Distributed photovoltaic cluster cooperative control method based on output prediction and pre-distribution
By dividing and pre-allocating the support capabilities of photovoltaic inverters, the problem of poor grid stability caused by output fluctuations of photovoltaic inverters is solved, enabling photovoltaic inverters to provide rapid and effective support to the grid, thereby improving grid stability and control efficiency.
Patent Information
- Application Number
- CN202511554412.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-29
- Publication Date
- 2026-01-27
AI Technical Summary
Traditional photovoltaic control schemes are difficult to meet the grid's demand for active voltage/frequency support. The output fluctuation of photovoltaic inverters leads to poor grid stability. Existing technology control processes are time-consuming and easily miss the optimal support window, resulting in voltage recovery failure or frequency collapse.
The distributed photovoltaic cluster collaborative control method based on output prediction and pre-allocation divides the photovoltaic inverters into support capabilities, pre-allocates photovoltaic inverter sets for each demand, and uses factors such as historical active and reactive power fluctuation coefficients and electrical distance to determine the support inverter sets corresponding to active and reactive power demands, and performs rapid control when there are frequency or voltage deviations.
It improves the stability and efficiency of photovoltaic inverters in supporting the power grid, reduces the output fluctuation of photovoltaic clusters, and ensures the rapid response and effectiveness of power grid support.
Smart Images
Figure CN121417367A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of power system control technology, specifically relating to a distributed photovoltaic cluster collaborative control method based on output prediction and pre-allocation. Background Technology
[0002] Currently, driven by the "dual carbon" goal, the penetration rate of distributed photovoltaics continues to increase. However, traditional photovoltaic control schemes are difficult to meet the grid's demand for active voltage / frequency support. Low-voltage grid disconnection of photovoltaics will exacerbate the power deficit at grid nodes, and photovoltaic inverters are unable to provide effective support.
[0003] Furthermore, due to the influence of environmental factors such as sunlight and temperature, the active and reactive power output fluctuation characteristics of photovoltaic inverters vary in different geographical locations. Controlling photovoltaic inverters can easily lead to secondary disturbances in grid voltage / frequency, or even trigger protection disconnection. The stability of photovoltaic inverters in supporting the grid is poor. In addition, existing technologies usually require real-time calculation of grid reactive / active power deficits, screening of available photovoltaic inverters, and then allocating control commands to the photovoltaic inverters. The control process is time-consuming and may miss the optimal support window, resulting in voltage recovery failure or frequency collapse. The efficiency of photovoltaic inverters in supporting the grid is low.
[0004] Therefore, improving the stability and efficiency of photovoltaic inverters in supporting the power grid has become an urgent problem to be solved. Summary of the Invention
[0005] To address the shortcomings of existing technologies, this invention provides a distributed photovoltaic cluster collaborative control method based on output prediction and pre-allocation.
[0006] To solve one or more or all of the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A distributed photovoltaic (PV) cluster collaborative control method based on output prediction and pre-allocation includes: dividing the support capacity of each PV inverter in the target PV inverter set according to the fluctuation coefficients of the historical active power and historical reactive power of each PV inverter; determining the active power demand and reactive power demand based on the support capacity, historical active power, historical reactive power, and electrical distance to the target grid node of each PV inverter; determining the active power support inverter set corresponding to the active power demand and the reactive power support inverter set corresponding to the reactive power demand; when the frequency deviation value of the target grid node at a preset time point is greater than a preset threshold, selecting the active power support inverter set corresponding to the active power demand with the closest frequency deviation value, and controlling the PV inverters in the selected active power support inverter set according to the frequency deviation value using a preset control algorithm; when the voltage deviation value of the target grid node at a preset time point is greater than a preset threshold, selecting the reactive power support inverter set corresponding to the reactive power demand with the closest voltage deviation value, and controlling the PV inverters in the selected reactive power support inverter set according to the voltage deviation value using a preset control algorithm.
[0007] Furthermore, the volatility coefficient is the ratio of the mean to the standard deviation.
[0008] Furthermore, the method for classifying the support capabilities of each photovoltaic inverter in the target photovoltaic inverter set includes: classifying a specified target number of photovoltaic inverters as active power stable type according to the order of active power fluctuation coefficient from smallest to largest, and classifying the remaining photovoltaic inverters as active power flexible type; classifying a specified target number of photovoltaic inverters as reactive power stable type according to the order of reactive power fluctuation coefficient from smallest to largest, and classifying the remaining photovoltaic inverters as reactive power flexible type.
[0009] Furthermore, methods for determining active and reactive power demand include: performing cluster analysis on the historical active power demand of the target power grid nodes to obtain multiple active power demand quantities; and performing cluster analysis on the historical reactive power demand of the target power grid nodes to obtain multiple reactive power demand quantities.
[0010] Furthermore, the method for determining the set of active power support inverters corresponding to each active power demand includes: obtaining the priority value of each photovoltaic inverter; traversing each active power demand, using the current active power demand as the active power demand target; for active power stable photovoltaic inverters, if their historical average active power output is less than the active power demand target and their priority value is greater than a preset threshold, then the photovoltaic inverter is determined as an active power candidate inverter; for active power flexible photovoltaic inverters, if their priority value is greater than a preset threshold, then the photovoltaic inverter is determined as a candidate inverter. The system first selects temporary active power inverters and predicts their active power output based on their historical active power sampling data. It then selects several active power candidate inverters, ensuring that the sum of their historical active power output averages is not greater than and is closest to the current active power demand target, and determines the current active power gap. Finally, it selects several temporary active power inverters, ensuring that the sum of their predicted active power outputs is not less than and is closest to the current active power gap. The selected active power candidate inverters and temporary active power inverters form the active power support inverter set corresponding to the active power demand.
[0011] Furthermore, the method for determining the set of reactive power-supporting inverters corresponding to each reactive power demand includes: obtaining the priority value of each photovoltaic inverter; traversing each reactive power demand, using the current reactive power demand as the reactive power demand target; for reactive power-stable photovoltaic inverters, if their historical average reactive power output is less than the reactive power demand target and their priority value is greater than a preset threshold, then the photovoltaic inverter is determined to be a reactive power candidate inverter; for reactive power-flexible photovoltaic inverters, if their priority value is greater than a preset threshold, then the photovoltaic inverter is determined to be a reactive power candidate inverter. The system first selects a temporary reactive power inverter and predicts its reactive power output based on its historical reactive power sampling data. Then, it selects several candidate reactive power inverters such that the sum of their historical average reactive power outputs is not greater than and is closest to the current reactive power demand target, and determines the current reactive power gap. Finally, it selects several temporary reactive power inverters such that the sum of their predicted reactive power outputs is not less than and is closest to the current reactive power gap. The selected candidate and temporary reactive power inverters form a set of reactive power support inverters corresponding to the reactive power demand.
[0012] Furthermore, the method for obtaining the priority value of each photovoltaic inverter includes: determining the corresponding distance evaluation value based on the electrical distance between each photovoltaic inverter and the target grid node; determining the corresponding selection evaluation value based on the number of times each photovoltaic inverter is included in the active power support inverter set and the reactive power support inverter set; and determining the corresponding priority value based on the product of the distance evaluation value and the selection evaluation value of each photovoltaic inverter.
[0013] A distributed photovoltaic (PV) cluster collaborative control device based on output prediction and pre-allocation includes: a classification module for classifying the support capabilities of each PV inverter in the target PV inverter set according to the fluctuation coefficients of the historical active power and historical reactive power of each PV inverter; a set construction module for determining the active power demand and reactive power demand based on the support capabilities, historical active power, historical reactive power, and electrical distance to the target grid node of each PV inverter, and determining the active power support inverter set and the reactive power support inverter set corresponding to the active power demand and the reactive power support inverter set respectively; and a frequency support control module. The first module is used to select the set of active power support inverters corresponding to the active power demand with the closest frequency deviation value when the frequency deviation value of the target grid node at a preset time point is greater than a preset threshold value; and to control the photovoltaic inverters in the selected set of active power support inverters according to the frequency deviation value using a preset control algorithm. The second module is used to select the set of reactive power support inverters corresponding to the reactive power demand with the closest voltage deviation value when the voltage deviation value of the target grid node at a preset time point is greater than a preset threshold value; and to control the photovoltaic inverters in the selected set of reactive power support inverters according to the voltage deviation value using a preset control algorithm.
[0014] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the control method when executing the computer program.
[0015] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the control method.
[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention categorizes photovoltaic (PV) inverters by output fluctuation analysis, pre-allocating corresponding PV inverter sets to each preset demand. This effectively reduces the output fluctuation of PV clusters, improving the stability of PV inverters' grid support. The pre-allocation of PV inverter sets also allows for direct control of the PV inverter sets based on preset demand when grid support needs arise, thus improving the efficiency of PV inverters' grid support. Attached Figure Description
[0017] The present invention will now be described in further detail with reference to the accompanying drawings.
[0018] Figure 1 : Flowchart of the control method of the present invention; Figure 2 : Schematic diagram of the control device of the present invention; Figure 3: Schematic diagram of the computer device of the present invention. Detailed Implementation
[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0020] In one embodiment, such as Figure 1 As shown, a method for coordinated control of distributed photovoltaic clusters based on output prediction and pre-allocation is provided, which includes the following steps: Step S101: Based on the fluctuation coefficients of the historical active power and historical reactive power of each photovoltaic inverter, the support capabilities of each photovoltaic inverter in the target photovoltaic inverter set are divided.
[0021] A photovoltaic (PV) inverter set is formed by identifying PV inverters in the target power grid capable of output control. Based on historical active and reactive power sampling data for each PV inverter in the set, the active power fluctuation coefficient and reactive power fluctuation coefficient for each inverter are obtained. The target PV inverter set is then categorized according to its active and reactive power support capabilities based on the historical fluctuation coefficients. Based on the active power fluctuation coefficient, the target PV inverter set is further divided into active power stable and active power flexible types; and based on the reactive power fluctuation coefficient, it is divided into reactive power stable and reactive power flexible types.
[0022] The volatility coefficient is an indicator that quantifies the degree of fluctuation in the active and reactive power output of a photovoltaic (PV) inverter. It is the ratio of the average value to the standard deviation; specifically, the active power volatility coefficient is the ratio of the historical average active power output to the historical standard deviation, and the reactive power volatility coefficient is the ratio of the historical average reactive power output to the historical standard deviation. Using the volatility coefficient can eliminate the impact of PV inverter capacity variations and improve the accuracy of quantifying the degree of fluctuation in the active and reactive power output of PV inverters.
[0023] Methods for classifying the support capabilities of each photovoltaic inverter in the target photovoltaic inverter set include: According to the order of active power fluctuation coefficient from smallest to largest, a specified number of photovoltaic inverters are classified as active power stable type, and the remaining photovoltaic inverters are classified as active power flexible type. Based on the reactive power fluctuation coefficient from smallest to largest, a specified number of photovoltaic inverters are classified as reactive power stable type, and the remaining photovoltaic inverters are classified as reactive power flexible type.
[0024] In this embodiment, active power stable and reactive power stable photovoltaic inverters take priority in meeting demand and providing stable support, while active power flexible and reactive power flexible photovoltaic inverters fill the gap through subsequent forecasting and provide flexible support.
[0025] Step S102: Based on the support capacity, historical active power, historical reactive power, and electrical distance to the target grid node of each photovoltaic inverter, determine the active power demand and reactive power demand, and determine the active power support inverter set corresponding to the active power demand and the reactive power support inverter set corresponding to the reactive power demand.
[0026] Based on the active power support capability category, historical active power sampling data, historical reactive power sampling data, and electrical distance between each photovoltaic inverter in the target photovoltaic inverter set and the target grid node, determine the active power support inverter set corresponding to the N active power demand of the target grid node, and determine the reactive power support inverter set corresponding to the N reactive power demand of the target grid node.
[0027] Electrical distance can be characterized by the voltage / power sensitivity of the photovoltaic inverter to the target grid node, thus quantifying supporting efficiency; higher sensitivity indicates a shorter electrical distance. Electrical distance can also be characterized by line impedance; lower line impedance indicates a shorter electrical distance.
[0028] The active power demand and reactive power demand are preset output demand values, which can be set by the implementer in conjunction with expert experience.
[0029] The active or reactive power demand can be obtained by clustering analysis based on the historical active or reactive power demand of the target power grid node. In the clustering analysis, N is used as the number of cluster sets. The K-means clustering algorithm can be used to cluster the data to obtain N cluster sets. The cluster centers of the N cluster sets are used as multiple (N) active or reactive power demand values.
[0030] Methods for determining the set of active power support inverters corresponding to the N active power demands of a target grid node include: Iterate through each active power demand and use the current active power demand as the active power demand target. For any photovoltaic inverter, the distance evaluation value of the photovoltaic inverter is determined based on the electrical distance between the photovoltaic inverter and the target grid node; the selection evaluation value of the photovoltaic inverter is determined based on the number of times the photovoltaic inverter is included in the active power support inverter set and the number of times the reactive power support inverter set. The priority value of the photovoltaic inverter is determined based on its distance evaluation value and selection evaluation value. For a photovoltaic inverter with stable active power output, if its historical average active power output is less than the active power demand target and its priority value is greater than a preset threshold, then the photovoltaic inverter is determined to be an active power candidate inverter. For a flexible active photovoltaic inverter, if its priority value is greater than a preset threshold, the photovoltaic inverter is determined to be a temporary active inverter, and the predicted active output of the photovoltaic inverter at L preset time points is predicted based on its historical active power sampling data. Select several active power candidate inverters such that the sum of the historical active power output averages of the selected active power candidate inverters is not greater than and is closest to the current active power demand target. The difference between the active power demand target and the historical active power output averages of the selected active power candidate inverters is the current active power gap. Select several active temporary inverters such that the sum of the predicted active output of the selected active temporary inverters is not less than and is closest to the current active power gap. The selected active power candidate inverters and active power temporary inverters form the active power support inverter set corresponding to the active power demand.
[0031] The method for determining the reactive power demand corresponding to the target grid node and the corresponding reactive power support inverter set is the same as the method described above. It only requires the use of the historical reactive power sampling data and historical reactive power output average data of the corresponding inverter, and will not be elaborated further.
[0032] Electrical distance can also be characterized by line impedance; the smaller the line impedance, the closer the electrical distance.
[0033] The distance evaluation value can be calculated by the reciprocal of the electrical distance to achieve normalization, and to make the distance evaluation value larger the closer the electrical distance.
[0034] The selection evaluation value reflects the availability of photovoltaic (PV) inverters, preventing over-allocation and conflicts caused by multiple grid nodes calling the same PV inverter when they have support needs. The selection evaluation value for a PV inverter is determined based on the number of times it is included in the active power support inverter set and the reactive power support inverter set. The included active power support inverter set can correspond to any active power demand of any grid node, and similarly, the included reactive power support inverter set can correspond to any reactive power demand of any grid node.
[0035] The implementer can preset the maximum number of allocations C. Let A be the number of times the photovoltaic inverter is included in the active power support inverter set and B be the number of times it is included in the reactive power support inverter set. The selection evaluation value of the photovoltaic inverter can be calculated by 1-(A+B) / C.
[0036] Priority values can be the product of distance evaluation values and selection evaluation values, or a sum of weighted values.
[0037] When predicting active power output based on historical active power sampling data, or when predicting reactive power output based on historical reactive power sampling data, existing time series prediction models can be used, such as long short-term memory network models, temporal convolutional models, etc. The architecture and training process of time series prediction models will not be elaborated here.
[0038] When selecting active power candidate inverters, active power candidate inverters are pre-selected in descending order of selection priority value and their historical active power output average value is accumulated. After each accumulation, if the accumulated result is not greater than the current active power demand target, the pre-selected active power candidate inverter will be officially selected.
[0039] Correspondingly, when selecting active temporary inverters, active temporary inverters are pre-selected in descending order of selection priority value and their historical active output average value is accumulated. After each accumulation, if the accumulated result is not less than the current active power gap, the pre-selected active temporary inverter will be officially selected.
[0040] Step S103: When the frequency deviation value of the target grid node at a preset time point is greater than a preset threshold, select the active power support inverter set corresponding to the active power demand that is closest to the frequency deviation value, and control the photovoltaic inverter in the selected active power support inverter set according to the frequency deviation value using a preset control algorithm.
[0041] The frequency deviation value is the difference between the real-time frequency and the rated frequency of the target power grid node at a preset time point.
[0042] The threshold set for the frequency deviation value is the upper limit of the frequency deviation allowed by the power grid, for example, it can be set to 0.1Hz.
[0043] The preset control algorithm in this step can adopt the capacity ratio allocation method. When the photovoltaic inverter receives the frequency support control signal, it can provide fast primary frequency regulation by actively reducing active power or calling the energy storage unit.
[0044] In another implementation, a set of active power support inverters is selected that corresponds to the active power demand that is closest to but greater than the frequency deviation value. Selecting an active power demand greater than the frequency deviation value ensures sufficient active power support and prevents a further drop in frequency due to insufficient support.
[0045] Step S104: When the voltage deviation of the target grid node at a preset time point is greater than a preset threshold, select the set of reactive power support inverters corresponding to the reactive power demand that is closest to the voltage deviation value, and control the photovoltaic inverters in the selected set of reactive power support inverters according to the voltage deviation value using a preset control algorithm.
[0046] The voltage deviation value is the difference between the real-time voltage and the rated voltage U of the target grid node at a preset time point.
[0047] The threshold set for the voltage deviation value is the upper limit of the voltage deviation allowed by the power grid, for example, it can be set to 0.05×U.
[0048] The preset control algorithm in this step can adopt the capacity ratio allocation method. When the photovoltaic inverter receives the voltage support control signal, it can switch to reactive power priority mode to provide capacitive reactive power support voltage recovery.
[0049] In another implementation, a set of reactive power support inverters is selected that corresponds to the reactive power demand that is closest to but greater than the voltage deviation value. Selecting a reactive power demand greater than the voltage deviation value ensures sufficient reactive power support and prevents a further voltage drop due to insufficient reactive power.
[0050] In this embodiment, photovoltaic inverters are categorized by their support capabilities through output fluctuation analysis. A corresponding set of photovoltaic inverters is pre-allocated to each preset demand, which effectively reduces the output fluctuation of the photovoltaic cluster and improves the stability of the photovoltaic inverters' support for the grid. The pre-allocation of photovoltaic inverter sets also means that when grid support demand arises, the photovoltaic inverter sets can be directly called and controlled according to the preset demand, thus improving the efficiency of the photovoltaic inverters' support for the grid.
[0051] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
[0052] In one embodiment, a distributed photovoltaic cluster collaborative control device based on output prediction and pre-allocation is provided, which corresponds to the control method in the above embodiments. For example... Figure 2 As shown, the functional modules of this control device are described in detail below: The classification module 201 is used to classify the support capabilities of each photovoltaic inverter in the target photovoltaic inverter set according to the fluctuation coefficients of the historical active power and historical reactive power of each photovoltaic inverter. The assembly module 202 is used to determine the active power demand and reactive power demand based on the support capacity, historical active power, historical reactive power and electrical distance to the target grid node of each photovoltaic inverter, and to determine the active power support inverter set and the reactive power support inverter set corresponding to the active power demand and the reactive power support inverter set respectively. The frequency support control module 203 is used to select the set of active power support inverters corresponding to the active power demand that is closest to the frequency deviation value when the frequency deviation value of the target grid node at a preset time point is greater than a preset threshold; and to control the photovoltaic inverters in the selected set of active power support inverters according to the frequency deviation value using a preset control algorithm. The voltage support control module 204 is used to select the set of reactive power support inverters corresponding to the reactive power demand that is closest to the voltage deviation value when the voltage deviation value of the target grid node at a preset time point is greater than a preset threshold; and to control the photovoltaic inverters in the selected set of reactive power support inverters according to the voltage deviation value using a preset control algorithm.
[0053] For specific limitations regarding the control device, please refer to the limitations on the control method above, which will not be repeated here. Each module in the aforementioned control device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in hardware or independently of the processor in the computer device, or stored in software in the memory of the computer device, so that the processor can call and execute the operations corresponding to each module.
[0054] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 3 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing 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 database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a control method.
[0055] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the control method described in the above embodiment; to avoid repetition, this will not be repeated here. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in this embodiment of the control device; to avoid repetition, this will not be repeated here.
[0056] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When executed by a processor, the computer program implements the control method described in the above embodiment; to avoid repetition, this will not be repeated here. Alternatively, when executed by a processor, the computer program implements the functions of each module / unit in this embodiment of the control device; to avoid repetition, this will not be repeated here.
[0057] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This 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 of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0058] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.
[0059] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.
Claims
1. A method for coordinated control of distributed photovoltaic clusters based on output prediction and pre-allocation, characterized in that, include Based on the fluctuation coefficients of the historical active power and historical reactive power of each photovoltaic inverter, the support capabilities of each photovoltaic inverter in the target photovoltaic inverter set are divided. Based on the support capacity, historical active power, historical reactive power, and electrical distance to the target grid node of each photovoltaic inverter, the active power demand and reactive power demand are determined, and the active power support inverter set and the reactive power support inverter set corresponding to the active power demand and the reactive power demand are determined respectively. When the frequency deviation of the target grid node at a preset time point is greater than a preset threshold, the active power support inverter set corresponding to the active power demand closest to the frequency deviation value is selected, and the photovoltaic inverters in the selected active power support inverter set are controlled according to the frequency deviation value using a preset control algorithm. When the voltage deviation of the target grid node at a preset time point is greater than a preset threshold, the set of reactive power support inverters corresponding to the reactive power demand closest to the voltage deviation value is selected, and the photovoltaic inverters in the selected set of reactive power support inverters are controlled according to the voltage deviation value using a preset control algorithm.
2. The distributed photovoltaic cluster collaborative control method based on output prediction and pre-allocation according to claim 1, characterized in that, The volatility coefficient is the ratio of the mean to the standard deviation.
3. The distributed photovoltaic cluster collaborative control method based on output prediction and pre-allocation according to claim 1, characterized in that, Methods for classifying the support capabilities of each photovoltaic inverter in the target photovoltaic inverter set include: According to the order of active power fluctuation coefficient from smallest to largest, a specified number of photovoltaic inverters are classified as active power stable type, and the remaining photovoltaic inverters are classified as active power flexible type. Based on the reactive power fluctuation coefficient from smallest to largest, a specified number of photovoltaic inverters are classified as reactive power stable type, and the remaining photovoltaic inverters are classified as reactive power flexible type.
4. The distributed photovoltaic cluster collaborative control method based on output prediction and pre-allocation according to claim 3, characterized in that, Methods for determining active power demand and reactive power demand include: Cluster analysis is performed on the historical active power demand of the target power grid node to obtain multiple active power demand values; Cluster analysis is performed on the historical reactive power demand of the target power grid node to obtain multiple reactive power demand values.
5. The distributed photovoltaic cluster collaborative control method based on output prediction and pre-allocation according to claim 4, characterized in that, Methods for determining the set of active power support inverters corresponding to each active power demand include: Obtain the priority value of each photovoltaic inverter; Iterate through each active power demand and use the current active power demand as the active power demand target. For a photovoltaic inverter with stable active power output, if its historical average active power output is less than the active power demand target and its priority value is greater than a preset threshold, then the photovoltaic inverter is determined to be an active power candidate inverter. For a flexible active photovoltaic inverter, if its priority value is greater than a preset threshold, the photovoltaic inverter is determined to be a temporary active inverter, and the predicted active output is obtained based on its historical active power sampling data. Select several active power candidate inverters such that the sum of the historical active power output averages of the selected active power candidate inverters is not greater than and is closest to the current active power demand target, and determine the current active power gap. Select several active temporary inverters such that the sum of the predicted active output of the selected active temporary inverters is not less than and is closest to the current active power gap. The selected active candidate inverters and active temporary inverters form the active support inverter set corresponding to the active demand.
6. The distributed photovoltaic cluster collaborative control method based on output prediction and pre-allocation according to claim 4, characterized in that, Methods for determining the set of reactive power support inverters corresponding to each reactive power demand include: Obtain the priority value of each photovoltaic inverter; Iterate through each reactive power demand and use the current reactive power demand as the reactive power demand target. For a reactive power stabilization photovoltaic inverter, if its historical average reactive power output is less than the reactive power demand target and its priority value is greater than a preset threshold, then the photovoltaic inverter is determined to be a reactive power candidate inverter. For a reactive flexible photovoltaic inverter, if its priority value is greater than a preset threshold, the photovoltaic inverter is determined to be a reactive temporary inverter, and the predicted reactive output is obtained based on its historical reactive power sampling data. Select several reactive power candidate inverters such that the sum of the historical average reactive power output of the selected reactive power candidate inverters is not greater than and is closest to the current reactive power demand target, and determine the current reactive power gap. Select several reactive power temporary inverters such that the sum of the predicted reactive power output of the selected reactive power temporary inverters is not less than and is closest to the current reactive power gap. The selected reactive power candidate inverters and reactive power temporary inverters form a set of reactive power support inverters corresponding to the reactive power demand.
7. The distributed photovoltaic cluster collaborative control method based on output prediction and pre-allocation according to claim 5 or 6, characterized in that, Methods for obtaining the priority values of each photovoltaic inverter include: Based on the electrical distance between each photovoltaic inverter and the target grid node, the corresponding distance evaluation value is determined; based on the number of times each photovoltaic inverter is included in the active power support inverter set and the reactive power support inverter set, the corresponding selection evaluation value is determined. The priority value is determined by multiplying the distance evaluation value and the selection evaluation value of each photovoltaic inverter.
8. A distributed photovoltaic cluster collaborative control device based on output prediction and pre-allocation, characterized in that, include: The category classification module is used to classify the support capabilities of each photovoltaic inverter in the target photovoltaic inverter set according to the fluctuation coefficients of the historical active power and historical reactive power of each photovoltaic inverter. The assembly module is used to determine the active power demand and reactive power demand based on the support capacity, historical active power, historical reactive power and electrical distance to the target grid node of each photovoltaic inverter, and to determine the active power support inverter set and the reactive power support inverter set corresponding to the active power demand and the reactive power support inverter set respectively. The frequency support control module is used to select the set of active power support inverters that corresponds to the active power demand closest to the frequency deviation value when the frequency deviation value of the target grid node at a preset time point is greater than a preset threshold. Used to control the photovoltaic inverters in the selected active power support inverter set according to the frequency deviation value using a preset control algorithm; The voltage support control module is used to select the set of reactive power support inverters that corresponds to the reactive power demand closest to the voltage deviation value when the voltage deviation value of the target grid node at a preset time point is greater than a preset threshold. Used to control the photovoltaic inverters in the selected set of reactive power support inverters according to the voltage deviation value using a preset control algorithm.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the control method according to any one of claims 1-7.
10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the control method according to any one of claims 1-7.