Power grid optimization dispatching method and system based on high-proportion photovoltaic

Abstract

The present invention discloses a method and system for optimizing and dispatching a power grid based on high-proportion photovoltaics. The method relates to the technical field of optimizing and dispatching a power grid with high-proportion photovoltaics, and includes the following steps: classifying users into groups based on modularization according to historical electricity consumption data, and obtaining historical load response data of each module; constructing a user load behavior model according to the historical load response data of each module, and extracting electricity consumption behavior characteristics; analyzing the difference data between each module and the overall module, and comparing it with a preset standard threshold, and obtaining a target module with difference data greater than the standard threshold; inputting the historical load response data of the target module into a preset personalized user electricity consumption prediction model, and obtaining load prediction data of personalized groups; using the load prediction data of the personalized group as the ideal output target of photovoltaic system equipment, and combining the system benefit with the ideal output target of photovoltaic system equipment to obtain the operation distribution of photovoltaic system equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of high-proportion photovoltaic power grid optimization dispatching, and in particular to a high-proportion photovoltaic power grid optimization dispatching method and system. Background Art

[0002] With the large-scale access of new energy, the power grid faces new challenges. Optimized dispatching can adapt to the intermittent and volatile nature of new energy, and ensure the stable operation of the power system through strategies such as peak-shifting and valley-filling. Optimized dispatching of power grids manages power production, storage, transmission and consumption, ensuring that the total amount of power production, transmission and use is balanced at any moment, thereby ensuring that the power quality indicators meet the national standards. Through optimized dispatching, while meeting the power supply demand, it is possible to minimize energy waste and environmental pollution and achieve green and low-carbon development.

[0003] With the acceleration of the construction of new power systems, the scale of power grids is expanding, and traditional power grids are transforming into AC / DC hybrid power grids. Optimized dispatching can effectively improve the regulation capacity of power systems, solve the problems caused by the "double high" characteristics, and ensure the safe and stable operation of new power systems.

[0004] The existing wide-area source-load coordinated optimization dispatching system with a high proportion of renewable energy access to the power grid is realized through the timing coupling and randomness of load regulation. In the existing technical research, the load regulation process takes random responses into account and establishes an adjustable capacity model for aggregated civil loads.

[0005] However, the regulation of civil loads is often affected by the laws of daily life, and there is obvious time coupling and randomness, which may make regulation more difficult. The time coupling of loads may not be fully considered, such as the collective behavior patterns of users during peak hours. This behavior may not be fully captured by existing models. The random response regulation model may ignore the differences between individual consumers, resulting in load forecasting errors.

[0006] The collective behavior patterns of users, especially during peak hours, usually have certain regularity and timing. For example, many families will turn on or off household appliances in the same time period. The clustering effect of this behavior may lead to a peak in load. The existing load regulation model may not fully consider the timing coupling of this collective behavior, resulting in large errors in peak load forecasting. In addition, the load response behavior of each family or user is different, affected by many factors such as living habits, family size, and appliance usage. These differences may not be fully considered in the traditional regulation model, resulting in errors in the random response regulation model, and the failure to fully refine the diversity of user groups, resulting in the inability to accurately capture the behavioral differences of personalized groups during load regulation and forecasting.

[0007] In the prior art, there is a problem that load regulation cannot accurately capture the behavioral differences of personalized groups. Summary of the invention

[0008] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide a method and system for optimizing the dispatching of a power grid based on a high proportion of photovoltaics, which refines the prediction method of load regulation to solve the problem in the prior art that load regulation cannot accurately capture the behavioral differences of personalized groups.

[0009] To achieve the above object, the present invention provides the following technical solutions:

[0010] A method for optimizing and dispatching a power grid based on a high proportion of photovoltaic power generation, comprising the following steps: classifying users into groups based on modularization according to historical power consumption data, and obtaining historical load response data of each module; constructing a user load behavior model according to the historical load response data of each module, and extracting power consumption behavior characteristics from the user load behavior model; analyzing the difference data between each module and the overall module, comparing the difference data with a preset standard threshold, and obtaining a target module with difference data greater than the standard threshold; inputting the historical load response data of the target module into a preset personalized user power consumption prediction model, and obtaining load prediction data of a personalized group; using the load prediction data of the personalized group as the ideal output target of photovoltaic system equipment, combining the system benefit with the ideal output target of photovoltaic system equipment, and obtaining the operation distribution of photovoltaic system equipment.

[0011] In a preferred embodiment, the modular classification of users according to historical electricity consumption data includes the following: modular division of the photovoltaic system transmission range area, and division of users in the module area according to residential, commercial, and industrial types; obtaining the daily historical electricity consumption data of users in each module, and dividing the users in the module area into holidays and non-holidays according to the maximum daily historical electricity consumption data within a first preset time interval; obtaining the daily historical electricity consumption data of users in each module within a second preset time interval, and dividing the users in the module area into daytime and nighttime according to the peak electricity consumption within a day; and performing data preprocessing on the historical electricity consumption data after classification to obtain historical load response data of each module.

[0012] In a preferred embodiment, a user load behavior model is constructed based on the historical load response data of each module, specifically: identifying the power load change trend in the historical load response data of each module, and calculating the change amplitude of the power load change trend; comparing the change amplitude of the power load change trend with the standard deviation of the load change trend, and marking the historical load response data of the module whose change amplitude of the power load change trend exceeds the standard deviation of the load change trend; inputting the historical load response data of the marked module into a preset load behavior analysis model, and obtaining the user load behavior model after training.

[0013] In a preferred embodiment, the analysis of the difference data between each module and the overall module is specifically as follows: the power consumption behavior characteristics and each module are numbered separately, and the average value of each power consumption behavior characteristic in all modules within the same timestamp is obtained; based on the deviation between the independent power consumption behavior characteristics of each module and the average value of the corresponding power consumption behavior characteristics, the difference data of the analyzed module is calculated.

[0014] In a preferred embodiment, the load forecast data of the personalized group is used as the ideal output target of the photovoltaic system equipment, and the operation distribution of the photovoltaic system equipment is obtained by combining the system benefit with the ideal output target of the photovoltaic system equipment, specifically: the load forecast data of the personalized group is converted into a line graph, and the peak point and inflection point in the line graph are identified; the actual maximum output power of the photovoltaic system equipment is obtained based on the system benefit, and the peak point is compared with the actual maximum output power of the photovoltaic system equipment; when the peak point exceeds the actual maximum output power of the photovoltaic system equipment, the excess is satisfied through the system reserve equipment transmission plan; when the peak point is lower than the actual maximum output power of the photovoltaic system equipment, the inflection point in the line graph is analyzed, and when the inflection point is on an upward trend, the photovoltaic system extracts and increases power output, and if it is on a downward trend, the photovoltaic system reduces the output power in advance.

[0015] In a preferred embodiment, the system of the power grid optimization dispatching method based on a high proportion of photovoltaics is characterized by including a user classification module, a behavior feature extraction module, a personalized user screening module, a personalized user analysis module, and a photovoltaic system equipment adjustment module, specifically: a user classification module, which is used to classify users into groups based on modularization according to historical electricity consumption data, and obtain historical load response data of each module; a behavior feature extraction module, which is used to build a user load behavior model according to the historical load response data of each module, and extract electricity consumption behavior characteristics from the user load behavior model; a personalized user screening module, which is used to analyze the difference data between each module and the overall module, compare the difference data with a preset standard threshold, and obtain a target module with difference data greater than the standard threshold; a personalized user analysis module, which is used to input the historical load response data of the target module into a preset personalized user electricity consumption prediction model to obtain load prediction data of a personalized group; a photovoltaic system equipment adjustment module, which is used to use the load prediction data of the personalized group as the ideal output target of the photovoltaic system equipment, and combine the system benefit with the ideal output target of the photovoltaic system equipment to obtain the operation distribution of the photovoltaic system equipment.

[0016] The technical effects and advantages of the power grid optimization dispatching method and system based on high-proportion photovoltaic power generation of the present invention are as follows:

[0017] 1. By classifying user groups based on historical electricity consumption data and modularization, combined with personalized load forecasting models, we can accurately predict the electricity demand of personalized groups for different types of users, effectively solving the problem in existing technologies that load regulation cannot accurately capture the behavioral differences of personalized groups.

[0018] 2. By taking the load forecast data of individual groups as the ideal output target of the photovoltaic system equipment and optimizing the output power of the photovoltaic system in combination with the system benefits, it is possible to reasonably adjust the power output of photovoltaic power generation and energy storage equipment, thereby achieving effective utilization of the power generation capacity of the photovoltaic system and reducing unnecessary losses of energy storage equipment.

[0019] 3. By analyzing the difference data between each module and the overall module, we can more accurately screen out groups of users with special electricity consumption, thereby reducing the blindness of power grid dispatching and making power dispatching more accurate and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flowchart of a power grid optimization dispatching method based on a high proportion of photovoltaic power provided in an embodiment of the present application;

[0021] Figure 2A schematic diagram of the system structure of a high-proportion photovoltaic power grid optimization scheduling method provided in an embodiment of the present application; DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0024] Embodiment 1, Figure 1 The present invention provides a method for optimizing the dispatching of a power grid based on high-proportion photovoltaic power generation, comprising the following steps:

[0025] S1, classify users into groups based on modularization according to historical electricity consumption data, and obtain historical load response data of each module;

[0026] S2, constructing a user load behavior model based on the historical load response data of each module, and extracting power consumption behavior characteristics from the user load behavior model;

[0027] S3, analyzing the difference data between each module and the whole module, comparing the difference data with a preset standard threshold, and obtaining a target module with a difference data greater than the standard threshold;

[0028] S4, inputting the historical load response data of the target module into a preset personalized user power consumption prediction model to obtain personalized group load prediction data;

[0029] S5, taking the load forecast data of the personalized group as the ideal output target of the photovoltaic system equipment, combining the system benefit with the ideal output target of the photovoltaic system equipment, and obtaining the operation distribution of the photovoltaic system equipment.

[0030] Load regulation refers to the dynamic adjustment process of energy, resources or load in the system. In the power system, load regulation usually refers to the balancing and regulation of power demand to ensure that the system can maintain stable operation when the load changes.

[0031] In reality, load changes are often not completely predictable. System load is affected by many factors, such as weather changes, fluctuations in user demand, equipment failures, etc., which are inherently random. Therefore, many parameters in the load regulation process (such as the magnitude and timing of changes in power demand) are not completely predictable, making it difficult for the system to accurately predict future load changes at certain times.

[0032] The user group classification based on modularization according to historical electricity consumption data includes the following:

[0033] The transmission range of the photovoltaic system is divided into modules, and the users in the module area are divided into residential, commercial and industrial types;

[0034] Acquire the daily historical electricity consumption data of users in each module, and divide the users in the area of ​​the module into holidays and non-holidays according to the maximum daily historical electricity consumption data within the first preset time interval;

[0035] According to the daily historical electricity consumption data of users in each module obtained within the second preset time interval, the users in the area of ​​the module are divided into daytime and nighttime according to the peak electricity consumption in a day;

[0036] The classified historical electricity consumption data is preprocessed to obtain the historical load response data of each module.

[0037] It should be noted that the user's location information is obtained, and if a commercial building may also have residential functions, it is classified according to the business type; the historical electricity consumption data is distinguished by holidays and non-holidays, and the holidays include Saturdays, Sundays and national statutory holidays; daytime refers to the electricity consumption data generated between 8 am and 8 pm, and night refers to the electricity consumption data generated between 8 pm and 8 am.

[0038] The purpose of preprocessing the classified historical power consumption data is to clean and organize the data so that the results obtained in the subsequent analysis of the historical load response data of each module are more accurate. The data preprocessing includes the following steps:

[0039] Missing value processing: select historical electricity consumption data with the same classification as the missing value data to fill in the missing value data;

[0040] Abnormal value detection and processing: Identify and process load values ​​and abnormal power consumption data that deviate significantly from the normal range, as well as incorrect power metering, equipment failures, etc.

[0041] Data normalization: Convert data to a uniform scale to facilitate subsequent analysis.

[0042] The historical load response data of each module refers to the way in which users adjust their electricity consumption under specific conditions in response to electricity demand, price fluctuations or other external factors. The historical load response data of each module includes timestamp, electricity consumption data, and group classification results. The electricity consumption data includes daily electricity consumption data, weekly electricity consumption data, monthly electricity consumption data, and quarterly electricity consumption data.

[0043] According to the historical load response data of each module, a user load behavior model is constructed, specifically:

[0044] Identify the power load change trend in the historical load response data of each module, and calculate the change amplitude of the power load change trend;

[0045] Compare the change amplitude of the power load change trend with the standard deviation of the load change trend, and mark the historical load response data of the module whose change amplitude of the power load change trend exceeds the standard deviation of the load change trend;

[0046] The historical load response data of the marked modules are input into the preset load behavior analysis model, and the user load behavior model is obtained after training.

[0047] It should be noted that the power load change trend includes the power consumption trend of peak periods and off-peak periods in the historical load response data of each module. The peak periods and off-peak periods are measured based on the power consumption in the period, and the average value of all power consumption in the same period within the third preset time interval is obtained. The variance of the average value of all power consumption in the measurement period and the same period within the third preset time interval is calculated. If the variance is greater than 1, the measurement period is a peak period. If the variance is less than 1, the measurement period is a off-peak period.

[0048] The deep learning-based model extracts the power consumption behavior features from the user load behavior model, specifically:

[0049] Decompose the time series of the user load behavior prediction model to obtain the trend, seasonality and residual components of the decomposed time series;

[0050] The power consumption behavior characteristics are extracted from the trend, seasonality and residual components of the time series, and the power consumption behavior characteristics include periodic fluctuation rate, sudden load peak, load increase rate and load decrease rate.

[0051] Identify the power load change trend in the historical load response data of each module, and calculate the change amplitude of the power load change trend, specifically:

[0052]

[0053] Where: is the change range of the power load change trend, is the time period value corresponding to the peak period, is the value corresponding to the low period, is the timestamp interval of the power load change trend, is the load in period t, is the load in period t+1, and m is the total number of periods in the changing trend

[0054] The electricity consumption behavior characteristics include periodic fluctuation rate, sudden load peak, load increase rate and load decrease rate, specifically:

[0055] The periodic volatility is calculated as follows:

[0056]

[0057] Where: is the periodic fluctuation rate; i is the total number of time points in the period, that is, the total number of load data; is the average value of the load within the cycle; Indicates the number of modules involved in the calculation; The load value at the i-th moment in the cycle.

[0058] The sudden load peak is calculated as follows:

[0059]

[0060] Where: G is the peak value of sudden load, is the load in period t, is the load in period t-1.

[0061] The load increase rate is calculated as follows:

[0062]

[0063] The load drop rate is calculated as follows:

[0064]

[0065] Where: s is the rate of increase of the load, J is the rate of decrease of the load, is the load in period t, is the load in period t-1, It is the timestamp interval of the power load change trend.

[0066] It should be noted that the load increase rate indicates the speed at which the load increases in unit time, reflecting the growth trend of the power load, and the load decrease rate indicates the speed at which the load decreases in unit time, reflecting the reduction trend of the power load.

[0067] The analysis of the difference data between each module and the overall module is specifically as follows:

[0068] Number the power consumption behavior features and each module respectively, and obtain the average value of each power consumption behavior feature in all modules within the same timestamp;

[0069] According to the deviation between the independent power consumption behavior characteristics of each module and the average value of the corresponding power consumption behavior characteristics, the difference data of the analyzed module is calculated.

[0070] The difference data is calculated by the following formula:

[0071]

[0072] Where: is the difference data of electricity consumption behavior characteristics, z represents independent electricity consumption behavior characteristics, Indicates the number of modules involved in the calculation, is the characteristic value of the zth power consumption behavior feature of the ith module, It represents the average value of the zth electricity consumption behavior feature obtained in all modules.

[0073] It should be noted that a deep learning-based model method is used to evaluate the load change trend in different time periods, and this application uses slope changes to represent amplitude changes.

[0074] It should be noted that the standard deviation indicates the degree of dispersion of the user's electricity usage behavior during the time period. If the standard deviation is large, it means that the users' electricity usage behaviors vary greatly; otherwise, the differences are small.

[0075] Based on the personalized load regulation model, the power consumption of different groups is predicted to obtain user load prediction data, specifically:

[0076] Use the trained personalized load regulation model to predict the future electricity consumption of different groups;

[0077] Combine user behavior characteristics to independently predict the electricity demand of different groups and adjust the electricity load in a personalized manner;

[0078] The model outputs the predicted value of electricity demand for each group in a certain period of time in the future and obtains user load forecast data.

[0079] The user load forecast data is based on the load characteristics of historical data (including time, date, and climate factors). After algorithm training, power demand forecasts for different time periods (hours, days, and weeks) are generated.

[0080] The user load forecast data is used as the output target of the photovoltaic system equipment to obtain the power value that the photovoltaic system needs to output, specifically:

[0081] Taking the predicted user load demand data as the target, combined with the photovoltaic system power demand calculation formula, the output power of the photovoltaic system is adjusted according to factors such as the photovoltaic system's power generation capacity and solar radiation.

[0082] The power value that the photovoltaic system needs to output is calculated as follows:

[0083]

[0084] Where: P is the output power of the photovoltaic system, G is the solar radiation intensity in the photovoltaic system, A is the effective area of ​​the photovoltaic panel, is the conversion efficiency of the photovoltaic system.

[0085] The load forecast data of the personalized group is used as the ideal output target of the photovoltaic system equipment, and the operation distribution of the photovoltaic system equipment is obtained by combining the system benefit with the ideal output target of the photovoltaic system equipment, which is specifically:

[0086] Converting the load forecast data of the personalized group into a line graph, and identifying peak points and inflection points in the line graph;

[0087] Obtaining the actual maximum output power of the photovoltaic system equipment based on the system benefit, and comparing the peak point with the actual maximum output power of the photovoltaic system equipment;

[0088] When the peak point exceeds the actual maximum output power of the photovoltaic system equipment, the excess power is met through the system reserve equipment transmission plan;

[0089] When the peak point is lower than the actual maximum output power of the photovoltaic system equipment, the inflection point in the line graph is analyzed. When the inflection point shows an upward trend, the photovoltaic system extracts increased power output. If it shows a downward trend, the photovoltaic system reduces the output power in advance.

[0090] The system of the power grid optimization dispatching method based on high-proportion photovoltaics is characterized by comprising a user classification module, a behavior feature extraction module, a personalized user screening module, a personalized user analysis module, and a photovoltaic system equipment adjustment module, specifically:

[0091] A user classification module is used to classify users into groups based on modularization according to historical power consumption data and obtain historical load response data of each module;

[0092] A behavior feature extraction module, used to construct a user load behavior model based on the historical load response data of each module, and extract power consumption behavior features from the user load behavior model;

[0093] A personalized user screening module is used to analyze the difference data between each module and the overall module, compare the difference data with a preset standard threshold, and obtain a target module with a difference data greater than the standard threshold;

[0094] A personalized user analysis module, used to input the historical load response data of the target module into a preset personalized user power consumption prediction model to obtain load prediction data of a personalized group;

[0095] The photovoltaic system equipment adjustment module is used to use the load forecast data of the personalized group as the ideal output target of the photovoltaic system equipment, combine the system benefit with the ideal output target of the photovoltaic system equipment, and obtain the operation distribution of the photovoltaic system equipment.

[0096] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0097] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0098] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0099] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0100] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0101] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A grid optimization dispatching method based on high-proportion photovoltaic power generation, characterized in that: The following steps are involved: According to the historical electricity consumption data, users are grouped based on modularization, and the historical load response data of each module is obtained; Building a user load behavior model based on the historical load response data of each module, and extracting power consumption behavior features from the user load behavior model; Number the power consumption behavior features and each module respectively, and obtain the average value of each power consumption behavior feature in all modules within the same timestamp; Calculate the difference data between each module and the overall module based on the deviation between the independent power consumption behavior characteristics of each module and the average value of the corresponding power consumption behavior characteristics; Compare the difference data with a preset standard threshold value to obtain a target module with difference data greater than the standard threshold value; Input the historical load response data of the target module into a preset personalized user power consumption prediction model to obtain load prediction data for a personalized group; The load forecast data of the personalized group is used as the ideal output target of the photovoltaic system equipment, and the operation distribution of the photovoltaic system equipment is obtained by combining the system benefit with the ideal output target of the photovoltaic system equipment.

2. The method for optimizing the dispatching of a power grid based on high-proportion photovoltaic power generation according to claim 1, characterized in that: The user group classification based on modularization according to historical electricity consumption data includes the following: The transmission range of the photovoltaic system is divided into modules, and the users in the module area are divided into residential, commercial and industrial types; Acquire the daily historical electricity consumption data of users in each module, and divide the users in the area of ​​the module into holidays and non-holidays according to the maximum daily historical electricity consumption data within the first preset time interval; According to the daily historical electricity consumption data of users in each module obtained within the second preset time interval, the users in the area of ​​the module are divided into daytime and nighttime according to the peak electricity consumption in a day; The classified historical electricity consumption data is preprocessed to obtain the historical load response data of each module.

3. The method for optimizing the dispatching of a power grid based on high-proportion photovoltaic power generation as claimed in claim 2, characterized in that: According to the historical load response data of each module, a user load behavior model is constructed, specifically: Identify the power load change trend in the historical load response data of each module, and calculate the change amplitude of the power load change trend; Compare the change amplitude of the power load change trend with the standard deviation of the load change trend, and mark the historical load response data of the module whose change amplitude of the power load change trend exceeds the standard deviation of the load change trend; The historical load response data of the marked modules are input into the preset load behavior analysis model, and the user load behavior model is obtained after training.

4. The method for optimizing the dispatching of a power grid based on high-proportion photovoltaic power generation as claimed in claim 3, characterized in that: The load forecast data of the personalized group is used as the ideal output target of the photovoltaic system equipment, and the operation distribution of the photovoltaic system equipment is obtained by combining the system benefit with the ideal output target of the photovoltaic system equipment, which is specifically: Converting the load forecast data of the personalized group into a line graph, and identifying peak points and inflection points in the line graph; Obtaining the actual maximum output power of the photovoltaic system equipment based on the system benefit, and comparing the peak point with the actual maximum output power of the photovoltaic system equipment; When the peak point exceeds the actual maximum output power of the photovoltaic system equipment, the excess power is met through the system reserve equipment transmission plan; When the peak point is lower than the actual maximum output power of the photovoltaic system equipment, the inflection point in the line graph is analyzed. When the inflection point shows an upward trend, the photovoltaic system extracts increased power output. If it shows a downward trend, the photovoltaic system reduces the output power in advance.

5. The method for optimizing the dispatching of a power grid based on high-proportion photovoltaic power generation as claimed in claim 4, characterized in that: The difference data is calculated by the following formula: Where: is the difference data of electricity consumption behavior characteristics, z represents independent electricity consumption behavior characteristics, Indicates the number of modules involved in the calculation, is the characteristic value of the zth power consumption behavior feature of the ith module, It represents the average value of the zth electricity consumption behavior feature obtained in all modules.

6. The method for optimizing the dispatching of a power grid based on high-proportion photovoltaic power generation as claimed in claim 5, characterized in that: The identification of the power load change trend in the historical load response data of each module and the calculation of the change amplitude of the power load change trend are specifically as follows: Where: is the change range of the power load change trend, is the time period value corresponding to the peak period, is the value corresponding to the low period, is the timestamp interval of the power load change trend, is the load in period t, is the load in period t+1, and m is the total number of periods in the changing trend.

7. The method for optimizing the dispatching of a power grid based on high-proportion photovoltaic power generation as claimed in claim 6, characterized in that: The electricity consumption behavior characteristics include periodic fluctuation rate, sudden load peak, load increase rate and load decrease rate, specifically: The periodic volatility is calculated as follows: Where: is the periodic fluctuation rate; i is the total number of time points in the period, that is, the total number of load data; is the average value of the load within the cycle; Indicates the number of modules involved in the calculation; The load value at the i-th moment in the cycle; The sudden load peak is calculated as follows: Where: is the load in period t, is the load in period t-1; The load increase rate is calculated as follows: The load drop rate is calculated as follows: Where: is the load in period t, is the load in period t-1, It is the timestamp interval of the power load change trend.

8. A system using the power grid optimization dispatching method based on high-proportion photovoltaics as described in any one of claims 1 to 7, characterized in that: It includes user classification module, behavior feature extraction module, personalized user screening module, personalized user analysis module, and photovoltaic system equipment adjustment module, specifically: A user classification module is used to classify users into groups based on modularization according to historical power consumption data and obtain historical load response data of each module; A behavior feature extraction module, used to construct a user load behavior model based on the historical load response data of each module, and extract power consumption behavior features from the user load behavior model; A personalized user screening module is used to analyze the difference data between each module and the overall module, compare the difference data with a preset standard threshold, and obtain a target module with a difference data greater than the standard threshold; A personalized user analysis module, used to input the historical load response data of the target module into a preset personalized user power consumption prediction model to obtain load prediction data of a personalized group; The photovoltaic system equipment adjustment module is used to use the load forecast data of the personalized group as the ideal output target of the photovoltaic system equipment, combine the system benefit with the ideal output target of the photovoltaic system equipment, and obtain the operation distribution of the photovoltaic system equipment.

Images