A cloud platform-based energy storage soc cooperative calibration method and system

Abstract

The application provides a kind of energy storage SOC collaborative calibration method and system based on cloud platform, and relates to energy storage battery technical field.The application breaks through the restriction of local computing power bottleneck of energy storage equipment on high-precision SOC estimation algorithm by cloud platform to undertake complex computing task, and the cloud platform completes fine calculation based on multi-dimensional correction coefficient, improves the estimation accuracy;At the same time, the cloud platform gathers multi-energy storage station data, dynamically allocates weight based on operating state and carries out weighted fusion, realizes cross-device collaborative calibration, overcomes the inherent defects of single-device independent operation and information fragmentation, makes similar sites as reference to each other, and eliminates individual bias.The application solves the problem of large deviation of current energy storage SOC calibration result, and improves the estimation accuracy of energy storage SOC.

Description

Technical Field

[0001] This invention relates to the field of energy storage battery technology, and in particular to a cloud-based energy storage SOC collaborative calibration method and system. Background Technology

[0002] Energy storage systems are playing an increasingly important role in applications such as power peak shaving, renewable energy consumption, and peak-valley arbitrage on the user side. The state of charge (SOC) of a battery, as a key parameter characterizing the remaining capacity of an energy storage battery, directly impacts the energy management decisions, charge / discharge control strategies, and the safety and economy of the system. Accurate and reliable SOC estimation is a fundamental prerequisite for the efficient utilization and safe operation of energy storage systems.

[0003] Currently, SOC estimation for energy storage systems primarily employs the ampere-hour integration method, open-circuit voltage method, Kalman filtering method, and their improved algorithms. In practical engineering applications, data acquisition, processing, and calibration output are typically performed locally on the energy storage device (i.e., at the battery management system or energy storage converter side). However, energy storage devices usually have limited embedded computing resources, with processor power and storage capacity far inferior to high-performance computing platforms. When SOC estimation requires processing high-frequency sampled data, executing complex filtering algorithms, or constructing multi-order battery equivalent circuit models, the local device's computing resources are insufficient to support the real-time operation of high-precision algorithms. This necessitates the use of simplified models or reduced computation frequency, leading to a decrease in SOC estimation accuracy. Especially under complex operating conditions such as high-power charging and discharging and drastic temperature changes, the error of simplified models increases significantly.

[0004] On the other hand, existing energy storage SOC calibration schemes all operate in a single-device independent mode, with each energy storage device relying solely on its own collected voltage, current, and temperature data for independent estimation and calibration. There is a lack of data exchange and collaboration mechanisms between different energy storage devices, making it impossible to reference and correct their own estimation results using operational data from other devices at the same site or similar devices at neighboring sites. Even within the same energy storage power station, multiple energy storage devices commissioned at the same time, with the same rated power and similar operating conditions, will exhibit inconsistent SOC estimation results due to individual differences and independent calculations, making it difficult for maintenance personnel to determine which device's estimation result is closer to the true value. Due to the lack of cross-device data fusion and collaborative optimization techniques, existing schemes fail to fully leverage the complementary value of multi-source data, thus hindering further improvements in SOC estimation accuracy.

[0005] In summary, existing energy storage SOC calibration methods suffer from large errors in SOC estimation due to limitations in local computing power, independent operation of equipment, and data fragmentation, making it difficult to meet the actual needs of high-precision operation control and safety management of energy storage systems. Summary of the Invention

[0006] This invention provides a cloud platform-based collaborative calibration method and system for energy storage SOC, which solves the problem of large deviations in current energy storage SOC calibration results and improves the estimation accuracy of energy storage SOC.

[0007] In a first aspect, the present invention provides a cloud-based energy storage SOC collaborative calibration method, applied to a collaborative calibration system comprising an energy storage station and a cloud platform. The method includes: acquiring the operating status, real-time operating data, and local SOC estimates of each energy storage station; the operating data including voltage, current, temperature, and charge / discharge data; performing anomaly repair on the real-time operating data of each energy storage station to obtain repaired basic data; calculating the cloud-based SOC estimate of each energy storage station based on the repaired basic data; determining the weighting coefficient of each energy storage station based on its operating status; calculating the cloud-based SOC calibration value based on the cloud-based SOC estimate and weighting coefficient of each energy storage station; and comparing and determining the final SOC value of each energy storage station based on its local SOC estimate and the cloud-based SOC calibration value.

[0008] In a second aspect, embodiments of the present invention provide an electronic device including a memory and a processor. The memory stores a computer program, and the processor is configured to call and run the computer program stored in the memory to perform the steps of the method as described in the first aspect and any possible implementation thereof.

[0009] Thirdly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, it implements the steps of the method as described in the first aspect and any possible implementation thereof.

[0010] This invention provides a cloud-based collaborative calibration method and system for energy storage SOC. By centralizing complex computational tasks through a cloud platform, this invention overcomes the limitation of local computing power bottlenecks in energy storage devices on high-precision SOC estimation algorithms. The cloud platform performs fine calculations based on multiple maintenance correction coefficients, improving estimation accuracy. Simultaneously, the cloud platform aggregates data from multiple energy storage stations, dynamically assigns weights based on operating status, and performs weighted fusion to achieve cross-device collaborative calibration. This overcomes the inherent defects of independent operation and information fragmentation of single devices, allowing similar sites to serve as references and eliminating individual biases. This invention solves the problem of large deviations in current energy storage SOC calibration results and improves the estimation accuracy of energy storage SOC. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a schematic flowchart of a cloud-based energy storage SOC collaborative calibration method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0013] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0014] To address the issue of large deviations in current energy storage SOC calibration results, this invention provides a cloud-based collaborative calibration system for energy storage SOCs. This system achieves accurate SOC calibration through a three-layer cloud-edge-device architecture, overcoming the limitations of computing power and insufficient accuracy inherent in traditional local calibration methods.

[0015] The collaborative calibration system provided by this invention includes a three-layer architecture: an edge data acquisition layer, a cloud platform computing layer, and a data fusion layer.

[0016] End-side data acquisition layer: The energy storage device locally collects basic data (voltage, current, temperature, charge / discharge status, etc.), performs preliminary screening, and then sends it to the cloud platform via the IEMS system or directly. The end-side retains a simplified local calibration algorithm to ensure basic operational requirements.

[0017] Cloud platform computing layer: Receives data from multiple energy storage sites, performs multi-site collaborative computing, data cleaning, and model optimization to generate high-precision SOC calibration results. The core components include a multi-site collaborative computing module, a data cleaning module, and an SOC calibration model.

[0018] Data fusion layer: The IEMS system receives calibration results from the cloud platform, compares and dynamically fuses them with local calculation results, and selects the final SOC value or performs weighted fusion based on preset conditions.

[0019] like Figure 1As shown, the present invention provides a cloud platform-based collaborative calibration method for energy storage SOCs, which is applied to a collaborative calibration system, including an energy storage station and a cloud platform; the method includes steps S101-S105.

[0020] S101. Obtain the operating status, real-time operating data, and local SOC estimate of each energy storage station.

[0021] In some embodiments, the operating data includes voltage, current, temperature, and charge / discharge data; For example, the energy storage stations are those commissioned at the same time and with equal rated power. Since energy storage stations commissioned at the same time have similar health conditions and degradation characteristics in terms of commissioning time, battery aging, and cycle life consumption, and equal rated power ensures that the depth of charge and discharge, current rate, and thermal characteristics of each station are basically consistent under the same operating conditions, thus providing a unified and comparable benchmark for the SOC estimation of each energy storage station. Based on this, when the cloud platform performs collaborative SOC calculation and weighted fusion for each station, it can effectively eliminate systematic biases caused by differences in battery model, commissioning time, and aging degree, ensuring the rationality of multi-site data fusion and the reliability of calibration results.

[0022] For example, the energy storage station collects operational data in real time through the end-side data acquisition layer. Specifically, the energy storage device locally collects basic data such as voltage, current, and temperature of each battery cell or battery module through the battery management system, while also recording the current charging and discharging status (including charging, discharging, and resting states). After initial screening, the collected data is sent to the cloud platform either through the IEMS system or directly.

[0023] For example, the method for calculating the local SOC estimate includes steps A1-A5.

[0024] A1. For any energy storage station, based on the temperature and open-circuit voltage of the energy storage station, query the temperature-related open-circuit voltage-state-of-charge two-dimensional curve to determine the reference SOC of the energy storage station.

[0025] For example, when the energy storage device is in a static state, it measures the open-circuit voltage of the battery and, using a temperature-open-circuit voltage-SOC two-dimensional mapping table or fitted curve pre-stored in local memory, obtains the reference SOC value at the current temperature. This two-dimensional curve is obtained through laboratory calibration and covers the correspondence between open-circuit voltage and SOC at different temperature points (e.g., -20℃, -10℃, 0℃, 10℃, 25℃, 40℃, 55℃, etc.).

[0026] A2. Based on the temperature of the energy storage station, determine the temperature correction coefficient of the energy storage station.

[0027] For example, a temperature correction factor is used to compensate for the impact of temperature on battery capacity and voltage characteristics. A preset temperature correction factor table, obtained by fitting experimental data, can be consulted based on the current temperature. For instance, the correction factor is 1.0 at a reference temperature of 25°C, less than 1.0 at low temperatures, and greater than 1.0 at high temperatures.

[0028] A3. Based on the current of the energy storage station, determine the current ratio correction factor of the energy storage station.

[0029] For example, the current rate correction factor is used to compensate for the impact of different charge / discharge rates on SOC estimation. A preset current rate correction factor table can be consulted based on the ratio of the current charge / discharge current to the battery's rated capacity (i.e., the charge / discharge rate). For instance, the correction factor is 1.0 at 0.5C, 0.98 at 1C, and 0.95 at 2C.

[0030] A4. Based on the rated battery capacity, temperature correction factor, current rate correction factor, and current of the energy storage station, the cumulative integral is performed to obtain the ampere-hour integral correction value of the energy storage station.

[0031] A5. Based on the baseline SOC and the integral correction value of the ampere-hour for the energy storage station, determine the estimated local SOC value for the energy storage station.

[0032] S102. Perform anomaly repair on the real-time operation data of each energy storage station to obtain the repaired basic data.

[0033] As one possible implementation, step S102 can be specifically implemented as steps S1021-S1025.

[0034] S1021. Based on the real-time operating data of each energy storage station, calculate the mean and standard deviation of each operating data.

[0035] In some embodiments, after receiving real-time operating data uploaded by each energy storage station, the cloud platform calculates the mean and standard deviation for each parameter type, such as voltage, current, and temperature. The mean and standard deviation can be calculated using a sliding window method, and the window size can be set according to the data sampling frequency and calculation accuracy requirements. For example, the mean and standard deviation can be calculated using data from the most recent N sampling points.

[0036] It should be noted that during equipment initialization, the initial mean μ and standard deviation σ are calculated in advance based on historical normal operating data under ideal conditions (e.g., fault-free voltage and current data for 1-2 consecutive weeks under the same season and charge / discharge rate). Then, during power plant operation, the mean μ and standard deviation σ are dynamically updated as new data is continuously collected. The mean μ and standard deviation σ for temperature, voltage, and current are all different values.

[0037] S1022. Based on the mean and standard deviation of each operating data, anomaly detection is performed on the real-time operating data of each energy storage station to obtain abnormal data in the real-time operating data.

[0038] For example, this invention is based on the improved 3σ principle and dynamically adjusts the threshold according to battery characteristics. For voltage data, the range [μ-2σ, μ+2σ] is used as the normal data range; for current and temperature data, the range [μ-3σ, μ+3σ] is used as the normal data range. When a data point exceeds the corresponding normal data range, that data point is determined to be abnormal data.

[0039] S1023. Calculate the degree of deviation of abnormal data.

[0040] For example, the deviation can be calculated according to the following formula in the embodiments of the present invention.

[0041] Degree of deviation = |X - μ| / σ; Where X is the actual value of the outlier, μ is the mean of the outlier, and σ is the standard deviation of the outlier.

[0042] S1024. Based on the degree of deviation of the abnormal data, the abnormal data is repaired to obtain the repaired basic data.

[0043] For example, for any abnormal data, the embodiments of the present invention can achieve repair through steps B1-B5.

[0044] B1. If the deviation of the abnormal data is less than the first deviation threshold, the abnormal data is classified as a minor abnormality, and the average value of the historical data within the sliding window in which the abnormal data is located is used to correct the abnormal data to obtain the corrected data.

[0045] In this embodiment of the invention, the first deviation threshold is preferably 3 (corresponding to voltage data) or 4.5 (corresponding to current and temperature data). That is, when the deviation of the voltage data is less than 3, or the deviation of the current / temperature data is less than 4.5, it is determined to be a slight abnormality. The correction method is: taking the abnormal data point as the center, take the average value of several normal data points before and after it (for example, 5 each) to replace the abnormal data.

[0046] B2. If the deviation of the abnormal data is greater than or equal to the first deviation threshold and less than the second deviation threshold, the abnormal data is classified as moderately abnormal, and interpolation data is obtained by forward single-step linear extrapolation. The difference data is used to replace the abnormal data to obtain the repaired data.

[0047] In this embodiment of the invention, the second deviation threshold is preferably 6 (corresponding to voltage data) or 9 (corresponding to current and temperature data). That is, when the deviation of the voltage data is between 3 and 6, or the deviation of the current / temperature data is between 4.5 and 9, it is determined to be a moderate anomaly. The repair method is to use several normal data points preceding the abnormal data point for linear extrapolation to predict the data value at the current moment, and replace the abnormal data with the predicted value. For example, linear extrapolation is performed using the first two normal data points.

[0048] B3. If the deviation of the abnormal data is greater than or equal to the second deviation threshold, the abnormal data is classified as a severe anomaly, marked as invalid data, and replaced with the historical data of the same period corresponding to the abnormal data to obtain the repaired data.

[0049] In this embodiment of the invention, a severe anomaly is identified when the voltage data deviation is greater than or equal to 6, or the current / temperature data deviation is greater than or equal to 9. The repair method is to replace the abnormal data with historical data from the same period (e.g., the same time the previous day, the same time the previous week, etc.). This historical data refers to data from the same or similar states during historical charging and discharging processes. For example, values ​​with equal or adjacent SOC values.

[0050] B4. Perform cross-validation on the repaired data to obtain the cross-validation results.

[0051] For example, cross-validation refers to verifying the consistency of parameters such as voltage, current, and temperature. For instance, during charging, the voltage should gradually increase with charging time, and the current should be positive and within a certain range; during discharging, the voltage should gradually decrease with discharging time, and the current should be negative and within a certain range. If the trend of a certain parameter does not match the trends of other parameters, the data is deemed unreasonable.

[0052] B5. If there is unreasonable data in the cross-validation results, the unreasonable data will be corrected a second time to obtain the corrected basic data.

[0053] For example, unreasonable data discovered by cross-validation is corrected again according to the three-level processing strategy described above until all data passes cross-validation.

[0054] S103. Based on the repaired basic data, calculate the cloud-based SOC estimate for each energy storage station.

[0055] As one possible implementation, step S103 can be specifically implemented as steps S1031-S1035.

[0056] S1031. Based on the temperature and open-circuit voltage of each energy storage station in the repaired basic data, query the temperature-related open-circuit voltage-state-of-charge two-dimensional curve preset in the cloud platform to determine the reference SOC of each energy storage station.

[0057] For example, the cloud platform stores a finely calibrated temperature-open-circuit voltage-SOC three-dimensional mapping table, which has higher accuracy and more calibration points compared to the locally stored version. Based on the received temperature and open-circuit voltage data, the cloud platform uses an interpolation algorithm to look up the table and obtain the accurate baseline SOC value.

[0058] S1032. Based on the temperature of each energy storage station, determine the temperature correction coefficient for each energy storage station.

[0059] For example, the cloud platform queries a preset temperature correction factor table based on the temperature data reported by each energy storage station to determine the temperature correction factor for each station. This temperature correction factor table has also been finely calibrated and has higher accuracy than the local version.

[0060] S1033. Based on the current of each energy storage station, determine the current ratio correction coefficient for each energy storage station.

[0061] For example, the cloud platform calculates the charge / discharge rate based on the current data reported by each energy storage station, and queries a preset current rate correction coefficient table to determine the current rate correction coefficient for each energy storage station.

[0062] S1034. Based on the rated battery capacity, temperature correction coefficient, current rate correction coefficient, and current of each energy storage station, perform cumulative integration to obtain the ampere-hour integral correction value of each energy storage station.

[0063] For example, the cloud platform calculates the ampere-hour integral correction value more frequently and with higher accuracy, and can utilize more historical data for smoothing.

[0064] S1035. Based on the baseline SOC and ampere-hour integral correction value of each energy storage station, determine the cloud SOC estimate value of each energy storage station.

[0065] As another possible implementation, embodiments of the present invention can calculate the cloud-based SOC estimate of each energy storage station using the following formula.

[0066] ; ; in, The SOC is obtained by interpolation from a table using the temperature-dependent OCV-SOC two-dimensional curve; for the same open-circuit voltage, the corresponding SOC value is different at different temperatures; OCV is the battery open-circuit voltage. Temperature correction factor; at low temperatures The actual usable capacity of the battery decreases, and the SOC corresponding to the same ampere-hours changes even more; during operation, it is all obtained by looking up tables and linear interpolation, which comes from previous experimental calculations. : Current ratio correction factor. Obtained entirely through table lookup and linear interpolation during operation, derived from prior experimental calculations. Rated capacity of the battery. : ampere-hour integral → from the starting time Up to the current t, the cumulative amount of electricity flowing through the battery. : Charge and discharge current that changes in real time. SOC uses decimal arithmetic from 0 to 1, and is multiplied by 100 to convert to a percentage when displayed externally. This is the cloud-based SOC estimate at the current time t. OCV is the battery open-circuit voltage (unit: V), which refers to the battery's terminal voltage in a quiescent state (no charging or discharging current or current approaching zero), and has a definite mapping relationship with the current SOC and temperature. This is the OCV-SOC-T mapping function, representing the correspondence between open-circuit voltage, temperature, and SOC. It is obtained through laboratory calibration experiments and is pre-stored in the cloud platform in the form of a two-dimensional lookup table or fitted curve. T is the current battery temperature. The starting point for SOC estimation is typically the time when the last SOC calibration was performed or when the baseline SOC value was obtained after the battery had been left to rest.

[0067] S104. Based on the operating status of each energy storage station, determine the weighting coefficient of each energy storage station.

[0068] As one possible implementation, step S104 can be specifically implemented as steps S1041-S1042.

[0069] S1041. Based on the operating status of each energy storage station, query the correspondence between the preset operating status and the original weight range, and determine the original weight value of each energy storage station.

[0070] In some embodiments, the operating states include standby state, constant power charge / discharge state, variable power charge / discharge state, and high power charge / discharge state. The correspondence between the operating states and the original weight range is shown in Table 1.

[0071] Table 1 Among them, the battery is stable in standby mode, and the data is the most reliable, so it has the highest weight; the state changes smoothly in constant power charge and discharge mode, and the data is relatively reliable; the state changes frequently in variable power charge and discharge mode, and the data fluctuates greatly; the polarization effect is obvious in high power charge and discharge mode, and the data is the least reliable.

[0072] S1042. Normalize the original weight values ​​of each energy storage station to determine the weight coefficient of each energy storage station.

[0073] For example, suppose there are n energy storage stations participating in collaborative computing, and the original weight value of the i-th energy storage station is w'. i Then the normalized weight coefficients are: w i =w' i / Σ(w' j ), j=1,2,...,n; After normalization, Σw is satisfied i =1.

[0074] For example, suppose there are four energy storage stations in standby mode, constant power charging / discharging mode, variable power charging / discharging mode, and high power charging / discharging mode, respectively, with their original weight values ​​of 0.9, 0.6, 0.35, and 0.15, respectively. Then the normalized weight coefficients are as follows: w1=0.9 / (0.9+0.6+0.35+0.15)=0.45; w2=0.6 / (0.9+0.6+0.35+0.15)=0.30; w3=0.35 / (0.9+0.6+0.35+0.15)=0.175; w4=0.15 / (0.9+0.6+0.35+0.15)=0.075.

[0075] The normalized weights w1+w2+w3+w4=0.45+0.3+0.175+0.075=1.

[0076] S105. Calculate the cloud SOC calibration value based on the cloud SOC estimate and weighting coefficient of each energy storage station.

[0077] As one possible implementation, embodiments of the present invention can perform weighted summation based on the cloud-based SOC estimates and weighting coefficients of each energy storage station to obtain the cloud-based SOC calibration value.

[0078] For example, the cloud-based SOC calibration value = Σ(w i ×SOC i ), where SOC i Let w be the cloud-based SOC estimate for the i-th energy storage station. i is the weighting coefficient for the i-th energy storage station.

[0079] S106. Based on the local SOC estimate of each energy storage station and the cloud SOC calibration value, compare and determine the final SOC value of each energy storage station.

[0080] As one possible implementation, step S106 can be specifically implemented as steps S1061-S1064.

[0081] S1061. For any energy storage station, calculate the absolute value of the difference between the local SOC estimate and the cloud-based SOC calibration value.

[0082] For example, the absolute value of the difference Δ = |SOC 本地 -SOC 云端 |. Among them, SOC 本地 This is a local SOC estimate. 云端 This is the cloud-based SOC calibration value.

[0083] S1062. If the absolute value of the difference is less than or equal to the first judgment threshold, the cloud SOC calibration value is determined as the final SOC value of the energy storage station.

[0084] In this embodiment of the invention, the default value of the first determination threshold a is 2%. When Δ≤2%, it indicates that the local estimated value is very close to the cloud calibration value, and the cloud calibration value has higher accuracy. Therefore, the cloud calibration value is directly used as the final SOC value.

[0085] S1063. If the absolute value of the difference is greater than the first judgment threshold and less than or equal to the second judgment threshold, then a weighted fusion is performed based on the local SOC estimate and cloud SOC calibration value of the energy storage station, as well as the first weight combination, and the weighted fusion result is determined as the final SOC value of the energy storage station.

[0086] In this embodiment of the invention, the default value of the second determination threshold b is 5%. When 2% < Δ ≤ 5%, the first weight combination is used for weighted fusion: SOC 最终 =0.7×SOC 云端 +0.3×SOC 本地 Among them, SOC 最终 SOC final value S1064. If the absolute value of the difference is greater than the second judgment threshold, the abnormal reporting mechanism is triggered, and the local SOC estimate, cloud SOC calibration value, and the second weight combination of the energy storage station are weighted and fused, and the fusion result is used as the final SOC value of the energy storage station.

[0087] In the first weight combination, the weight of the local SOC estimate is less than the weight of the local SOC estimate in the second weight combination.

[0088] Specifically, when Δ > 5%, an anomaly reporting mechanism is triggered, and a second weighted combination is used for weighted fusion: SOC 最终 =0.6×SOC 云端 +0.4×SOC 本地At the same time, the system records this abnormality report event.

[0089] This invention provides a cloud-based collaborative calibration method for the State of Charge (SOC) of energy storage. By centralizing complex computational tasks on the cloud platform, it overcomes the limitation of local computing power bottlenecks in energy storage devices on high-precision SOC estimation algorithms. The cloud platform performs fine calculations based on multiple maintenance correction coefficients, improving estimation accuracy. Simultaneously, the cloud platform aggregates data from multiple energy storage stations, dynamically assigns weights based on operating status, and performs weighted fusion to achieve cross-device collaborative calibration. This overcomes the inherent defects of independent operation and information fragmentation of single devices, allowing similar sites to serve as references and eliminating individual biases. This invention solves the problem of large deviations in current energy storage SOC calibration results and improves the estimation accuracy of energy storage SOC.

[0090] Optionally, the cloud-based energy storage SOC collaborative calibration method provided in this embodiment of the invention further includes steps S201-S205 after step S106.

[0091] S201. Record the number of consecutive triggers of the anomaly reporting mechanism for each energy storage station.

[0092] For example, the cloud platform maintains an anomaly trigger counter for each energy storage station, and the counter is incremented by 1 each time the anomaly reporting mechanism is triggered.

[0093] It should be noted that the State of Charge (SOC) is calculated continuously and periodically during charging and discharging. For example, if the previous SOC calculation triggered an anomaly report, the counter increments by 1; if the next SOC calculation does not trigger an anomaly report, the counter will be reset to 0. Similarly, when charging or discharging stops, the counter will also be reset to 0 after SOC calculation ceases; there is no need to preset a time threshold to reset it to 0.

[0094] S202. If an energy storage station triggers the abnormal reporting mechanism only once and does not trigger it continuously, it is determined to be a calculation error caused by a sudden change in data, and the normal operation of the energy storage station is maintained.

[0095] For example, if an energy storage station only triggers an anomaly report once, and does not trigger it continuously in a short period of time, it may be due to a calculation error or data distortion at the device end caused by a sudden change in data such as voltage and current, rather than a fault in the device itself. Therefore, no special handling is required, and normal operation is maintained.

[0096] S203. If the number of times a certain energy storage station triggers the abnormal reporting mechanism reaches a preset threshold, the energy storage station will be marked as an abnormal device to be inspected, and a maintenance prompt message will be sent to the operation and maintenance terminal.

[0097] In this embodiment of the invention, the preset threshold for the number of times is 3. If an energy storage station triggers the anomaly reporting mechanism three times consecutively, it indicates that the device may have a persistent measurement error or hardware failure. The cloud platform marks the device as an abnormal device to be inspected and sends maintenance prompts to the operation and maintenance personnel via SMS, APP push, or other means.

[0098] S204. Before the abnormal equipment to be inspected is repaired, the first judgment threshold and the second judgment threshold are dynamically adjusted. Each time the abnormal reporting mechanism is triggered, the first judgment threshold and the second judgment threshold are increased by a preset ratio, and the first judgment threshold does not exceed the first upper limit value, and the second judgment threshold does not exceed the second upper limit value.

[0099] Specifically, after a device is marked as an abnormal device awaiting inspection until maintenance is completed, for each additional trigger of the abnormal reporting mechanism, the first judgment threshold 'a' and the second judgment threshold 'b' increase by 5%. However, the increase in 'a' shall not exceed 5% (i.e., 'a' ≤ 5%), and the increase in 'b' shall not exceed 12% (i.e., 'b' ≤ 12%). The purpose of this is to appropriately relax the fusion judgment conditions in cases where the device may be faulty, avoiding frequent triggering of abnormal reporting due to data deviations of the device itself, while still maintaining a certain level of sensitivity in abnormal detection.

[0100] S205. After the abnormal equipment to be inspected has been repaired, the first judgment threshold and the second judgment threshold are restored to their default values.

[0101] For example, after the maintenance personnel complete the equipment inspection and confirm it in the system, the cloud platform resets the abnormal trigger counter of the energy storage station to 0, restores the first judgment threshold a to the default value of 2%, and restores the second judgment threshold b to the default value of 5%.

[0102] Thus, this invention, through an anomaly reporting count and continuous trigger judgment mechanism, can effectively distinguish between sporadic data disturbances and persistent equipment failures, avoiding false alarms from interfering with operation and maintenance. When three consecutive triggers occur, the abnormal equipment is promptly marked and maintenance is notified, ensuring system security. Dynamically increasing the judgment threshold before maintenance reduces invalid reports during fault periods, while setting an upper limit prevents excessive leniency. After maintenance, the default value is restored, balancing anomaly detection sensitivity and system operational stability. This improves fault tolerance and operational efficiency.

[0103] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0104] Figure 2This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention; the electronic device 200 includes: a processor 201, a memory 202, and a computer program 203 stored in the memory 202 and executable on the processor 201; when the processor 201 executes the computer program 203, it implements the steps in the above method embodiments; or, when the processor 201 executes the computer program 203, it implements the functions of each module / unit in the above device embodiments; For example, the computer program 203 may be divided into one or more modules / units, one or more modules / units are stored in the memory 202 and executed by the processor 201 to complete the present invention; one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, which are used to describe the execution process of the computer program 203 in the electronic device 200.

[0105] The processor 201 may be a central processing unit, or it may be other general-purpose processors, digital signal processors, application-specific integrated circuits, field-programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.; the general-purpose processor may be a microprocessor or any conventional processor, etc. The memory 202 can be an internal storage unit of the electronic device 200, such as a hard disk or RAM of the electronic device 200; the memory 202 can also be an external storage device of the electronic device 200, such as a plug-in hard disk, smart memory card, secure digital card, flash memory card, etc., equipped on the electronic device 200; furthermore, the memory 202 can include both internal storage units and external storage devices of the electronic device 200. The memory 202 is used to store computer programs and other programs and data required by the terminal; the memory 202 can also be used to temporarily store data that has been output or will be output.

[0106] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A cloud-based collaborative calibration method for energy storage SOCs, characterized in that, The method is applied to a collaborative calibration system, which includes an energy storage station and a cloud platform; the method includes: The system acquires the operating status, real-time operating data, and local SOC estimate of each energy storage station. The operating data includes voltage, current, temperature, and charge / discharge data. Anomalies were corrected in the real-time operating data of each energy storage station to obtain the corrected basic data; Based on the repaired basic data, the cloud-based SOC estimate of each energy storage station is calculated; Based on the operating status of each energy storage station, the weighting coefficient of each energy storage station is determined. Based on the cloud SOC estimate and weighting coefficient of each energy storage station, the cloud SOC calibration value is calculated; Based on the local SOC estimate of each energy storage station and the cloud-based SOC calibration value, a comparison is made to determine the final SOC value of each energy storage station.

2. The cloud-based energy storage SOC collaborative calibration method according to claim 1, characterized in that, The process of anomaly repairing the real-time operating data of each energy storage station to obtain repaired basic data includes: Based on the real-time operating data of each energy storage station, the mean and standard deviation of each operating data are calculated; Based on the mean and standard deviation of each operational data, anomaly detection is performed on the real-time operational data of each energy storage station to obtain abnormal data in the real-time operational data. Calculate the degree of deviation of the abnormal data; Based on the degree of deviation of the abnormal data, the abnormal data is repaired to obtain the repaired basic data.

3. The cloud-based energy storage SOC collaborative calibration method according to claim 1, characterized in that, The abnormal data is repaired based on the degree of deviation to obtain the repaired basic data, including: For any abnormal data, if the deviation of the abnormal data is less than the first deviation threshold, the abnormal data is classified as a minor abnormality, and the average value of the historical data within the sliding window in which the abnormal data is located is used to correct the abnormal data to obtain the corrected data. If the deviation of the abnormal data is greater than or equal to the first deviation threshold and less than the second deviation threshold, the abnormal data is classified as moderately abnormal, and interpolation data is obtained by forward single-step linear extrapolation. The difference data is then used to replace the abnormal data to obtain the repaired data. If the deviation of the abnormal data is greater than or equal to the second deviation threshold, the abnormal data is classified as a severe anomaly, marked as invalid data, and replaced with the historical data of the same period corresponding to the abnormal data to obtain the repaired data. Cross-validation was performed on the repaired data to obtain the cross-validation results; If the cross-validation results contain unreasonable data, the unreasonable data will be corrected a second time to obtain the corrected basic data.

4. The cloud-based energy storage SOC collaborative calibration method according to claim 1, characterized in that, Based on the repaired basic data, the cloud-based SOC estimate for each energy storage station is calculated, including: Based on the temperature and open-circuit voltage of each energy storage station in the repaired basic data, the temperature-related open-circuit voltage-state of charge two-dimensional curves preset in the cloud platform are queried to determine the baseline SOC of each energy storage station. Based on the temperature of each energy storage station, determine the temperature correction coefficient for each energy storage station; Based on the current of each energy storage station, determine the current ratio correction factor for each energy storage station; Based on the rated battery capacity, temperature correction factor, current rate correction factor of each energy storage station, and the current of each energy storage station, the cumulative integral is performed to obtain the ampere-hour integral correction value of each energy storage station. Based on the baseline SOC and ampere-hour integral correction value of each energy storage station, the cloud SOC estimate value of each energy storage station is determined.

5. The cloud-based energy storage SOC collaborative calibration method according to claim 1, characterized in that, The determination of the weighting coefficient for each energy storage station based on its operating status includes: Based on the operating status of each energy storage station, the correspondence between the preset operating status and the original weight range is queried to determine the original weight value of each energy storage station; the operating status includes standby status, constant power charging and discharging status, variable power charging and discharging status and high power charging and discharging status. The original weight values ​​of each energy storage station are normalized to determine the weight coefficient of each energy storage station.

6. The cloud-based energy storage SOC collaborative calibration method according to claim 1, characterized in that, The calculation of the cloud-based SOC calibration value based on the cloud-based SOC estimate and weighting coefficient of each energy storage station includes: Based on the cloud-based SOC estimates and weighting coefficients of each energy storage station, a weighted sum is performed to obtain the cloud-based SOC calibration value.

7. The cloud-based energy storage SOC collaborative calibration method according to claim 1, characterized in that, The process of comparing and determining the final SOC value of each energy storage station based on the local SOC estimate of each station and the cloud-based SOC calibration value includes: For any energy storage station, calculate the absolute value of the difference between the local SOC estimate of the energy storage station and the cloud-based SOC calibration value; If the absolute value of the difference is less than or equal to the first determination threshold, then the cloud SOC calibration value is determined as the final SOC value of the energy storage station. If the absolute value of the difference is greater than the first determination threshold and less than or equal to the second determination threshold, then a weighted fusion is performed based on the local SOC estimate of the energy storage station, the cloud SOC calibration value, and the first weight combination, and the weighted fusion result is determined as the final SOC value of the energy storage station. If the absolute value of the difference is greater than the second judgment threshold, an anomaly reporting mechanism is triggered, and the local SOC estimate of the energy storage station, the cloud SOC calibration value, and the second weight combination are weighted and fused, and the fusion result is used as the final SOC value of the energy storage station. In the first weight combination, the weight of the local SOC estimate is less than the weight of the local SOC estimate in the second weight combination.

8. The cloud-based energy storage SOC collaborative calibration method according to claim 1, characterized in that, The method for calculating the local SOC estimate includes: For any energy storage station, based on the station's temperature and open-circuit voltage, query the temperature-related open-circuit voltage-state-of-charge two-dimensional curve to determine the station's baseline SOC. Based on the temperature of the energy storage station, determine the temperature correction factor for the energy storage station; Based on the current of the energy storage station, determine the current ratio correction factor of the energy storage station; Based on the rated battery capacity, temperature correction factor, current rate correction factor, and current of the energy storage station, the ampere-hour integral correction value of the energy storage station is obtained by cumulative integration. Based on the baseline SOC and the ampere-hour integral correction value of the energy storage station, the local SOC estimate of the energy storage station is determined.

9. The cloud-based energy storage SOC collaborative calibration method according to claim 1, characterized in that, After comparing and determining the final SOC value of each energy storage station based on the local SOC estimate and the cloud-based SOC calibration value, the method further includes: Record the number of consecutive triggers of the anomaly reporting mechanism at each energy storage station; If an energy storage station triggers the anomaly reporting mechanism only once and not continuously, it is determined to be a calculation error caused by a sudden change in data, and the normal operation of the energy storage station is maintained. If the number of times a certain energy storage station triggers the abnormal reporting mechanism reaches the preset threshold, the energy storage station will be marked as an abnormal device to be inspected, and a maintenance prompt message will be sent to the operation and maintenance terminal. Before the abnormal equipment to be inspected is repaired, the first judgment threshold and the second judgment threshold are dynamically adjusted. Each time the abnormal reporting mechanism is triggered, the first judgment threshold and the second judgment threshold are increased by a preset ratio, and the first judgment threshold does not exceed the first upper limit value and the second judgment threshold does not exceed the second upper limit value. After the abnormal equipment to be inspected has been repaired, the first judgment threshold and the second judgment threshold are restored to their default values.

10. A cloud-based energy storage SOC collaborative calibration system, characterized in that, The collaborative calibration system includes an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor is used to call and run the computer program stored in the memory to perform the method as described in any one of claims 1 to 9.

Images