A photovoltaic power station power supply scheduling method based on data analysis

Abstract

The present invention relates to the technical field of power supply scheduling, and specifically to a photovoltaic power station power supply scheduling method based on data analysis. The present invention obtains power supply data of the power station and analyzes the power supply data to obtain scheduling prediction information, and analyzes power generation-related data to obtain power generation status information. Then, the scheduling prediction information and power generation status information are comprehensively analyzed to obtain an initial scheduling strategy. According to the initial scheduling strategy, power supply scheduling of the photovoltaic power station can be realized, facilitating power supply scheduling. The present invention obtains scheduling optimization information by analyzing response action information, and then optimizes the initial scheduling strategy according to the scheduling optimization information to obtain power supply scheduling instructions, and monitors the response status of each device so that the power supply scheduling can be adjusted according to the real-time response status to avoid failures caused by abnormal responses.

Description

Technical Field

[0001] The present invention relates to the technical field of power supply scheduling, and in particular to a photovoltaic power station power supply scheduling method based on data analysis. Background Art

[0002] Photovoltaic power stations, also known as photovoltaic power stations, are facilities that use solar cell panels to convert light energy into electrical energy. These power stations usually contain a large number of photovoltaic panels, inverters, distribution equipment and possible energy storage systems. Photovoltaic power stations can be distributed or centralized. The former are usually installed at the user end, such as on rooftops or open spaces, for self-sufficiency or to supply power to the grid; the latter are usually built in open areas, such as deserts or Gobi, for large-scale power generation and input into the grid.

[0003] As the proportion of renewable energy in the energy structure continues to increase, the power supply scheduling of photovoltaic power stations helps to improve their electricity absorption rate. Through effective coordination with the power grid and flexible scheduling strategies, more photovoltaic power generation can be absorbed and utilized by the power grid, reducing the phenomenon of abandoned light and promoting the sustainable development of renewable energy.

[0004] Currently, when photovoltaic power stations conduct power supply scheduling, each device module is required to respond according to the scheduling information. However, most of the device modules are independently controlled. Therefore, there will be a delay in the scheduling response, making it impossible for the photovoltaic output power to keep up with the load changes in time. The frequency fluctuation may exceed the allowable range, affecting the normal operation of sensitive equipment in the power grid, such as precision instruments and electronic equipment. Summary of the Invention

[0005] The present invention provides a photovoltaic power station power supply scheduling method based on data analysis, which is used to solve the above technical problems.

[0006] A first aspect of the present invention provides a photovoltaic power station power supply scheduling method based on data analysis, comprising:

[0007] Step 1: Obtain the operating data of each device module and the power supply data of the photovoltaic power station; use sensors to collect the corresponding operating data of each device module of the photovoltaic power station. The operating data includes power generation-related data and response action information. Among them, power generation-related data includes photovoltaic module data, inverter data and energy storage system data. Response action information includes instruction execution time, response delay and online status; power supply data includes power generation data, load data, meteorological data and electricity price data.

[0008] Step 2: Obtain the power supply data of the power station and analyze the power supply data to obtain scheduling forecast information.

[0009] As a further improvement of the present invention, the power supply data is analyzed, and the specific analysis method is as follows:

[0010] The power supply data includes power generation data, load data, meteorological data and electricity price data; the power generation data includes power generation power and photovoltaic performance data, and a pre-set photovoltaic health performance range is obtained, and the photovoltaic performance data is matched with the photovoltaic health performance range. When the photovoltaic performance data is in the photovoltaic health performance range, photovoltaic health information is generated; conversely, when the photovoltaic performance data exceeds the photovoltaic health performance range, photovoltaic abnormality information is generated; based on the photovoltaic health information, the corresponding power generation power is obtained, and multiple historical power generation powers of the corresponding date are extracted from the database. The multiple historical power generation powers are calculated to obtain the average value, and the average value is marked as the normal reference power. The power generation power corresponding to the current date is calculated to obtain the power difference with the normal reference power, and the corresponding power adjustment information is obtained according to the power difference.

[0011] Based on the database, load data of different time periods are obtained. The time periods include weekdays and weekends. When the corresponding time period is a weekday, the peak power consumption corresponds to industrial and commercial activities, and the valley power consumption corresponds to residential electricity consumption. Conversely, when the corresponding time period is a weekend, the peak power consumption corresponds to residential electricity consumption and commercial activities, and the valley power consumption corresponds to industrial activities. The power consumption objects corresponding to each time period are marked to obtain the power supply object information corresponding to each time period.

[0012] Meteorological data is identified to obtain light intensity data and duration of illumination. Based on the corresponding daily light intensity data and duration of illumination, power generation forecast information can be obtained, and the power supply efficiency and energy storage capacity of the photovoltaic power station can be obtained. The power supply efficiency and energy storage capacity are comprehensively judged with the power generation forecast information to obtain the corresponding energy supply and storage information; for example, when the power generation forecast information exceeds the power supply efficiency and energy storage capacity, the corresponding electric energy will be mainly supplied to the power grid and the power generation power will be reduced.

[0013] The electricity price data is identified to obtain the time-of-use electricity price, which includes peak, flat and valley electricity prices. The duration of each time-of-use electricity price is obtained, and the corresponding change in electricity consumption is obtained. The change in electricity consumption is compared with the preset change threshold. When the corresponding change threshold is exceeded, the corresponding duration adjustment is made to obtain the corresponding electricity price-related adjustment. For example, in summer, due to the high demand for electricity for air conditioning, the peak electricity price period may be extended or the electricity price may increase. This can help consider supplying more electricity when the electricity price is high during scheduling, thereby improving economic benefits.

[0014] The power regulation information, power supply object information, energy storage information and electricity price related adjustments are combined to obtain scheduling forecast information.

[0015] Step 3: Identify the operating data to obtain power generation related data and analyze the power generation related data to obtain power generation status information.

[0016] As a further improvement of the present invention, the power generation related data is analyzed, and the specific analysis method is as follows:

[0017] A1: PV module data, inverter data, and energy storage system data are obtained by identifying power generation-related data;

[0018] A2: Obtain the solar power generation power and module temperature based on the PV module data; divide the solar power generation power into multiple power generation intervals, each power generation interval corresponding to a power generation capacity value; match the solar power generation power corresponding to each PV panel with the multiple power generation intervals to obtain the corresponding power generation capacity value; calculate the power generation capacity value corresponding to each PV panel to obtain the total power generation capacity value; obtain a pre-set module temperature threshold, compare the module temperature corresponding to each PV panel with the module temperature threshold; when the module temperature is greater than the module temperature threshold, calculate the difference between the module temperature and the module temperature threshold to obtain a temperature anomaly value; add up the temperature anomalies corresponding to each PV panel to obtain the total temperature anomaly value.

[0019] A3: Obtain the inverter output power and input power based on the inverter data, calculate the difference between the output power and input power to obtain the power loss value, calculate the conversion efficiency based on the power loss value, and obtain a preset conversion reference efficiency. When the conversion efficiency is less than the conversion reference efficiency, calculate the difference between the conversion efficiency corresponding to the current inverter and the conversion reference efficiency to obtain a conversion abnormality value.

[0020] A4: Obtain energy storage margin and energy storage efficiency based on energy storage system data. Obtain a pre-set minimum margin limit and compare the energy storage margin of each energy storage module with the minimum margin limit. When the energy storage margin is less than the minimum margin limit, mark the corresponding energy storage module as a full storage module. Count the number of full storage modules and record it as the number of full storage modules. Calculate the ratio of the number of full storage modules to the total number of energy storage modules to obtain the full storage percentage.

[0021] A5: Construct two circles with the total temperature anomaly value and the conversion anomaly value as the radius of each circle. Draw a straight line perpendicular to the two circles with the centers of the two circles as the starting point and the end point. The length of the line is equal to the full storage ratio. Then construct a frustum with the two circles and the straight line. Construct a sphere with the midpoint of the straight line as the center and the total power generation capacity as the radius. Cut off the part of the frustum corresponding to the sphere and calculate the volume of the remaining part of the frustum. Mark the volume value as the power generation impact value.

[0022] A6: The power generation impact value is compared with a preset power generation impact threshold. When the power generation impact value is greater than the power generation impact threshold, corresponding power generation status information is generated as abnormal power generation status.

[0023] Step 4: Comprehensively analyze the scheduling forecast information and power generation status information to obtain the initial scheduling strategy, specifically: identify the power generation status information, and when the power generation status information indicates that the power generation status is abnormal, generate the corresponding initial scheduling strategy for abnormality investigation; conversely, when the power generation status information corresponds to a normal power generation status, generate the corresponding initial scheduling strategy for maintaining the original scheduling forecast information.

[0024] Step 5: Identify the operation data to obtain response action information and analyze the response action information to obtain scheduling optimization information.

[0025] As a further improvement of the present invention, the response action information is analyzed, and the specific analysis method is as follows:

[0026] S1: By identifying the response action information, the command execution time, response delay and online status of each device module of the photovoltaic power station are obtained;

[0027] S2: Based on the historical execution time of multiple response actions corresponding to each device module extracted from the database, the multiple historical execution times are calculated to obtain the average value, and the average value is marked as the execution reference time. The execution time corresponding to each current device module is obtained. When the execution time is greater than the execution reference time, the difference between the execution time and the corresponding execution reference time is calculated to obtain the execution timeout. The execution timeouts corresponding to each device module are added together to obtain the total execution shadow value, which is recorded as zxy.

[0028] S3: Obtain the response delays corresponding to multiple response actions of the device module, and obtain a pre-set response delay threshold. Compare each response delay with the response delay threshold. When the response delay is greater than the response delay threshold, mark the corresponding device module as a response abnormal module. Count the number of response abnormal modules and record it as the number of abnormal response modules, which is also recorded as ycx.

[0029] S4: Obtain a preset detection time period, identify the online status of each device module corresponding to the detection time period, and when the online status corresponds to offline, the reasons for offline include but are not limited to communication failure, device failure, environmental impact, and human operation. Count the offline duration corresponding to each device module, record the offline duration as offline, and calculate the ratio of the offline duration to the duration corresponding to the detection time period to obtain the offline impact value, which is recorded as dxy.

[0030] S5: Normalize the total execution shadow value, the number of abnormal response modules, and the offline impact value and take their values. Use the formula Calculate the response status value XY; where ycx' represents the number of abnormal response modules allowed; the value of n is a positive integer. It is represented as the sum of the offline impact values ​​corresponding to n detection time periods. p1, p2, and p3 are preset weight factors, with values ​​of 2.726, 1.145, and 2.367, respectively.

[0031] S6: When the response state value is greater than a preset response state threshold, a corresponding response state mark is generated as abnormal, and corresponding scheduling optimization information is generated as an instruction adjustment delay.

[0032] Step 6: Optimize the initial scheduling strategy according to the scheduling optimization information to obtain the power supply scheduling instruction; identify the scheduling optimization information, and when the response state corresponding to the scheduling optimization information is abnormal, perform scheduling optimization according to the abnormality, and use the optimized scheduling instruction as the power supply scheduling instruction.

[0033] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0034] 1. The present invention obtains the power supply data of the power station and analyzes the power supply data to obtain scheduling prediction information, and obtains power generation related data by identifying the operation data and analyzes the power generation related data to obtain power generation status information, and then obtains the initial scheduling strategy by comprehensively analyzing the scheduling prediction information and the power generation status information. According to the initial scheduling strategy, the power supply scheduling of the photovoltaic power station can be realized, which facilitates power supply scheduling.

[0035] 2. The present invention obtains response action information by identifying the operating data and analyzes the response action information to obtain scheduling optimization information. Then, the initial scheduling strategy is optimized according to the scheduling optimization information to obtain power supply scheduling instructions, and the response status of each device is monitored so that the power supply scheduling can be adjusted according to the real-time response status to avoid failures caused by abnormal responses. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for the description of the embodiments. The following drawings are not intentionally scaled to the actual size, and the focus is on illustrating the main purpose of the present application.

[0037] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0038] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making any creative efforts shall fall within the scope of protection of the present invention.

[0039] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 In an embodiment of the present invention, a method for power supply scheduling of a photovoltaic power station based on data analysis includes the following steps:

[0040] Step 1: Obtain the operating data of each device module and the power supply data of the photovoltaic power station; use sensors to collect the corresponding operating data of each device module of the photovoltaic power station. The operating data includes power generation-related data and response action information. Among them, power generation-related data includes photovoltaic module data, inverter data and energy storage system data. Response action information includes instruction execution time, response delay and online status; power supply data includes power generation data, load data, meteorological data and electricity price data.

[0041] Step 2: Obtain the power supply data of the power station and analyze the power supply data to obtain scheduling forecast information; the power supply data is analyzed in the following specific analysis methods:

[0042] The power supply data includes power generation data, load data, meteorological data and electricity price data; the power generation data includes power generation power and photovoltaic performance data, and a pre-set photovoltaic health performance range is obtained, and the photovoltaic performance data is matched with the photovoltaic health performance range. When the photovoltaic performance data is in the photovoltaic health performance range, photovoltaic health information is generated; conversely, when the photovoltaic performance data exceeds the photovoltaic health performance range, photovoltaic abnormality information is generated; based on the photovoltaic health information, the corresponding power generation power is obtained, and multiple historical power generation powers of the corresponding date are extracted from the database. The multiple historical power generation powers are calculated to obtain the average value, and the average value is marked as the normal reference power. The power generation power corresponding to the current date is calculated to obtain the power difference with the normal reference power, and the corresponding power adjustment information is obtained according to the power difference.

[0043] Based on the database, load data of different time periods are obtained. The time periods include weekdays and weekends. When the corresponding time period is a weekday, the peak power consumption corresponds to industrial and commercial activities, and the valley power consumption corresponds to residential electricity consumption. Conversely, when the corresponding time period is a weekend, the peak power consumption corresponds to residential electricity consumption and commercial activities, and the valley power consumption corresponds to industrial activities. The power consumption objects corresponding to each time period are marked to obtain the power supply object information corresponding to each time period.

[0044] Meteorological data is identified to obtain light intensity data and duration of illumination. Based on the corresponding daily light intensity data and duration of illumination, power generation forecast information can be obtained, and the power supply efficiency and energy storage capacity of the photovoltaic power station can be obtained. The power supply efficiency and energy storage capacity are comprehensively judged with the power generation forecast information to obtain the corresponding energy supply and storage information; for example, when the power generation forecast information exceeds the power supply efficiency and energy storage capacity, the corresponding electric energy will be mainly supplied to the power grid and the power generation power will be reduced.

[0045] The electricity price data is identified to obtain the time-of-use electricity price, which includes peak, flat and valley electricity prices. The duration of each time-of-use electricity price is obtained, and the corresponding change in electricity consumption is obtained. The change in electricity consumption is compared with the preset change threshold. When the corresponding change threshold is exceeded, the corresponding duration adjustment is made to obtain the corresponding electricity price-related adjustment. For example, in summer, due to the high demand for electricity for air conditioning, the peak electricity price period may be extended or the electricity price may increase. This can help consider supplying more electricity when the electricity price is high during scheduling, thereby improving economic benefits.

[0046] The power regulation information, power supply object information, energy storage information and electricity price related adjustments are combined to obtain scheduling forecast information.

[0047] Step 3: Identify the operation data to obtain power generation related data and analyze the power generation related data to obtain power generation status information; analyze the power generation related data in the following specific analysis methods:

[0048] A1: PV module data, inverter data, and energy storage system data are obtained by identifying power generation-related data;

[0049] A2: Obtain the solar power generation power and module temperature based on the PV module data; divide the solar power generation power into multiple power generation intervals, each power generation interval corresponding to a power generation capacity value; match the solar power generation power corresponding to each PV panel with the multiple power generation intervals to obtain the corresponding power generation capacity value; calculate the power generation capacity value corresponding to each PV panel to obtain the total power generation capacity value; obtain a pre-set module temperature threshold, compare the module temperature corresponding to each PV panel with the module temperature threshold; when the module temperature is greater than the module temperature threshold, calculate the difference between the module temperature and the module temperature threshold to obtain a temperature anomaly value; add up the temperature anomalies corresponding to each PV panel to obtain the total temperature anomaly value.

[0050] A3: Obtain the inverter output power and input power based on the inverter data, calculate the difference between the output power and input power to obtain the power loss value, calculate the conversion efficiency based on the power loss value, and obtain a preset conversion reference efficiency. When the conversion efficiency is less than the conversion reference efficiency, calculate the difference between the conversion efficiency corresponding to the current inverter and the conversion reference efficiency to obtain a conversion abnormality value.

[0051] A4: Obtain energy storage margin and energy storage efficiency based on energy storage system data. Obtain a pre-set minimum margin limit and compare the energy storage margin of each energy storage module with the minimum margin limit. When the energy storage margin is less than the minimum margin limit, mark the corresponding energy storage module as a full storage module. Count the number of full storage modules and record it as the number of full storage modules. Calculate the ratio of the number of full storage modules to the total number of energy storage modules to obtain the full storage percentage.

[0052] A5: Construct two circles with the total temperature anomaly value and the conversion anomaly value as the radius of each circle. Draw a straight line perpendicular to the two circles with the centers of the two circles as the starting point and the end point. The length of the line is equal to the full storage ratio. Then construct a frustum with the two circles and the straight line. Construct a sphere with the midpoint of the straight line as the center and the total power generation capacity as the radius. Cut off the part of the frustum corresponding to the sphere and calculate the volume of the remaining part of the frustum. Mark the volume value as the power generation impact value.

[0053] A6: The power generation impact value is compared with a preset power generation impact threshold. When the power generation impact value is greater than the power generation impact threshold, corresponding power generation status information is generated as abnormal power generation status.

[0054] Step 4: Comprehensively analyze the scheduling forecast information and power generation status information to obtain the initial scheduling strategy, specifically: identify the power generation status information, and when the power generation status information indicates that the power generation status is abnormal, generate the corresponding initial scheduling strategy for abnormality investigation; conversely, when the power generation status information corresponds to a normal power generation status, generate the corresponding initial scheduling strategy for maintaining the original scheduling forecast information.

[0055] Step 5: Identify the operation data to obtain response action information and analyze the response action information to obtain scheduling optimization information; analyze the response action information, and the specific analysis method is as follows:

[0056] S1: By identifying the response action information, the command execution time, response delay and online status of each device module of the photovoltaic power station are obtained;

[0057] S2: Based on the historical execution time of multiple response actions corresponding to each device module extracted from the database, the multiple historical execution times are calculated to obtain the average value, and the average value is marked as the execution reference time, and the execution time corresponding to each current device module is obtained. When the execution time is greater than the execution reference time, the execution time and the corresponding execution reference time are calculated as the difference to obtain the execution timeout, and the execution timeouts corresponding to each device module are added together to obtain the total execution value.

[0058] S3: Obtain the response delays corresponding to multiple response actions of the device module, and obtain a pre-set response delay threshold, compare each response delay with the response delay threshold, and when the response delay is greater than the response delay threshold, mark the corresponding device module as a response abnormal module, count the number of response abnormal modules, and record it as the number of abnormal response modules.

[0059] S4: Obtain a preset detection time period, identify the online status of each device module corresponding to the detection time period, when the online status corresponds to offline, the reasons for offline include but are not limited to communication failure, equipment failure, environmental impact, and human operation, count the offline duration corresponding to each device module, record the offline duration as offline, and calculate the ratio of the offline duration to the duration corresponding to the detection time period to obtain the offline impact value.

[0060] S5: Normalize the total execution shadow value, the number of abnormal response modules, and the offline impact value and take their values. Use the formula Calculate the response status value XY; where zxy, ycx, and dxy represent the execution total impact value, the number of abnormal response modules, and the offline impact value, respectively; ycx' represents the allowed number of abnormal response modules; n is a positive integer. It is represented as the sum of the offline impact values ​​corresponding to n detection time periods. p1, p2, and p3 are preset weight factors, with values ​​of 2.726, 1.145, and 2.367, respectively.

[0061] S6: When the response state value is greater than a preset response state threshold, a corresponding response state mark is generated as abnormal, and corresponding scheduling optimization information is generated as an instruction adjustment delay.

[0062] Step 6: Optimize the initial scheduling strategy according to the scheduling optimization information to obtain the power supply scheduling instruction; identify the scheduling optimization information, and when the response state corresponding to the scheduling optimization information is abnormal, perform scheduling optimization according to the abnormality, and use the optimized scheduling instruction as the power supply scheduling instruction.

[0063] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A photovoltaic power station power supply scheduling method based on data analysis, characterized in that: The steps include: Step 1: Obtain the operating data of each device module and the power supply data of the photovoltaic power station; Step 2: Obtain the power supply data of the power station and analyze the power supply data to obtain scheduling prediction information; The power supply data is analyzed in the following specific manner: Power supply data includes power generation data, load data, meteorological data and electricity price data; power generation data includes power generation power and photovoltaic performance data, obtains a pre-set photovoltaic health performance range, matches the photovoltaic performance data with the photovoltaic health performance range, and generates photovoltaic health information when the photovoltaic performance data is in the photovoltaic health performance range; obtains the corresponding power generation power based on the photovoltaic health information, extracts multiple historical power generation powers of the corresponding date from the database, calculates the average value of the multiple historical power generation powers, and marks the average value as the normal reference power, calculates the difference between the power generation power corresponding to the current date and the normal reference power to obtain the power difference, and obtains the corresponding power adjustment information based on the power difference; Obtain load data for different time periods based on the database. The time periods include weekdays and weekends. When the corresponding time period is a weekday, the peak value corresponds to industrial and commercial activities, and the valley value corresponds to residential electricity consumption. Mark the electricity users corresponding to each time period to obtain the power supply object information corresponding to each time period; Meteorological data is identified to obtain light intensity data and duration. Based on the daily light intensity data and duration, power generation forecast information can be obtained. The power supply efficiency and energy storage capacity of the photovoltaic power station can also be obtained. The power supply efficiency and energy storage capacity are comprehensively judged with the power generation forecast information to obtain the corresponding energy supply and storage information. Identify electricity price data to obtain time-of-use electricity prices, which include peak, flat, and valley prices. Obtain the duration of each time-of-use electricity price and the corresponding change in electricity consumption. Compare the change in electricity consumption with a preset change threshold. When the corresponding change threshold is exceeded, adjust the duration accordingly to obtain the corresponding electricity price-related adjustment. Combine power adjustment information, power supply object information, energy storage information, and electricity price-related adjustment to obtain scheduling prediction information. Step 3: Identify the operation data to obtain power generation related data and analyze the power generation related data to obtain power generation status information; The specific analysis method for analyzing the power generation related data is as follows: A1: PV module data, inverter data, and energy storage system data are obtained by identifying power generation-related data; A2: Obtain the solar power generation power and module temperature based on the PV module data; divide the solar power generation power into multiple power generation intervals, each power generation interval corresponding to a power generation capacity value; match the solar power generation power corresponding to each PV panel with the multiple power generation intervals to obtain the corresponding power generation capacity value; calculate the power generation capacity value corresponding to each PV panel to obtain the total power generation capacity value; obtain a pre-set module temperature threshold, compare the module temperature corresponding to each PV panel with the module temperature threshold; when the module temperature is greater than the module temperature threshold, calculate the difference between the module temperature and the module temperature threshold to obtain a temperature anomaly value; add up the temperature anomalies corresponding to each PV panel to obtain the total temperature anomaly value; A3: Obtain the inverter output power and input power based on the inverter data. Calculate the difference between the output power and input power to obtain the power loss value. Calculate the conversion efficiency based on the power loss value and obtain a preset conversion reference efficiency. When the conversion efficiency is less than the conversion reference efficiency, calculate the difference between the conversion efficiency corresponding to the current inverter and the conversion reference efficiency to obtain a conversion abnormality value. A4: Obtain energy storage margin and energy storage efficiency based on energy storage system data. Obtain a pre-set minimum margin limit and compare the energy storage margin of each energy storage module with the minimum margin limit. When the energy storage margin is less than the minimum margin limit, mark the corresponding energy storage module as a full storage module. Count the number of full storage modules and record it as the number of full storage modules. Calculate the ratio of the number of full storage modules to the total number of energy storage modules to obtain the full storage percentage. A5: Construct two circles with the total temperature anomaly value and the conversion anomaly value as their radii. Draw a line perpendicular to the two circles with the centers of the two circles as the starting and ending points, with the length of the line equal to the full storage ratio. Construct a frustum with the two circles and the line. Construct a sphere with the midpoint of the line as the center and the total power generation capacity as the radius. Cut off the portion of the frustum corresponding to the sphere and calculate the volume of the remaining portion. This volume is labeled as the power generation impact value. A6: Compare the power generation impact value with a preset power generation impact threshold. When the power generation impact value is greater than the power generation impact threshold, generate corresponding power generation status information indicating abnormal power generation status. Step 4: Comprehensively analyze the dispatch forecast information and power generation status information to obtain the initial dispatch strategy; Step 5: Identify the operation data to obtain response action information and analyze the response action information to obtain scheduling optimization information; Step 6: Optimize the initial dispatch strategy based on the dispatch optimization information to obtain the power supply dispatch instruction.

2. A photovoltaic power station power supply scheduling method based on data analysis according to claim 1, characterized in that: The obtaining of the operating data of each device module and the power supply data of the photovoltaic power station is specifically as follows: the corresponding operating data is collected from each device module of the photovoltaic power station through sensors, and the operating data includes power generation-related data and response action information, wherein the power generation-related data includes photovoltaic module data, inverter data and energy storage system data, and the response action information includes instruction execution time, response delay and online status; the power supply data includes power generation data, load data, meteorological data and electricity price data.

3. The photovoltaic power station power supply scheduling method based on data analysis according to claim 1 is characterized in that: The scheduling prediction information and power generation status information are comprehensively analyzed to obtain the initial scheduling strategy, specifically: the power generation status information is identified, and when the power generation status information indicates that the power generation status is abnormal, the corresponding initial scheduling strategy is generated to be abnormality investigation; conversely, when the power generation status information corresponds to a normal power generation status, the corresponding initial scheduling strategy is generated to maintain the original scheduling prediction information.

4. The photovoltaic power station power supply scheduling method based on data analysis according to claim 1 is characterized in that: The specific analysis method of the response action information is as follows: S1: By identifying the response action information, the command execution time, response delay and online status of each device module of the photovoltaic power station are obtained; S2: Extract the historical execution times of multiple response actions corresponding to each device module from the database, calculate the average value of the multiple historical execution times, and mark the average value as the execution reference time. Obtain the execution time corresponding to each device module. When the execution time is greater than the execution reference time, calculate the difference between the execution time and the corresponding execution reference time to obtain the execution timeout. Add the execution timeouts corresponding to each device module to obtain the total execution shadow value. S3: Obtain response delays corresponding to multiple response actions of the device module, obtain a preset response delay threshold, compare each response delay with the response delay threshold, and when the response delay is greater than the response delay threshold, mark the corresponding device module as a response abnormal module, count the number of response abnormal modules, and record it as the number of abnormal response modules; S4: Obtain a preset detection time period, identify the online status of each device module corresponding to the detection time period, and if the online status corresponds to offline, record the offline duration as offline time, and calculate the ratio of the offline duration to the duration corresponding to the detection time period to obtain an offline impact value; S5: Calculate the total execution shadow value, the number of abnormal response modules, and the offline impact value to obtain a response status value; S6: When the response state value is greater than a preset response state threshold, a corresponding response state mark is generated as abnormal, and corresponding scheduling optimization information is generated as an instruction adjustment delay.

Images