A big data-based distributed photovoltaic system load management method and system

Abstract

The application provides a big data-based distributed photovoltaic system load management and control method and system, and belongs to the technical field of power systems, and specifically comprises the following steps: using updated data of a load management and control weather type and on-grid power of the distributed photovoltaic system in the load management and control weather type to determine a management and control processing strategy of the distributed photovoltaic system, performing management and control processing of the distributed photovoltaic system based on the management and control processing strategy, using management and control processing data of a subsystem and the load management and control subsystem in the load management and control weather type to determine a management and control optimization target of the load management and control weather type, and improving the reliability of frequency modulation response processing based on the photovoltaic system.

Description

Technical Field

[0001] This invention belongs to the field of power system technology, and in particular relates to a method and system for load management of distributed photovoltaic systems based on big data. Background Technology

[0002] With the advancement of the "dual carbon" goals, distributed photovoltaic (PV) power is being connected to the grid on a large scale at the user side. However, distributed PV is characterized by intermittency and fluctuation, and its output is greatly affected by weather conditions. Large-scale, high-proportion grid connection can lead to problems such as voltage exceeding limits in the distribution network, power flow reversal, and transformer overload.

[0003] To address the aforementioned technical issues, for example, the invention patent application CN202511527095.7, "Adaptive Control Method and System for Distributed Photovoltaic Power Grid Access," intelligently controls the output power of the photovoltaic system to achieve grid load balance and voltage stability, optimize photovoltaic power generation efficiency, and reduce power waste. Through real-time monitoring and automatic adjustment, the system can flexibly respond to grid load changes, improve the collaborative working capability between photovoltaic power generation and the grid, enhance the grid's ability to cope with abnormal fluctuations, reduce equipment damage risks, and improve the system's flexibility and scalability. However, the following shortcomings still exist: With the increasing number of photovoltaic subsystems connected to the grid, especially newly connected photovoltaic subsystems, it is difficult to effectively obtain their operating load and frequency regulation performance. Therefore, determining the load management scheme under different weather conditions and thus achieving a reliable assessment of frequency regulation capability and operating performance has become an urgent technical problem to be solved.

[0004] To address the aforementioned technical issues, this application provides a method and system for load management of distributed photovoltaic systems based on big data. Summary of the Invention

[0005] To achieve the objectives of this invention, the following technical solution is adopted: Specifically, this application provides a load management method for distributed photovoltaic systems based on big data, which includes: S1 uses the new data of the distributed photovoltaic system to determine the operation data of different subsystems in the distributed photovoltaic system under different weather types, uses the operation data to determine the operation demand type of the subsystem, and determines the load control weather type update strategy based on the operation demand type of different subsystems in the target area and the degree of overlap of the demand matching weather types of the subsystems. S2 determines the management and control strategy for the distributed photovoltaic system based on the updated data of the load control weather type and the on-grid electricity of the distributed photovoltaic system in the load control weather type. S3 performs management and control processing of the distributed photovoltaic system based on the management and control processing strategy, and determines the management and control optimization target of the load control weather type using the management and control processing data of the subsystem and the load control subsystem in the load control weather type.

[0006] The beneficial effects of this invention are as follows: Based on the operational demand types of different subsystems in the target area and the degree of overlap between subsystem demand and weather types, the update strategy for load control weather types is determined. A hierarchical, multi-condition judgment mechanism comprehensively assesses the data scarcity and distribution characteristics of subsystems within the area, thereby deciding whether to adopt a "relaxed control strategy" or a "strict control strategy." When there are many subsystems with scarce data (one type of demand) in the area, it indicates a weak overall data foundation, requiring a strict strategy to expand the control coverage. When there are fewer data-scarce subsystems, the degree of overlap of subsystems with insufficient data under different weather types is further analyzed. This ensures that the load control strategy can adaptively match the regional data accumulation status, achieving efficient allocation of control resources and reducing the impact on internet access.

[0007] By using the control and management data of the subsystems and the load control subsystems in the load control weather types, the control optimization targets for the load control weather types are determined. Based on the results of the executed load control processing, the control impact of different subsystems under each load control weather type is evaluated. This determines the actual response capability of a type of demand subsystem participating in frequency regulation under what conditions, and identifies which load control weather types need to be listed as control optimization targets, i.e., to stop load control processing on them. The system can dynamically identify weather types that require cessation of load control processing when the control impact is significant, thereby reducing the impact of load control on the overall power supply reliability.

[0008] Furthermore, the new data for the distributed photovoltaic system includes newly added subsystems in the target area and the operating data of the newly added subsystems.

[0009] Furthermore, the operating data of the subsystem under different weather types is determined based on the operating time of the subsystem under different weather types.

[0010] Furthermore, the method for determining the operational requirement type of the subsystem is as follows: S11 determines the operating time of the subsystem under different weather types based on the operating data; S12 uses the runtime to determine the weather type whose runtime of the subsystem is less than a preset runtime threshold, and uses the weather type whose runtime of the subsystem is less than the preset runtime threshold as the demand matching weather type of the subsystem. S13 determines the operational requirement type of the subsystem by matching the weather type with the requirements of the subsystem.

[0011] Furthermore, the control and processing data of the subsystem is determined according to the load control weather type that the subsystem needs to perform load control processing on.

[0012] Furthermore, the method for determining the control optimization target for the load control weather type is as follows: S41 uses the control and processing data of the subsystem to determine the load control weather type that the subsystem needs to perform load control processing, and uses the load control weather type that the subsystem needs to perform load control processing as the matching control weather type; S42 determines the load control subsystem in the load control weather type based on the load control subsystem in the load control weather type; S43 determines the control optimization objective of the load control weather type based on the load control subsystem in the load control weather type and the matching control weather types of different load control subsystems.

[0013] Secondly, the present invention provides a computer system comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-described method for load management of a distributed photovoltaic system based on big data when running the computer program.

[0014] Other features and advantages will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0015] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0016] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0017] Figure 1 This is a flowchart of a load management method for distributed photovoltaic systems based on big data. Figure 2 This is a flowchart illustrating the method for determining the operational requirements type of a subsystem; Figure 3 This is a flowchart illustrating the method for determining the update strategy for load control weather types; Figure 4This is a flowchart illustrating the method for determining the management and control strategies for distributed photovoltaic systems. Detailed Implementation

[0018] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, they are provided so that the invention will be thorough and complete, and the concept of the exemplary embodiments will be fully conveyed to those skilled in the art. The same reference numerals in the drawings denote the same or similar structures, and therefore their detailed description will be omitted.

[0019] The terms “a,” “one,” “the,” and “the” are used to indicate the existence of one or more elements / components / etc.; the terms “including” and “having” are used to indicate an open-ended meaning of inclusion and that other elements / components / etc. may exist in addition to the listed elements / components / etc.

[0020] Example 1 To solve the above problems, according to one aspect of the present invention, such as Figure 1 As shown, a load management method for distributed photovoltaic systems based on big data is provided, specifically including: S1 uses the new data of the distributed photovoltaic system to determine the operation data of different subsystems in the distributed photovoltaic system under different weather types, uses the operation data to determine the operation demand type of the subsystem, and determines the load control weather type update strategy based on the operation demand type of different subsystems in the target area and the degree of overlap of the demand matching weather types of the subsystems. Furthermore, the new data for the distributed photovoltaic system includes newly added subsystems in the target area and the operating data of the newly added subsystems.

[0021] Furthermore, the operating data of the subsystem under different weather types is determined based on the operating time of the subsystem under different weather types.

[0022] It should be noted that the subsystem is divided based on whether the photovoltaic equipment uses the same grid-connected device.

[0023] Specifically, such as Figure 2 As shown, the method for determining the operational requirement type of the subsystem is as follows: For newly added subsystems in distributed photovoltaic systems, by analyzing their runtime data under different refined weather types, the system accurately identifies weather types with insufficient operational data accumulation, thereby determining the subsystem's operational needs. The core logic is that the shorter the runtime of a weather type, the scarcer the historical data for that subsystem under such weather conditions, and the more urgent its need for targeted data collection and model updates. By quantifying the number of "demand-matched weather types," the subsystem's needs are categorized into different levels, thus providing a basis for developing differentiated management and control strategies to ensure reliable operation of the subsystem under certain weather types and to provide a decision-making basis for identifying the reliability of frequency regulation processing.

[0024] S11 determines the operating time of the subsystem under different weather types based on the operating data; The operational data refers to the time-series data, encompassing various operational status parameters, automatically recorded and reported by the system after the new subsystem is put into operation. The weather type refers to a refined weather category generated based on three meteorological elements: temperature, wind speed, and solar irradiance, through cluster analysis or interval division methods. Examples include "high temperature, strong sunlight, light wind," "medium temperature, medium sunlight, medium wind," and "low temperature, weak sunlight, strong wind," used to comprehensively characterize the environmental state of the photovoltaic equipment. The operating time refers to the total duration for which the photovoltaic equipment of the subsystem is in power generation or grid-connected operation under a specific weather type.

[0025] This step is fundamental to all subsequent analyses. Refining the weather types to more than 10 allows for a more accurate reflection of the operational characteristics of photovoltaic equipment under different environmental combinations, avoiding the obscuring of data gaps under specific operating conditions due to overly coarse weather classifications. This involves extracting quantifiable indicators reflecting the operational activity of subsystems from the raw, continuous operational data, correlated with the refined weather types.

[0026] The significance of this step lies in achieving ultra-high precision and multi-dimensional scenario-based characterization of the operational status of the newly added subsystems. It links the "abundance" of accumulated data with the complex external environment, including temperature, wind speed, and light intensity, providing an objective and measurable basis for identifying data shortcomings. This avoids analytical biases caused by singular or generalized weather classifications, making subsequent identification of data shortcomings more targeted and practically significant.

[0027] S12 uses the runtime to determine the weather type whose runtime of the subsystem is less than a preset runtime threshold, and uses the weather type whose runtime of the subsystem is less than the preset runtime threshold as the demand matching weather type of the subsystem. The preset runtime threshold is a pre-defined time threshold used to determine whether data accumulation is sufficient under a specific refined weather type. The demand-matching weather type refers to weather types where the runtime of the subsystem is lower than the preset runtime threshold. Under such weather types, due to insufficient subsystem runtime, the sample size of operational data under this specific environmental combination is small, making it difficult to support accurate analysis and model construction of the subsystem's operational characteristics under this condition.

[0028] Setting a preset runtime threshold introduces an objective evaluation standard, transforming continuous runtime values ​​into discrete judgments of "sufficient" or "insufficient." Each weather type naturally occurs at a different frequency. By identifying all weather types with runtimes less than the threshold, the system can accurately pinpoint "weak links" in data accumulation, regardless of whether the weather type is common or rare. Defining these weather types as "demand-matching weather types" clarifies the areas requiring focused attention and reinforcement.

[0029] The significance of this step lies in concretizing the abstract problem of insufficient data into a series of clear, actionable, and detailed weather type lists. It provides a direct input variable for subsequent demand grading, namely the number of "demand-matching weather types." Through this list, system administrators can clearly understand under which specific combinations of temperature, wind speed, and sunlight conditions the subsystem is "data-scarce."

[0030] S13 determines the operational requirement type of the subsystem by matching the weather type with the requirements of the subsystem.

[0031] It is understood that the demand-matched weather type is a weather type with a runtime shorter than a preset runtime threshold. In this case, due to the short runtime, the subsystem needs to run sufficiently in the above-mentioned weather type in order to accumulate more operating data, which will lay the foundation for the later construction of load characteristics of the photovoltaic equipment in the subsystem and the identification of the reliability of frequency regulation processing.

[0032] The operational requirement types are categories derived from classifying the operational requirements of a subsystem based on the urgency and complexity of its data collection and model updates. These include three types: Type 1, Type 2, and Type 3, with Type 1 having the highest priority, followed by Type 2, and then Type 3, which has the lowest priority.

[0033] Different subsystems face varying degrees of data scarcity (i.e., the number of weather types matching demand). Therefore, subsystems need to be categorized based on the severity of the scarcity (reflected by the number of weather types matching demand) to implement differentiated response strategies. Without categorization, all subsystems will face the same processing methods, leading to inefficient and inefficient resource allocation.

[0034] The significance of this step lies in transforming objective, refined data analysis results into actionable decision-making instructions. By mapping the "number of weather types matching demand" to "operational demand types," the system can automatically determine the current status of each subsystem and provide clear priority guidance for subsequent resource scheduling and planning. Subsystems with one type of demand (corresponding to a large number of weather types matching demand) should be prioritized for data collection and model update tasks, while subsystems with three types of demand (corresponding to fewer or zero weather types matching demand) can operate in normal mode. This optimizes the allocation of management resources and ensures that data collection and model updates accurately cover data-scarce operating conditions.

[0035] Furthermore, based on the weather type matching the needs of the subsystem, the operational needs type of the subsystem is determined, specifically including: Based on the number of weather types that the subsystem's needs match, the operational needs type of the subsystem is determined according to the needs type corresponding to the number of weather types that the needs match.

[0036] It is understood that the operational requirement types include a first type of requirement, a second type of requirement, and a third type of requirement, wherein the first type of requirement is greater than the second type of requirement, and the second type of requirement is greater than the third type of requirement.

[0037] This embodiment provides a method for determining the subsystem operation requirement type of a distributed photovoltaic system, applied to a photovoltaic management system that has been connected to a cloud platform, which contains multiple distributed photovoltaic subsystems.

[0038] First, the cloud platform acquires the operational data of the newly added subsystem D in the target area. Subsystem D is defined based on whether the photovoltaic equipment uses the same grid-connected device; all its photovoltaic modules are connected to the grid through the same inverter. Since subsystem D was put into operation, the cloud platform has continuously collected its daily operational data, such as power generation, voltage, and current, and performs spatiotemporal matching with meteorological data provided by meteorological service providers. Meteorological data includes three dimensions: temperature, wind speed, and sunshine duration. Based on these three dimensions, the cloud platform uses a K-means clustering algorithm to classify weather types into 12 categories: Type 1 (high temperature, strong sunlight, light wind), Type 2 (high temperature, strong sunlight, moderate wind), Type 3 (moderate temperature, moderate sunlight, light wind), Type 4 (moderate temperature, moderate sunlight, moderate wind), Type 5 (low temperature, weak sunlight, light wind), Type 6 (low temperature, weak sunlight, strong wind), Type 7 (high temperature, weak sunlight, light wind), Type 8 (high temperature, weak sunlight, strong wind), Type 9 (moderate temperature, strong sunlight, strong wind), Type 10 (moderate temperature, weak sunlight, strong wind), Type 11 (low temperature, strong sunlight, light wind), and Type 12 (low temperature, strong sunlight, strong wind). The cloud platform then tags and archives the operational data according to these 12 weather types.

[0039] Secondly, the cloud platform executes step S11: based on the operational data of subsystem D, it calculates its cumulative runtime under each of the 12 weather types. Statistics show that subsystem D's runtime is 200 hours under weather type 1, 150 hours under type 2, 100 hours under type 3, 90 hours under type 4, 4 hours under type 5, 3 hours under type 6, 60 hours under type 7, 90 hours under type 8, 80 hours under type 9, 60 hours under type 10, 60 hours under type 11, and 90 hours under type 12.

[0040] Then, the cloud platform executes step S12: the system reads a preset runtime threshold, which is set to 50 hours in this embodiment. The cloud platform compares the runtime of subsystem D under each weather type with this threshold. The comparison shows that the runtime of subsystem D under weather type 5 (low temperature, low light, light wind) (4 hours) is less than 50 hours, and the runtime under weather type 6 (low temperature, low light, strong wind) (3 hours) is less than 50 hours. The runtime under the remaining 10 weather types is greater than or equal to 50 hours. Therefore, the cloud platform determines weather types 5 and 6 as the required weather types for subsystem D.

[0041] Next, the cloud platform executes step S13: The cloud platform counts the number of weather types that match the demand of subsystem D, and the result is 2 (i.e., weather type 5 and weather type 6). The system's internal preset mapping rule is: when the number of weather types that match the demand is 0, it corresponds to a third-class demand type; when the number is 1, it corresponds to a second-class demand type; and when the number is greater than or equal to 2, it corresponds to a first-class demand type. According to this rule, the cloud platform determines the operational demand type of subsystem D as a first-class demand type.

[0042] The significance and value of this invention lies in providing an adaptive mechanism for assessing the operational needs of newly added photovoltaic subsystems based on refined weather type classification. First, it defines weather types as at least 10 refined categories based on temperature, wind speed, and solar irradiance. This allows the assessment of subsystem data accumulation to move beyond simple sunny / rainy weather classifications, accurately capturing operational characteristics under the coupled effects of multiple environmental factors, significantly improving the accuracy and scenario coverage of the assessment. Second, by utilizing "preset operating time thresholds" and "demand-matched weather types," an objective data sufficiency criterion is established within a refined weather framework. This accurately identifies weak data points under specific temperature, wind speed, and solar irradiance combinations, eliminating the data omissions caused by the coarse weather classifications of traditional methods. Finally, by mapping the number of "demand-matched weather types" to "operational demand types," scientific hierarchical management of the subsystem is achieved. This ensures that limited resources such as data acquisition equipment and computing resources are precisely allocated to specific weather conditions with the most scarce data, greatly improving the targeting of data acquisition.

[0043] Specifically, such as Figure 3 As shown, the method for determining the update strategy for the load control weather type is as follows: Based on the operational demand types of each subsystem in the target area, a dynamic update strategy for load control weather types is formulated to guide the initiation of load control measures under specific weather conditions. The core logic lies in using a hierarchical, multi-condition judgment mechanism to comprehensively assess the data scarcity and distribution characteristics of subsystems within the region, thereby determining whether to adopt a "lenient control strategy" or a "strict control strategy." When a large number of subsystems in the region exhibit data scarcity (a type of demand), it indicates a weak overall data foundation, necessitating a strict strategy to expand control coverage. When the number of data-scarce subsystems is small, the overlap of data-deficient subsystems under different weather types is further analyzed. High overlap indicates widespread data gaps in certain weather types, requiring focused control with a strict strategy; otherwise, a more refined judgment is needed based on the distribution of subsystems exhibiting a single demand type. This mechanism ensures that the load control strategy can adaptively match the regional data accumulation status, achieving efficient allocation of control resources.

[0044] S21 Based on the operational requirement type of the subsystem, determine a subsystem of a certain requirement type in the target area; The operational demand type refers to the category categorized based on the insufficient operating time of subsystems under different weather types. This includes three types: Type 1, Type 2, and Type 3. Type 1 demand indicates that the subsystem has insufficient data under many weather types, and its data acquisition and model updates have the highest priority. The target area refers to the geographical area where load management strategies need to be formulated, which includes multiple distributed photovoltaic subsystems. The Type 1 demand subsystems are those classified as Type 1 in the operational demand type determination; these subsystems have the most urgent need for data acquisition and model updates.

[0045] This step is the starting point for the entire strategy formulation. By selecting subsystems of a specific type of demand, we can focus on the group of subsystems in the region that have the most scarce data and require the highest priority to ensure operational reliability. The operational status of these subsystems directly determines the shortcomings of the overall data quality in the region, and therefore should serve as the core basis for subsequent strategy judgments.

[0046] The significance of this step lies in achieving a leap from assessing the needs of a single subsystem to evaluating the overall situation of the region. By statistically analyzing the number of subsystems of a particular type of demand, the severity of data scarcity within the region can be quickly grasped, providing the first layer of decision-making basis for subsequently selecting a "lenient" or "strict" control strategy.

[0047] It should be noted that if the number of subsystems of a certain demand type in the target area is greater than the preset threshold for the number of a certain demand type, the demand for load control is greater, so that the subsystems of a certain demand type can operate reliably in different weather types. Therefore, the update strategy for the load control weather type is a strict control strategy. As long as the proportion of the number of days with load control in the weather type is above the first proportion threshold, the weather type will be used as the load control weather type.

[0048] The preset threshold for the number of demand types is a pre-defined critical value used to determine whether the level of data scarcity in a region has reached a high level. The load control weather type refers to the weather type determined by the strategy to require load control measures (such as power rationing, power regulation, etc.). The lenient control strategy refers to a broader method for determining control types, characterized by relatively relaxed triggering conditions, aiming to expand the control coverage and ensure that data-scarce subsystems can obtain operational support under more weather conditions. The percentage of days requiring load control refers to the proportion of days requiring load control under a certain weather type out of the total number of days that weather type occurs. The first percentage threshold is used to determine whether this proportion has reached the critical value for activating a strict control strategy.

[0049] When there are a large number of subsystems of a particular demand type within a region, it indicates widespread data gaps, with many subsystems facing insufficient data under various weather conditions. In this situation, adopting an overly lenient control strategy may result in some weather types with insufficient data being excluded from the control scope, thus affecting the reliable operation of the subsystems. Therefore, adopting a strict control strategy and lowering the threshold for determining the controllable weather type (requiring only the fulfillment of a certain percentage of dates) can expand the control coverage, allowing more weather types to be included in the control, thereby ensuring the operational stability of the subsystems of that demand type.

[0050] The significance of this situation lies in realizing the "macro-priority" principle of strategy formulation. When the overall data quality of the region is poor, a more lenient strategy with broader coverage is prioritized to avoid subsystem operational risks due to insufficient control, reflecting a safety net approach to risk prevention.

[0051] Additionally, it should be noted that if the number of subsystems of a certain demand type in the target area is not greater than the preset threshold for the number of a certain demand type, the weight value of the operating demand in different weather types is determined based on the operating demand type of the subsystems in the target area. It is then determined whether the sum of the operating demand weight values ​​in different weather types is greater than the preset weight threshold. If yes, the process proceeds to step S22. If no, the update strategy for the load control weather type is a lenient control strategy, that is, as long as the proportion of the number of days under load control in the weather type is above the first proportion threshold and the load control duration is above the target duration, the weather type is used as the load control weather type.

[0052] The operational demand weight value refers to the weight assigned to each weather type based on the operational demand type of the subsystem. This weight quantifies the severity of data insufficiency under that weather type. For example, a subsystem with a first-type demand has the highest weight under a given weather type, followed by the second-type, and the third-type has the lowest. The preset weight threshold is a pre-defined critical value used to determine whether the accumulated data insufficiency for a particular weather type within the region has reached a level requiring further refined analysis. The load control duration refers to the length of time a single load control measure needs to last under a given weather type. The target duration is a pre-defined critical value used to determine whether the control duration has reached the execution condition.

[0053] When the number of subsystems for a particular demand type is small, the overall data quality of the region is good, but there may be a relatively concentrated shortage of data for certain weather types. By calculating the sum of the operational demand weights for each weather type, the degree of clustering of data shortages along the weather dimension can be quantified. If the sum of the weights is high, it indicates that some weather types are carrying a large demand for insufficient data, requiring further analysis (proceed to step S22); if the sum of the weights is low, it indicates that the distribution of insufficient data is relatively dispersed. In this case, a strict control strategy with added duration conditions can be adopted to ensure necessary control coverage while avoiding excessive control that would waste resources.

[0054] The significance of this situation lies in achieving a "micro-level refinement" transition in strategy formulation. When the macro-level situation is not urgent, the introduction of a weighted evaluation mechanism shifts the decision-making focus from "the number of subsystems" to "the concentration of demand for weather types," providing a scientific basis for subsequent refined judgments.

[0055] S22 determines the subsystems that belong to the demand-matching weather type in the weather type based on the degree of overlap between the demand-matching weather types of different subsystems, and identifies the subsystems that belong to the demand-matching weather type in the weather type as associated subsystems; In the above steps, the number of associated subsystems in the weather type is obtained, and it is determined whether the number of associated subsystems in the weather type is a preset subsystem number threshold. If so, the update strategy of the weather type as a load control weather type is a strict control strategy, that is, as long as the proportion of the number of days under load control in the weather type is above the first proportion threshold, the weather type is used as a load control weather type. If not, proceed to step S23.

[0056] The demand-matching weather type refers to the weather type in the subsystem whose runtime is less than a preset runtime threshold, reflecting the data bottleneck of the subsystem under that weather type. The overlap degree refers to the degree of overlap between demand-matching weather types of different subsystems, used to measure whether a certain weather type is a data bottleneck for multiple subsystems simultaneously. The associated subsystem refers to a subsystem that uses a specific weather type as its demand-matching weather type. The preset subsystem quantity threshold is a pre-set critical value used to determine whether the number of associated subsystems under a certain weather type has reached a level requiring a strict control strategy. The strict control strategy refers to a relatively strict method for determining the control type, characterized by relatively strict triggering conditions, aiming to accurately control weather types with concentrated data bottlenecks.

[0057] When step S22 is reached, it indicates that there is a high concentration of demand for certain weather types within the region. By analyzing the degree of overlap between the demand and weather types of different subsystems, it is possible to identify which weather types are common data bottlenecks for multiple subsystems. If the number of associated subsystems under a certain weather type reaches a preset threshold, it indicates that this weather type has a wide-ranging impact on the overall operation of the region, and strict control strategies should be adopted to ensure that the operation of all related subsystems can be effectively guaranteed when this weather type occurs.

[0058] The significance of this step lies in realizing the identification of "common needs of multiple subsystems" from "single subsystem requirements". By introducing overlap analysis and the concept of related subsystems, it is possible to accurately locate common data gaps in weather types within the region and implement strict control strategies to focus on covering them, thereby improving the pertinence and effectiveness of the control strategies.

[0059] S23 uses the subsystems of a demand type in the target area and the operational demand types of the associated subsystems in the weather type to determine the update strategy for the load control weather type.

[0060] Furthermore, it is determined whether the proportion of weather types with strict control strategies in the weather types is greater than a preset weather type proportion threshold. If so, the update strategy for the weather type as a load control weather type is a lenient control strategy. If not, the update strategy for the weather type as a load control weather type is determined based on the number of associated subsystems of a demand type in the weather type.

[0061] Furthermore, based on the number of associated subsystems of one type of demand within the weather type, the update strategy for the weather type as the load management weather type is determined, specifically including: If the number of associated subsystems for one type of demand within the weather type exceeds a preset value for the number of associated systems, then the weather type is determined to be a strict control strategy for load control weather type updates; otherwise, it is a lenient control strategy.

[0062] The weather type subject to the strict control strategy refers to the weather type that has been determined to be subject to the strict control strategy in the preceding steps. The preset weather type proportion threshold is a pre-set critical value used to determine whether the proportion of weather types subject to the strict control strategy among all weather types is too high. The related subsystem of a certain demand type refers to a subsystem within a certain weather type's related subsystem whose own operational demand type is a certain demand type.

[0063] When the number of associated subsystems in step S22 fails to reach the threshold and the process proceeds to step S23, it indicates that the data bottleneck's distribution across weather types is neither widespread nor highly concentrated. At this point, further judgment is needed by considering the distribution of a specific type of demand subsystem. First, if the proportion of weather types subject to strict control strategies is too high, it suggests that previous judgments may have led to overly broad control scope. In this case, the remaining weather types will be subject to lenient control strategies to avoid over-control. If the proportion is low, further analysis is needed on the number of specific demand subsystems within the associated subsystems to determine the final strategy.

[0064] The significance of this step lies in the fact that by assessing the proportion of weather types under strict control strategies, it is possible to avoid the problem of excessively comprehensive control, thus ensuring the rationality and balance of the final strategy. At the same time, prioritizing a particular type of demand subsystem demonstrates a focus on ensuring the weakest link in the system.

[0065] Furthermore, based on the number of associated subsystems of one type of demand in the weather type, the update strategy for the weather type as a load control weather type is determined. Specifically, if the number of associated subsystems of one type of demand in the weather type is greater than a preset value for the number of associated systems, then the update strategy for the weather type as a load control weather type is determined to be a strict control strategy; otherwise, it is a lenient control strategy.

[0066] The preset value for the number of associated systems is a pre-set threshold value used to determine whether the number of associated subsystems for a certain type of demand under a certain weather type has reached a level that requires strict control strategies.

[0067] In the final stage of strategy determination, the focus is on the related subsystems of a specific demand type. This is because these subsystems represent the group with the scarcest data and the greatest need for protection. If, under a certain weather type, the number of related subsystems of a specific demand type reaches a preset value, it indicates that this weather type has a significant impact on the operation of the weakest link subsystem, and a strict control strategy should be adopted to ensure its protection. If the number does not reach this value, it indicates that the impact is limited, and a more lenient control strategy is sufficient.

[0068] The significance of this step lies in achieving a "priority safety net" in strategy formulation. By focusing on a specific type of demand subsystem, it ensures that control measures are implemented most strictly in the most critical subsystem groups and the most vulnerable weather types, thereby maximizing the overall operational reliability of the system.

[0069] This embodiment provides a method for determining the update strategy of weather type for load management of distributed photovoltaic systems. It is applied to a photovoltaic management system that has been connected to a cloud platform. The system contains a target area with 30 distributed photovoltaic subsystems. The weather type is divided into 12 refined types based on three dimensions: temperature, wind speed, and irradiance, through cluster analysis.

[0070] First, the cloud platform executes step S21: The cloud platform obtains the operational requirement types of each subsystem within the target area. Statistics show that there are 3 subsystems with one type of requirement within the target area. The preset threshold for the number of subsystems with one type of requirement is 5. The cloud platform determines that 3 is not greater than 5, therefore proceeding to the next level of judgment.

[0071] The cloud platform calculates operational demand weights for 12 weather types based on the operational demand types of 30 subsystems in the target area. For example, for weather type 5 (low temperature, low light, light wind), the system counts 4 subsystems that match this weather type as a demand type. Among them, 2 are subsystems of type 1 demand (weight 2), 2 are subsystems of type 2 demand (weight 1), and 0 are subsystems of type 3 demand (weight 0). Therefore, the sum of the operational demand weights for this weather type is 2×2 + 2×1 = 6. After traversing all 12 weather types, the system calculates the total weight of all weather types to be 45. The preset weight threshold is 40. Since 45 is greater than 40, the system proceeds to step S22.

[0072] The cloud platform executes step S22: The system analyzes the degree of overlap between the demand and weather types of different subsystems. For weather type 5 (low temperature, low light, light wind), the number of associated subsystems that match it as a demand-matching weather type is counted as 4. The preset threshold for the number of subsystems is 5. Since 4 is less than 5, the threshold is not met, therefore this weather type is not directly subject to a strict control strategy, and the process proceeds to step S23.

[0073] The cloud platform executes step S23: The system first counts the number of weather types marked as subject to strict control strategies during the current judgment process. The count shows that out of 12 weather types, 3 have been marked as subject to strict control strategies because the number of associated subsystems has reached a threshold. The preset threshold for the proportion of weather types is 30%, or 3.6 types. Since 3 is less than 3.6, the threshold has not been exceeded, and the next judgment step continues.

[0074] The system further analyzes the number of subsystems with the same demand type within the associated subsystems under weather type 5 (low temperature, low light, light wind). Statistics show that two of the four associated subsystems under weather type 5 have the same operational demand type. The preset value for the number of associated systems is 1. Since 2 is above the preset value, the system determines that this weather type will be subject to a strict control strategy for load management weather type updates.

[0075] Ultimately, the cloud platform outputs the following judgment: For weather type 5 (low temperature, low light, light wind), a strict control strategy will be adopted. That is, when this weather type occurs, as long as the proportion of days subject to load control in this weather type reaches the first proportion threshold (e.g., 30%), the cloud platform will treat this weather type as a load control weather type.

[0076] S2 determines the management and control strategy for the distributed photovoltaic system based on the updated data of the load control weather type and the on-grid electricity of the distributed photovoltaic system in the load control weather type. Specifically, such as Figure 4 As shown, the method for determining the management and control strategy of the distributed photovoltaic system is as follows: Based on the identified load control weather types, a dynamic control strategy for distributed photovoltaic (PV) systems is formulated. This strategy fully considers the impact of load control on overall grid connection reliability, as well as the consistency of frequency regulation curves between different subsystems and other subsystems. Priority is given to controlling subsystems with lower consistency, while ensuring frequency regulation reliability during load control. This determines which subsystems should be connected to the grid for frequency regulation under load control weather conditions. The core logic involves a hierarchical, multi-condition judgment mechanism to comprehensively evaluate indicators such as the number of load control weather types, grid connection matching coefficients, and the proportion of reliable grid connection weather types. This determines the appropriate load demand satisfaction standard as a constraint for subsystem connection. When there are many load control weather types, indicating a wider control scope, a preset multiple standard is used for subsystem selection. When the quantity is small, the matching degree of grid-connected electricity is further analyzed. If the matching coefficient is high, a higher multiple standard is adopted; if the matching coefficient is low, the proportion of reliable grid-connected weather types is analyzed. If the proportion is high, a second preset multiple standard is adopted; if the proportion is low, the grid-connected demand value is calculated by combining the quantity and the matching coefficient, and a preset multiple or a third preset multiple standard is adopted according to the demand value. This mechanism ensures that the control and processing strategy can adaptively match the grid load demand, reduce the impact of load control on distributed photovoltaic systems, and achieve the optimal configuration of the number of connected subsystems.

[0077] S31 uses the updated data of the load control weather type to determine the number of the load control weather types; The updated data for load control weather types refers to the set of weather types that currently require load control measures, determined after the load control weather type update strategy is implemented. The number of load control weather types refers to the total number of weather types included in this set.

[0078] This step is the starting point for formulating the entire control and management strategy. The number of weather types for load control reflects the breadth of the current control scope and is the primary basis for determining which handling strategy to adopt. The greater the number, the more operating scenarios require control, the greater the scheduling pressure on the system, and the more efficient the screening mechanism needs to be adopted.

[0079] The significance of this step lies in achieving a transition from "which weather conditions need to be controlled" to "how to implement control measures." By statistically analyzing the number of weather types requiring control, the scale of the control task can be quickly grasped, providing the first layer of decision-making basis for subsequently selecting different multiples of load demand satisfaction standards.

[0080] It should be noted that in the above steps, if the number of load-controlled weather types is greater than the preset threshold for the number of load-controlled weather types, then among the load-controlled weather types, the load demand of the load-controlled weather types that can meet a preset multiple must be included, with the constraint that the number of subsystems connected is minimized. The subsystems connected to the network are sorted by the average deviation rate between the frequency regulation response curve of the subsystem in the load-controlled weather type and the frequency regulation response curve of other subsystems, from smallest to largest. The remaining subsystems are no longer connected to the network, thereby determining whether the subsystem of the first demand type can reliably regulate the frequency under different load conditions.

[0081] In another embodiment, if the number of load-controlled weather types is not greater than a preset threshold for the number of controlled weather types, then proceed to step S32.

[0082] The preset threshold for the number of controlled weather types is a pre-set critical value used to determine whether the controlled area has reached a high level. The preset multiple refers to a pre-set load demand satisfaction ratio coefficient used to determine the load demand level that must be met when selecting subsystems for access. The first demand type subsystem refers to a subsystem classified as a specific demand type in the operational demand type determination; this type of subsystem has the most urgent need for data acquisition and model updates, and must be guaranteed in the control process. The frequency regulation response curve refers to the characteristic curve of the subsystem's output power changing over time when participating in grid frequency regulation, reflecting the subsystem's frequency regulation capability. The average deviation rate is an average measure of the difference between the frequency regulation response curve of a certain subsystem and the frequency regulation response curves of all other subsystems; the smaller the average deviation rate, the higher the frequency regulation reliability when the subsystem is retained.

[0083] When there are many weather types subject to load control, it indicates a wide control scope. To reduce the impact of load control on overall grid reliability, the following constraint is applied: "meeting a preset multiple of load demand" and "minimizing the number of connected subsystems." Subsystems are selected in ascending order of the average deviation rate of the frequency regulation response curve. This ensures that the connected subsystems have good frequency regulation reliability. Simultaneously, including the subsystem of the highest demand type is a constraint to ensure that the subsystem with the least data can reliably monitor the aforementioned weather types.

[0084] The significance of this is that it enables an "efficient screening" mechanism when the control scope is wide. By prioritizing the access of subsystems with the most representative frequency regulation characteristics, the number of accesses can be minimized while meeting load requirements, thereby improving the efficiency and reliability of control processing.

[0085] S32 Based on the grid-connected electricity of the distributed photovoltaic system in the load control weather type, determine the average value of the ratio of the grid-connected electricity of the distributed photovoltaic system in the load control weather type to the grid-connected electricity in other load control weather types, and use it as the grid-connected electricity matching coefficient of the load control weather type; The grid-connected electricity refers to the total electricity transmitted to the grid by the distributed photovoltaic subsystem under a specific weather type. Other load-controlled weather types refer to all load-controlled weather types other than the one currently being analyzed. The grid-connected electricity matching coefficient is a quantitative indicator obtained by averaging the ratios of the grid-connected electricity under a certain load-controlled weather type to the grid-connected electricity under all other load-controlled weather types; it reflects the relative importance of that weather type in the overall grid-connected electricity.

[0086] This step forms the basis for subsequent refined judgments. The amount of electricity generated online may vary significantly under different weather conditions requiring load control. The online electricity generation matching coefficient can quantify the relative importance of each weather type in terms of electricity contribution, providing a basis for determining whether stricter access standards need to be adopted.

[0087] The significance of this step lies in incorporating electricity contribution factors into the formulation of control strategies, so that decisions not only consider the number of weather types, but also their actual electricity impact, thereby improving the scientific and economic efficiency of strategy formulation.

[0088] The above steps include the following: S321 determines whether the power matching coefficient of the load control weather type is greater than the preset matching coefficient threshold. If so, in the load control weather type, the load demand of the load control weather type that can meet the preset multiple must be included as a constraint. The goal is to minimize the number of subsystems connected. The subsystems are sorted from smallest to largest by the average deviation rate of the frequency regulation response curve of the subsystem in the load control weather type and the frequency regulation response curve of other subsystems. The remaining subsystems are no longer connected to the grid. This determines whether the subsystem of the first demand type can reliably regulate the frequency under different load conditions. If not, proceed to step S322. The preset matching coefficient threshold is a pre-set critical value used to determine whether the power consumption matching coefficient for a certain load control weather type has reached a high level. The second preset multiple is a pre-set load demand satisfaction ratio coefficient that is greater than the preset multiple, used to adopt stricter access standards when the matching coefficient is high.

[0089] When the on-grid electricity matching coefficient for a certain weather type under load control is high, it indicates that this weather type plays an important role in the on-grid electricity volume, and the effectiveness of its control and processing has a significant impact on the entire system. Therefore, using a larger multiple (preset multiple) as the standard for meeting load demand can ensure that the access subsystem has stronger frequency regulation capabilities under important weather types, thereby guaranteeing the stable operation of the power grid.

[0090] The significance of this step lies in establishing a "priority protection" mechanism for critical weather types. By adopting a higher load demand fulfillment multiple, it ensures that the management and handling strategies are more robust and reliable during weather types that contribute significantly to electricity generation.

[0091] S322 defines a load control weather type that can meet the load demand of the load control weather type by a preset multiple as a reliable internet access weather type. It then determines whether the proportion of the reliable internet access weather type in the load control weather type is greater than a preset proportion threshold. If so, it uses the constraint that the load control weather type can meet the load demand of the load control weather type by a second preset multiple and must include a subsystem of the first demand type. With the goal of minimizing the number of connected subsystems, it sorts the subsystems for internet access by the average deviation rate of the frequency modulation response curve of the subsystem in the load control weather type from smallest to largest. The remaining subsystems are no longer connected to the grid. This determines whether the subsystem of the first demand type can reliably perform frequency modulation under different load conditions. If not, it proceeds to step S33.

[0092] The "reliable internet access weather type" refers to those weather types among all load control weather types that can meet a preset multiple of load demand, reflecting that these weather types have good power supply guarantee capabilities. The preset ratio threshold is a pre-set critical value used to determine whether the proportion of reliable internet access weather types among all load control weather types has reached a high level.

[0093] When the power supply matching coefficient is low, further analysis of the proportion of reliable internet access weather types is needed. If the proportion is high, it indicates that most load control weather types can meet the load demand of the preset multiple, and the overall power supply capacity of the system is strong. In this case, a lower multiple (the second preset multiple) can be used as the access standard to further improve the control effect. If the proportion is low, a more refined judgment is needed.

[0094] The significance of this step lies in introducing the assessment dimension of "power supply capacity ratio," which enables decisions to reflect the overall power supply reliability level of the system and avoids affecting the rationality of the overall strategy due to power supply shortages caused by local weather conditions.

[0095] S33 uses the number of load control weather types and the matching coefficient of the on-grid electricity for the load control weather types to determine the control and processing strategy for the distributed photovoltaic system.

[0096] Furthermore, based on the power matching coefficient for the load-controlled weather type and the proportion of reliable internet access weather types within the load-controlled weather type, the internet access demand value for the load-controlled weather type is determined. It is then determined whether the internet access demand value for the load-controlled weather type exceeds a preset demand threshold. If so, subsystems within the load-controlled weather type are selected based on constraints such as meeting a preset multiple of the load demand for that load-controlled weather type, and must include subsystems of the first demand type. The goal is to minimize the number of connected subsystems. Subsystems are sorted from smallest to largest by the average deviation rate between their frequency response curves and those of other subsystems within the load-controlled weather type. The remaining subsystems will not be connected to the grid. This determines whether the subsystems of the first demand type can reliably regulate frequency under different load conditions. If not, then in the load control weather type, the subsystems that can meet the load demand of the load control weather type by a third preset multiple and must include the subsystems of the first demand type are selected as the constraint. The goal is to minimize the number of connected subsystems. The subsystems are sorted from smallest to largest by the average deviation rate of the frequency regulation response curve of the subsystem in the load control weather type and the frequency regulation response curve of other subsystems. The remaining subsystems will not be connected to the grid. This determines whether the subsystems of the first demand type can reliably regulate frequency under different load conditions.

[0097] It should be noted that the third preset multiple is greater than the second preset multiple, and the third preset multiple is less than the preset multiple.

[0098] The internet access demand value refers to a comprehensive quantitative indicator calculated by combining the internet access power matching coefficient for weather types under comprehensive load control and the proportion of reliable internet access weather types. It reflects the overall internet access demand level of the system. The preset demand threshold is a pre-set critical value used to determine whether the internet access demand value has reached a high level. The third preset multiple is a pre-set load demand fulfillment ratio coefficient that is greater than the second preset multiple but less than the preset multiple, used to adopt a relatively lenient access standard when the internet access demand value is low.

[0099] If the strategy remains undetermined after the aforementioned steps, a comprehensive judgment needs to be made based on both quantity and matching coefficient dimensions. By constructing a comprehensive indicator called "Internet Connection Demand Value," the actual demand for internet access power can be more fully reflected. A high Internet Connection Demand Value indicates stricter overall control, and a greater demand for internet access power under load control weather conditions. A preset multiplier standard is used to ensure basic load requirements are met. Conversely, a low Internet Connection Demand Value indicates a more lenient demand for power. A third preset multiplier standard is adopted, with the multiplier falling between the second and third preset multipliers. This allows for an appropriate increase in access requirements while ensuring access for subsystems with the first demand type.

[0100] A "comprehensive trade-off" mechanism for strategy formulation has been implemented. By constructing a comprehensive indicator of internet access demand value, information from both quantity and power consumption dimensions is integrated to ensure that the final management and control strategy can fully reflect the actual operational needs of the system, thus enabling scientific decision-making under complex conditions.

[0101] This embodiment provides a method for determining the management and control strategy for distributed photovoltaic systems. It is applied to a photovoltaic management system that has been connected to a cloud platform. The system contains a target area with 30 distributed photovoltaic subsystems. The weather types are divided into 12 refined types based on three dimensions: temperature, wind speed, and solar irradiance, through cluster analysis. The update strategy for load control weather types has been determined using the aforementioned method.

[0102] First, the cloud platform executes step S31: The cloud platform obtains updated data on load control weather types and determines that there are currently four weather types requiring load control measures: Weather Type 1 (low temperature, weak light, strong wind), Weather Type 2 (moderate temperature, moderate light, moderate wind), Weather Type 3 (high temperature, strong light, light wind), and Weather Type 4 (low temperature, weak light, light wind). The preset threshold for the number of controllable weather types is 5. The cloud platform determines that 4 is not greater than 5, therefore proceeding to step S32.

[0103] The cloud platform executes step S32: Based on the online power consumption data of 30 subsystems across four weather types under load control, the cloud platform calculates the online power consumption matching coefficient for each weather type. Taking weather type 1 as an example, its online power consumption is 1500 kWh. The online power consumption for the other three weather types are 2000 kWh, 2500 kWh, and 1000 kWh, respectively. The average ratio of the online power consumption of this weather type to the other weather types is calculated as (1500 / 2000 + 1500 / 2500 + 1500 / 1000) / 3 = 0.6167, meaning the online power consumption matching coefficient is 0.6167. After iterating through the four weather types, the cloud platform obtains matching coefficients of 0.62, 0.58, 0.71, and 0.45 for each weather type.

[0104] The cloud platform executes step S321: The cloud platform reads the preset matching coefficient threshold as 0.6. After comparison, weather type 1 (0.62) is greater than 0.6, weather type 2 (0.58) is not greater than 0.6, weather type 3 (0.71) is greater than 0.6, and weather type 4 (0.45) is not greater than 0.6. For weather type 1 and weather type 3, whose matching coefficients are greater than 0.6, the cloud platform directly determines to adopt the preset multiple standard. In this embodiment, the preset multiple is 1.2 times. Therefore, for weather type 1 and weather type 3, the system selects the subsystems to be accessed according to the constraint of meeting 1.2 times the load demand, aiming to minimize the number of access subsystems, and in ascending order of the average deviation rate of the subsystem frequency modulation response curve, and must include the subsystem of the first demand type.

[0105] For weather types 2 and 4 with a matching coefficient of no more than 0.6, the cloud platform proceeds to step S322.

[0106] The cloud platform executes step S322: First, the cloud platform determines the reliable internet access weather types, i.e., the load control weather types that can meet the preset multiple (1.2 times) of the load demand. System calculations show that among the four load control weather types, weather type 1, weather type 2, and weather type 3 can all meet the 1.2 times load demand, while weather type 4 cannot. Therefore, the reliable internet access weather types are weather type 1, weather type 2, and weather type 3, a total of three, with a reliable internet access weather type ratio of 75%. The preset ratio threshold is 80%. Since 75% is not greater than 80%, the cloud platform proceeds to step S33.

[0107] The cloud platform executes step S33: Based on the power consumption matching coefficient for the load-controlled weather type and the proportion of reliable internet access weather types, the cloud platform calculates the internet access demand value. In this embodiment, the calculation formula for the internet access demand value is: Internet access demand value = Power consumption matching coefficient for the load-controlled weather type / Proportion of reliable internet access weather type in the load-controlled weather type. After calculation, the power consumption matching coefficient is 0.45, and the internet access demand value = 0.45 × 4 / 3 = 0.6. The preset demand threshold is 0.7. Since 0.6 is not greater than 0.7, the cloud platform determines that a third preset multiple standard is used for weather type 4. In this embodiment, the third preset multiple is 1.1 times.

[0108] The significance and value of this invention lies in providing an adaptive control and management strategy generation mechanism based on the characteristics of load control weather types. First, through a hierarchical decision-making architecture, it organically combines multiple factors such as the number of load control weather types, the matching coefficient of grid-connected electricity, and the proportion of reliable grid-connected weather types, achieving a complete logical closed loop from "control scope identification" to "access standard formulation." Second, by introducing three different levels of load demand satisfaction standards—preset multiples, second preset multiples, and third preset multiples—a refined access screening mechanism is established, enabling the control and management strategy to dynamically adjust the number of access subsystems and frequency regulation capability requirements according to actual needs. Third, the constraint of "must include subsystems of the first demand type" ensures that subsystems with the scarcest data and the most urgent needs are prioritized in control and management, reflecting a bottom-line approach to risk prevention. Finally, by using a sorting mechanism that aims for "the fewest access subsystems" combined with the average deviation rate of the frequency regulation response curve, efficient screening of access subsystems is achieved, reducing the difficulty and reliability of frequency regulation processing. This method provides scientific, efficient, and adaptive decision support for the management and control of distributed photovoltaic systems in complex operating environments, laying the foundation for the reliable identification of the operating status of a certain type of demand.

[0109] S3 performs management and control processing of the distributed photovoltaic system based on the management and control processing strategy, and determines the management and control optimization target of the load control weather type using the management and control processing data of the subsystem and the load control subsystem in the load control weather type.

[0110] Furthermore, the control and processing data of the subsystem is determined according to the load control weather type that the subsystem needs to perform load control processing on.

[0111] Specifically, the method for determining the control optimization target for the load control weather type is as follows: Based on the control and management data of the subsystems and the load control subsystems in the load control weather types, the control optimization targets for the load control weather types are determined. Based on the results of the executed load control processing, the control impact of different subsystems under each load control weather type is evaluated, and it is determined which load control weather types need to be listed as control optimization targets, that is, stop load control processing on them. The system can dynamically identify the weather types that need to be stopped when the control impact is significant, thereby reducing the impact of load control on the overall energy supply reliability.

[0112] S41 uses the control and processing data of the subsystem to determine the load control weather type that the subsystem needs to perform load control processing, and uses the load control weather type that the subsystem needs to perform load control processing as the matching control weather type; The subsystem's control and processing data refers to historical data recording which subsystems actually performed load control processing under which load control weather types. The load control weather types requiring load control processing refer to the set of weather types under which the subsystem is required to participate in load control according to the aforementioned control and processing strategy. The matched control weather types refer to the weather types that a single subsystem actually participates in load control for, reflecting the control association between that subsystem and specific weather types.

[0113] This step forms the basis for determining the overall control optimization objectives. By extracting the actual load control weather types involved in each subsystem from the control processing data, the control impact on the subsystem can be established, laying the foundation for further reducing the load control impact on the subsystem.

[0114] S42 determines the load control subsystem in the load control weather type based on the load control subsystem in the load control weather type; The load control subsystem in the aforementioned load control weather type refers to the set of all subsystems that actually perform load control processing under a specific load control weather type. The load control subsystem refers to the subsystem identified as participating in load control processing under a specific load control weather type.

[0115] This step, from a weather type perspective, compiles a list of all subsystems actually involved in control under that weather type, forming a control participation list for each load control weather type. This list forms the basis for subsequent assessments of the impact of control measures for that weather type on the overall subsystems.

[0116] The significance of this step lies in realizing the transformation from a "subsystem perspective" to a "weather type perspective," which enables a quantitative assessment of the control impact of each load control weather type on the subsystem.

[0117] S43 determines the control optimization objective of the load control weather type based on the load control subsystem in the load control weather type and the matching control weather types of different load control subsystems.

[0118] Furthermore, if the number of load control subsystems in the load control weather type is greater than the preset threshold for the number of control subsystems, then if the proportion of dates with frequency regulation response not meeting the requirements after load control processing for the load control weather type is greater than the threshold for the proportion of dates with frequency regulation deviation, then the load control weather type is determined to be a control optimization target, i.e., no more load control processing will be performed.

[0119] The preset threshold for the number of control subsystems is a pre-defined critical value used to determine whether the number of subsystems participating in control under a certain load control weather type has reached a high level. The frequency regulation period where the frequency regulation response does not meet requirements refers to the period during load control processing where the deviation between the actual and expected frequency regulation response of a subsystem exceeds the allowable range. This deviation directly reflects that a type of demand subsystem cannot effectively respond to frequency regulation commands under that weather type. The frequency regulation deviation date percentage threshold is a pre-defined critical value used to determine whether the percentage of dates where the frequency regulation response does not meet requirements has reached a level requiring the cessation of control, i.e., whether the frequency regulation capability deficiency of a type of demand subsystem has been fully confirmed.

[0120] When a large number of subsystems participate in frequency regulation under a specific load control weather type, it indicates a wide control coverage and the participation of numerous subsystems (including one type of demand subsystem) in frequency regulation response. If, under these circumstances, a high proportion (greater than the threshold) of frequency regulation deviation dates still exist, it suggests that the demand-type subsystems are generally unable to effectively respond to frequency regulation commands under this weather type, and their frequency regulation capability deficiencies have been fully verified. Continuing control in this situation will not improve frequency regulation performance; instead, it will consume resources due to ineffective control. Therefore, this should be the target for control optimization, and subsequent control should be discontinued.

[0121] The significance of this situation lies in establishing a mechanism for "timely exiting control measures when widespread frequency modulation failure occurs." When control measures have covered a sufficient number of subsystems and it has been confirmed that a certain type of demand subsystem generally suffers from frequency modulation capability deficiencies, ineffective control measures are decisively stopped, thereby reducing the impact on the power consumption of the subsystems.

[0122] Additionally, it is understood that if the number of load control subsystems in the load control weather type is not greater than a preset threshold for the number of control subsystems, the following content is also included: Case 1: Based on the matching control weather type of the load control subsystem in the load control weather type, determine the weight value of the load control subsystem. If the sum of the weight values ​​of the load control subsystem in the load control weather type is greater than the preset weight threshold, then if the proportion of the number of dates with frequency regulation response not meeting the requirements after load control processing of the load control weather type is greater than the frequency regulation deviation date proportion threshold, then it is determined that the load control weather type belongs to the control optimization target, that is, no longer perform load control processing.

[0123] The weight value refers to the quantitative weight assigned to the load control subsystem based on the number of matched controlled weather types or other attributes, reflecting the degree of impact of the control measures received by the subsystem. A higher number of matched controlled weather types indicates that the frequency regulation needs of the subsystem are controlled under various weather conditions, and the greater the control impact. The preset weight threshold is a pre-defined critical value used to determine whether the sum of the weight values ​​of the load control subsystem has reached a high level, i.e., whether it has severely impacted the network load of a large number of subsystems.

[0124] When the number of subsystems involved in the control is small, if the sum of the weight values ​​of these subsystems is high, the frequency regulation demand under various weather types will be controlled, and the control impact will be higher. In this case, if the proportion of frequency regulation deviation dates exceeds the threshold, it indicates that a representative type of demand subsystem cannot effectively respond to frequency regulation under that weather type, and control should be stopped.

[0125] Scenario 2: If the sum of the weight values ​​of the load control subsystems in the load control weather type is not greater than the preset weight threshold, determine whether the number of load control subsystems in the load control weather type with weight values ​​greater than the preset weight value is less than the preset control subsystem number threshold. If not, if after load control processing, the load control weather type has frequency regulation periods where the frequency regulation response does not meet the requirements on different dates, then the load control weather type is determined to be a control optimization target, i.e., no more load control processing is performed. If yes, then the load control weather type is determined not to be a control optimization target.

[0126] The preset weight value is a pre-defined weight threshold used to filter out subsystems with higher weight values, i.e., subsystems that are more significantly affected by control measures. The statement that frequency modulation response failures exist on different dates indicates that the frequency modulation deviation occurs on multiple different control dates, rather than being concentrated on a few dates. This reflects that the frequency modulation capability deficiency of a certain type of demand subsystem is persistent rather than sporadic.

[0127] When the weight values ​​of the subsystems involved in the control are not high, it is necessary to further analyze the number of subsystems with higher weight values ​​and greater control impact. If the number of subsystems with greater control impact is small (less than the threshold), the control impact is small, and all load control weather types are not part of the control optimization target. If the number of subsystems with greater control impact is large, and frequency regulation deviation is prevalent in different dates for the aforementioned load control weather type, the persistence of its frequency regulation capability defect has sufficiently confirmed the existence of the problem, and control should be stopped.

[0128] It should be noted that once the duration of control for all load control weather types has met the requirements, the frequency regulation matching characteristics of different subsystems have been fully understood, and thus they can be used as the control optimization target.

[0129] The requirement that the control duration meets the requirements means that the cumulative duration of load control processing under each load control weather type has reached the preset target duration, ensuring that the system has a full understanding of the frequency regulation capability defects of a certain type of demand subsystem, so that targeted optimization measures can be taken based on the diagnostic results, and ultimately stop ineffective control.

[0130] This statement clarifies the final criteria for determining the control optimization target: control can only be safely stopped when all load control weather types have undergone sufficient control duration and the system has fully diagnosed the frequency regulation capability of a certain type of demand subsystem. At this point, stopping control is a scientific decision based on the fact that the capability defects have been fully confirmed.

[0131] This embodiment provides a method for determining the control optimization target of load management weather type for distributed photovoltaic systems. It is applied to a photovoltaic management system that has been connected to a cloud platform. The system contains a target area with 30 distributed photovoltaic subsystems. The weather type is divided into 12 refined types based on three dimensions: temperature, wind speed, and irradiance, through cluster analysis. The load management weather type has been determined by the aforementioned method and the corresponding control processing has been performed.

[0132] First, the cloud platform executes step S41: The cloud platform obtains the control and processing data of each subsystem and determines the load control weather type for each subsystem. Taking subsystem 1 as an example, its control and processing data shows that it performed load control processing under weather type A, weather type B, and weather type C. These weather types are the matching control weather types for subsystem 1. The cloud platform then iterates through all 30 subsystems and establishes a list of matching control weather types for each subsystem.

[0133] Secondly, the cloud platform executes step S42: From the perspective of weather type, the cloud platform summarizes the subsystems that actually performed load control processing under each load control weather type. Taking weather type A (low temperature, weak light, strong wind) as an example, according to statistics, there are a total of 8 subsystems participating in load control under this weather type. These subsystems are the load control subsystems for weather type A.

[0134] Then, the cloud platform executes step S43: the cloud platform determines the control and optimization targets for each load control weather type one by one.

[0135] For weather type A, the cloud platform obtains 8 load control subsystems. The preset threshold for the number of control subsystems is 5. Since 8 is greater than 5, the cloud platform proceeds to condition 1. The cloud platform retrieves the frequency modulation response data for weather type A after control processing and counts the number of dates where the frequency modulation response does not meet the requirements. The statistics show that out of 10 controlled dates, 4 dates have frequency modulation deviations, accounting for 40%. The threshold for the percentage of dates with frequency modulation deviations is 30%. Since 40% is greater than 30%, it indicates that a certain type of demand subsystem generally cannot effectively respond to frequency modulation commands under this weather type, and its frequency modulation capability deficiency has been fully verified. The cloud platform determines that weather type A belongs to the control optimization target, i.e., no further load control processing is performed.

[0136] For weather type B (medium temperature, medium sunshine, medium wind), the cloud platform identifies 3 load control subsystems. The preset threshold for the number of control subsystems is 5. Since 3 is not greater than 5, the cloud platform enters the second branch. The cloud platform first determines the weight value of each load control subsystem based on the number of matching control weather types. This weight value reflects the control impact on the subsystem. Subsystem 2 has 5 matching control weather types, assigned a weight of 5; subsystem 3 has 4 matching control weather types, assigned a weight of 4; and subsystem 4 has 3 matching control weather types, assigned a weight of 3. The sum of the weight values ​​of the load control subsystems under weather type B is 12. The preset weight threshold is 10. Since 12 is greater than 10, the cloud platform calculates that the frequency regulation deviation date percentage for weather type B is 35%. The frequency regulation deviation date percentage threshold is 30%. Since 35% is greater than 30%, it indicates that a representative demand type subsystem cannot effectively respond to frequency regulation under this weather type, and its frequency regulation capability deficiency has been fully confirmed by the representative subsystem. The cloud platform determines that weather type B belongs to the control optimization target, meaning that load control processing will no longer be performed.

[0137] For weather type C (high temperature, strong light, light wind), the cloud platform identifies two load management subsystems. The preset threshold for the number of management subsystems is 5. Since 2 is not greater than 5, the cloud platform enters the second branch. The cloud platform calculates the sum of the weight values ​​of the load management subsystems under weather type C to be 6. The preset weight threshold is 10. Since 6 is not greater than 10, the cloud platform further counts the number of load management subsystems with weight values ​​greater than the preset weight value (set to 4). The count shows that one of the two load management subsystems has a weight value greater than 4. The preset threshold for the number of management subsystems is 5. Since 1 is less than 5, weather type C is determined not to be a management optimization target.

[0138] Example 2 Secondly, the present invention provides a computer system comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-described method for load management of a distributed photovoltaic system based on big data when running the computer program.

[0139] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0140] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0141] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. A big data-based distributed photovoltaic system load management method, characterized in that, Specifically, it includes: Using the newly added data of the distributed photovoltaic system, the operating data of different subsystems in the distributed photovoltaic system under different weather types are determined. The operating data is used to determine the operating demand type of the subsystem. Based on the operating demand type of different subsystems in the target area and the degree of overlap of the demand matching weather types of the subsystems, the update strategy of load control weather type is determined. Based on the updated data of the load control weather type and the on-grid electricity of the distributed photovoltaic system in the load control weather type, the control and processing strategy of the distributed photovoltaic system is determined. Based on the aforementioned control and processing strategy, the distributed photovoltaic system is controlled and processed. Using the control and processing data of the subsystem and the load control subsystem in the load control weather type, the control optimization target for the load control weather type is determined.

2. The big data based distributed photovoltaic system load management method of claim 1, wherein, The new data for the distributed photovoltaic system includes the newly added subsystems in the target area and the operating data of the newly added subsystems.

3. The big data based distributed photovoltaic system load management method of claim 2, wherein, The operating data of the subsystem under different weather types is determined based on the operating time of the subsystem under different weather types.

4. The big data based distributed photovoltaic system load management method of claim 1, wherein, The subsystem is divided based on whether the photovoltaic devices use the same grid-connected device.

5. The big data based distributed photovoltaic system load management method of claim 1, wherein, The method for determining the operational requirement type of the subsystem is as follows: Based on the operational data, the operating time of the subsystem under different weather types is determined; Using the runtime, determine the weather types for which the runtime of the subsystem is less than a preset runtime threshold, and use the weather types for which the runtime of the subsystem is less than the preset runtime threshold as the demand-matching weather types for the subsystem; Based on the weather type matching of the subsystem's requirements, the operational requirement type of the subsystem is determined.

6. The big data based distributed photovoltaic system load management method of claim 5, wherein, The weather type to be matched with the demand is a weather type whose runtime is less than a preset runtime threshold.

7. The load management method for distributed photovoltaic systems based on big data as described in claim 5, characterized in that, Based on the demand matching weather type of the subsystem, the operational demand type of the subsystem is determined, specifically including: Based on the number of weather types that the subsystem's needs match, the subsystem's operational needs types are determined according to the corresponding needs types. The operational needs types include three types: Type 1, Type 2, and Type 3. Type 1 needs are greater than Type 2 needs, and Type 2 needs are greater than Type 3 needs. Type 1 needs have the highest priority, followed by Type 2 needs, and Type 3 needs have the lowest priority.

8. The load management method for distributed photovoltaic systems based on big data as described in claim 1, characterized in that, The control and processing data of the subsystem is determined according to the load control weather type that the subsystem needs to control and process.

9. The load management method for distributed photovoltaic systems based on big data as described in claim 1, characterized in that, The method for determining the control optimization objectives for the aforementioned load control weather type is as follows: Based on the control and processing data of the subsystem, determine the load control weather type that the subsystem needs to control and process, and use the load control weather type that the subsystem needs to control and process as the matching control weather type; The load control subsystem in the load control weather type is determined based on the load control subsystem in the load control weather type; Based on the load control subsystem in the load control weather type and the matching control weather types of different load control subsystems, the control optimization target of the load control weather type is determined.

10. A computer system, comprising: A memory and processor connected by communication, and a computer program stored on the memory and capable of running on the processor, characterized in that, when the processor runs the computer program, it executes a load management method for a distributed photovoltaic system based on big data as described in any one of claims 1-9.

Images