Wind and light storage station power reporting and tracking method based on energy storage rapid power response

Abstract

The invention discloses a wind and light storage station power reporting and tracking method based on energy storage rapid power response. The method comprises the following steps: acquiring wind and light power probability prediction data, power grid real-time operation state data, station operation state data and power grid examination allowable power deviation data in a current scheduling period; generating a dynamic power reporting interval of the next scheduling period based on the wind-solar power probability prediction data, the power grid real-time operation state data, the station operation state data and the power grid examination allowable power deviation data; and based on the dynamic power reporting interval and the real-time operation data of the wind and light storage station, a flexible control strategy is adopted to control the charging and discharging behaviors of an energy storage system in the station. By generating the dynamic power reporting interval with self-adaptive change and adopting the flexible control strategy, the operation condition of the energy storage system can be obviously optimized, the fluctuation of the wind and light power can be effectively stabilized, and the economy and reliability of the operation of the station can be improved on the premise of ensuring that the wind and light storage station tracks the power grid assessment requirement.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of smart grid control, and particularly relates to a wind-solar-storage power station power reporting tracking method based on fast power response of energy storage. BACKGROUND

[0002] With the increasing penetration of renewable energy such as wind and solar energy in power systems, wind-solar-storage combined power stations have become an important component of supporting the stable operation of new power systems. Such stations are usually composed of wind power generation, photovoltaic power generation and energy storage systems, and realize power smoothing, tracking of planned output and participation in grid auxiliary services through coordinated control. However, wind-solar power output has significant intermittency and volatility, and power prediction has great uncertainty, which makes it a serious challenge for the station to execute grid dispatch instructions and meet power assessment requirements.

[0003] Currently, wind-solar-storage power stations usually report power plans based on fixed power intervals or deterministic prediction results, and lack dynamic response capability to wind-solar uncertainty, real-time grid state and energy storage operating state. In actual operation, if the reporting interval is too wide, although the assessment risk can be reduced, it may lead to frequent action of energy storage, imbalance of state of charge, and shorten the service life of equipment; if the reporting interval is too narrow, it is easy to cause assessment penalties due to wind-solar fluctuations exceeding the interval, and the energy storage system needs to bear more pressure to compensate for the deviation, affecting its long-term operation economy and reliability.

[0004] In addition, existing control strategies mostly use rigid threshold control, that is, when the power at the grid connection point exceeds the limit, the energy storage is forced to charge and discharge to pull back into the interval, and lack the ability to coordinate and optimize multiple factors such as state of charge of energy storage, grid frequency change and dispatch instructions. This control method is difficult to adapt to the higher requirements of grid on station operation flexibility, fast response capability and intelligent level under high proportion of renewable energy access. SUMMARY

[0005] The purpose of the embodiments of the present application is to provide a wind-solar-storage power station power reporting tracking method based on fast power response of energy storage, which can significantly optimize the operating conditions of the energy storage system, effectively smooth the wind-solar power fluctuations, and improve the economy and reliability of the station operation by generating dynamically adaptive power reporting intervals and using a flexible control strategy while ensuring that the wind-solar-storage power station tracks the grid assessment requirements.

[0006] To solve the above technical problems, the first aspect of the embodiments of the present application provides a wind-solar-storage power station power reporting tracking method based on fast power response of energy storage, comprising the following steps:

[0007] acquire wind and light power probability prediction data, grid real-time operation state data, station own operation state data and grid examination allowed power deviation data in a current scheduling period;

[0008] generate a dynamic power reporting interval of a next scheduling period based on the wind and light power probability prediction data, the grid real-time operation state data, the station own operation state data and the grid examination allowed power deviation data, the dynamic power reporting interval being a safe operation power range defined by an adaptive upper boundary value and a lower boundary value;

[0009] based on the dynamic power reporting interval and real-time operation data of the wind and light storage station, a flexible control strategy is used to control the charge and discharge behavior of the energy storage system in the station, the flexible control strategy being used to optimize the operation condition of the energy storage system on the premise that the grid-connected point power of the station is maintained between the upper limit power value and the lower limit power value.

[0010] Further, the generation of the dynamic power reporting interval of the next scheduling period based on the wind and light power probability prediction data, the grid real-time operation state data, the station own operation state data and the grid examination allowed power deviation data comprises:

[0011] determining a basic prediction interval defined by a confidence quantile point based on the wind and light power probability prediction data;

[0012] adjusting the upper boundary value of the basic prediction interval according to grid scheduling instruction information and grid frequency monitoring information in the grid real-time operation state data;

[0013] adjusting the bandwidth of the basic prediction interval based on the state of charge information of the energy storage system in the station own operation state data;

[0014] adjusting the bandwidth of the basic prediction interval in combination with the grid examination allowed power deviation data to obtain the dynamic power reporting interval.

[0015] Further, the determination of the basic prediction interval defined by the confidence quantile point based on the wind and light power probability prediction data comprises:

[0016] acquiring probability distribution data of wind and light power at a predetermined confidence level;

[0017] acquiring a quantile value corresponding to a preset high confidence level from the probability distribution data as an interval upper boundary reference value;

[0018] extracting a quantile value corresponding to a preset low confidence level from the probability distribution data as an interval lower boundary reference value;

[0019] An initial bandwidth of the basic prediction interval is obtained from a difference between the upper boundary reference value and the lower boundary reference value of the interval.

[0020] Further, the adjusting the upper boundary value of the basic prediction interval according to the power grid scheduling instruction information and the power grid frequency monitoring information in the real-time operation state data of the power grid comprises:

[0021] analyzing the active power control instruction in the power grid scheduling instruction information, when the active power control instruction is a power reduction operation instruction, calculating a first adjustment amount for reducing the upper boundary of the basic prediction interval according to a power reduction ratio required by the active power control instruction;

[0022] monitoring a real-time frequency value in the power grid frequency monitoring information, when it is detected that the power grid frequency exceeds a preset frequency threshold value, calculating a second adjustment amount for reducing the upper boundary of the basic prediction interval based on a difference between the real-time frequency value and the preset frequency threshold value;

[0023] based on the first adjustment amount and the second adjustment amount, performing a reduction processing on the upper boundary value of the basic prediction interval.

[0024] Further, the adjusting the bandwidth of the basic prediction interval based on the state of charge information of the energy storage system in the self-operation state data of the field station comprises:

[0025] real-time obtaining a current state of charge value of the energy storage system;

[0026] when the current state of charge value is lower than a first preset threshold value, calculating a lower boundary down adjustment amount according to a capacity deficiency, and accordingly down adjusting a lower boundary value of the basic prediction interval;

[0027] when the current state of charge value is higher than a second preset threshold value, calculating an upper boundary up adjustment amount according to a capacity margin, and accordingly up adjusting an upper boundary value of the basic prediction interval;

[0028] wherein, the first preset threshold value is lower than the second preset threshold value.

[0029] Further, the adjusting the bandwidth of the basic prediction interval in combination with the power grid examination allowed power deviation data comprises:

[0030] obtaining a power deviation cumulative value that has occurred in a current examination period and a period allowed deviation total amount in the power grid examination allowed power deviation data;

[0031] calculating a ratio between the power deviation cumulative value and the period allowed deviation total amount to obtain a real-time examination risk coefficient;

[0032] calculate a bandwidth adjustment weight based on the real-time examination risk coefficient, wherein the bandwidth adjustment weight is negatively correlated with the real-time examination risk coefficient;

[0033] dynamically scale the bandwidth of the basic prediction interval according to the bandwidth adjustment weight, and a scaling direction of the bandwidth is opposite to a change direction of the real-time examination risk coefficient.

[0034] Further, the dynamically scaling the bandwidth of the basic prediction interval according to the bandwidth adjustment weight comprises:

[0035] calculate a scaling ratio based on the bandwidth adjustment weight and a current bandwidth value of the basic prediction interval, wherein the scaling ratio is positively correlated with the bandwidth adjustment weight;

[0036] equi-proportionally scale upper and lower boundary values of the basic prediction interval according to the scaling ratio, and keep a center position of the interval unchanged;

[0037] set a minimum bandwidth protection threshold, and when a scaled bandwidth value is less than the minimum bandwidth protection threshold, forcibly adjust the bandwidth to the minimum bandwidth protection threshold while keeping the center position of the interval unchanged;

[0038] update the upper and lower boundary values of the basic prediction interval.

[0039] Further, after the bandwidth of the basic prediction interval is adjusted in combination with the grid examination allowed power deviation data, the method further comprises:

[0040] construct a conflict decision objective function, wherein input parameters of the conflict decision objective function comprise an upper boundary adjustment amount required by a grid dispatching instruction, an upper boundary adjustment amount required by a grid frequency out-of-limit, a bandwidth adjustment amount required by a state of charge of the energy storage system, and a bandwidth adjustment amount required by the examination risk coefficient;

[0041] calculate an optimal solution of the conflict decision objective function based on a multi-objective optimization algorithm, to obtain a coordination weight coefficient of each adjustment amount;

[0042] weight and fuse each adjustment amount according to the coordination weight coefficient, to generate a final upper boundary adjustment instruction and a final bandwidth adjustment instruction;

[0043] update the basic prediction interval based on the final upper boundary adjustment instruction and the final bandwidth adjustment instruction, to obtain an optimized dynamic power reporting interval.

[0044] Further, the calculating the optimal solution of the conflict decision objective function based on the multi-objective optimization algorithm to obtain the coordination weight coefficient of each adjustment amount comprises:

[0045] Determine the priority order of each adjustment amount, and the priority of the adjustment amount is from high to low: the adjustment amount required by the power grid dispatching instruction, the adjustment amount required by the power grid frequency out-of-limit, the adjustment amount required by the risk coefficient, and the adjustment amount required by the state of charge of the energy storage system;

[0046] According to the priority order, all the adjustment amounts are assigned initial weight coefficients, and the higher the priority of the adjustment amount, the greater the initial weight coefficient assigned to the adjustment amount;

[0047] An optimization solving model containing all the initial weight coefficients is established, and the optimization solving model is solved by an iterative optimization algorithm to obtain the coordinated weight coefficients of all the adjustment amounts that satisfy all the constraint conditions and make the objective function optimal.

[0048] Further, based on the dynamic power reporting interval and the real-time operation data of the wind-solar-storage station, a flexible control strategy is used to control the charging and discharging behavior of the energy storage system in the station, including:

[0049] The actual value of the grid-connected point power of the wind-solar-storage station is collected in real time, and the instantaneous deviation amount of the actual value from the upper and lower boundary values of the dynamic power reporting interval is calculated;

[0050] According to the size and direction of the instantaneous deviation amount, a power regulation preliminary instruction of the energy storage system is generated;

[0051] The real-time state of charge value of the energy storage system is obtained, and the power regulation preliminary instruction is adaptively corrected based on the real-time state of charge value to obtain a power regulation correction instruction;

[0052] According to the power regulation correction instruction, the power output of the energy storage converter is controlled to make the actual value of the grid-connected point power approach the center value of the dynamic power reporting interval;

[0053] When the actual value of the grid-connected point power continuously stays within the range of the dynamic power reporting interval within a preset time threshold, the power regulation strength of the energy storage system is reduced, and the state of charge of the energy storage system is preferentially maintained stable.

[0054] Correspondingly, a second aspect of an embodiment of the present application provides an electronic device, comprising: at least one processor; and a memory connected with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to make the at least one processor execute the wind-solar-storage station power reporting tracking method based on the energy storage fast power response.

[0055] Accordingly, a third aspect of the embodiment of the present application provides a computer readable storage medium having stored thereon computer instructions which, when executed by a processor, implement the wind and light storage station power reporting tracking method based on fast power response of energy storage.

[0056] The above technical solution of the embodiment of the present application has the following beneficial technical effects:

[0057] 1. By comprehensively considering wind and light probability prediction, real-time state of the power grid, SOC of energy storage, and historical assessment deviation data, a power reporting interval is dynamically generated and optimized in real time, which not only avoids the risk of assessment punishment caused by too narrow interval, but also prevents unnecessary frequent action of the energy storage system caused by too wide interval. The dynamic optimization mechanism fundamentally realizes the maximum reduction of the number and depth of charging and discharging of the energy storage under the premise of meeting the requirements of the power grid assessment, thereby prolonging the service life of the energy storage, reducing the operation and maintenance cost, and improving the economy and reliability of the entire station.

[0058] 2. The traditional control strategy is difficult to coordinate the mutually conflicting goals such as dispatching instruction power reduction, frequency out-of-limit, SOC recovery, and assessment risk avoidance. By constructing a conflict decision objective function and using a multi-objective optimization algorithm, the adjustment requirements from different sources are quantified and weightedly fused to generate a coordinated optimal adjustment instruction. The method can intelligently determine which demand should be met with what priority in different scenarios, thereby making a comprehensive decision that is most beneficial to the stability of the power grid and the operation of the station itself, and avoiding the performance deterioration in other aspects caused by a single strategy.

[0059] 3. Through the flexible control strategy, not only the instantaneous power deviation is adjusted, but also the real-time SOC of the energy storage is introduced as a correction factor, and the control intensity after long-time stable operation is optimized, so that the adjustment behavior of the energy storage is more smooth and accurate. The method not only ensures that the grid-connected point power is effectively maintained within the dynamic interval, but also actively avoids overcharging and overdischarging of the energy storage, and preferentially maintains the SOC of the energy storage within the healthy interval, thereby completing the power support task while effectively protecting the energy storage device itself, and improving the intelligent and refined level of control. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 is a flow chart of the wind and light storage station power reporting tracking method based on fast power response of energy storage provided by the embodiment of the present application. DETAILED DESCRIPTION

[0061] To make the object, technical solutions and advantages of the present application clearer and more comprehensible, the present application is further described in detail below in conjunction with specific implementation manners and with reference to the drawings. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present application. In addition, in the following description, the description of well-known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present application.

[0062] Referring to Figure 1 The first aspect of the embodiments of the present application provides a wind-solar storage station power reporting tracking method based on energy storage fast power response, comprising the following steps:

[0063] In step S100, wind-solar power probability prediction data, real-time operation state data of the power grid, self-operation state data of the station, and power grid examination allowable power deviation data in the current dispatching period are obtained.

[0064] Specifically, multi-dimensional data needs to be obtained and integrated in real time from multiple data sources such as a station monitoring system (SCADA), a power prediction system, a power grid dispatching mechanism, and a battery management system (BMS). The wind-solar power probability prediction data not only includes point prediction values of wind farms and photovoltaic power stations in the future dispatching period, but more importantly, it includes probability distribution information, for example, prediction intervals given in the form of different confidence levels (such as 10% and 90%), which provides a core basis for quantifying the uncertainty of wind-solar output. The real-time operation state data of the power grid mainly includes the active power control instructions (such as power limit operation requirements) issued by the superior dispatching master station, as well as the current real-time frequency value of the power grid, which directly reflects the real-time operation demand and stable state of the power grid. The key of the self-operation state data of the station lies in the real-time state of charge (SOC) of the energy storage system, which represents the real-time available energy level of the energy storage and is the key constraint to determine its regulation capacity. The power grid examination allowable power deviation data is usually derived from the grid connection protocol or dispatching regulations, which clearly defines the allowable range of the power deviation between the station grid connection point and the reported plan value within a specific examination period (such as 15 minutes or 1 hour), which is an external standard for measuring whether the station operation is qualified.

[0065] In step S200, based on the wind-solar power probability prediction data, the real-time operation state data of the power grid, the self-operation state data of the station, and the power grid examination allowable power deviation data, a dynamic power reporting interval for the next dispatching period is generated, and the dynamic power reporting interval is a safe operation power range defined by an adaptive upper boundary value and a lower boundary value.

[0066] Step S200 dynamically optimizes a static, history-statistics-based power prediction interval into a safe and economic power reporting interval that can respond to multi-dimensional real-time operating conditions. The process begins with determining an initial basic prediction interval based on wind and solar power probability prediction data, which is bounded by high and low quantile points and preliminarily covers the main uncertainty of wind and solar output. Then, real-time grid operating state data is introduced for upper boundary adjustment. For example, when receiving a grid power reduction instruction or monitoring that the grid frequency is too high, the upper boundary of the power interval needs to be actively reduced to show the response capability of the station to grid frequency regulation or standby demand. Then, combined with the station's own state, mainly the SOC of the energy storage, the interval bandwidth is preliminarily adjusted: if the SOC is low, it indicates that the energy storage charging demand is urgent, and the lower boundary of the interval can be adjusted downward to increase the opportunity to absorb excess wind and solar power through energy storage charging; on the contrary, if the SOC is high, the upper boundary can be adjusted upward to reserve space for energy storage discharging. Finally, the grid examination allowed power deviation data is introduced to secondarily optimize the interval bandwidth: if the cumulative deviation in the current period has approached the allowed limit, the interval bandwidth is contracted to strictly control subsequent power fluctuations and avoid examination risks; if the cumulative deviation space is sufficient, the bandwidth can be appropriately relaxed to reduce the invalid action of the energy storage. The above adjustment targets may conflict, so a built-in multi-objective optimization algorithm is finally used to coordinate and weigh to calculate the comprehensive optimal dynamic power reporting interval upper and lower boundary values.

[0067] Step S300, based on the dynamic power reporting interval and the real-time operating data of the wind and solar energy storage station, a flexible control strategy is used to control the charging and discharging behavior of the energy storage system in the station. The flexible control strategy is used to optimize the operating conditions of the energy storage system while ensuring that the station grid point power is maintained between the upper and lower power values.

[0068] The rapid power regulation capability of the energy storage system is utilized to ensure that the power at the grid connection point of the station is strictly controlled within the dynamic power reporting interval generated by step S200, while the optimization of the energy storage itself is also considered. The flexible control strategy first continuously monitors the actual value of the power at the grid connection point and calculates the instantaneous deviation of the actual value from the upper and lower boundaries of the dynamic interval. Once the actual power has a tendency to exceed the limit or has exceeded the limit, the controller will immediately generate a preliminary power regulation instruction according to the size and direction of the deviation, instructing the energy storage system to compensate the power in the corresponding direction. Thereafter, instead of directly executing the preliminary instruction, the strategy introduces the real-time SOC of the energy storage as a key correction factor: if the SOC is already in an extremely low or high state, the amplitude of the charge and discharge instruction needs to be limited or corrected to prevent the energy storage from entering an unsafe state of overcharging or overdischarging, thereby protecting the energy storage device itself. After the accurate power instruction is issued to control the action of the PCS (Power Conversion System) of the energy storage, the controller will also continuously evaluate the control effect. If the power at the grid connection point remains stable within the interval for more than a predetermined time, the strategy will actively reduce the power regulation strength of the energy storage or put it into standby state, which aims to reduce the mechanical wear and energy loss of the energy storage, and to maintain the SOC stable, thereby optimizing the operating conditions and life of the energy storage.

[0069] By dynamically generating the power reporting interval by fusing multi-source information and adopting the flexible control strategy of collaborative optimization, the present application can fully respond to the real-time dispatching requirements and frequency support requirements of the power grid, effectively smooth the random fluctuations of wind and light output, ensure that the power at the grid connection point of the station meets the grid evaluation standard, and significantly reduce the evaluation risk. The method also considers the economy and safety of the station's own operation, reduces the action frequency and depth of the energy storage system through intelligent decision-making, avoids the drastic changes and out-of-limit risks of the state of charge, thereby optimizing the operating conditions of the energy storage system, prolonging its service life, and improving the comprehensive operating efficiency and friendliness to the power grid of the entire wind-solar-storage combined station.

[0070] Further, the dynamic power reporting interval for the next dispatching period is generated in step S200 based on the wind and light power probability prediction data, the real-time operating state data of the power grid, the operating state data of the station itself, and the power grid evaluation allowed power deviation data, which includes:

[0071] In step S210, the basic prediction interval defined by the confidence quantile is determined based on the wind and light power probability prediction data.

[0072] The rich uncertainty information provided by the probability prediction establishes a scientific and reasonable initial benchmark for subsequent adjustment. In specific implementation, the energy management system (EMS) of the station obtains the wind-solar combined output probability prediction result of a specific scheduling period (such as the next 15 minutes) from the power prediction server. The result is usually presented in the form of a probability distribution function or a large number of scenario sets. According to the pre-set high and low confidence levels (for example, the 90% and 10% quantile points can be selected), the corresponding power values are extracted from the probability distribution. The power value of the high confidence quantile point (such as 90%) is determined as the initial benchmark value of the upper boundary of the interval, which means that the actual wind-solar output has a higher probability of being lower than this value. Similarly, the power value of the low confidence quantile point (such as 10%) is taken as the initial benchmark value of the lower boundary of the interval. Thus, the range between the two quantile points constitutes the basic prediction interval, which statistically covers the wind-solar output values of the predetermined probability (such as 80%), providing an initial decision framework with safety and economy for the station's power plan reporting.

[0073] In step S220, the upper boundary value of the basic prediction interval is adjusted according to the grid scheduling instruction information and grid frequency monitoring information in the real-time operation state data of the grid.

[0074] The power reporting interval can actively respond to the real-time operation requirements and abnormal states of the grid, supporting the station as a friendly power source. The adjustment is mainly based on two aspects: one is to analyze the active power control instructions issued by the upper dispatching agency through the remote terminal unit (RTU) or communication gateway. When the instruction requires the station to operate at a lower power (for example, due to line congestion or grid balance needs), a corresponding adjustment amount is calculated according to the specified power reduction ratio or absolute value in the instruction, which is used to reduce the upper boundary of the basic prediction interval, thereby ensuring that the upper limit of the reported power interval meets the scheduling requirements of the grid. The second is to monitor the frequency measurement value of the grid in real time. When the detected frequency exceeds the pre-set safe operation threshold (such as 50.2 Hz), it indicates that there is a power shortage in the grid, and the station needs to reduce the output to assist in restoring the frequency. At this time, an additional adjustment amount (second adjustment amount) is calculated according to the size of the frequency deviation (the difference between the real-time frequency value and the rated frequency) with a pre-set adjustment coefficient, which further reduces the upper boundary of the interval, reserving necessary adjustment space for the station to quickly respond to primary frequency modulation. Finally, the upper boundary value of the basic prediction interval is reduced according to the two types of adjustment amounts.

[0075] In step S230, the bandwidth of the basic prediction interval is adjusted based on the state of charge information of the energy storage system in the station's own operation state data.

[0076] Step S230 focuses on optimizing the bandwidth of the basic prediction interval according to the energy state (SOC) of the energy storage system itself, with the purpose of maintaining the healthy operation and continuous regulation capability of the energy storage system while meeting external requirements. The average SOC value of the energy storage unit is obtained in real time through the battery management system (BMS). If the current SOC value is lower than the preset first safety threshold (for example, 20%), it indicates that the energy storage is insufficient in discharging energy and needs to be charged to restore capacity. At this time, a lower boundary downshift amount is calculated according to the capacity shortage, and the lower boundary value of the basic prediction interval is actively adjusted downward; when the actual wind and light output is low, the station is allowed to operate at a lower power value, thereby creating more charging opportunities (absorbing power from the grid) for the energy storage system. Conversely, if the current SOC value is higher than the preset second safety threshold (for example, 80%), it indicates that the energy storage has limited charging space and has discharging margin. An upper boundary upshift amount is calculated according to the capacity margin, and the upper boundary value of the basic prediction interval is adjusted upward accordingly, providing a larger operating space for the energy storage to discharge when the wind and light output is large, avoiding curtailment of wind and light, and also creating conditions for participating in grid peak regulation or standby power support. Through this two-way adjustment, deep charging and discharging of the energy storage is effectively avoided, and its SOC is maintained within the desired intermediate range.

[0077] Step S240 adjusts the bandwidth of the basic prediction interval in combination with the grid examination allowed power deviation data to obtain a dynamic power reporting interval.

[0078] The goal of step S240 is to internalize the grid examination rules as constraint conditions for interval generation, achieving economic management of operational risks. First, the accumulated value of the deviation between the station grid point power and the reported plan value within the current examination period (such as an electricity settlement period) is obtained from the data center, and compared with the total allowed deviation amount specified by the grid examination standard for that period. The ratio of the two is calculated to obtain a real-time examination risk coefficient. The higher the coefficient, the smaller the allowed deviation space in the remaining period, and the greater the examination risk. Based on this risk coefficient, a bandwidth adjustment weight negatively related to it is calculated. Subsequently, the bandwidth of the basic prediction interval adjusted in the previous steps is dynamically scaled according to this weight: when the risk coefficient is high, the bandwidth is reduced to strictly control subsequent power fluctuations and avoid penalties; when the risk coefficient is low, the bandwidth can be appropriately relaxed to reduce the frequency of energy storage intervention. The scaling process is usually carried out in proportion to the interval center value, and a minimum bandwidth protection threshold is set to prevent the interval from being too narrow and causing control failure. Finally, the interval obtained after this step adjustment is the dynamic power reporting interval that takes into account wind and light uncertainty, real-time grid demand, energy storage state, and examination risk management.

[0079] Through step-by-step optimization and collaborative adjustment, the generated dynamic power reporting interval is no longer an isolated prediction result, but a comprehensive decision product that fully internalizes wind and light uncertainty, real-time grid dispatching and frequency support demand, energy storage system's own energy state constraints, and grid assessment risk management. This method enables the reporting interval to adaptively track changes in various external conditions and internal states, thereby ensuring that the station meets the grid connection assessment requirements, effectively reducing the assessment risk, while reserving reasonable adjustment space for the operation of the energy storage system, guiding the SOC to maintain a healthy level, reducing the action frequency and depth, and ultimately achieving an effective balance between safe and stable operation of the station, maximum economic benefit, and equipment life extension.

[0080] Further, the step S210 of determining the basic prediction interval bounded by the confidence quantile points based on the wind and light power probability prediction data comprises:

[0081] Step S211, obtain the probability distribution data of wind and light power under a predetermined confidence level.

[0082] In the specific implementation process, the station energy management system (EMS) calls or receives the probability prediction results from the wind and light power prediction server at a preset period (such as every 15 minutes). The prediction data is usually generated based on numerical weather prediction (NWP), historical power data, and advanced probability prediction algorithms (such as quantile regression, Bootstrap method, or deep learning method), and its performance can be in the form of power probability density function (PDF), cumulative distribution function (CDF) at a specific time point or period in the future (such as the next 15-minute period), or in the form of a large number of equally probable scenario sets. Obtain the joint output probability distribution data described at a certain time resolution (such as 5 minutes or 15 minutes) covering the entire dispatching period, which fully considers the uncertainty of input variables such as wind speed, irradiance, temperature, and the uncertainty of the prediction model itself, and provides a data basis for subsequent determination of the statistically significant safe operation range.

[0083] Step S212, based on the probability distribution data, obtain the quantile value corresponding to the preset high confidence level as the upper boundary reference value of the interval.

[0084] According to the pre-configured high confidence level parameter (for example, β high =95%), extract the corresponding upper quantile value from the probability distribution data obtained in step S211. If the data is provided in the form of cumulative distribution function, the corresponding power point is obtained by querying the CDF value equal to β high ; if provided in the form of a large number of scenario sets, arrange the predicted output values of all scenarios in the period in ascending order, and take the corresponding β highThe percentile of the scenario power value. This quantile value is set as the upper bound reference value (P upper_base ) of the interval. high The physical meaning of this is that, from a prediction perspective, there is a (1-β low )*100% probability (in this case, a 5% probability) that the actual wind-solar combined output will exceed this value. Selecting a higher confidence level (e.g., above 90%) aims to construct a sufficiently safe upper limit to cover the actual wind-solar output with a high probability, leaving room for subsequent possible upper bound reduction, while the initial value itself also reflects a conservative estimate of extreme high output.

[0085] Step S213 extracts the quantile value corresponding to the preset low confidence level from the probability distribution data as the lower bound reference value of the interval.

[0086] According to the pre-configured low confidence level parameter (e.g., β low = 5%), the corresponding lower quantile value is extracted from the same probability distribution data. The acquisition method is similar to S212, querying the power point corresponding to the CDF value equal to β low , or taking the value corresponding to β low percentile from the sorted scenario output values. This quantile value is set as the lower bound reference value (P lower_base ) of the interval. The physical meaning of this is that there is a β low *100% probability (in this case, a 5% probability) that the actual wind-solar combined output will be below this value. Selecting a lower confidence level aims to construct a sufficiently safe lower limit to cover the actual wind-solar output with a high probability, leaving room for subsequent possible lower bound reduction (e.g., to create opportunities for energy storage charging), while the initial value itself also reflects a conservative estimate of extreme low output.

[0087] Step S214 obtains the initial bandwidth of the basic prediction interval from the difference between the upper bound reference value and the lower bound reference value of the interval.

[0088] By quantifying the uncertainty range of wind-solar output, the above initial bandwidth is the basis for all subsequent adjustment operations. Calculate the arithmetic difference between the upper bound reference value (P upper_base ) and the lower bound reference value (P lower_base ) of the interval, i.e., ΔP initial = P upper_base -P lower_base . This difference ΔP initial is the initial bandwidth of the basic prediction interval. This bandwidth value directly reflects the selected confidence level ((β high -β low*100%, 90% in this example) under which the wind and solar power output is predicted to fluctuate. A larger initial bandwidth indicates that the prediction uncertainty is large and the wind and solar power output can fluctuate dramatically; conversely, a smaller initial bandwidth indicates that the prediction confidence is high and the wind and solar power output is expected to be stable. The initial bandwidth obtained based on the statistical characteristics of the pure probability prediction provides an original and objective reference benchmark for subsequent interval scaling that comprehensively considers the grid instruction, frequency demand, energy storage state, and actual operation factors such as risk assessment.

[0089] By means of the statistical theory based on the probability prediction, an initial basic prediction interval is constructed. The interval is not a single point prediction or a fixed-width interval, but is strictly defined by statistical quantiles under a specific confidence level. The upper and lower boundary reference values and the initial bandwidth of the interval have clear probability significance, and the inherent uncertainty of the wind and solar power output is quantified. This provides an objective, reliable, and interpretable initial decision benchmark for the entire dynamic power reporting interval generation process, ensuring that all subsequent optimization adjustment operations are based on a solid statistical prediction, thereby enhancing the scientificity and adaptability of the entire power reporting strategy and providing core data support for the station to cope with wind and solar fluctuations.

[0090] Further, the adjustment of the upper boundary value of the basic prediction interval according to the grid dispatch instruction information and the grid frequency monitoring information in the grid real-time operation state data in step S220 includes:

[0091] In step S221, the active power control instruction in the grid dispatch instruction information is analyzed. When the active power control instruction is a power reduction operation instruction, a first adjustment amount for reducing the upper boundary of the basic prediction interval is calculated according to the power reduction ratio required by the active power control instruction.

[0092] In response to the direct control instruction of the grid dispatch mechanism, it is ensured that the power interval reported by the station meets the global dispatching demand of the grid. In specific implementation, the station monitoring system receives and analyzes the active power control instruction message from the dispatching master station in real time through the remote device or data communication gateway. When the analyzed instruction type is a power reduction operation instruction (usually issued due to overload of the power transmission line, blockage of the local grid, or system balance requirement), the key parameters in the instruction need to be further extracted. These parameters can be given in the form of absolute power value (unit: megawatt) or proportion (unit: percentage) relative to the current output / capacity. According to the parameter, the power absolute value that needs to be reduced, i.e., the first adjustment amount (ΔP adjust1 ), is calculated. For example, if the instruction requires a power reduction to 70% of the rated capacity and the rated capacity of the station is 100 MW, then ΔP adjust1 = 100 MW * (1-70%) = 30 MW. This adjustment amount directly reflects the mandatory constraint requirement of the grid on the upper limit of the station output power.

[0093] Step S222, monitoring the real-time frequency value in the grid frequency monitoring information, when detecting that the grid frequency exceeds the preset frequency threshold, calculating a second adjustment amount for reducing the upper boundary of the basic prediction interval based on the difference between the real-time frequency value and the preset frequency threshold.

[0094] Make the field station can actively respond to the frequency anomaly of the power grid, the body field station participates in the auxiliary service ability of the power grid primary frequency modulation, supports the stable operation of the power grid. Through the frequency measurement device (such as PMU or smart meter) connected to the field station and the grid point, the real-time frequency value (f real ) of the power grid is continuously monitored, and the data is collected at a high refresh rate (such as 50 times per second). A frequency action threshold (f set ) is preset, which is usually set according to the operation rules of the power grid, for example, the upper limit is set to 50.2Hz. When f real >f set , it indicates that there is power surplus in the power grid, the frequency is rising, and the field station needs to reduce the output to help restore system balance. At this time, the positive deviation (Δf = f real -f set ) of the real-time frequency and the set threshold is calculated. Based on this frequency deviation value, according to the preset frequency-power adjustment coefficient (K f , the unit may be MW / Hz), the required second adjustment amount (ΔP adjust2 = K f *Δf) is calculated. The adjustment amount is a dynamic value that increases with the increase of the frequency deviation, which ensures the rapid and proportional response of the field station to the change of the power grid frequency.

[0095] Step S223, based on the first adjustment amount and the second adjustment amount, the upper boundary value of the basic prediction interval is reduced.

[0096] The first adjustment amount (ΔP adjust1 , derived from the dispatching instruction) obtained in step S221 and the second adjustment amount (ΔP adjust2 , derived from the frequency overrun) obtained in step S222 are superimposed to obtain the total down adjustment amount (ΔP total_adjust = ΔP adjust1 + ΔP adjust2 ) for the upper boundary adjustment. Then, the total down adjustment amount is applied to the basic prediction interval upper boundary reference value (P upper_base ) determined by step S210, that is, the operation of P upper_new =P upper_base -ΔP total_adjust is performed. This reduction processing ensures that the new interval upper boundary (P upper_new)At the same time, the direct power control command of the grid dispatching institution and the primary frequency modulation demand for coping with abnormal grid frequency are met, so that the upper limit value of the finally generated power reporting interval can not only avoid the risk of exceeding the limit due to wind and light prediction uncertainty, but also reflect the support role of the wind and light storage station as a controllable unit for real-time operation state.

[0097] Through the continuous processing of the above steps, the adjustment of the upper boundary of the basic prediction interval under dynamic and multiple constraints is realized, so that the wind and light storage station is no longer passively following the wind and light prediction output, but can actively internalize the explicit dispatching instruction (power reduction requirement) and implicit stability demand (frequency overrun) of the grid as part of its power reporting strategy. This adjustment mechanism significantly enhances the coordination and synchronization between the station operation and the global operation state of the grid, ensures that the upper limit of the reported power interval is always within the range allowed by the safe operation of the grid, thereby improving the friendliness of the station to the grid and its ability to participate in grid frequency modulation and standby auxiliary services as a controllable resource, providing effective support for the safe and stable operation of the grid.

[0098] Further, the adjustment of the bandwidth of the basic prediction interval based on the state of charge information of the energy storage system in the station's own operation state data in step S230 includes:

[0099] In step S231, the current state of charge value of the energy storage system is obtained in real time.

[0100] The station energy management system continuously obtains the voltage, current, temperature and other original parameters of each battery unit in the energy storage system through real-time data interaction with the battery management system through a high-speed communication network at a specific sampling period. The algorithm built-in BMS based on ampere-hour integration method or combined with open circuit voltage method, Kalman filter and other advanced algorithms, fuses and calculates these original data, and finally obtains the state of charge value which can accurately reflect the percentage of the current available energy of the energy storage system to the total available capacity. The SOC value is updated at a high refresh rate (such as once per second or once every few seconds) and uploaded to the station-level control system, providing the most critical data input for judging the real-time energy level and regulation capacity of the energy storage.

[0101] In step S232, when the current state of charge value is lower than the first preset threshold, the lower boundary downshift amount is calculated according to the capacity shortage, and the lower boundary value of the basic prediction interval is accordingly downshifted.

[0102] When the energy level of the energy storage system is monitored to be low, its core goal is to create charging opportunities for the energy storage system by adjusting the operating interval, preventing it from entering a protection state or damaging its life due to excessive discharge. The current SOC value obtained in step S231 is compared with the first safety threshold (for example, SOC low= 20%) is compared. When the current SOC value is determined to be lower than this threshold, it indicates that the energy available for the energy storage system is insufficient and needs to be replenished urgently. Then, the lower bound adjustment amount is calculated according to the capacity shortage, which is usually in a positive proportional relationship with the difference between (SOC low - current SOC value) and can be determined by a predefined function or lookup table in combination with the rated power of the energy storage system and the remaining adjustment time. After the lower bound adjustment amount (AP downward ) is calculated, the lower bound value of the basic prediction interval is adjusted, i.e., P lower_new = P lower_original - AP downward . The lower bound of the adjusted interval means that the power plant is allowed to operate at a lower power value when the wind and solar output is low, thereby increasing the power difference between the grid and the energy storage system and effectively and quickly replenishing the energy of the energy storage system.

[0103] In step S234, when the current state of charge value is higher than a second preset threshold, the upper bound adjustment amount is calculated according to the capacity margin, and the upper bound value of the basic prediction interval is adjusted accordingly. The first preset threshold is lower than the second preset threshold.

[0104] When the energy level of the energy storage system is monitored to be high, the operation interval is adjusted to create a discharging opportunity for the energy storage system, avoid triggering protection due to overcharging, and improve the ability to respond to the grid demand for discharging. The current SOC value is compared with a second preset safety threshold (e.g., SOC high = 80%). When the current SOC value is determined to be higher than this threshold, it indicates that the energy stored in the energy storage system is excessive and the available charging space is limited. Then, the upper bound adjustment amount is calculated according to the capacity margin, which is usually in a positive proportional relationship with the difference between (current SOC value - SOC high ) and can be determined by a predefined function or lookup table in combination with the rated power of the energy storage system and the subsequent wind and solar prediction output. After the upper bound adjustment amount (AP upward ) is calculated, the upper bound value of the basic prediction interval is adjusted, i.e., P upper_new = P upper_original + AP upward . The upper bound of the adjusted interval means that the power plant is allowed to operate at a higher power value when the wind and solar output is high, thereby providing more discharging space for the energy storage system, avoiding possible curtailment of wind and solar power, creating conditions for the energy storage system to participate in grid peak shaving or provide backup auxiliary services, and helping to maintain the SOC at a healthy level.

[0105] By closed-loop adjustment, the real-time energy state (SOC) of the energy storage system itself is intelligently integrated into the generation strategy of the power reporting interval, realizing the transition from passive response to active management: by dynamically adjusting the lower boundary of the interval, it actively creates favorable conditions for charging the energy storage when the SOC is low; by dynamically adjusting the upper boundary of the interval, it actively reserves necessary space for discharging the energy storage when the SOC is high. This self-optimization mechanism based on the internal state of the energy storage system effectively avoids deep charging and discharging of the energy storage, guides the SOC to operate in the desired intermediate range, thereby ensuring that the station completes the external power control task while significantly optimizing the operating conditions of the energy storage system, reducing its cycle loss, and making substantial contributions to extending the service life of the energy storage equipment, maintaining its long-term regulation capacity, and improving the economy and reliability of the entire station operation.

[0106] Further, the combination of the grid assessment allowed power deviation data in step S240 adjusts the bandwidth of the basic prediction interval, including:

[0107] In step S241, the accumulated value of the power deviation that has occurred in the current assessment period and the period allowed deviation total in the grid assessment allowed power deviation data are obtained.

[0108] The above steps are the data preparation stage of the assessment risk quantification evaluation, aiming to accurately master the actual situation of the station power deviation in the current period and the remaining capacity of the grid assessment standard. The station energy management system (EMS) queries and obtains the absolute value cumulative value (Σ|ΔP actual |) of the deviation between the actual value and the reported planned value of the station grid point power from the start time of the current assessment period to the present time from the historical database. This data is usually recorded and stored in real time by the station internal power quality monitoring device or the remote terminal. At the same time, the total amount of power deviation allowed in the current assessment period (ΔP allowable ) is obtained from the grid company's issued grid management specification or dispatching instructions. This total amount may be a fixed value (such as 10 MWh) or a dynamic value specified in proportion to the rated capacity of the station. Obtaining these two types of data provides accurate input for subsequent calculation of the real-time risk level.

[0109] In step S242, the ratio of the power deviation cumulative value and the period allowed deviation total is calculated to obtain the real-time assessment risk coefficient.

[0110] By comparing the deviation that has occurred with the allowed deviation total, the assessment risk degree at the current time is quantified. The ratio of the cumulative value of the power deviation that has occurred (Σ|ΔP actual |) and the period allowed deviation total (ΔP allowable ), i.e., the real-time assessment risk coefficient (K risk =Σ|ΔP actual | / ΔP allowableThis coefficient is a dimensionless number, and its value ranges from 0 upwards. risk The closer the value is to 1, the more deviation allowance has been consumed, the smaller the allowable deviation space for the remaining time in this cycle, and the higher the risk that the station will trigger assessment penalties due to subsequent power fluctuations; if K risk A value much less than 1 indicates a low risk and sufficient margin of error.

[0111] Step S243: Calculate the bandwidth adjustment weight based on the real-time assessment risk coefficient, wherein the bandwidth adjustment weight is negatively correlated with the real-time assessment risk coefficient.

[0112] The risk coefficient is transformed into an operational weight that can be used to directly control the interval bandwidth, based on the calculated real-time assessment risk coefficient (K). risk The bandwidth adjustment weight (W) is calculated by mapping the corresponding bandwidth adjustment weights through a predefined, monotonically decreasing function or lookup table. adjust This weight is negatively correlated with the risk coefficient, that is: when K risk When W is high (close to or above 1), adjust It will tend towards 0 or a small positive number, meaning that a significant reduction in bandwidth is needed to strictly control subsequent fluctuations; while when K risk At lower levels, W adjust This value tends to 1 or a large number, meaning that bandwidth can be maintained or appropriately relaxed. This negative correlation is the core of risk control strategies, ensuring that the intensity of control matches the actual risk.

[0113] Step S244: Dynamically scale the bandwidth of the basic prediction interval according to the bandwidth adjustment weight. The scaling direction of the bandwidth is opposite to the change direction of the real-time assessment risk coefficient.

[0114] To achieve risk control, the interval bandwidth is dynamically scaled based on adjusted weights. First, the current center value (P) of the basic prediction interval is calculated. center =(P upper +P lower ) / 2) and current bandwidth (ΔP) current =P upper -P lower Then, the bandwidth is used to adjust the weights (W). adjust The current bandwidth is scaled to obtain the new bandwidth (ΔP). new =W adjust *ΔP current The scaling direction is opposite to the change in the risk coefficient: when the risk increases (K... risk ↑) leads to a decrease in weight (W) adjust ↓), which in turn leads to bandwidth contraction (ΔP) new ↓); When the risk decreases (K) risk ↓) leads to an increase in weight (W)adjust ↑), thus widening the bandwidth (ΔP) new ↑). Next, adhering to the principle of keeping the center value unchanged, the upper and lower boundaries are recalculated based on the new bandwidth: P upper_new =P center +ΔP new / 2,P lower_new =P center -ΔP new / 2. To prevent control failure due to excessively narrow bandwidth, a minimum bandwidth protection threshold (ΔP) is also preset. min If ΔP new <ΔP min Then force ΔP new =ΔP min And recalculate the boundary; the final result is the dynamic power reporting interval that is integrated into the assessment risk management.

[0115] Through continuous calculation and adjustment, the power grid's assessment rules are dynamically and precisely internalized into the generation process of the power reporting interval, achieving proactive risk management: when the deviation is small at the beginning of the assessment cycle, the interval is appropriately widened to reduce ineffective actions of energy storage; as time goes on and the deviation accumulates, the risk coefficient increases, so the interval bandwidth is gradually narrowed, and increasingly strict control is implemented on the power of the power station, thereby intelligently controlling the total deviation of the entire cycle within the allowable range. This dynamic bandwidth adjustment mechanism based on real-time risk assessment significantly reduces the risk of power stations being penalized for exceeding power fluctuation limits, ensuring the economic efficiency of power station operation, while also optimizing the operation strategy of the energy storage system while meeting the assessment requirements.

[0116] Furthermore, step S244, which involves dynamically scaling the bandwidth of the basic prediction interval based on the bandwidth adjustment weights, includes:

[0117] Step S2441: Calculate the scaling ratio based on the bandwidth adjustment weight and the current bandwidth value of the basic prediction interval, wherein the scaling ratio is positively correlated with the bandwidth adjustment weight.

[0118] This is achieved by translating abstract bandwidth adjustment weights into concrete scaling operation instructions. The bandwidth adjustment weights (W) calculated in previous steps are read. adjust This weight is a coefficient between 0 and a positive number, and its magnitude directly reflects the strength of bandwidth adjustment required based on real-time risk assessment. Based on this weight value, a predefined linear or non-linear mapping relationship is used (e.g., directly mapping W...). adjust As a scaling factor S, i.e., S = W adjust ), and combined with the current bandwidth value (ΔP) of the basic prediction interval. current This calculates the precise scaling factor used for the actual scaling operation. This scaling factor (S) is correlated with the bandwidth adjustment weight (W).adjust ) is positively correlated, meaning that when the risk of assessment is low and the weight value is large, the scaling ratio is also large, allowing the bandwidth to expand; when the risk of assessment is high and the weight value is small, the scaling ratio is also small, resulting in bandwidth contraction. This step provides a quantitative basis for the subsequent boundary value adjustment.

[0119] In step S2442, the upper boundary value and the lower boundary value of the basic prediction interval are scaled proportionally according to the scaling ratio, and the interval center position is kept unchanged.

[0120] According to the calculated scaling ratio, the range of the basic prediction interval is reset, and the interval center value is kept unchanged to ensure the stability of the adjustment. First, the center power value (P center ) of the current basic prediction interval is calculated, which is determined by the arithmetic mean of the upper boundary value (P upper ) and the lower boundary value (P lower ). Then, the current bandwidth value (ΔP current ) is multiplied by the scaling ratio (S) to obtain the new target bandwidth value (ΔP new = ΔP current *S). Under the premise of keeping the center value (P center ) constant, the upper boundary value and the lower boundary value of the interval are recalculated according to the new bandwidth value. The specific calculation formulas are: P upper_new = P center + ΔP new / 2 and P lower_new = P center - ΔP new / 2. This proportional scaling method ensures that the interval symmetrically expands or contracts around the center point, thereby maintaining the statistical characteristics and center trend of the original prediction interval to the maximum extent while adjusting the bandwidth to cope with the risk of assessment.

[0121] In step S2443, a minimum bandwidth protection threshold is set, and when the scaled bandwidth value is less than the minimum bandwidth protection threshold, the bandwidth is forced to adjust to the minimum bandwidth protection threshold while keeping the interval center position unchanged.

[0122] By pre-setting a minimum bandwidth protection threshold (ΔP min ), which is a technical lower limit determined by factors such as the maximum sustained power regulation capacity of the energy storage system in the field station, the minimum fluctuation range of wind and light output, and measurement error. After calculating the new target bandwidth value (ΔP new ) in step S2442, it is compared with the minimum bandwidth protection threshold (ΔP min ). If ΔP new < ΔP minIf so, it indicates that the calculated new bandwidth is too narrow, and there is a risk of control failure. At this time, the calculated value is ignored, and the bandwidth is forcibly set to the minimum protection threshold, that is, let ΔP new = ΔP min . Similarly, when this operation is enforced, the interval center value (P center ) remains unchanged, and the boundary values are updated according to P upper_new = P center + ΔP min / 2 and P lower_new = P center - ΔP min / 2, so as to ensure that the interval has the most basic power fluctuation accommodation capacity and controllability in any case.

[0123] Step S2444, update the upper boundary value and the lower boundary value of the basic prediction interval.

[0124] The new upper boundary value (P upper_new ) and the new lower boundary value (P lower_new ) finally determined after scaling calculation and minimum bandwidth protection check are written into the data structure of the dynamic power reporting interval, replacing the old boundary values. This update operation marks the completion of the bandwidth dynamic scaling adjustment process based on real-time risk assessment. The updated boundary values will be directly used as the basis for the next stage of flexible control strategy to control the charging and discharging behavior of the energy storage system, so as to ensure that the power at the grid connection point of the station runs within a new interval that is more consistent with the current risk assessment situation.

[0125] Through the bandwidth dynamic scaling mechanism described in steps S2441 to S2444, intelligent and adaptive risk assessment management is achieved, which can accurately and smoothly adjust the width of the power reporting interval according to the real-time risk level caused by the deviation accumulation in the period: when the risk is low, the control is relaxed to reduce the energy storage action and save its cycle life; when the risk is high, the control is tightened to strictly restrict the power fluctuation and avoid the assessment penalty. The minimum bandwidth protection mechanism ensures the feasibility and reliability of the control. Ultimately, the mechanism enables the station to dynamically optimize its operation strategy throughout the assessment period, achieving the best balance between economy (avoiding assessment penalties) and device life (reducing energy storage consumption).

[0126] Further, after adjusting the bandwidth of the basic prediction interval in step S240 by combining the power deviation data allowed by the grid assessment, it further includes:

[0127] Step S251, construct a conflict decision objective function, and the input parameters of the conflict decision objective function include the upper boundary adjustment amount required by the grid dispatching instruction, the upper boundary adjustment amount required by the grid frequency overrun, the bandwidth adjustment amount required by the state of charge of the energy storage system, and the bandwidth adjustment amount required by the assessment risk coefficient.

[0128] By systematically integrating the potentially conflicting adjustment demands generated by the aforementioned independent adjustment steps, and incorporating them into a unified mathematical framework for coordinated optimization, the adjustment quantities from different sources are first aggregated: one is the upper boundary adjustment quantity (ΔP) required by the power grid dispatching command. dispatch The first is the amount directly derived from the mandatory commands of the dispatching agency; the second is the adjustment amount (ΔP) at the upper boundary of the grid frequency over-limit requirement. frequency The third factor is the bandwidth adjustment required by the state of charge (SOC) of the energy storage system, which may include the amount of lowering the lower boundary (ΔP) for charging. soc_lower Or, the amount of the upper boundary (ΔP) is increased due to discharge. soc_upper This reflects the maintenance requirements for the internal equipment status; fourthly, it is the bandwidth adjustment amount (ΔP) required by the assessment risk coefficient. risk This quantity represents the risk control needs based on economic considerations. These adjustments may have different directions and objectives. Based on these input parameters, a conflict decision objective function is constructed. The design goal of this function is usually to minimize the total adjustment cost or maximize overall satisfaction. Its mathematical expression includes sub-items representing the costs of different adjustment needs, such as the cost of violating scheduling instructions, the cost of insufficient frequency support, the cost of SOC deviating from the ideal range, and the cost of exceeding assessment risk limits. A coordination weight coefficient to be optimized is introduced for each sub-item.

[0129] Step S252: Calculate the optimal solution of the conflict decision objective function based on the multi-objective optimization algorithm, and obtain the coordination weight coefficients of each adjustment quantity.

[0130] By solving the aforementioned conflict decision objective function, a set of optimal coordination weight coefficients is found to achieve the best balance among multiple conflicting objectives. A multi-objective optimization algorithm (e.g., weighted sum method, ε-constraint method, or intelligent optimization algorithm such as NSGA-II) is employed to solve this problem. The solution process needs to satisfy a series of constraints, including energy storage power limits, SOC safe operating range, grid assessment requirements, and mandatory constraints on dispatch commands. Through iterative calculations, the space of decision variables is explored, ultimately outputting a set of Pareto optimal solutions or an optimal solution determined according to preset preferences. This solution contains the coordination weight coefficient (λ) corresponding to each adjustment quantum term in the objective function. dispatch ,λ frequency ,λ soc ,λ risk These coefficients quantify the optimal allocation of the relative importance of each adjustment requirement under the current operating conditions.

[0131] Step S253: Weight and merge each adjustment amount according to the coordination weight coefficient to generate the final upper boundary adjustment instruction and the final bandwidth adjustment instruction.

[0132] Using the optimal coordination weight coefficients obtained in step S252, various adjustment quantities are weighted and fused to generate the final executable adjustment instruction. For the upper boundary adjustment, the two upper boundary adjustment quantities (ΔP) originating from the scheduling instruction and frequency exceeding the limit are respectively... dispatch ,ΔP frequency According to its corresponding coordination weight coefficient (λ) dispatch ,λ frequency A weighted synthesis is performed to generate a final upper boundary adjustment instruction (ΔP). upper_final =λ dispatch *ΔP dispatch +λ frequency *ΔP frequency The bandwidth adjustment will be based on two separate adjustments (ΔP) derived from SOC status and assessment risk. soc_bandwidth ,ΔP risk According to its corresponding coordination weight coefficient (λ) soc ,λ risk The weighted summation is performed to generate a final bandwidth adjustment command (ΔP). bandwidth_final =λ soc *ΔP soc_bandwidth +λ risk *ΔP risk This process integrates scattered and potentially conflicting adjustment requirements into two unified, coordinated control commands.

[0133] Step S254: Based on the final upper boundary adjustment instruction and the final bandwidth adjustment instruction, the basic prediction interval is updated synchronously to obtain the optimized dynamic power reporting interval.

[0134] First, apply the final upper boundary adjustment command (ΔP). upper_final The current upper boundary value (P) of the basic prediction interval upper_base On ), the adjusted upper boundary (P) is obtained. upper_adjusted =P upper_base -ΔP upper_final Next, the final bandwidth adjustment command (ΔP) is applied. bandwidth_final This instruction may require changing the width of the entire interval. While keeping the interval center value or other specific constraints (as described in step S244) constant, based on ΔP... bandwidth_final Calculate the new lower boundary value (e.g., P). lower_adjusted =P upper_adjusted -(Initial_Bandwidth+ΔP bandwidth_final Finally, the synchronized updated upper boundary value (P)upper_adjusted ) and the lower boundary value (P lower_adjusted ) together constitute an optimized dynamic power reporting interval, which is a synergistic optimization result considering all external instructions, internal states, real-time demand of power grid and economic risks.

[0135] Through the conflict decision and coordination optimization mechanism, the conflict problem between multiple heterogeneous adjustment demands is successfully solved, and suboptimal decisions caused by simple rules or priority sorting are avoided. Instead, by establishing a mathematical model and using optimization theory, a global optimal solution or satisfactory solution that meets all constraint conditions and maximizes overall benefits under a specific operating scenario is obtained. This greatly enhances the intelligence, adaptability and robustness of the power reporting strategy, ensuring that the wind-solar-storage station can respond to multiple goals such as grid dispatching, frequency stability, device state maintenance and economic risk control in a complex and changing operating environment, thereby achieving safe, stable and economic optimal operation.

[0136] Further, the optimal solution of the conflict decision objective function is calculated based on the multi-objective optimization algorithm in step S252, and the coordination weight coefficients of each adjustment quantity are obtained, including:

[0137] In step S2521, the priority order of each adjustment quantity is determined, and the priority of the adjustment quantity is from high to low as follows: the adjustment quantity required by the grid dispatching instruction, the adjustment quantity required by the grid frequency limit, the adjustment quantity required by the risk coefficient, and the adjustment quantity required by the state of charge of the energy storage system.

[0138] The above steps are the preprocessing stage before multi-objective optimization solution, and according to the objective laws and rules of power system operation, a clear decision priority is set for each type of adjustment demand to ensure that the system behavior conforms to the primary principle of grid safety when resources are limited or goals conflict. Specifically, a strict priority order is pre-set: the highest priority is the adjustment quantity required by the grid dispatching instruction, because it represents the mandatory command issued by the dispatching agency based on the global grid safety and stability operation demand, and the station must obey it first; the second priority is the adjustment quantity required by the grid frequency limit, because frequency is a key state quantity for measuring real-time balance and safety of the grid, and fast response to frequency deviation is the basic obligation of grid-connected power sources, which is related to the instantaneous safety of the entire grid; the third priority is the adjustment quantity required by the risk coefficient, which is mainly related to the economy of the station's own operation, although it is important, but its urgency and mandatory nature is inferior to the first two; the lowest priority is the adjustment quantity required by the state of charge of the energy storage system, which mainly involves the optimal operation and life management of the internal devices of the station, and is a local optimization goal at the station level, which usually submits to the above global and mandatory demands. This priority order provides an important decision direction and constraint basis for the subsequent initial allocation of weights and optimization solution.

[0139] Step S2522, initial weight coefficients are assigned to all adjustment amounts according to the priority order, and the higher the priority of the adjustment amount, the larger the initial weight coefficient assigned to the adjustment amount.

[0140] Based on the determined priority order, initial weight coefficients are assigned to each adjustment amount in the conflict decision objective function, thereby providing a high-quality initial starting point for the iterative solution of the multi-objective optimization algorithm, accelerating the convergence process and improving the solution efficiency. An initial weight coefficient is assigned to each adjustment amount, and these coefficients constitute the initial vector of the optimization algorithm iteration. The principle of assignment is that the higher the priority of the adjustment amount, the larger the value of the corresponding initial weight coefficient. For example, the initial weight coefficient (λ dispatch_init ) corresponding to the grid dispatching instruction adjustment amount is assigned a relatively large value; the initial weight coefficient (λ frequency_init ) corresponding to the grid frequency out-of-limit adjustment amount is assigned a smaller value; the initial weight coefficient (λ risk_init ) of the risk assessment adjustment amount is relatively small; and the initial weight coefficient (λ soc_init ) of the energy storage SOC adjustment amount is assigned the smallest initial value. This assignment method is not the final solution, but it converts the qualitative judgment of priority into quantitative initial values of weight, enabling the optimization algorithm to start searching from a region closer to the global optimal solution, and also reflects the relative importance difference of each adjustment demand in decision-making.

[0141] Step S2523, an optimization solution model containing all initial weight coefficients is established, and the optimization solution model is solved by an iterative optimization algorithm to obtain the coordinated weight coefficients of all adjustment amounts that satisfy all constraint conditions and make the objective function optimal.

[0142] Through mathematical optimization method, the coordinated weight coefficient set that can simultaneously satisfy all hard constraint conditions and make the conflict decision objective function optimal is found. First, a complete optimization solution model is established, and the decision variables of the model are all to-be-solved coordinated weight coefficients (λ dispatch , λ frequency , λ risk , λ soc ), the objective function is the conflict decision objective function (for example, minimizing the total adjustment cost) constructed in step S251, and the constraint conditions include the non-negativity of weight coefficients, the normalization constraint (such as the sum being 1), and the inequality constraint derived from the priority order in step S2521 (for example, λ dispatch > = λ frequency > = λ risk > = λ soc) and mathematical expressions of various operation constraints. Subsequently, a suitable iterative optimization algorithm (such as gradient descent method, interior point method or evolutionary algorithm suitable for multi-objective problems) is used to solve the model. The algorithm iteratively searches in the solution space, constantly updating the weight coefficient vector, until a set of coordination weight coefficients is found that satisfies all constraint conditions and optimizes the objective function value (or reaches the preset convergence accuracy). These final solutions of the weight coefficients achieve a fine quantitative trade-off and global optimization of multiple conflicting objectives while strictly respecting the established priorities and operation constraints.

[0143] Through the multi-objective optimization decision mechanism based on priority guidance, the conflict problem between multiple adjustment objectives in the wind-solar-storage power reporting process is effectively solved. Combining qualitative priority judgment based on operation experience with quantitative weight calculation based on mathematical optimization, not only the highest priority objectives such as grid safety and dispatching instructions are strictly satisfied, but also the economic objectives and equipment optimization objectives are intelligently coordinated within the allowed range; the limitations of simple rules or fixed weight strategies are overcome, and the set of coordination weight coefficients generated is the global optimal or satisfactory solution under the current operating condition, so that the final generated dynamic power reporting interval is not only scientific and reliable, but also realizes the optimal balance between multiple operation objectives, significantly improving the intelligent level and overall efficiency of the station operation.

[0144] Further, in step S300, based on the dynamic power reporting interval and the real-time operation data of the wind-solar-storage station, a flexible control strategy is used to control the charging and discharging behavior of the energy storage system in the station, including:

[0145] In step S310, the actual value of the grid-connected point power of the wind-solar-storage station is collected in real time, and the instantaneous deviation amount from the upper and lower boundary values of the dynamic power reporting interval is calculated.

[0146] In step S310, the actual value of the grid-connected point power of the wind-solar-storage station is collected in real time, and the instantaneous deviation amount from the upper and lower boundary values of the dynamic power reporting interval is calculated. This step is implemented by deploying a multi-channel power measurement system at the grid-connected point, which usually includes a combination of high-precision Rogowski coil mutual inductors of Class A accuracy and 0.2S-level power transducers, or a merging unit (MU) that meets the IEC61850-9-2LE standard, to continuously collect the instantaneous values of three-phase active power at the grid-connected point at a sampling frequency of 10Hz to 50Hz; the collected power data is transmitted to the wind-solar-storage collaborative controller through the station control layer network, which uses IEEE C37.118.2 protocol or IEC61850 MMS protocol to interact with the upper energy management system and receive dynamic power reporting interval parameters from the previous calculation module in real time, including the upper boundary value P_upper and the lower boundary value P_lower updated at a second level.

[0147] In the deviation amount calculation link, the controller calculates the deviation amount ΔP actual Multiple logical comparisons with dynamic interval boundaries: when P actual >P upper , the positive deviation amount ΔP upper =P actual -P upper is calculated, which represents the degree to which the current power exceeds the upper limit value; when P actual <P lower , the negative deviation amount ΔP lower =P lower -P actual is calculated, which represents the deficiency of the current power below the lower limit value; when P actual is within the interval [P lower , P upper ], both types of deviation amounts are set to zero. To eliminate the effects of measurement noise and transient disturbances, a first-order low-pass digital filter is used to process the original power data before deviation calculation, with a cutoff frequency of 0.5-2 Hz, and a sliding time window algorithm is used for trend verification, with a window length of 5-30 seconds.

[0148] The deviation amount data packet calculated contains not only the deviation value, but also metadata information such as time stamp, data quality flag, and boundary overrun duration. These data are provided to the subsequent power regulation instruction generation module through shared memory or real-time database, providing accurate input reference for flexible control strategy. When communication interruption or data anomaly is detected, the system can automatically switch to the valid interval parameters of the previous period or enable the backup threshold to ensure the continuity of system control.

[0149] Step S320, generate the power regulation preliminary instruction of the energy storage system according to the size and direction of the instantaneous deviation amount.

[0150] Generate the power regulation preliminary instruction of the energy storage system according to the size and direction of the obtained instantaneous deviation amount, which is executed by the power instruction calculation module in the wind-solar-storage collaborative controller. This module uses a multi-parameter adaptive proportional-derivative (P-D) control algorithm to generate the preliminary instruction. When a positive deviation ΔP upper is detected, a negative power instruction P cmd_pre =-K p *ΔP upper -K d *d(ΔP upper ) / dt is generated, with a negative instruction value indicating that the energy storage system is required to discharge to reduce the grid point power; when a negative deviation ΔP lower is detected, a positive power instruction P cmd_pre =K p *ΔP lower +Kd d(ΔP lower ) / dt, positive command value indicates that the energy storage system is required to charge to increase the grid point power. Wherein, the proportional coefficient K p According to the size of the deviation segment setting: small deviation interval (0-5% rated power) using smaller coefficient to avoid frequent adjustment, large deviation interval (>15% rated power) using larger coefficient to ensure fast response; the differential coefficient K d According to the power change rate adaptive adjustment, enhance the system damping characteristics when the power fluctuates violently.

[0151] The dead zone control logic is also introduced in the command generation process: when the deviation is less than the preset dead zone threshold (usually set to 0.5%-1% of the rated power), no adjustment command is generated to avoid frequent start and stop of the energy storage system. At the same time, the command change rate limit is set, the ramp rate of charge and discharge command is limited within the maximum allowable power change rate range of the energy storage system, and the typical value is 30%-50% of the rated power per minute. The preliminary command also needs to be processed by the power limit value checking module to ensure that its absolute value does not exceed the current maximum output capability of the energy storage converter, which is dynamically calculated according to the real-time parameters such as DC bus voltage and battery temperature.

[0152] By monitoring the deviation trend of wind and light power prediction value and actual value, the compensation command is generated in advance at the initial stage of deviation. All generated preliminary commands are attached with time stamp, command priority flag and expected execution time window information, and are transmitted to the command correction module through real-time communication bus. The module also records the key parameters in the command generation process, including the reference deviation, the value of the adjustment coefficient, the amplitude limiting state, etc., which are used for system operation state analysis and control parameter optimization.

[0153] Step S330, obtaining the real-time state of charge value of the energy storage system, and adaptively correcting the power regulation preliminary command based on the real-time state of charge value to obtain the power regulation correction command.

[0154] By obtaining the current state of charge value of the energy storage system, and based on the value, the power regulation preliminary command is adaptively corrected in multiple dimensions. The correction process is completed by the energy storage management system (BMS) and the station collaborative controller, which obtains the state of charge (SOC) data calculated and verified by the BMS through CAN bus or industrial Ethernet with a period of not less than 100ms, which is usually calculated by integrating the ampere-hour method and open circuit voltage method, and is smoothed by Kalman filtering algorithm.

[0155] The correction strategy adopts a segmented adaptive adjustment mechanism: when the SOC is in a normal operation interval (e.g., 20%-80%), the preliminary power instruction is executed in principle; when the SOC enters a high alert interval (e.g., 80%-90% or 10%-20%), an instruction amplitude limiting algorithm is started, and a SOC feedback coefficient K soc is used to scale the instruction, where K soc = 1-|(SOC-SOC center )| / (SOC limit -SOC center ), SOC center is an ideal center value (usually 50%), and SOC limit is a boundary value of the current interval; when the SOC enters a dangerous interval (e.g., >90% or <10%), instruction direction intervention logic is triggered, if the preliminary instruction direction conflicts with the SOC protection direction (e.g., a charging instruction is received when the SOC is >90%), the instruction is forced to zero and a warning is triggered, if the directions are consistent, the execution is allowed but with additional power gradient limitation.

[0156] The correction process also considers the real-time operation characteristics of the energy storage device: the power limit value is dynamically adjusted according to the battery temperature, and the allowed power is reduced according to a pre-set derating curve when the temperature is lower than 5℃ or higher than 45℃; the recent charging and discharging efficiency map (usually stored in the form of a two-dimensional table) is referred to, the instruction is preferentially executed in the high efficiency area, and the instruction amplitude is appropriately reduced in the low efficiency area; a life decay model is integrated, and when the cumulative throughput approaches the design value, the instruction amplitude is automatically reduced to reduce the number of high-rate charging and discharging. The final corrected instruction is attached with an effective period label, a SOC state identifier and a correction reason code, and after safety verification, it is transmitted to the execution layer.

[0157] In step S340, the power output of the energy storage converter is controlled according to the corrected power regulation instruction, so that the actual power value of the grid connection point tends to the center value of the dynamic power reporting interval.

[0158] The power output of the energy storage converter is controlled according to the corrected power regulation instruction, which is realized through the energy storage converter controller (PCS). After receiving the corrected power regulation instruction from the upper layer controller, the PCS uses a double-loop control structure for accurate power execution: the outer loop is a power control loop, which uses a PI regulator with feedforward compensation, taking the deviation between the instruction power value and the actual output power value of the converter as input, and outputs a current reference value; the inner loop is a current control loop, which uses a proportional-resonant (PR) controller or a feedforward decoupling control algorithm to quickly track the current in the dq rotating coordinate system, generating a PWM modulation wave. The converter uses space vector pulse width modulation (SVPWM) technology to control the output voltage and phase of the AC side by adjusting the switching state and duty cycle of the power devices, thereby accurately controlling the active power size and direction injected into the grid.

[0159] The control instructions are transmitted at a frequency of no less than 10 kHz through a high-speed communication network (such as EtherCAT or CANopen), and parameters such as the voltage on the DC side of the converter, the current on the AC side, and the phase of the grid voltage are collected in real time. The grid synchronization is achieved by using a phase-locked loop (PLL) technology. The power control loop is provided with an anti-saturation integrator and output amplitude limiting protection to ensure that the output current does not exceed the safety limit of the device. At the same time, the modulation ratio is automatically adjusted according to the grid voltage fluctuation, and the reactive power support capability is provided when the grid voltage drops.

[0160] To improve the response speed, power feedforward compensation can be introduced to adjust the current reference value in advance by monitoring the wind and light power mutation trend. During the execution, the key parameters of the converter (such as IGBT junction temperature and DC link voltage fluctuation) are continuously monitored, and when an abnormality is detected, the system is automatically switched to a degraded operation mode and the fault information is reported. The entire control process ensures that the actual value of the grid-connected point power approaches the center value of the dynamic power reporting interval with minimum overshoot and fast response characteristics, and the typical response time is less than 100 milliseconds, and the control accuracy is maintained within ±1% of the rated power.

[0161] In step S350, when the actual value of the grid-connected point power continuously stays within the range of the dynamic power reporting interval within a preset time threshold, the power regulation strength of the energy storage system is reduced, and the state of charge of the energy storage system is preferentially maintained stable.

[0162] When the interval state monitoring module of the wind-solar-storage station cooperative controller detects that the actual value of the grid-connected point power continuously stays within the range of the dynamic power reporting interval [P lower ,P upper ] within a continuous time T (T is a configurable parameter, usually set to 5-15 minutes), and the power fluctuation amplitude is less than ±2% of the rated power, the energy storage operation mode switching logic is automatically triggered to switch from the active power regulation mode to the state of charge maintenance mode. In this mode, the power instruction generation module adopts a hysteresis control strategy to reduce the regulation strength: the power instruction dead zone range is expanded from ±0.5% to ±3% of the rated power, reducing the charge-discharge switching caused by small fluctuations; at the same time, the proportional regulation coefficient K p is dynamically reduced to 30%-50% of the normal value, and the differential regulation term is disabled, significantly reducing the variation amplitude and frequency of the power instruction.

[0163] In addition, the synchronous starting SOC equalization management strategy is started, the optimal target SOC value (usually set to the interval of 50%-60%) is calculated by querying the historical operation data of the energy storage system, and the single-point power injection method with adaptive step is used to gradually adjust the SOC to the target interval. The adjustment process is slow charging and discharging with small power not exceeding 5% of the rated power, and if the grid-connected point power exceeds the limit during the adjustment process, the equalization process is interrupted immediately and the active regulation mode is restored. In order to improve the stability of the system, the power smoothing filter is enabled in this mode to flexibly suppress the wind and light power fluctuations instead of completely compensating them, allowing the grid-connected point power to fluctuate moderately within the interval.

[0164] The mode switching process adopts a gradual dynamic rate transition algorithm to gradually complete the control parameter adjustment within 60 seconds, avoiding power command jumps. Continuous monitoring of the operating state, when detecting a trend of continuously expanding power deviation or SOC deviation from the safe interval, automatically restores the active regulation mode in advance. All mode switching events, parameter adjustment records and SOC change trajectories are recorded in the operation log for subsequent operation strategy optimization. The intelligent mode management mechanism effectively balances the power regulation requirements and the need to extend the life of the energy storage system, while ensuring the qualified rate of the grid-connected point power and significantly reducing the number of actions of the energy storage system.

[0165] Correspondingly, a second aspect of the embodiment of the application provides an electronic device, comprising: at least one processor; and a memory connected with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the wind and light storage station power reporting tracking method based on the fast power response of the energy storage.

[0166] Correspondingly, a third aspect of the embodiment of the application provides a computer readable storage medium having computer instructions stored thereon, the instructions being executed by a processor to implement the wind and light storage station power reporting tracking method based on the fast power response of the energy storage.

[0167] The embodiment of the application aims to protect a wind and light storage station power reporting tracking method based on the fast power response of the energy storage, which has the following effects:

[0168] 1. By comprehensively considering the wind and light probability prediction, the real-time state of the power grid, the SOC of the energy storage and the historical deviation data, the power reporting interval is dynamically generated and optimized in real time, which avoids the risk of examination punishment caused by too narrow interval and prevents unnecessary frequent action of the energy storage system caused by too wide interval; the dynamic optimization mechanism fundamentally realizes the maximum reduction of the charging and discharging times and depth of the energy storage under the premise of meeting the requirements of the power grid examination, thereby prolonging the life of the energy storage, reducing the operation and maintenance cost, and improving the economy and reliability of the entire station.

[0169] 2. The traditional control strategy is difficult to coordinate the conflicting goals such as scheduling instruction power reduction, frequency out-of-limit, SOC recovery, and risk avoidance; by constructing a conflict decision objective function and using a multi-objective optimization algorithm, the adjustment requirements from different sources are quantified and weighted to generate a coordinated optimal adjustment instruction; it can intelligently judge which priority should be given to meet which demand in different scenarios, so as to make a comprehensive decision that is most beneficial to the stability of the power grid and the operation of the station itself, avoiding the performance deterioration in other aspects caused by a single strategy;

[0170] 3. Through the flexible control strategy, not only the instantaneous power deviation is adjusted, but also the real-time SOC of the energy storage is introduced as a correction factor, and the control strength after long-time stable operation is optimized to make the adjustment behavior of the energy storage more smooth and accurate, which not only ensures that the grid-connected point power is effectively maintained within the dynamic interval, but also actively avoids overcharging and overdischarging of the energy storage, and preferentially maintains the SOC of the energy storage within the healthy interval, thereby effectively protecting the energy storage device itself while completing the power support task, and improving the intelligent and refined level of control.

[0171] Those skilled in the art will understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0172] The present application is described with reference to flowcharts and / or block diagrams according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in a flow or multiple flows and / or blocks Figure 1 The functions specified in a flow or multiple flows and / or blocks

[0173] These computer program instructions can also be stored in a computer-readable storage medium that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable storage medium produce a manufactured product including instruction devices that implement the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in a flow or multiple flows and / or blocksFigure 1 the function specified in the one or more blocks.

[0174] These computer program instructions can also be loaded into computer or other programmable data processing devices, so that a series of operation steps are performed on the computer or other programmable data processing devices to generate computer-implemented processes, thus the instructions executed on the computer or other programmable data processing devices provide a process for implementing the flow Figure 1 the flow or flows and / or blocks Figure 1 the steps of the function specified in the one or more blocks.

[0175] Finally, it should be noted that: the above examples are only used to illustrate the technical solutions of the present application, but not to limit it, although the present application has been described in detail with reference to the above examples, those skilled in the art should understand: the specific embodiments of the present application can still be modified or replaced by the equivalent, without departing from the spirit and scope of the present application, any modification or equivalent replacement, which should be covered within the protection scope of the claims of the present application.

Claims

1. A wind-solar-storage power station power reporting tracking method based on energy storage fast power response, characterized in that, The method comprises the following steps: obtaining wind and light power probability prediction data, real-time power grid operation state data, power station operation state data and power grid examination allowable power deviation data in a current scheduling period; based on the wind and light power probability prediction data, real-time power grid operation state data, power station operation state data and power grid examination allowable power deviation data, generating a dynamic power reporting interval for the next scheduling period, the dynamic power reporting interval being a safe operation power range defined by an adaptive upper boundary value and a lower boundary value; based on the dynamic power reporting interval and real-time operation data of the wind and light storage power station, using a flexible control strategy to control the charging and discharging behavior of the energy storage system in the power station, the flexible control strategy being used to optimize the operating conditions of the energy storage system on the premise that the power at the power station grid connection point is maintained between the upper limit power value and the lower limit power value.

2. The wind-solar-storage station power reporting tracking method based on energy storage fast power response according to claim 1, characterized in that, The method comprises the following steps: determining a basic prediction interval defined by a confidence quantile point based on the wind and light power probability prediction data; adjusting the upper boundary value of the basic prediction interval according to the power grid scheduling instruction information and power grid frequency monitoring information in the real-time power grid operation state data; adjusting the bandwidth of the basic prediction interval based on the state of charge information of the energy storage system in the power station operation state data; adjusting the bandwidth of the basic prediction interval in combination with the power grid examination allowable power deviation data to obtain the dynamic power reporting interval.

3. The wind-solar-storage station power reporting tracking method based on energy storage fast power response according to claim 2, characterized in that, The method comprises the following steps: obtaining probability distribution data of wind and light power at a predetermined confidence level; obtaining a quantile value corresponding to a preset high confidence level from the probability distribution data as an interval upper boundary reference value; obtaining a quantile value corresponding to a preset low confidence level from the probability distribution data as an interval lower boundary reference value; obtaining an initial bandwidth of the basic prediction interval from the difference between the interval upper boundary reference value and the interval lower boundary reference value.

4. The wind-solar-storage station power reporting tracking method based on energy storage fast power response according to claim 2, characterized in that, The method comprises the following steps: analyzing the active power control instruction in the power grid scheduling instruction information, when the active power control instruction is a power reduction operation instruction, calculating a first adjustment amount for reducing the upper boundary of the basic prediction interval according to the power reduction ratio required by the active power control instruction; monitoring the real-time frequency value in the power grid frequency monitoring information, when the power grid frequency is detected to exceed a preset frequency threshold, calculating a second adjustment amount for reducing the upper boundary of the basic prediction interval based on the difference between the real-time frequency value and the preset frequency threshold; based on the first adjustment amount and the second adjustment amount, performing a reduction process on the upper boundary value of the basic prediction interval.

5. The wind-solar-storage station power reporting tracking method based on energy storage fast power response according to claim 4, characterized in that, The bandwidth of the basic prediction interval is adjusted based on the state of charge information of the energy storage system in the field station self-operation state data, including: real-time acquisition of the current state of charge value of the energy storage system; when the current state of charge value is lower than a first preset threshold, calculating a lower boundary downshift amount according to a capacity shortage, and accordingly downshifting the lower boundary value of the basic prediction interval; when the current state of charge value is higher than a second preset threshold, calculating an upper boundary upshift amount according to a capacity margin, and accordingly upshifting the upper boundary value of the basic prediction interval; wherein the first preset threshold is lower than the second preset threshold.

6. The wind-solar-storage station power reporting tracking method based on energy storage fast power response according to claim 5, characterized in that, The bandwidth of the basic prediction interval is adjusted based on the grid examination allowable power deviation data, including: acquiring the power deviation cumulative value that has occurred in the current examination period and the period allowable deviation total amount in the grid examination allowable power deviation data; calculating the ratio of the power deviation cumulative value to the period allowable deviation total amount to obtain a real-time examination risk coefficient; calculating a bandwidth adjustment weight based on the real-time examination risk coefficient, wherein the bandwidth adjustment weight is negatively correlated with the real-time examination risk coefficient; dynamically scaling the bandwidth of the basic prediction interval according to the bandwidth adjustment weight, and the scaling direction of the bandwidth is opposite to the change direction of the real-time examination risk coefficient.

7. The wind-solar-storage station power reporting tracking method based on energy storage fast power response according to claim 6, characterized in that, The dynamically scaling the bandwidth of the basic prediction interval according to the bandwidth adjustment weight, including: calculating a scaling ratio based on the bandwidth adjustment weight and the current bandwidth value of the basic prediction interval, wherein the scaling ratio is positively correlated with the bandwidth adjustment weight; performing equal-proportion scaling on the upper boundary value and the lower boundary value of the basic prediction interval according to the scaling ratio, and keeping the interval center position unchanged; setting a minimum bandwidth protection threshold, when the scaled bandwidth value is smaller than the minimum bandwidth protection threshold, forcibly adjusting the bandwidth to the minimum bandwidth protection threshold while keeping the interval center position unchanged; updating the upper boundary value and the lower boundary value of the basic prediction interval.

8. The wind-solar-storage station power reporting tracking method based on energy storage fast power response according to claim 6, characterized in that, After adjusting the bandwidth of the basic prediction interval based on the grid examination allowable power deviation data, further including: building a conflict decision objective function, the input parameters of the conflict decision objective function including an upper boundary adjustment amount required by the grid dispatching instruction, an upper boundary adjustment amount required by the grid frequency out-of-limit, a bandwidth adjustment amount required by the state of charge of the energy storage system, and a bandwidth adjustment amount required by the examination risk coefficient; calculating the optimal solution of the conflict decision objective function based on a multi-objective optimization algorithm to obtain a coordination weight coefficient of each adjustment amount; performing weighted fusion on each adjustment amount according to the coordination weight coefficient to generate a final upper boundary adjustment instruction and a final bandwidth adjustment instruction; based on the final upper boundary adjustment instruction and the final bandwidth adjustment instruction, synchronously updating the basic prediction interval to obtain an optimized dynamic power reporting interval.

9. The wind-solar-storage station power reporting tracking method based on energy storage fast power response according to claim 8, characterized in that, The optimal solution of the conflict decision objective function is calculated based on a multi-objective optimization algorithm to obtain a coordination weight coefficient of each adjustment amount, including: The priority order of each adjustment amount is determined, and the priority of the adjustment amount is from high to low: an adjustment amount required by a power grid dispatching instruction, an adjustment amount required by a power grid frequency out-of-limit, an adjustment amount required by an assessment risk coefficient, and an adjustment amount required by a state of charge of the energy storage system; An initial weight coefficient is assigned to all the adjustment amounts according to the priority order, and the initial weight coefficient assigned to the adjustment amount with a higher priority is larger; An optimization solving model containing all the initial weight coefficients is established, the optimization solving model is solved by an iterative optimization algorithm, and a coordinated weight coefficient of all the adjustment amounts that meets all the constraint conditions and optimizes a target function is obtained.

10. The wind-solar-storage power station power reporting tracking method based on energy storage fast power response according to any one of claims 1-9, characterized in that, The flexible control strategy is adopted to control the charging and discharging behavior of the energy storage system in the station based on the dynamic power reporting interval and real-time operation data of the wind-solar-storage station, including: The actual value of the grid-connected point power of the wind-solar-storage station is collected in real time, and the instantaneous deviation amount of the actual value from the upper and lower boundary values of the dynamic power reporting interval is calculated; The power regulation preliminary instruction of the energy storage system is generated according to the size and direction of the instantaneous deviation amount; The real-time state of charge value of the energy storage system is obtained, and the power regulation preliminary instruction is adaptively corrected based on the real-time state of charge value to obtain a power regulation correction instruction; The power output of the energy storage converter is controlled according to the power regulation correction instruction, so that the actual value of the grid-connected point power tends to the center value of the dynamic power reporting interval; When the actual value of the grid-connected point power continuously stays within the range of the dynamic power reporting interval within a preset time threshold, the power regulation strength of the energy storage system is reduced, and the state of charge of the energy storage system is preferentially maintained stable.

Images