Photovoltaic grid-connected inverter adaptive control method, device and computer equipment based on running mode adaptive switching hybrid control

Abstract

The application relates to a photovoltaic grid-connected inverter adaptive control method, device and computer equipment based on operation mode adaptive switching hybrid control. The method comprises the following steps: acquiring a first correlation parameter between a photovoltaic grid-connected inverter and a photovoltaic power station, and a second correlation parameter between the photovoltaic grid-connected inverter and a to-be-analyzed power grid corresponding to the photovoltaic grid-connected inverter; inputting the first correlation parameter and the second correlation parameter into a plurality of trained operation mode prediction models to obtain a historical operation mode of the to-be-analyzed power grid; determining a future operation mode of the to-be-analyzed power grid according to a current operation mode and the historical operation mode; determining an adaptive control instruction corresponding to the photovoltaic grid-connected inverter according to the current operation mode, the historical operation mode and the future operation mode; and performing corresponding adaptive control processing on the photovoltaic grid-connected inverter according to the adaptive control instruction. By adopting the method, the adaptive control accuracy of the photovoltaic grid-connected inverter can be improved.

Description

Technical Field

[0001] This application relates to the field of power grid technology, and in particular to a photovoltaic grid-connected inverter adaptation control method, device, computer equipment, computer-readable storage medium and computer program product based on hybrid control with adaptive switching of operating modes. Background Technology

[0002] Currently, in order to improve the adaptability and operational stability of the power grid, it is crucial to accurately adapt and control the photovoltaic grid-connected inverter.

[0003] In traditional technologies, fixed parameters and preset model control are generally used when adapting and controlling photovoltaic grid-connected inverters. However, this method is difficult to adapt to grids with different characteristics, resulting in low accuracy of adaptive control of photovoltaic grid-connected inverters. Summary of the Invention

[0004] Based on this, it is necessary to provide a photovoltaic grid-connected inverter adaptation control method, device, computer equipment, computer-readable storage medium, and computer program product that can improve the accuracy of adaptation control of photovoltaic grid-connected inverters based on hybrid control of operating mode adaptive switching, in order to address the above-mentioned technical problems.

[0005] Firstly, this application provides a photovoltaic grid-connected inverter adaptation control method based on hybrid control with adaptive switching of operating modes, including:

[0006] In response to the adaptation control request for the photovoltaic grid-connected inverter, the hardware status parameters of the photovoltaic grid-connected inverter and the effective output coefficient of the photovoltaic power station corresponding to the photovoltaic grid-connected inverter are obtained.

[0007] When the hardware status parameters meet the preset parameter constraints and the effective output coefficient is greater than or equal to the preset output coefficient, the first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station, and the second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter are obtained.

[0008] The first correlation parameter and the second correlation parameter are respectively input into multiple trained operation mode prediction models to obtain multiple predicted operation modes corresponding to the power grid to be analyzed, and the prediction probability corresponding to each predicted operation mode output by each trained operation mode prediction model.

[0009] For each predicted operating mode, the predicted probabilities corresponding to each predicted operating mode are summed according to the model weights of each trained operating mode prediction model to obtain the target predicted probability corresponding to the predicted operating mode output by each trained operating mode prediction model. From each predicted operating mode, the predicted operating mode with the highest target predicted probability is selected as the current operating mode of the power grid to be analyzed.

[0010] The historical operating modes of the power grid to be analyzed are obtained. Based on the current operating mode and the historical operating modes, the future operating mode of the power grid to be analyzed is determined. Based on the current operating mode, the historical operating mode and the future operating mode, the corresponding adaptation control command for the photovoltaic grid-connected inverter is determined. The photovoltaic grid-connected inverter is then subjected to corresponding adaptation control processing according to the adaptation control command.

[0011] In one embodiment, determining the adaptation control command corresponding to the photovoltaic grid-connected inverter based on the current operating mode, the historical operating mode, and the future operating mode includes:

[0012] The system obtains a first predicted probability corresponding to the historical operating mode and a second predicted probability corresponding to the future operating mode, and determines a first probability difference between the target predicted probability and the first predicted probability, and a second probability difference between the target predicted probability and the second predicted probability. The system also obtains a first inverter parameter of the photovoltaic grid-connected inverter in the current operating mode, a second inverter parameter in the historical operating mode, and a third inverter parameter in the future operating mode.

[0013] When the first probability difference is greater than the first preset probability difference and the second probability difference is greater than the second preset probability difference, the corresponding adaptation control command for the photovoltaic grid-connected inverter is determined based on the first inverter parameters.

[0014] When the first probability difference is less than or equal to the first preset probability difference and the second probability difference is greater than the second preset probability difference, the adaptation control command corresponding to the photovoltaic grid-connected inverter is determined according to the first inverter parameters and the second inverter parameters.

[0015] When the first probability difference is greater than the first preset probability difference and the second probability difference is less than or equal to the second preset probability difference, the adaptation control command corresponding to the photovoltaic grid-connected inverter is determined according to the first inverter parameters and the third inverter parameters.

[0016] When the first probability difference is less than or equal to the first preset probability difference, and the second probability difference is less than or equal to the second preset probability difference, the adaptation control command corresponding to the photovoltaic grid-connected inverter is determined based on the first inverter parameters, the second inverter parameters, and the third inverter parameters.

[0017] In one embodiment, determining the adaptation control command corresponding to the photovoltaic grid-connected inverter based on the first inverter parameters, the second inverter parameters, and the third inverter parameters includes:

[0018] The first feature vector of the first inverter parameter, the second feature vector of the second inverter parameter, and the third feature vector of the third inverter parameter are extracted respectively.

[0019] The first feature vector, the second feature vector, and the third feature vector are fused to obtain the fused feature vector corresponding to the photovoltaic grid-connected inverter.

[0020] The fused feature vector is input into the trained adaptive control command prediction model to obtain the adaptive control command corresponding to the photovoltaic grid-connected inverter.

[0021] In one embodiment, the trained adaptive control command prediction model includes a power control command prediction network, a modulation wave control command prediction network, and an auxiliary control command prediction network.

[0022] The step of inputting the fused feature vector into the trained adaptive control command prediction model to obtain the adaptive control command corresponding to the photovoltaic grid-connected inverter includes:

[0023] The fused feature vector is input into the power control command prediction network to obtain the power control command corresponding to the photovoltaic grid-connected inverter.

[0024] The fused feature vector is input into the modulation wave control command prediction network to obtain the modulation wave control command corresponding to the photovoltaic grid-connected inverter.

[0025] The fused feature vector is input into the auxiliary control command prediction network to obtain the auxiliary control command corresponding to the photovoltaic grid-connected inverter.

[0026] Based on the power control command, the modulation wave control command, and the auxiliary control command, the corresponding adaptation control command for the photovoltaic grid-connected inverter is obtained.

[0027] In one embodiment, obtaining the first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station, and the second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter, includes:

[0028] The power matching degree, response delay value, and voltage adaptation deviation between the photovoltaic grid-connected inverter and the photovoltaic power station are obtained, as well as the impedance matching degree, phase synchronization deviation value, and voltage improvement rate between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter.

[0029] Based on the power matching degree, the response delay value, and the voltage adaptation deviation, a first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station is obtained. Based on the impedance matching degree, the phase synchronization deviation value, and the voltage improvement rate, a second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter is obtained.

[0030] In one embodiment, the step of inputting the first correlation parameter and the second correlation parameter into multiple trained operation mode prediction models to obtain multiple predicted operation modes corresponding to the power grid to be analyzed, and the prediction probability corresponding to each predicted operation mode, output by each trained operation mode prediction model, includes:

[0031] Based on the power matching degree, the response delay value, and the voltage adaptation deviation, a fourth feature vector corresponding to the first correlation parameter is constructed; and based on the impedance matching degree, the phase synchronization deviation value, and the voltage improvement rate, a fifth feature vector corresponding to the second correlation parameter is constructed.

[0032] The fourth feature vector and the fifth feature vector are concatenated to obtain a concatenated feature vector;

[0033] The concatenated feature vector is input into multiple trained operation mode prediction models to obtain multiple predicted operation modes corresponding to the power grid to be analyzed, and the prediction probability corresponding to each predicted operation mode, as output by each trained operation mode prediction model.

[0034] Secondly, this application also provides a photovoltaic grid-connected inverter adaptation control device based on adaptive switching hybrid control of operating modes, comprising:

[0035] The first acquisition module is used to acquire the hardware status parameters of the photovoltaic grid-connected inverter and the effective output coefficient of the photovoltaic power station corresponding to the photovoltaic grid-connected inverter in response to the adaptation control request for the photovoltaic grid-connected inverter.

[0036] The second acquisition module is used to acquire, when the hardware status parameters meet the preset parameter constraints and the effective output coefficient is greater than or equal to the preset output coefficient, a first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station, and a second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter.

[0037] The mode prediction module is used to input the first correlation parameter and the second correlation parameter into multiple trained operation mode prediction models respectively, so as to obtain multiple predicted operation modes corresponding to the power grid to be analyzed output by each trained operation mode prediction model, and the prediction probability corresponding to each predicted operation mode.

[0038] The mode determination module is used to sum the prediction probabilities corresponding to the predicted operating mode output by each trained operating mode prediction model according to the model weights of each predicted operating mode prediction model, to obtain the target prediction probability corresponding to each predicted operating mode, and to select the predicted operating mode with the highest target prediction probability from each predicted operating mode as the current operating mode of the power grid to be analyzed.

[0039] The inverter control module is used to acquire the historical operating modes of the power grid to be analyzed, determine the future operating mode of the power grid to be analyzed based on the current operating mode and the historical operating modes, determine the corresponding adaptation control command for the photovoltaic grid-connected inverter based on the current operating mode, the historical operating mode and the future operating mode, and perform corresponding adaptation control processing on the photovoltaic grid-connected inverter according to the adaptation control command.

[0040] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0041] In response to the adaptation control request for the photovoltaic grid-connected inverter, the hardware status parameters of the photovoltaic grid-connected inverter and the effective output coefficient of the photovoltaic power station corresponding to the photovoltaic grid-connected inverter are obtained.

[0042] When the hardware status parameters meet the preset parameter constraints and the effective output coefficient is greater than or equal to the preset output coefficient, the first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station, and the second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter are obtained.

[0043] The first correlation parameter and the second correlation parameter are respectively input into multiple trained operation mode prediction models to obtain multiple predicted operation modes corresponding to the power grid to be analyzed, and the prediction probability corresponding to each predicted operation mode output by each trained operation mode prediction model.

[0044] For each predicted operating mode, according to the model weight of each trained operating mode prediction model, the predicted probabilities corresponding to the predicted operating mode output by each trained operating mode prediction model are summed to obtain the target predicted probability corresponding to each predicted operating mode. From each predicted operating mode, the predicted operating mode with the highest target predicted probability is selected as the current operating mode of the power grid to be analyzed.

[0045] The historical operating modes of the power grid to be analyzed are obtained. Based on the current operating mode and the historical operating modes, the future operating mode of the power grid to be analyzed is determined. Based on the current operating mode, the historical operating mode and the future operating mode, the corresponding adaptation control command for the photovoltaic grid-connected inverter is determined. The photovoltaic grid-connected inverter is then subjected to corresponding adaptation control processing according to the adaptation control command.

[0046] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0047] In response to the adaptation control request for the photovoltaic grid-connected inverter, the hardware status parameters of the photovoltaic grid-connected inverter and the effective output coefficient of the photovoltaic power station corresponding to the photovoltaic grid-connected inverter are obtained.

[0048] When the hardware status parameters meet the preset parameter constraints and the effective output coefficient is greater than or equal to the preset output coefficient, the first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station, and the second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter are obtained.

[0049] The first correlation parameter and the second correlation parameter are respectively input into multiple trained operation mode prediction models to obtain multiple predicted operation modes corresponding to the power grid to be analyzed, and the prediction probability corresponding to each predicted operation mode output by each trained operation mode prediction model.

[0050] For each predicted operating mode, according to the model weight of each trained operating mode prediction model, the predicted probabilities corresponding to the predicted operating mode output by each trained operating mode prediction model are summed to obtain the target predicted probability corresponding to each predicted operating mode. From each predicted operating mode, the predicted operating mode with the highest target predicted probability is selected as the current operating mode of the power grid to be analyzed.

[0051] The historical operating modes of the power grid to be analyzed are obtained. Based on the current operating mode and the historical operating modes, the future operating mode of the power grid to be analyzed is determined. Based on the current operating mode, the historical operating mode and the future operating mode, the corresponding adaptation control command for the photovoltaic grid-connected inverter is determined. The photovoltaic grid-connected inverter is then subjected to corresponding adaptation control processing according to the adaptation control command.

[0052] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0053] In response to the adaptation control request for the photovoltaic grid-connected inverter, the hardware status parameters of the photovoltaic grid-connected inverter and the effective output coefficient of the photovoltaic power station corresponding to the photovoltaic grid-connected inverter are obtained.

[0054] When the hardware status parameters meet the preset parameter constraints and the effective output coefficient is greater than or equal to the preset output coefficient, the first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station, and the second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter are obtained.

[0055] The first correlation parameter and the second correlation parameter are respectively input into multiple trained operation mode prediction models to obtain multiple predicted operation modes corresponding to the power grid to be analyzed, and the prediction probability corresponding to each predicted operation mode output by each trained operation mode prediction model.

[0056] For each predicted operating mode, the predicted probabilities corresponding to each predicted operating mode are summed according to the model weights of each trained operating mode prediction model to obtain the target predicted probability corresponding to the predicted operating mode output by each trained operating mode prediction model. From each predicted operating mode, the predicted operating mode with the highest target predicted probability is selected as the current operating mode of the power grid to be analyzed.

[0057] The historical operating modes of the power grid to be analyzed are obtained. Based on the current operating mode and the historical operating modes, the future operating mode of the power grid to be analyzed is determined. Based on the current operating mode, the historical operating mode and the future operating mode, the corresponding adaptation control command for the photovoltaic grid-connected inverter is determined. The photovoltaic grid-connected inverter is then subjected to corresponding adaptation control processing according to the adaptation control command.

[0058] The aforementioned photovoltaic grid-connected inverter adaptation control method, device, computer equipment, storage medium, and computer program product based on adaptive switching hybrid control of operating modes, responds to an adaptation control request for the photovoltaic grid-connected inverter by acquiring the hardware status parameters of the photovoltaic grid-connected inverter and the effective output coefficient of the photovoltaic power station corresponding to the photovoltaic grid-connected inverter. If the hardware status parameters meet preset parameter constraints and the effective output coefficient is greater than or equal to the preset output coefficient, it acquires a first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station, and a second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter. Then, it inputs the first and second correlation parameters into multiple trained operating mode prediction models, respectively, to obtain multiple predictions of the grid to be analyzed output by each trained operating mode prediction model. The system first identifies the operating modes and the prediction probabilities corresponding to each predicted operating mode. Then, for each predicted operating mode, the prediction probabilities corresponding to each predicted operating mode are summed according to the model weights of each trained operating mode prediction model to obtain the target prediction probability corresponding to the predicted operating mode output by each trained operating mode prediction model. From each predicted operating mode, the predicted operating mode with the highest target prediction probability is selected as the current operating mode of the power grid to be analyzed. Finally, the historical operating modes of the power grid to be analyzed are obtained. Based on the current operating mode and the historical operating modes, the future operating mode of the power grid to be analyzed is determined. Based on the current operating mode, the historical operating mode, and the future operating mode, the corresponding adaptation control command for the photovoltaic grid-connected inverter is determined, and the photovoltaic grid-connected inverter is subjected to corresponding adaptation control processing according to the adaptation control command. In this way, when performing adaptive control of photovoltaic grid-connected inverters, the reliability of the control premise is ensured by first verifying the hardware status and the effective output coefficient of the photovoltaic power station. Then, based on the correlation parameters between the inverter and the power station, and between the inverter and the grid, multiple trained operation mode prediction models are used to make probabilistic predictions. The optimal current grid operation mode is obtained by weighted fusion according to the model weights. Furthermore, the adaptive control command is determined by combining historical operation modes and future operation modes. This approach breaks away from the limitations of traditional fixed parameters and preset models, and can accurately match different operating conditions and grids with different characteristics, which is conducive to improving the accuracy of adaptive control of photovoltaic grid-connected inverters. Attached Figure Description

[0059] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0060] Figure 1 This is a flowchart illustrating a photovoltaic grid-connected inverter adaptation control method based on hybrid control with adaptive switching of operating modes in one embodiment.

[0061] Figure 2 This is a flowchart illustrating a photovoltaic grid-connected inverter adaptation control method based on hybrid control with adaptive switching of operating modes, as described in another embodiment.

[0062] Figure 3 This is a schematic diagram of the main circuit of a photovoltaic grid-connected inverter in one embodiment;

[0063] Figure 4 This is a schematic diagram illustrating the principle of the network control strategy in one embodiment;

[0064] Figure 5 This is a schematic diagram illustrating the principle of network control strategy in one embodiment;

[0065] Figure 6 This is a schematic diagram of the state switching controller of a photovoltaic grid-connected inverter in one embodiment;

[0066] Figure 7 This is a schematic diagram of the active and reactive current decoupling controller of a photovoltaic grid-connected inverter in one embodiment;

[0067] Figure 8 This is a schematic diagram of the process for generating active and reactive current commands for grid-connected control of a photovoltaic grid-connected inverter in one embodiment.

[0068] Figure 9 This is a schematic diagram illustrating the process of generating active and reactive current commands for grid-connected photovoltaic inverters in one embodiment.

[0069] Figure 10 This is a schematic diagram of the grid connection state switching control of a photovoltaic grid-connected inverter in one embodiment;

[0070] Figure 11 This is a schematic diagram of the decoupling control process of active and reactive current in a photovoltaic grid-connected inverter in one embodiment.

[0071] Figure 12This is a schematic diagram of the SVPWM (Space Vector Pulse Width Modulation) modulation method of a photovoltaic grid-connected inverter in one embodiment;

[0072] Figure 13 This is a schematic diagram of the SVPWM modulation process of a photovoltaic grid-connected inverter in one embodiment;

[0073] Figure 14 This is a schematic diagram of the voltage waveform during grid-connected control of a photovoltaic grid-connected inverter in one embodiment;

[0074] Figure 15 This is a schematic diagram of the frequency waveform during grid-connected control of a photovoltaic grid-connected inverter in one embodiment;

[0075] Figure 16 This is a schematic diagram of the active current setpoint waveform during grid-connected control of a photovoltaic grid-connected inverter in one embodiment;

[0076] Figure 17 This is a schematic diagram of the reactive current setpoint waveform during grid-connected control of a photovoltaic grid-connected inverter in one embodiment;

[0077] Figure 18 This is a schematic diagram of the voltage waveform during grid connection control of a photovoltaic grid-connected inverter in one embodiment;

[0078] Figure 19 This is a schematic diagram of the frequency waveform during grid connection control of a photovoltaic grid-connected inverter in one embodiment;

[0079] Figure 20 This is a schematic diagram of the active current setpoint waveform during grid connection control of a photovoltaic grid-connected inverter in one embodiment;

[0080] Figure 21 This is a schematic diagram of the reactive current setpoint waveform during grid connection control of a photovoltaic grid-connected inverter in one embodiment;

[0081] Figure 22 This is a schematic diagram of the voltage waveform during grid-connected inverter switching control in one embodiment;

[0082] Figure 23 This is a schematic diagram of the frequency waveform during grid-connected inverter switching control in one embodiment;

[0083] Figure 24 This is a schematic diagram of the active current setpoint waveform during grid-connected inverter grid-connected switching control in one embodiment;

[0084] Figure 25This is a schematic diagram of the reactive current setpoint waveform during grid-connected inverter grid-connected switching control in one embodiment;

[0085] Figure 26 This is a schematic diagram of the three-phase SPWM (Sinusoidal Pulse Width Modulation) modulation wave of a photovoltaic grid-connected inverter in one embodiment;

[0086] Figure 27 This is a schematic diagram illustrating the increase in the modulation wave of a photovoltaic grid-connected inverter in one embodiment.

[0087] Figure 28 This is a schematic diagram of the three-phase SVPWM modulation waveform of a photovoltaic grid-connected inverter in one embodiment;

[0088] Figure 29 This is a structural block diagram of a photovoltaic grid-connected inverter adaptation control device based on hybrid control with adaptive switching of operating modes in one embodiment.

[0089] Figure 30 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0090] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0091] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0092] In one exemplary embodiment, such as Figure 1 As shown, a photovoltaic grid-connected inverter adaptation control method based on adaptive switching hybrid control of operating modes is provided. This embodiment illustrates the application of this method to a server; it is understood that this method can also be applied to a terminal, or to a system including a terminal and a server, and is implemented through interaction between the terminal and the server. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, and tablets; the server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. In this embodiment, the method includes the following steps:

[0093] Step S101: In response to the adaptation control request for the photovoltaic grid-connected inverter, obtain the hardware status parameters of the photovoltaic grid-connected inverter and the effective output coefficient of the photovoltaic power station corresponding to the photovoltaic grid-connected inverter.

[0094] In this context, a photovoltaic (PV) grid-connected inverter refers to a power electronic conversion device used to achieve energy exchange and coordinated control between a PV power plant and the power grid under analysis. An adaptive control request is a request used to trigger adaptive grid-connected control of the PV grid-connected inverter. Hardware status parameters represent the hardware characteristics and operating conditions of the PV grid-connected inverter, including but not limited to rated capacity, grid-connected inductance, switching frequency, operating temperature, device health status, filtering parameters, and real-time operating conditions. A PV power plant refers to a new energy power generation system that converts solar energy into electrical energy and connects to the grid for power generation, consisting of a PV module array, combiner box, DC distribution unit, and PV grid-connected inverter. The effective output coefficient represents the ratio between the current actual output power and the theoretical maximum output power of the PV power plant.

[0095] For example, the server receives the adaptation control request for the photovoltaic grid-connected inverter sent by the terminal through the network path between the server and the terminal; then, it performs a reasonableness verification on the adaptation control request and obtains the verification result of the adaptation control request; if the verification result indicates that the adaptation control request has passed the verification, in response to the adaptation control request for the photovoltaic grid-connected inverter, it collects the hardware status parameters of the photovoltaic grid-connected inverter and the effective output coefficient of the photovoltaic power station corresponding to the photovoltaic grid-connected inverter in real time.

[0096] Step S102: If the hardware status parameters meet the preset parameter constraints and the effective output coefficient is greater than or equal to the preset output coefficient, obtain the first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station, and the second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter.

[0097] The preset parameter constraints refer to the pre-set hardware normal operation judgment conditions for the photovoltaic grid-connected inverter, including rated capacity range, grid-connected inductance threshold, operating temperature range, and switching device health status. The preset output coefficient refers to the pre-set output threshold used to determine whether the photovoltaic power station's power generation capacity meets the stable operation requirements. The first correlation parameter is used to characterize the electrical matching relationship between the photovoltaic grid-connected inverter and the photovoltaic power station. The grid to be analyzed refers to the grid that is connected to the photovoltaic grid-connected inverter and requires operation mode identification and stability control, such as a regional grid or public grid. The second correlation parameter is used to characterize the grid-connected interaction characteristics between the photovoltaic grid-connected inverter and the grid to be analyzed.

[0098] For example, the server determines preset parameter constraints by comprehensively analyzing and calibrating the rated operating parameters, historical operating data, and device characteristic curves of the photovoltaic grid-connected inverter. Then, based on the installed capacity of the photovoltaic power station, component degradation characteristics, power generation efficiency under preset weather conditions, and the minimum allowable stable output threshold of the photovoltaic grid-connected inverter, an initial output coefficient is determined. This initial output coefficient is then corrected based on seasonal variation information, solar radiation distribution information, and system loss information corresponding to the photovoltaic power station, resulting in a preset output coefficient. Next, based on the preset parameter constraints and the preset output coefficient, the hardware status parameters and effective output coefficient are determined. If the hardware status parameters meet the preset parameter constraints and the effective output coefficient is greater than or equal to the preset output coefficient, the server obtains the first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station, and the second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter.

[0099] Step S103: Input the first correlation parameter and the second correlation parameter into multiple trained operation mode prediction models respectively to obtain multiple predicted operation modes corresponding to the power grid to be analyzed, and the prediction probability corresponding to each predicted operation mode output by each trained operation mode prediction model.

[0100] The operation mode prediction model refers to a network model, such as a recurrent neural network model, that can obtain the predicted operation mode of the power grid to be analyzed using the first and second correlation parameters. The predicted operation mode refers to the operation mode of the power grid to be analyzed predicted by the trained operation mode prediction model, including grid-following mode or grid-connecting mode. The prediction probability refers to the likelihood that the trained operation mode prediction model correctly determines the predicted operation mode.

[0101] For example, the server inputs a first correlation parameter and a second correlation parameter into an importance prediction model to obtain a first importance of the first correlation parameter and a second importance of the second correlation parameter. When the first importance is greater than the second importance, the first correlation parameter is used as primary data and the second correlation parameter as auxiliary data, and input into multiple trained operation mode prediction models to obtain multiple predicted operation modes corresponding to the power grid to be analyzed, and the prediction probability corresponding to each predicted operation mode, output by each trained operation mode prediction model. When the first importance is less than the second importance, the second correlation parameter is used as primary data and the first correlation parameter as auxiliary data, and input into multiple trained operation mode prediction models to obtain multiple predicted operation modes corresponding to the power grid to be analyzed, and the prediction probability corresponding to each predicted operation mode, output by each trained operation mode prediction model. When the first importance is equal to the second importance, the first correlation parameter and the second correlation parameter are fused to obtain a fused correlation parameter, and the fused correlation parameter is input into multiple trained operation mode prediction models to obtain multiple predicted operation modes corresponding to the power grid to be analyzed, and the prediction probability corresponding to each predicted operation mode, output by each trained operation mode prediction model.

[0102] Step S104: For each predicted operating mode, sum the predicted probabilities corresponding to the predicted operating mode output by each trained operating mode prediction model according to the model weights of each trained operating mode prediction model to obtain the target prediction probability corresponding to each predicted operating mode. Then, select the predicted operating mode with the highest target prediction probability from each predicted operating mode as the current operating mode of the power grid to be analyzed.

[0103] Here, model weights refer to the decision weight coefficients pre-assigned to each trained prediction model for each operating mode. The target prediction probability represents the final overall prediction probability determined for each predicted operating mode. The current operating mode refers to the predicted operating mode with the highest target prediction probability, including either the following network mode or the network construction mode.

[0104] For example, the server determines the model weight of each trained operation mode prediction model based on the prediction accuracy of each trained operation mode prediction model; then, for each predicted operation mode, the prediction probability corresponding to the predicted operation mode output by each trained operation mode prediction model is summed according to the model weight of each trained operation mode prediction model to obtain the target prediction probability corresponding to each predicted operation mode; next, from each predicted operation mode, the predicted operation mode with the highest target prediction probability is selected, and this predicted operation mode is taken as the current operation mode of the power grid to be analyzed.

[0105] Step S105: Obtain the historical operating mode of the power grid to be analyzed; determine the future operating mode of the power grid to be analyzed based on the current operating mode and the historical operating mode; determine the corresponding adaptation control command for the photovoltaic grid-connected inverter based on the current operating mode, the historical operating mode and the future operating mode; and perform corresponding adaptation control processing on the photovoltaic grid-connected inverter according to the adaptation control command.

[0106] The historical operating mode refers to the actual operating mode of the power grid under analysis within a preset time period prior to the current moment, including grid-following mode or grid-connected mode. The future operating mode refers to the estimated operating mode of the power grid under analysis within a future period, including grid-following mode or grid-connected mode. The adaptive control command refers to the adaptive control command corresponding to the photovoltaic grid-connected inverter.

[0107] For example, the server retrieves the historical operating modes of the power grid to be analyzed from the database; then, based on the current operating mode and the historical operating modes, it determines the transition probability corresponding to each preset operating mode of the power grid to be analyzed within a future period; next, based on the hardware status parameters corresponding to the photovoltaic grid-connected inverter and the effective output coefficient of the photovoltaic power station corresponding to the photovoltaic grid-connected inverter, it corrects the transition probability corresponding to each preset operating mode of the power grid to be analyzed within a future period, obtaining the corrected transition probability corresponding to each preset operating mode of the power grid to be analyzed within a future period; then, from each preset operating mode, the preset operating mode with the highest corrected transition probability is selected as the future operating mode of the power grid to be analyzed; then, based on the current operating mode, the historical operating mode, and the future operating mode, the corresponding adaptation control command for the photovoltaic grid-connected inverter is determined; finally, according to the adaptation control command, the photovoltaic grid-connected inverter is subjected to the corresponding adaptation control processing.

[0108] In the aforementioned photovoltaic grid-connected inverter adaptation control method based on hybrid control with adaptive switching of operating modes, in response to an adaptation control request for the photovoltaic grid-connected inverter, the hardware state parameters of the photovoltaic grid-connected inverter and the effective output coefficient of the photovoltaic power station corresponding to the photovoltaic grid-connected inverter are obtained. If the hardware state parameters meet preset parameter constraints and the effective output coefficient is greater than or equal to the preset output coefficient, a first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station, and a second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter are obtained. Then, the first and second correlation parameters are respectively input into multiple trained operating mode prediction models to obtain multiple predicted operating modes corresponding to the grid to be analyzed output by each trained operating mode prediction model, and each predicted operating mode... The system calculates the predicted probability corresponding to each operating mode. Then, for each predicted operating mode, it sums the predicted probabilities output by each trained operating mode prediction model according to the model weights of each model, obtaining the target predicted probability for each predicted operating mode. From each predicted operating mode, the predicted operating mode with the highest target predicted probability is selected as the current operating mode of the power grid to be analyzed. Finally, it obtains the historical operating modes of the power grid to be analyzed, determines the future operating mode of the power grid to be analyzed based on the current operating mode and the historical operating mode, determines the corresponding adaptation control command for the photovoltaic grid-connected inverter based on the current operating mode, the historical operating mode, and the future operating mode, and performs corresponding adaptation control processing on the photovoltaic grid-connected inverter according to the adaptation control command. In this way, when performing adaptive control of photovoltaic grid-connected inverters, the reliability of the control premise is ensured by first verifying the hardware status and the effective output coefficient of the photovoltaic power station. Then, based on the correlation parameters between the inverter and the power station, and between the inverter and the grid, multiple trained operation mode prediction models are used to make probabilistic predictions. The optimal current grid operation mode is obtained by weighted fusion according to the model weights. Furthermore, the adaptive control command is determined by combining historical operation modes and future operation modes. This approach breaks away from the limitations of traditional fixed parameters and preset models, and can accurately match different operating conditions and grids with different characteristics, which is conducive to improving the accuracy of adaptive control of photovoltaic grid-connected inverters.

[0109] In an exemplary embodiment, step S105, which determines the adaptation control command corresponding to the photovoltaic grid-connected inverter based on the current operating mode, historical operating mode, and future operating mode, specifically includes the following: obtaining the first predicted probability corresponding to the historical operating mode and the second predicted probability corresponding to the future operating mode, and determining the first probability difference between the target predicted probability and the first predicted probability, and the second probability difference between the target predicted probability and the second predicted probability; obtaining the first inverter parameters of the photovoltaic grid-connected inverter in the current operating mode, the second inverter parameters in the historical operating mode, and the third inverter parameters in the future operating mode; and, if the first probability difference is greater than the first preset probability difference and the second probability difference is greater than the second preset probability difference, determining the adaptation control command corresponding to the photovoltaic grid-connected inverter based on the first inverter parameters. The system determines the appropriate control command for the photovoltaic grid-connected inverter. When the first probability difference is less than or equal to the first preset probability difference and the second probability difference is greater than the second preset probability difference, the system determines the appropriate control command for the photovoltaic grid-connected inverter based on the first inverter parameters and the second inverter parameters. When the first probability difference is greater than the first preset probability difference and the second probability difference is less than or equal to the second preset probability difference, the system determines the appropriate control command for the photovoltaic grid-connected inverter based on the first inverter parameters and the third inverter parameters. When the first probability difference is less than or equal to the first preset probability difference and the second probability difference is less than or equal to the second preset probability difference, the system determines the appropriate control command for the photovoltaic grid-connected inverter based on the first inverter parameters, the second inverter parameters, and the third inverter parameters.

[0110] The first predicted probability represents the reliability of the historical operating mode. The second predicted probability represents the reliability of the future operating mode. The first probability difference is the absolute difference between the target predicted probability and the first predicted probability. The second probability difference is the absolute difference between the target predicted probability and the second predicted probability. The first inverter parameters refer to the inverter parameters of the photovoltaic grid-connected inverter in the current operating mode, including but not limited to grid-connected power, voltage regulation coefficient, current control parameters, phase-locked loop parameters, and harmonic suppression parameters. The second inverter parameters refer to the inverter parameters of the photovoltaic grid-connected inverter in the historical operating mode, including but not limited to grid-connected power, voltage regulation coefficient, current control parameters, phase-locked loop parameters, and harmonic suppression parameters. The third inverter parameters refer to the inverter parameters of the photovoltaic grid-connected inverter in the future operating mode, including but not limited to grid-connected power, voltage regulation coefficient, current control parameters, phase-locked loop parameters, and harmonic suppression parameters. The first preset probability difference is a pre-set threshold used to determine whether the reliability difference between the current operating mode and the historical operating mode is significant. The second preset probability difference refers to a pre-set threshold used to determine whether the difference in credibility between the current operating mode and the future operating mode is significant. It should be noted that the first preset probability difference and the second preset probability difference can be the same or different.

[0111] For example, the server retrieves the first predicted probability corresponding to the historical operating mode from the database, and performs a weighted summation of the target predicted probability and the first predicted probability to obtain the second predicted probability corresponding to the future operating mode. Next, it subtracts the target predicted probability from the first predicted probability to obtain a first probability difference, and subtracts the target predicted probability from the second predicted probability to obtain a second probability difference. Then, it obtains the first inverter parameters of the photovoltaic grid-connected inverter in the current operating mode, the second inverter parameters in the historical operating modes, and the third inverter parameters in the future operating mode. Finally, if the first probability difference is greater than a first preset probability difference and the second probability difference is greater than a second preset probability difference, the first inverter parameters are input into the trained adaptive control command prediction model to obtain the photovoltaic grid-connected inverter. The corresponding adaptive control commands are obtained as follows: When the first probability difference is less than or equal to the first preset probability difference, and the second probability difference is greater than the second preset probability difference, the first inverter parameters and the second inverter parameters are input into the trained adaptive control command prediction model to obtain the adaptive control command corresponding to the photovoltaic grid-connected inverter; when the first probability difference is greater than the first preset probability difference, and the second probability difference is less than or equal to the second preset probability difference, the first inverter parameters and the third inverter parameters are input into the trained adaptive control command prediction model to obtain the adaptive control command corresponding to the photovoltaic grid-connected inverter; when the first probability difference is less than or equal to the first preset probability difference, and the second probability difference is less than or equal to the second preset probability difference, the first inverter parameters, the second inverter parameters, and the third inverter parameters are input into the trained adaptive control command prediction model to obtain the adaptive control command corresponding to the photovoltaic grid-connected inverter.

[0112] In this embodiment, by introducing the first predicted probability of the historical operating mode and the second predicted probability of the future operating mode, and by comparing the probability difference with the preset threshold, the reliability difference of the current, historical and future operating modes is dynamically determined, and then the inverter parameters in the corresponding mode are selected in a differentiated manner to determine the adaptation control command, which is beneficial to improving the reliability of the photovoltaic grid-connected inverter adaptation control.

[0113] In an exemplary embodiment, the adaptive control command corresponding to the photovoltaic grid-connected inverter is determined based on the first inverter parameters, the second inverter parameters, and the third inverter parameters. Specifically, this includes: extracting the first feature vector of the first inverter parameters, the second feature vector of the second inverter parameters, and the third feature vector of the third inverter parameters; fusing the first, second, and third feature vectors to obtain a fused feature vector corresponding to the photovoltaic grid-connected inverter; and inputting the fused feature vector into the trained adaptive control command prediction model to obtain the adaptive control command corresponding to the photovoltaic grid-connected inverter.

[0114] Here, the first feature vector refers to the representation vector corresponding to the first inverter parameter. The second feature vector refers to the representation vector corresponding to the second inverter parameter. The third feature vector refers to the representation vector corresponding to the third inverter parameter. The fused feature vector is a unified vector obtained by integrating the first, second, and third feature vectors according to a preset fusion rule (such as weighted concatenation, feature cross-validation, or dimensionality fusion). The adaptive control command prediction model refers to a network model, such as a random forest model, used to predict the adaptive control commands corresponding to the photovoltaic grid-connected inverter.

[0115] For example, the server uses the first inverter parameters as primary data and the second and third inverter parameters as auxiliary data, inputting them into a feature extraction model for feature extraction processing to obtain a first feature vector of the first inverter parameters. Next, the second inverter parameters are used as primary data, with the first and third inverter parameters used as auxiliary data, inputting them into the feature extraction model for feature extraction processing to obtain a second feature vector of the second inverter parameters. Finally, the third inverter parameters are used as primary data, with the first and second inverter parameters used as auxiliary data, inputting them into the feature extraction model for feature extraction processing. The process begins by obtaining the third eigenvector of the third inverter parameters. Then, the first, second, and third eigenvectors are input into the attention mechanism model to obtain the first weight corresponding to the first eigenvector, the second weight corresponding to the second eigenvector, and the third weight corresponding to the third eigenvector. Next, the first, second, and third eigenvectors are summed according to their respective weights to obtain the fused eigenvector corresponding to the photovoltaic grid-connected inverter. Finally, the fused eigenvector is input into the trained adaptive control command prediction model to obtain the adaptive control command corresponding to the photovoltaic grid-connected inverter.

[0116] In this embodiment, by extracting the feature vectors corresponding to the current, historical and future inverter parameters respectively, and performing multi-dimensional feature fusion, it is possible to fully explore the continuous change pattern of the grid operation mode in the time series and the inverter control characteristics. Then, by using the trained adaptive control command prediction model to intelligently generate control commands, it is beneficial to improve the prediction accuracy of the adaptive control commands corresponding to the photovoltaic grid-connected inverter.

[0117] In an exemplary embodiment, the trained adaptive control command prediction model includes a power control command prediction network, a modulation wave control command prediction network, and an auxiliary control command prediction network.

[0118] Then, the fused feature vector is input into the trained adaptive control command prediction model to obtain the adaptive control command corresponding to the photovoltaic grid-connected inverter. Specifically, this includes the following: inputting the fused feature vector into the power control command prediction network to obtain the power control command corresponding to the photovoltaic grid-connected inverter; inputting the fused feature vector into the modulation wave control command prediction network to obtain the modulation wave control command corresponding to the photovoltaic grid-connected inverter; inputting the fused feature vector into the auxiliary control command prediction network to obtain the auxiliary control command corresponding to the photovoltaic grid-connected inverter; and obtaining the adaptive control command corresponding to the photovoltaic grid-connected inverter based on the power control command, modulation wave control command, and auxiliary control command.

[0119] The power control command prediction network refers to the network in the trained adaptive control command prediction model used to predict the power control commands corresponding to the photovoltaic grid-connected inverter. The modulation wave control command prediction network is the network in the trained adaptive control command prediction model used to predict the modulation wave control commands corresponding to the photovoltaic grid-connected inverter. The auxiliary control command prediction network is the network in the trained adaptive control command prediction model used to predict the auxiliary control commands corresponding to the photovoltaic grid-connected inverter. Power control commands refer to commands used to regulate the power output characteristics of the photovoltaic grid-connected inverter (including active power setpoint, reactive power compensation value, power ramp rate limit, grid power factor setting, etc.). Modulation wave control commands refer to commands used to adjust the waveform output characteristics of the photovoltaic grid-connected inverter (including carrier frequency, modulation ratio, dead time, phase compensation value, etc.). Auxiliary control commands refer to commands used to adjust the operating characteristics of the photovoltaic grid-connected inverter (including temperature protection threshold adjustment, filter parameter optimization, harmonic mitigation strategy switching, grid fault ride-through parameter setting, etc.).

[0120] For example, the server performs further feature extraction on the fused feature vector to obtain a processed fused feature vector. Then, the processed fused feature vector is input into the power control command prediction network to obtain the power control command corresponding to the photovoltaic grid-connected inverter. Next, the processed fused feature vector is input into the modulation wave control command prediction network to obtain the modulation wave control command corresponding to the photovoltaic grid-connected inverter. Then, the processed fused feature vector is input into the auxiliary control command prediction network to obtain the auxiliary control command corresponding to the photovoltaic grid-connected inverter. Finally, the power control command, modulation wave control command, and auxiliary control command are integrated to obtain the adaptive control command corresponding to the photovoltaic grid-connected inverter.

[0121] In this embodiment, by inputting the fused feature vector into the power control command prediction network, the modulation wave control command prediction network, and the auxiliary control command prediction network respectively, corresponding control commands are generated respectively, and then the final adaptive control command is obtained by combining them. This realizes the multi-dimensional generation of control commands, thereby improving the integrity and stability of the adaptive control of the photovoltaic grid-connected inverter.

[0122] In an exemplary embodiment, step S102, which involves obtaining the first correlation parameters between the photovoltaic grid-connected inverter and the photovoltaic power station, and the second correlation parameters between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter, specifically includes the following: obtaining the power matching degree, response delay value, and voltage adaptation deviation between the photovoltaic grid-connected inverter and the photovoltaic power station, and the impedance matching degree, phase synchronization deviation value, and voltage improvement rate between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter; obtaining the first correlation parameters between the photovoltaic grid-connected inverter and the photovoltaic power station based on the power matching degree, response delay value, and voltage adaptation deviation, and obtaining the second correlation parameters between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter based on the impedance matching degree, phase synchronization deviation value, and voltage improvement rate.

[0123] The power matching degree characterizes the degree of power transmission matching between the photovoltaic grid-connected inverter and the photovoltaic power station, for example, 0.96. The response delay value refers to the time required for the photovoltaic grid-connected inverter to complete power regulation and reach a stable state when the output power of the photovoltaic power station changes. The voltage adaptation deviation is the difference between the actual voltage on the photovoltaic power station side and the inverter's rated operating voltage, used to assess whether the DC-side voltage is within a safe and efficient operating range, for example, 12 volts. The impedance matching degree characterizes the degree of impedance matching between the photovoltaic grid-connected inverter and the grid under analysis, for example, 0.92. The phase synchronization deviation value is the difference between the phase of the output voltage of the photovoltaic grid-connected inverter and the phase of the grid voltage, for example, 2.5 electrical degrees. The voltage improvement rate is the degree to which voltage distortion, voltage fluctuation, or voltage deviation is improved at the grid connection point in the grid under analysis after the photovoltaic grid-connected inverter is connected, for example, 88%.

[0124] For example, the server collects real-time operating data between the photovoltaic grid-connected inverter and the photovoltaic power station, and calculates the operating data to obtain the power matching degree, which characterizes the DC-side energy interaction state, the response delay value, which characterizes the dynamic adjustment speed, and the voltage adaptation deviation, which characterizes the voltage adaptation level. The above three indicators are normalized and feature fusion processed to form a first correlation parameter that can comprehensively reflect the collaborative operation characteristics of the inverter and the photovoltaic power station. At the same time, the impedance matching degree, phase synchronization deviation value, and voltage improvement rate between the inverter and the grid to be analyzed are collected and calculated. The three sets of grid-side characteristic indicators are standardized and feature integrated to form a second correlation parameter that can accurately characterize the grid-connected interaction characteristics and grid support capabilities.

[0125] In this embodiment, by collecting the power matching degree, response delay value, and voltage adaptation deviation between the photovoltaic grid-connected inverter and the photovoltaic power station, as well as the impedance matching degree, phase synchronization deviation value, and voltage improvement rate between the inverter and the grid, the operating status of the grid-connected system can be comprehensively and multidimensionally reflected, providing accurate and reliable data support for subsequent operation mode prediction and adaptation control command generation.

[0126] In an exemplary embodiment, step S103 above, which involves inputting the first correlation parameter and the second correlation parameter into multiple trained operation mode prediction models to obtain multiple predicted operation modes corresponding to the power grid to be analyzed output by each trained operation mode prediction model, and the prediction probability corresponding to each predicted operation mode, specifically includes the following: constructing a fourth feature vector corresponding to the first correlation parameter based on power matching degree, response delay value, and voltage adaptation deviation, and constructing a fifth feature vector corresponding to the second correlation parameter based on impedance matching degree, phase synchronization deviation value, and voltage improvement rate; concatenating the fourth feature vector and the fifth feature vector to obtain a concatenated feature vector; and inputting the concatenated feature vector into multiple trained operation mode prediction models to obtain multiple predicted operation modes corresponding to the power grid to be analyzed output by each trained operation mode prediction model, and the prediction probability corresponding to each predicted operation mode.

[0127] The fourth eigenvector is a feature vector constructed based on power matching degree, response delay value, and voltage adaptation deviation. The fifth eigenvector is a feature vector constructed based on impedance matching degree, phase synchronization deviation value, and voltage improvement rate. The concatenated feature vector is obtained by directly concatenating or weighting the fourth and fifth eigenvectors according to their dimensional order.

[0128] For example, the server performs feature encoding processing on power matching degree, response delay value, and voltage adaptation deviation respectively to obtain the feature encoding results of power matching degree, response delay value, and voltage adaptation deviation. These feature encoding results are then concatenated to obtain the fourth feature vector corresponding to the first correlation parameter. Simultaneously, feature encoding processing is performed on impedance matching degree, phase synchronization deviation value, and voltage improvement rate respectively to obtain the feature encoding results of impedance matching degree, phase synchronization deviation value, and voltage improvement rate. These feature encoding results are then concatenated to obtain the fifth feature vector corresponding to the second correlation parameter. Next, the fourth and fifth feature vectors are concatenated according to a preset concatenation order to obtain a concatenated feature vector. Then, the concatenated feature vector undergoes further feature extraction processing to obtain a processed concatenated feature vector. This processed concatenated feature vector is then input into multiple trained operation mode prediction models to obtain multiple predicted operation modes corresponding to the power grid to be analyzed, and the prediction probability corresponding to each predicted operation mode, output by each trained operation mode prediction model.

[0129] In this embodiment, two types of feature vectors are concatenated to form a concatenated feature vector containing comprehensive operational information from both the photovoltaic and grid sides. This concatenated feature vector serves as the input for multiple operational mode prediction models. This approach fully integrates the key state characteristics of both the photovoltaic power plant and the grid, making the predicted operational modes and corresponding probabilities output by each prediction model more accurate and reliable. This provides a solid data foundation for subsequent grid mode judgment and inverter adaptive control.

[0130] In one exemplary embodiment, such as Figure 2 As shown, another photovoltaic grid-connected inverter adaptation control method based on adaptive switching hybrid control of operating mode is provided. Taking the application of this method to a server as an example, the specific steps include:

[0131] Step S201: In response to the adaptation control request for the photovoltaic grid-connected inverter, obtain the hardware status parameters of the photovoltaic grid-connected inverter and the effective output coefficient of the photovoltaic power station corresponding to the photovoltaic grid-connected inverter.

[0132] Step S202: Under the condition that the hardware status parameters meet the preset parameter constraints and the effective output coefficient is greater than or equal to the preset output coefficient, obtain the power matching degree, response delay value and voltage adaptation deviation between the photovoltaic grid-connected inverter and the photovoltaic power station, as well as the impedance matching degree, phase synchronization deviation value and voltage improvement rate between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter.

[0133] Step S203: Based on power matching degree, response delay value and voltage adaptation deviation, the first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station is obtained, and based on impedance matching degree, phase synchronization deviation value and voltage improvement rate, the second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter is obtained.

[0134] Step S204: Input the first correlation parameter and the second correlation parameter into multiple trained operation mode prediction models respectively to obtain multiple predicted operation modes corresponding to the power grid to be analyzed, and the prediction probability corresponding to each predicted operation mode output by each trained operation mode prediction model.

[0135] Step S205: For each predicted operating mode, sum the predicted probabilities corresponding to the predicted operating mode output by each trained operating mode prediction model according to the model weights of each trained operating mode prediction model to obtain the target prediction probability corresponding to each predicted operating mode. Then, select the predicted operating mode with the highest target prediction probability from each predicted operating mode as the current operating mode of the power grid to be analyzed.

[0136] Step S206: Obtain the historical operating modes of the power grid to be analyzed, and determine the future operating modes of the power grid to be analyzed based on the current operating modes and historical operating modes.

[0137] Step S207: Obtain the first predicted probability corresponding to the historical operating mode and the second predicted probability corresponding to the future operating mode, and determine the first probability difference between the target predicted probability and the first predicted probability, and the second probability difference between the target predicted probability and the second predicted probability, and obtain the first inverter parameters of the photovoltaic grid-connected inverter in the current operating mode, the second inverter parameters in the historical operating mode, and the third inverter parameters in the future operating mode.

[0138] In step S208, if the first probability difference is greater than the first preset probability difference and the second probability difference is greater than the second preset probability difference, the corresponding adaptation control command for the photovoltaic grid-connected inverter is determined based on the first inverter parameters.

[0139] Step S209: If the first probability difference is less than or equal to the first preset probability difference and the second probability difference is greater than the second preset probability difference, determine the corresponding adaptation control command for the photovoltaic grid-connected inverter based on the first inverter parameters and the second inverter parameters.

[0140] Step S210: When the first probability difference is greater than the first preset probability difference and the second probability difference is less than or equal to the second preset probability difference, the corresponding adaptation control command for the photovoltaic grid-connected inverter is determined based on the first inverter parameters and the third inverter parameters.

[0141] Step S211: When the first probability difference is less than or equal to the first preset probability difference and the second probability difference is less than or equal to the second preset probability difference, the corresponding adaptation control command for the photovoltaic grid-connected inverter is determined based on the first inverter parameters, the second inverter parameters and the third inverter parameters.

[0142] Step S212: Perform corresponding adaptation control processing on the photovoltaic grid-connected inverter according to the adaptation control instructions.

[0143] In the above-mentioned photovoltaic grid-connected inverter adaptation control method based on adaptive switching hybrid control of operating modes, when performing adaptation control of the photovoltaic grid-connected inverter, the reliability of the control premise is ensured by first verifying the hardware status and the effective output coefficient of the photovoltaic power station. Then, based on the correlation parameters between the inverter and the power station, and between the inverter and the grid, multiple trained operating mode prediction models are used to make probabilistic predictions, and the optimal current grid operating mode is obtained by weighted fusion according to the model weights. Furthermore, the adaptation control command is determined by combining historical operating modes and future operating modes. This method gets rid of the limitations of traditional fixed parameters and preset models, and can accurately match different operating conditions and grids with different characteristics, which is conducive to improving the accuracy of adaptation control of photovoltaic grid-connected inverters.

[0144] In an exemplary embodiment, to more clearly illustrate the photovoltaic grid-connected inverter adaptation control method based on adaptive switching hybrid control of operating modes provided in this application, the following specific embodiment will be used to describe this photovoltaic grid-connected inverter adaptation control method based on adaptive switching hybrid control of operating modes. In one embodiment, this application also provides a photovoltaic grid-connected inverter grid-connected operation mode adaptive switching hybrid control method based on SCR (Short Circuit Ratio). Specifically, it includes the following:

[0145] This scheme is based on a hybrid control method for adaptive switching between grid-connected and grid-connected operation modes of photovoltaic grid-connected inverters using SCR. First, based on system parameters, grid-connected controllers and grid-connected controllers are designed separately to obtain the corresponding active current and reactive current setpoints. Next, the short-circuit ratio (SCR) of the power grid is identified. Based on the SCR value, hysteresis control is used to switch to grid-connected mode under strong grid conditions and to grid-connected mode under weak grid conditions. Then, through active and reactive current decoupling control and voltage feedforward control, SPWM modulation waves are obtained. Finally, the maximum and minimum values ​​of the three-phase modulation waves are calculated to obtain the modulation waves of the SVPWM modulation method, resulting in the drive pulse signals for each half-bridge.

[0146] This solution is based on a hybrid control method for adaptive switching of SCR photovoltaic grid-connected inverter and grid-connected operation mode, and is implemented according to the following steps:

[0147] Step 1: Calculate the rated AC phase voltage and current, rated DC voltage, AC inductance, and AC equivalent resistance based on the rated line voltage and rated power of the photovoltaic grid-connected inverter system.

[0148] Figure 3 This is the main circuit block diagram of the photovoltaic grid-connected inverter for which the control method of this scheme is applied. Figure 3 middle, This represents the effective value of the three-phase phase voltage. I represents the effective value of the three-phase line voltage. These are the effective values ​​of the three-phase currents. Given the rated line voltage and rated power of the photovoltaic grid-connected inverter system, the rated AC phase voltage, rated AC current, rated DC voltage, AC inductance, and AC equivalent resistance are calculated separately. The specific process is as follows:

[0149] 1.1 Calculate the rated AC phase voltage based on the rated line voltage. The expression is:

[0150] Equation (1)

[0151] in, This is the rated line voltage.

[0152] 1.2 Calculate the rated AC current based on the rated power and rated AC phase voltage. The expression is:

[0153] Equation (2)

[0154] in, Rated power, This is the rated AC phase voltage.

[0155] 1.3 Calculate the rated DC voltage based on the rated AC phase voltage. The expression is:

[0156] Equation (3)

[0157] in, The DC boost factor ranges from 1.1 to 1.5, with 1.414 being preferred.

[0158] 1.4 Calculate the connection inductance based on the known system parameters. The expression is:

[0159] Equation (4)

[0160] in, ω is the reactance, ranging from 0.05 to 0.2, preferably 0.1; ω is the grid angular frequency.

[0161] 1.5 Calculate the connection resistance based on the system parameters. The expression is:

[0162] Equation (5)

[0163] in, For system efficiency, the range is between 0.98 and 0.995, with 0.99 being preferred.

[0164] Step 2, based on the rated AC current of the photovoltaic grid-connected inverter system and rated DC voltage Select the rated voltage of the power device IGBT (Insulated Gate Bipolar Transistor) and the DC voltage of the module; calculate the DC bus capacitance parameters.

[0165] The specific process is as follows:

[0166] 2.1. Based on the numerical range of the rated AC current, select the rated voltage of the IGBT power device. The expression is:

[0167] Equation (6)

[0168] 2.2. Based on the rated voltage range of the IGBT power device, select the DC voltage of the module. The expression is:

[0169] Equation (7)

[0170] 2.3 Calculate the DC bus capacitance parameters based on the known system parameters:

[0171] When a film capacitor is used for the DC bus capacitance, the expression is as follows:

[0172] Equation (8)

[0173] in, This is the capacitance value of the thin-film capacitor 1; This represents the minimum value of the DC bus voltage; a typical value is taken as... ; This represents the maximum value of the DC bus voltage; a typical value is taken as... ; The energy storage time ranges from 0.2s to 30s, with a typical value of 0.5s. The proportionality coefficient representing the alternating current and direct current ranges from 1.1 to 1.5, preferably 1.23.

[0174] Step 3, based on the system parameters of the photovoltaic grid-connected inverter system, see... Figure 2 The grid control unit is equipped with grid-connected active and reactive current command generators. The grid active current setpoint is calculated by the grid active current regulator, and the grid reactive current setpoint is calculated by the grid reactive current regulator.

[0175] Figure 4 This is a block diagram illustrating the principle of the grid-connected control strategy for a photovoltaic grid-connected inverter system; correspondingly, Figure 8 This is a flowchart of the grid-connected control process for a photovoltaic grid-connected inverter system. The specific process is as follows:

[0176] 3.1. The instantaneous voltage value uabc and the instantaneous current value iabc of the power grid are obtained by AD (Analog-to-Digital) sampling.

[0177] 3.2. Using the MPPT (Maximum Power Point Tracking) algorithm, the active power setpoint is obtained based on the DC voltage Udc and DC current Idc. .

[0178] 3.3 Obtain the system's active power by calculating instantaneous reactive power. and reactive power The actual value of is expressed as follows:

[0179] Equation (9)

[0180] In equation (9), These are the instantaneous values ​​of the three-phase phase voltages. i represents the instantaneous value of the three-phase line voltage. This represents the instantaneous value of the three-phase current.

[0181] 3.4. Obtain the actual frequency f and phase angle wt1 of the power grid through a phase-locked loop (PLL).

[0182] 3.5. The abc / pq converter is used to perform abc / pq conversion to obtain up1 and uq1 of the power grid.

[0183] 3.6 Based on the actual power grid frequency f and the power grid frequency setpoint f*, combined with the actual active power value P and the active power setpoint... The active current setpoint ip1* during grid-connected control is calculated using the grid-connected active power regulator. The calculation formula is as follows:

[0184] Equation (10)

[0185] Where J is the torque inertia, denoted as the damping coefficient, ud as the rated phase voltage of the system, and s as the differential operator.

[0186] 3.7 Based on the actual value U and the given value U* of the grid AC voltage, and combined with the actual value Q and the given value Q* of the reactive power, the grid-connected reactive current given value iq1* during grid-connected control is calculated using the grid-connected reactive power regulator. The calculation formula is as follows:

[0187] Equation (11)

[0188] in, Given the voltage value, The given value for reactive current. This is the reactive power droop factor. This is the voltage droop factor.

[0189] The calculation process in step 3 is as follows: Figure 4 See the principle block diagram. Figure 8 .

[0190] Step 4: Based on the system parameters of the photovoltaic grid-connected inverter system, calculate the given values ​​of the grid active current and reactive current through the grid control unit.

[0191] Reference Figure 5 This is a schematic diagram of the grid-connected inverter system's control unit; refer to... Figure 9 This is a flowchart of the grid connection control of a photovoltaic grid-connected inverter system. The specific process is as follows:

[0192] 4.1. Obtain the instantaneous voltage value uabc and the instantaneous current value iabc of the power grid through AD sampling.

[0193] 4.2. Using the MPPT algorithm, the active power setpoint is obtained based on the DC voltage Udc and DC current Idc. .

[0194] 4.3. Using the instantaneous reactive power calculation formula, obtain the actual active power P and the actual reactive power Q of the system, as shown in the following expressions:

[0195] Equation (12)

[0196] 4.4, Based on the system's grid active power setpoint The phase angle of the system is obtained through the grid active power regulator. The expression is as follows:

[0197] Equation (13)

[0198] Where J is the torque inertia, The damping coefficient is... This is the second frequency modulation coefficient. This represents the actual value of the active power of the power grid. The given value for the active power of the grid.

[0199] 4.5, Based on the system's grid reactive power setpoint and AC voltage setpoint The output voltage setpoint e of the system is obtained through the grid reactive power regulator, and the expression is as follows:

[0200] Equation (14)

[0201] in, As a voltage reference, This is the reactive power droop factor. This is the voltage droop factor.

[0202] 4.6, based on the system's phase angle and output voltage setpoint The instantaneous setpoint eabc of the system output voltage is obtained through Ew / abc transformation, and the expression is as follows:

[0203] Equation (15)

[0204] 4.7 Based on the system's voltage output setpoint and feedback value, and AC current setpoint and feedback value, the output current setpoint of the photovoltaic grid-connected inverter system is obtained through the grid-connected mechanical regulator. The expression is as follows:

[0205] Equation (16)

[0206] Where Ls is the AC connection inductance, This is the collected grid voltage value. The system's equivalent resistance. This is the given value for the instantaneous output current of the photovoltaic grid-connected inverter system.

[0207] 4.8, based on the system's phase angle and the given value of the instantaneous output current Through the new abc / pq transformation, the active current setpoint ip2* and reactive current setpoint iq2* during grid-connected control are obtained.

[0208] Step 5: Execute the mode adaptive switching hybrid control strategy. Specifically, the SCR of the photovoltaic grid-connected inverter system is calculated based on the operating data of the system, and hysteresis control is adopted. Based on the SCR value, the state switch STA (Station) is performed between the grid connection and the grid connection.

[0209] Reference Figure 6 and Figure 7 This is a block diagram illustrating the principle of the grid-connected hybrid control strategy for a photovoltaic grid-connected inverter system; refer to... Figure 10 This is a flowchart of the grid connection status switching control of a photovoltaic grid-connected inverter system. The specific process is as follows:

[0210] 5.1 Using the state switching controller, based on the grid voltage acquisition value uabc and the grid current acquisition value iabc, the effective value of the grid voltage U and the effective value of the grid current I are obtained through the RMS (Root Mean Square) algorithm, as shown in the following expressions:

[0211] Equation (17)

[0212] Equation (18)

[0213] Where N is the number of data points used in the root mean square algorithm; ua(i) is the i-th data point when performing the root mean square algorithm.

[0214] 5.2 Calculate the grid connection inductive reactance based on the grid voltage RMS value U, the grid voltage rated value U*, and the grid current RMS value I. The expression is as follows:

[0215] Equation (19)

[0216] 5.3, based on the line voltage drop dU and the grid connection inductive reactance Given the grid current rating I*, calculate the short-circuit ratio SCR using the following expression:

[0217] Equation (20)

[0218] Where Ls is the line connection inductance and Rs is the line connection resistance.

[0219] 5.4 In the state switching controller, the switching state STA is calculated based on the short-circuit ratio SCR. When the short-circuit ratio SCR is large and the power grid is in a strong grid state, the controller switches to the grid-following control unit; when the short-circuit ratio SCR is small and the power grid is in a weak grid state, the controller switches to the grid-forming control unit.

[0220] To reduce phenomena such as current overcurrent, voltage surges, and power oscillations during switching, hysteresis control is introduced during state switching. The expression for the control logic is as follows:

[0221] Equation (21)

[0222] Where h is the switching midpoint, with a value of 5; dh is the hysteresis width. , dh is the hysteresis width; when the value is 0.4, then dh is 2.

[0223] 5.5. Select the appropriate phase angle based on the value of the switching state STA. The active current setpoint ip* and reactive current setpoint iq* are given. Since this is a photovoltaic grid-connected inverter, the range of ip* is limited to [0, 3*ia], and the expression is as follows:

[0224] Equation (22)

[0225] Equation (23)

[0226] Equation (24)

[0227] Step 6: Based on the current state of the switch determined in Step 5, obtain the three-phase modulation waves Sa, Sb, Sc of the photovoltaic grid-connected inverter system during SPWM modulation through the active current and reactive current decoupling controller.

[0228] Reference Figure 6 and Figure 7 The schematic diagram, and then refer to Figure 11 The flowchart shown is for the decoupling control of active and reactive currents in a photovoltaic grid-connected inverter system. The specific process is as follows:

[0229] 6.1, based on the grid voltage uabc and the photovoltaic grid-connected inverter system current iabc, and based on the currently used grid phase angle... In the active and reactive current decoupling controller, one abc / pq transformation is used to calculate the voltages up and uq; simultaneously, another abc / pq transformation is used to calculate the actual values ​​of the active current ip and reactive current iq, as shown in the following expressions:

[0230] Equation (25)

[0231] Equation (26)

[0232] 6.2 Based on the active current setpoint ip* and actual value ip, the voltage deviation dup is obtained through a PI regulator, and based on the reactive current setpoint iq* and actual value iq, the voltage deviation duq is obtained through another PI regulator. The expressions are as follows:

[0233] Equation (27)

[0234] Equation (28)

[0235] 6.3, through By performing current decoupling and combining it with voltage feedforward of up and uq, the voltage setpoint is obtained. and The expression is as follows:

[0236] Equation (29)

[0237] 6.4, using the voltage setpoint and Combined with the current power grid phase angle The instantaneous values ​​of the three-phase voltage setpoints are obtained through dq / abc transformation. Dividing each by the DC voltage, we obtain the three-phase modulation waves Sa1, Sb1, and Sc1, as expressed below:

[0238] Equation (30)

[0239] Step 7: Based on the three-phase modulation wave of the photovoltaic grid-connected inverter system obtained in Step 6, calculate the maximum and minimum values ​​of the three-phase SPWM, obtain the modulation wave during SVPWM modulation, and obtain the switching pulse values ​​of each power unit.

[0240] Reference Figure 12 This is a block diagram illustrating the principle of a hybrid modulation strategy combining CPSPWM (Carrier Phase Shifted Pulse Width Modulation) and NLM (Nearest Level Modulation) in a photovoltaic grid-connected inverter system; refer to... Figure 13 This is the control flowchart of the hybrid modulation method of CPSPWM and NLM in a photovoltaic grid-connected inverter system; the specific process is as follows:

[0241] 7.1 Calculate the maximum value of the three-phase modulation wave based on the modulation wave during three-phase SPWM modulation. and minimum value The expression is as follows:

[0242] Equation (31)

[0243] Equation (32)

[0244] 7.2 Based on the maximum and minimum values ​​of the modulation wave in the three-phase bridge arm, the increase in modulation wave is obtained:

[0245] Equation (33)

[0246] 7.3, based on the three-phase SPWM modulation wave and the increase in modulation wave of the photovoltaic grid-connected inverter system. The three-phase modulation wave under SVPWM modulation is obtained.

[0247] Equation (34)

[0248] Equation (35)

[0249] Equation (36)

[0250] 7.4 Based on the three-phase modulation waveform during SVPWM modulation, the two switching pulse values ​​of each half-bridge power module are calculated. The expressions for the switching pulse values ​​S1 and S2 are as follows:

[0251] Equation (37)

[0252] Equation (38)

[0253] in, A triangular wave with frequency f is used for each half-bridge power module.

[0254] Based on the states of each half-bridge power module obtained by the above method, corresponding switching transistor drive pulses are generated, thereby realizing state control of the photovoltaic grid-connected inverter system.

[0255] Example 1:

[0256] The rated line voltage of the photovoltaic grid-connected inverter system is 380V, and the rated power is 100kW. Calculate the rated AC phase voltage and current, rated DC voltage, AC inductance, and AC equivalent resistance. The specific process is as follows:

[0257] 1.1 Calculate the rated AC phase voltage based on the rated line voltage. The expression is:

[0258] Equation (39)

[0259] in, This is the rated line voltage.

[0260] 1.2 Calculate the rated AC current based on the rated power and rated AC phase voltage. The expression is:

[0261] Equation (40)

[0262] in, Rated power, This is the rated AC phase voltage.

[0263] 1.3 Calculate the total DC voltage of the bridge arms based on the rated AC phase voltages. The expression is:

[0264] Equation (41)

[0265] in, The DC boost factor is set to 1.414.

[0266] 1.4 Calculate the connection inductance based on the system parameters. The expression is:

[0267] Equation (42)

[0268] in, ω is the reactance, taken as 0.1; ω is the angular frequency of the power grid.

[0269] 1.5 Calculate the connection resistance based on the system parameters. The expression is:

[0270] Equation (43)

[0271] in, For system efficiency, we take 0.99.

[0272] Example 2:

[0273] Step 2, based on the rated AC current of the photovoltaic grid-connected inverter system and rated DC voltage The rated voltage of the IGBT power device and the DC voltage of the module are selected; the DC bus capacitor parameters are calculated. The specific process is as follows:

[0274] 2.1. Based on the rated AC current, select the rated voltage of the power devices IGBTs (a total of six IGBTs). The expression is:

[0275] Equation (44)

[0276] 2.2 Select the module DC voltage based on the rated voltage of the IGBT power device. The expression is:

[0277] Equation (45)

[0278] 2.3 Calculate the DC bus capacitance parameters based on the known system parameters.

[0279] When a film capacitor is used for the DC bus capacitance, the expression is as follows:

[0280] Equation (46)

[0281] in, The proportionality coefficient representing the alternating current and direct current is taken as 1.23; The energy storage time is taken as 0.5s.

[0282] Example 3:

[0283] Based on Example 2, the following further design is made:

[0284] The grid voltage reference value is 1 pu, which drops to 0.6 pu between 0.1s and 0.4s, simulating a voltage drop; and rises to 1.3 pu between 0.5s and 0.6s, simulating a voltage surge. The grid frequency reference value is 50Hz, which drops to 45Hz between 0.5s and 0.6s, and rises to 55Hz between 0.7s and 0.8s.

[0285] Figure 14 , Figure 15 , Figure 16 and Figure 17 The graph displays the waveforms of grid voltage, grid frequency, active current setpoint, and reactive current setpoint when grid-connected control is employed. From top to bottom, the four variables are represented as follows:

[0286] When the grid voltage drops, the reactive current setpoint increases to provide voltage support for the grid.

[0287] When the grid voltage rises, the reactive current setpoint decreases, causing the voltage to return to the normal range.

[0288] When the grid frequency decreases, the active current setpoint increases to prevent the frequency from decreasing further.

[0289] When the grid frequency increases, the active current setpoint decreases to suppress further frequency increases.

[0290] Example 4:

[0291] Based on Example 3, the following further design is made:

[0292] The grid voltage reference value is 1 pu, which drops to 0.9 pu between 0.9 s and 1.2 s, resulting in a slight drop in the analog voltage. The grid frequency reference value is 50 Hz, which drops to 49.9 Hz between 0.3 s and 0.6 s, resulting in a slight decrease in the analog frequency.

[0293] Figure 18 , Figure 19 , Figure 20 and Figure 21 The graph displays the waveforms of grid voltage, grid frequency, active current setpoint, and reactive current setpoint when grid-connected control is employed. From top to bottom, the four variables are represented as follows:

[0294] When the grid voltage drops, the reactive current absorbed by the system decreases, but it can still provide support for the grid voltage.

[0295] When the grid frequency decreases, the active current setpoint increases in order to maintain grid frequency stability.

[0296] Example 5:

[0297] Based on Example 4, the following further design is made:

[0298] The grid voltage reference is 1 pu. The system short-circuit capacity ratio (SCR) is 2 during 0s–0.3s, jumps to 8 during 0.3s–0.6s, and returns to 2 during 0.6s–1s. The switching threshold is set to 5, and the hysteresis width is 1.

[0299] Figure 22 , Figure 23 , Figure 24 and Figure 25 The waveforms of grid voltage, detected grid SCR, grid-connected / network-connected status switching flag STA, and system output current are displayed. The variables are listed from top to bottom in the figure. The results show that:

[0300] At 0.3s, the system SCR jumps from 2 to 8, and the detection value gradually increases from 2 to 8. When the SCR detection value reaches 6, the status flag STA switches from 0 (following the network) to 1 (forming the network).

[0301] At 0.6s, the system SCR recovered from 8 to 2, and the detection value gradually decreased from 8 to 2. When the SCR detection value dropped to 4, the status flag STA switched back from 1 to 0.

[0302] Due to the introduction of hysteresis control, no voltage surge occurred during the state switching process, and there was no overcurrent in the system current, verifying the effectiveness of the hysteresis switching strategy.

[0303] Example 6:

[0304] Based on Example 5, the following further design is made:

[0305] For step 7, the following implementation plan is designed: the system defaults to operating in the rated inductive current output condition, and switches from rated inductive current to rated capacitive current operating condition in 0.6 seconds.

[0306] Figure 26 , Figure 27 and Figure 28 In the middle, from top to bottom, are the original value of the three-phase SPWM modulation wave, the increase in modulation wave, and the three-phase SVPWM modulation wave. Figure 26 , Figure 27 and Figure 28 It can be seen that the amplitude of the SVPWM modulation wave is lower after increasing the modulation wave amount, which can improve the DC voltage utilization rate, indicating the correctness of the proposed SVPWM modulation method.

[0307] In the above embodiments, when performing adaptive control of photovoltaic grid-connected inverters, the reliability of the control premise is ensured by first verifying the hardware status and the effective output coefficient of the photovoltaic power station. Then, based on the correlation parameters between the inverter and the power station, and between the inverter and the grid, multiple trained operation mode prediction models are used to perform probabilistic prediction. The optimal current grid operation mode is obtained by weighted fusion according to the model weights. Furthermore, the adaptive control command is determined by combining historical operation modes and future operation modes. This approach overcomes the limitations of traditional fixed parameters and preset models, and can accurately match different operating conditions and grids with different characteristics, which is beneficial to improving the accuracy of adaptive control of photovoltaic grid-connected inverters.

[0308] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0309] Based on the same inventive concept, this application also provides a photovoltaic grid-connected inverter adaptation control device for implementing the photovoltaic grid-connected inverter adaptation control method based on adaptive switching hybrid control of operating modes as described above. The solution provided by this device is similar to the implementation scheme described in the above method. Therefore, the specific limitations of one or more embodiments of the photovoltaic grid-connected inverter adaptation control device based on adaptive switching hybrid control of operating modes provided below can be found in the limitations of the photovoltaic grid-connected inverter adaptation control method based on adaptive switching hybrid control of operating modes described above, and will not be repeated here.

[0310] In one exemplary embodiment, such as Figure 29 As shown, a photovoltaic grid-connected inverter adaptation control device based on adaptive switching hybrid control of operating modes is provided, including: a first acquisition module 2901, a second acquisition module 2902, a mode prediction module 2903, a mode determination module 2904, and an inverter control module 2905, wherein:

[0311] The first acquisition module 2901 is used to respond to the adaptation control request for the photovoltaic grid-connected inverter, acquire the hardware status parameters of the photovoltaic grid-connected inverter, and the effective output coefficient of the photovoltaic power station corresponding to the photovoltaic grid-connected inverter.

[0312] The second acquisition module 2902 is used to acquire the first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station, and the second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter, when the hardware status parameters meet the preset parameter constraints and the effective output coefficient is greater than or equal to the preset output coefficient.

[0313] The mode prediction module 2903 is used to input the first correlation parameter and the second correlation parameter into multiple trained operation mode prediction models respectively, to obtain multiple predicted operation modes corresponding to the power grid to be analyzed output by each trained operation mode prediction model, and the prediction probability corresponding to each predicted operation mode.

[0314] The mode determination module 2904 is used to sum the prediction probabilities corresponding to the prediction operation mode output by each trained operation mode prediction model according to the model weight of each trained operation mode prediction model, to obtain the target prediction probability corresponding to each prediction operation mode, and to select the prediction operation mode with the highest target prediction probability from each prediction operation mode as the current operation mode of the power grid to be analyzed.

[0315] The inverter control module 2905 is used to acquire the historical operating modes of the power grid to be analyzed, determine the future operating mode of the power grid to be analyzed based on the current operating mode and the historical operating mode, determine the corresponding adaptation control command for the photovoltaic grid-connected inverter based on the current operating mode, the historical operating mode and the future operating mode, and perform corresponding adaptation control processing on the photovoltaic grid-connected inverter according to the adaptation control command.

[0316] In an exemplary embodiment, the inverter control module 2905 is further configured to acquire a first predicted probability corresponding to a historical operating mode and a second predicted probability corresponding to a future operating mode, and determine a first probability difference between a target predicted probability and the first predicted probability, and a second probability difference between the target predicted probability and the second predicted probability, and acquire first inverter parameters of the photovoltaic grid-connected inverter in the current operating mode, second inverter parameters in the historical operating mode, and third inverter parameters in the future operating mode; when the first probability difference is greater than a first preset probability difference and the second probability difference is greater than a second preset probability difference, determine the adaptation control command corresponding to the photovoltaic grid-connected inverter based on the first inverter parameters; in the case that ... When the probability difference is less than or equal to the first preset probability difference and the second probability difference is greater than the second preset probability difference, the corresponding adaptation control command for the photovoltaic grid-connected inverter is determined based on the first inverter parameters and the second inverter parameters. When the first probability difference is greater than the first preset probability difference and the second probability difference is less than or equal to the second preset probability difference, the corresponding adaptation control command for the photovoltaic grid-connected inverter is determined based on the first inverter parameters and the third inverter parameters. When the first probability difference is less than or equal to the first preset probability difference and the second probability difference is less than or equal to the second preset probability difference, the corresponding adaptation control command for the photovoltaic grid-connected inverter is determined based on the first inverter parameters, the second inverter parameters, and the third inverter parameters.

[0317] In an exemplary embodiment, the inverter control module 2905 is further configured to extract the first feature vector of the first inverter parameter, the second feature vector of the second inverter parameter, and the third feature vector of the third inverter parameter; perform fusion processing on the first feature vector, the second feature vector, and the third feature vector to obtain the fused feature vector corresponding to the photovoltaic grid-connected inverter; and input the fused feature vector into the trained adaptive control command prediction model to obtain the adaptive control command corresponding to the photovoltaic grid-connected inverter.

[0318] In an exemplary embodiment, the inverter control module 2905 is further configured to input the fused feature vector into a power control command prediction network to obtain a power control command corresponding to the photovoltaic grid-connected inverter; input the fused feature vector into a modulation wave control command prediction network to obtain a modulation wave control command corresponding to the photovoltaic grid-connected inverter; input the fused feature vector into an auxiliary control command prediction network to obtain an auxiliary control command corresponding to the photovoltaic grid-connected inverter; and obtain an adaptation control command corresponding to the photovoltaic grid-connected inverter based on the power control command, the modulation wave control command, and the auxiliary control command.

[0319] In an exemplary embodiment, the second acquisition module 2902 is further configured to acquire the power matching degree, response delay value, and voltage adaptation deviation between the photovoltaic grid-connected inverter and the photovoltaic power station, as well as the impedance matching degree, phase synchronization deviation value, and voltage improvement rate between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter; based on the power matching degree, response delay value, and voltage adaptation deviation, a first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station is obtained; and based on the impedance matching degree, phase synchronization deviation value, and voltage improvement rate, a second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter is obtained.

[0320] In an exemplary embodiment, the mode prediction module 2903 is further configured to construct a fourth feature vector corresponding to the first correlation parameter based on the power matching degree, response delay value, and voltage adaptation deviation, and to construct a fifth feature vector corresponding to the second correlation parameter based on the impedance matching degree, phase synchronization deviation value, and voltage improvement rate; to concatenate the fourth feature vector and the fifth feature vector to obtain a concatenated feature vector; and to input the concatenated feature vector into multiple trained operation mode prediction models to obtain multiple predicted operation modes corresponding to the power grid to be analyzed output by each trained operation mode prediction model, and the prediction probability corresponding to each predicted operation mode.

[0321] The modules in the aforementioned photovoltaic grid-connected inverter adaptation control device based on adaptive switching of operating modes can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0322] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 30As shown, this computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores data such as hardware status parameters and effective output coefficients. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a photovoltaic grid-connected inverter adaptation control method based on adaptive switching hybrid control of operating modes.

[0323] Those skilled in the art will understand that Figure 30 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0324] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0325] In one exemplary embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above-described method embodiments.

[0326] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0327] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0328] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0329] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A photovoltaic grid-connected inverter adaptation control method based on adaptive switching hybrid control of operating modes, characterized in that, The method includes: In response to the adaptation control request for the photovoltaic grid-connected inverter, the hardware status parameters of the photovoltaic grid-connected inverter and the effective output coefficient of the photovoltaic power station corresponding to the photovoltaic grid-connected inverter are obtained. When the hardware status parameters meet the preset parameter constraints and the effective output coefficient is greater than or equal to the preset output coefficient, the first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station, and the second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter are obtained. The first correlation parameter and the second correlation parameter are respectively input into multiple trained operation mode prediction models to obtain multiple predicted operation modes corresponding to the power grid to be analyzed, and the prediction probability corresponding to each predicted operation mode output by each trained operation mode prediction model. For each predicted operating mode, according to the model weight of each trained operating mode prediction model, the predicted probabilities corresponding to the predicted operating mode output by each trained operating mode prediction model are summed to obtain the target predicted probability corresponding to each predicted operating mode. From each predicted operating mode, the predicted operating mode with the highest target predicted probability is selected as the current operating mode of the power grid to be analyzed. The historical operating modes of the power grid to be analyzed are obtained. Based on the current operating mode and the historical operating modes, the future operating mode of the power grid to be analyzed is determined. Based on the current operating mode, the historical operating mode and the future operating mode, the corresponding adaptation control command for the photovoltaic grid-connected inverter is determined. The photovoltaic grid-connected inverter is then subjected to corresponding adaptation control processing according to the adaptation control command.

2. The method according to claim 1, characterized in that, The step of determining the adaptation control command corresponding to the photovoltaic grid-connected inverter based on the current operating mode, the historical operating mode, and the future operating mode includes: The system obtains a first predicted probability corresponding to the historical operating mode and a second predicted probability corresponding to the future operating mode, and determines a first probability difference between the target predicted probability and the first predicted probability, and a second probability difference between the target predicted probability and the second predicted probability. The system also obtains a first inverter parameter of the photovoltaic grid-connected inverter in the current operating mode, a second inverter parameter in the historical operating mode, and a third inverter parameter in the future operating mode. When the first probability difference is greater than the first preset probability difference and the second probability difference is greater than the second preset probability difference, the corresponding adaptation control command for the photovoltaic grid-connected inverter is determined based on the first inverter parameters. When the first probability difference is less than or equal to the first preset probability difference and the second probability difference is greater than the second preset probability difference, the corresponding adaptation control command for the photovoltaic grid-connected inverter is determined based on the first inverter parameters and the second inverter parameters. When the first probability difference is greater than the first preset probability difference and the second probability difference is less than or equal to the second preset probability difference, the adaptation control command corresponding to the photovoltaic grid-connected inverter is determined according to the first inverter parameters and the third inverter parameters. When the first probability difference is less than or equal to the first preset probability difference, and the second probability difference is less than or equal to the second preset probability difference, the adaptation control command corresponding to the photovoltaic grid-connected inverter is determined based on the first inverter parameters, the second inverter parameters, and the third inverter parameters.

3. The method according to claim 2, characterized in that, The step of determining the adaptation control command corresponding to the photovoltaic grid-connected inverter based on the first inverter parameters, the second inverter parameters, and the third inverter parameters includes: The first feature vector of the first inverter parameter, the second feature vector of the second inverter parameter, and the third feature vector of the third inverter parameter are extracted respectively. The first feature vector, the second feature vector, and the third feature vector are fused to obtain the fused feature vector corresponding to the photovoltaic grid-connected inverter. The fused feature vector is input into the trained adaptive control command prediction model to obtain the adaptive control command corresponding to the photovoltaic grid-connected inverter.

4. The method according to claim 3, characterized in that, The trained adaptive control command prediction model includes a power control command prediction network, a modulation wave control command prediction network, and an auxiliary control command prediction network. The step of inputting the fused feature vector into the trained adaptive control command prediction model to obtain the adaptive control command corresponding to the photovoltaic grid-connected inverter includes: The fused feature vector is input into the power control command prediction network to obtain the power control command corresponding to the photovoltaic grid-connected inverter. The fused feature vector is input into the modulation wave control command prediction network to obtain the modulation wave control command corresponding to the photovoltaic grid-connected inverter. The fused feature vector is input into the auxiliary control command prediction network to obtain the auxiliary control command corresponding to the photovoltaic grid-connected inverter. Based on the power control command, the modulation wave control command, and the auxiliary control command, the corresponding adaptation control command for the photovoltaic grid-connected inverter is obtained.

5. The method according to any one of claims 1 to 4, characterized in that, The process of obtaining the first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station, and the second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter, includes: The power matching degree, response delay value, and voltage adaptation deviation between the photovoltaic grid-connected inverter and the photovoltaic power station are obtained, as well as the impedance matching degree, phase synchronization deviation value, and voltage improvement rate between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter. Based on the power matching degree, the response delay value, and the voltage adaptation deviation, a first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station is obtained. Based on the impedance matching degree, the phase synchronization deviation value, and the voltage improvement rate, a second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter is obtained.

6. The method according to claim 5, characterized in that, The step of inputting the first correlation parameter and the second correlation parameter into multiple trained operation mode prediction models to obtain multiple predicted operation modes corresponding to the power grid to be analyzed, and the prediction probability corresponding to each predicted operation mode, as output by each trained operation mode prediction model, includes: Based on the power matching degree, the response delay value, and the voltage adaptation deviation, a fourth feature vector corresponding to the first correlation parameter is constructed; and based on the impedance matching degree, the phase synchronization deviation value, and the voltage improvement rate, a fifth feature vector corresponding to the second correlation parameter is constructed. The fourth feature vector and the fifth feature vector are concatenated to obtain a concatenated feature vector; The concatenated feature vector is input into multiple trained operation mode prediction models to obtain multiple predicted operation modes corresponding to the power grid to be analyzed, and the prediction probability corresponding to each predicted operation mode, as output by each trained operation mode prediction model.

7. A photovoltaic grid-connected inverter adaptation control device based on adaptive switching hybrid control of operating modes, characterized in that, The device includes: The first acquisition module is used to acquire the hardware status parameters of the photovoltaic grid-connected inverter and the effective output coefficient of the photovoltaic power station corresponding to the photovoltaic grid-connected inverter in response to the adaptation control request for the photovoltaic grid-connected inverter. The second acquisition module is used to acquire, when the hardware status parameters meet the preset parameter constraints and the effective output coefficient is greater than or equal to the preset output coefficient, a first correlation parameter between the photovoltaic grid-connected inverter and the photovoltaic power station, and a second correlation parameter between the photovoltaic grid-connected inverter and the grid to be analyzed corresponding to the photovoltaic grid-connected inverter. The mode prediction module is used to input the first correlation parameter and the second correlation parameter into multiple trained operation mode prediction models respectively, so as to obtain multiple predicted operation modes corresponding to the power grid to be analyzed output by each trained operation mode prediction model, and the prediction probability corresponding to each predicted operation mode. The mode determination module is used to sum the prediction probabilities corresponding to the predicted operating mode output by each trained operating mode prediction model according to the model weights of each predicted operating mode prediction model, to obtain the target prediction probability corresponding to each predicted operating mode, and to select the predicted operating mode with the highest target prediction probability from each predicted operating mode as the current operating mode of the power grid to be analyzed. The inverter control module is used to acquire the historical operating modes of the power grid to be analyzed, determine the future operating mode of the power grid to be analyzed based on the current operating mode and the historical operating modes, determine the corresponding adaptation control command for the photovoltaic grid-connected inverter based on the current operating mode, the historical operating mode and the future operating mode, and perform corresponding adaptation control processing on the photovoltaic grid-connected inverter according to the adaptation control command.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images