Clustering frequency modulation control method and system for distributed photovoltaic participated primary frequency modulation

Abstract

The invention discloses a clustering frequency modulation control method and system for distributed photovoltaic participating in primary frequency modulation, and the method comprises the steps: screening out photovoltaic inverters, meeting the basic frequency modulation requirements, of a region where distributed photovoltaic participates in primary frequency modulation; grouping the screened photovoltaic inverters according to the rated power capacities and geographic positions of the photovoltaic inverters by using a weighted fuzzy C-means clustering algorithm; setting and regularly changing a primary frequency modulation starting sequence of each group of photovoltaic inverters according to the group of each photovoltaic inverter; and optimizing the primary frequency modulation parameters of each group of photovoltaic inverters by using a particle swarm algorithm, and performing frequency modulation control by using an optimization result. According to the invention, the primary frequency modulation response sequence of each group of photovoltaic inverters can be changed in turn, and the balance of the service life of each group of photovoltaic inverters is considered.

Description

Technical Field

[0001] The present invention belongs to the technical field of distributed photovoltaic frequency modulation control, and relates to a cluster frequency modulation control method and system for distributed photovoltaics participating in primary frequency modulation. Background Art

[0002] With the continuous growth of global energy demand and the improvement of environmental protection awareness, the development and utilization of clean energy has become an important part of the energy policies of various countries. Among them, photovoltaic power generation has attracted widespread attention due to its renewability and environmental friendliness. However, the intermittent and volatile nature of photovoltaic power generation also poses challenges to the stable operation of the power grid. In this context, studying the primary frequency modulation technology of distributed photovoltaics will help to smoothly promote energy transformation and provide support for achieving the "dual carbon" goal.

[0003] In traditional power systems, primary frequency regulation relies on units in large centralized power plants. However, with the widespread application of distributed photovoltaics, the structure of the power system is changing. Without frequency regulation capabilities, the output fluctuations of photovoltaic power generation may cause the power system to frequently start and stop standby units, increasing system operating costs. Through primary frequency regulation technology, photovoltaic systems can participate in grid frequency regulation while meeting their own power generation needs, which not only improves the absorption capacity of photovoltaic power, but also reduces the peak-shaving pressure of fossil energy units, and promotes the optimization of energy structure.

[0004] However, unlike centralized photovoltaic power stations, distributed photovoltaics are huge in number and scattered in geographical locations. Simply setting the frequency modulation parameters uniformly will cause the output power of distributed photovoltaics to drop sharply when the grid is overclocked, and laying communication lines on a large scale between distributed photovoltaics will cause huge economic costs. Therefore, for a large number of distributed photovoltaic inverters, it is difficult to coordinate and control their primary frequency modulation. There is an urgent need for a low-cost, effective distributed photovoltaic inverter frequency modulation control method and system to continuously promote the application of distributed photovoltaics. Summary of the invention

[0005] In order to solve the deficiencies in the prior art, the present invention provides a clustered frequency modulation control method and system for distributed photovoltaics participating in primary frequency modulation, which can coordinate distributed photovoltaic inverters to perform orderly primary frequency modulation tasks, and use a particle swarm algorithm to finely adjust and optimize the parameters of photovoltaic inverters in each group, thereby improving the rough and simple inefficient frequency modulation operation, ensuring that the entire frequency modulation process has good controllability and stability, and by changing the primary frequency modulation response order of each group of photovoltaic inverters in rotation, taking into account the balance of the service life of each group of photovoltaic inverters.

[0006] The present invention adopts the following technical solution.

[0007] The first aspect of the present invention provides a cluster frequency modulation control method for distributed photovoltaic participating in primary frequency modulation, comprising: For areas where distributed photovoltaics participate in primary frequency regulation, select photovoltaic inverters that meet the basic requirements of frequency regulation in the area; The weighted fuzzy C-means clustering algorithm is used to group the selected PV inverters according to their rated power capacity and geographical location. According to the groups of photovoltaic inverters, set and regularly change the frequency modulation start-up sequence of each group of photovoltaic inverters; The particle swarm algorithm is used to optimize the primary frequency modulation parameters of each group of photovoltaic inverters, and the optimization results are used for frequency modulation control.

[0008] Preferably, the basic requirements for frequency modulation are: , in, , , They are the average shortest primary frequency regulation response time, average rated power capacity, and average maximum droop coefficient absolute value of photovoltaic inverters in the region; , They are the comprehensive performance index threshold of the photovoltaic inverter and the maximum reaction time limit of the local power grid to the photovoltaic inverter participating in the primary frequency regulation; In the primary frequency regulation, the photovoltaic inverter i The shortest frequency modulation response time that can be set within its configurable range; For photovoltaic inverter i Rated power capacity; In the primary frequency regulation, the photovoltaic inverter i The maximum absolute value of the droop coefficient that can be set within the settable range.

[0009] Preferably, the weighted fuzzy C-means clustering algorithm is used to group the screened photovoltaic inverters according to the rated power capacity and geographical location of the photovoltaic inverters, as follows: First, randomly set the geographical location of each vector center, and then calculate the membership degree of each photovoltaic inverter to each vector center: (5) in, For photovoltaic inverter i Geographical location With j The geographical location of the center of the vector The Euclidean distance of For photovoltaic inverter i Geographical location With r The geographical location of the center of the vector The Euclidean distance of μ ij For photovoltaic inverter i For j The degree of membership of the vector center is j The vector center corresponds to the j PV inverter groups; m is the fuzzy weighted index; c Indicates that n The number of groups into which the PV inverters are clustered; n The total number of photovoltaic inverters that meet the basic frequency regulation requirements in the region; And according to the membership degree, the geographical location of each vector center is updated as: (6) Based on formulas (5)-(6), the membership degree and the geographical location of each vector center are updated through continuous iteration until the change of the geographical location of the vector center in the iteration is less than the set condition. At this time, the clustering result is obtained according to formula (7): : (7) The value is 0 or 1. If it is 1, it means the photovoltaic inverter i Belong to j PV inverter groups, If it is 1, it means the photovoltaic inverter i Does not belong to j PV inverter groups.

[0010] Preferably, the objective function for iteratively updating the membership degree and the geographical location of each vector center is: (2) (3) in, w i For photovoltaic inverter i The weight coefficient of p i For photovoltaic inverter i Rated power capacity; p lFor photovoltaic inverter l Rated power capacity; | k | i max For photovoltaic inverter i The absolute value of the maximum droop coefficient that can be set; | k | l max For photovoltaic inverter l The absolute value of the maximum droop factor that can be set.

[0011] Preferably, the The following constraints are met: (4) in, is the fuzzy weighted partitioning matrix; is a real matrix.

[0012] Preferably, the setting and periodic change of the frequency modulation start-up sequence of each group of photovoltaic inverters according to the group of each photovoltaic inverter includes: The groups of c photovoltaic inverters are respectively recorded as group 1, group 2, ..., group c; The primary frequency modulation start-up sequence of the photovoltaic inverters in group c is set to group 1, group 2, ..., group c; After the set period, the primary frequency modulation start-up order of the photovoltaic inverters of group c is changed to: group 2, ..., group c, group 1; And so on, every set period, the frequency modulation start-up sequence of each group of photovoltaic inverters is changed once.

[0013] Preferably, when the startup sequence is set to group 1, group 2, ..., group c, if the grid frequency deviates and requires the photovoltaic inverter to perform a frequency modulation, the photovoltaic inverter in group 1 will start the frequency modulation first; if the grid frequency still does not return to the normal range within the set time, the photovoltaic inverter in group 2 will also start the frequency modulation; and so on, until all groups have started the frequency modulation.

[0014] Preferably, the primary frequency modulation parameters include a primary frequency modulation reaction time, a frequency modulation start frequency threshold, a frequency modulation end frequency threshold and an absolute value of a droop coefficient.

[0015] Preferably, the objective function of optimizing the primary frequency modulation parameters of each group of photovoltaic inverters using the particle swarm algorithm is: (8) in, t recThe time required for the grid frequency to recover from abnormal to normal; The recovery time specified for the power grid frequency to recover from abnormal to normal.

[0016] Preferably, when the particle swarm algorithm is used to optimize the primary frequency modulation parameters of each group of photovoltaic inverters, each particle X Represents the primary frequency regulation parameters of all PV inverters.

[0017] The second aspect of the present invention provides a cluster frequency modulation control system in which distributed photovoltaics participate in primary frequency modulation, comprising: The inverter screening module is used to screen out photovoltaic inverters that meet the basic requirements of frequency regulation in areas where distributed photovoltaics participate in primary frequency regulation; An inverter grouping module is used to group the selected photovoltaic inverters according to the rated power capacity and geographical location of the photovoltaic inverters by using a weighted fuzzy C-means clustering algorithm; The start-up sequence adjustment module is used to set and regularly change the primary frequency modulation start-up sequence of each group of photovoltaic inverters according to the group of each photovoltaic inverter; The frequency modulation control module is used to optimize the primary frequency modulation parameters of each group of photovoltaic inverters using a particle swarm algorithm, and use the optimization results to perform frequency modulation control.

[0018] A third aspect of the present invention provides a terminal, comprising a processor and a storage medium; the storage medium is used to store instructions; the processor is used to operate according to the instructions to execute the steps of the method.

[0019] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method when executed by a processor.

[0020] Compared with the prior art, the beneficial effects of the present invention include at least: The present invention screens out photovoltaic inverters that meet the basic requirements of frequency regulation in the region and groups them, integrates and coordinates the operation of multiple groups of photovoltaic inverters, and effectively realizes precise frequency regulation within the overfrequency range. It can avoid the problem of a collective and simultaneous sharp drop in distributed photovoltaic output power when the power grid is overclocked. It is particularly suitable for large-scale and scattered distributed photovoltaic systems, and saves the cost of using communication network equipment between inverters.

[0021] The basic frequency modulation requirements of the present invention not only take into account the rigid requirements, but also introduce performance indicators that comprehensively consider the shortest primary frequency modulation response time, rated power capacity, and the maximum absolute value of the droop coefficient of the photovoltaic inverter. Photovoltaic inverters with excellent performance can be screened out, which is beneficial to improving the working efficiency of the primary frequency modulation.

[0022] The present invention utilizes a weighted fuzzy C-means clustering algorithm to divide distributed photovoltaic inverters into multiple groups according to their geographical locations and capacities, and then responds to frequency modulation services one by one in turn, which helps to stabilize the frequency modulation effect when the subsequent groups rotate the frequency modulation response sequence.

[0023] The weighted fuzzy C-means clustering algorithm of the present invention introduces a weight coefficient for each photovoltaic inverter. w i , and this coefficient fully considers the dual performance of the photovoltaic inverter, namely the rated power capacity and the absolute value of the maximum droop coefficient. The traditional weighted fuzzy C-means clustering method is only a fuzzy weighted index m Adjustments are made, so the differences between each PV inverter cannot be taken into account in the weighting.

[0024] The present invention sets and regularly changes the primary frequency modulation start-up sequence of each group of photovoltaic inverters according to the group of each photovoltaic inverter, and can rotate and change the primary frequency modulation response sequence of each group of photovoltaic inverters, which can better average the working intensity of each group of photovoltaic inverters, thereby facilitating the extension of the service life of the photovoltaic inverters and balancing the service life of each group of photovoltaic inverters.

[0025] The present invention utilizes a particle swarm algorithm to specifically and finely optimize the parameters of each group of photovoltaic inverters, which is beneficial to improving frequency modulation efficiency and controllability. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a flow chart of a cluster frequency modulation control method for distributed photovoltaic participating in primary frequency modulation according to the present invention; Figure 2 It is a frequency modulation control curve diagram of the photovoltaic inverter of the present invention in the overfrequency range. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the scope of protection of the present invention.

[0028] like Figure 1 As shown, Embodiment 1 of the present invention provides a cluster frequency modulation control method for distributed photovoltaic participating in primary frequency modulation, which is applicable to primary frequency modulation in the overfrequency field, and specifically includes the following steps: Step 1: For areas where distributed photovoltaics participate in primary frequency regulation, select photovoltaic inverters that meet the basic requirements of frequency regulation in the area; Further preferably, for areas where distributed photovoltaics participate in primary frequency modulation, the frequency modulation response time and comprehensive frequency modulation performance of distributed photovoltaic inverters participating in primary frequency modulation in the area are screened according to the frequency modulation needs of the power grid company in the area, specifically involving the average shortest primary frequency modulation response time of photovoltaic inverters in the area. , the average rated power capacity of photovoltaic inverters in the region , the average maximum absolute value of the droop coefficient of photovoltaic inverters in this region , Comprehensive performance index threshold of photovoltaic inverters in this region , the maximum response time limit of the local power grid to the PV inverter participating in the primary frequency regulation In this embodiment, there are two requirements. If one of them is not met, the photovoltaic inverter cannot participate in the primary frequency modulation work. The two requirements can be expressed by the formula: (1) The meanings of these two requirements from top to bottom are: 1) In the primary frequency modulation, the photovoltaic inverter is within its configurable range, and its shortest configurable primary frequency modulation response time Must be less than ; 2) In the primary frequency modulation, the photovoltaic inverter is within its adjustable range, and its comprehensive performance index is Must be greater than or equal to the comprehensive performance index threshold of the PV inverter in this area .

[0029] In the formula, , , They are the average shortest primary frequency regulation response time, average rated power capacity, and average maximum droop coefficient absolute value of photovoltaic inverters in the region; It is the comprehensive performance index of PV inverter participating in primary frequency regulation; , They are the comprehensive performance index threshold of the photovoltaic inverter and the maximum reaction time limit of the local power grid to the photovoltaic inverter participating in the primary frequency regulation; In the primary frequency regulation, the photovoltaic inverter i The shortest frequency modulation response time that can be set within the settable range; For photovoltaic inverter i Rated power capacity; In the primary frequency regulation, the photovoltaic inverter iThe maximum absolute value of the droop coefficient that can be set within its settable range.

[0030] The photovoltaic inverter that meets the above two basic requirements can be used as the photovoltaic inverter participating in the primary frequency regulation in the region.

[0031] It should be pointed out that For rigid requirements, please refer to the relevant regulations of the regional power grid; It reflects the frequency modulation performance of a photovoltaic inverter. When the value is greater than 1, it means that the frequency regulation performance and corresponding contribution of the photovoltaic inverter are high, exceeding the average value of the local area; When the value is significantly less than 1 and close to 0, it means that the frequency regulation comprehensive performance of the photovoltaic inverter is too low and is not suitable for participating in primary frequency regulation. If it participates in frequency regulation, frequent operations may affect its service life. Therefore, according to the specific conditions of the local power grid, The value of is 0.1).

[0032] It can be seen that the basic frequency modulation requirements of the present invention not only take into account the rigid requirements, but also introduce the comprehensive performance indicators of the photovoltaic inverter participating in the primary frequency modulation, so as to screen out photovoltaic inverters with excellent performance, which is beneficial to improving the working efficiency of the primary frequency modulation.

[0033] Step 2: Using the weighted fuzzy C-means clustering algorithm, cluster and group the selected photovoltaic inverters according to their rated power capacity and geographical location; Further preferably, among the photovoltaic inverters that can meet the basic requirements of frequency modulation, the photovoltaic inverters are clustered using weighted fuzzy C-means clustering according to the rated power capacity and geographical location of the photovoltaic inverters, as follows: First, suppose there are n PV inverters that meet the requirements for participating in primary frequency regulation (i.e., the inverters selected in step 1) n PV inverters), and these n The photovoltaic inverters are clustered into c Groups, is the geographical location set of each photovoltaic inverter, and the objective function of the weighted fuzzy C-means clustering method is: (2) in, μ ij For the i PV inverters for the j The degree of membership of the vector center is w i For the iThe weight coefficient of a photovoltaic inverter takes into account the rated capacity of the photovoltaic inverter and the absolute value of the maximum droop coefficient that can be set: (3) w i For photovoltaic inverter i The weight coefficient of p i For photovoltaic inverter i Rated power capacity; p l For photovoltaic inverter l Rated power capacity; | k | i max For photovoltaic inverter i The absolute value of the maximum droop coefficient that can be set; | k | l max For photovoltaic inverter l The absolute value of the maximum droop factor that can be set.

[0034] is the fuzzy weighted partitioning matrix, which satisfies the constraints: (4) in, is a real matrix with n rows and c columns respectively; Then c A set of fuzzy cluster center vectors; For the i PV inverter and j The Euclidean distance between the centers of the vectors. m It is a fuzzy weighted index, and the present invention sets it to a default value of 2.

[0035] The weighted fuzzy C-means clustering algorithm of the present invention introduces a weight coefficient for each photovoltaic inverter. w i , and this coefficient fully considers the dual performance of the photovoltaic inverter, namely the rated power capacity and the absolute value of the maximum droop coefficient. The traditional weighted fuzzy C-means clustering method only considers the parameters m Adjustments are made, so the weights cannot take into account the differences between each photovoltaic inverter. It should be noted that since step 1 has screened out all inverters that meet the maximum response time limit for a frequency modulation, w i The setting does not include the frequency modulation response time of each inverter.

[0036] Based on the above settings, the present invention first randomly sets the geographical location of each vector center according to the weighted fuzzy C-means clustering method, and then calculates the membership degree of each photovoltaic inverter to each vector center: (5) The geographical location of each vector center can be updated based on the membership degree as follows: (6) Therefore, the membership and the geographical location of each vector center are updated through continuous iteration until the change of the geographical location of the vector center in the iteration can be ignored. At this time, the membership has also tended to a constant value. Therefore, the clustering results need to be further determined to complete the clustering. To this end, the present invention can further obtain the clustering results according to the following formula .

[0037] (7) The value of can only be 0 or 1 if If it is equal to 1, it means i The photovoltaic inverter belongs to j Groups.

[0038] It can be understood that formula (7) is to determine the clustering result Z after formulas (2)-(6) are optimized. ij Because after the optimization of formulas (2)-(6) in the weighted fuzzy C-means clustering method, only the membership degree U ij , there is still one step to determine the clustering result, so the assistance of formula (7) is needed.

[0039] It can be seen that the final clustering result The membership and weight of all photovoltaic inverters are comprehensively considered.

[0040] Step 3: Set and regularly change the frequency modulation start-up sequence of each group of photovoltaic inverters according to the group of each photovoltaic inverter; Further preferably, the primary frequency modulation response order of the photovoltaic inverters is arranged according to the grouping of the clustering results, and a plan is made to rotate the primary frequency modulation response order of each group after a certain period of time in the future, so as to avoid the photovoltaic inverters in a certain group being always in the first start-up round state, thereby affecting the service life of the photovoltaic inverters in this group.

[0041] For example, assume that in step 2, the photovoltaic inverters participating in the primary frequency modulation are clustered into three groups, which are recorded as group 1, group 2, and group 3. Then, when the grid frequency deviates and the photovoltaic inverter needs a primary frequency modulation, the photovoltaic inverters in group 1 will start the primary frequency modulation first. If the grid frequency still does not return to the normal range within a certain period of time (for example, 1 second), the photovoltaic inverters in group 2 will also start the primary frequency modulation. If the grid frequency still does not return within a certain period of time, the photovoltaic inverters in group 3 will also start the primary frequency modulation. And so on, until all groups start the primary frequency modulation.

[0042] It should be noted that the weighted fuzzy C-means clustering method tends to cluster all photovoltaic inverters into groups with similar total power according to their geographical locations. Therefore, in the above example, the photovoltaic inverters of Group 1, Group 2, and Group 3 actually have very close total rated power, which is convenient for subsequent sequential rotation, that is, after a certain period of time (such as 6 months), the startup sequence of a frequency modulation is changed to: Group 2, Group 3, Group 1, and then it can be changed to: Group 3, Group 1, Group 2. The purpose of such rotation is to avoid the photovoltaic inverters in a certain group being in the first startup round state during a frequency modulation, which in turn affects the service life of the photovoltaic inverters in this group. The clustering method that considers geographical location has the advantages of being clear and convenient and conducive to unified management and control in the later stage.

[0043] Step 4: Use the particle swarm algorithm to optimize the primary frequency modulation parameters of each group of photovoltaic inverters, and use the optimization results to perform frequency modulation control.

[0044] It is understandable that the inverters of each group join the frequency modulation work in sequence. For example, if the grid frequency has not recovered after group 1 joins the work, group 2 will join next, and so on. However, step 4 is not to optimize the frequency modulation parameters of only the group that is starting up, which may be too late, but to optimize the parameters of all the inverters of all groups (whose startup order has been arranged) at the moment.

[0045] Further preferably, the frequency modulation reaction time of each group of photovoltaic inverters is calculated by using the particle swarm. t g , Frequency modulation start frequency threshold , FM termination frequency threshold , the absolute value of the droop coefficient Optimize to get the best FM control effect.

[0046] Specifically, the objective function optimized by particle swarm optimization can be expressed as: (8) in, t rec It is the time required for the grid frequency to recover from abnormal to normal. The recovery time of the grid frequency from abnormal to normal is specified for the grid. It should be noted that t rec There is no explicit analytical relationship between the parameters of each group of inverters contained in the particles in the particle swarm algorithm, so it cannot be directly given in the form of a formula. t rec Based on the regional power grid model, the corresponding results can be obtained by using simulation software such as Matlab / Simulink when the inverter group, reaction sequence and reaction time are given.

[0047] When using the particle swarm algorithm, the present invention sets: X represents the sum of information of all photovoltaic inverters. Assuming that the photovoltaic inverters participating in primary frequency modulation are clustered into two groups, the particle X The settings are as follows: (9) in, t g1 Represents the response time of the PV inverter in the first group to a frequency modulation. f g1,1 It is the abnormal frequency threshold when the photovoltaic inverter in the first group starts to reduce power during the primary frequency regulation process; f g1,2 It is the maximum abnormal frequency threshold corresponding to the power reduction during the primary frequency modulation process of the photovoltaic inverter in the first group; |k| g1 is the absolute value of the droop coefficient of the PV inverter in the first group during the primary frequency modulation. g2 The parameters are the corresponding information of the photovoltaic inverters in the second group.

[0048] The optimization process of the particle swarm algorithm is continuously iterated until the reduction of the objective function value is gradually reduced to a negligible value. At this point, the iteration stops and the final photovoltaic inverter layering and the optimization results of each layer parameter are output. All the photovoltaic inverters mentioned above refer to all groups of photovoltaic inverters after the grouping and startup sequence are completed. Each group of photovoltaic inverters uses the same set of frequency modulation parameters.

[0049] Embodiment 2 of the present invention provides a cluster frequency modulation control system in which distributed photovoltaics participate in primary frequency modulation, including: The inverter screening module is used to screen out photovoltaic inverters that meet the basic requirements of frequency regulation in areas where distributed photovoltaics participate in primary frequency regulation; An inverter grouping module is used to group the selected photovoltaic inverters according to the rated power capacity and geographical location of the photovoltaic inverters by using a weighted fuzzy C-means clustering algorithm; The start-up sequence adjustment module is used to set and regularly change the primary frequency modulation start-up sequence of each group of photovoltaic inverters according to the group of each photovoltaic inverter; The frequency modulation control module is used to optimize the primary frequency modulation parameters of each group of photovoltaic inverters using a particle swarm algorithm, and use the optimization results to perform frequency modulation control.

[0050] Embodiment 3 of the present invention provides a terminal, including a processor and a storage medium; the storage medium is used to store instructions; the processor is used to operate according to the instructions to execute the steps of the method.

[0051] Embodiment 4 of the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method when executed by a processor.

[0052] Compared with the prior art, the beneficial effects of the present invention include at least: The present invention screens out photovoltaic inverters that meet the basic requirements of frequency regulation in the region and groups them, integrates and coordinates the operation of multiple groups of photovoltaic inverters, and effectively realizes precise frequency regulation within the overfrequency range. It can avoid the problem of a collective and simultaneous sharp drop in distributed photovoltaic output power when the power grid is overclocked. It is particularly suitable for large-scale and scattered distributed photovoltaic systems, and saves the cost of using communication network equipment between inverters.

[0053] The basic frequency modulation requirements of the present invention not only take into account the rigid requirements, but also introduce performance indicators that comprehensively consider the shortest primary frequency modulation response time, rated power capacity, and the maximum absolute value of the droop coefficient of the photovoltaic inverter. Photovoltaic inverters with excellent performance can be screened out, which is beneficial to improving the working efficiency of the primary frequency modulation.

[0054] The present invention utilizes a weighted fuzzy C-means clustering algorithm to divide distributed photovoltaic inverters into multiple groups according to their geographical locations and capacities, and then responds to frequency modulation services one by one in turn, which helps to stabilize the frequency modulation effect when the subsequent groups rotate the frequency modulation response sequence.

[0055] The weighted fuzzy C-means clustering algorithm of the present invention introduces a weight coefficient for each photovoltaic inverter. w i , and this coefficient fully considers the dual performance of the photovoltaic inverter, namely the rated power capacity and the absolute value of the maximum droop coefficient. The traditional weighted fuzzy C-means clustering method only considers the parameters m Adjustments are made, so the differences between each PV inverter cannot be taken into account in the weighting.

[0056] The present invention sets and regularly changes the primary frequency modulation start-up sequence of each group of photovoltaic inverters according to the group of each photovoltaic inverter, and can rotate and change the primary frequency modulation response sequence of each group of photovoltaic inverters, which can better average the working intensity of each group of photovoltaic inverters, thereby facilitating the extension of the service life of the photovoltaic inverters and balancing the service life of each group of photovoltaic inverters.

[0057] The present invention utilizes a particle swarm algorithm to specifically and finely optimize the parameters of each group of photovoltaic inverters, which is beneficial to improving frequency modulation efficiency and controllability.

[0058] The present invention utilizes the characteristic that the weighted fuzzy C-means clustering algorithm tends to balanced clustering, and can change the primary frequency modulation response sequence of each group of photovoltaic inverters in rotation, so as to balance the service life of each group of photovoltaic inverters.

[0059] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0060] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the above. The computer-readable storage medium used herein is not to be interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0061] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0062] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect through the Internet). In some embodiments, by using the state information of the computer-readable program instructions to personalize an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), the electronic circuit may execute the computer-readable program instructions, thereby implementing various aspects of the present disclosure.

[0063] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A cluster frequency modulation control method for distributed photovoltaic participating in primary frequency modulation, characterized in that: include: For areas where distributed photovoltaics participate in primary frequency regulation, select photovoltaic inverters that meet the basic requirements of frequency regulation in the area; The weighted fuzzy C-means clustering algorithm is used to group the selected PV inverters according to their rated power capacity and geographical location. According to the groups of photovoltaic inverters, set and regularly change the frequency modulation start-up sequence of each group of photovoltaic inverters; The particle swarm algorithm is used to optimize the primary frequency modulation parameters of each group of photovoltaic inverters, and the optimization results are used for frequency modulation control.

2. A cluster frequency modulation control method for distributed photovoltaic participating in primary frequency modulation according to claim 1, characterized in that: The basic requirements for frequency modulation are: ， in, , , They are the average shortest primary frequency regulation response time, average rated power capacity, and average maximum droop coefficient absolute value of photovoltaic inverters in the region; , They are the comprehensive performance index threshold of the photovoltaic inverter and the maximum reaction time limit of the local power grid to the photovoltaic inverter participating in the primary frequency regulation; In the primary frequency regulation, the photovoltaic inverter i The shortest frequency modulation response time that can be set within its configurable range; For photovoltaic inverter i Rated power capacity; In the primary frequency regulation, the photovoltaic inverter i The maximum absolute value of the droop coefficient that can be set within the settable range.

3. The cluster frequency modulation control method for distributed photovoltaic participating in primary frequency modulation according to claim 1, characterized in that: The weighted fuzzy C-means clustering algorithm is used to group the selected photovoltaic inverters according to the rated power capacity and geographical location of the photovoltaic inverters, as follows: First, randomly set the geographical location of each vector center, and then calculate the membership degree of each photovoltaic inverter to each vector center: （5） in, For photovoltaic inverter i Geographical location With j The geographical location of the center of the vector The Euclidean distance of For photovoltaic inverter i Geographical location With r The geographical location of the center of the vector The Euclidean distance of μ ij For photovoltaic inverter i For j The degree of membership of the vector center is j The vector center corresponds to the j PV inverter groups; m is the fuzzy weighted index; c Indicates that n The number of groups into which the PV inverters are clustered; n The total number of photovoltaic inverters that meet the basic frequency regulation requirements in the region; And according to the membership degree, the geographical location of each vector center is updated as: （6） in, w i For photovoltaic inverter i The weight coefficient of Based on formulas (5)-(6), the membership degree and the geographical location of each vector center are updated through continuous iteration until the change of the geographical location of the vector center in the iteration is less than the set condition. At this time, the clustering result is obtained according to formula (7): : （7） The value is 0 or 1. If it is 1, it means the photovoltaic inverter i Belong to j PV inverter groups, If it is 1, it means the photovoltaic inverter i Does not belong to j PV inverter groups.

4. A cluster frequency modulation control method for distributed photovoltaic participating in primary frequency modulation according to claim 3, characterized in that: The objective function of the loop iterative update of the membership degree and the geographical location of each vector center is: （2） （3） in, p i For photovoltaic inverter i Rated power capacity; p l For photovoltaic inverter l Rated power capacity; | k | i max For photovoltaic inverter i The absolute value of the maximum droop factor that can be set; | k | l max For photovoltaic inverter l The absolute value of the maximum droop factor that can be set.

5. A cluster frequency modulation control method for distributed photovoltaic participating in primary frequency modulation according to claim 4, characterized in that: Said The following constraints are met: （4） in, is the fuzzy weighted partitioning matrix; is a real matrix with n rows and c columns respectively.

6. The cluster frequency modulation control method for distributed photovoltaic participating in primary frequency modulation according to claim 1, characterized in that: The setting and periodic change of the frequency modulation start-up sequence of each group of photovoltaic inverters according to the group of each photovoltaic inverter includes: The groups of c photovoltaic inverters are respectively recorded as group 1, group 2, ..., group c; The primary frequency modulation start-up sequence of the photovoltaic inverters in group c is set to group 1, group 2, ..., group c; After the set period, the primary frequency modulation start-up order of the photovoltaic inverters of group c is changed to: group 2, ..., group c, group 1; And so on, every set period, the frequency modulation start-up sequence of each group of photovoltaic inverters is changed once.

7. A cluster frequency modulation control method for distributed photovoltaic participating in primary frequency modulation according to claim 6, characterized in that: When the startup sequence is set to group 1, group 2, ..., group c, if the grid frequency deviates and requires a frequency modulation of the photovoltaic inverter, the photovoltaic inverter in group 1 will start the frequency modulation first; if the grid frequency still does not return to the normal range within the set time, the photovoltaic inverter in group 2 will also start the frequency modulation; and so on, until all groups have started the frequency modulation.

8. The cluster frequency modulation control method for distributed photovoltaic participating in primary frequency modulation according to claim 1, characterized in that: The primary frequency modulation parameters include a primary frequency modulation reaction time, a frequency modulation start frequency threshold, a frequency modulation end frequency threshold and an absolute value of a droop coefficient.

9. The cluster frequency modulation control method for distributed photovoltaic participating in primary frequency modulation according to claim 1, characterized in that: The objective function of optimizing the primary frequency modulation parameters of each group of photovoltaic inverters using the particle swarm algorithm is: （8） in, t rec The time required for the grid frequency to recover from abnormal to normal; The recovery time specified for the power grid frequency to recover from abnormal to normal.

10. The cluster frequency modulation control method for distributed photovoltaic participating in primary frequency modulation according to claim 1, characterized in that: When the particle swarm algorithm is used to optimize the primary frequency modulation parameters of each group of photovoltaic inverters, each particle X Represents the primary frequency regulation parameters of all PV inverters.

11. A distributed photovoltaic cluster frequency modulation control system for primary frequency modulation, using the method described in any one of claims 1 to 10, characterized in that: The system comprises: The inverter screening module is used to screen out photovoltaic inverters that meet the basic requirements of frequency regulation in areas where distributed photovoltaics participate in primary frequency regulation; An inverter grouping module is used to group the selected photovoltaic inverters according to the rated power capacity and geographical location of the photovoltaic inverters by using a weighted fuzzy C-means clustering algorithm; The start-up sequence adjustment module is used to set and regularly change the primary frequency modulation start-up sequence of each group of photovoltaic inverters according to the group of each photovoltaic inverter; The frequency modulation control module is used to optimize the primary frequency modulation parameters of each group of photovoltaic inverters using a particle swarm algorithm, and use the optimization results to perform frequency modulation control.

12. A terminal comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1-10.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

Images