Micro-inverter output power control method and system

Abstract

The present application relates to the technical field of inverter output power control, and particularly relates to a micro inverter output power control method and system, the method comprising the following steps: acquiring a grid-connected point alternating voltage signal; based on the grid-connected point alternating voltage signal, performing in parallel: first phase-locked loop processing, extracting first phase information, the first phase information containing a voltage phase affected by an interference component with a frequency higher than a fundamental frequency; second phase-locked loop processing, extracting second phase information, the second phase information containing a fundamental phase after filtering out the interference component; calculating a phase disturbance amount according to the first phase information and the second phase information; generating a correction phase according to the second phase information and the phase disturbance amount; generating a corrected current command based on the correction phase; and controlling a micro inverter output alternating current according to the corrected current command. The above can improve power quality.

Description

Technical Field

[0001] This invention relates to the technical field of inverter output power control, and specifically to a method and system for controlling the output power of a micro inverter. Background Technology

[0002] In distributed photovoltaic power generation systems, microinverters serve as key grid-connected units, and their main responsibility is to efficiently convert the DC power generated by photovoltaic modules into AC power that meets the grid requirements.

[0003] Traditional microinverters typically employ phase-locked loop (PLL) technology to accurately obtain the phase information of the grid voltage in order to achieve synchronous grid connection. However, existing PLL designs often focus primarily on robustness to common low-order harmonics in the grid (such as the third and fifth harmonics), while failing to adequately consider the impact of voltage disturbances. Summary of the Invention

[0004] The purpose of this invention is to address the aforementioned shortcomings by proposing a method and system for controlling the output power of a micro inverter.

[0005] The present invention adopts the following technical solution:

[0006] A method for controlling the output power of a micro inverter, the method comprising the following steps:

[0007] Obtain the AC voltage signal at the grid connection point;

[0008] Based on the AC voltage signal at the grid connection point, the following parallel processes are performed: First phase-locked loop processing is performed to extract first phase information, which includes the voltage phase affected by interference components with frequencies higher than the fundamental frequency; Second phase-locked loop processing is performed to extract second phase information, which includes the fundamental phase after filtering out interference components.

[0009] Calculate the phase disturbance based on the first phase information and the second phase information;

[0010] Based on the second phase information and the phase disturbance, a corrected phase is generated;

[0011] A corrected current command is generated based on the corrected phase.

[0012] The micro inverter is controlled to output AC current according to the correction current command.

[0013] The above solution can effectively identify and compensate for phase oscillations caused by high-frequency interference, thereby suppressing sideband harmonics in the output current of the micro-inverter and improving power quality.

[0014] Optionally, this application also proposes that the steps for calculating the phase perturbation include:

[0015] Obtain the phase difference between the first phase information and the second phase information;

[0016] Spectral analysis is performed on the phase difference between the first phase information and the second phase information to identify multiple perturbation frequency components contained in the phase difference between the first phase information and the second phase information.

[0017] Based on multiple perturbation frequency components, determine multiple original perturbation frequencies that cause the phase difference between the first phase information and the second phase information;

[0018] Based on multiple original disturbance frequencies, a composite phase correction command is generated;

[0019] The composite phase correction command is used as the phase perturbation quantity.

[0020] The above scheme defines in detail the calculation method of phase disturbance and accurately identifies the disturbance frequency components through spectrum analysis, providing a foundation for subsequent accurate compensation.

[0021] Optionally, this application also proposes a step of performing spectral analysis on the phase difference between the first phase information and the second phase information to identify multiple perturbation frequency components contained in the phase difference between the first phase information and the second phase information, including:

[0022] Perform continuous short-time spectrum analysis on the phase difference between the first phase information and the second phase information;

[0023] Based on the continuous short-time spectrum analysis results, the instantaneous disturbance frequency in the phase difference between the first phase information and the second phase information is extracted, and the instantaneous disturbance frequency is used as multiple disturbance frequency components.

[0024] By introducing continuous short-time spectrum analysis through the above scheme, the instantaneous disturbance frequency in the phase difference can be captured in real time and dynamically, which improves the real-time performance and accuracy of disturbance identification.

[0025] Optionally, this application also proposes that the steps for performing continuous short-time spectrum analysis include:

[0026] The spectral characteristics of the signal that monitors the phase difference between the first phase information and the second phase information;

[0027] Adjust the window length and / or window overlap rate of continuous short-time spectrum analysis based on spectral characteristics;

[0028] Based on the adjusted window length and / or window overlap rate, perform continuous short-time spectral analysis on the phase difference between the first phase information and the second phase information.

[0029] The above scheme, by adaptively adjusting the spectrum analysis parameters, further optimizes the identification accuracy and robustness of instantaneous disturbance frequencies, adapting to the complex and ever-changing power grid environment.

[0030] Optionally, this application also proposes a step of extracting the instantaneous disturbance frequency from the phase difference between the first phase information and the second phase information based on continuous short-time spectrum analysis results, and using the instantaneous disturbance frequency as multiple disturbance frequency components, including:

[0031] Based on continuous short-time spectrum analysis results, identify the energy peak in the spectrum of the phase difference between the first phase information and the second phase information;

[0032] Analyze the spectral characteristics of the energy peak, which include peak width and / or peak shape;

[0033] Based on the spectral characteristics of the energy peak, determine whether the spectral characteristics of the energy peak are formed by the superposition of multiple close instantaneous disturbance frequencies;

[0034] If it is determined that the disturbance is formed by the superposition of multiple close instantaneous disturbance frequencies, separate the multiple close instantaneous disturbance frequencies;

[0035] The multiple instantaneous disturbance frequencies after separation, as well as the instantaneous disturbance frequencies corresponding to the energy peaks that are not determined to be superimposed, are taken as multiple disturbance frequency components contained in the phase difference between the first phase information and the second phase information.

[0036] The above scheme provides a method for identifying and separating superimposed disturbance frequencies, ensuring that each individual disturbance source can be accurately identified even when multiple close frequencies exist simultaneously.

[0037] Optionally, this application also proposes a step for separating multiple close instantaneous disturbance frequencies, including:

[0038] Perform curve fitting on the energy peak;

[0039] Based on the fitting results, the energy peak is decomposed into multiple independent sub-peaks;

[0040] The center frequency corresponding to each sub-peak is extracted as multiple close instantaneous perturbation frequencies.

[0041] The above scheme, using curve fitting technology, achieves accurate decomposition of superimposed energy peaks, further improving the precision of perturbation frequency identification.

[0042] Optionally, this application also proposes a step for generating a composite phase correction command based on multiple original perturbation frequencies, including:

[0043] Based on multiple original disturbance frequencies, corresponding phase correction components are generated;

[0044] The phase correction components are superimposed to generate a composite phase correction command.

[0045] The above scheme clarifies the generation method of composite phase correction commands, and achieves comprehensive compensation for multiple disturbance frequencies by superimposing multiple correction components.

[0046] Optionally, this application also proposes that the step of generating the corresponding phase correction component includes:

[0047] For multiple original disturbance frequencies, the instantaneous amplitude and instantaneous phase of the component corresponding to the original disturbance frequency in the phase difference between the first phase information and the second phase information are tracked in real time.

[0048] Based on the instantaneous amplitude and instantaneous phase, a compensation signal with the same frequency, the same instantaneous amplitude, and opposite phase as the component corresponding to the original disturbance frequency is generated as the phase correction component.

[0049] The above scheme provides a method for generating accurate phase correction components. By tracking the amplitude and phase of the disturbance components in real time, the accuracy and effectiveness of the compensation signal are ensured.

[0050] Optionally, this application also proposes a step of real-time tracking of the instantaneous amplitude and instantaneous phase of the component corresponding to the original disturbance frequency in the phase difference between the first phase information and the second phase information, including:

[0051] For each of the multiple original disturbance frequencies, an independent phase-locked loop is constructed.

[0052] The instantaneous amplitude and instantaneous phase of the component corresponding to the original disturbance frequency in the phase difference between the first phase information and the second phase information are tracked in real time using a phase-locked loop.

[0053] The above scheme uses an independent phase-locked loop to track each disturbance frequency, which improves the parallelism and accuracy of multi-frequency disturbance tracking.

[0054] Optionally, this application also proposes a micro-inverter output power control system applied to the aforementioned micro-inverter output power control method, the system comprising:

[0055] The acquisition module is used to acquire the AC voltage signal at the grid connection point;

[0056] The first phase-locked loop module extracts the first phase information based on the AC voltage signal at the grid connection point. The first phase information includes the voltage phase affected by interference components with frequencies higher than the fundamental frequency.

[0057] The second phase-locked loop module extracts the second phase information based on the AC voltage signal at the grid connection point. The second phase information includes the fundamental phase after filtering out interference components.

[0058] The calculation module is used to calculate the phase disturbance based on the first phase information and the second phase information;

[0059] The correction module is used to generate a corrected phase based on the second phase information and the phase perturbation amount;

[0060] The correction instruction generation module is used to generate correction current instructions based on the correction phase;

[0061] The current control module is used to control the output AC current of the micro inverter according to the correction current command.

[0062] The above solution, through its modular design, enables the method to be practically deployed and applied, demonstrating good feasibility.

[0063] As can be seen from the above, the micro-inverter output power control method and system provided in this application, by parallel processing of the AC voltage signal at the grid connection point, extracting the first phase information containing interference and the second phase information filtering out interference, and calculating the phase disturbance based on the two, thereby generating correction phase and correction current commands to control the output AC current of the micro-inverter, can effectively suppress the interference of high-frequency voltage disturbances of the grid on the phase-locked loop of the micro-inverter, avoid the resulting output current sideband harmonics, thereby improving the grid-connected power quality, and is especially suitable for the advantages of multiple micro-inverters operating in parallel.

[0064] To further understand the features and technical content of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are for reference and illustration only and are not intended to limit the present invention. Attached Figure Description

[0065] Figure 1 This is a flowchart of a method for controlling the output power of a micro inverter according to the present invention;

[0066] Figure 2 This is a schematic diagram of the output power control system for a micro inverter according to the present invention. Detailed Implementation

[0067] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can understand the advantages and effects of the present invention from the content disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the spirit of the present invention. Furthermore, the accompanying drawings of the present invention are for simple illustrative purposes only and are not depictions of actual dimensions; this is stated in advance. The following embodiments will further describe the relevant technical content of the present invention in detail, but the disclosed content is not intended to limit the scope of protection of the present invention.

[0068] This embodiment provides a method and system for controlling the output power of a micro inverter, combined with Figure 1 and Figure 2 As shown.

[0069] refer to Figure 1 A method for controlling the output power of a micro-inverter includes the following steps: acquiring the AC voltage signal at the grid connection point; based on the AC voltage signal at the grid connection point, performing in parallel the following: a first phase-locked loop (PLL) process to extract first phase information, the first phase information including the voltage phase affected by interference components with frequencies higher than the fundamental frequency; a second PLL process to extract second phase information, the second phase information including the fundamental phase after filtering out interference components; calculating the phase disturbance based on the first phase information and the second phase information; generating a correction phase based on the second phase information and the phase disturbance; generating a correction current command based on the correction phase; and controlling the output AC current of the micro-inverter according to the correction current command.

[0070] This application aims to address the problem of improving the grid-connected power quality of distributed photovoltaic systems composed of a large number of micro-inverters, where high-frequency voltage noise in the local power grid causes interference to the internal phase-locked loop of the micro-inverter, resulting in specific sideband harmonic currents generated through phase modulation mechanisms.

[0071] To achieve the above objectives, this application proposes a method for controlling the output power of a micro-inverter. This method first acquires the AC voltage signal at the grid connection point and then performs a first phase-locked loop (PLL) process and a second PLL process in parallel based on this signal. The first PLL process is a procedure for extracting phase information from the AC voltage signal at the grid connection point. Its characteristic is that its design or configuration enables it to capture the voltage phase containing interference components with frequencies higher than the fundamental frequency, i.e., the first phase information. This process can be implemented using a wideband PLL or a PLL with weak low-pass filtering characteristics. Its purpose is to obtain the raw phase information of the grid voltage, including superimposed high-frequency disturbance components, providing raw data for subsequent interference analysis. Simultaneously, the second PLL process is another procedure for extracting phase information from the AC voltage signal at the grid connection point. Its characteristic is that its design or configuration enables it to effectively filter out interference components with frequencies higher than the fundamental frequency, thereby extracting the pure fundamental phase, i.e., the second phase information. This process can be implemented using a narrowband phase-locked loop, a phase-locked loop with strong low-pass filtering characteristics, or a phase-locked loop with an integrated notch filter. The purpose is to provide an accurate grid fundamental phase reference that is not affected by high-frequency interference, as a reference for current control.

[0072] After acquiring the first and second phase information, the method calculates the phase disturbance based on them. The phase disturbance refers to the phase deviation or difference between the first and second phase information. This quantifies the instantaneous impact of high-frequency interference components on the grid voltage phase. The phase disturbance can be obtained by directly calculating the phase difference between the first and second phase information, aiming to accurately identify and quantify the phase distortion caused by high-frequency interference, providing a basis for subsequent phase correction. Subsequently, a corrected phase is generated based on the second phase information and the phase disturbance. The corrected phase refers to the phase information obtained by correcting the second phase information by incorporating the phase disturbance. This phase information reflects the accurate phase of the grid fundamental wave while eliminating the influence of high-frequency interference components on the phase. The corrected phase can be generated by performing appropriate addition and subtraction operations on the second phase information and the phase disturbance, aiming to provide a clean, accurate phase reference unaffected by high-frequency interference, used to generate the output current command of the micro-inverter. Finally, a corrected current command is generated based on the corrected phase, and the micro-inverter output AC current is controlled according to this corrected current command. The corrected current command is a command signal generated based on the corrected phase and used to control the output AC current of the microinverter. The phase of this command signal is consistent with the corrected phase, ensuring that the AC current output by the microinverter is accurately synchronized with the grid fundamental voltage, and that harmonic components caused by high-frequency interference in its waveform are effectively suppressed. The corrected current command can be generated by inputting the corrected phase into the current controller, with the aim of guiding the microinverter to output high-quality AC current and avoiding the injection of unwanted harmonics.

[0073] The proposed solution uses the grid-connected AC voltage signal as input to initiate a parallel processing flow. Specifically, this grid-connected AC voltage signal is simultaneously fed into two independent phase-locked loops (PLLs) for processing. First, the first PLL is designed to capture the voltage phase affected by interference components with frequencies higher than the fundamental frequency, thereby extracting the first phase information. This means that the bandwidth or filtering characteristics of the first PLL allow high-frequency disturbance components to pass through, ensuring that its output phase signal reflects the true instantaneous phase of the grid voltage, including high-frequency oscillations. Simultaneously, the second PLL is designed to effectively filter out these interference components, focusing on extracting the pure fundamental phase, thus obtaining the second phase information. The second PLL, through its internal filtering mechanism, ensures that its output phase signal is a smooth and accurate fundamental phase reference. It is precisely because these two PLLs operate in parallel and have different filtering characteristics that subsequent calculation of the phase disturbance becomes possible. By comparing the first and second phase information, the system can accurately calculate the phase disturbance, which directly reflects the instantaneous impact of high-frequency interference on the internal phase synchronization of the micro-inverter. Once the phase disturbance is determined, it is used to correct the second phase information, thereby generating a corrected phase. This corrected phase is a compensated, more accurate phase reference that includes both the synchronization information of the grid fundamental wave and effectively eliminates the phase deviation caused by high-frequency interference. Based on this, the microinverter uses this corrected phase to generate a corrected current command. This means that the microinverter's current controller no longer relies on the disturbed original phase information, but instead generates its output current reference waveform based on the precisely corrected phase. Finally, the microinverter controls its output AC current according to this corrected current command. In this way, the microinverter can output an AC current that is precisely synchronized with the grid fundamental voltage and has a pure waveform, thereby effectively suppressing the internal phase modulation effect caused by high-frequency voltage disturbances and avoiding the injection of sideband harmonic currents into the grid. The entire process forms a closed-loop control, ensuring the stable and high-quality grid-connected operation of the microinverter in complex grid environments.

[0074] In some preferred embodiments, this application is implemented as follows. First, the AC voltage signal at the grid connection point can be acquired in real time by a voltage sensor and converted into a digital signal by an analog-to-digital converter for processing by a digital signal processor (DSP) or microcontroller (MCU). The first phase-locked loop (PLL) processing can employ a wideband synchronous reference coordinate system (SRF-PLL) with a relatively high low-pass filter cutoff frequency to ensure that the phase oscillations caused by high-frequency voltage disturbances can be accurately tracked and reflected in the first phase information. For example, its bandwidth can be set sufficient to cover the grid fundamental frequency and the high-frequency disturbance frequencies in its vicinity. Simultaneously, the second PLL processing can employ another synchronous reference coordinate system PLL, but with a lower low-pass filter cutoff frequency or integrated with a notch filter for known high-frequency disturbance frequencies, thereby effectively filtering out high-frequency interference and extracting only the pure grid fundamental phase as the second phase information. Subsequently, the step of calculating the phase disturbance can be achieved by directly subtracting the first phase information from the second phase information in real time to obtain a signal reflecting the instantaneous phase deviation. For example, if the first phase information is theta_1(t) and the second phase information is theta_2(t), then the phase disturbance can be simply expressed as Deltatheta(t) = theta_1(t) - theta_2(t). Next, the step of generating the corrected phase applies this phase disturbance to the second phase information. Specifically, the corrected phase can be expressed as theta_{corr}(t) = theta_2(t) - Deltatheta(t), or by appropriate addition or subtraction operations based on the definition of the disturbance and the compensation direction to counteract its effects. Based on this corrected phase, the generation of the corrected current command can be achieved through a proportional resonant (PR) controller, which uses the corrected phase as its internal reference to generate a sinusoidal current command in phase with the grid voltage. Finally, the pulse width modulation (PWM) module inside the microinverter drives the inverter's power switching devices according to this corrected current command, thereby controlling the microinverter to output an AC current that is synchronized with the grid and has a clean waveform. The entire control algorithm can be integrated into the digital controller inside the micro-inverter to achieve real-time and efficient power output control.

[0075] This application further proposes steps for calculating the phase perturbation amount, including:

[0076] Obtain the phase difference between the first phase information and the second phase information;

[0077] Spectral analysis is performed on the phase difference between the first phase information and the second phase information to identify multiple perturbation frequency components contained in the phase difference between the first phase information and the second phase information.

[0078] Based on multiple perturbation frequency components, determine multiple original perturbation frequencies that cause the phase difference between the first phase information and the second phase information;

[0079] Based on multiple original disturbance frequencies, a composite phase correction command is generated;

[0080] The composite phase correction command is used as the phase perturbation quantity.

[0081] Spectrum analysis refers to the process of converting a time-domain signal into a frequency-domain signal to reveal the various frequency components contained in the signal and their corresponding amplitude and phase information. It can be implemented using various mathematical methods such as Fast Fourier Transform, Discrete Fourier Transform, Wavelet Transform, or Short-Time Fourier Transform. Its purpose is to decompose complex phase difference signals into identifiable single frequency components, thereby identifying the perturbation frequency components present in the signal. The original perturbation frequency refers to the frequency that actually causes the phase information difference, which is further selected and determined from the identified perturbation frequency components after spectral analysis of the phase difference. The components, which can be obtained by processing the spectrum analysis results through noise filtering, threshold judgment, or pattern recognition, aim to extract the truly physically meaningful disturbance source frequencies that need compensation from the complex spectrum information; the composite phase correction command refers to a comprehensive correction signal formed by superimposing the corresponding phase correction components of multiple original disturbance frequencies. It can be constructed by linearly superimposing the amplitude and phase tracking of each original disturbance frequency component and generating the corresponding reverse compensation signal. Its purpose is to provide a precise phase correction amount that can simultaneously cancel the effects of multiple frequency disturbances.

[0082] The proposed solution obtains the phase difference between first and second phase information. This phase difference directly reflects the impact of interference components with frequencies higher than the fundamental frequency in the power grid on the voltage phase, thus providing fundamental data for subsequent disturbance analysis. Based on this, spectral analysis of the phase difference decomposes it into multiple independent disturbance frequency components, revealing various disturbances hidden within complex signals. Further, based on these identified disturbance frequency components, multiple original disturbance frequencies causing the phase difference are determined. This process aims to accurately pinpoint the actual interference source causing the phase deviation, avoiding the processing of irrelevant noise. Subsequently, based on these original disturbance frequencies, a composite phase correction command is generated. This command is a comprehensive compensation signal for multiple disturbance frequencies, capable of simultaneously and accurately canceling phase disturbances of different frequencies. Finally, this composite phase correction command is used as the phase disturbance quantity for subsequent corrected phase generation.

[0083] This scheme overcomes the inaccuracy problem that may exist in directly calculating phase disturbance quantities through this refined phase disturbance quantity calculation mechanism. In the micro-inverter output power control method, the phase information extracted by the first phase-locked loop (PLL) includes the voltage phase affected by the disturbance component, while the phase information extracted by the second PLL is the fundamental phase after filtering out the disturbance component. When the two are used to calculate the phase difference, this difference accurately characterizes the phase disturbance caused by high-frequency noise in the power grid. This scheme performs in-depth spectrum analysis and original disturbance frequency identification on this phase difference, which can accurately capture these small but cumulative phase oscillations. These accurately identified disturbance frequencies are used to generate composite phase correction commands and used as phase disturbance quantities, so that the subsequently generated correction phase can more accurately correct the current command. This enables the micro-inverter to output a purer AC current, effectively suppressing specific sideband harmonic currents generated by high-frequency noise interference in the internal PLL, thereby improving the accuracy and stability of the micro-inverter output power control. Especially when multiple micro-inverters are connected in parallel, it can reduce the harmonic superposition problem caused by the "group effect".

[0084] In some preferred embodiments, this application is implemented as follows: First, in a digital signal processor, the first phase information and the second phase information acquired in real time are subtracted point by point to obtain a continuous phase difference signal. For example, if the first phase information is theta_1(t) and the second phase information is theta_2(t), then the phase difference signal is Deltatheta(t) = theta_1(t) - theta_2(t). Subsequently, periodic spectral analysis is performed on the phase difference signal Deltatheta(t). For example, a fast Fourier transform algorithm can be used to process the acquired phase difference data within each fixed time window to obtain the spectrum of the phase difference signal within that time window. By analyzing the energy peaks in the spectrum, multiple perturbation frequency components can be identified. These peaks represent periodic perturbations of different frequencies present in the phase difference signal.

[0085] Furthermore, based on the identified multiple disturbance frequency components, an energy threshold or frequency range can be set to filter out components with significant energy and frequencies within a specific range, identifying them as the multiple original disturbance frequencies causing the phase difference. For example, if noise around 2 kHz is known to exist in the power grid, then spectral peaks near 2 kHz can be the focus. Once these original disturbance frequencies are determined, their instantaneous amplitude and instantaneous phase in the phase difference signal can be tracked in real time for each original disturbance frequency. This can be achieved by constructing multiple independent digital notch filters or adaptive filters, each designed for one original disturbance frequency.

[0086] Based on the instantaneous amplitude and phase of each original disturbance frequency, a compensation signal with the same frequency and instantaneous amplitude but opposite phase is generated as the corresponding phase correction component. For example, if a certain original disturbance frequency component is A sin(omega t + phi), then the corresponding phase correction component is -A sin(omega t + phi). Finally, all generated phase correction components are linearly superimposed to form a composite phase correction command. This composite phase correction command contains accurate compensation information for all identified original disturbance frequencies and is directly used as the phase disturbance quantity, input into the subsequent correction phase generation module to achieve accurate correction of the current command.

[0087] This application further proposes a step of performing spectral analysis on the phase difference between the first phase information and the second phase information to identify multiple perturbation frequency components contained in the phase difference between the first phase information and the second phase information, including:

[0088] Perform continuous short-time spectrum analysis on the phase difference between the first phase information and the second phase information;

[0089] Based on the continuous short-time spectrum analysis results, the instantaneous disturbance frequency in the phase difference between the first phase information and the second phase information is extracted, and the instantaneous disturbance frequency is used as multiple disturbance frequency components.

[0090] Among them, continuous short-time spectrum analysis refers to the method of dividing a long time-domain signal into a series of short time periods or "windows" and performing spectrum analysis on the signal within each window. Specifically, techniques such as short-time Fourier transform (STFT) or wavelet transform can be used. Its purpose is to provide information on the frequency changes of the signal in the time domain, thereby capturing the dynamic characteristics of the signal frequency changing over time. Instantaneous perturbation frequency refers to the perturbation frequency that reflects the frequency components of the signal at that moment, identified by spectrum analysis within a specific short time window. Its purpose is to more accurately describe the frequency characteristics of the phase difference signal over a short period of time, in order to adapt to the situation where the perturbation frequency changes rapidly over time.

[0091] The proposed solution overcomes the limitations of traditional overall spectrum analysis in handling non-steady-state signals by performing continuous short-time spectrum analysis on the phase difference between the first and second phase information. By dividing the phase difference signal into continuous time windows and performing spectrum analysis on each window, the system can track the dynamic evolution of frequency components in the phase difference signal in real time, thereby capturing those instantaneous disturbance frequencies that change rapidly over time. Based on this, these instantaneous disturbance frequencies are extracted according to the results of the continuous short-time spectrum analysis and treated as multiple disturbance frequency components. This processing method allows the identified disturbance frequency components to more accurately reflect the true frequency characteristics of the phase difference signal at different times. Through this improved spectrum analysis method, the system can more comprehensively and accurately identify disturbance frequencies in the phase difference that may have time-varying characteristics. This provides a more precise input for subsequently determining the original disturbance frequency based on these disturbance frequencies and generating composite phase correction commands, thereby effectively improving the accuracy and real-time performance of the overall phase disturbance calculation. This, in turn, enables the micro-inverter to more effectively cancel internally generated harmonics caused by high-frequency grid noise, ensuring the waveform quality of the output current.

[0092] In some preferred embodiments, continuous short-time spectral analysis is performed on the phase difference between the first and second phase information, specifically using Short-Time Fourier Transform (STFT). For example, the phase difference signal can be acquired at a fixed sampling rate, and then a Fast Fourier Transform (FFT) can be performed on the data within each window by sliding a fixed-length time window. This time window can be set to, for example, 256 sampling points, and a certain overlap rate can be set between windows, such as 50%, to ensure the continuity and smoothness of the spectral analysis. In this way, a time-frequency graph can be generated, which visually shows the frequency components of the phase difference signal changing over time. Furthermore, based on the results of the continuous short-time spectral analysis, instantaneous perturbation frequencies in the phase difference are extracted. This can be specifically achieved by identifying frequency peaks with significant energy in the spectral results of each time window. For example, an energy threshold can be set, and frequency components higher than this threshold can be identified as instantaneous perturbation frequencies. These identified instantaneous perturbation frequencies, such as 2kHz, 4kHz, etc., will be used as multiple perturbation frequency components for subsequent calculation of phase perturbation.

[0093] This application further proposes steps for performing continuous short-time spectrum analysis, including:

[0094] The spectral characteristics of the signal that monitors the phase difference between the first phase information and the second phase information;

[0095] Adjust the window length and / or window overlap rate of continuous short-time spectrum analysis based on spectral characteristics;

[0096] Based on the adjusted window length and / or window overlap rate, perform continuous short-time spectral analysis on the phase difference between the first phase information and the second phase information.

[0097] The monitoring of the spectral characteristics of the signal with phase difference between the first and second phase information involves real-time or near-real-time spectral analysis of the phase difference signal to obtain information such as its frequency distribution, energy concentration region, dominant frequency components, and spectral trend over time. This can be achieved using techniques such as Fast Fourier Transform (FFT), wavelet transform, or adaptive filtering, with the aim of providing a data foundation for subsequent dynamic adjustment of short-time spectral analysis parameters. The window length refers to the duration of each time interval used to extract the signal for Fourier transform in short-time spectral analysis. It can be set according to the signal sampling rate and the required time / frequency resolution, aiming to balance time and frequency resolution. A longer window length provides better frequency resolution but reduces time resolution, and vice versa. The window overlap rate refers to the proportion of overlap between two adjacent short-time spectral analysis windows. It can be set according to the smoothness of the analysis and computational efficiency, aiming to ensure the continuity and smoothness of the spectral analysis results and avoid information loss or discontinuity caused by window switching.

[0098] The proposed solution dynamically adjusts the parameters of short-time spectrum analysis to adapt to the constantly changing spectral characteristics of the phase difference signal. Specifically, firstly, the system continuously monitors the spectral characteristics of the signal with respect to the phase difference between the first and second phase information. This monitoring process forms the basis for dynamic adjustment, enabling the system to perceive the signal's frequency composition, energy distribution, and its changes over time in real time. For example, when a change in the signal spectrum or the appearance of new frequency components is detected, the system can acquire this information promptly. Based on this, the system adjusts the window length and / or window overlap rate of the continuous short-time spectrum analysis according to the monitored spectral characteristics. For example, when the signal spectrum changes rapidly and instantaneous frequency identification capability needs to be improved, the system can reduce the window length to improve time resolution; conversely, when the signal spectrum is relatively stable and frequency resolution capability needs to be improved, the system can increase the window length to improve frequency resolution. Simultaneously, adjusting the window overlap rate helps balance the smoothness of the analysis and computational efficiency. Finally, the system performs continuous short-time spectrum analysis on the phase difference between the first and second phase information based on these adjusted parameters. This adaptive spectrum analysis method allows the system to analyze with parameters suitable for the current signal state, thereby obtaining accurate spectral information. In this way, even in complex power grid environments where the spectral characteristics of the phase difference signal are constantly changing, the system can accurately identify the multiple disturbance frequency components contained therein. This provides a reliable basis for subsequent calculation of phase disturbance, generation of composite phase correction commands, and final control of the micro-inverter output current. It also suppresses specific sideband harmonic currents caused by interference with the internal phase-locked loop, avoiding the problems of reduced accuracy and insufficient identification effect that may be caused by fixed parameter analysis.

[0099] In some preferred embodiments, this application is implemented as follows: A real-time spectrum analysis module can be set up in the digital signal processor (DSP) of the microinverter. This module continuously performs Fast Fourier Transform (FFT) or Power Spectral Density (PSD) estimation on the phase difference signal between the first phase information and the second phase information to monitor its spectral characteristics. For example, this module can calculate the instantaneous bandwidth of the signal, the energy concentration of the dominant frequency, or the flatness of the spectrum. When a significant increase in the instantaneous bandwidth of the spectrum is detected, indicating the presence of new or rapidly changing frequency components in the signal, the control algorithm can determine that the time resolution needs to be increased, thereby dynamically reducing the window length of the short-time spectrum analysis from a default longer value (e.g., 256 sampling points) to a shorter value (e.g., 128 sampling points). Simultaneously, to maintain the continuity and smoothness of the spectrum analysis, the window overlap rate can be adjusted from 50% to 75%. Conversely, when the spectral characteristics show that the signal energy is concentrated on a few stable frequencies and the spectral flatness decreases, it indicates that the signal is relatively stable. In this case, the window length can be increased (e.g., from 256 sampling points to 512 sampling points) to improve frequency resolution, while the window overlap rate can be appropriately adjusted to optimize computational efficiency. Once the window length and / or window overlap rate are adjusted, the short-time spectrum analysis module will immediately use these new parameters to process subsequent phase difference signals, thereby ensuring high-precision spectrum analysis results under any operating conditions and providing support for accurately identifying disturbance frequency components.

[0100] This application further proposes a step of extracting the instantaneous disturbance frequency of the phase difference between the first phase information and the second phase information based on continuous short-time spectrum analysis results, and using the instantaneous disturbance frequency as multiple disturbance frequency components, including:

[0101] Based on continuous short-time spectrum analysis results, identify the energy peak of the spectrum of the phase difference between the first phase information and the second phase information;

[0102] Analyze the spectral characteristics of the energy peak, which include peak width and / or peak shape;

[0103] Based on the spectral characteristics of the energy peak, determine whether the spectral characteristics of the energy peak are formed by the superposition of multiple close instantaneous disturbance frequencies;

[0104] If it is determined that the disturbance is formed by the superposition of multiple close instantaneous disturbance frequencies, separate the multiple close instantaneous disturbance frequencies;

[0105] The multiple instantaneous disturbance frequencies after separation, as well as the instantaneous disturbance frequencies corresponding to the energy peaks that are not determined to be superimposed, are taken as multiple disturbance frequency components contained in the phase difference between the first phase information and the second phase information.

[0106] The spectral characteristics of an energy peak refer to the inherent properties of the energy concentration region in the spectrum, which can be characterized by parameters such as peak width, peak shape, peak symmetry, peak slope, or peak sidelobe structure. The purpose is to provide a basis for determining whether the energy peak is composed of a single frequency component or a superposition of multiple frequency components. Separating multiple close instantaneous disturbance frequencies refers to decomposing a composite energy peak formed by the superposition of multiple frequency components into multiple independent, more refined frequency components. This can be achieved using signal decomposition algorithms, blind source separation techniques, high-resolution spectrum estimation methods, or parameter estimation methods based on model fitting. The aim is to accurately identify and extract each independent instantaneous disturbance frequency, avoiding mistaking multiple close frequencies for a single frequency, thereby improving the accuracy of frequency analysis.

[0107] The present application's solution performs continuous short-time spectral analysis on the phase difference between the first and second phase information, and further optimizes the extraction process of instantaneous disturbance frequencies based on this analysis. Specifically, the solution first identifies energy peaks in the phase difference spectrum based on the results of the continuous short-time spectral analysis. These energy peaks are preliminary indicators of potential disturbance frequencies because they represent frequency regions where signal energy is concentrated. Subsequently, the solution performs in-depth spectral characteristic analysis on these identified energy peaks, such as evaluating their peak width and / or peak shape. Peak width and shape can provide important information about whether the peak is caused by a single frequency or is formed by the superposition of multiple close instantaneous disturbance frequencies. For example, an unusually wide peak or an irregular peak shape usually suggests that it may contain multiple closely adjacent frequency components. Based on the analysis of the spectral characteristics of the energy peaks, the solution further determines whether the energy peak is formed by the superposition of multiple close instantaneous disturbance frequencies. This determination is a crucial step, enabling the system to distinguish between simple single-frequency disturbances and complex superimposed disturbances. If the determination result indicates that the energy peak is indeed formed by the superposition of multiple close instantaneous disturbance frequencies, the system will perform a separation operation to decompose these superimposed frequency components into multiple independent instantaneous disturbance frequencies. This separation process can significantly improve the accuracy of frequency identification and avoid confusing multiple actual disturbance frequencies into a single frequency. Finally, both the multiple instantaneous disturbance frequencies obtained after separation and those instantaneous disturbance frequencies corresponding to the energy peak that were not determined to be superimposed will be regarded as multiple disturbance frequency components contained in the phase difference between the first phase information and the second phase information. Through this refined processing, this scheme overcomes the inaccuracy caused by simply using the center frequency of the energy peak as the instantaneous disturbance frequency in traditional methods. It can more comprehensively and accurately capture the complex disturbance frequency components contained in the phase difference, especially those disturbances that are superimposed due to their close frequencies. This accurate identification of disturbance frequency components provides a more reliable and refined basis for subsequent phase correction. When these more accurate disturbance frequency components are used to determine the original disturbance frequency and generate composite phase correction commands, the accuracy and performance of micro-inverter output power control can be significantly improved, thereby effectively suppressing specific sideband harmonic currents caused by high-frequency noise interference in the internal phase-locked loop, and thus improving grid-connected power quality.

[0108] In some preferred embodiments, this application is implemented as follows: After obtaining continuous short-time spectrum analysis results, a threshold-based peak detection algorithm can be used to identify the energy peak of the spectrum of the phase difference between the first phase information and the second phase information. For example, an energy threshold can be set, and any frequency point higher than the threshold is initially identified as an energy peak. Subsequently, in order to analyze the spectral characteristics of these energy peaks, the full width at half maximum (FWHM) of each identified peak can be calculated as the peak width, and the slope change rate on both sides of the peak can be calculated or the residual of Gaussian fitting can be used to characterize the peak shape. For example, if the FWHM of a peak exceeds a preset threshold, or its shape deviates significantly from an ideal single-frequency peak (such as a Gaussian peak or a Lorentz peak), it can be preliminarily determined that it may be formed by the superposition of multiple close instantaneous disturbance frequencies. Further, based on the peak width and / or peak shape obtained from the analysis, a machine learning classifier or a rule-based expert system can be used to determine whether the spectral characteristics of the energy peak are formed by the superposition of multiple close instantaneous disturbance frequencies. For example, a Support Vector Machine (SVM) model can be trained, taking features such as peak width and peak symmetry as input, and outputting a judgment result on whether it is a superimposed peak. If it is determined to be formed by the superposition of multiple close instantaneous disturbance frequencies, signal decomposition techniques such as Independent Component Analysis (ICA) or Non-negative Matrix Factorization (NFF) can be used to separate the multiple close instantaneous disturbance frequencies. For example, the local spectral region where the superimposed peak is located can be regarded as a mixed signal, and decomposed into multiple independent frequency components using the ICA algorithm. Finally, the multiple independent instantaneous disturbance frequencies obtained through the above separation process, as well as the instantaneous disturbance frequencies corresponding to the energy peaks that were not identified as superimposed in the judgment step, are collectively used as multiple disturbance frequency components contained in the phase difference between the first phase information and the second phase information for use by the subsequent phase correction module.

[0109] This application further proposes steps for separating multiple close instantaneous disturbance frequencies, including:

[0110] Perform curve fitting on the energy peak;

[0111] Based on the fitting results, the energy peak is decomposed into multiple independent sub-peaks;

[0112] The center frequency corresponding to each sub-peak is extracted as multiple close instantaneous perturbation frequencies.

[0113] Curve fitting refers to approximating or describing the shape and distribution of energy peaks using mathematical models. Methods such as Gaussian fitting, Lorentz fitting, polynomial fitting, or spline fitting can be employed. The aim is to provide a mathematical basis for subsequent peak decomposition, thereby accurately identifying superimposed frequency components. Decomposing the energy peak into multiple independent sub-peaks involves using algorithms to break down a composite, broad energy peak into several narrower peaks with independent characteristics. Each sub-peak represents an independent frequency component. Iterative fitting, peak detection and separation algorithms, or model-based decomposition methods can be used. The goal is to identify multiple instantaneous disturbance frequencies that were originally superimposed. Extracting the center frequency corresponding to each sub-peak involves determining the frequency point with the largest energy or amplitude from each independent sub-peak. This can be done by calculating the peak position of the fitted curve or directly finding the maximum value point from the decomposed sub-peak data. The purpose is to obtain the value of each instantaneous disturbance frequency for subsequent phase correction.

[0114] The proposed solution captures the shape and potential superposition structure of energy peaks in the spectrum by performing curve fitting on the energy peaks. This fitting process provides a data foundation for subsequent decomposition, avoiding errors caused by simple thresholding or coarse segmentation. Based on the fitting results, the energy peaks are decomposed into multiple independent sub-peaks, separating multiple transient disturbance frequencies that were originally superimposed. Each sub-peak represents an independent disturbance frequency component, thus solving the problem that traditional methods struggle to distinguish these closely adjacent frequencies. Subsequently, the center frequencies corresponding to each sub-peak are extracted. These center frequencies, obtained after fitting and decomposition, represent estimates of each transient disturbance frequency. This transient disturbance frequency information, as multiple disturbance frequency components contained in the phase difference between the first and second phase information, can be identified. In the entire micro-inverter output power control method, this separation process is a crucial step in identifying phase disturbance quantities. By separating these close transient disturbance frequencies, raw disturbance frequency information can be provided for the generation of subsequent composite phase correction commands. For example, when generating the corresponding phase correction component, the instantaneous amplitude and phase of each separated instantaneous disturbance frequency can be tracked in real time, and a compensation signal with the same frequency, the same instantaneous amplitude, and opposite phase can be generated. The composite phase correction command formed by superimposing these compensation signals can cancel the phase modulation effect caused by high-frequency voltage noise, thereby reducing the sideband harmonic components in the micro-inverter output current. Therefore, this solution can improve the accuracy of phase correction, thereby improving the power quality of the micro-inverter's grid-connected output current, avoiding under-correction or over-correction problems caused by inaccurate frequency identification, and ensuring the stable operation and purity of the micro-inverter's output power in complex grid environments.

[0115] In some preferred embodiments, when the system detects an energy peak in the spectrum of the phase difference between the first and second phase information, formed by the superposition of multiple close instantaneous perturbation frequencies—for example, a wide peak near 2 kHz—it indicates the possible presence of multiple perturbation components with close frequencies such as 2000 Hz, 2010 Hz, and 2020 Hz. To separate these components, curve fitting can be performed on the energy peak. Specifically, a Gaussian Mixture Model (GMM) can be used to fit the energy peak. This model decomposes a complex peak into a superposition of multiple Gaussian distributions, each representing an independent sub-peak. For example, if the fitting result shows that the peak can be formed by the superposition of three Gaussian distributions, then based on this fitting result, the energy peak is decomposed into three independent sub-peaks. Subsequently, the corresponding center frequency is extracted from each decomposed sub-peak. For example, the center frequency of the first sub-peak might be 2000 Hz, the center frequency of the second sub-peak might be 2010 Hz, and the center frequency of the third sub-peak might be 2020 Hz. The extracted center frequencies, namely 2000Hz, 2010Hz, and 2020Hz, are used as multiple close instantaneous disturbance frequencies for subsequent phase correction processing. In this way, even if multiple disturbance frequencies are close together, they can be identified and separated, thus providing frequency information for subsequent compensation.

[0116] This application further proposes a step for generating a composite phase correction command based on multiple original disturbance frequencies, including:

[0117] Based on multiple original disturbance frequencies, corresponding phase correction components are generated;

[0118] The phase correction components are superimposed to generate a composite phase correction command.

[0119] Among them, the original disturbance frequency refers to the interference signal component with specific frequency characteristics that causes the phase deviation of the system. It can originate from high-frequency noise in the power grid, harmonics generated by nonlinear loads, or oscillations within the system. Its purpose is to identify and quantify the specific frequency component that causes the phase disturbance. The phase correction component refers to a compensation signal calculated or generated separately for each original disturbance frequency to cancel the frequency disturbance. It can be a sine wave signal with the same frequency and amplitude as the disturbance frequency but opposite in phase, or it can be a correction amount calculated based on a specific frequency response model. Its purpose is to provide customized compensation for each independent disturbance frequency. The composite phase correction command refers to the final correction signal formed by combining all independent phase correction components. It can be obtained by linear superposition, weighted summation, or other signal synthesis methods. Its purpose is to provide a comprehensive command that can compensate for multiple frequency disturbances simultaneously. Generating corresponding phase correction components refers to calculating or constructing a compensation signal that can effectively counteract the impact of each identified original disturbance frequency. Specifically, this can be done by looking up a preset frequency-compensation mapping relationship or by calculating it in real time using an adaptive algorithm. The purpose is to provide an accurate correction basis for subsequent superposition operations. Superimposing phase correction components refers to combining the independent phase correction components generated for different original disturbance frequencies into a unified correction signal. This can be achieved through simple arithmetic addition or by weighted summation considering the weights of each component. The purpose is to integrate the scattered compensation information into a single command that can be directly applied to system control.

[0120] The proposed solution decomposes complex phase disturbances into multiple independent original disturbance frequencies and generates specific phase correction components for each frequency. These components are then superimposed to form a composite phase correction command, thereby achieving precise compensation for phase disturbances. Specifically, after acquiring the AC voltage signal at the grid connection point, the system executes a first phase-locked loop (PLL) and a second PLL in parallel, extracting first phase information containing interference components and second phase information after filtering out interference components, respectively. Subsequently, by calculating the phase difference between the first and second phase information and performing spectral analysis on this phase difference, multiple disturbance frequency components can be identified, thus determining the multiple original disturbance frequencies causing the phase difference. Based on this, the proposed solution further generates corresponding phase correction components for these identified multiple original disturbance frequencies. This means that the system no longer treats all disturbances as a whole for coarse compensation but can identify and process disturbances at each specific frequency. For example, if there are disturbance frequencies of 2kHz and 3kHz, the system will generate independent correction signals for 2kHz and 3kHz, respectively. This refined processing method ensures that each correction component precisely matches the characteristics of its target disturbance frequency, including its amplitude and phase. Subsequently, these phase correction components generated for different original disturbance frequencies are precisely superimposed to form a unified composite phase correction command. The superposition operation integrates the individual correction signals into a comprehensive command that simultaneously contains compensation information for multiple frequency disturbances. This superposition method ensures that the final correction command covers the entire disturbance spectrum, thereby achieving comprehensive and accurate cancellation of complex phase disturbances. It is precisely this strategy of "decomposing – processing independently – precisely superimposing" complex disturbances that allows the generated composite phase correction command to more accurately reflect and cancel the actual amount of phase disturbance. Compared to correction methods based solely on a single frequency or coarse estimation, this scheme can effectively suppress harmonic components caused by high-frequency voltage noise, such as the 1950Hz and 2050Hz sideband harmonics mentioned in the background art, thereby significantly improving the grid-connected power quality of micro-inverters, avoiding the in-phase superposition effect of harmonics when multiple inverters are connected in parallel, and solving the problem of how to effectively utilize the original disturbance frequency to construct a composite correction command to achieve more accurate phase compensation.

[0121] In some preferred embodiments, the specific process of generating a composite phase correction command based on multiple original perturbation frequencies can be implemented as follows: After the system determines multiple original perturbation frequencies, such as f1, f2, and f3, through spectrum analysis, the system can configure an independent signal generator or algorithm module for each original perturbation frequency. For example, for the original perturbation frequency f1, the system can monitor in real time the instantaneous amplitude and instantaneous phase of the component corresponding to f1 in the phase difference between the first phase information and the second phase information. Based on this real-time tracked information, the system can generate a sinusoidal compensation signal with the same frequency and instantaneous amplitude as the f1 component but with a completely opposite phase, and use it as the phase correction component for f1. Similarly, for the original perturbation frequencies f2 and f3, the system can also use a similar method to generate corresponding phase correction components respectively. Once all identified original perturbation frequencies have generated their respective phase correction components, these components are then fed into a summing unit or digital accumulator. This summing unit linearly superimposes all independent phase correction components. For example, if the phase correction components are represented as time functions C1(t), C2(t), and C3(t), then the composite phase correction instruction can be simply expressed as C_composite(t) = C1(t) + C2(t) + C3(t). This superposition operation can be implemented in a digital signal processor using simple addition. The final superimposed signal is the composite phase correction instruction, which contains comprehensive compensation information for all identified original disturbance frequencies. It can be directly used in subsequent phase correction stages to cancel the phase deviation caused by these disturbance frequencies.

[0122] Through the above technical solution, this application can generate independent phase correction components for each original disturbance frequency and accurately superimpose them to construct a composite phase correction command capable of simultaneously compensating for multiple frequency disturbances. This method avoids single or coarse compensation for complex phase disturbances, instead achieving refined and customized correction for different frequency disturbance components. Therefore, this solution can significantly improve the accuracy and effectiveness of phase compensation, enabling micro-inverters to output cleaner AC current even in complex grid environments, despite facing multi-band high-frequency voltage noise, effectively suppressing sideband harmonics generated by phase modulation, thereby improving grid-connected power quality.

[0123] This application further proposes steps for generating the corresponding phase correction component, including:

[0124] For multiple original disturbance frequencies, the instantaneous amplitude and instantaneous phase of the component corresponding to the original disturbance frequency in the phase difference between the first phase information and the second phase information are tracked in real time.

[0125] Based on the instantaneous amplitude and instantaneous phase, a compensation signal with the same frequency, the same instantaneous amplitude, and opposite phase as the component corresponding to the original disturbance frequency is generated as the phase correction component.

[0126] Real-time tracking refers to the system's ability to continuously and dynamically monitor and acquire the amplitude and phase information of the target signal at a specific point in time. This can be achieved using digital signal processing techniques, such as adaptive filtering algorithms or Kalman filtering, with the aim of capturing the dynamic characteristics of the signal. The compensation signal is an artificially generated signal designed to cancel or reduce the influence of disturbance signals. This can be achieved through inverse superposition, aiming to ensure accurate cancellation. The phase correction component is the signal component used to correct the system's phase error. It can be a combination of one or more compensation signals, and its purpose is to provide correction for subsequent phase correction.

[0127] This application optimizes the output power control method of micro-inverters to address the problem that the static form of correction commands in existing technologies is ill-suited to dynamic phase disturbances. Specifically, in generating composite phase correction commands, this solution no longer relies on fixed correction parameters but introduces a dynamic compensation mechanism. First, for multiple identified original disturbance frequencies, the system tracks in real time the instantaneous amplitude and instantaneous phase of the components corresponding to these original disturbance frequencies in the phase difference between the first and second phase information. This means the system can continuously sense the intensity and phase changes of each disturbance component at different times, thereby acquiring its dynamic characteristics. Subsequently, based on these real-time tracked instantaneous amplitudes and phases, the system generates a compensation signal with the same frequency, the same instantaneous amplitude, and opposite phase to the component corresponding to the original disturbance frequency. This compensation signal is used as the phase correction component. Since the compensation signal perfectly matches the disturbance component in instantaneous characteristics but has an opposite phase, when they are superimposed, they can cancel out the original disturbance component. In this way, each phase correction component can dynamically adapt to the changes in the specific disturbance frequency component it targets. When these dynamically generated phase correction components are superimposed to form a composite phase correction command, the composite command has the ability to compensate for dynamically changing phase disturbances in the power grid. The dynamism of this compensation mechanism significantly improves the accuracy of phase disturbance calculation, thereby enabling the correction current command generated based on the corrected phase to more effectively suppress harmonics in the output current of the micro-inverter, thus improving power quality and solving the problem that the static form of the correction command cannot effectively cope with dynamic disturbances.

[0128] In some embodiments, the process of generating the corresponding phase correction component can be implemented as follows: For each original perturbation frequency identified in the spectrum analysis, such as 2kHz and 4kHz, the system can construct an independent digital phase-locked loop (PLL). These independent PLLs are configured to specifically track the component corresponding to its specific original perturbation frequency in the phase difference between the first phase information and the second phase information. Through these PLLs, the system can extract the instantaneous amplitude and instantaneous phase of each perturbation frequency component in real time. For example, for a 2kHz perturbation component, its PLL will output a signal representing the instantaneous amplitude and instantaneous phase of that component. Subsequently, based on these real-time acquired instantaneous amplitudes and instantaneous phases, the system can use a signal generator or digital synthesizer to generate a sine wave signal with the same frequency and the same instantaneous amplitude as the perturbation component, but with an exact opposite phase. This inverted sine wave signal serves as the corresponding phase correction component. For example, if the instantaneous amplitude of the 2kHz disturbance component is A and the instantaneous phase is φ at a certain moment, the generated compensation signal will be a sine wave with a frequency of 2kHz, an instantaneous amplitude of A, and an instantaneous phase of φ+π (or φ-π). In this way, for each original disturbance frequency, an inverse compensation signal matching its dynamic characteristics can be generated, thus providing a dynamically adjustable component for the generation of subsequent composite phase correction commands.

[0129] Through the above technical solution, this application provides a precise and dynamic solution to the harmonic problem caused by phase disturbances in the output current of micro-inverters. By tracking the instantaneous amplitude and phase of each original disturbance frequency component in real time, and generating a compensation signal with the same instantaneous amplitude and opposite phase as a phase correction component, this application overcomes the limitation of traditional static correction commands that cannot adapt to dynamic changes in disturbance signals. This allows the generated composite phase correction command to cancel the time-varying phase disturbances caused by the original disturbance frequency, significantly improving the accuracy and effectiveness of phase correction. Ultimately, this helps suppress harmonic components in the output current of micro-inverters, improving the quality of grid-connected power.

[0130] This application further proposes a step for real-time tracking of the instantaneous amplitude and instantaneous phase of the component corresponding to the original disturbance frequency in the phase difference between the first phase information and the second phase information, including:

[0131] For each of the multiple original disturbance frequencies, an independent phase-locked loop is constructed.

[0132] The instantaneous amplitude and instantaneous phase of the component corresponding to the original disturbance frequency in the phase difference between the first phase information and the second phase information are tracked in real time using a phase-locked loop.

[0133] Independent phase-locked loops (PLLs) refer to configuring a dedicated PLL unit for each original perturbation frequency component. These PLL units operate independently of each other structurally or logically, without interfering with each other. This can be achieved by constructing multiple independent PLL circuits in hardware or by instantiating multiple independent PLL algorithm modules in software. The purpose is to ensure that the tracking process of each perturbation frequency is not affected by other frequency components, improving the specificity and accuracy of the tracking. Real-time tracking of the instantaneous amplitude and phase of the component corresponding to the original perturbation frequency in the phase difference between the first and second phase information using PLLs refers to utilizing the synchronous tracking capability of the PLL to not only lock and output the instantaneous phase of the corresponding original perturbation frequency component, but also, in conjunction with internal or external signal processing mechanisms of the PLL, such as synchronous coordinate system transformation or generalized integrators, to extract the instantaneous amplitude of this component in real time. The purpose is to provide accurate instantaneous amplitude and phase information for the subsequent generation of phase correction components.

[0134] This application's solution achieves precise and real-time tracking of the instantaneous amplitude and phase of the component corresponding to the original disturbance frequency in the phase difference between the first and second phase information by constructing independent phase-locked loops (PLLs) for multiple original disturbance frequencies. Specifically, after the system identifies multiple original disturbance frequencies, it no longer employs a single, potentially limited tracking mechanism, but instead assigns a dedicated PLL to each identified disturbance frequency. Each PLL is optimized to focus on its specific frequency component, effectively avoiding crosstalk between different frequency components and improving the extraction accuracy of the instantaneous amplitude and phase of each component. It is precisely this decoupled and specialized tracking mechanism that allows each PLL to quickly and accurately lock and output the instantaneous amplitude and phase of its corresponding component, even when the disturbance frequency is high or changes rapidly. This high-precision instantaneous information is then used to generate a compensation signal with the same frequency and instantaneous amplitude as the original disturbance component but with an opposite phase—the phase correction component. By superimposing these precisely generated phase correction components, a more accurate composite phase correction command can be formed. This composite phase correction command, as a phase disturbance, can more effectively correct the output current command of the micro-inverter, thereby canceling the internally generated harmonics caused by high-frequency noise from the power grid at the source, ensuring the purity of the micro-inverter's output current, and solving the problem that traditional methods are difficult to accurately and quickly extract instantaneous information in complex disturbance environments, thus affecting the accuracy and real-time performance of phase correction.

[0135] In some preferred embodiments, when the system identifies two main original disturbance frequencies, such as 2kHz and 4kHz, through spectrum analysis, two independent phase-locked loops (PLLs) can be constructed. The first PLL can be configured specifically to track the 2kHz disturbance component, with its loop filter and phase detector parameters optimized for the 2kHz signal to ensure rapid locking and accurate tracking of this frequency component. Simultaneously, this PLL can be integrated with or used in conjunction with a synchronous rotating coordinate system transformation module to transform the 2kHz component to the DC domain, thereby facilitating the extraction of its instantaneous amplitude and phase. Similarly, the second independent PLL can be configured specifically to track the 4kHz disturbance component, extracting its instantaneous amplitude and phase in a similar manner. These two PLLs operate in parallel, independently, each outputting the instantaneous amplitude and phase information of its corresponding frequency component. This information is then fed into a phase correction component generation module to generate respective compensation signals, which are ultimately superimposed to form a composite phase correction command, achieving accurate correction of the micro-inverter output current.

[0136] Through the above technical solution, this application can achieve accurate and real-time instantaneous amplitude and phase tracking for multiple original disturbance frequencies using independent phase-locked loops. This method effectively solves the problem that traditional direct tracking methods are difficult to accurately and quickly extract the required instantaneous information in complex power grid environments, especially when the disturbance frequency is high or changes rapidly. By providing a dedicated tracking mechanism for each disturbance frequency component, the extraction accuracy and response speed of instantaneous amplitude and phase are improved, thereby ensuring the accuracy and real-time performance of subsequent phase correction component generation. Ultimately, this effectively suppresses specific sideband harmonic currents generated by interference in the internal phase-locked loop of the micro-inverter, improving the grid-connected power quality.

[0137] refer to Figure 2 This application further proposes a micro-inverter output power control system, applied to a micro-inverter output power control method, the system comprising:

[0138] The acquisition module is used to acquire the AC voltage signal at the grid connection point;

[0139] The first phase-locked loop module extracts the first phase information based on the AC voltage signal at the grid connection point. The first phase information includes the voltage phase affected by interference components with frequencies higher than the fundamental frequency.

[0140] The second phase-locked loop module extracts the second phase information based on the AC voltage signal at the grid connection point. The second phase information includes the fundamental phase after filtering out interference components.

[0141] The calculation module is used to calculate the phase disturbance based on the first phase information and the second phase information;

[0142] The correction module is used to generate a corrected phase based on the second phase information and the phase perturbation amount;

[0143] The correction instruction generation module is used to generate correction current instructions based on the correction phase;

[0144] The current control module is used to control the output AC current of the micro inverter according to the correction current command.

[0145] The acquisition module refers to a hardware or software unit used to collect AC voltage signals at the grid connection point. It can be implemented using a voltage sensor, analog-to-digital converter, or data interface, and its purpose is to provide raw data for subsequent signal processing. The first phase-locked loop (PLL) module is a circuit or algorithm unit used to extract phase information from the AC signal. It can be implemented using a digital PLL, analog PLL, or software-based PLL algorithm, and its purpose is to acquire voltage phase information containing interference components with frequencies higher than the fundamental frequency. The second PLL module is another circuit or algorithm unit used to extract phase information from the AC signal. It can be implemented using different filtering characteristics or algorithms than the first PLL module, and its purpose is to acquire the fundamental phase information after filtering out interference. The calculation module refers to a unit used to perform data processing and logical operations. The microinverter employs microcontrollers, digital signal processors, or application-specific integrated circuits (ASICs) to determine the phase disturbance based on two types of phase information. The correction module, implemented using digital logic circuits, software algorithms, or analog circuits, generates a corrected phase based on the fundamental phase and the phase disturbance. The correction command generation module generates control commands based on specific inputs, using lookup tables, mathematical models, or proportional-integral controllers. It generates commands to control the output current based on the corrected phase. The current control module regulates and controls the current output, employing pulse-width modulation controllers, power semiconductor switches, or current sensors. It controls the microinverter's output AC current based on the corrected current command.

[0146] The proposed solution utilizes a modular system architecture to concretely implement the aforementioned micro-inverter output power control method. The system first acquires the AC voltage signal at the grid connection point in real time via an acquisition module, providing fundamental data for the entire control process. Subsequently, this voltage signal is fed in parallel into a first phase-locked loop (PLL) module and a second PLL module. The first PLL module is specifically used to extract the first phase information containing high-frequency interference components, while the second PLL module focuses on extracting the fundamental phase information after interference filtering. This parallel processing mechanism ensures that two phase references with different characteristics can be obtained simultaneously, laying the foundation for subsequent disturbance analysis.

[0147] Next, the calculation module receives phase information from the two phase-locked loop modules and calculates the phase disturbance based on the difference between them. This phase disturbance characterizes the specific impact of high-frequency interference on the voltage phase. The correction module then uses the second phase information (fundamental phase) and the calculated phase disturbance to generate a corrected phase signal. This corrected phase signal cancels out the negative impact of high-frequency interference on the synchronization phase, ensuring the accuracy of subsequent current commands.

[0148] Based on this, the correction command generation module generates a correction current command according to this corrected phase signal. This command takes into account the actual phase condition of the power grid and compensates for the phase deviation caused by high-frequency disturbances. Finally, the current control module receives and executes the correction current command to control the output AC current of the microinverter. In this way, the microinverter can output a current that is synchronized with and in phase with the grid voltage, thereby reducing output current harmonics caused by high-frequency voltage disturbances.

[0149] This system architecture provides hardware or software platform support for the actual deployment and operation of the aforementioned control methods. By decomposing the control logic into functionally defined modules, the system's feasibility is improved, its operational stability is enhanced, and the control methods can be transformed from theoretical concepts into operational entities. This solves the problem of the lack of specific system architecture support for the control methods, enabling their practical application and ultimately improving the grid-connected power quality of micro-inverters in complex grid environments.

[0150] In some embodiments, this application is specifically implemented as follows. The acquisition module may consist of a voltage sensor and an analog-to-digital converter. The voltage sensor is used to convert the AC voltage signal at the grid connection point into an analog electrical signal, and the analog-to-digital converter converts the analog electrical signal into a digital signal for subsequent digital processing.

[0151] Both the first and second phase-locked loop (PLL) modules can be implemented in software on a digital signal processor (DSP) or microcontroller (MCU). The first PLL module can employ a wideband PLL algorithm, with its internal filters designed to allow higher-frequency interference components to pass through, ensuring that the first phase information reflects the impact of high-frequency disturbances. The second PLL module can employ a narrowband PLL algorithm, with its internal filters designed to effectively filter out interference components with frequencies higher than the fundamental frequency, thereby extracting the fundamental phase.

[0152] The calculation module and the correction module can also run as software modules on the same DSP or MCU. The calculation module can perform phase difference calculation and disturbance analysis algorithms, such as determining the phase disturbance by comparing the phase angles of the outputs of two phase-locked loops. The correction module can generate a corrected phase by superimposing or subtracting the calculated phase disturbance from the second phase information according to a preset correction logic.

[0153] The correction instruction generation module can be a current reference generator based on the correction phase and the desired output power. It can also be implemented in software on a DSP or MCU, for example, by using a sine wave generator whose phase is determined by the correction phase and whose amplitude is determined by the power instruction, thereby generating the correction current instruction.

[0154] The current control module can consist of a proportional resonant (PR) controller or a proportional-integral (PI) controller and a pulse-width modulation (PWM) generator. These components are typically integrated into a DSP or MCU to drive the inverter's power semiconductor switches, such as insulated-gate bipolar transistors (IGBTs) or metal-oxide-semiconductor field-effect transistors (MOSFETs), thereby controlling the output AC current of the micro-inverter according to corrected current commands. The entire system can exchange data and receive commands from an external monitoring system via a communication interface.

[0155] Through the above technical solution, this application provides a specific system structure that transforms the micro-inverter output power control method from a theoretical level into a practically deployable solution. The system, through modular design, clarifies the responsibilities and collaborative relationships of each functional unit, thereby solving the problem of the lack of a concrete implementation carrier for the control method. This allows the control method to be effectively integrated into micro-inverter products, realizing the sensing of the AC voltage signal at the grid connection point, the extraction and correction of phase information, and the control of the output current. Ultimately, the system ensures that the micro-inverter can still output AC current that meets power quality requirements in grid environments with high-frequency voltage disturbances, suppressing the generation of harmonic components and improving grid-connected power quality and system operational stability.

[0156] The content disclosed above is only a preferred and feasible embodiment of the present invention, and is not intended to limit the scope of protection of the present invention. Therefore, all equivalent technical changes made based on the content of the present invention specification and drawings are included within the scope of protection of the present invention. Furthermore, the elements therein can be updated as technology develops.

Claims

1. A micro-inverter output power control method, characterized by, The method comprises the following steps: obtaining a grid-connected point AC voltage signal; based on the grid-connected point AC voltage signal, performing in parallel: first phase-locked loop processing to extract first phase information, the first phase information containing a voltage phase affected by an interference component with a frequency higher than a fundamental frequency; and second phase-locked loop processing to extract second phase information, the second phase information containing a fundamental phase after filtering out the interference component; calculating a phase disturbance quantity according to the first phase information and the second phase information; generating a correction phase according to the second phase information and the phase disturbance quantity; generating a modified current instruction based on the correction phase; controlling the micro-inverter to output an AC current according to the modified current instruction; the step of calculating the phase disturbance quantity comprises: obtaining a phase difference between the first phase information and the second phase information; performing frequency spectrum analysis on the phase difference between the first phase information and the second phase information to identify a plurality of disturbance frequency components contained in the phase difference between the first phase information and the second phase information; determining a plurality of original disturbance frequencies causing the phase difference between the first phase information and the second phase information according to the plurality of disturbance frequency components; generating a composite phase correction instruction according to the plurality of original disturbance frequencies; taking the composite phase correction instruction as the phase disturbance quantity.

2. The micro-inverter output power control method of claim 1, wherein, The step of performing frequency spectrum analysis on the phase difference between the first phase information and the second phase information to identify a plurality of disturbance frequency components contained in the phase difference between the first phase information and the second phase information comprises: performing continuous short-time spectrum analysis on the phase difference between the first phase information and the second phase information; extracting an instantaneous disturbance frequency in the phase difference between the first phase information and the second phase information according to the result of the continuous short-time spectrum analysis, and taking the instantaneous disturbance frequency as the plurality of disturbance frequency components.

3. The micro-inverter output power control method of claim 2, wherein, The step of performing continuous short-time spectrum analysis comprises: monitoring a frequency spectrum characteristic of a signal of the phase difference between the first phase information and the second phase information; adjusting a window length and / or a window overlap rate of the continuous short-time spectrum analysis according to the frequency spectrum characteristic; performing the continuous short-time spectrum analysis on the phase difference between the first phase information and the second phase information according to the adjusted window length and / or window overlap rate.

4. The micro-inverter output power control method of claim 2, wherein, The step of extracting an instantaneous disturbance frequency in the phase difference between the first phase information and the second phase information according to the result of the continuous short-time spectrum analysis, and taking the instantaneous disturbance frequency as the plurality of disturbance frequency components comprises: identifying an energy peak in a frequency spectrum of the phase difference between the first phase information and the second phase information according to the result of the continuous short-time spectrum analysis; analyzing a frequency spectrum characteristic of the energy peak, the frequency spectrum characteristic of the energy peak including a peak width and / or a peak shape; judging whether the frequency spectrum characteristic of the energy peak is formed by superposition of a plurality of close instantaneous disturbance frequencies according to the frequency spectrum characteristic of the energy peak; if it is judged that the frequency spectrum characteristic of the energy peak is formed by superposition of the plurality of close instantaneous disturbance frequencies, separating the plurality of close instantaneous disturbance frequencies; taking the separated plurality of instantaneous disturbance frequencies and an instantaneous disturbance frequency corresponding to an energy peak not judged to be formed by superposition as the plurality of disturbance frequency components contained in the phase difference between the first phase information and the second phase information. ​ 5. The micro-inverter output power control method of claim 4, wherein, The step of separating the plurality of close-in transient disturbance frequencies comprises: performing curve fitting processing on the energy peaks; based on the fitting result, decomposing the energy peaks into a plurality of independent sub-peaks; extracting the center frequency corresponding to each sub-peak as the plurality of close-in transient disturbance frequencies.

6. The micro-inverter output power control method of claim 1, wherein, The step of generating a composite phase correction instruction according to the plurality of original disturbance frequencies comprises: generating a corresponding phase correction component according to the plurality of original disturbance frequencies; superimposing the phase correction components to generate a composite phase correction instruction.

7. The micro-inverter output power control method of claim 6, wherein, The step of generating a corresponding phase correction component comprises: tracking the instantaneous amplitude and instantaneous phase of the component corresponding to the original disturbance frequency in the phase difference between the first phase information and the second phase information in real time for the plurality of original disturbance frequencies; generating a compensation signal with the same frequency, the same instantaneous amplitude and the opposite phase as the component corresponding to the original disturbance frequency as the phase correction component according to the instantaneous amplitude and the instantaneous phase.

8. The micro-inverter output power control method of claim 7, wherein, The step of tracking the instantaneous amplitude and instantaneous phase of the component corresponding to the original disturbance frequency in the phase difference between the first phase information and the second phase information in real time comprises: constructing an independent phase-locked loop for each of the plurality of original disturbance frequencies; tracking the instantaneous amplitude and instantaneous phase of the component corresponding to the original disturbance frequency in the phase difference between the first phase information and the second phase information in real time through the phase-locked loop.

9. A micro-inverter output power control system applied to the micro-inverter output power control method of claim 1, characterized in that, The system comprises: an acquisition module configured to acquire a grid-connected point alternating voltage signal; a first phase-locked loop module configured to extract first phase information based on the grid-connected point alternating voltage signal, the first phase information containing a voltage phase affected by an interference component with a frequency higher than a fundamental frequency; a second phase-locked loop module configured to extract second phase information based on the grid-connected point alternating voltage signal, the second phase information containing a fundamental phase after filtering out the interference component; a calculation module configured to calculate a phase disturbance amount according to the first phase information and the second phase information; a correction module configured to generate a correction phase according to the second phase information and the phase disturbance amount; a correction instruction generation module configured to generate a correction current instruction based on the correction phase; a current control module configured to control an alternating current output by the micro-inverter according to the correction current instruction. The calculation module is further configured to: acquire a phase difference between the first phase information and the second phase information; perform frequency spectrum analysis on the phase difference between the first phase information and the second phase information to identify a plurality of disturbance frequency components contained in the phase difference between the first phase information and the second phase information; determine a plurality of original disturbance frequencies causing the phase difference between the first phase information and the second phase information according to the plurality of disturbance frequency components; generate a composite phase correction instruction according to the plurality of original disturbance frequencies; use the composite phase correction instruction as the phase disturbance amount.

Images