A fragile variable observer-based optical storage charging station abnormal data detection method and system

Abstract

The application discloses a kind of based on fragile variable observer's light storage fills abnormal data detection method and system of station, it is related to data processing technical field, including data window generation module, fragile variable role observation module and abnormal data determination module.Data window generation module is used to obtain the multi-source operating data of light storage fills station and forms continuous data detection window;Fragile variable role observation module is used to identify the fragile working condition window that photovoltaic available energy is higher than vehicle charging energy and energy storage remaining absorption capacity is insufficient, and distributes corresponding role for photovoltaic, vehicle, energy storage, consumption branch, photovoltaic reduction and DC bus data;Abnormal data determination module is used to judge whether photovoltaic available energy and each legal data direction are closed, and the abnormal data category is positioned in combination with bus state.The application can identify the abnormality that single item data is in normal range but multi-source data energy relationship is not closed.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a method and system for detecting abnormal data in photovoltaic energy storage and charging stations based on a vulnerable variable observer. Background Technology

[0002] A typical photovoltaic-storage-charging (PV-SGC) power station includes PV power generation units, energy storage units, charging units, a DC bus, an energy management platform, and a station operation data acquisition platform. During operation, it continuously generates multi-source operational data, including available PV power, actual PV power connected to the grid, energy storage state of charge, energy storage charging power, vehicle charging power, power input to the grid's absorption branch, PV power reduction, DC bus voltage, and vehicle connection status. This data is typically uploaded to the system by the station's energy management platform, inverters, energy storage converters, charging pile controllers, and DC bus acquisition devices. The system performs status identification, anomaly alerts, operation record generation, and data quality analysis. A vulnerability variable observer, a data processing unit executed by the system, is used to classify and verify the role of multi-source operational data in the energy destination closure judgment of the PV-SGC power station.

[0003] Existing methods for detecting abnormal data often rely on individual variable exceeding limits, sudden changes in historical curves, abnormal differences between adjacent time points, and reconstruction errors in general time series models. However, under specific operating conditions such as surplus photovoltaic power, near-full energy storage, sudden reduction in vehicle charging load, and the need for power stations to explain the destination of remaining energy through absorption branches and photovoltaic reduction data, even if a few key data points are replaced with values ​​within the normal range, it may not cause single-point exceeding limits or obvious time series changes. In this case, each variable may appear reasonable when observed individually, but the energy source, vehicle absorption, energy storage absorption, absorption, and source reduction described by these variables cannot be closed, making it difficult for the system to detect hidden anomalies where the remaining photovoltaic energy has no clear destination at the data level, and further making it difficult to determine which type of vulnerable variable the anomaly mainly originates from. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for detecting abnormal data in photovoltaic energy storage and charging stations based on a vulnerable variable observer, so as to solve the problems mentioned in the background art.

[0005] To address the aforementioned technical problems, this invention provides the following technical solution: an abnormal data detection system for photovoltaic-storage-charging stations based on a vulnerable variable observer, comprising a data window generation module, a vulnerable variable role observation module, and an abnormal data determination module. The data window generation module acquires multi-source operational data from the photovoltaic-storage-charging station and forms a continuous data detection window according to a unified time scale, enabling data uploaded by different devices to be analyzed within the same time window. The vulnerable variable role observation module determines the energy source role, energy absorption role, source-end reduction role, and status verification role of various operational data under the current operating conditions based on the photovoltaic surplus status, energy storage charge status, vehicle charging load, absorption branch input status, photovoltaic reduction status, and DC bus voltage status within the current data detection window. The abnormal data determination module determines whether the photovoltaic surplus energy has a closed-loop destination at the data level based on the role relationships of the various operational data, and outputs an abnormal data window, abnormal data category, and abnormal cause prompt when the energy destination is not closed. The closed-loop destination refers to the state where the available photovoltaic energy within the same data detection window can be explained by vehicle charging energy, energy storage charging energy, absorption branch absorbed energy, and photovoltaic reduction energy.

[0006] According to the above technical solution, the data window generation module includes a data receiving submodule, a time alignment submodule, and a window energy conversion submodule; the data receiving submodule is used to receive photovoltaic available power, actual photovoltaic power connected to the grid, energy storage state of charge, energy storage charging power, vehicle charging power, power input of the absorption branch, DC bus voltage, and vehicle access status; the time alignment submodule is used to align data from different sampling periods to the same detection window; the window energy conversion submodule is used to convert the power data in the detection window into corresponding energy data; The vulnerable variable role observation module includes a vulnerable condition identification submodule, a variable role allocation submodule, and a role succession relationship generation submodule. The vulnerable condition identification submodule is used to identify detection windows where the available photovoltaic energy is higher than the vehicle charging energy and the remaining energy storage absorption capacity is insufficient. The variable role allocation submodule is used to assign available photovoltaic data as the source role, vehicle charging data, energy storage charging data, and absorption branch data as the absorption role, photovoltaic reduction data as the source-end reduction role, and DC bus voltage data as the status verification role. The role succession relationship generation submodule is used to generate the data verification order between vehicle absorption, energy storage absorption, absorption, and source-end reduction. The abnormal data determination module includes an energy destination closure submodule, a bus status verification submodule, and an abnormal output submodule. The energy destination closure submodule is used to determine whether there is a closure between the available photovoltaic energy and various types of energy destinations. The bus status verification submodule is used to confirm whether an abnormal energy destination is accompanied by an abnormal bus status based on changes in DC bus voltage. The abnormal output submodule is used to output the abnormal window number, abnormal data category, and abnormal cause description. The absorption branch refers to the load branch used to absorb surplus electrical energy from the photovoltaic-storage-charging station and capable of uploading input power data.

[0007] A method for detecting anomaly data in a photovoltaic energy storage and charging station based on a vulnerable variable observer includes the following steps: S1. Acquire and organize multi-source operation data of photovoltaic-storage-charging stations: The system receives available photovoltaic power, actual photovoltaic power connected to the grid, energy storage state of charge, energy storage charging power, vehicle charging power, power input of the absorption branch, DC bus voltage and vehicle access status. The system aligns the data from different sources according to a unified sampling time and forms a continuous data detection window. The system saves source data, absorbed data, reduced data and status verification data in each data detection window. S2. Identify specific operating conditions that require vulnerable variable observation: The system determines whether there is a state in the current data detection window where the available photovoltaic energy is higher than the vehicle charging energy. The system further determines whether the energy storage state of charge is close to the preset upper limit and whether the remaining energy storage absorption capacity is insufficient to receive the surplus photovoltaic energy. When the available photovoltaic energy in the current data detection window is higher than the vehicle charging energy and the remaining energy storage absorption capacity is insufficient to receive the surplus photovoltaic energy after the vehicle is charged, the system marks the current data detection window as a vulnerable operating condition window. S3. Generate variable role relationships within the vulnerable operating condition window: The system determines the data corresponding to the available photovoltaic power as the energy source role, the data corresponding to the vehicle charging power, energy storage charging power, and power input of the absorption branch as the energy absorption role, the data corresponding to the photovoltaic power reduction as the source-end reduction role, and the data corresponding to the DC bus voltage as the status verification role. The system forms a role succession sequence for detecting data consistency based on vehicle load changes, energy storage absorption capacity, and the status of the absorption branch. S4. Determine whether the energy destination is closed based on the role relationship of variables: The system calculates the available photovoltaic energy, vehicle charging energy, energy storage absorbed energy, energy absorption and consumption energy, and photovoltaic reduction energy within the vulnerable operating condition window. The system takes the vehicle charging energy, energy storage absorbed energy, energy absorption and consumption energy, and photovoltaic reduction energy as the legitimate data destinations of the available photovoltaic energy. If there is still residual energy exceeding the preset tolerance after deducting the above legitimate data destinations from the available photovoltaic energy, the system determines that there is an energy destination non-closed phenomenon in the current data detection window. S5. Locate the abnormal data category and output the detection results according to the role succession order: In the detection window where the energy destination is not closed, the system verifies whether each role has undertaken the corresponding data destination in the order of vehicle absorption, energy storage absorption, consumption absorption and source reduction. The system combines the DC bus voltage change to determine which category the abnormality mainly comes from: source role data, absorption role data, source reduction role data and status verification role data. The system outputs the abnormal data window, abnormal data category, abnormal judgment basis and abnormal reason prompt.

[0008] According to the above technical solution, step S1 specifically includes: S1-1. Establishing a data detection window: The system establishes the first data detection window at a fixed time interval. Each data detection window contains [number] data detection windows. There are sampling points, and the time interval between adjacent sampling points is . ,in, Indicates the data detection window number. This indicates the number of sampling points within a data detection window. Indicates the time length between adjacent sampling points; S1-2. Window Energy Conversion of Power Data: Since this method determines whether the available photovoltaic energy within a data detection window is jointly interpreted by vehicle charging, energy storage charging, consumption branches, and photovoltaic reduction data, the system first converts various instantaneous power sampling values ​​into cumulative energy within the same window. To enable different power data to be closed-loop judged on the same energy scale, the system calculates the following formula: The first data detection window Window energy corresponding to power-like data: ,in, Indicates the first The first data detection window The window energy corresponding to the power data. Indicates the first The first data detection window The sampling point of the first sampling point Class of power data, Indicates the sampling point number. Corresponding in sequence , , , , , Indicates the available power of photovoltaics. Indicates the vehicle's charging power. Indicates the power absorbed during energy storage charging. Indicates the power input of the absorption branch. The formula represents the photovoltaic power reduction. The significance of this formula is that it converts the power sampling value within the detection window into an energy value, so that the photovoltaic source data and various destination data can be compared on the same scale. The photovoltaic available power refers to the available power generation before reduction provided by the data acquisition platform of the power station. The photovoltaic power reduction refers to the reduction part of the available power generation that has not entered the process of vehicle charging, energy storage charging, and consumption branch line input. S1-3, Forming basic data records: The system in the... Records in each data detection window , , , , Energy storage state of charge DC bus voltage maximum value Minimum DC bus voltage and the number of vehicles connected ,in, Indicates the first The state of charge of the energy storage at the start of each data detection window is used to calculate the remaining absorption capacity of the energy storage to absorb excess photovoltaic energy within that data detection window. Indicates the first The maximum DC bus voltage within each data detection window. Indicates the first Minimum DC bus voltage within each data detection window Indicates the first The system calculates and records the actual photovoltaic power connected to the grid at the end of each data detection window, based on the same window energy conversion method as steps S1-2, as the actual photovoltaic power connected to the grid within the k-th data detection window. Used to verify the available energy of photovoltaic power. Excluding solar energy reduction Whether the energy subsequently integrated with the photovoltaic system is consistent with the actual energy output. Not used as energy to reach surplus quantity Independent vectors in computation.

[0009] According to the above technical solution, step S2 specifically includes: S2-1. Calculating the Remaining Absorption Capacity of Energy Storage: To determine whether energy storage can still play a major absorption role, the system calculates the remaining absorption capacity according to the following formula. Remaining energy absorption capacity within each data detection window: ,in, Indicates the first The remaining energy that can be absorbed within each data detection window. This indicates the preset upper limit of the energy storage state of charge. Indicates the first The energy storage charge state at the start of each data detection window. This represents the rated capacity of the energy storage. The significance of this formula lies in determining the surplus energy that the energy storage can still handle by comparing the energy storage's state of charge with its rated capacity. and Using the same dimensionless state of charge representation, when the collected state of charge is recorded as a percentage, the system first converts it to a decimal form before participating in the calculation of the remaining energy storage absorption capacity. The remaining energy storage absorption capacity is calculated based on the energy storage state of charge at the beginning of the data detection window, in order to avoid the reverse calculation of the energy storage charging process that has already occurred within the window into the remaining absorption capacity judgment. S2-2, Determining the surplus state of photovoltaic power: When Greater than At that time, the system determines the first Several data detection windows show a state where the available photovoltaic energy is higher than the vehicle charging energy. Indicates the available energy from photovoltaics. This indicates the vehicle's charging energy, and this judgment is used to confirm whether the current window has surplus photovoltaic energy that requires further explanation; S2-3, Determine the vulnerable condition window: When Greater than ,and Less than At the same time, the number of vehicles connected If no addition is made, the system will [the following will be displayed:] the first [number]. The data detection window was identified as a vulnerable operating condition window, and the number of vehicles connected... "No increase" means that the number of vehicles in the access state at the end of the k-th data detection window is no greater than the number of vehicles in the access state at the end of the (k-1)-th data detection window. This judgment indicates that vehicle charging cannot fully absorb the available photovoltaic energy, and the remaining absorption capacity of energy storage is also insufficient to fully accept the surplus energy. The data from the power station should show clear data manifestations of absorption, photovoltaic reduction, and changes in bus status. Less than This indicates that the remaining absorption capacity of the energy storage is insufficient to independently absorb the remaining usable photovoltaic energy after the vehicle is charged. It does not mean that the energy storage cannot absorb any energy at all within the current data detection window. When the k-th data detection window does not meet the judgment conditions of the vulnerable operating condition window, the system marks the data detection window as a non-vulnerable operating condition window and proceeds to the acquisition and processing steps of the next data detection window.

[0010] According to the above technical solution, step S3 specifically includes: S3-1. Determine the energy source role: The system will... The corresponding photovoltaic available data is identified as the energy source role, which is used to represent the first... The energy sources within each data detection window need to be interpreted jointly by vehicle, energy storage, consumption branch, and photovoltaic reduction data; S3-2, Determine the energy absorption role: The system will... The corresponding vehicle charging data is identified as the primary absorption role, and will The corresponding energy storage charging data identifies it as the second absorption role. The corresponding absorption branch input data is determined as the end absorption role. The first absorption role is used to represent the energy actually consumed by the vehicle, the second absorption role is used to represent the energy actually absorbed by the energy storage, and the end absorption role is used to represent the energy destination undertaken by the absorption branch when the vehicle and the energy storage cannot fully absorb the excess energy. S3-3. Determine the source reduction role and status verification role: The system will... The corresponding photovoltaic reduction data is identified as the source-end reduction role, and the DC bus voltage data is identified as the status verification role. The source-end reduction role is used to represent the surplus photovoltaic energy that has not entered the DC side of the station, and the status verification role is used to verify whether the change in energy destination forms a corresponding status manifestation in the DC bus voltage. S3-4. Forming a role succession sequence: Within the vulnerable operating condition window, the system forms a role succession sequence for vehicle absorption, energy storage absorption, absorption absorption, source reduction, and bus status verification. This role succession sequence is only used for system data consistency verification and does not send control commands to the photovoltaic power generation unit, energy storage unit, charging unit, and absorption branch. The role succession sequence is the data processing sequence adopted by the system when performing consistency verification on the collected operating data.

[0011] According to the above technical solution, step S4 specifically includes: S4-1, Calculate the remaining energy destination: To determine whether the available photovoltaic energy has complete data destination, the system calculates the remaining energy destination according to the following formula. Energy destination and remaining amount within each data detection window: ,in, Indicates the first Energy within a data detection window that has not yet been explained by the legitimate data destination. Indicates the available energy from photovoltaics. Indicates the vehicle's charging energy. This indicates that the energy storage and charging process absorbs energy. This indicates that the energy is absorbed by the absorption branch. This formula represents the energy reduction from photovoltaic (PV) power. Its significance lies in treating vehicle charging, energy storage charging, grid connection, and PV power reduction as legitimate data destinations for usable PV energy, thereby checking whether the energy relationships described by the data are closed. When the value is positive, it indicates that there is a remaining portion of the available photovoltaic energy that has not yet been explained by vehicle charging, energy storage charging, grid connection input, and photovoltaic reduction data. This remaining portion is the missing energy destination object on which subsequent anomaly judgment is based. S4-2. Calculate the bus voltage fluctuation: To verify whether abnormal energy destination is accompanied by changes in the DC bus status, the system calculates the following formula: Bus voltage fluctuation within each data detection window: ,in, Indicates the first The bus voltage fluctuation within each data detection window. This indicates the maximum DC bus voltage within the detection window. This indicates the minimum DC bus voltage within the detection window. The significance of this formula is that it determines whether the bus state exhibits a state performance corresponding to the change in energy destination by measuring the voltage change amplitude within the same window. The bus voltage fluctuation is not used as a separate basis for judging abnormal data, but rather as an auxiliary method to determine whether the non-closed phenomenon has formed a corresponding performance in the DC bus state when the energy destination is not closed. S4-3, Determining the direction of energy in an unclosed window: When At that time, the system determines the first In some data detection windows, the available photovoltaic energy was not fully explained by the legitimate data destinations. This indicates the allowable energy calculation error tolerance, used to exclude minor differences caused by sampling and conversion errors. The judgment result serves as input for subsequent anomaly data category localization. When the system determines that there is no energy out-of-closed phenomenon in the k-th data detection window, it outputs a normal window mark and proceeds to the acquisition and processing steps of the next data detection window.

[0012] According to the above technical solution, step S5 specifically includes: S5-1, Identifying Missing Destinations: Effective energy destinations refer to the corresponding roles that generate energy exceeding the preset energy tolerance within the current data detection window. The window energy, and that window energy is used as the energy destination for the remaining amount. The deduction term in the calculation participates in the energy closure judgment, when the first... Each data detection window satisfies At that time, the system checks whether the corresponding role has formed an effective energy destination in the order of vehicle absorption, energy storage absorption, consumption absorption, and source reduction. If the vehicle absorption role corresponds to... Below The corresponding role of energy storage and absorption Not greater than the preset energy tolerance Absorbing and absorbing the corresponding role Not greater than the preset energy tolerance Furthermore, the source side reduced the corresponding roles. Not greater than the preset energy tolerance If so, the system determines that the current window has a missing destination role; S5-2, Combining busbar status verification anomaly categories: When the... Each data detection window satisfies ,and Less than the preset voltage fluctuation threshold At that time, the system determined that there was an anomaly in the current data detection window where the energy destination was not closed despite no significant fluctuation in the bus voltage. The anomaly message indicated an inconsistency in the data between available photovoltaic power, vehicle charging power, energy storage charging power, power input of the absorption branch, and photovoltaic power reduction. This represents the threshold used to determine whether the bus voltage has significant fluctuations; S5-3, Identify the category of anomalous data with priority: When ,and ,at the same time and At that time, the system will identify the input data of the absorption branch and the photovoltaic reduction data as the priority categories of abnormal data for verification; when ,and ,at the same time When this happens, the system will prioritize verifying the energy storage state of charge data and energy storage charging power data as the categories of abnormal data; when ,and At that time, the system identifies available photovoltaic power data, actual photovoltaic power input data, and photovoltaic power reduction data as priority categories for verification of abnormal data. The continuous large-scale absorption of energy by energy storage refers to... This indicates that the energy absorbed during charging has exceeded the sum of the remaining absorption capacity calculated from the energy storage state of charge and the allowable error. Each data detection window satisfies However, if the priority verification conditions for the input data of the absorption branch and the photovoltaic reduction data, the energy storage state of charge data and the energy storage charging power data, and the photovoltaic available power data and the actual photovoltaic power connected in this step are not met, the system will mark the data detection window as having an abnormal comprehensive data consistency and will participate in the verification. Calculated , , , , as well as As an output for review. When the same data detection window simultaneously meets multiple priority verification conditions, the system generates an abnormal data category set according to the data category corresponding to each priority verification condition, and outputs the entire abnormal data category set. S5-4. Output anomaly detection results: The system outputs the... The window number, abnormal data category, energy destination, and remaining amount of each data detection window. Bus voltage fluctuation The system provides error message prompts, which include specific reasons for issues such as unclosed photovoltaic surplus energy, insufficient energy storage absorption capacity, lack of absorption and utilization, lack of source reduction, and bus status mismatch. This allows maintenance personnel to verify the corresponding data source based on the output results.

[0013] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: Starting from the energy destination relationship between multi-source operational data, this invention detects the data self-consistency under specific operating conditions such as surplus photovoltaic power, limited energy storage absorption, and insufficient vehicle load. By defining photovoltaic data as the source role, vehicle charging, energy storage charging, absorption branches, and photovoltaic reduction as destination roles at different levels, and bus voltage as the status verification role, the system can identify abnormal windows where individual data values ​​are all within the normal range but the overall energy relationship cannot be closed. Furthermore, it can locate the abnormal data category based on the missing destination roles and changes in bus voltage, thereby improving the targeting, interpretability, and hidden anomaly identification capabilities of abnormal data detection in photovoltaic-energy storage-charging stations. Attached Figure Description

[0014] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart illustrating the present invention; Figure 2 This is a schematic diagram of the overall modular structure of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] Please see Figure 1 and Figure 2This invention provides a technical solution: an abnormal data detection system for photovoltaic-storage-charging stations based on a vulnerable variable observer, comprising a data window generation module, a vulnerable variable role observation module, and an abnormal data determination module. The data window generation module is used to acquire multi-source operating data of the photovoltaic-storage-charging station and form a continuous data detection window according to a unified time scale, so that data uploaded by different devices can be analyzed in the same time window. The vulnerable variable role observation module is used to determine the energy source role, energy absorption role, source-end reduction role, and status verification role of various operating data under the current operating conditions based on the photovoltaic surplus status, energy storage charge status, vehicle charging load, absorption branch input status, photovoltaic reduction status, and DC bus voltage status within the current data detection window. The abnormal data determination module is used to determine whether the photovoltaic surplus energy has a closed destination at the data level based on the role relationship of various operating data, and outputs an abnormal data window, abnormal data category, and abnormal cause prompt when the energy destination is not closed. A closed destination refers to the state in which the photovoltaic available energy within the same data detection window can be explained by vehicle charging energy, energy storage charging energy, absorption branch absorbed energy, and photovoltaic reduction energy. The data window generation module includes a data receiving submodule, a time alignment submodule, and a window energy conversion submodule. The data receiving submodule is used to receive available photovoltaic power, actual photovoltaic power connected to the grid, energy storage state of charge, energy storage charging power, vehicle charging power, power input of the absorption branch, DC bus voltage, and vehicle access status. The time alignment submodule is used to align data from different sampling periods to the same detection window. The window energy conversion submodule is used to convert the power data within the detection window into the corresponding energy data. The vulnerable variable role observation module includes a vulnerable condition identification submodule, a variable role allocation submodule, and a role succession relationship generation submodule. The vulnerable condition identification submodule is used to identify detection windows where the available photovoltaic energy is higher than the vehicle charging energy and the remaining energy storage absorption capacity is insufficient. The variable role allocation submodule is used to assign the available photovoltaic data as the source role, the vehicle charging data, energy storage charging data, and absorption branch data as the absorption role, the photovoltaic reduction data as the source-end reduction role, and the DC bus voltage data as the status verification role. The role succession relationship generation submodule is used to generate the data verification order between vehicle absorption, energy storage absorption, absorption, and source-end reduction. The abnormal data determination module includes an energy destination closure submodule, a bus status verification submodule, and an abnormal output submodule. The energy destination closure submodule is used to determine whether the available photovoltaic energy and various destination energy are closed. The bus status verification submodule is used to confirm whether the abnormal energy destination is accompanied by the abnormal bus status based on the change of DC bus voltage. The abnormal output submodule is used to output the abnormal window number, abnormal data category, and abnormal cause description. The absorption branch refers to the load branch used to absorb the surplus power of the photovoltaic storage and charging station and can upload the input power data. A method for detecting anomaly data in a photovoltaic energy storage and charging station based on a vulnerable variable observer includes the following steps: S1. Acquire and organize multi-source operation data of photovoltaic-storage-charging stations: The system receives available photovoltaic power, actual photovoltaic power connected to the grid, energy storage state of charge, energy storage charging power, vehicle charging power, power input of the absorption branch, DC bus voltage and vehicle access status. The system aligns the data from different sources according to a unified sampling time and forms a continuous data detection window. The system saves source data, absorbed data, reduced data and status verification data in each data detection window. S2. Identify specific operating conditions that require vulnerable variable observation: The system determines whether there is a state in the current data detection window where the available photovoltaic energy is higher than the vehicle charging energy. The system further determines whether the energy storage state of charge is close to the preset upper limit and whether the remaining energy storage absorption capacity is insufficient to receive the surplus photovoltaic energy. When the available photovoltaic energy in the current data detection window is higher than the vehicle charging energy and the remaining energy storage absorption capacity is insufficient to receive the surplus photovoltaic energy after the vehicle is charged, the system marks the current data detection window as a vulnerable operating condition window. S3. Generate variable role relationships within the vulnerable operating condition window: The system determines the data corresponding to the available photovoltaic power as the energy source role, the data corresponding to the vehicle charging power, energy storage charging power, and power input of the absorption branch as the energy absorption role, the data corresponding to the photovoltaic power reduction as the source-end reduction role, and the data corresponding to the DC bus voltage as the status verification role. The system forms a role succession sequence for detecting data consistency based on vehicle load changes, energy storage absorption capacity, and the status of the absorption branch. S4. Determine whether the energy destination is closed based on the role relationship of variables: The system calculates the available photovoltaic energy, vehicle charging energy, energy storage absorbed energy, energy absorption and consumption energy, and photovoltaic reduction energy within the vulnerable operating condition window. The system takes the vehicle charging energy, energy storage absorbed energy, energy absorption and consumption energy, and photovoltaic reduction energy as the legitimate data destinations of the available photovoltaic energy. If there is still residual energy exceeding the preset tolerance after deducting the above legitimate data destinations from the available photovoltaic energy, the system determines that there is an energy destination non-closed phenomenon in the current data detection window. S5. Locate the abnormal data category and output the detection results according to the role succession order: In the detection window where the energy destination is not closed, the system verifies whether each role has undertaken the corresponding data destination in the order of vehicle absorption, energy storage absorption, consumption absorption and source reduction. The system combines the DC bus voltage change to determine which category the abnormality mainly comes from: source role data, absorption role data, source reduction role data and status verification role data. The system outputs the abnormal data window, abnormal data category, abnormal judgment basis and abnormal reason prompt. Step S1 specifically includes: S1-1. Establishing a data detection window: The system establishes the first data detection window at a fixed time interval. Each data detection window contains [number] data detection windows. There are sampling points, and the time interval between adjacent sampling points is . ,in, Indicates the data detection window number. This indicates the number of sampling points within a data detection window. Indicates the time length between adjacent sampling points; S1-2. Window Energy Conversion of Power Data: Since this method determines whether the available photovoltaic energy within a data detection window is jointly interpreted by vehicle charging, energy storage charging, consumption branches, and photovoltaic reduction data, the system first converts various instantaneous power sampling values ​​into cumulative energy within the same window. To enable different power data to be closed-loop judged on the same energy scale, the system calculates the following formula: The first data detection window Window energy corresponding to power-like data: ,in, Indicates the first The first data detection window The window energy corresponding to the power data. Indicates the first The first data detection window The sampling point of the first sampling point Class of power data, Indicates the sampling point number. Corresponding in sequence , , , , , Indicates the available power of photovoltaics. Indicates the vehicle's charging power. Indicates the power absorbed during energy storage charging. Indicates the power input of the absorption branch. The formula represents the reduction in photovoltaic power. Its significance lies in converting the power sampling value within the detection window into an energy value, enabling photovoltaic source data and various destination data to be compared on the same scale. The available photovoltaic power refers to the available power generation before reduction provided by the data acquisition platform of the power station. The photovoltaic power reduction refers to the reduction portion of the available power generation that has not entered the process of vehicle charging, energy storage charging, and consumption branch line input. S1-3, Forming basic data records: The system in the... Records in each data detection window , , , , Energy storage state of charge DC bus voltage maximum value Minimum DC bus voltage and the number of vehicles connected ,in, Indicates the first The state of charge of the energy storage at the start of each data detection window is used to calculate the remaining absorption capacity of the energy storage to absorb excess photovoltaic energy within that data detection window. Indicates the first The maximum DC bus voltage within each data detection window. Indicates the first Minimum DC bus voltage within each data detection window Indicates the first The system calculates and records the actual photovoltaic power connected to the grid at the end of each data detection window, based on the same window energy conversion method as steps S1-2, as the actual photovoltaic power connected to the grid within the k-th data detection window. Used to verify the available energy of photovoltaic power. Excluding solar energy reduction Whether the energy subsequently integrated with the photovoltaic system is consistent with the actual energy output. Not used as energy to reach surplus quantity Independent vectors in computation; The purpose of this step is to establish a common time reference for subsequent judgments. Data collected from photovoltaic, energy storage, charging piles, absorption branches, and busbars in a photovoltaic-storage-charging station typically comes from different devices. Directly comparing raw collected values ​​can easily lead to discrepancies between data at the same moment. This invention first organizes various types of data into a single data detection window, so that subsequent judgments no longer rely on a single instantaneous value, but rather observe whether there is a match between the photovoltaic energy source and various energy destinations within a short time window. This processing is not a simple data aggregation, but rather establishes a comparable data foundation for subsequent energy destination closure judgments, allowing power, status, and vehicle access information from different sources to participate in anomaly identification within the same window.

[0017] Step S2 specifically includes: S2-1. Calculating the Remaining Absorption Capacity of Energy Storage: To determine whether energy storage can still play a major absorption role, the system calculates the remaining absorption capacity according to the following formula. Remaining energy absorption capacity within each data detection window: ,in, Indicates the first The remaining energy that can be absorbed within each data detection window. This indicates the preset upper limit of the energy storage state of charge. Indicates the first The energy storage charge state at the start of each data detection window. This represents the rated capacity of the energy storage. The significance of this formula lies in determining the surplus energy that the energy storage can still handle by comparing the energy storage's state of charge with its rated capacity. and Using the same dimensionless state of charge (SOC) representation, when the collected SOC is recorded as a percentage, the system first converts it to a decimal form before participating in the calculation of remaining energy storage capacity. The remaining energy storage capacity is calculated based on the SOC at the start of the data detection window to avoid retroactively including charging processes that have already occurred within the window in the remaining capacity assessment. The remaining energy storage capacity is also limited by the maximum allowable charging power. When the maximum allowable charging power corresponds to less absorbable energy within the current data detection window than the calculated result, the system uses the absorbable energy as the threshold. .

[0018] S2-2, Determining the surplus state of photovoltaic power: When Greater than At that time, the system determines the first Several data detection windows show a state where the available photovoltaic energy is higher than the vehicle charging energy. Indicates the available energy from photovoltaics. This indicates the vehicle's charging energy, and this judgment is used to confirm whether the current window has surplus photovoltaic energy that requires further explanation; S2-3, Determine the vulnerable condition window: When Greater than ,and Less than At the same time, the number of vehicles connected If no addition is made, the system will [the following will be displayed:] the first [number]. The data detection window was identified as a vulnerable operating condition window, and the number of vehicles connected... "No increase" means that the number of vehicles in the access state at the end of the k-th data detection window is no greater than the number of vehicles in the access state at the end of the (k-1)-th data detection window. This judgment indicates that vehicle charging cannot fully absorb the available photovoltaic energy, and the remaining absorption capacity of energy storage is also insufficient to fully accept the surplus energy. The data from the power station should show clear data manifestations of absorption, photovoltaic reduction, and changes in bus status. Less than This indicates that the remaining absorption capacity of the energy storage is insufficient to independently absorb the remaining usable photovoltaic energy after the vehicle is charged. It does not mean that the energy storage cannot absorb any energy at all within the current data detection window. When the k-th data detection window does not meet the judgment conditions of the vulnerable operating condition window, the system marks the data detection window as a non-vulnerable operating condition window and proceeds to the acquisition and processing steps of the next data detection window. The real potential for hidden data inconsistencies arises in specific scenarios where there is a significant surplus of available photovoltaic energy, insufficient vehicle charging load, and the energy storage system cannot continue to absorb energy sufficiently. Under these conditions, the data from the power plant should clearly show one of the following: absorption by the absorption branch, photovoltaic load reduction, or changes in the busbar status. If these characteristics are not present in the data, even if individual data points remain within normal ranges, it indicates a high risk of data inconsistency in the current window. Compared to conventional methods that directly apply a uniform threshold or model to all time periods, this invention first identifies this specific vulnerable operating condition before proceeding with subsequent observation and judgment. This allows the detection focus to be concentrated on the operational gaps most easily disguised as anomalies, reducing the interference of irrelevant operating conditions on the judgment results.

[0019] Step S3 specifically includes: S3-1. Determine the energy source role: The system will... The corresponding photovoltaic available data is identified as the energy source role, which is used to represent the first... The energy sources within each data detection window need to be interpreted jointly by vehicle, energy storage, consumption branch, and photovoltaic reduction data; S3-2, Determine the energy absorption role: The system will... The corresponding vehicle charging data is identified as the primary absorption role, and will The corresponding energy storage charging data identifies it as the second absorption role. The corresponding absorption branch input data is determined as the end absorption role. The first absorption role is used to represent the energy actually consumed by the vehicle, the second absorption role is used to represent the energy actually absorbed by the energy storage, and the end absorption role is used to represent the energy destination undertaken by the absorption branch when the vehicle and the energy storage cannot fully absorb the excess energy. S3-3. Determine the source reduction role and status verification role: The system will... The corresponding photovoltaic reduction data is identified as the source-end reduction role, and the DC bus voltage data is identified as the status verification role. The source-end reduction role is used to represent the surplus photovoltaic energy that has not entered the DC side of the station, and the status verification role is used to verify whether the change in energy destination forms a corresponding status manifestation in the DC bus voltage. S3-4. Forming a role succession sequence: Within the vulnerable operating condition window, the system forms a role succession sequence for vehicle absorption, energy storage absorption, absorption, source reduction, and bus status verification. This role succession sequence is only used for system data consistency verification and does not send control commands to photovoltaic power generation units, energy storage units, charging units, and absorption branches. The role succession sequence is the data processing sequence used by the system when performing consistency verification on the collected operating data. Conventional anomaly detection typically treats photovoltaic power, energy storage power, charging power, and bus voltage as independent numerical sequences for judgment. However, this invention does not first determine whether a single value is abnormal, but rather first determines the role that each type of data should play under the current operating conditions. Photovoltaic available data indicates where the energy is generated; vehicle charging data, energy storage charging data, and absorption branch data indicate who absorbs the energy; photovoltaic reduction data indicates whether excess energy is reduced at the source; and bus voltage data verifies whether changes in energy relationships result in corresponding state manifestations on the DC side. Through this classification, the computer equipment can determine which part of the energy relationship each type of data should interpret within the current window. Its originality lies in the fact that the meaning of variables is not fixed but rather its role in determining energy destination is determined according to the current operating conditions, thus avoiding the limitations of judging solely based on the similarity of single-variable curves or exceeding limits.

[0020] Step S4 specifically includes: S4-1, Calculate the remaining energy destination: To determine whether the available photovoltaic energy has complete data destination, the system calculates the remaining energy destination according to the following formula. Energy destination and remaining amount within each data detection window: ,in, Indicates the first Energy within a data detection window that has not yet been explained by the legitimate data destination. Indicates the available energy from photovoltaics. Indicates the vehicle's charging energy. This indicates that the energy storage and charging process absorbs energy. This indicates that the energy is absorbed by the absorption branch. This formula represents the energy reduction from photovoltaic (PV) power. Its significance lies in treating vehicle charging, energy storage charging, grid connection, and PV power reduction as legitimate data destinations for usable PV energy, thereby checking whether the energy relationships described by the data are closed. When the value is positive, it indicates that there is a remaining portion of the available photovoltaic energy that has not yet been explained by vehicle charging, energy storage charging, grid connection input, and photovoltaic reduction data. This remaining portion is the missing energy destination object on which subsequent anomaly judgment is based. S4-2. Calculate the bus voltage fluctuation: To verify whether abnormal energy destination is accompanied by changes in the DC bus status, the system calculates the following formula: Bus voltage fluctuation within each data detection window: ,in, Indicates the first The bus voltage fluctuation within each data detection window. This indicates the maximum DC bus voltage within the detection window. This indicates the minimum DC bus voltage within the detection window. The significance of this formula is that it determines whether the bus status exhibits a state corresponding to the change in energy destination by measuring the voltage change amplitude within the same window. The bus voltage fluctuation is not used as a separate basis for judging abnormal data, but rather as an auxiliary method to determine whether the non-closed phenomenon has formed a corresponding manifestation in the DC bus status when the energy destination is not closed. S4-3, Determining the direction of energy in an unclosed window: When At that time, the system determines the first In some data detection windows, the available photovoltaic energy was not fully explained by the legitimate data destinations. This indicates the allowable energy calculation error tolerance, used to exclude minor differences caused by sampling and conversion errors. The judgment result serves as input for subsequent anomaly data category localization. When the system determines that there is no energy destination closure phenomenon in the k-th data detection window, it outputs a normal window mark and enters the acquisition and processing steps of the next data detection window; This step treats available photovoltaic energy within the same data detection window as a source requiring interpretation, and then examines whether data from vehicle charging, energy storage charging, absorption by grid connection branches, and photovoltaic load reduction collectively provide a complete destination. If available photovoltaic energy is not explained by these legitimate destinations at the data level, it indicates that although the multi-source data appear reasonable on the surface, they cannot form a self-consistent relationship when combined. The purpose of this step is to transform the question of whether a hidden anomaly is an anomaly in a single value into the question of whether the energy relationship described by multiple source data is valid. Compared to conventional threshold detection, reconstruction error detection, and historical curve similarity detection, this step can identify anomalies where individual data points do not exceed limits and curves do not show obvious abrupt changes, but the overall data relationship cannot be established.

[0021] Step S5 specifically includes: S5-1, Identifying Missing Destinations: Effective energy destinations refer to the corresponding roles that generate energy exceeding the preset energy tolerance within the current data detection window. The window energy, and that window energy is used as the energy destination for the remaining amount. The deduction term in the calculation participates in the energy closure judgment, when the first... Each data detection window satisfies At that time, the system checks whether the corresponding role has formed an effective energy destination in the order of vehicle absorption, energy storage absorption, consumption absorption, and source reduction. If the vehicle absorption role corresponds to... Below The corresponding role of energy storage and absorption Not greater than the preset energy tolerance Absorbing and absorbing the corresponding role Not greater than the preset energy tolerance Furthermore, the source side reduced the corresponding roles. Not greater than the preset energy tolerance If so, the system determines that the current window has a missing destination role; S5-2, Combining busbar status verification anomaly categories: When the... Each data detection window satisfies ,and Less than the preset voltage fluctuation threshold At that time, the system determined that there was an anomaly in the current data detection window where the energy destination was not closed despite no significant fluctuation in the bus voltage. The anomaly message indicated an inconsistency in the data between available photovoltaic power, vehicle charging power, energy storage charging power, power input of the absorption branch, and photovoltaic power reduction. This represents the threshold used to determine whether the bus voltage has significant fluctuations; The deviation is determined based on the metering errors of each data acquisition device, the time alignment errors of different data sources, the energy consumption of station auxiliary equipment, and the total deviation of energy conversion losses within the normal operating window. Determined based on the normal voltage regulation range of the DC bus and the voltage acquisition error.

[0022] S5-3, Identify the category of anomalous data with priority: When ,and ,at the same time and At that time, the system will identify the input data of the absorption branch and the photovoltaic reduction data as the priority categories of abnormal data for verification; when ,and ,at the same time When this happens, the system will prioritize verifying the energy storage state of charge data and energy storage charging power data as the categories of abnormal data; when ,and At that time, the system identifies available photovoltaic power data, actual photovoltaic power input data, and photovoltaic power reduction data as priority categories for verification of abnormal data. The continuous large-scale absorption of energy by energy storage refers to... This indicates that the energy absorbed during charging has exceeded the sum of the remaining absorption capacity calculated from the energy storage state of charge and the allowable error. Each data detection window satisfies However, if the priority verification conditions for the input data of the absorption branch and the photovoltaic reduction data, the energy storage state of charge data and the energy storage charging power data, and the photovoltaic available power data and the actual photovoltaic power connected in this step are not met, the system will mark the data detection window as having an abnormal comprehensive data consistency and will participate in the verification. Calculated , , , , as well as As an output for review. When the same data detection window simultaneously meets multiple priority verification conditions, the system generates an abnormal data category set according to the data category corresponding to each priority verification condition, and outputs the entire abnormal data category set. S5-4. Output anomaly detection results: The system outputs the... The window number, abnormal data category, energy destination, and remaining amount of each data detection window. Bus voltage fluctuation The system provides error message prompts, which include specific reasons for issues such as unclosed photovoltaic surplus energy, insufficient energy storage capacity, lack of absorption and utilization, lack of source reduction, and bus status mismatch. This allows maintenance personnel to verify the corresponding data source based on the output results.

[0023] This step further categorizes the reasons for the non-closed energy flow into verifiable data categories. Once the current window is determined to have an incomplete energy flow path, the computer analyzes the data in the order of vehicle absorption, energy storage absorption, energy consumption absorption, source reduction, and bus status verification to determine which type of data is failing to fulfill its explanatory role under the current operating conditions. If the bus voltage does not fluctuate significantly, but the energy relationship is already incomplete, the anomaly is more likely due to static consistency between the source and destination data. If the bus voltage has changed, but the energy consumption branch and photovoltaic reduction data do not show corresponding changes, then the energy consumption branch input data and photovoltaic reduction data are more suitable as priority verification targets. Compared to conventional anomaly detection methods that only output anomaly scores or labels, this step emphasizes the traceability of the anomaly's cause, enabling maintenance personnel to directly identify which data source needs verification, thereby improving the interpretability and executability of the detection results.

[0024] Through the above processing, this invention examines whether the energy relationship jointly described by multi-source data can hold true under specific vulnerable operating conditions. The key lies in first determining whether the current window belongs to a state of photovoltaic surplus and limited conventional absorption capacity, then determining the interpretive role that various types of data should play in this state, subsequently judging whether the available photovoltaic energy has a complete data destination, and finally mapping the non-closed phenomena to specific abnormal data categories. In this way, even if abnormal data is disguised as values ​​within the normal range, this invention can still discover problems from contradictions in the relationships between data and provide verifiable causes of the anomalies. This method avoids the insufficient identification problems caused by simply relying on thresholds, historical curve similarity, and black-box anomaly scoring, and can more accurately identify hidden abnormal data under specific operating conditions in photovoltaic-storage-charging stations.

[0025] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0026] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended 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 make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A system for detecting abnormal data in photovoltaic storage and charging stations based on a vulnerable variable observer, characterized in that: The system includes a data window generation module, a vulnerable variable role observation module, and an anomaly data determination module. The data window generation module acquires multi-source operational data from the photovoltaic-storage-charging station and forms a continuous data detection window according to a unified time scale, enabling data uploaded by different devices to be analyzed within the same time window. The vulnerable variable role observation module determines the energy source role, energy absorption role, source-end reduction role, and status verification role of various operational data under the current operating conditions based on the photovoltaic surplus status, energy storage charge status, vehicle charging load, absorption branch input status, photovoltaic reduction status, and DC bus voltage status within the current data detection window. The anomaly data determination module determines whether the photovoltaic surplus energy has a closed-loop destination at the data level based on the role relationships of the various operational data, and outputs an anomaly data window, anomaly data category, and anomaly cause prompt when the energy destination is not closed.

2. The abnormal data detection system for photovoltaic storage and charging stations based on a vulnerable variable observer according to claim 1, characterized in that: The data window generation module includes a data receiving submodule, a time alignment submodule, and a window energy conversion submodule. The data receiving submodule receives available photovoltaic power, actual photovoltaic power connected to the grid, energy storage state of charge, energy storage charging power, vehicle charging power, power input of the absorption branch, DC bus voltage, and vehicle access status. The time alignment submodule aligns data from different sampling periods to the same detection window. The window energy conversion submodule converts the power data within the detection window into corresponding energy data. The vulnerable variable role observation module includes a vulnerable condition identification submodule, a variable role allocation submodule, and a role succession relationship generation submodule; the vulnerable condition identification submodule is used to identify detection windows where the available photovoltaic energy is higher than the vehicle charging energy and the remaining energy storage absorption capacity is insufficient. The variable role allocation submodule is used to allocate available photovoltaic data as the source role, vehicle charging data, energy storage charging data, and absorption branch data as the absorption role, photovoltaic reduction data as the source-end reduction role, and DC bus voltage data as the status verification role; the role succession relationship generation submodule is used to generate the data verification order between vehicle absorption, energy storage absorption, absorption, and source-end reduction. The abnormal data determination module includes an energy destination closure submodule, a bus status verification submodule, and an abnormal output submodule. The energy destination closure submodule is used to determine whether the available photovoltaic energy and various types of energy destinations are closed. The bus status verification submodule is used to confirm whether an abnormal energy destination is accompanied by an abnormal bus status based on changes in DC bus voltage. The abnormal output submodule is used to output the abnormal window number, abnormal data category, and abnormal cause description.

3. A method for detecting anomalous data in a photovoltaic-storage-charging station based on a vulnerable variable observer, characterized in that: Includes the following steps: S1. Acquire and organize multi-source operation data of photovoltaic-storage-charging stations: The system receives available photovoltaic power, actual photovoltaic power connected to the grid, energy storage state of charge, energy storage charging power, vehicle charging power, power input of the absorption branch, DC bus voltage and vehicle access status. The system aligns the data from different sources according to a unified sampling time and forms a continuous data detection window. The system saves source data, absorbed data, reduced data and status verification data in each data detection window. S2. Identify specific operating conditions that require vulnerable variable observation: The system determines whether there is a state in the current data detection window where the available photovoltaic energy is higher than the vehicle charging energy. The system further determines whether the energy storage state of charge is close to the preset upper limit and whether the remaining energy storage absorption capacity is insufficient to receive the surplus photovoltaic energy. When the available photovoltaic energy in the current data detection window is higher than the vehicle charging energy and the remaining energy storage absorption capacity is insufficient to receive the surplus photovoltaic energy after the vehicle is charged, the system marks the current data detection window as a vulnerable operating condition window. S3. Generate variable role relationships within the vulnerable operating condition window: The system determines the data corresponding to the available photovoltaic power as the energy source role, the data corresponding to the vehicle charging power, energy storage charging power, and power input of the absorption branch as the energy absorption role, the data corresponding to the photovoltaic power reduction as the source-end reduction role, and the data corresponding to the DC bus voltage as the status verification role. The system forms a role succession sequence for detecting data consistency based on vehicle load changes, energy storage absorption capacity, and the status of the absorption branch. S4. Determine whether the energy destination is closed based on the role relationship of variables: The system calculates the available photovoltaic energy, vehicle charging energy, energy storage absorbed energy, energy absorption and consumption energy, and photovoltaic reduction energy within the vulnerable operating condition window. The system takes the vehicle charging energy, energy storage absorbed energy, energy absorption and consumption energy, and photovoltaic reduction energy as the legitimate data destinations of the available photovoltaic energy. If there is still residual energy exceeding the preset tolerance after deducting the above legitimate data destinations from the available photovoltaic energy, the system determines that there is an energy destination non-closed phenomenon in the current data detection window. S5. Locate the abnormal data category and output the detection results according to the role succession order: In the detection window where the energy destination is not closed, the system verifies whether each role has undertaken the corresponding data destination in the order of vehicle absorption, energy storage absorption, consumption absorption and source reduction. The system combines the DC bus voltage change to determine which category the abnormality mainly comes from: source role data, absorption role data, source reduction role data and status verification role data. The system outputs the abnormal data window, abnormal data category, abnormal judgment basis and abnormal reason prompt.

4. The method for detecting abnormal data in a photovoltaic storage and charging station based on a vulnerable variable observer according to claim 3, characterized in that: Step S1 specifically includes: S1-1. Establishing a data detection window: The system establishes the first data detection window at a fixed time interval. Each data detection window contains [number] data detection windows. There are sampling points, and the time interval between adjacent sampling points is . ,in, Indicates the data detection window number. This indicates the number of sampling points within a data detection window. Indicates the time length between adjacent sampling points; S1-2, Perform window energy conversion on the power data: The system calculates the power data according to the following formula. The first data detection window Window energy corresponding to power-like data: ,in, Indicates the first The first data detection window The window energy corresponding to the power data. Indicates the first The first data detection window The sampling point of the first sampling point Class of power data, Indicates the sampling point number. Corresponding in sequence , , , , , Indicates the available power of photovoltaics. Indicates the vehicle's charging power. Indicates the power absorbed during energy storage charging. Indicates the power input of the absorption branch. Indicates the reduction in photovoltaic power; S1-3, Forming basic data records: The system in the... Records in each data detection window , , , , Energy storage state of charge DC bus voltage maximum value Minimum DC bus voltage and the number of vehicles connected ,in, Indicates the first The state of charge of the energy storage at the start of each data detection window is used to calculate the remaining absorption capacity of the energy storage to absorb excess photovoltaic energy within that data detection window. Indicates the first The maximum DC bus voltage within each data detection window. Indicates the first Minimum DC bus voltage within each data detection window Indicates the first The system calculates and records the actual photovoltaic power connected to the grid at the end of each data detection window, based on the same window energy conversion method as steps S1-2, as the actual photovoltaic power connected to the grid within the k-th data detection window. Used to verify the available energy of photovoltaic power. Excluding solar energy reduction Whether it is consistent with the actual photovoltaic energy input, and is not used as an independent destination vector in the subsequent calculation of the remaining energy destination.

5. The method for detecting abnormal data in a photovoltaic storage and charging station based on a vulnerable variable observer according to claim 4, characterized in that: Step S2 specifically includes: S2-1. Calculating the Remaining Absorption Capacity of Energy Storage: To determine whether energy storage can still play a major absorption role, the system calculates the remaining absorption capacity according to the following formula. Remaining energy absorption capacity within each data detection window: ,in, Indicates the first The remaining energy that can be absorbed within each data detection window. This indicates the preset upper limit of the energy storage state of charge. Indicates the first The energy storage charge state at the start of each data detection window. Indicates the rated capacity of energy storage; S2-2, Determining the surplus state of photovoltaic power: When Greater than At that time, the system determines the first Several data detection windows show a state where the available photovoltaic energy is higher than the vehicle charging energy. Indicates the available energy from photovoltaics. Indicates the vehicle's charging energy; S2-3, Determine the vulnerable condition window: When Greater than ,and Less than At the same time, the number of vehicles connected If no addition is made, the system will [the following will be displayed:] the first [number]. The data detection window was identified as a vulnerable operating condition window, and the number of vehicles connected... "No increase" means that the number of vehicles in the access state at the end of the k-th data detection window is no greater than the number of vehicles in the access state at the end of the (k-1)-th data detection window. This judgment indicates that vehicle charging cannot fully absorb the available photovoltaic energy, and the remaining absorption capacity of energy storage is also insufficient to fully accept the surplus energy. The data from the power station should show clear data manifestations of absorption, photovoltaic reduction, and changes in bus status. Less than This indicates that the remaining absorption capacity of the energy storage is insufficient to independently absorb the remaining photovoltaic energy after the vehicle is charged. When the k-th data detection window does not meet the judgment conditions of the vulnerable operating condition window, the system marks the data detection window as a non-vulnerable operating condition window and proceeds to the acquisition and processing steps of the next data detection window.

6. The method for detecting abnormal data in a photovoltaic storage and charging station based on a vulnerable variable observer according to claim 5, characterized in that: Step S3 specifically includes: S3-1. Determine the energy source role: The system will... The corresponding photovoltaic available data is identified as the energy source role, which is used to represent the first... The energy sources within each data detection window need to be interpreted jointly by vehicle, energy storage, consumption branch, and photovoltaic reduction data; S3-2. Determine the energy absorption role: The system will... The corresponding vehicle charging data is identified as the primary absorption role, and will The corresponding energy storage charging data identifies it as the second absorption role. The corresponding absorption branch input data is determined as the end absorption role. The first absorption role is used to represent the energy actually consumed by the vehicle, the second absorption role is used to represent the energy actually absorbed by the energy storage, and the end absorption role is used to represent the energy destination undertaken by the absorption branch when the vehicle and the energy storage cannot fully absorb the excess energy. S3-3. Determine the source reduction role and status verification role: The system will... The corresponding photovoltaic reduction data is identified as the source-end reduction role, and the DC bus voltage data is identified as the status verification role. S3-4. Forming a role succession sequence: Within the vulnerable operating condition window, the system forms a role succession sequence for vehicle absorption, energy storage absorption, absorption absorption, source reduction, and bus status verification. This role succession sequence is only used for system data consistency verification and does not send control commands to photovoltaic power generation units, energy storage units, charging units, and absorption branches.

7. The method for detecting abnormal data in a photovoltaic storage and charging station based on a vulnerable variable observer according to claim 6, characterized in that: Step S4 specifically includes: S4-1, Calculate the remaining energy destination: The system calculates the energy destination using the following formula: Energy destination and remaining amount within each data detection window: ,in, Indicates the first Energy within a data detection window that has not yet been explained by the legitimate data destination. Indicates the available energy from photovoltaics. Indicates the vehicle's charging energy. This indicates that the energy storage and charging process absorbs energy. This indicates that the absorption branch absorbs energy. This indicates that photovoltaics reduce energy consumption. S4-2. Calculate the bus voltage fluctuation: The system calculates the following formula: Bus voltage fluctuation within each data detection window: ,in, Indicates the first The bus voltage fluctuation within each data detection window. This indicates the maximum DC bus voltage within the detection window. This indicates the minimum DC bus voltage within the detection window; S4-3, Determining the direction of energy in an unclosed window: When At that time, the system determines the first In some data detection windows, the available photovoltaic energy was not fully explained by the legitimate data destinations. This indicates the allowable energy calculation error tolerance, used to exclude minor differences caused by sampling and conversion errors. The judgment result serves as input for subsequent anomaly data category localization. When the system determines that there is no energy out-of-closed phenomenon in the k-th data detection window, it outputs a normal window mark and proceeds to the acquisition and processing steps of the next data detection window.

8. The method for detecting abnormal data in a photovoltaic storage and charging station based on a vulnerable variable observer according to claim 7, characterized in that: Step S5 specifically includes: S5-1, Identifying Missing Destinations: Effective energy destinations refer to the corresponding roles that generate energy exceeding the preset energy tolerance within the current data detection window. The window energy, and that window energy is used as the energy destination for the remaining amount. The deduction term in the calculation participates in the energy closure judgment, when the first... Each data detection window satisfies At that time, the system checks whether the corresponding role has formed an effective energy destination in the order of vehicle absorption, energy storage absorption, consumption absorption, and source reduction. If the vehicle absorption role corresponds to... Below The corresponding role of energy storage and absorption Not greater than the preset energy tolerance Absorbing and absorbing the corresponding role Not greater than the preset energy tolerance Furthermore, the source side reduced the corresponding roles. Not greater than the preset energy tolerance If so, the system determines that the current window has a missing destination role; S5-2, Combining busbar status verification anomaly categories: When the... Each data detection window satisfies ,and Less than the preset voltage fluctuation threshold At that time, the system determined that there was an anomaly in the current data detection window where the energy destination was not closed despite no significant fluctuation in the bus voltage. The anomaly message indicated an inconsistency in the data between available photovoltaic power, vehicle charging power, energy storage charging power, power input of the absorption branch, and photovoltaic power reduction. This represents the threshold used to determine whether the bus voltage has significant fluctuations; S5-3, Identify the category of anomalous data with priority: When ,and ,at the same time and At that time, the system will identify the input data of the absorption branch and the photovoltaic reduction data as the priority categories of abnormal data for verification; when ,and ,at the same time When this happens, the system will prioritize verifying the energy storage state of charge data and energy storage charging power data as the categories of abnormal data; when ,and At that time, the system identifies the available photovoltaic power data, the actual photovoltaic power input data, and the photovoltaic power reduction data as the priority categories of abnormal data for verification. The continuous large-scale absorption by energy storage refers to... This indicates that the energy absorbed during charging has exceeded the sum of the remaining absorption capacity calculated from the energy storage state of charge and the allowable error. Each data detection window satisfies However, if the priority verification conditions for the input data of the absorption branch and the photovoltaic reduction data, the energy storage state of charge data and the energy storage charging power data, and the photovoltaic available power data and the actual photovoltaic power connected in this step are not met, the system will mark the data detection window as having an abnormal comprehensive data consistency and will participate in the verification. Calculated , , , , as well as Output as a review object; S5-4. Output anomaly detection results: The system outputs the... The window number, abnormal data category, energy destination, and remaining amount of each data detection window. Bus voltage fluctuation The abnormality cause prompts include specific reasons among the following: unclosed photovoltaic surplus energy, insufficient energy storage absorption capacity, lack of absorption and consumption, lack of source-end reduction, and bus status mismatch.

Images