A data intelligent analysis system and method based on source-grid-load-storage integration

Abstract

The present invention discloses a data intelligent analysis system and method based on the integration of source, grid, load and storage, which relates to the technical field of power analysis. The present invention counts the historical average power consumption in a certain time period and determines the characteristic time period; sets a training cycle, collects the power consumption of the load end at the sampling time point, and calculates the power consumption fluctuation rate of the load end in the characteristic time period; draws the power consumption fluctuation diagram of the load end in the characteristic time period within the training cycle, and determines the target load end; generates an energy storage end group, collects the operating parameters of the energy storage end group, calculates the operating score of the energy storage end group, and draws the characteristic operating score diagram of the analysis period corresponding to the energy storage end group; obtains the real-time operating score of the energy storage end group, calculates the real-time analysis period of the energy storage end group, and judges the status of the abnormal analysis of the energy storage end group to ensure the efficient and stable operation of the source, grid, load and storage integrated power system.

Description

Technical Field

[0001] The present invention relates to the field of power analysis technology, and in particular to a data intelligent analysis system and method based on source-grid-load-storage integration. Background Art

[0002] In recent years, with the large-scale integration of renewable energy and the diversification of power loads, the operating environment of power systems has become increasingly complex. Traditional power dispatching and operation and maintenance models are facing unprecedented challenges. Under the trend of integrated power generation, grid, load and storage, the uncertainty and volatility of system operation have increased significantly, and the demand for abnormal monitoring has also increased significantly.

[0003] Traditional anomaly analysis mechanisms generally use a fixed-cycle offline evaluation model. This analysis frequency is severely mismatched with the actual system fluctuation intensity. In high-volatility scenarios such as sudden load changes and sudden drops in renewable energy output, the fixed cycle can easily lead to analysis lags and anomaly propagation paths that cannot be captured in a timely manner. Furthermore, in low-volatility, stable operating conditions, frequent analysis can lead to redundant computing resources.

[0004] In this context, how to make full use of the massive data generated by the integrated power system of source, grid, load and storage, and combine it with data analysis methods to achieve adaptive optimization of intensive analysis in high-risk periods and sparse monitoring in low-risk periods, can reduce downtime caused by faults, improve the reliability and resource utilization of the power grid, and ensure the efficient and stable operation of the integrated power system of source, grid, load and storage. Summary of the Invention

[0005] The purpose of the present invention is to provide a data intelligent analysis system and method based on source-grid-load-storage integration to solve the problems raised in the prior art.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a data intelligent analysis method based on source-grid-load-storage integration, the method comprising:

[0007] Step S100: Obtain historical electricity consumption records of the load side in the source-grid-load-storage integrated power system, divide a day into several time periods, calculate the historical average electricity consumption of a certain time period, and determine a characteristic time period based on the historical average electricity demand;

[0008] Step S200: setting a training cycle, collecting the power consumption of the load end at a sampling time point, and calculating the power consumption fluctuation rate of the load end in a characteristic time period;

[0009] Step S300: drawing a power consumption fluctuation graph of the load end in a characteristic time period during a training cycle, and counting abnormal adjacent dates based on the slope of the power consumption fluctuation graph to determine the target load end;

[0010] Step S400: Generate an energy storage terminal group based on the target load terminal, collect operating parameters of the energy storage terminal group, calculate the operating score of the energy storage terminal group, and draw a characteristic operating score diagram of the analysis period corresponding to the energy storage terminal group;

[0011] Step S500: obtaining a real-time operation score of the energy storage terminal group, calculating a real-time analysis period of the energy storage terminal group, and determining a status of abnormal analysis of the energy storage terminal group.

[0012] Furthermore, step S100 includes:

[0013] Step S101: obtaining historical power consumption records of a certain load terminal, summarizing the historical power consumption of the load terminal in a certain time period, and calculating the historical average power consumption of the load terminal in the certain time period;

[0014] Step S102: Summarize the historical average power consumption of all load terminals in all time periods, calculate the mean of the historical average power consumption to be A and the standard deviation to be B, and calculate the threshold of the historical average power consumption to be C=A+B*K, where C represents the threshold of the historical average power consumption and K represents the preset value;

[0015] Step S103: comparing the historical average power consumption of a load terminal in a certain time period with a threshold value of the historical average power consumption; if the historical average power consumption of the load terminal is higher than the threshold value of the historical average power consumption, setting the load terminal as an abnormal load terminal;

[0016] Step S104: Counting the number of abnormal load terminals in a certain time period, setting a threshold for the number of abnormal load terminals, comparing the number of abnormal load terminals in the time period with the threshold for the number of abnormal load terminals, and if the number exceeds the threshold for the number of normal load terminals, setting the time period as a characteristic time period;

[0017] By calculating the mean and standard deviation of historical average power consumption and setting thresholds, abnormal loads can be effectively screened out to avoid misjudgment due to single high power consumption.

[0018] Using standard deviation as an adjustment factor makes the threshold dynamically adaptable and can adapt to the power consumption characteristics of different load ends, thereby improving the flexibility and applicability of anomaly detection;

[0019] Counting the number of abnormal load terminals and comparing them with the threshold can effectively determine whether there is a concentrated outbreak of abnormalities, helping operation and maintenance personnel quickly identify critical time periods and improve fault early warning capabilities;

[0020] Use historical data for statistical analysis instead of single-point data judgment to reduce false positives caused by accidental factors and improve the accuracy and reliability of anomaly detection.

[0021] Furthermore, step S200 includes:

[0022] Step S201: Select a number of consecutive days as a training cycle, obtain a time range corresponding to a certain characteristic time period, and preset a number of sampling time points within the time range;

[0023] Step S202: Collect the power consumption of a load end at a certain sampling time point, summarize the power consumption of the load end at all sampling time points within a certain characteristic time period, and mark the sampling time points in order to obtain the power consumption set of all sampling time points of the load end within the certain characteristic time period as D={d1, d2, ..., da}, where d1, d2, ..., da represent the power consumption of the 1st, 2nd, ..., ath sampling time points respectively;

[0024] Step S203: Calculate the power consumption fluctuation rate of the load end in a certain characteristic time period according to the following formula:

[0025] ;

[0026] Where, e is a positive integer and e∈[1,a-1], E represents the power consumption fluctuation rate of the load end in a certain characteristic time period, d e d e+1 They are respectively represented as the power consumption of the load end at the eth and e+1th sampling time points in a certain characteristic time period;

[0027] By calculating the power consumption fluctuation rate, we can capture the power consumption fluctuations at the load end during a characteristic time period. The level of the fluctuation rate reflects the power consumption stability at the load end, which is important for determining whether there is abnormal power consumption behavior.

[0028] In addition to detecting anomalies in historical power consumption, the power consumption fluctuation rate can be used to identify loads with large power consumption fluctuations. This can detect abnormal behaviors such as sudden load changes and frequent fluctuations, rather than relying solely on historical power consumption.

[0029] This method does not rely on a single data point, but instead evaluates fluctuations through changes in electricity consumption at consecutive sampling time points, further reducing false alarms due to accidental factors or occasional anomalies and enhancing the reliability of overall monitoring.

[0030] Furthermore, step S300 includes:

[0031] Step S301: Summarize all power consumption fluctuation rates of a certain load end in a certain characteristic time period during a training cycle, and draw a power consumption fluctuation graph of the certain load end in the certain characteristic time period, wherein the power consumption fluctuation graph uses date as the horizontal axis and power consumption fluctuation rate as the vertical axis;

[0032] Step S302: Calculate the slope of a certain adjacent date as H=|E1-E2| based on the power consumption fluctuation graph, preset a slope threshold, compare the slope with the slope threshold, and if the slope exceeds the slope threshold, set the adjacent date as an abnormal adjacent date;

[0033] Step S303: Counting the number of abnormal adjacent dates of a certain load terminal in a certain characteristic time period during the training cycle, presetting a threshold for the number of abnormal adjacent dates, comparing the number of abnormal adjacent dates of a certain load terminal in the certain characteristic time period with the threshold for the number of abnormal adjacent dates, and setting the load terminal as a target load terminal if the number exceeds the threshold for the number of abnormal adjacent dates;

[0034] By drawing a power consumption fluctuation graph, we can clearly show the power consumption fluctuations at the load end in different time periods, making the trend of power consumption anomalies more intuitive. Combined with slope calculation, we can effectively identify time periods with large power consumption fluctuations, avoiding misjudgments that may result from relying solely on a single fluctuation rate value.

[0035] Setting a slope threshold can detect dramatic changes in load consumption between adjacent dates and identify possible sudden load anomalies. This method can capture sudden changes in load consumption habits.

[0036] By counting the number of abnormal adjacent dates, false alarms caused by accidental fluctuations at a certain time point are avoided, and the accuracy of anomaly detection is improved. Combined with the number threshold of abnormal adjacent dates, load ends with real long-term abnormal trends can be effectively screened out, rather than load ends with only short-term fluctuations, thus improving monitoring reliability.

[0037] This method uses time series analysis, combined with slope calculation and quantitative statistics, to make anomaly detection more scientific and effectively reduce false alarms caused by short-term fluctuations or data noise.

[0038] Furthermore, step S400 includes:

[0039] Step S401: Obtain the load terminals associated with a certain energy storage terminal, match and screen them with the target load terminals in a certain characteristic time period, identify the target load terminals corresponding to the energy storage terminal, and generate an energy storage terminal group corresponding to the certain characteristic time period. The energy storage terminal group is a plurality of target load terminals associated with one energy storage terminal.

[0040] Step S402: Obtain the operating parameters of all energy storage end groups in all characteristic time periods during the training cycle, obtain a numerical set of a certain operating parameter, obtain the operating parameters of a certain energy storage end group in a certain characteristic time period on a certain day during the training cycle, and normalize the operating parameters based on the numerical set of the certain operating parameter. Calculate the operating score of the energy storage end group in the characteristic time period according to the following formula:

[0041] ;

[0042] Among them, F represents the operation score of the energy storage terminal group in the characteristic time period, G f It is represented by the normalized value of the fth operating parameter of the energy storage terminal group in the characteristic time period. It is represented as the weight of the fth operating parameter, and g is represented as the total number of operating parameters;

[0043] Step S403: Obtain an analysis period for the energy storage terminal group, collect all operation scores of the energy storage terminal group in a certain characteristic time period within a certain analysis period, and calculate an average operation score of the energy storage terminal group in a certain characteristic time period within the certain analysis period;

[0044] Step S404: Summarize the average operation scores of the energy storage end groups in a certain characteristic time period in all analysis periods within the training period, calculate the characteristic operation score of the energy storage end group in the certain characteristic time period, and draw a characteristic operation score graph for the analysis period corresponding to the energy storage end group, wherein the characteristic operation score graph uses the characteristic time period as the horizontal axis and the characteristic operation score as the vertical axis;

[0045] By normalizing operating parameters and calculating weights, this method can comprehensively evaluate the performance of energy storage terminal groups within different characteristic time periods, thereby deriving a more representative operational score. This multi-dimensional scoring system can more accurately reflect the operating efficiency and stability of energy storage terminals.

[0046] By drawing a characteristic operation score graph, the operating status of the energy storage terminal group can be dynamically displayed. The analysis cycle and operation score of the energy storage terminal group are combined with the data graphical display, allowing operation and maintenance personnel to more intuitively monitor the operating status of the energy storage terminal.

[0047] Because this method relies on the comprehensive scoring of multiple operating parameters rather than a single indicator, it avoids false positives caused by accidental events or local problems. Through normalization and weight distribution, it can ensure a fairer and more stable scoring system and improve the accuracy of anomaly detection.

[0048] Furthermore, step S500 includes:

[0049] Step S501: obtaining a real-time operation score of a certain energy storage terminal group in a characteristic time period, drawing a real-time operation score graph of the energy storage terminal group, and mapping the real-time operation score graph to a characteristic operation score graph of the energy storage terminal group. If the real-time operation score in a certain characteristic time period is higher than the characteristic operation score, the characteristic time period is set as a high-risk time period; if the real-time operation score in a certain characteristic time period is lower than the characteristic operation score, the characteristic time period is set as a low-risk time period;

[0050] Step S502: Calculate the absolute value of the difference in a certain characteristic time period, and calculate the real-time analysis period of the energy storage terminal group according to the following formula:

[0051] ;

[0052] Among them, T' represents the real-time analysis period of the energy storage terminal group, T represents the analysis period of the energy storage terminal group, and X n It is expressed as the absolute value of the difference between the nth characteristic time period in the high-risk time period, J is expressed as the weight of the high-risk time period, and Y m It is represented by the absolute value of the difference between the mth characteristic time period in the low-risk time period, L is represented by the weight of the low-risk time period, N is represented by the total number of high-risk time periods, and M is represented by the total number of low-risk time periods;

[0053] Step S503: Compare the real-time analysis period with the time interval of the next characteristic time period. If the time interval is greater than the time interval, execute step S501; if the time interval is less than the time interval, perform abnormality analysis on the energy storage terminal group.

[0054] By plotting a real-time operation score chart and comparing it with the characteristic operation score chart, high-risk and low-risk time periods can be quickly identified, ensuring accurate understanding of the operating status of the energy storage terminal group. If the real-time score is significantly higher than the characteristic score, there may be overload operation or abnormal fluctuations. If the real-time score is significantly lower than the characteristic score, there may be decreased energy storage efficiency, equipment aging, or failure.

[0055] Calculates the real-time analysis cycle of the energy storage end group and dynamically adjusts the monitoring frequency based on the weight changes of high-risk and low-risk time periods to ensure reasonable resource allocation. When there are more high-risk time periods, the system automatically shortens the analysis cycle and increases the monitoring frequency to respond to abnormal situations more quickly. When there are more low-risk time periods, the system appropriately extends the analysis cycle to reduce unnecessary calculations and resource consumption, thereby improving monitoring efficiency.

[0056] By combining the difference calculation between the real-time score and the feature score, a refined analysis of the operating status of the energy storage end group is performed to avoid misjudgments that may result from relying solely on static thresholds. By calculating the absolute value, the degree of deviation of the energy storage end group in a specific time period can be accurately measured, thereby more accurately identifying abnormal operating conditions.

[0057] In order to better implement the above method, a data intelligent analysis system based on the integration of source, grid, load and storage is proposed. The system includes a characteristic time period module, a power consumption fluctuation rate module, a target load terminal module, a characteristic operation scoring diagram module and a real-time analysis module.

[0058] Characteristic time period module: obtains historical electricity consumption records of the load side in the integrated power system of source, grid, load and storage, divides a day into several time periods, calculates the historical average electricity consumption of a certain time period, and determines the characteristic time period based on the historical average electricity consumption;

[0059] Power consumption fluctuation rate module: Set the training cycle, collect the power consumption of the load end at the sampling time point, and calculate the power consumption fluctuation rate of the load end in the characteristic time period;

[0060] Target load end module: draws a power consumption fluctuation graph of the load end in a characteristic time period during the training cycle, and based on the slope of the power consumption fluctuation graph, counts abnormal adjacent dates to determine the target load end;

[0061] Characteristic Operation Score Diagram Module: Generates energy storage end groups based on target load ends, collects operating parameters of the energy storage end groups, calculates the operation scores of the energy storage end groups, and draws characteristic operation score diagrams for the analysis periods corresponding to the energy storage end groups.

[0062] Real-time analysis module: obtains the real-time operation score of the energy storage end group, calculates the real-time analysis cycle of the energy storage end group, and determines the status of the energy storage end group's abnormal analysis.

[0063] Furthermore, the characteristic time period module includes a historical average power consumption threshold calculation unit and a characteristic time period determination unit:

[0064] Historical average power consumption threshold unit: obtains the historical power consumption record of a certain load end, summarizes the historical power consumption of the load end in a certain time period, calculates the historical average power consumption of the load end in the certain time period, summarizes the historical average power consumption of all load ends in all time periods, calculates the mean and standard deviation of the historical average power consumption, and calculates the threshold of the historical average power consumption;

[0065] Determine the characteristic time period unit: compare the historical average power consumption of a certain load end in a certain time period with the threshold of the historical average power consumption; if the historical average power consumption of the load end is higher than the threshold of the historical average power consumption, set the load end as an abnormal load end; count the number of abnormal load ends in a certain time period, set the number threshold of abnormal load ends, compare the number of abnormal load ends in the time period with the number threshold of abnormal load ends; if it exceeds the number threshold of abnormal load ends, set the time period as a characteristic time period.

[0066] Furthermore, the target load end module includes a unit for drawing a power consumption fluctuation diagram and a unit for determining a target load end:

[0067] Drawing electricity consumption fluctuation graph unit: summarizes all electricity consumption fluctuation rates of a certain load end in a certain characteristic time period during the training cycle, and draws the electricity consumption fluctuation graph of a certain load end in a certain characteristic time period;

[0068] Determine the target load end unit: according to the power consumption fluctuation graph, calculate the slope of a certain adjacent date, preset a slope threshold, compare the slope with the slope threshold, if it exceeds the slope threshold, set the adjacent date as an abnormal adjacent date, count the number of abnormal adjacent dates of a certain load end in a certain characteristic time period during the training cycle, preset a number threshold of abnormal adjacent dates, compare the number of abnormal adjacent dates of a certain load end in a certain characteristic time period with the number threshold of abnormal adjacent dates, if it exceeds the number threshold of abnormal adjacent dates, set the load end as the target load end.

[0069] Furthermore, the feature operation score graph module includes a calculation operation score unit and a drawing feature operation score graph unit:

[0070] Calculating operation scoring unit: obtaining the load terminal associated with a certain energy storage terminal, matching and screening the target load terminal in a certain characteristic time period, identifying the target load terminal corresponding to the energy storage terminal, and generating an energy storage terminal group corresponding to a certain characteristic time period, wherein the energy storage terminal group is a plurality of target load terminals associated with one energy storage terminal, obtaining the operating parameters of all energy storage terminal groups in all characteristic time periods during a training cycle, obtaining a numerical set of a certain operating parameter, obtaining the operating parameters of a certain energy storage terminal group in a certain characteristic time period on a certain day during the training cycle, and normalizing the operating parameters according to the numerical set of the certain operating parameter, and calculating the operation score of the energy storage terminal group in the characteristic time period;

[0071] The unit for drawing a characteristic operation score graph is used to obtain the analysis period of the energy storage end group, collect all the operation scores of the energy storage end group in a certain characteristic time period within a certain analysis period, calculate the average operation score of the energy storage end group in a certain characteristic time period within a certain analysis period, summarize the average operation scores of the energy storage end group in a certain characteristic time period across all analysis periods within the training period, calculate the characteristic operation score of the energy storage end group in a certain characteristic time period, and draw a characteristic operation score graph for the analysis period corresponding to the energy storage end group.

[0072] Compared with the prior art, the present invention has the following beneficial effects:

[0073] By calculating the mean, standard deviation, and threshold of historical average power consumption, we can accurately screen out abnormal load terminals. By combining the threshold of the number of abnormal load terminals, we can determine the characteristic time period and improve the accuracy of data analysis.

[0074] Using time series analysis to calculate the power consumption fluctuation rate of the load end during a characteristic time period can dynamically reflect the load change trend and help accurately predict load anomalies. Through power consumption fluctuation charts and slope calculations, statistics of abnormal adjacent dates are collected to identify the target load end. Combined with historical data for trend analysis, this improves the accuracy of target load end identification and avoids misjudgment.

[0075] Through data-driven methods, dynamic optimization of energy storage systems can be achieved, the stability and economy of the power system can be improved, and historical data, real-time data and predictive analysis can be combined to optimize energy storage scheduling and improve the new energy absorption capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 This is a flow chart of a data intelligent analysis method based on source-grid-load-storage integration according to the present invention;

[0077] Figure 2 This is a structural schematic diagram of a data intelligent analysis system based on source-grid-load-storage integration according to the present invention. DETAILED DESCRIPTION

[0078] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0079] See also Figure 1 and Figure 2 The present invention provides a technical solution: a data intelligent analysis method based on source-grid-load-storage integration, the method comprising:

[0080] Step S100: Obtain historical electricity consumption records of the load side in the source-grid-load-storage integrated power system, divide a day into several time periods, calculate the historical average electricity consumption of a certain time period, and determine a characteristic time period based on the historical average electricity demand;

[0081] Wherein, step S100 includes:

[0082] Step S101: obtaining historical power consumption records of a certain load terminal, summarizing the historical power consumption of the load terminal in a certain time period, and calculating the historical average power consumption of the load terminal in the certain time period;

[0083] Step S102: Summarize the historical average power consumption of all load terminals in all time periods, calculate the mean of the historical average power consumption to be A and the standard deviation to be B, and calculate the threshold of the historical average power consumption to be C=A+B*K, where C represents the threshold of the historical average power consumption and K represents the preset value;

[0084] Step S103: comparing the historical average power consumption of a load terminal in a certain time period with a threshold value of the historical average power consumption; if the historical average power consumption of the load terminal is higher than the threshold value of the historical average power consumption, setting the load terminal as an abnormal load terminal;

[0085] Step S104: Counting the number of abnormal load terminals in a certain time period, setting a threshold for the number of abnormal load terminals, comparing the number of abnormal load terminals in the time period with the threshold for the number of abnormal load terminals, and if the number exceeds the threshold for the number of normal load terminals, setting the time period as a characteristic time period;

[0086] For example, the historical power consumption of the first load end in the first time period is 50kWh, 150kWh, 75kWh, 45kWh, and 100kWh, and the historical power consumption of the first load end in the second time period is 25kWh, 20kWh, 35kWh, 40kWh, and 55kWh;

[0087] The calculated historical average power consumption has a mean of 55 kWh and a standard deviation of 34.24 kWh. The preset K is 1.5, and the calculated threshold of the historical average power consumption is 106.36 kWh.

[0088] The abnormal load end in the first time period is 1, and the abnormal load end in the second time period is 0, so the first time period is the characteristic time period.

[0089] Step S200: setting a training cycle, collecting the power consumption of the load end at a sampling time point, and calculating the power consumption fluctuation rate of the load end in a characteristic time period;

[0090] Wherein, step S200 includes:

[0091] Step S201: Select a number of consecutive days as a training cycle, obtain a time range corresponding to a certain characteristic time period, and preset a number of sampling time points within the time range;

[0092] Step S202: Collect the power consumption of a load end at a certain sampling time point, summarize the power consumption of the load end at all sampling time points within a certain characteristic time period, and mark the sampling time points in order to obtain the power consumption set of all sampling time points of the load end within the certain characteristic time period as D={d1, d2, ..., da}, where d1, d2, ..., da represent the power consumption of the 1st, 2nd, ..., ath sampling time points respectively;

[0093] Step S203: Calculate the power consumption fluctuation rate of the load end in a certain characteristic time period according to the following formula:

[0094] ;

[0095] Where, e is a positive integer and e∈[1,a-1], E represents the power consumption fluctuation rate of the load end in a certain characteristic time period, d e d e+1 They are respectively represented as the power consumption of the load end at the eth and e+1th sampling time points in a certain characteristic time period;

[0096] For example, the electricity consumption at the sampling time points is 50kWh, 55kWh, 60kWh, 58kWh, and 65kWh, and the calculated electricity consumption fluctuation rate is 8.62%.

[0097] Step S300: drawing a power consumption fluctuation graph of the load end in a characteristic time period during a training cycle, and counting abnormal adjacent dates based on the slope of the power consumption fluctuation graph to determine the target load end;

[0098] Wherein, step S300 includes:

[0099] Step S301: Summarize all power consumption fluctuation rates of a certain load end in a certain characteristic time period during a training cycle, and draw a power consumption fluctuation graph of the certain load end in the certain characteristic time period, wherein the power consumption fluctuation graph uses date as the horizontal axis and power consumption fluctuation rate as the vertical axis;

[0100] Step S302: Calculate the slope of a certain adjacent date as H=|E1-E2| based on the power consumption fluctuation graph, preset a slope threshold, compare the slope with the slope threshold, and if the slope exceeds the slope threshold, set the adjacent date as an abnormal adjacent date;

[0101] Step S303: Counting the number of abnormal adjacent dates of a certain load terminal in a certain characteristic time period during the training cycle, presetting a threshold for the number of abnormal adjacent dates, comparing the number of abnormal adjacent dates of a certain load terminal in the certain characteristic time period with the threshold for the number of abnormal adjacent dates, and setting the load terminal as a target load terminal if the number exceeds the threshold for the number of abnormal adjacent dates;

[0102] For example, the volatility is 5.2, 6.1, 5.8, 7.5, 8.0, 10.2, 12.5, 15.0, 14.2, 16.8;

[0103] The slopes calculated for adjacent dates are 0.9, 0.3, 1.7, 0.5, 2.2, 2.3, 2.5, 0.8, and 2.6;

[0104] Assuming that the slope threshold is set to 2.0, the total number of abnormal adjacent dates is 4. If the number threshold of abnormal adjacent dates is set to 3, the load end is set as the target load end.

[0105] Step S400: Generate an energy storage terminal group based on the target load terminal, collect operating parameters of the energy storage terminal group, calculate the operating score of the energy storage terminal group, and draw a characteristic operating score diagram of the analysis period corresponding to the energy storage terminal group;

[0106] Wherein, step S400 includes:

[0107] Step S401: Obtain the load terminals associated with a certain energy storage terminal, match and screen them with the target load terminals in a certain characteristic time period, identify the target load terminals corresponding to the energy storage terminal, and generate an energy storage terminal group corresponding to the certain characteristic time period. The energy storage terminal group is a plurality of target load terminals associated with one energy storage terminal.

[0108] Step S402: Obtain the operating parameters of all energy storage end groups in all characteristic time periods during the training cycle, obtain a numerical set of a certain operating parameter, obtain the operating parameters of a certain energy storage end group in a certain characteristic time period on a certain day during the training cycle, and normalize the operating parameters based on the numerical set of the certain operating parameter. Calculate the operating score of the energy storage end group in the characteristic time period according to the following formula:

[0109] ;

[0110] Among them, F represents the operation score of the energy storage terminal group in the characteristic time period, G f It is represented by the normalized value of the fth operating parameter of the energy storage terminal group in the characteristic time period. It is represented as the weight of the fth operating parameter, and g is represented as the total number of operating parameters;

[0111] Step S403: Obtain an analysis period for the energy storage terminal group, collect all operation scores of the energy storage terminal group in a certain characteristic time period within a certain analysis period, and calculate an average operation score of the energy storage terminal group in a certain characteristic time period within the certain analysis period;

[0112] Step S404: Summarize the average operation scores of the energy storage end groups in a certain characteristic time period in all analysis periods within the training period, calculate the characteristic operation score of the energy storage end group in the certain characteristic time period, and draw a characteristic operation score graph for the analysis period corresponding to the energy storage end group, wherein the characteristic operation score graph uses the characteristic time period as the horizontal axis and the characteristic operation score as the vertical axis;

[0113] For example, the energy storage terminal A is connected to the load terminals X, Y, and Z, where X and Y are the target load terminals in the first characteristic time period. Then, the energy storage terminal group A in the first characteristic time period includes X and Y.

[0114] During the first characteristic time period of the energy storage end group, the discharge power was 550 kW, the charging power was 300 kW, and the energy storage efficiency was 95%. After normalization, the discharge power was 0.9, the charging power was 0.6, and the energy storage efficiency was 0.95. The weight of the discharge power was 0.4, the weight of the charging power was 0.3, and the weight of the energy storage efficiency was 0.3. The calculated operation score was 0.81.

[0115] Step S500: obtaining a real-time operation score of the energy storage terminal group, calculating a real-time analysis period of the energy storage terminal group, and determining the status of abnormal analysis of the energy storage terminal group;

[0116] Wherein, step S500 includes:

[0117] Step S501: obtaining a real-time operation score of a certain energy storage terminal group in a characteristic time period, drawing a real-time operation score graph of the energy storage terminal group, and mapping the real-time operation score graph to a characteristic operation score graph of the energy storage terminal group. If the real-time operation score in a certain characteristic time period is higher than the characteristic operation score, the characteristic time period is set as a high-risk time period; if the real-time operation score in a certain characteristic time period is lower than the characteristic operation score, the characteristic time period is set as a low-risk time period;

[0118] Step S502: Calculate the absolute value of the difference in a certain characteristic time period, and calculate the real-time analysis period of the energy storage terminal group according to the following formula:

[0119] ;

[0120] Among them, T' represents the real-time analysis period of the energy storage terminal group, T represents the analysis period of the energy storage terminal group, and X n It is expressed as the absolute value of the difference between the nth characteristic time period in the high-risk time period, J is expressed as the weight of the high-risk time period, and Y m It is represented by the absolute value of the difference between the mth characteristic time period in the low-risk time period, L is represented by the weight of the low-risk time period, N is represented by the total number of high-risk time periods, and M is represented by the total number of low-risk time periods;

[0121] Step S503: Compare the real-time analysis period with the time interval of the next characteristic time period. If the time interval is greater than the time interval, execute step S501; if the time interval is less than the time interval, perform abnormality analysis on the energy storage terminal group.

[0122] For example, the energy storage end group's discharge power in the second characteristic time period is 500 kW, the charging power is 400 kW, and the energy storage efficiency is 92%. The normalized discharge power is 0.85, the normalized charging power is 0.65, and the normalized energy storage efficiency is 0.90. The calculated real-time operation score is 0.82.

[0123] The real-time analysis period is calculated and compared with the time interval of the next characteristic time period. If the time interval is greater than the real-time analysis period, step S501 needs to be re-executed to update the real-time analysis period. If the time interval is less than the real-time analysis period, the energy storage end group enters the abnormal analysis mode.

[0124] In order to better implement the above method, a data intelligent analysis system based on the integration of source, grid, load and storage is proposed. The system includes a characteristic time period module, a power consumption fluctuation rate module, a target load terminal module, a characteristic operation scoring diagram module and a real-time analysis module.

[0125] Characteristic time period module: obtains historical electricity consumption records of the load side in the integrated power system of source, grid, load and storage, divides a day into several time periods, calculates the historical average electricity consumption of a certain time period, and determines the characteristic time period based on the historical average electricity consumption;

[0126] The characteristic time period module includes a unit for calculating the historical average power consumption threshold and a unit for determining the characteristic time period:

[0127] Historical average power consumption threshold unit: obtains the historical power consumption record of a certain load end, summarizes the historical power consumption of the load end in a certain time period, calculates the historical average power consumption of the load end in the certain time period, summarizes the historical average power consumption of all load ends in all time periods, calculates the mean and standard deviation of the historical average power consumption, and calculates the threshold of the historical average power consumption;

[0128] Determine the characteristic time period unit: compare the historical average power consumption of a certain load end in a certain time period with the threshold of the historical average power consumption; if the historical average power consumption of the load end is higher than the threshold of the historical average power consumption, set the load end as an abnormal load end; count the number of abnormal load ends in a certain time period, set the number threshold of abnormal load ends, compare the number of abnormal load ends in the time period with the number threshold of abnormal load ends; if it exceeds the number threshold of abnormal load ends, set the time period as a characteristic time period.

[0129] Power consumption fluctuation rate module: Set the training cycle, collect the power consumption of the load end at the sampling time point, and calculate the power consumption fluctuation rate of the load end in the characteristic time period;

[0130] Target load end module: draws a power consumption fluctuation graph of the load end in a characteristic time period during the training cycle, and based on the slope of the power consumption fluctuation graph, counts abnormal adjacent dates to determine the target load end;

[0131] The target load end module includes a unit for drawing a power consumption fluctuation diagram and a unit for determining the target load end:

[0132] Drawing electricity consumption fluctuation graph unit: summarizes all electricity consumption fluctuation rates of a certain load end in a certain characteristic time period during the training cycle, and draws the electricity consumption fluctuation graph of a certain load end in a certain characteristic time period;

[0133] Determine the target load end unit: according to the power consumption fluctuation graph, calculate the slope of a certain adjacent date, preset a slope threshold, compare the slope with the slope threshold, if it exceeds the slope threshold, set the adjacent date as an abnormal adjacent date, count the number of abnormal adjacent dates of a certain load end in a certain characteristic time period during the training cycle, preset a number threshold of abnormal adjacent dates, compare the number of abnormal adjacent dates of a certain load end in a certain characteristic time period with the number threshold of abnormal adjacent dates, if it exceeds the number threshold of abnormal adjacent dates, set the load end as the target load end.

[0134] Characteristic Operation Score Diagram Module: Generates energy storage end groups based on target load ends, collects operating parameters of the energy storage end groups, calculates the operation scores of the energy storage end groups, and draws characteristic operation score diagrams for the analysis periods corresponding to the energy storage end groups.

[0135] The feature operation score graph module includes a calculation operation score unit and a feature operation score graph drawing unit:

[0136] Calculating operation scoring unit: obtaining the load terminal associated with a certain energy storage terminal, matching and screening the target load terminal in a certain characteristic time period, identifying the target load terminal corresponding to the energy storage terminal, and generating an energy storage terminal group corresponding to a certain characteristic time period, wherein the energy storage terminal group is a plurality of target load terminals associated with one energy storage terminal, obtaining the operating parameters of all energy storage terminal groups in all characteristic time periods during a training cycle, obtaining a numerical set of a certain operating parameter, obtaining the operating parameters of a certain energy storage terminal group in a certain characteristic time period on a certain day during the training cycle, and normalizing the operating parameters according to the numerical set of the certain operating parameter, and calculating the operation score of the energy storage terminal group in the characteristic time period;

[0137] The unit for drawing a characteristic operation score graph is used to obtain the analysis period of the energy storage end group, collect all the operation scores of the energy storage end group in a certain characteristic time period within a certain analysis period, calculate the average operation score of the energy storage end group in a certain characteristic time period within a certain analysis period, summarize the average operation scores of the energy storage end group in a certain characteristic time period across all analysis periods within the training period, calculate the characteristic operation score of the energy storage end group in a certain characteristic time period, and draw a characteristic operation score graph for the analysis period corresponding to the energy storage end group.

[0138] Real-time analysis module: obtains the real-time operation score of the energy storage end group, calculates the real-time analysis cycle of the energy storage end group, and determines the status of the energy storage end group's abnormal analysis.

[0139] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A data intelligent analysis method based on source-grid-load-storage integration, characterized in that: Methods include: Step S100: Obtain historical electricity consumption records of the load side in the source-grid-load-storage integrated power system, divide a day into several time periods, calculate the historical average electricity consumption of a certain time period, and determine a characteristic time period based on the historical average electricity consumption; Furthermore, step S101: obtaining historical power consumption records of a certain load terminal, summarizing the historical power consumption of the load terminal in a certain time period, and calculating the historical average power consumption of the load terminal in the certain time period; Step S102: Summarize the historical average power consumption of all load terminals in all time periods, calculate the mean of the historical average power consumption to be A and the standard deviation to be B, and calculate the threshold of the historical average power consumption to be C=A+B*K, where C represents the threshold of the historical average power consumption and K represents the preset value; Step S103: comparing the historical average power consumption of a load terminal in a certain time period with a threshold value of the historical average power consumption; if the historical average power consumption of the load terminal is higher than the threshold value of the historical average power consumption, setting the load terminal as an abnormal load terminal; Step S104: Counting the number of abnormal load terminals in a certain time period, setting a threshold for the number of abnormal load terminals, comparing the number of abnormal load terminals in the time period with the threshold for the number of abnormal load terminals, and setting the time period as a characteristic time period if the number of abnormal load terminals exceeds the threshold. Step S200: setting a training cycle, collecting the power consumption of the load end at a sampling time point, and calculating the power consumption fluctuation rate of the load end in a characteristic time period; Step S300: drawing a power consumption fluctuation graph of the load end in a characteristic time period during a training cycle, and counting abnormal adjacent dates based on the slope of the power consumption fluctuation graph to determine the target load end; Step S400: Generate an energy storage terminal group based on the target load terminal, collect operating parameters of the energy storage terminal group, calculate the operating score of the energy storage terminal group, and draw a characteristic operating score diagram of the analysis period corresponding to the energy storage terminal group; Step S500: obtaining a real-time operation score of the energy storage terminal group, calculating a real-time analysis period of the energy storage terminal group, and determining a status of abnormal analysis of the energy storage terminal group.

2. The data intelligent analysis method based on source-grid-load-storage integration according to claim 1 is characterized in that: The step S200 includes the following steps: Step S201: Select a number of consecutive days as a training cycle, obtain a time range corresponding to a certain characteristic time period, and preset a number of sampling time points within the time range; Step S202: Collect the power consumption of a load end at a certain sampling time point, summarize the power consumption of the load end at all sampling time points within a certain characteristic time period, and mark the sampling time points in order to obtain the power consumption set of all sampling time points of the load end within the certain characteristic time period as D={d1, d2, ..., da}, where d1, d2, ..., da represent the power consumption of the 1st, 2nd, ..., ath sampling time points respectively, and a represents the total number of sampling time points; Step S203: Calculate the power consumption fluctuation rate of the load end in a certain characteristic time period according to the following formula: ； Where, e is a positive integer and e∈[1,a-1], E represents the power consumption fluctuation rate of the load end in a certain characteristic time period, d e d e+1 They are respectively represented as the power consumption of the load end at the eth and e+1th sampling time points in a certain characteristic time period.

3. The data intelligent analysis method based on source-grid-load-storage integration according to claim 2 is characterized in that: The step S300 includes the following steps: Step S301: Summarize all power consumption fluctuation rates of a certain load end in a certain characteristic time period during a training cycle, and draw a power consumption fluctuation graph of the certain load end in the certain characteristic time period, wherein the power consumption fluctuation graph uses date as the horizontal axis and power consumption fluctuation rate as the vertical axis; Step S302: Calculate the slope of a certain adjacent date as H=|E1-E2| based on the power consumption fluctuation graph, preset a slope threshold, compare the slope with the slope threshold, and if the slope exceeds the slope threshold, set the adjacent date as an abnormal adjacent date; Step S303: Count the number of abnormal adjacent dates of a certain load end in a certain characteristic time period during the training cycle, preset a threshold value for the number of abnormal adjacent dates, compare the number of abnormal adjacent dates of a certain load end in a certain characteristic time period with the threshold value for the number of abnormal adjacent dates, and if it exceeds the threshold value for the number of abnormal adjacent dates, set the load end as the target load end.

4. The data intelligent analysis method based on source-grid-load-storage integration according to claim 3 is characterized in that: The step S400 includes the following steps: Step S401: Obtain the load terminals associated with a certain energy storage terminal, match and screen them with the target load terminals in a certain characteristic time period, identify the target load terminals corresponding to the energy storage terminal, and generate an energy storage terminal group corresponding to the certain characteristic time period. The energy storage terminal group is a plurality of target load terminals associated with one energy storage terminal. Step S402: Obtain the operating parameters of all energy storage end groups in all characteristic time periods during the training cycle, obtain a numerical set of a certain operating parameter, obtain the operating parameters of a certain energy storage end group in a certain characteristic time period on a certain day during the training cycle, and normalize the operating parameters based on the numerical set of the certain operating parameter. Calculate the operating score of the energy storage end group in the characteristic time period according to the following formula: ； Among them, F represents the operation score of the energy storage terminal group in the characteristic time period, G f It is represented by the normalized value of the fth operating parameter of the energy storage terminal group in the characteristic time period. It is represented as the weight of the fth operating parameter, and g is represented as the total number of operating parameters; Step S403: Obtain an analysis period for the energy storage terminal group, collect all operation scores of the energy storage terminal group in a certain characteristic time period within a certain analysis period, and calculate an average operation score of the energy storage terminal group in a certain characteristic time period within the certain analysis period; Step S404: Summarize the average operation scores of the energy storage end groups in a certain characteristic time period in all analysis periods within the training period, calculate the characteristic operation scores of the energy storage end groups in the certain characteristic time period, and draw a characteristic operation score graph for the analysis period corresponding to the energy storage end group. The characteristic operation score graph uses the characteristic time period as the horizontal axis and the characteristic operation score as the vertical axis.

5. The data intelligent analysis method based on source-grid-load-storage integration according to claim 4 is characterized in that: The step S500 includes the following steps: Step S501: obtaining a real-time operation score of a certain energy storage terminal group in a characteristic time period, drawing a real-time operation score graph of the energy storage terminal group, and mapping the real-time operation score graph to a characteristic operation score graph of the energy storage terminal group. If the real-time operation score in a certain characteristic time period is higher than the characteristic operation score, the characteristic time period is set as a high-risk time period; if the real-time operation score in a certain characteristic time period is lower than the characteristic operation score, the characteristic time period is set as a low-risk time period; Step S502: Calculate the absolute value of the difference in a certain characteristic time period, and calculate the real-time analysis period of the energy storage terminal group according to the following formula: ； Among them, T' represents the real-time analysis period of the energy storage terminal group, T represents the analysis period of the energy storage terminal group, and X n It is expressed as the absolute value of the difference between the nth characteristic time period in the high-risk time period, J is expressed as the weight of the high-risk time period, and Y m It is represented by the absolute value of the difference between the mth characteristic time period in the low-risk time period, L is represented by the weight of the low-risk time period, N is represented by the total number of high-risk time periods, and M is represented by the total number of low-risk time periods; Step S503: compare the real-time analysis period with the time interval of the next characteristic time period. If it is greater than the time interval, execute step S501; if it is less than the time interval, perform abnormal analysis on the energy storage terminal group.

6. A data intelligent analysis system based on the integration of source, grid, load and storage, used to implement the data intelligent analysis method based on the integration of source, grid, load and storage as described in any one of claims 1 to 5, characterized in that: The system includes a characteristic time period module, a power consumption fluctuation rate module, a target load terminal module, a characteristic operation scoring diagram module and a real-time analysis module; The characteristic time period module: obtains the historical power consumption records of the load end in the source-grid-load-storage integrated power system, divides a day into several time periods, calculates the historical average power consumption of a certain time period, and determines the characteristic time period based on the historical average power consumption; The power consumption fluctuation rate module: sets a training cycle, collects the power consumption of the load end at the sampling time point, and calculates the power consumption fluctuation rate of the load end in the characteristic time period; The target load end module draws a power consumption fluctuation graph of the load end in a characteristic time period during a training cycle, and determines the target load end by counting abnormal adjacent dates based on the slope of the power consumption fluctuation graph; The characteristic operation score diagram module generates an energy storage terminal group according to the target load terminal, collects the operating parameters of the energy storage terminal group, calculates the operation score of the energy storage terminal group, and draws a characteristic operation score diagram for the analysis period corresponding to the energy storage terminal group; The real-time analysis module is configured to obtain a real-time operation score of the energy storage terminal group, calculate a real-time analysis period of the energy storage terminal group, and determine a status of abnormal analysis of the energy storage terminal group.

7. The data intelligent analysis system based on source-grid-load-storage integration according to claim 6 is characterized in that: The characteristic time period module includes a historical average power consumption threshold calculation unit and a characteristic time period determination unit: The historical average power consumption threshold unit: obtains the historical power consumption record of a certain load terminal, summarizes the historical power consumption of the load terminal in a certain time period, calculates the historical average power consumption of the load terminal in the certain time period, summarizes the historical average power consumption of all load terminals in all time periods, calculates the mean and standard deviation of the historical average power consumption, and calculates the threshold of the historical average power consumption; The characteristic time period determination unit compares the historical average power consumption of a load terminal in a certain time period with a threshold value of the historical average power consumption, and sets the load terminal as an abnormal load terminal if the historical average power consumption of the load terminal is higher than the threshold value of the historical average power consumption; Count the number of abnormal load ends in a certain time period, set a number threshold of abnormal load ends, compare the number of abnormal load ends in the time period with the number threshold of abnormal load ends, and if the number exceeds the number threshold of abnormal load ends, set the time period as a characteristic time period.

8. The data intelligent analysis system based on source-grid-load-storage integration according to claim 6 is characterized in that: The target load end module includes a unit for drawing a power consumption fluctuation diagram and a unit for determining a target load end: The unit for drawing a power consumption fluctuation graph: summarizes all power consumption fluctuation rates of a certain load end in a certain characteristic time period during a training cycle, and draws a power consumption fluctuation graph of a certain load end in a certain characteristic time period; The target load end determination unit: calculates the slope of a certain adjacent date according to the power consumption fluctuation graph, presets a slope threshold, compares the slope with the slope threshold, and if it exceeds the slope threshold, sets the adjacent date as an abnormal adjacent date, counts the number of abnormal adjacent dates of a certain load end in a certain characteristic time period within a training cycle, presets a number threshold of abnormal adjacent dates, compares the number of abnormal adjacent dates of a certain load end in a certain characteristic time period with the number threshold of abnormal adjacent dates, and if it exceeds the number threshold of abnormal adjacent dates, sets the load end as a target load end.

9. The data intelligent analysis system based on source-grid-load-storage integration according to claim 6 is characterized in that: The feature operation score graph module includes a calculation operation score unit and a feature operation score graph drawing unit: The operation scoring calculation unit: obtains the load terminal associated with a certain energy storage terminal, matches and screens it with the target load terminal in a certain characteristic time period, identifies the target load terminal corresponding to the energy storage terminal, and generates an energy storage terminal group corresponding to the certain characteristic time period, wherein the energy storage terminal group is a plurality of target load terminals associated with one energy storage terminal, obtains the operating parameters of all energy storage terminal groups in all characteristic time periods within a training cycle, obtains a numerical set of a certain operating parameter, obtains the operating parameters of a certain energy storage terminal group in a certain characteristic time period on a certain day within the training cycle, and normalizes the operating parameters according to the numerical set of the certain operating parameters to calculate the operation score of the energy storage terminal group in the characteristic time period; The unit for drawing a characteristic operation score graph is as follows: obtaining an analysis period of the energy storage end group, collecting all operation scores of the energy storage end group in a characteristic time period within a certain analysis period, calculating an average operation score of the energy storage end group in a certain characteristic time period within a certain analysis period, summarizing the average operation scores of the energy storage end group in a certain characteristic time period in all analysis periods within a training period, calculating a characteristic operation score of the energy storage end group in a certain characteristic time period, and drawing a characteristic operation score graph of the analysis period corresponding to the energy storage end group.

Images