implementation process of virtual impedance technology is as follows: The inverter's output current is collected and multiplied by the introduced virtual impedance value to obtain the voltage drop generated at the virtual impedance. This voltage drop is then subtracted from the given voltage value to generate a reference value for the inverter's output voltage. The introduced virtual impedance operates in series with the inverter's equivalent output impedance, thus reshaping the inverter's actual output impedance. By adding virtual impedance to the control loop, not only can the inverter's equivalent transmission impedance be adjusted to primarily exhibit inductive characteristics, but power decoupling is further achieved. Moreover, this control method does not introduce additional electrical components; it is equivalent to introducing a virtual impedance in series with the external circuit. It allows for flexible adjustment of the inverter's transmission impedance characteristics without increasing external circuit voltage and power losses. It is easy to implement in practical engineering, reducing system costs.

[0067] like Figure 3 The diagram shows the topology of one of the inverters based on CAN multi-inverter module 3. Its main circuit adopts a three-phase T-type three-level four-wire inverter circuit, consisting of three completely symmetrical bridge arms, each corresponding to one phase of the inverter (phase A, phase B, and phase C), and four power switching transistors Q arranged in a T-shape. _x1 ~Q _x4 (x=u,v,w) and its parasitic diode D k1 ~D k4 (k=u,v,w) is composed of two DC capacitors C connected in series. dc1 and C dc2 Composition, forming the DC bus voltage V bus The midpoint of its voltage is denoted as point O; L s and C su C sv C sw These are the AC filter inductor and filter capacitor, respectively, where R... s For filter inductor L s parasitic resistance; usa , u sb ,u sc These are the AC outputs for phases A, B, and C, respectively, with the midpoint of their voltages denoted as point N.

[0068] A specific embodiment of the present invention is as follows:

[0069] like Figure 4 As shown, this invention is applied to a parallel scenario of three-phase T-type three-level four-wire inverters. In terms of specific hardware implementation, all inverters in the CAN-based multi-inverter module 3 use the TMS320F280025 DSP chip, while the main controller uses the TMS320F280039 DSP chip. Both DSP chips have built-in CAN hardware interfaces and are fully supported by TI's official C2000Ware driver. The multi-inverter parallel system communicates via CAN bus 2, with a baud rate set to 500kbps. Regarding communication mailbox configuration, both the main controller 1 and the CAN-based multi-inverter module 3 are configured with a transmitting mailbox 5 and a receiving mailbox 1. Specifically, the main controller 1 sends information through mailbox 5 and receives information from the CAN-based multi-inverter module 3 through mailbox 1; similarly, the CAN-based multi-inverter module 3 sends information between itself and the main controller 1 through mailbox 5 and receives information between itself and the main controller 1 through mailbox 1. The main controller 1 and the CAN-based multi-inverter module 3 communicate via mailbox 5 as the transmitting mailbox and mailbox 1 as the receiving mailbox, with the receive interrupt set to high priority mode. Mailbox 1 of the CAN-based multi-inverter module 3 supports receiving data via interrupt or interrupt flag polling. The main controller 1 uses mailbox 5 to send active and reactive power request signals, and the CAN-based multi-inverter module 3 processes the data according to the requests and requirements of the main controller 1. In terms of control strategy and function allocation, the main controller 1 mainly implements the average calculation and processing of active and reactive power. The CAN-based multi-inverter module 3 implements local voltage and current dual closed-loop control, traditional droop control, and virtual impedance control. The adaptive power decoupling droop control strategy 4 is implemented by the cooperation of the main controller 1, CAN bus 2, and CAN-based multi-inverter module 3.

[0070] The formulas for the active power-frequency correction Δf and reactive power-voltage correction ΔE generated by the adaptive power decoupling droop control module 4 in the multi-inverter parallel system are as follows:

[0071] (5)

[0072] In the formula, k p and k iThese are the proportional and integral parameters of the power decoupling PI regulator, respectively. P and Q represent the active and reactive power outputs of each inverter in the CAN multi-inverter module, respectively. and These are the average active power and average reactive power outputs of all inverters in the CAN multi-inverter module, respectively. Δf and ΔE are the active-frequency correction and reactive-voltage correction values ​​of each inverter in the CAN multi-inverter module, respectively.

[0073] The frequency and amplitude setpoints of the AC output voltage of each inverter in the CAN multi-inverter module are generated, thus forming a new voltage reference value for each inverter in the system based on the CAN multi-inverter module, as shown in the following formula:

[0074] (6)

[0075] In the formula, f0 and E0 are the given values ​​of frequency and amplitude of AC output voltage of each inverter in the CAN multi-inverter module, respectively, and m and n are the active power-frequency droop coefficient and reactive power-voltage droop coefficient, respectively. and These are the reference power for the output active and reactive power of each inverter in the CAN multi-inverter module, respectively; f and E are the actual phase and actual voltage amplitude of the inverter output, respectively; P and Q are the active and reactive power output of each inverter in the CAN multi-inverter module, respectively; and Δf and ΔE are the active-frequency correction and reactive-voltage correction of each inverter in the CAN multi-inverter module, respectively.

[0076] In this example, the AC output voltage frequency setpoint f0 is 50Hz, the AC output voltage setpoint E0 is 325V, the active power-frequency droop factor m is 0.2125, the reactive power-voltage droop factor n is 0.000207417989148, the resistive component R in the virtual impedance is 0.04749717741575, the inductive component L in the virtual impedance is 0.003111685425, and the proportional parameter k of the power decoupling PI regulator is... p The integral parameter k of the power decoupling PI regulator is 0.00000015. i It is 0.00001.

[0077] To further verify the effectiveness and superiority of the adaptive power decoupling droop control strategy proposed in this invention under actual operating conditions, this embodiment constructs a parallel experimental platform consisting of four three-phase T-type three-level inverters for comparative testing. In the experiment, the rated output voltage RMS value of all four inverters was uniformly set to 230V, and the system load was configured as an inductive load with an apparent power of 40kVar and a power factor of 0.8, thereby simulating a complex microgrid operating environment.

[0078] First, experiments were conducted using a composite control mode that combined traditional droop control with fixed virtual impedance. Figure 5 The data measured by the power analyzer are shown in Table 1. Figure 7 The image shows the current waveforms of four inverters on an oscilloscope. Due to the oscilloscope's insufficient accuracy, its current sampling has errors. Therefore, the current imbalance data should be based on the power analyzer data, with the oscilloscope data used as a reference.

[0079] Table 1. Experimental data of the combined control of traditional droop control and fixed virtual impedance measured by power analyzer.

[0080]

[0081] Table 2. Experimental data of the combined control of traditional droop control and fixed virtual impedance measured by oscilloscope.

[0082]

[0083] As can be seen from Tables 1 and 2, despite the introduction of a fixed virtual impedance, there is still a significant circulating current phenomenon within the system due to the asymmetry of the physical impedance of the line and the residual power coupling effect, which makes it impossible to meet the requirements of high-reliability power supply in terms of power distribution accuracy.

[0084] Subsequently, while maintaining the same hardware environment and load conditions, the adaptive power decoupling droop control strategy 4 proposed in this invention was switched to for verification. Figure 6 The data measured by the power analyzer are shown in Table 2. Figure 8 The image shows the current waveforms of four inverters on an oscilloscope. Due to the oscilloscope's insufficient accuracy, its current sampling has errors. Therefore, the current imbalance data should be based on the power analyzer data, with the oscilloscope data used as a reference.

[0085] Table 3 shows the experimental data of the combined control strategy of adaptive power decoupling droop control and fixed virtual impedance measured by the power analyzer.

[0086]

[0087] Table 4 shows the experimental data of the combined control strategy of adaptive power decoupling droop control and fixed virtual impedance measured by oscilloscope.

[0088]

[0089] Comparing Tables 1 and 3, and Tables 2 and 4, it is evident that after adopting the adaptive power decoupling droop control strategy 4, the active power imbalance of the four inverters, measured by the power analyzer, decreased significantly from 4.5662% to 1.6911%, the reactive power imbalance from 6.2937% to 2.8942%, and the current imbalance from 5.0682% to 2.1047%. The current imbalance of the four inverters, measured by the oscilloscope, decreased significantly from 4.9188% to 2.8858%. Experiments show that the adaptive power decoupling droop control strategy 4 achieves deep decoupling of the active and reactive components of the inverters, thereby improving power sharing accuracy, effectively avoiding low-frequency oscillations caused by power coupling, significantly improving the stability of the multi-inverter parallel system, and significantly suppressing current deviations between modules. This proves that the control strategy of this invention can effectively eliminate the distribution deviation caused by sensor gain error and impedance mismatch, achieving truly high-precision current sharing.

[0090] In summary, the method of this invention provides current sharing performance far superior to traditional control strategies, and has extremely high engineering application value. The strategy described in this invention exhibits the following significant advantages:

[0091] (1) The method of the present invention achieves extremely high equalization accuracy of the output power of the three-phase T-type three-level three-phase four-wire inverter and high stability of the parallel system;

[0092] (2) The method of the present invention, combined with the control of virtual impedance, realizes the equal distribution of output current of four parallel inverter systems, and obtains extremely low output current imbalance, which is helpful for the construction of microgrids of new power systems.

[0093] (3) The method of the present invention has achieved a significant circulating current suppression effect, which helps to realize the modularization of high-power new photovoltaic energy storage inverters, and can effectively avoid the risk of overload operation of a single module, thereby improving the safety and robustness of the system;

[0094] (4) The method of the present invention is not limited to the parallel connection of three-phase inverters, but can also be extended to the parallel system of single-phase inverters and solid-state transformers in new power systems, thereby helping the stable operation of power distribution in the distribution area.

[0095] The above description of the embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. It should be noted that those skilled in the art can make several improvements and modifications to the present invention without departing from the principles of the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

Claims

1. An adaptive power decoupling droop control method for multiple inverters in parallel, characterized in that, Includes the following steps: 1) The main controller collects the active power and reactive power of each inverter in the parallel CAN-based multi-inverter module, calculates the average active power and average reactive power of all inverters in the CAN-based multi-inverter module in the main controller, and broadcasts them to each inverter in the CAN-based multi-inverter module via the CAN bus. 2) The power error value obtained by subtracting the average active power and average reactive power received from the main controller by all inverters in the CAN multi-inverter module from the active power and reactive power of each inverter in the CAN multi-inverter module is used to generate active-frequency correction and reactive-voltage correction through the power decoupling PI regulator algorithm. 3) The active-frequency correction and reactive-voltage correction are superimposed on the traditional droop control algorithm to obtain the AC output voltage amplitude and frequency setpoint of each inverter in the CAN multi-inverter module, thereby forming a new voltage reference.

2. The adaptive power decoupling droop control method for multiple inverters in parallel as described in claim 1, characterized in that, Step 2) specifically includes: In each inverter based on the CAN multi-inverter module, the power decoupling PI regulator algorithm is used to obtain the active-frequency correction Δf by subtracting the received average active power from the active power of each inverter in the CAN multi-inverter module; and to obtain the reactive-voltage correction ΔE by subtracting the received average reactive power from the reactive power of each inverter in the CAN multi-inverter module, as shown in equation (1). ； In the formula, k p and k i These are the proportional and integral parameters of the power decoupling PI regulator, respectively. P and Q represent the active and reactive power outputs of each inverter in the CAN multi-inverter module, respectively. and These are the average active power and average reactive power outputs of all inverters in the CAN multi-inverter module, respectively. Δf and ΔE are the active-frequency correction and reactive-voltage correction values ​​of each inverter in the CAN multi-inverter module, respectively.

3. The adaptive power decoupling droop control method for multiple inverters in parallel according to claim 2, characterized in that, Step 3) specifically includes: algebraically superimposing the active-frequency correction and reactive-voltage correction with the active-frequency droop and reactive-voltage droop formulas in the traditional droop control, respectively, to form the algorithm of the adaptive power decoupling droop control strategy as shown in the following formula (2), so as to generate the frequency and amplitude given values ​​of the AC output voltage of each inverter in the CAN multi-inverter module, thereby forming a new voltage reference value of each inverter in the CAN multi-inverter module in the system; ； In the formula, f0 and E0 are the given values ​​of frequency and amplitude of AC output voltage of each inverter in the CAN multi-inverter module, respectively, and m and n are the active power-frequency droop coefficient and reactive power-voltage droop coefficient, respectively. and These are the reference power for the output active and reactive power of each inverter in the CAN multi-inverter module, respectively; f and E are the actual phase and actual voltage amplitude of the inverter output, respectively; P and Q are the active and reactive power output of each inverter in the CAN multi-inverter module, respectively; and Δf and ΔE are the active-frequency correction and reactive-voltage correction of each inverter in the CAN multi-inverter module, respectively.

4. An adaptive power decoupling droop control system for multiple inverters in parallel, characterized in that, Includes main controller, CAN bus, CAN-based multi-inverter module, and adaptive power decoupling droop control module; In a multi-inverter parallel system, the main controller is used to request the active and reactive power of all inverters in the CAN-based multi-inverter module, calculate the average active and average reactive power based on the active and reactive power uploaded by all inverters in the CAN-based multi-inverter module, and broadcast them to all inverters in the CAN-based multi-inverter module via the CAN bus. The CAN bus is used to realize data transmission between the main controller and the CAN-based multi-inverter module via the CAN data line; The CAN-based multi-inverter module is composed of n identical PWM inverter power supply modules connected in parallel, and is used to control the power circuit. The adaptive power decoupling droop control module is used to generate given values ​​for the frequency and amplitude of the AC output voltage of each inverter in the CAN multi-inverter module through the power decoupling PI regulator algorithm, based on the data exchanged by the main controller, the CAN multi-inverter module and the CAN bus, thereby forming a new voltage reference value based on each inverter in the CAN multi-inverter module.

5. The adaptive power decoupling droop control system for multiple inverters in parallel according to claim 4, characterized in that, The two ends of the CAN bus are respectively connected to terminating resistors R. T It is used to eliminate signal reflection to ensure communication stability; the CAN bus is configured as a bidirectional interactive link, which not only transmits the control commands of the main controller, but also uploads the status data based on the CAN multi-inverter module, serving the adaptive power decoupling droop control module.

6. The adaptive power decoupling droop control system for multiple inverters in parallel according to claim 4, characterized in that, Each CAN-based multi-inverter module internally includes a DC bus, a PWM inverter circuit, a filter circuit, and an AC bus. The DC bus input is connected to a DC source, and the output is connected to the input of a three-phase inverter circuit, providing DC energy to each inverter module. The PWM inverter circuit includes power switching transistors and drive circuits, with its DC input connected to the DC bus and its AC output connected to the filter circuit. The filter circuit input is connected to the output of the PWM inverter circuit, and its output is connected to the AC bus. The AC bus connects to the output of the filter circuit of each PWM inverter circuit, collecting electrical energy and supplying it to the load.

Images