A transformer area source load storage intelligent adjustment system based on multi-scene cooperation

Abstract

The application provides a kind of based on multi-scene cooperation's transformer area source and load storage intelligent adjustment method, belongs to the technical field of smart grid regulation, comprising: data acquisition module: collect power supply, load and grid related data, and generate corresponding characteristic parameters;Data analysis module: determine the comprehensive adjustment parameter based on the characteristic parameter, obtain several scenes and generate corresponding scene label;Data prediction module: generate predicted load curve, and then determine the charging capacity parameter in load valley period;Strategy execution module: based on predicted load curve, load valley period charging capacity parameter and scene label, execute the control strategy corresponding to scene label;Strategy updating module: compare the real-time load curve after strategy execution with predicted load curve, determine the adjustment deviation rate, and correct predicted load curve based on adjustment deviation rate, and then re-execute the control strategy. Significantly improve the energy utilization rate and the adaptive ability of system.

Description

Technical Field

[0001] This invention relates to the field of smart grid regulation technology, and in particular to a smart regulation system for source-load-storage distribution areas based on multi-scenario collaboration. Background Technology

[0002] With the development of smart grids, the demand for power system regulation is increasing, especially under conditions of large fluctuations in power supply and load. How to achieve intelligent regulation of the power grid and optimize energy allocation has become a research hotspot. Currently, there are some load forecasting and power grid control systems based on traditional regulation algorithms, but these systems often can only cope with single load fluctuation situations and are difficult to flexibly regulate in complex power grid scenarios.

[0003] Current power regulation systems typically rely on simple load forecasting and regulation strategies. However, due to a lack of consideration for changes in various scenarios, these systems often fail to achieve accurate load forecasting and optimized energy regulation. Therefore, existing technologies suffer from low regulation accuracy and poor energy efficiency when facing drastic load fluctuations and complex environmental changes.

[0004] Therefore, the present invention provides a smart regulation system for source-load-storage in transformer substations based on multi-scenario collaboration. Summary of the Invention

[0005] This invention provides a multi-scenario collaborative intelligent regulation system for power distribution area source-load-storage, which determines and executes corresponding control strategies through real-time data acquisition, load forecasting, and scenario labeling to achieve precise regulation. It overcomes the shortcomings of poor adaptability and low regulation accuracy in single scenarios in existing technologies. It can dynamically adjust the regulation strategy according to different power grid conditions, improve the accuracy of load forecasting, optimize energy allocation, ensure the efficient and stable operation of the power grid, and significantly improve energy utilization and the system's self-adaptive capability.

[0006] This invention provides a multi-scenario collaborative intelligent regulation system for power distribution area source-load-storage, comprising:

[0007] Data acquisition module: Collects power supply, load and power grid related data through a preset intelligent sensor network and generates corresponding characteristic parameters;

[0008] Data analysis module: Determines comprehensive adjustment parameters based on feature parameters, generates the distribution of transformer area scenes by combining preset thresholds, obtains several scenes and generates corresponding scene labels;

[0009] Data prediction module: acquires historical load data for a preset time period collected through a preset smart sensor network, generates a predicted load curve, and then determines the charging quantity parameters during off-peak periods.

[0010] Strategy execution module: Executes the control strategy corresponding to the scenario label based on the predicted load curve, charging quantity parameters during off-peak periods, and scenario label;

[0011] Strategy update module: Compare the real-time load curve after strategy execution with the predicted load curve, determine the adjustment deviation rate, correct the predicted load curve based on the adjustment deviation rate, and then re-execute the control strategy.

[0012] Preferred scenario tags include: deep underload scenario, light underload scenario, steady-state equilibrium scenario, light overload scenario, and deep overload scenario.

[0013] Preferably, the data acquisition module includes:

[0014] Parameter acquisition unit: Collects parameters of preset types, including: distributed power source data, energy storage system data, load classification data, and environmental data;

[0015] Feature generation unit: Generates corresponding feature parameters based on preset types of parameters. The feature parameters include: photovoltaic output correction coefficient, wind power availability, energy storage available capacity, charge and discharge efficiency correction coefficient, adjustable load ratio, load fluctuation index, voltage deviation rate, frequency deviation rate, and environmental impact factor.

[0016] Preferably, the data analysis module includes:

[0017] Parameter determination unit: determines comprehensive adjustment parameters based on characteristic parameters;

[0018] Parameter comparison unit: compares the comprehensive adjustment parameters with the preset threshold to generate several first candidate sets for preliminary screening;

[0019] Cross-validation unit: Cross-validation is performed based on each first candidate set and several feature parameters corresponding to each candidate set to generate several valid scene sets;

[0020] Tag generation unit: For the set of valid scenarios that have passed cross-validation, generate unique scenario tags according to preset priorities and generate corresponding scenario tags.

[0021] Preferably, the parameter determination unit includes:

[0022] Data preprocessing subunit: Normalizes the feature parameters;

[0023] Supply determination subunit: Determine the supply-side stability index based on the normalized photovoltaic output correction coefficient, wind power availability, energy storage available capacity, and charge / discharge efficiency correction coefficient;

[0024] ;

[0025] in, As a supply-side stability index, This is a correction factor for photovoltaic output. Wind power availability This represents the percentage of available energy storage capacity. This is a correction factor for charge / discharge efficiency;

[0026] Demand Determination Subunit: Determine the demand-side adjustment potential index based on the adjustable load ratio, load fluctuation index, voltage deviation rate, and frequency deviation rate;

[0027] ;

[0028] in, This is a potential index for demand-side adjustment. To allow for adjustable load proportions, For load fluctuation index, This is a correction value for the load fluctuation index. These are correction values ​​for voltage deviation rate and frequency deviation rate. Voltage deviation rate, Frequency deviation rate;

[0029] Parameter determination sub-unit: Comprehensive adjustment parameters are determined based on the supply-side stability index, the demand-side adjustment potential index, and environmental impact factors.

[0030] ;

[0031] in, To comprehensively adjust parameters, These are environmental impact factors.

[0032] Preferably, the data prediction module includes:

[0033] Curve generation unit: acquires historical load data for a preset time period collected through a preset smart sensor network, and generates a predicted load curve based on a preset model;

[0034] Time Period Determination Unit: Determines off-peak periods based on predicted load curves and energy storage system data;

[0035] Charging quantity determination unit: Determines the basic charging quantity during off-peak hours based on energy storage system data and distributed power source data;

[0036] Parameter adjustment unit: Adjusts the basic charging amount based on scene tags to determine the charging amount parameters during off-peak periods.

[0037] Preferably, the strategy update module includes:

[0038] Deviation determination unit: acquires the real-time load curve after strategy execution and determines the adjustment deviation rate based on the predicted load curve;

[0039] Curve adjustment unit: Based on the adjustment deviation rate, the predicted load curve of the data prediction module is corrected. If the deviation rate is positive, it indicates that the actual load is higher than the prediction, and the prediction curve is adjusted upward. If the deviation rate is negative, the prediction curve is adjusted downward.

[0040] Strategy Re-execution Subunit: Based on the corrected predicted load curve, combined with the charging quantity parameters during off-peak periods and scenario labels, the control strategy in the strategy execution module is re-executed.

[0041] Preferably, the strategy re-execution subunit includes:

[0042] Strategy matching block: Based on the modified predicted load curve scenario labels, it matches the corresponding control strategy from the preset strategy library;

[0043] Parameter determination block: Based on the control strategy obtained through matching and the charging quantity parameters during off-peak hours, the specific control parameters of each source-load-storage device are determined.

[0044] Instruction execution block: Based on specific control parameters, it sends control instructions to each source, load, and storage device, thereby re-executing the strategy.

[0045] Compared with the prior art, the beneficial effects of this application are as follows:

[0046] By determining and executing corresponding control strategies through real-time data acquisition, load forecasting, and scenario labeling, precise regulation is achieved. This overcomes the shortcomings of existing technologies, such as poor adaptability to single scenarios and low regulation accuracy. It can dynamically adjust regulation strategies according to different power grid conditions, improve the accuracy of load forecasting, optimize energy allocation, ensure the efficient and stable operation of the power grid, and significantly improve energy utilization and the system's self-adaptability. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0048] Figure 1 This is a schematic diagram of the structure of a power distribution area source-load-storage intelligent regulation system based on multi-scenario collaboration, provided by an embodiment of the present invention. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0050] Example 1:

[0051] This invention provides a smart regulation system for power distribution area source-load-storage based on multi-scenario collaboration, such as... Figure 1 As shown, it includes:

[0052] Data acquisition module: Collects power supply, load and power grid related data through a preset intelligent sensor network and generates corresponding characteristic parameters;

[0053] Data analysis module: Determines comprehensive adjustment parameters based on feature parameters, generates the distribution of transformer area scenes by combining preset thresholds, obtains several scenes and generates corresponding scene labels;

[0054] Data prediction module: acquires historical load data for a preset time period collected through a preset smart sensor network, generates a predicted load curve, and then determines the charging quantity parameters during off-peak periods.

[0055] Strategy execution module: Executes the control strategy corresponding to the scenario label based on the predicted load curve, charging quantity parameters during off-peak periods, and scenario label;

[0056] Strategy update module: Compare the real-time load curve after strategy execution with the predicted load curve, determine the adjustment deviation rate, correct the predicted load curve based on the adjustment deviation rate, and then re-execute the control strategy.

[0057] In this embodiment, the pre-defined intelligent sensor network consists of several data acquisition devices: Distributed power source data acquisition devices: Photovoltaic dedicated: irradiance sensor (accuracy ±2%), thermocouple temperature sensor (accuracy ±1℃), DC energy meter (compliant with GB / T 17215 standard); Wind power dedicated: ultrasonic anemometer (accuracy ±0.5m / s), speed encoder (resolution 0.1rpm); Energy storage system data acquisition devices: Battery Management System (BMS): supports 10Hz high-frequency acquisition, built-in SOC / SOH estimation algorithm (pre-defined ampere-hour integration method + Kalman filter); Temperature sensor array: 3 NTC temperature probes deployed per battery cluster, accuracy ±0.5℃. Load classification data acquisition devices: Smart meter: supports DL / T 645 protocol, with load classification metering function (accuracy 0.5S level); Load contract database: pre-defined adjustable load response time, adjustment range, and other parameters. Data Acquisition Equipment: Voltage / Frequency Transmitter: Compliant with GB / T 13850 standard, accuracy class 0.2; Power Quality Monitoring Device: Supports IEC 61000-4-3 standard, real-time monitoring of harmonic components. Environmental Data Acquisition Equipment: Six-Element Weather Station: Includes temperature and humidity sensors (accuracy ±0.5℃ / ±2% RH) and barometric pressure sensors (accuracy ±0.3hPa); Meteorological Data Interface: Pre-connected to the National Meteorological Administration API to obtain hourly forecast data.

[0058] In this embodiment, the control strategy corresponding to the scenario label is executed based on the predicted load curve, the charging quantity parameters during the off-peak period, and the scenario label. This includes determining and executing the content of the control strategy corresponding to the scenario label based on the predicted load curve, the charging quantity parameters during the off-peak period, and the scenario label. Specifically, this includes directly inputting the energy storage charging strategy and using it as a benchmark value for evaluating the energy storage regulation effect in the closed-loop feedback. Basic regulation strategy: For the mild overload scenario (F3), combined with the generated next day's load prediction curve, the full-power operation range of distributed photovoltaics is planned in advance, and the upper limit of energy storage charging power is dynamically adjusted according to the energy storage charging quantity during the predicted off-peak period (from the energy storage charging and discharging path planning). In the deep underload scenario (F0), referring to the energy storage charging and discharging path planning results, energy storage capacity is prioritized to be supplemented during the off-peak period to ensure the availability of energy storage when the backup power supply is connected. Dynamic compensation mechanism: When the voltage deviation rate exceeds 5%, the reactive power compensation device is triggered, and the compensation amount is calculated based on the rated apparent power of the transformer area and the real-time voltage deviation. The strategy execution module implements hierarchical control and forms a closed-loop feedback based on the scenario labels (F3 / F0) from the data analysis module, the load prediction curve from the data prediction module, and the energy storage charging parameters: For the mild overload scenario (F3): peak and valley periods are identified based on the predicted load curve, and the full-power operation range of distributed photovoltaics is defined. Simultaneously, based on the energy storage charging parameters output by the data prediction module, the upper limit of energy storage charging power is dynamically determined (matching the duration of the valley period and the safe capacity of energy storage), ensuring that the charging process does not exceed the safe upper limit of energy storage SOC (preset threshold). The relevant photovoltaic operation range and charging power parameters are used as benchmark values ​​and input to the strategy update module (S4) to evaluate the adjustment effect. If the actual charging amount deviates from the planned value by more than a preset range, the charging strategy for subsequent periods is automatically corrected. For the deep underload scenario (F0): combined with the energy storage charging and discharging path planning from the data prediction module, energy storage charging is prioritized during the low-load period. The charging rhythm is adjusted with reference to the predicted load curve to ensure the availability of energy storage when backup power is connected. In the control logic, the energy storage charging capacity planning result is directly used as the target parameter for strategy execution, linked with the real-time acquired energy storage SOC, dynamically adjusting the charging power to ensure that the energy storage state meets the preset coordination conditions; dynamic compensation mechanism: real-time monitoring of the voltage deviation rate of the data acquisition module, triggering reactive power compensation equipment when the limit is exceeded, the compensation amount is dynamically calculated based on the rated apparent power of the transformer area and the real-time deviation, quickly responding to grid fluctuations. The execution data of the compensation strategy (such as the actual compensation amount and voltage recovery effect) is fed back to the strategy update module, compared with the grid stability parameters in the prediction model, forming the basis for evaluating the regulation effect.

[0059] In this embodiment, the preset time period is the historical load data collection cycle pre-set by the data prediction module to generate the predicted load curve. Its setting needs to be combined with load change patterns and prediction targets. For example, when predicting the load for the next day, considering the daily cycle of residential electricity consumption and weekend differences, the preset time period can be set to the most recent 7 days, covering the complete weekly cycle and capturing the load fluctuation characteristics of weekdays and weekends. If predicting stable loads in industrial areas, a preset time period of the most recent 3 days is sufficient to reflect the regular production electricity consumption pattern. During special holidays such as Spring Festival and National Day, the preset time period can include data from the previous similar holiday, making the historical load data more relevant to the specific scenario, ensuring the accuracy of the predicted load curve, and providing a reliable basis for system adjustment. By flexibly setting the preset time period, historical data can be made more representative, improving the effectiveness of prediction and subsequent adjustment strategies.

[0060] In this embodiment, the adjustment deviation rate is calculated as follows: Adjustment deviation rate = (real-time load power − predicted load power) / predicted load power;

[0061] In this embodiment, the predicted load curve is corrected, and then the control strategy is re-executed. The correction logic is as follows: if the adjustment deviation rate exceeds 10% for three consecutive time periods, the daily load prediction model of S5 is called to update the load prediction for future time periods, and the charging amount of the energy storage charging and discharging path planning in S3 is adjusted simultaneously to ensure that the energy storage SOC is maintained in the safe range of 20%-80%.

[0062] The beneficial effects of the above technical solution are as follows: By performing precise regulation through real-time data acquisition, load forecasting, and scenario labeling, it overcomes the shortcomings of poor adaptability to single scenarios and low regulation accuracy in existing technologies. It can dynamically adjust the regulation strategy according to different power grid conditions, improve the accuracy of load forecasting, optimize energy allocation, ensure the efficient and stable operation of the power grid, and significantly improve energy utilization and system adaptability.

[0063] Example 2:

[0064] This invention provides a multi-scenario collaborative intelligent regulation system for source, load and storage in transformer substations. The scenario tags include: deep underload scenario, mild underload scenario, steady-state balance scenario, mild overload scenario, and deep overload scenario.

[0065] In this embodiment, the intelligent regulation system for power generation, load, and energy storage in a distribution area based on multi-scenario collaboration is tagged with five scenarios: Deep underload scenario refers to power supply being far below load demand, such as in remote mountain villages at night when photovoltaic power stops, power supply is scarce, and energy storage is low, resulting in severe deviations in grid voltage and frequency; Mild underload scenario refers to slightly insufficient supply, such as in residential areas during the day when photovoltaic output is limited, load is moderate, energy storage is moderately low, and grid parameters are slightly abnormal; Steady-state balance scenario refers to a basic match between supply and demand, such as during normal working days when power supply is stable, load is moderate, energy storage is within a reasonable range, and grid parameters have no significant deviations; Mild overload scenario refers to supply being slightly less than demand, such as in commercial areas in the evening when increased load leads to slightly insufficient power supply, which is supplemented by energy storage discharge, resulting in slight fluctuations in grid parameters; Deep overload scenario refers to supply being far below demand, such as in industrial areas during the high-temperature period in summer when a surge in load leads to insufficient power supply and energy storage, resulting in severe imbalances in grid voltage and frequency. These scenario labels accurately depict the operating status of the distribution area by comprehensively evaluating power output, energy storage status, load level and grid parameters, providing a basis for intelligent adjustment and ensuring the safe and efficient operation of the system. For example, in the event of deep overload, priority is given to protecting critical loads, and in steady state, energy utilization efficiency is optimized.

[0066] The beneficial effects of the above technical solution are as follows: By introducing multiple scenario labels (such as deep underload, slight underload, steady-state balance, slight overload, and deep overload scenarios), accurate identification and dynamic adjustment of different load states of the power grid are achieved. Compared with existing technologies, this method can more accurately respond to different load conditions that may occur during power grid operation, improve the flexibility and adaptability of adjustment strategies, ensure effective optimization of energy allocation under changing power grid conditions, reduce the risk of overload or underload, and improve the stability and reliability of the power grid.

[0067] The data acquisition module of the intelligent regulation system for source-load-storage in transformer substations based on multi-scenario collaboration includes:

[0068] Parameter acquisition unit: Collects parameters of preset types, including: distributed power source data, energy storage system data, load classification data, and environmental data;

[0069] Feature generation unit: Generates corresponding feature parameters based on preset types of parameters. The feature parameters include: photovoltaic output correction coefficient, wind power availability, energy storage available capacity, charge and discharge efficiency correction coefficient, adjustable load ratio, load fluctuation index, voltage deviation rate, frequency deviation rate, and environmental impact factor.

[0070] In this embodiment, the distributed power source data includes: real-time irradiance and module temperature of distributed photovoltaic power, and real-time wind speed and turbine speed of distributed wind power; the energy storage system data includes: state of charge, health status, and surface temperature of energy storage batteries; the load classification data includes: real-time power and adjustable load capacity of the load; the grid operation data includes: voltage and frequency of the transformer substation bus, and active power flow; and the environmental data includes: real-time temperature, humidity, air pressure, and weather forecast for the next 2 hours (light intensity and wind speed prediction).

[0071] In this embodiment, the corresponding characteristic parameters generated based on preset types of parameters include: generating a photovoltaic output correction coefficient based on irradiance and module temperature; generating wind power availability based on wind speed and turbine speed; generating energy storage available capacity (SOC×SOH×rated capacity) based on SOC, SOH, and preset rated capacity; generating a charge / discharge efficiency correction coefficient (temperature's influence on energy storage efficiency) based on the deviation between surface temperature and preset optimal temperature (25℃); generating an adjustable load ratio (the proportion of adjustable load to total load) based on real-time power and adjustable load capacity; generating a load fluctuation index (reflecting load stability) based on power fluctuation data within a 10-minute sliding window; generating a voltage deviation rate (reflecting voltage stability) based on the deviation between bus voltage and rated voltage (220V); generating a frequency deviation rate (reflecting frequency stability) based on the deviation between frequency and rated frequency (50Hz); generating an environmental impact factor (the comprehensive impact of the environment on power output) based on air temperature and humidity; and generating a weather risk label (sunny / ...) based on weather forecast data. (Rainy days, etc., for auxiliary scene classification); Photovoltaic output correction coefficient: Based on irradiance and module temperature, the correction coefficient is higher when the temperature is close to 25℃ and the irradiance is high. It can be expressed as "Photovoltaic output correction coefficient = 1 - 0.005 × |Module temperature - 25℃| (adjusted based on this when the irradiance meets certain conditions)," for example, "The photovoltaic output correction coefficient is equal to 1 minus 0.005 multiplied by the absolute value of the difference between the module temperature and 25℃. When the irradiance is sufficient, it is appropriately increased on this basis, and vice versa." Wind power availability: With reasonable wind speed and normal turbine speed, the actual output is divided by the rated output. It can be written as "Wind power availability equals the actual wind power output divided by the rated output. When the wind speed is within a reasonable range and the turbine speed is normal, the higher this ratio, the higher the availability."

[0072] In this embodiment, the environmental impact factor is generated by normalizing parameters such as temperature and humidity. Assuming the temperature is T, its effective range is [Tmin, Tmax], and the humidity is H, its effective range is [Hmin, Hmax]. First, Tnorm and Hnorm are normalized to obtain Tnorm and Hnorm. Then, weights α and β (α+β=1) are assigned according to the degree of environmental support for power sources (such as photovoltaic and wind power). The formula for the environmental impact factor can be expressed as: Eenv(t)=α⋅Tnorm+β⋅Hnorm. For example, on a sunny day with sufficient sunlight, the temperature T is close to Tmax, and the humidity H is low and close to Hmin. Eenv(t) approaches 1, indicating that the environment strongly supports power output. On a rainy day with low temperature and high humidity, Tnorm is low and Hnorm is high. If α>β (assuming that temperature has a greater impact on photovoltaics), Eenv(t) approaches 0, reflecting weak environmental support. This formula transforms temperature and humidity into the degree of support for power output through normalization and weighting. "The higher the value, the stronger the environmental support for power output."

[0073] The beneficial effects of the above technical solution are as follows: By collecting data from distributed power sources, energy storage systems, load classification, and the environment, and generating multiple characteristic parameters, it provides comprehensive and accurate data support. Compared with existing technologies, it fully considers the real-time operation of various energy sources such as photovoltaics, wind power, and energy storage, as well as the impact of environmental factors, significantly improving regulation precision and load forecasting accuracy. This multi-dimensional data fusion enhances the system's adaptability and regulation effect in complex power grid environments, improving power grid stability and energy utilization efficiency.

[0074] The data analysis module of the intelligent regulation system for source-load-storage in transformer substations based on multi-scenario collaboration includes:

[0075] Parameter determination unit: determines comprehensive adjustment parameters based on characteristic parameters;

[0076] Parameter comparison unit: compares the comprehensive adjustment parameters with the preset threshold to generate several first candidate sets for preliminary screening;

[0077] Cross-validation unit: Cross-validation is performed based on each first candidate set and several feature parameters corresponding to each candidate set to generate several valid scene sets;

[0078] Tag generation unit: For the set of valid scenarios that have passed cross-validation, generate unique scenario tags according to preset priorities and generate corresponding scenario tags.

[0079] In this embodiment, the preset thresholds are as follows: For the deep underload scenario (F0), it is triggered when the comprehensive adjustment parameter is lower than 0.3, and the energy storage state of charge (SOC) is lower than 20%, the load fluctuation index is less than 10%, and the deviation rates of voltage and frequency both exceed 15%, indicating a serious shortage of supply and the power grid on the verge of collapse; for the mild underload scenario (F1), it is determined when the comprehensive adjustment parameter is in the range of 0.3 - 0.5, the energy storage SOC is in the range of 20% - 50%, the load fluctuation index is in the range of 10% - 20%, and any one of the deviation rates of voltage or frequency exceeds 10%, reflecting insufficient supply accompanied by medium load fluctuations and power grid anomalies; for the steady-state balance scenario (F2), the comprehensive adjustment parameter is 0.5 - 0.7, the energy storage SOC remains in the ideal range of 50% - 80%, the load fluctuation index is lower than 20%, and the deviation rates of voltage and frequency both do not exceed 10%, representing balanced links and a stable power grid; for the light overload scenario (F3), it is triggered when the comprehensive adjustment parameter is 0.7 - 0.9, the energy storage SOC is higher than 50%, the load fluctuation index exceeds 20%, and any one of the deviation rates of voltage or frequency exceeds 15%, indicating sufficient supply but severe load fluctuations and power grid warnings; for the deep overload scenario (F4), it is determined when the comprehensive adjustment parameter is higher than 0.9, or the deviation rates of voltage and frequency both exceed 20%, and a device trip signal is detected. Regardless of the energy storage and load states, it is preferentially identified as a serious anomaly or emergency fault scenario of the power grid;

[0080] In this embodiment, several first candidate sets for preliminary screening are generated: Preliminary screening: If K(t) < 0.3, enter the deep underload F0 candidate set; if 0.3 ≤ K(t) < 0.5, enter the mild underload F1 candidate set; if 0.5 ≤ K(t) ≤ 0.7, enter the steady-state balance F2 candidate set; if 0.7 < K(t) ≤ 0.9, enter the light overload F3 candidate set; if K(t) > 0.9, enter the deep overload F4 candidate set.

[0081] In this embodiment, several valid scenario sets are generated: Multi-condition cross-validation: F0 scenario validation: Based on the initial screening, the following must be met simultaneously: Energy storage SOC < 20% (preset energy storage safety lower limit to avoid over-discharge); Load fluctuation index < 10% (stable load but severely insufficient supply); Voltage deviation rate > 15% and frequency deviation rate > 15% (grid on the verge of collapse). F1 scenario validation: 20% ≤ Energy storage SOC < 50% (energy storage has some adjustment space but is insufficient); 10% ≤ Load fluctuation index ≤ 20% (moderate load fluctuation); Voltage deviation rate > 10% or frequency deviation rate > 10% (single grid parameter anomaly). F2 Scenario Verification: 50% ≤ Energy Storage SOC ≤ 80% (ideal energy storage range, supporting bidirectional adjustment); Load fluctuation index < 20% (stable load); Voltage deviation rate ≤ 10% and frequency deviation rate ≤ 10% (stable grid operation); F3 Scenario Verification: Energy Storage SOC > 50% (sufficient energy storage, supporting discharge); Load fluctuation index > 20% (severe load fluctuation, requiring rapid response); Voltage deviation rate > 15% or frequency deviation rate > 15% (grid issues warning signal); F4 Scenario Verification: No energy storage / load conditions required, as long as: Voltage deviation rate > 20% and frequency deviation rate > 20% (severe grid anomaly), or a device tripping signal is detected (emergency fault).

[0082] In this embodiment, the preset priority is: (F4 > F3 > F2 > F1 > F0).

[0083] In this embodiment, the boundary scenario handling is as follows: when the comprehensive adjustment parameter falls on the threshold boundary (such as 0.3 / 0.5 / 0.7 / 0.9), the grid deviation condition is used as the main judgment basis (safety priority principle); when the energy storage SOC is at the critical value (such as 20% / 80%), the scenario level is adjusted in combination with the load fluctuation index (high fluctuation is classified into a higher level).

[0084] In this embodiment, the parameter comparison unit first compares the comprehensive adjustment parameters in the 0-1 range with preset five-level scenario thresholds to quickly generate a preliminary candidate set. The preset thresholds are based on power system safety specifications (such as the 20%-80% SOC range for energy storage and the national standard voltage / frequency deviation allowable values) and are divided into: deep underload (<0.3), slight underload (0.3-0.5), steady-state balance (0.5-0.7), slight overload (0.7-0.9), and deep overload (>0.9). For example, a comprehensive parameter of 0.6 is included in the "steady-state balance candidate set," and 0.85 is included in the "slight overload candidate set." The cross-validation unit performs multi-condition verification on each candidate set, combining characteristic parameters such as energy storage status (SOC / SOH), load fluctuation index, and grid deviation rate to eliminate misjudgments based on a single parameter. For example, the deep underload candidate set must simultaneously meet the following conditions: energy storage SOC < 20% (insufficient energy storage), load fluctuation index < 10% (stable load but insufficient supply), and voltage and frequency deviation rates both > 15% (severe grid anomaly); the mild overload candidate set must meet the following conditions: energy storage SOC > 50% (sufficient energy storage), load fluctuation index > 20% (severe load fluctuation), and voltage or frequency deviation rate > 15% (single grid early warning). If a transformer substation has a comprehensive parameter of 0.8 (mild overload candidate) but an energy storage SOC of 40%, it will be excluded because it does not meet the energy storage conditions. The tag generation unit generates a unique scenario tag for the verified scenarios according to a preset priority (deep overload F4 > mild overload F3 > steady-state balance F2 > mild underload F1 > deep underload F0) to ensure that emergency scenarios take priority. If the comprehensive parameter is at the threshold boundary (e.g., 0.3), the grid deviation condition is the main criterion for judgment; if multiple scenario conditions are met simultaneously, the highest priority is used (e.g., if a device tripping signal is detected, it is marked as F4 even if the comprehensive parameter is high). For example, if the comprehensive parameter is 0.45 and meets the requirements of 35% SOC for energy storage, 18% load fluctuation, and 12% voltage deviation, a "slight underload F1" label is generated; if a frequency deviation of 25% also occurs, it is preferentially marked as "deep overload F4".

[0085] The beneficial effects of the above technical solution are as follows: Through a multi-level analysis method involving parameter comparison, cross-validation, and label generation, the accuracy and adaptability of scene labels are significantly improved. Compared with existing technologies, cross-validation is used to screen effective scenes, avoiding the problems of inaccurate scene identification or poor adaptability in traditional methods. The generated unique scene labels can more accurately reflect the real-time status of the power grid, ensuring more refined and flexible regulation strategies, and effectively improving the precision and stability of power grid regulation.

[0086] The parameter determination unit of the intelligent regulation system for source-load-storage in transformer substations based on multi-scenario collaboration includes:

[0087] Data preprocessing subunit: Normalizes the feature parameters;

[0088] Supply determination subunit: Determine the supply-side stability index based on the normalized photovoltaic output correction coefficient, wind power availability, energy storage available capacity, and charge / discharge efficiency correction coefficient;

[0089] ;

[0090] in, As a supply-side stability index, This is a correction factor for photovoltaic output. Wind power availability This represents the percentage of available energy storage capacity. This is a correction factor for charge / discharge efficiency;

[0091] Demand Determination Subunit: Determine the demand-side adjustment potential index based on the adjustable load ratio, load fluctuation index, voltage deviation rate, and frequency deviation rate;

[0092] ;

[0093] in, This is a potential index for demand-side adjustment. To allow for adjustable load proportions, For load fluctuation index, This is a correction value for the load fluctuation index. These are correction values ​​for voltage deviation rate and frequency deviation rate. Voltage deviation rate, Frequency deviation rate;

[0094] Parameter determination sub-unit: Comprehensive adjustment parameters are determined based on the supply-side stability index, the demand-side adjustment potential index, and environmental impact factors.

[0095] ;

[0096] in, To comprehensively adjust parameters, These are environmental impact factors.

[0097] In this embodiment, the supply-side stability index primarily measures the stability of the coordinated operation of the power source (such as photovoltaic and wind power) and energy storage. It focuses on the temperature-induced correction of photovoltaic power, the actual availability of wind power, and the available capacity ratio and charging / discharging efficiency of energy storage. By combining the minimum power source stability (reflecting the weakest link effect) with the effectiveness of energy storage, it reflects the ability of the power source and energy storage to maintain system stability in synergy. A higher value indicates a more stable and reliable supply side.

[0098] In this embodiment, the demand-side adjustment potential index reflects the demand side's ability to support system regulation. It comprehensively considers the proportion of adjustable load in the total load (the higher the proportion, the greater the adjustment flexibility), load power fluctuations (the smaller the fluctuations, the easier it is to release the adjustment potential), and the degree of deviation of grid voltage and frequency (the smaller the deviation, the stronger the grid stability's support for regulation). It is the product of load adjustability, stability, and grid stability. The larger the value, the more powerful the support that the demand side can provide in regulation, and the more conducive it is to the optimized operation of the system.

[0099] In this embodiment, The minimum value of the stability of photovoltaic and wind power (the weakest link effect) is taken, and then multiplied by the energy storage effectiveness (the product of the available capacity ratio and the efficiency correction) to reflect the stability of the coordinated operation of the power source side (photovoltaic and wind power) and energy storage.

[0100] In this embodiment, the percentage of available energy storage capacity is the ratio of available energy storage capacity to rated capacity, ranging from [0,1].

[0101] In this embodiment, the supply-side stability index follows the weakest link effect and takes the photovoltaic output correction coefficient. (Affected by temperature, through) The minimum of the normalized (normalized) and wind power availability (actual output to rated output ratio) is multiplied by the product of the energy storage available capacity ratio (available capacity to rated capacity ratio) and the charge / discharge efficiency correction coefficient (efficiency after temperature influence). This reflects the stability of the power supply side (photovoltaic, wind power) and energy storage coordination. Weakness in any link will lower the overall stability. The demand-side adjustment potential index consists of three parts: adjustable load ratio (adjustable load to total load ratio, reflecting adjustment flexibility), load fluctuation index correction value (complement of 10-minute power fluctuations, the greater the fluctuation, the lower the potential), and voltage / frequency deviation correction value (deviation rate complement, normalized to reflect grid stability). The product of these three reflects the demand side's support capability for adjustment. When the load is highly adjustable, the fluctuation is small, and the grid is stable, the potential is greater. The environmental impact factor is generated by normalizing parameters such as temperature and humidity. The higher the value, the stronger the environmental support for power output (such as photovoltaic and wind power conditions). For example, it is close to 1 when there is sufficient sunshine on a sunny day. The final comprehensive adjustment parameter is the product of the above three factors, which integrates supply-side stability, demand-side adjustment potential, and environmental support. It comprehensively reflects the integrated adjustment capability of the source-load-storage system. The larger the value, the more conducive the system is to optimized adjustment under the synergy of multiple factors, providing a quantitative basis for the operation and control of the distribution area.

[0102] The beneficial effects of the above technical solution are as follows: By introducing precise definitions of data preprocessing, supply-side stability index, and demand-side adjustment potential index, the shortcomings of imprecise single-factor regulation in existing technologies are overcome. The combination of supply-side stability and demand-side adjustment potential index makes the regulation of the power grid under different load conditions more scientific and reasonable, while taking into account the impact of environmental factors, further improving the accuracy and flexibility of power grid regulation. This dynamic determination of comprehensive regulation parameters ensures the stability of the power grid and the optimization of energy distribution under changing conditions.

[0103] The data prediction module of the intelligent regulation system for power generation, load and storage in transformer substations based on multi-scenario collaboration includes:

[0104] Curve generation unit: acquires historical load data for a preset time period collected through a preset smart sensor network, and generates a predicted load curve based on a preset model;

[0105] Time Period Determination Unit: Determines off-peak periods based on predicted load curves and energy storage system data;

[0106] Charging quantity determination unit: Determines the basic charging quantity during off-peak hours based on energy storage system data and distributed power source data;

[0107] Parameter adjustment unit: Adjusts the basic charging amount based on scene tags to determine the charging amount parameters during off-peak periods.

[0108] In this embodiment, the load trough period is determined based on the predicted load curve and energy storage system data by extracting the period in the predicted load curve where the load is less than 60% of the rated capacity, and using this period as the load trough period.

[0109] In this embodiment, the basic charging amount during off-peak hours is determined based on energy storage system data and distributed power source data: Charging amount calculation: Safety capacity limit: Charging limit = 0.8Erated−Eavail (preset energy storage safety charging range 20%-80% SOC); Power difference constraint: Photovoltaic chargeable power = Photovoltaic predicted output − predicted load (to avoid charging affecting real-time power supply); Charging amount parameter: Echa_plan = min(charging limit, photovoltaic chargeable power × off-peak period duration). Safety capacity limit parameter (charging limit = 0.8Erated−Eavail) Erated (rated energy storage capacity) Source: Inherent parameters of the energy storage device (hardware nameplate parameters, such as battery pack rated capacity) input during system initialization, which are preset fixed values. Example: If the rated capacity of the energy storage system is 100 kWh, then Erated = 100 kWh. Eavail (available energy storage capacity) Source: The current available capacity of energy storage (unit: kWh) collected in real time by the data acquisition module, reflecting the remaining space that the energy storage can currently use for charging. Data Link: Smart Sensors (e.g., Energy Storage BMS Sensors) → Data Acquisition Module → Real-time Synchronization to Charging Planning Subunit. Photovoltaic Rechargeable Power = Photovoltaic Predicted Output − Predicted Load. Basic Constraint Parameters: Off-Peak Period Duration Source: The length of a continuous time period (unit: hours) extracted from the off-peak periods (periods when load < 60% of rated capacity) marked by the load prediction subunit. Example: If the off-peak period is 2:00-5:00, then the duration is 3 hours. Rated Capacity (used for defining off-peak periods) Source: The system's preset rated power supply capacity for the distribution area (unit: kW), i.e., the rated power of the distribution transformer, which is a system initial configuration parameter (e.g., 500 kW); Charging Parameter: Echa_plan = min(Charging Upper Limit, Photovoltaic Rechargeable Power × Off-Peak Period Duration).

[0110] In this embodiment, the charging quantity parameter for the off-peak period is determined by adjusting the basic charging quantity based on the scene label, including: scene linkage adjustment: if the scene of the transformer area is deep underload (F0) or slight underload (F1), the charging quantity parameter is increased by 20% (prioritizing energy storage); if it is slight overload (F3) or deep overload (F4), the charging quantity parameter is decreased by 30% (to avoid energy storage overload).

[0111] The beneficial effects of the above technical solution are as follows: By introducing a curve generation unit, a time period determination unit, and a charging quantity determination unit, the load curve is accurately predicted and the off-peak period is determined, thereby improving the accuracy of load forecasting and the rationality of charging quantity. Compared with existing technologies, it can dynamically adjust charging quantity parameters and optimize the adjustment strategy by combining scenario tags, ensuring accurate charging under different grid conditions. This avoids the problem of inaccurate charging quantity estimation in traditional methods and significantly improves the energy utilization efficiency and adjustment response speed of the power grid.

[0112] The strategy update module of the intelligent regulation system for source-load-storage in transformer substations based on multi-scenario collaboration includes:

[0113] Deviation determination unit: acquires the real-time load curve after strategy execution and determines the adjustment deviation rate based on the predicted load curve;

[0114] Curve adjustment unit: Based on the adjustment deviation rate, the predicted load curve of the data prediction module is corrected. If the deviation rate is positive, it indicates that the actual load is higher than the prediction, and the prediction curve is adjusted upward. If the deviation rate is negative, the prediction curve is adjusted downward.

[0115] Strategy Re-execution Subunit: Based on the corrected predicted load curve, combined with the charging quantity parameters during off-peak periods and scenario labels, the control strategy in the strategy execution module is re-executed.

[0116] In this embodiment, obtaining the real-time load curve after strategy execution and determining the adjustment deviation rate based on the predicted load curve includes: dividing the difference between the real-time load curve and the predicted load curve at the same time point by the predicted load value to obtain the deviation rate at that time point, and then determining the average adjustment deviation rate for the entire period. This deviation rate reflects the gap between the strategy execution effect and the expectation, providing key data for subsequent sub-units.

[0117] The beneficial effects of the above technical solution are as follows: By analyzing the deviation between the real-time load curve and the forecast curve, dynamic correction of load forecasting is achieved. Compared with existing technologies, it can automatically adjust the forecast curve according to the regulation deviation rate, thereby optimizing subsequent regulation strategies and avoiding the regulation failure or inefficiency caused by forecast errors in traditional methods. Through strategy re-execution, the system can adapt to changes in the power grid in real time, improving the accuracy and flexibility of regulation and ensuring the efficient and stable operation of the power grid.

[0118] The strategy re-execution subunit of the intelligent regulation system for source-load-storage in transformer substations based on multi-scenario collaboration includes:

[0119] Strategy matching block: Based on the modified predicted load curve scenario labels, it matches the corresponding control strategy from the preset strategy library;

[0120] Parameter determination block: Based on the control strategy obtained through matching and the charging quantity parameters during off-peak hours, the specific control parameters of each source-load-storage device are determined.

[0121] Instruction execution block: Based on specific control parameters, it sends control instructions to each source, load, and storage device, thereby re-executing the strategy.

[0122] In this embodiment, the corresponding control strategy is matched: different scenario labels (such as deep underload, light overload, etc.) correspond to different basic control logic, while the corrected predicted load curve provides more accurate load peak and valley information, thereby determining the adjustment direction and approximate range of each link of source, load and storage.

[0123] In this embodiment, the preset strategy library is a pre-built "strategy toolbox" that stores corresponding source-load-storage coordinated control strategies for different scenario labels (such as deep underload, slight overload, etc.). For example, when the scenario label is "slight overload scenario", the pre-stored strategy in the strategy library may be: prioritize increasing the energy storage discharge power to supplement the power gap, while reducing the power demand of non-critical adjustable loads, and appropriately limiting the unplanned additional output of distributed power sources (such as photovoltaics) when necessary. If the corrected predicted load curve shows that a slight overload will occur during a certain period, the strategy matching block quickly matches this strategy from the library; then the parameter determination block combines the charging quantity parameters during the off-peak period to accurately calculate the specific power of energy storage discharge, the reduction amount of adjustable load, etc.; finally, the instruction execution block sends an instruction to increase the discharge power to the energy storage device and an instruction to reduce the power consumption to the adjustable load device, completing the strategy re-execution and alleviating the overload situation. For example, in a "deep underload scenario", the strategy in the strategy library may be to prioritize the use of backup power and increase the energy storage charging power (if conditions permit). The parameter determination block calculates the backup power access power, energy storage charging parameters, etc., based on this, and the instruction execution block drives the device to execute, thereby realizing system adjustment.

[0124] In this embodiment, determining the specific control parameters for each source-load-storage device includes: Power Parameter Determination Sub-block: Based on the matched control strategy and the charging quantity parameters during off-peak hours, the specific parameters of the distributed power source are determined. If the system is in a mild overload scenario and the strategy requires full-power operation of the photovoltaic system, the upper limit of photovoltaic output in each time period is calculated, considering the availability of surplus electricity during off-peak hours for energy storage charging. If the charging demand is high during off-peak hours, the photovoltaic output in the corresponding time period can be appropriately increased to meet the energy storage charging demand. This parameter provides a basis for subsequent judgment on whether the power source meets system requirements. Energy Storage Parameter Determination Sub-block: Based on the power output obtained from the power parameter determination sub-block, combined with the control strategy and the charging quantity parameters during off-peak hours, the charging and discharging parameters of the energy storage device are determined. If the power source has sufficient output and surplus electricity during a certain time period, the charging power and duration of the energy storage are determined based on the charging quantity during off-peak hours; if the system is under load and the power source output is insufficient, the discharging power and duration of the energy storage are determined. These parameters ensure proper charging and discharging of energy storage, maintaining system stability. The load parameter determination sub-block: Based on the results of the power supply parameter determination sub-block and the energy storage parameter determination sub-block, combined with the control strategy, the parameters of the adjustable load are determined. If the total power of the power supply and energy storage still cannot meet peak load demand, the adjustable load is adjusted according to the strategy, determining the load reduction amount and reduction period to ensure system supply and demand balance; for example, for energy storage devices, the charging power and discharging power at different times are determined based on charging parameters and the corrected load curve; for distributed power sources, the magnitude of their output adjustment is determined, etc. These parameters are the key basis for the accurate execution of the control strategy.

[0125] In this embodiment, the strategy is re-executed. During the execution process, the operating status of the equipment is monitored in real time, and feedback information is sent back. If a deviation is found between the actual execution and the expected parameters, it is promptly fed back to the strategy matching sub-unit so that the strategy can be adjusted again, forming a dynamic closed-loop adjustment process to ensure that the system can always adjust accurately according to actual load changes.

[0126] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; 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; and these 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.

Claims

1. A smart regulation system for source-load-storage in transformer substations based on multi-scenario collaboration, characterized in that, include: Data acquisition module: Collects power supply, load and power grid related data through a preset intelligent sensor network and generates corresponding characteristic parameters; Data analysis module: Determines comprehensive adjustment parameters based on feature parameters, generates the distribution of transformer area scenes by combining preset thresholds, obtains several scenes and generates corresponding scene labels; Data prediction module: acquires historical load data for a preset time period collected through a preset smart sensor network, generates a predicted load curve, and then determines the charging quantity parameters during off-peak periods. Strategy execution module: Executes the control strategy corresponding to the scenario label based on the predicted load curve, charging quantity parameters during off-peak periods, and scenario label; Strategy update module: Compare the real-time load curve after strategy execution with the predicted load curve, determine the adjustment deviation rate, correct the predicted load curve based on the adjustment deviation rate, and then re-execute the control strategy; The data analysis module includes: Parameter determination unit: determines comprehensive adjustment parameters based on characteristic parameters; Parameter comparison unit: compares the comprehensive adjustment parameters with the preset threshold to generate several first candidate sets for preliminary screening; Cross-validation unit: Cross-validation is performed based on each first candidate set and several feature parameters corresponding to each candidate set to generate several valid scene sets; Tag generation unit: For the set of valid scenarios that have passed cross-validation, generate unique scenario tags according to preset priorities and generate corresponding scenario tags; The parameter determination unit includes: Data preprocessing subunit: Normalizes the feature parameters; Supply determination subunit: Determine the supply-side stability index based on the normalized photovoltaic output correction coefficient, wind power availability, energy storage available capacity, and charge / discharge efficiency correction coefficient; ； in, As a supply-side stability index, This is a correction factor for photovoltaic output. Wind power availability This represents the percentage of available energy storage capacity. This is a correction factor for charge / discharge efficiency; Demand Determination Subunit: Determine the demand-side adjustment potential index based on the adjustable load ratio, load fluctuation index, voltage deviation rate, and frequency deviation rate; ； in, This is a potential index for demand-side adjustment. To allow for adjustable load proportions, For load fluctuation index, This is a correction value for the load fluctuation index. These are correction values ​​for voltage deviation rate and frequency deviation rate. Voltage deviation rate, Frequency deviation rate; Parameter determination sub-unit: Determining comprehensive adjustment parameters based on supply-side stability index, demand-side adjustment potential index, and environmental impact factors. ； in, To comprehensively adjust parameters, These are environmental impact factors.

2. The intelligent regulation system for source-load-storage in transformer substations based on multi-scenario collaboration according to claim 1, characterized in that, Scene tags include: deep underload scene, light underload scene, steady-state equilibrium scene, light overload scene, and deep overload scene.

3. The intelligent regulation system for source-load-storage in transformer substations based on multi-scenario collaboration according to claim 1, characterized in that, The data acquisition module includes: Parameter acquisition unit: Collects parameters of preset types, including: distributed power source data, energy storage system data, load classification data, and environmental data; Feature generation unit: Generates corresponding feature parameters based on preset type parameters. The feature parameters include: photovoltaic output correction coefficient, wind power availability, energy storage available capacity, charge and discharge efficiency correction coefficient, adjustable load ratio, load fluctuation index, voltage deviation rate, frequency deviation rate, and environmental impact factor.

4. The intelligent regulation system for source-load-storage in transformer substations based on multi-scenario collaboration according to claim 3, characterized in that, The data prediction module includes: Curve generation unit: acquires historical load data for a preset time period collected through a preset smart sensor network, and generates a predicted load curve based on a preset model; Time Period Determination Unit: Determines off-peak periods based on predicted load curves and energy storage system data; Charging quantity determination unit: Determines the basic charging quantity during off-peak hours based on energy storage system data and distributed power source data; Parameter adjustment unit: Adjusts the basic charging amount based on scene tags to determine the charging amount parameters during off-peak periods.

5. The intelligent regulation system for source-load-storage in transformer substations based on multi-scenario collaboration according to claim 1, characterized in that, The policy update module includes: Deviation determination unit: acquires the real-time load curve after strategy execution and determines the adjustment deviation rate based on the predicted load curve; Curve adjustment unit: Based on the adjustment deviation rate, the predicted load curve of the data prediction module is corrected. If the deviation rate is positive, it indicates that the actual load is higher than the prediction, and the prediction curve is adjusted upward. If the deviation rate is negative, the prediction curve is adjusted downward. Strategy Re-execution Subunit: Based on the corrected predicted load curve, combined with the charging quantity parameters during off-peak periods and scenario labels, the control strategy in the strategy execution module is re-executed.

6. The intelligent regulation system for source-load-storage in transformer substations based on multi-scenario collaboration according to claim 4, characterized in that, The policy re-execution subunit includes: Strategy matching block: Based on the modified predicted load curve scenario labels, it matches the corresponding control strategy from the preset strategy library; Parameter determination block: Based on the control strategy obtained through matching and the charging quantity parameters during off-peak hours, the specific control parameters of each source-load-storage device are determined. Instruction execution block: Based on specific control parameters, it sends control instructions to each source, load, and storage device, thereby re-executing the strategy.

Images