A Distributed Photovoltaic Management Method and System Based on a Four-Fusion Terminal

Abstract

The present invention discloses a distributed photovoltaic management method and system based on a four-in-one fusion terminal, which relates to the technical field of energy management. The system of the present invention includes: a data acquisition and classification module, a data analysis and feature extraction module, a real-time data analysis and strategy matching module, and a control operation and feedback evaluation module; the data acquisition and classification module obtains historical data from a database and classifies it based on switching and flexible control records, forms data units and extracts features; the data analysis and feature extraction module analyzes the features of historical data to identify the key features of switching and flexible control strategies; the real-time data analysis and strategy matching module obtains real-time data, extracts real-time features and compares them with the features of historical data to match corresponding switching and flexible control strategies; the control operation and feedback evaluation module executes corresponding operations according to the matched control strategies, analyzes the feedback data, evaluates the control effect and generates adjustment strategies.

Description

Technical Field

[0001] The present invention relates to the technical field of energy management, and particularly to a distributed photovoltaic management method and system based on a four-in-one fusion terminal. Background Art

[0002] With the increasing global demand for renewable energy, distributed photovoltaic power generation, as a clean and green energy form, has become an important part of microgrids and smart energy systems. Especially in urban and rural areas, distributed photovoltaic systems have been widely used due to their flexibility and environmental friendliness. However, due to the intermittency and instability of photovoltaic power generation, its power generation is affected by various factors such as weather, time, and season, which easily leads to instability of power output. This instability makes distributed photovoltaic systems often face challenges in access and disconnection control when connected to the grid. Especially in microgrids, the instability of photovoltaic power generation may have a negative impact on the frequency, stability, and power quality of the grid.

[0003] Existing distributed photovoltaic management methods mostly rely on traditional single control methods. These methods are difficult to achieve precise scheduling and flexible control of the system in the face of factors such as grid load fluctuations, unstable photovoltaic power generation, and weather changes. In addition, existing management systems often cannot make full use of historical data of distributed photovoltaic systems for scientific analysis, lacking intelligent and personalized control strategies, resulting in the inability to effectively improve photovoltaic power generation efficiency and reduce system operation risks. Summary of the Invention

[0004] The purpose of the present invention is to provide a distributed photovoltaic management method and system based on a four-in-one fusion terminal to solve the problems raised in the above background art.

[0005] To solve the above technical problems, the present invention provides the following technical solutions:

[0006] A distributed photovoltaic management method based on a four-in-one fusion terminal includes the following steps:

[0007] Step S100. Obtain historical data of the distributed photovoltaic system in the low-voltage distribution area and relevant switching and flexible control records from the database, analyze the corresponding historical data according to the switching and flexible control records to obtain historical data characteristics; classify the switching and flexible control records based on the historical data characteristics.

[0008] Step S200. According to the classification results of the switching and flexible control records, for the switching and flexible control records of the same category, obtain the corresponding historical data characteristics, identify key characteristics related to the switching and flexible control strategy by analyzing the historical data characteristics of the switching and flexible control records of the same category and combining the corresponding switching and flexible control records.

[0009] Step S300. Use the four-in-one fusion terminal to obtain the real-time data of the low-voltage distribution photovoltaic system, analyze the real-time data to obtain real-time key features; compare and analyze the real-time key features with the key features corresponding to the switching and flexible control records of each category, so as to match the corresponding switching and flexible control strategies.

[0010] Step S400. According to the matched switching and flexible control strategies, perform corresponding switching and flexible control operations on the current distributed photovoltaic system, and obtain the feedback data after switching and flexible control; analyze the feedback data, evaluate the effect of the current switching and flexible control, generate corresponding adjustment strategies according to the evaluation results, and let relevant personnel handle the generated adjustment strategies accordingly.

[0011] Among them, the four-in-one fusion terminal has the characteristics of "observable, measurable, adjustable, and controllable", and supports real-time data exchange and control with devices from different manufacturers (such as photovoltaic power generation systems, energy storage devices, load management devices, etc.). Through a unified data transmission protocol and communication interface, terminal devices can work effectively in coordination and transmit device data to the cloud or edge computing nodes for processing. This compatibility makes the system more adaptable in terms of device diversity and technology upgrade, facilitating future expansion and upgrade.

[0012] Further, step S100 includes:

[0013] S101. Obtain the historical data of the low-voltage distribution photovoltaic system and related switching and flexible control records from the database. The historical data includes photovoltaic power generation data, environmental data, and load data. The historical data consists of several historical data segments, and each historical data segment has a one-to-one correspondence with a switching and flexible control record; the switching and flexible control record refers to the control behavior record of the switching and flexible control strategy that meets the corresponding system state and requirements during the operation of the low-voltage distribution photovoltaic system; according to the correspondence between the historical data segment and the switching and flexible control record, form a historical data unit with the corresponding historical data segment and the switching and flexible control record.

[0014] S102. For each historical data unit, analyze the corresponding historical data to extract the corresponding historical data features, normalize the values corresponding to the extracted historical data features, thereby forming a historical data feature vector V, and V = [v1, v2,..., vn], where n represents the dimension of the historical data feature vector V, v1 represents the eigenvalue corresponding to the first dimension of the historical data feature vector, v2 represents the eigenvalue corresponding to the second dimension of the historical data feature vector, and so on, vn represents the eigenvalue corresponding to the nth dimension of the historical data feature vector; summarize the historical data feature vectors V of all historical data units, and use the support vector machine method to classify the historical data feature vectors V to obtain switching and flexible control records of several categories.

[0015] Further, step S200 includes:

[0016] S201. According to the classification results of the switching and flexible control records, for the switching and flexible control records of the same category, extract the corresponding historical data feature vector V; for each dimension of the historical data feature vector V, calculate the mean μi and standard deviation σi of the corresponding eigenvalue vi, where i ranges from 1 to n; analyze the degree of difference between the eigenvalue vi and the corresponding mean μi and standard deviation σi, thereby calculating the corresponding difference index Ci, and the specific calculation formula is: Ci = (vi - μi) / σi; summarize the difference indexes Ci of all dimension eigenvalues to form a difference index set CY of the historical data feature vector V, and CY = {C1, C2,..., Cn}, where C1 represents the difference index of the eigenvalue of the first dimension of the historical data feature vector V, C2 represents the difference index of the eigenvalue of the first dimension of the historical data feature vector V, C2 represents the difference index of the eigenvalue of the second dimension of the historical data feature vector V, and so on, Cn represents the difference index of the eigenvalue of the nth dimension of the historical data feature vector V;

[0017] S202. For the difference index set CY of the historical data feature vector V corresponding to the switching and flexible control records of the same category, perform intersection calculations in sequence, summarize all intersections, classify the same intersections into one category, count the number of each intersection category, extract the intersection category with the largest number, and obtain the corresponding historical data feature vector V; according to the dimension of the historical data feature vector corresponding to the elements in the intersection, extract the historical data of the corresponding dimension and mark it as a key feature;

[0018] S203. Extract the switching and flexible control records of the historical data feature vector V corresponding to the intersection category, perform semantic analysis on the switching and flexible control records to obtain the corresponding switching and flexible control strategies, and convert the format of the switching and flexible control strategies to obtain the corresponding switching and flexible control strategy feature vector K, and K = [k1, k2,..., km], where m represents the dimension of the switching and flexible control strategy feature vector K, k1 represents the first dimension eigenvalue of the switching and flexible control strategy feature vector, k2 represents the second dimension eigenvalue of the switching and flexible control strategy feature vector, and so on, km represents the mth dimension eigenvalue of the switching and flexible control strategy feature vector; perform intersection calculation on all the switching and flexible control strategy feature vectors K, summarize the dimension features of the switching and flexible control strategy feature vector corresponding to the intersection calculation result, and correspond the key features to the dimension features of the corresponding switching and flexible control strategy feature vector; traverse the switching and flexible control records of all categories in turn to obtain the key features related to the switching and flexible control strategies.

[0019] Further, step S300 includes:

[0020] S301. Use the four-in-one fusion terminal to obtain the real-time data of the low-voltage distribution photovoltaic system, analyze the real-time data with reference to the analysis method of historical data to obtain the real-time data features, normalize the values corresponding to the real-time data features to form the real-time data feature vector V'; extract the real-time key features of the corresponding category from the real-time data feature vector V' with reference to the key features corresponding to the switching and flexible control records of each category.

[0021] S302. For the key features corresponding to the switching and flexible control records of each category, calculate the similarity between the real-time key features of the corresponding category and the key features corresponding to each switching and flexible control record in turn, and take the average value as the average similarity of the key features of the corresponding category. Select the category with the largest average similarity as the real-time matching category, and use the switching and flexible control strategy corresponding to the key features of the real-time matching category as the currently matched switching and flexible control strategy.

[0022] Further, step S400 includes:

[0023] S401. According to the matched switching and flexible control strategy, perform corresponding switching and flexible control operations on the current distributed photovoltaic system, and obtain the feedback data after switching and flexible control; the switching and flexible control operations include switching operations and flexible control. Among them, the switching operation, according to the control strategy, executes to turn on or off some photovoltaic components, or adjusts their output power. The flexible control adjusts the photovoltaic power generation or the charge and discharge operations of energy storage devices according to the load demand or system status; the feedback data includes photovoltaic power generation data, load data, and environmental data;

[0024] S402. According to the feedback data, analyze the change in photovoltaic power generation, compare the photovoltaic power generation data before and after control, calculate the power change rate ΔP, and ΔP = (P_after - P_before) / P_before, where P_after represents the power generation after control, and P_before represents the power generation before control; compare the load change situation, calculate the change ratio ΔL of the load, and ΔL = (L_after - L_before) / L_before, where L_after and L_before represent the load after control and the load before control respectively; calculate the environmental adjustment coefficient E, and E = P_actual / P_theoretical, where P_actual represents the actual power generation of the photovoltaic components under the current actual environmental conditions, and P_theoretical represents the expected power generation of the photovoltaic components under ideal environmental conditions;

[0025] S403. Comprehensively evaluate the effect of the current switching and flexible control according to the power change rate ΔP, the change ratio ΔL of the load, and the environmental adjustment coefficient E, calculate the corresponding comprehensive evaluation index R, and R = w1×ΔP + w2×ΔL + w3×E, where w1, w2, and w3 represent different weight coefficients, and w1 + w2 + w3 = 1; compare the calculated comprehensive evaluation index R with the corresponding threshold R0. If R≥R0, it means that the currently matched switching and flexible control strategy is reasonable, and no operation is performed; if R<R0, it means that the currently matched switching and flexible control strategy is unreasonable, and a corresponding adjustment strategy is generated according to the values of the power change rate ΔP, the change ratio ΔL of the load, and the environmental adjustment coefficient E, and the adjustment strategy is output to the relevant personnel, and the relevant personnel perform corresponding processing on the generated adjustment strategy.

[0026] A distributed photovoltaic management system based on a four - fold fusion terminal, including: a data acquisition and classification module, a data analysis and feature extraction module, a real - time data analysis and strategy matching module, and a control operation and feedback evaluation module;

[0027] The data acquisition and classification module obtains the historical data of the low-voltage distribution network area distributed photovoltaic system and the relevant switching and flexible control records, analyzes the corresponding historical data according to the switching and flexible control records, so as to obtain the historical data characteristics; based on the historical data characteristics, classifies the switching and flexible control records;

[0028] According to the classification results of the switching and flexible control records, the data analysis and feature extraction module obtains the corresponding historical data characteristics for the switching and flexible control records of the same category, analyzes the historical data characteristics of the switching and flexible control records of the same category, and combines the corresponding switching and flexible control records to identify the key features related to the switching and flexible control strategy;

[0029] The real-time data analysis and strategy matching module obtains the real-time data of the low-voltage distribution network area distributed photovoltaic system, analyzes the real-time data, so as to obtain the real-time key features; compares and analyzes the real-time key features with the key features corresponding to each category of switching and flexible control records, so as to match the corresponding switching and flexible control strategy;

[0030] According to the matched switching and flexible control strategy, the control operation and feedback evaluation module performs corresponding switching and flexible control operations on the current distributed photovoltaic system, and obtains the feedback data after switching and flexible control; analyzes the feedback data, evaluates the effect of the current switching and flexible control, and generates a corresponding adjustment strategy according to the evaluation result, and the relevant personnel perform corresponding processing on the generated adjustment strategy.

[0031] Furthermore, the data acquisition and classification module includes a data acquisition unit and a data classification unit;

[0032] The data acquisition unit obtains the historical data of the low-voltage distribution network area distributed photovoltaic system and the corresponding switching and flexible control records from the database; the data classification unit analyzes the corresponding historical data according to the switching and flexible control records, so as to obtain the historical data characteristics; based on the historical data characteristics, classifies the switching and flexible control records.

[0033] Furthermore, the data analysis and feature extraction module includes a data analysis unit and a key feature extraction unit;

[0034] According to the classification results of the switching and flexible control records, the data analysis unit extracts the corresponding historical data feature vectors for the switching and flexible control records of the same category; and calculates the difference degree indexes of each dimension of the historical data feature vectors according to the historical data feature vectors; the key feature extraction unit establishes a difference index set according to the difference degree indexes of each dimension of the historical data feature vectors, and identifies the key features related to the switching and flexible control strategy through the analysis of the difference index set and the intersection calculation.

[0035] Furthermore, the real-time data analysis and strategy matching module includes a real-time data analysis unit and a strategy matching unit;

[0036] The real-time data analysis unit collects the real-time data of the photovoltaic system through the four-in-one fusion terminal, analyzes and processes the real-time data, extracts the real-time data features, and generates a real-time data feature vector; referring to the key features corresponding to the switching and flexible control records of each category, the real-time key features of the corresponding category are sequentially extracted from the real-time data feature vector;

[0037] The strategy matching unit calculates the similarity between the real-time key features of the corresponding category and the key features corresponding to each switching and flexible control record in sequence for the key features corresponding to the switching and flexible control records of each category, and calculates the average value as the average similarity of the key features of the corresponding category. The category with the largest average similarity is selected as the real-time matching category, and the switching and flexible control strategy corresponding to the key features of the real-time matching category is used as the currently matched switching and flexible control strategy.

[0038] Furthermore, the control operation and feedback evaluation module includes a control operation unit and a feedback evaluation unit;

[0039] The control operation unit executes corresponding operations according to the switched and flexibly controlled strategy matched in real time and collects the feedback data after control; the feedback evaluation unit analyzes the changes in photovoltaic power generation, load changes, and environmental adjustment coefficients based on the feedback data, calculates the comprehensive evaluation index, comprehensively evaluates the effect of switching and flexible control, generates an adjustment strategy based on the comprehensive evaluation result, and feeds back the adjustment strategy to relevant personnel for corresponding processing by relevant personnel.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows: By utilizing historical data features, real-time data analysis and matching, the present invention can accurately identify and match the switching and flexible control strategies most suitable for the current system state, realizing intelligent scheduling and flexible control of the distributed photovoltaic system; this method avoids the problems of relying only on simple rules or manual intervention in traditional methods, and improves the automation degree and response speed of the system. By deeply analyzing historical data features and real-time data features, the operating state of the photovoltaic system can be accurately grasped, and the best switching and flexible control strategies can be selected, thereby effectively improving the power generation efficiency of photovoltaic power generation and minimizing the power generation instability caused by factors such as weather changes and load fluctuations. The management method of the present invention combines the historical records and real-time feedback of switching and flexible control, and continuously monitors and optimizes the adjustment based on real-time data, effectively avoiding risks such as grid frequency fluctuations and power quality degradation caused by the instability of photovoltaic power generation in traditional methods, and ensuring the stability and safe operation of the power grid. Through the comprehensive evaluation of real-time feedback data, including the power change rate, load change ratio and environmental adjustment coefficient, the effect of the control strategy can be quantitatively evaluated, providing a scientific basis for subsequent decision-making; through the intelligent generation of adjustment strategies, the changes in grid demand and environmental conditions can be effectively responded to, avoiding the lag and inefficiency of traditional control strategies. The present invention combines feedback data for dynamic evaluation, and effectively improves the adaptability and stability of the distributed photovoltaic system by adjusting the working states of the photovoltaic power generation system and energy storage devices in real time; compared with the prior art, the present invention can timely adjust the control strategy when the system load fluctuates or the weather changes, ensuring the efficient operation of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation to the present invention. In the accompanying drawings:

[0042] Figure 1 is a schematic diagram of the modules of a distributed photovoltaic management system based on a four-in-one fusion terminal of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0044] Please refer to Figure 1 , the present invention provides the following technical solutions:

[0045] A distributed photovoltaic management system based on a four-integrated terminal, comprising: a data acquisition and classification module, a data analysis and feature extraction module, a real-time data analysis and strategy matching module, and a control operation and feedback evaluation module;

[0046] The data acquisition and classification module obtains the historical data of the distributed photovoltaic system in the low-voltage distribution area and the relevant switching and flexible control records, analyzes the corresponding historical data according to the switching and flexible control records, so as to obtain historical data features; based on the historical data features, classifies the switching and flexible control records;

[0047] According to the classification results of the switching and flexible control records, the data analysis and feature extraction module obtains the corresponding historical data features for the switching and flexible control records of the same category, analyzes the historical data features of the switching and flexible control records of the same category, and combines the corresponding switching and flexible control records to identify the key features related to the switching and flexible control strategy;

[0048] The real-time data analysis and strategy matching module obtains the real-time data of the distributed photovoltaic system in the low-voltage distribution area, analyzes the real-time data, so as to obtain real-time key features; compares and analyzes the real-time key features with the key features corresponding to each category of switching and flexible control records, so as to match the corresponding switching and flexible control strategy;

[0049] According to the matched switching and flexible control strategy, the control operation and feedback evaluation module performs corresponding switching and flexible control operations on the current distributed photovoltaic system, and obtains the feedback data after switching and flexible control; analyzes the feedback data, evaluates the effect of the current switching and flexible control, and generates a corresponding adjustment strategy according to the evaluation result, and relevant personnel perform corresponding processing on the generated adjustment strategy.

[0050] The data acquisition and classification module includes a data acquisition unit and a data classification unit;

[0051] The data acquisition unit obtains the historical data of the distributed photovoltaic system in the low-voltage distribution area from the database, as well as the corresponding switching and flexible control records; the data classification unit analyzes the corresponding historical data according to the switching and flexible control records, so as to obtain historical data features; based on the historical data features, classifies the switching and flexible control records.

[0052] The data analysis and feature extraction module includes a data analysis unit and a key feature extraction unit;

[0053] The data analysis unit extracts the corresponding historical data feature vectors for the switching and flexible control records of the same category according to the classification results of the switching and flexible control records; and calculates the difference degree indexes of each dimension of the historical data feature vectors according to the historical data feature vectors; the key feature extraction unit establishes a difference index set according to the difference degree indexes of each dimension of the historical data feature vectors, and identifies the key features related to the switching and flexible control strategies through the analysis of the difference index set and the intersection calculation.

[0054] The real-time data analysis and strategy matching module includes a real-time data analysis unit and a strategy matching unit;

[0055] The real-time data analysis unit collects the real-time data of the photovoltaic system through the four-in-one fusion terminal in real time, analyzes and processes the real-time data, extracts the real-time data features, and generates real-time data feature vectors; referring to the key features corresponding to the switching and flexible control records of each category, the real-time key features of the corresponding category are sequentially extracted from the real-time data feature vectors;

[0056] The strategy matching unit calculates the similarity between the real-time key features of the corresponding category and the key features corresponding to each switching and flexible control record in sequence for the key features corresponding to the switching and flexible control records of each category, and takes the average value as the average similarity of the key features of the corresponding category, selects the category with the largest average similarity as the real-time matching category, and uses the switching and flexible control strategy corresponding to the key features of the real-time matching category as the currently matched switching and flexible control strategy.

[0057] The control operation and feedback evaluation module includes a control operation unit and a feedback evaluation unit;

[0058] The control operation unit executes corresponding operations according to the real-time matched switching and flexible control strategy, and collects the feedback data after control; the feedback evaluation unit analyzes the changes in photovoltaic power generation, load changes, and environmental adjustment coefficients according to the feedback data, calculates the comprehensive evaluation index, comprehensively evaluates the switching and flexible control effect, generates an adjustment strategy according to the comprehensive evaluation result, and feeds back the adjustment strategy to relevant personnel for corresponding processing by relevant personnel.

[0059] A distributed photovoltaic management method based on a four-in-one fusion terminal includes the following steps:

[0060] Step S100. Obtain the historical data of the low-voltage distribution area photovoltaic system and the relevant switching and flexible control records from the database, analyze the corresponding historical data according to the switching and flexible control records, so as to obtain historical data features; classify the switching and flexible control records based on the historical data features;

[0061] Step S200. According to the classification results of switching and flexible control records, for switching and flexible control records of the same category, obtain the corresponding historical data features. By analyzing the historical data features of switching and flexible control records of the same category and combining with the corresponding switching and flexible control records, identify the key features related to the switching and flexible control strategy;

[0062] Step S300. Use the four-in-one fusion terminal to obtain the real-time data of the low-voltage distribution photovoltaic system, analyze the real-time data to obtain the real-time key features; compare and analyze the real-time key features with the key features corresponding to each category of switching and flexible control records, so as to match the corresponding switching and flexible control strategy;

[0063] Step S400. According to the matched switching and flexible control strategy, perform corresponding switching and flexible control operations on the current distributed photovoltaic system, and obtain the feedback data after switching and flexible control; analyze the feedback data, evaluate the effect of the current switching and flexible control, and generate corresponding adjustment strategies according to the evaluation results, and relevant personnel shall perform corresponding processing on the generated adjustment strategies.

[0064] Among them, the four-in-one fusion terminal has the characteristics of "observable, measurable, adjustable, and controllable", and supports real-time data exchange and control with equipment from different manufacturers (such as photovoltaic power generation systems, energy storage devices, load management devices, etc.). Through a unified data transmission protocol and communication interface, terminal devices can effectively cooperate with each other and transmit device data to the cloud or edge computing nodes for processing. This compatibility makes the system more adaptable in terms of equipment diversity and technology upgrade, facilitating future expansion and upgrade.

[0065] Step S100 includes:

[0066] S101. Obtain the historical data of the low-voltage distribution photovoltaic system and related switching and flexible control records from the database. The historical data includes photovoltaic power generation data, environmental data, and load data. The historical data is composed of several historical data segments, and each historical data segment has a one-to-one correspondence with a switching and flexible control record; the switching and flexible control record refers to the control behavior record of the switching and flexible control strategy that meets the corresponding system state and requirements during the operation of the low-voltage distribution photovoltaic system; according to the correspondence between the historical data segment and the switching and flexible control record, form a historical data unit with the corresponding historical data segment and the switching and flexible control record;

[0067] S102. For each historical data unit, analyze the corresponding historical data to extract the corresponding historical data features, normalize the values corresponding to the extracted historical data features, thereby forming a historical data feature vector V, and V = [v1, v2,..., vn], where n represents the dimension of the historical data feature vector V, v1 represents the eigenvalue corresponding to the first dimension of the historical data feature vector, v2 represents the eigenvalue corresponding to the second dimension of the historical data feature vector, and so on, vn represents the eigenvalue corresponding to the nth dimension of the historical data feature vector; summarize the historical data feature vectors V of all historical data units, and use the support vector machine method to classify the historical data feature vector V, thereby obtaining switching and flexible control records of several categories.

[0068] In this embodiment, the historical data features respectively include data features corresponding to photovoltaic power generation data, environmental data, and load data. It is assumed that the historical data features of photovoltaic power generation data include average power generation, power fluctuation characteristics, and power generation efficiency, etc., the historical data features of environmental data include the correlation between temperature and power generation, average value, and standard deviation, etc., the historical data features of load data include load fluctuation characteristics, average value, and standard deviation, etc., and the selection of specific historical data features is determined by relevant personnel.

[0069] Step S200 includes:

[0070] S201. According to the classification results of the switching and flexible control records, for the switching and flexible control records of the same category, extract the corresponding historical data feature vector V; for each dimension of the historical data feature vector V, calculate the mean μi and standard deviation σi of the corresponding eigenvalue vi, where i ranges from 1 to n; analyze the degree of difference between the eigenvalue vi and the corresponding mean μi and standard deviation σi, thereby calculating the corresponding difference index Ci, and the specific calculation formula is: Ci = (vi - μi) / σi; summarize the difference indexes Ci of all dimension eigenvalues, thereby forming a difference index set CY of the historical data feature vector V, and CY = {C1, C2,..., Cn}, where C1 represents the difference index of the eigenvalue of the first dimension of the historical data feature vector V, C2 represents the difference index of the eigenvalue of the first dimension of the historical data feature vector V, C2 represents the difference index of the eigenvalue of the second dimension of the historical data feature vector V, and so on, Cn represents the difference index of the eigenvalue of the nth dimension of the historical data feature vector V;

[0071] S202. For the switching and flexible control of the same category, calculate the difference index set CY of the corresponding historical data feature vectors V in sequence, perform intersection calculations in turn, summarize all intersections, classify the same intersections into one category, count the number of each intersection category, extract the intersection category with the largest number, and obtain the corresponding historical data feature vector V; according to the dimensions of the historical data feature vectors corresponding to the elements in the intersection, extract the historical data corresponding to the corresponding dimensions and mark it as key features;

[0072] S203. Extract the switching and flexible control records of the historical data feature vector V corresponding to the intersection category, perform semantic analysis on the switching and flexible control records to obtain the corresponding switching and flexible control strategies, and convert the format of the switching and flexible control strategies to obtain the corresponding switching and flexible control strategy feature vector K, and K = [k1, k2,..., km], where m represents the dimension of the switching and flexible control strategy feature vector K, k1 represents the first dimension eigenvalue of the switching and flexible control strategy feature vector, k2 represents the second dimension eigenvalue of the switching and flexible control strategy feature vector, and so on, km represents the mth dimension eigenvalue of the switching and flexible control strategy feature vector; perform intersection calculations on all the switching and flexible control strategy feature vectors K, summarize the dimension features of the switching and flexible control strategy feature vectors corresponding to the intersection calculation results, and correspond the key features to the dimension features of the corresponding switching and flexible control strategy feature vectors; traverse the switching and flexible control records of all categories in turn to obtain the key features related to the switching and flexible control strategies.

[0073] In this embodiment, assume that for a certain type of switching and flexible control records, the corresponding historical data feature vector V is extracted. Assume that there are 3 switching and flexible control records of this type, namely T1, T2, and T3, and the corresponding historical data feature vectors of the switching and flexible control records are V1, V2, and V3 respectively; for the eigenvalues of each dimension of the historical data feature vectors V1, V2, and V3, calculate the corresponding mean μ and standard deviation σ. Taking the eigenvalue v1 of the first dimension as an example, then the corresponding mean μ1 and standard deviation σ1; analyze the degree of difference between the eigenvalue v1 and the corresponding mean μ1 and standard deviation σ1, so as to calculate the corresponding difference index C11, and the specific calculation formula is: C11 = (v1 - μ1) / σ1; for the eigenvalues of the first dimension of the historical data feature vectors V1, V2, and V3, calculate the corresponding difference index Ci, so as to obtain C11, C12, and C13; summarize the difference indexes Ci of all dimensions of the historical data feature vectors V1, V2, and V3, and form the corresponding difference index sets CY1, CY2, and CY3; for the difference index sets CY1, CY2, and CY3, perform intersection calculations in turn, that is, CY1∩CY2, CY1∩CY3, and CY2∩CY3. Assume that CY1∩CY2 = CY1∩CY3 ≠ CY2∩CY3, then classify the same intersections into one category. Assume that CY1∩CY2 = CY1∩CY3, denoted as set A, and CY2∩CY3 is denoted as set B. Since the number of elements in set A is 2 and the number of elements in set B is 1, extract the intersection category with the largest number, and the intersection result is set A; assume that the dimensions of the historical data feature vectors corresponding to the elements in set A are 2, 3, and 5, extract the historical data features v2, v3, and v5 with dimensions 2, 3, and 5, and mark them as key features.

[0074] Respectively extract the switching and flexible control records T1, T2, and T3 of the historical data feature vector V corresponding to set A, perform semantic analysis on the switching and flexible control records, and convert the format of the switching and flexible control strategy to obtain the corresponding switching and flexible control strategy feature vector K, denoted as K1, K2, and K3 respectively; obtain the dimension features of the switching and flexible control strategy feature vector corresponding to K1∩K2∩K3. Assume they are k1, k2, and k4, then correspond the key features v2, v3, and v5 with k1, k2, and k4, so as to obtain the key features v2, v3, and v5 of the switching and flexible control strategy corresponding to k1, k2, and k4.

[0075] Step S300 includes:

[0076] S301. Obtain the real-time data of the low-voltage distribution photovoltaic system using a four-in-one fusion terminal, analyze the real-time data with reference to the analysis method of historical data to obtain the real-time data characteristics, normalize the values corresponding to the real-time data characteristics, and thus form a real-time data feature vector V'; With reference to the key features corresponding to the switching and flexible control records of each category, sequentially extract the real-time key features of the corresponding category from the real-time data feature vector V'.

[0077] S302. For the key features corresponding to the switching and flexible control records of each category, calculate the similarity between the real-time key features of the corresponding category and the key features of each switching and flexible control record in sequence, and take the average value as the average similarity of the key features of the corresponding category. Select the category with the largest average similarity as the real-time matching category, and use the switching and flexible control strategy corresponding to the key features of the real-time matching category as the currently matched switching and flexible control strategy.

[0078] In this embodiment, assume that the real-time data feature vector is V', and assume that the switching and flexible control records are divided into 2 categories, namely category 1 and category 2. Taking category 1 as an example, assume that the key features corresponding to category 1 are: "photovoltaic power generation" and "ambient temperature"; According to the key features corresponding to category 1, extract the features corresponding to "photovoltaic power generation" and "ambient temperature" from the real-time data feature vector V' as real-time key features; Assume that the historical data features corresponding to "photovoltaic power generation" and "ambient temperature" are calculated for similarity with the real-time key features, and the similarity calculation methods include Euclidean distance, cosine similarity, etc. Calculate the similarity between the historical data features corresponding to "photovoltaic power generation" and the real-time key features to obtain a similarity value S1, calculate the similarity between the historical data features corresponding to "ambient temperature" and the real-time key features to obtain a similarity value S2, and calculate the average value, that is, (S1 + S2) / 2 is the average similarity of category 1; Perform the same operation for category 2, compare the magnitude relationship between the average similarity of category 1 and the average similarity of category 2. Assume that the average similarity of category 1 is large, then use the switching and flexible control strategy corresponding to the key features of category 1 as the currently matched switching and flexible control strategy.

[0079] Step S400 includes:

[0080] S401. According to the matched switching and flexible control strategy, perform corresponding switching and flexible control operations on the current distributed photovoltaic system, and obtain the feedback data after switching and flexible control; the switching and flexible control operations include switching operations and flexible control. Among them, the switching operation, according to the control strategy, executes to turn on or off some photovoltaic modules, or adjust their output power, and the flexible control adjusts the photovoltaic power generation or the charge and discharge operation of the energy storage device according to the load demand or the system state; the feedback data includes photovoltaic power generation data, load data, and environmental data;

[0081] S402. According to the feedback data, analyze the change in photovoltaic power generation, compare the photovoltaic power generation data before and after control, calculate the power change rate ΔP, and ΔP = (P_after - P_before) / P_before, where P_after represents the power generation after control, and P_before represents the power generation before control; compare the load change situation, calculate the change ratio ΔL of the load, and ΔL = (L_after - L_before) / L_before, where L_after and L_before represent the load after control and the load before control respectively; calculate the environmental adjustment coefficient E, and E = P_actual / P_theoretical, where P_actual represents the actual power generation of the photovoltaic module under the current actual environmental conditions, and P_theoretical represents the expected power generation of the photovoltaic module under ideal environmental conditions;

[0082] S403. Comprehensively evaluate the effect of the current switching and flexible control according to the power change rate ΔP, the change ratio ΔL of the load, and the environmental adjustment coefficient E, calculate the corresponding comprehensive evaluation index R, and R = w1×ΔP + w2×ΔL + w3×E, where w1, w2, and w3 represent different weight coefficients, and w1 + w2 + w3 = 1; compare the calculated comprehensive evaluation index R with the corresponding threshold R0. If R ≥ R0, it means that the currently matched switching and flexible control strategy is reasonable, and no operation is performed; if R < R0, it means that the currently matched switching and flexible control strategy is unreasonable, then generate a corresponding adjustment strategy according to the values of the power change rate ΔP, the change ratio ΔL of the load, and the environmental adjustment coefficient E, and output the adjustment strategy to the relevant personnel, and the relevant personnel perform corresponding processing on the generated adjustment strategy.

[0083] In this embodiment, assume that the calculated comprehensive evaluation index R is compared with the corresponding threshold R0, and the corresponding size relationship is: R < R0, then obtain the corresponding power change rate ΔP, the change ratio ΔL of the load, and the environmental adjustment coefficient E. Assume there are the following situations:

[0084] For the power change rate (ΔP) and the adjustment strategy:

[0085] Case 1: The power change rate (ΔP) deviates positively (e.g., ΔP > 0.1)

[0086] Adjustment strategy: If the photovoltaic power generation increases significantly, it may cause grid overload or the load may not consume the excess generated power in time. At this time:

[0087] Increase the charging operation of energy storage devices: Store the excess photovoltaic power generation in the energy storage system to avoid power waste.

[0088] Turn off some photovoltaic modules: Turn off or reduce the output power of some photovoltaic modules to reduce power generation and make the system more matched with the load demand.

[0089] Case 2: The power change rate (ΔP) deviates negatively (e.g., ΔP < -0.1)

[0090] Adjustment strategy: If the photovoltaic power generation decreases significantly, it may be due to weather changes, component failures or other external factors. At this time:

[0091] Start standby photovoltaic modules: Turn on more photovoltaic modules to increase the power generation.

[0092] Increase the discharging operation of energy storage devices: Supplement the power shortage through the energy storage devices to maintain the stable operation of the system.

[0093] Adjust the load demand: Adjust the electricity consumption demand on the load side through flexible control and preferentially use the electricity in the energy storage devices.

[0094] For the load change ratio (ΔL) and adjustment strategy:

[0095] Case 1: The load change ratio (ΔL) deviates positively (e.g., ΔL > 0.1)

[0096] Adjustment strategy: The load demand increases, which may cause the system burden to increase. At this time:

[0097] Adjust the photovoltaic power generation: If the photovoltaic power generation is sufficient, appropriately increase the photovoltaic output to meet the additional load demand.

[0098] Start the discharging of energy storage devices: Provide power for the additional load through the energy storage devices to reduce the dependence on the grid.

[0099] Case 2: The load change ratio (ΔL) deviates negatively (e.g., ΔL < -0.1)

[0100] Adjustment strategy: The load demand decreases, which may cause power generation surplus. At this time:

[0101] Reduce photovoltaic power generation: properly shut down some photovoltaic panels to avoid excessive power output and maintain system balance.

[0102] Increase energy storage device charging operations: Absorb excess power through the energy storage system to prevent system overload.

[0103] For the environmental adjustment factor (E) and adjustment strategy:

[0104] Case 1: The environmental adjustment factor (E) is lower than expected (e.g. E < 0.8)

[0105] Adjustment strategy: If the actual power generated by the photovoltaic module is far lower than the expected power under ideal conditions, it means that the light or environmental conditions are poor. At this time:

[0106] Start backup PV panels or increase power output: If there are backup equipment or PV panels that can increase power, start these panels to supplement the insufficient power generation.

[0107] Increase the discharge of energy storage equipment: Under unfavorable environmental conditions, energy storage equipment can be used to supplement electricity to ensure that load demand is not affected.

[0108] Case 2: The environmental adjustment factor (E) is high (e.g. E>1.2)

[0109] Adjustment strategy: If the actual power generation of the PV panels exceeds expectations (which means that the environmental conditions are good, perhaps there is sufficient sunlight), then:

[0110] Increase energy storage device charging operation: store excess power generation for future use.

[0111] Adjust load demand or shut down some PV panels: If load demand does not increase accordingly, some PV panels need to be shut down to avoid excess electricity.

[0112] Comprehensive regulation strategy generation:

[0113] Based on the analysis of power change rate (ΔP), load change ratio (ΔL) and environmental adjustment coefficient (E), combined with the weight of each factor and the overall operation requirements of the system, a comprehensive regulation strategy is generated. For example:

[0114] If ΔP>0.1, ΔL>0.1, E<0.8, then: increase the charging operation of energy storage equipment, start more PV panels, increase the power generation capacity to meet the additional load demand, and if the weather is bad, adjust the output of PV panels appropriately.

[0115] If ΔP<-0.1, ΔL<-0.1, E>1.2, then: shut down some PV panels, reduce power generation, increase the discharge operation of energy storage equipment, maintain a stable power supply, adjust load demand, and give priority to the use of energy storage power.

[0116] According to the generated adjustment strategy above, pass it to relevant personnel or an automated control system to perform corresponding operations. These strategies may include: adjusting the component switches and power output of the photovoltaic system, controlling the charging and discharging behavior of the energy storage system, and adjusting the load management system to respond to changes in power demand.

[0117] It should be noted that in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device.

[0118] Finally, it should be noted that the above are only preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A distributed photovoltaic management method based on a four-integrated terminal, characterized in that: The method includes the following steps: Step S100. Obtain the historical data of the low-voltage distribution area distributed photovoltaic system and the relevant switching and flexible control records from the database, analyze the corresponding historical data according to the switching and flexible control records to obtain historical data characteristics; based on the historical data characteristics, classify the switching and flexible control records. Step S200. According to the classification result of the switching and flexible control records, for the switching and flexible control records of the same category, obtain the corresponding historical data characteristics, identify the key characteristics related to the switching and flexible control strategy by analyzing the historical data characteristics of the switching and flexible control records of the same category and combining the corresponding switching and flexible control records. The step S200 includes: S201. According to the classification result of the switching and flexible control records, for the switching and flexible control records of the same category, extract the corresponding historical data feature vector V; for each dimension of the historical data feature vector V, calculate the mean value μi and the standard deviation σi of the corresponding eigenvalue vi, where i ranges from 1 to n; analyze the degree of difference between the eigenvalue vi and the corresponding mean value μi and standard deviation σi, so as to calculate the corresponding difference index Ci, and the specific calculation formula is: Ci = (vi - μi) / σi; summarize the difference indexes Ci of all dimension eigenvalues to form the difference index set CY of the historical data feature vector V, and CY = {C1, C2,..., Cn}, where C1 represents the difference index of the first dimension eigenvalue of the historical data feature vector V, C2 represents the difference index of the first dimension eigenvalue of the historical data feature vector V, C2 represents the difference index of the second dimension eigenvalue of the historical data feature vector V, and so on, Cn represents the difference index of the nth dimension eigenvalue of the historical data feature vector V. S202. For the difference index set CY of the historical data feature vector V corresponding to the switching and flexible control records of the same category, perform intersection calculations in sequence, summarize all intersections, classify the same intersections into one category, count the number of each intersection category, extract the intersection category with the largest number, and obtain the corresponding historical data feature vector V; according to the dimension of the historical data feature vector corresponding to the elements in the intersection, extract the historical data features of the corresponding dimension and mark them as key features. S203. Extract the switching and flexible control records of the historical data feature vector V corresponding to the intersection category, perform semantic analysis on the switching and flexible control records, so as to obtain the corresponding switching and flexible control strategies, and convert the format of the switching and flexible control strategies to obtain the corresponding switching and flexible control strategy feature vector K, and K = [k1, k2,..., km], where m represents the dimension of the switching and flexible control strategy feature vector K, k1 represents the first dimension eigenvalue of the switching and flexible control strategy feature vector, k2 represents the second dimension eigenvalue of the switching and flexible control strategy feature vector, and so on, km represents the mth dimension eigenvalue of the switching and flexible control strategy feature vector; perform intersection calculation on all the switching and flexible control strategy feature vectors K, summarize the dimension features of the switching and flexible control strategy feature vector corresponding to the intersection calculation result, and correspond the key features to the dimension features of the corresponding switching and flexible control strategy feature vector; traverse the switching and flexible control records of all categories in turn, so as to obtain the key features related to the switching and flexible control strategy. Step S300. Use the four-in-one fusion terminal to obtain the real-time data of the low-voltage distribution network photovoltaic system, analyze the real-time data, so as to obtain the real-time key features; compare and analyze the real-time key features with the key features corresponding to the switching and flexible control records of each category, so as to match the corresponding switching and flexible control strategies. Step S400. According to the matched switching and flexible control strategies, perform corresponding switching and flexible control operations on the current distributed photovoltaic system, and obtain the feedback data after switching and flexible control; analyze the feedback data, evaluate the effect of the current switching and flexible control, and generate corresponding adjustment strategies according to the evaluation results, and relevant personnel perform corresponding processing on the generated adjustment strategies.

2. The distributed photovoltaic management method based on a four-in-one fusion terminal according to claim 1, wherein: The said step S100 includes: S101. Obtain the historical data of the low-voltage distribution network photovoltaic system and the relevant switching and flexible control records from the database. The historical data includes photovoltaic power generation data, environmental data, and load data. The historical data is composed of several historical data segments, and each historical data segment has a one-to-one correspondence with a switching and flexible control record; the switching and flexible control record refers to the control behavior record of the switching and flexible control strategy that meets the corresponding system state and requirements during the operation of the low-voltage distribution network photovoltaic system; according to the correspondence between the historical data segment and the switching and flexible control record, form a historical data unit with the corresponding historical data segment and the switching and flexible control record. S102. For each historical data unit, analyze the corresponding historical data to extract the corresponding historical data features, normalize the values corresponding to the extracted historical data features, thereby forming a historical data feature vector V, and V = [v1, v2,..., vn], where n represents the dimension of the historical data feature vector V, v1 represents the eigenvalue corresponding to the first dimension of the historical data feature vector, v2 represents the eigenvalue corresponding to the second dimension of the historical data feature vector, and so on, vn represents the eigenvalue corresponding to the nth dimension of the historical data feature vector; summarize the historical data feature vectors V of all historical data units, and use the support vector machine method to classify the historical data feature vector V, thereby obtaining switching and flexible control records of several categories.

3. A distributed photovoltaic management method based on a four-in-one fusion terminal according to claim 1, characterized in that: The step S300 includes: S301. Use the four-in-one fusion terminal to obtain the real-time data of the low-voltage distribution photovoltaic system, analyze the real-time data with reference to the analysis method of historical data to obtain real-time data features, normalize the values corresponding to the real-time data features, thereby forming a real-time data feature vector V'; with reference to the key features corresponding to the switching and flexible control records of each category, sequentially extract the real-time key features of the corresponding category from the real-time data feature vector V'; S302. For the key features corresponding to the switching and flexible control records of each category, calculate the similarity between the real-time key features of the corresponding category and the key features corresponding to each switching and flexible control record in sequence, and take the average value as the average similarity of the key features of the corresponding category. Select the category with the largest average similarity as the real-time matching category, and use the switching and flexible control strategy corresponding to the key features of the real-time matching category as the currently matched switching and flexible control strategy.

4. The distributed photovoltaic management method based on a four-in-one fusion terminal according to claim 3, characterized in that: The step S400 includes: S401. According to the matched switching and flexible control strategy, perform corresponding switching and flexible control operations on the current distributed photovoltaic system, and obtain the feedback data after switching and flexible control; the switching and flexible control operations include switching operations and flexible control. Among them, the switching operation, according to the control strategy, executes to turn on or off some photovoltaic modules, or adjusts their output power, and the flexible control adjusts the photovoltaic power generation or the charge and discharge operation of the energy storage device according to the load demand or the system state; the feedback data includes photovoltaic power generation data, load data, and environmental data; S402. Analyze the change in photovoltaic power generation according to the feedback data, compare the photovoltaic power generation data before and after control, calculate the power change rate ΔP, and ΔP = (P_after - P_before) / P_before, where P_after represents the power generation after control and P_before represents the power generation before control; compare the load change situation, calculate the change ratio ΔL of the load, and ΔL = (L_after - L_before) / L_before, where L_after and L_before represent the load after control and the load before control respectively; calculate the environmental adjustment coefficient E, and E = P_actual / P_theoretical, where P_actual represents the actual power generation of the photovoltaic module under the current actual environmental conditions and P_theoretical represents the expected power generation of the photovoltaic module under ideal environmental conditions; S403. Comprehensively evaluate the effect of the current switching and flexible control according to the power change rate ΔP, the change ratio ΔL of the load, and the environmental adjustment coefficient E, calculate the corresponding comprehensive evaluation index R, and R = w1×ΔP + w2×ΔL + w3×E, where w1, w2, and w3 represent different weight coefficients and w1 + w2 + w3 = 1; compare the calculated comprehensive evaluation index R with the corresponding threshold R0. If R ≥ R0, it means that the currently matched switching and flexible control strategy is reasonable, and no operation is performed; if R < R0, it means that the currently matched switching and flexible control strategy is unreasonable, and a corresponding adjustment strategy is generated according to the values of the power change rate ΔP, the change ratio ΔL of the load, and the environmental adjustment coefficient E, and the adjustment strategy is output to the relevant personnel, and the relevant personnel perform corresponding processing on the generated adjustment strategy.

5. A distributed photovoltaic management system based on a four-in-one fusion terminal, which is applied to a distributed photovoltaic management method based on a four-in-one fusion terminal described in any one of claims 1-4, and is characterized in that: The system includes: a data acquisition and classification module, a data analysis and feature extraction module, a real-time data analysis and strategy matching module, and a control operation and feedback evaluation module; The data acquisition and classification module obtains the historical data of the low-voltage distribution area distributed photovoltaic system and the relevant switching and flexible control records, analyzes the corresponding historical data according to the switching and flexible control records, so as to obtain historical data characteristics; based on the historical data characteristics, classifies the switching and flexible control records; The data analysis and feature extraction module, according to the classification results of the switching and flexible control records, for the switching and flexible control records of the same category, obtains the corresponding historical data characteristics, analyzes the historical data characteristics of the switching and flexible control records of the same category, and combines the corresponding switching and flexible control records to identify the key characteristics related to the switching and flexible control strategy; The real-time data analysis and strategy matching module obtains the real-time data of the low-voltage distribution area distributed photovoltaic system, analyzes the real-time data, so as to obtain real-time key characteristics; compares and analyzes the real-time key characteristics with the key characteristics corresponding to each category of switching and flexible control records, so as to match the corresponding switching and flexible control strategy; The control operation and feedback evaluation module performs corresponding switching and flexible control operations on the current distributed photovoltaic system according to the matched switching and flexible control strategy, and obtains the feedback data after switching and flexible control; analyzes the feedback data, evaluates the effects of the current switching and flexible control, generates corresponding adjustment strategies according to the evaluation results, and relevant personnel perform corresponding processing on the generated adjustment strategies.

6. The distributed photovoltaic management system based on a four-in-one fusion terminal according to claim 5, characterized in that: The data acquisition and classification module includes a data acquisition unit and a data classification unit; The data acquisition unit obtains the historical data of the distributed photovoltaic system in the low-voltage substation area and the corresponding switching and flexible control records from the database; the data classification unit analyzes the corresponding historical data according to the switching and flexible control records to obtain historical data characteristics; based on the historical data characteristics, classifies the switching and flexible control records.

7. A distributed photovoltaic management system based on a four-in-one fusion terminal according to claim 5, characterized in that: The data analysis and feature extraction module includes a data analysis unit and a key feature extraction unit; The data analysis unit extracts the corresponding historical data feature vectors for the switching and flexible control records of the same category according to the classification results of the switching and flexible control records; and calculates the difference degree indexes of each dimension of the historical data feature vectors according to the historical data feature vectors; the key feature extraction unit establishes a difference index set according to the difference degree indexes of each dimension of the historical data feature vectors, and identifies the key features related to the switching and flexible control strategy through the analysis of the difference index set and intersection calculation.

8. A distributed photovoltaic management system based on a four-in-one fusion terminal according to claim 5, characterized in that: The real-time data analysis and strategy matching module includes a real-time data analysis unit and a strategy matching unit; The real-time data analysis unit collects the real-time data of the photovoltaic system through the four-in-one fusion terminal in real time, analyzes and processes the real-time data, extracts the real-time data characteristics, and generates real-time data feature vectors; referring to the key features corresponding to the switching and flexible control records of each category, sequentially extracts the real-time key features of the corresponding category from the real-time data feature vectors; The strategy matching unit calculates the similarity between the real-time key features of each category and the key features corresponding to each switching and flexible control record for the key features corresponding to the switching and flexible control records of each category, and takes the average value as the average similarity of the key features of the corresponding category, selects the category with the largest average similarity as the real-time matching category, and uses the switching and flexible control strategy corresponding to the key features of the real-time matching category as the currently matched switching and flexible control strategy.

9. A distributed photovoltaic management system based on a four-in-one fusion terminal according to claim 5, characterized in that: The control operation and feedback evaluation module includes a control operation unit and a feedback evaluation unit; The control operation unit executes corresponding operations according to the real-time matched switching and flexible control strategy and collects the feedback data after control; the feedback evaluation unit analyzes the change of photovoltaic power generation, the change of load and the environmental adjustment coefficient according to the feedback data, calculates the comprehensive evaluation index, comprehensively evaluates the effects of the switching and flexible control, generates adjustment strategies according to the comprehensive evaluation results, and feeds back the adjustment strategies to relevant personnel for corresponding processing by relevant personnel.

Images