Power grid wind storage accurate energy management method based on dynamic optimization control

Abstract

The invention belongs to the technical field of power systems and automation thereof, and particularly discloses and provides a power grid wind storage accurate energy management method based on dynamic optimization control, which comprises the following steps of: performing collaborative decision-making on an energy storage charge state safety interval, wind power and instantaneous deviation of a power grid dispatching instruction, judging an initial working mode, and judging a power grid dispatching instruction according to the initial working mode; and generating a wind power prediction curve and an initial charging and discharging plan based on historical data, performing safety verification by simulating a future charge state track of energy storage, and if the verification is not passed, re-judging a safety working mode and generating an optimization plan according to trend association between a track crossing point and the prediction curve. And finally, by comparing the actual wind power with the predicted wind power, strategy mismatch judgment and decision closed-loop reset are realized. According to the method, the energy storage state is simulated and verified before the plan is executed, and when verification is not passed, the security mode is re-judged based on the association of the risk point and the prediction trend, so that security control upgrading from passive protection to active prevention is realized.

Description

Technical Field

[0001] This invention belongs to the field of power system and automation technology, and relates to a method for precise energy management of power grid wind and energy storage based on dynamic optimization control. Background Technology

[0002] With the energy structure shifting towards cleaner and lower-carbon energy, the installed capacity and power generation share of renewable energy sources such as wind power in the power system continue to rise. However, the inherent volatility and randomness of wind power output pose significant challenges to the real-time power balance and stable operation of the power grid. Therefore, coordinating the grid-based wind-storage system with wind farms to construct a dispatchable and adjustable wind-storage integrated system has become an important technical approach to enhance the absorption capacity of new energy sources and the resilience of the power grid.

[0003] For example, Chinese invention patent CN111864793B discloses a grid control method based on model prediction algorithms for the power energy internet. This method applies model prediction control to photovoltaic, wind power, and grid-connected wind-storage systems respectively, and collects the operating parameters of each unit in real time, coordinating control with the goal of system power balance. This scheme optimizes the power allocation of distributed power sources within the microgrid through a hierarchical control structure, aiming to improve response speed, efficiency, and grid stability.

[0004] The existing technologies mentioned above have the following shortcomings: 1. Existing technologies mainly rely on optimization algorithms to generate control commands, and lack a mechanism for collaborative decision-making based on the instantaneous deviation between wind power and dispatch commands, as well as the interval in which the wind and energy storage are located. This leads to the problem of mismatch between the initial control direction and the actual demand, reducing the timeliness of response.

[0005] 2. Existing technologies mainly focus on power balance optimization when performing safety control, lacking simulation and verification of the future trajectory of wind and energy storage charge state. It is difficult to identify in advance the risk that the plan may cause the wind and energy storage charge state to exceed the limit. Therefore, it lacks a dynamic re-judgment safety working mode based on the local trend of the crossing point location and the wind power prediction curve after the trajectory verification fails. At the same time, it lacks the ability to determine whether the current strategy is mismatched by comparing the actual wind power with the predicted value online. If mismatched, it will trigger the re-initialization of the entire decision-making process to form a closed loop feedback, making it difficult to correct in time when the situation deviates from the prediction. Summary of the Invention

[0006] In view of this, in order to solve the problems mentioned in the background technology, a precise energy management method for wind and energy storage based on dynamic optimization control is proposed.

[0007] The objective of this invention can be achieved through the following technical solution: This invention provides a precise energy management method for wind and energy storage based on dynamic optimization control, including: S1, determining the initial working mode by combining the set of charging and discharging actions determined by the state of charge of wind and energy storage, and making collaborative decisions based on the instantaneous deviation direction and magnitude of the current active power of wind power and the grid dispatch command.

[0008] S2. Based on the initial operating mode and the current active power of wind power, a short-term wind power prediction curve is generated by combining historical operating data sequences, and an initial charging and discharging power plan is generated accordingly.

[0009] S3. Predict the trajectory of wind and energy storage state of charge within the look-ahead optimization time window by simulating the execution of the plan, and verify whether the initial charge and discharge power plan passes. If the verification passes, execute the initial charge and discharge power plan.

[0010] S4. If the verification fails, determine the predicted crossing point of the trajectory crossing the safety boundary based on the wind-storage state of charge prediction trajectory, analyze the changing trend of the short-term wind power prediction curve before and after the predicted crossing point, re-determine the safe working mode based on the trend and generate an optimized charging and discharging power plan.

[0011] S5. Based on the predicted values ​​corresponding to the actual wind power and the optimized charging and discharging power plan, determine whether the strategy is mismatched. If there is no mismatch, execute the optimized charging and discharging power plan; otherwise, return to step S1.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention achieves rapid and safe determination of the initial working mode by coupling and coordinating the instantaneous deviation between the current active power of wind power and the grid dispatch command and the safe range of the real-time wind storage charge state, thereby avoiding the conflict between control requirements and equipment safety capabilities from the source, thus significantly improving the accuracy of the initial action of the system.

[0013] (2) After generating the initial charging and discharging power plan, the present invention predicts the future trajectory of the wind and energy storage charge state by simulating the execution of the plan and verifies whether it is in the safe range throughout the process, thereby avoiding the risk of overcharging or over-discharging of wind and energy storage due to improper planning, thus ensuring the safe operation of the power grid wind and energy storage system before the plan is executed.

[0014] (3) This invention proactively identifies overcharging and over-discharging risks by simulating the wind-storage charge state trajectory and verifying its safety before the plan is executed. When the verification fails, the safe working mode is re-evaluated by analyzing the correlation between the risk points and the wind power prediction trend, thus realizing the upgrade of safety control from passive protection to active prevention.

[0015] (4) This invention continuously compares the actual wind power with the predicted value on which the optimization plan is based online, dynamically determines whether the strategy is mismatched, and re-initializes the decision-making process when mismatch occurs, thereby ensuring that the management strategy can be self-corrected in a timely manner when the operating conditions such as wind power output deviate significantly from the predicted scenario, thus maintaining the adaptability and reliability of the system in long-term operation. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram showing the connections between the steps of the method of the present invention.

[0018] Figure 2 This is a schematic diagram of the connection steps for determining the initial working mode of the present invention.

[0019] Figure 3 This is a schematic diagram showing the connection steps for determining the safe operating mode of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see Figure 1 As shown, the present invention provides a precise energy management method for wind and energy storage based on dynamic optimization control. The method includes: S1, determining the initial working mode by combining the set of charging and discharging actions determined by the state of charge of wind and energy storage, and making collaborative decisions based on the instantaneous deviation direction and magnitude of the current active power of wind power and the grid dispatch command.

[0022] Please see Figure 2 As shown, for example, the determination of the initial working mode includes: S1-1, determining the set of charging and discharging actions of the grid wind and energy storage system based on the wind and energy storage state of charge value.

[0023] Furthermore, the determination of the charging and discharging action set of the grid wind and energy storage system includes: S1-1-1, obtaining the discharge prohibition threshold and the charging prohibition threshold of the grid wind and energy storage system, taking the values ​​below the discharge prohibition threshold as the first interval, the values ​​between the discharge prohibition threshold and the charging prohibition threshold as the second interval, and the values ​​above the charging prohibition threshold as the third interval.

[0024] It should be noted that the discharge prohibition threshold and the charging prohibition threshold are key operating parameters for ensuring the safe, efficient, and long-term stable operation of the grid-connected wind and energy storage system. These two thresholds together constitute the permissible operating range of the battery state of charge in the wind and energy storage system, i.e., the safe operating interval.

[0025] The discharge prohibition threshold defines the lower limit of the battery's permissible state of charge, primarily to prevent over-discharge. The specific value of this threshold is typically set according to the technical specifications of the batteries configured in the energy storage system. Taking widely used lithium-ion batteries as an example, to avoid over-discharge damage, the discharge prohibition threshold is generally set between 10% and 20% of the battery's total capacity. The charging prohibition threshold defines the upper limit of the battery's permissible state of charge, its core function being to prevent overcharging. This threshold is also set according to the technical standards provided by the battery manufacturer. For example, for lithium-ion batteries, to effectively prevent overcharging and provide necessary buffer space for operation, the charging prohibition threshold is typically set between 90% and 95% of the total capacity.

[0026] S1-1-2. If the wind and energy storage state of charge value is in the first interval, it indicates that the grid wind and energy storage system is in a deep discharge state. The internal battery voltage may be close to or has reached the damage threshold. At this time, continuing to discharge will likely lead to over-discharge of the battery, seriously damaging the battery health, shortening its cycle life, and may trigger a protective shutdown of the system. Therefore, the system must prohibit discharge behavior and limit the current executable safe action to charging. Based on this, the wind and energy storage state of charge is pulled back to the safe area, and the action set is defined as charging.

[0027] S1-1-3. If the wind-storage state of charge value is in the second range, it indicates that the grid wind-storage system has sufficient and safe charging and discharging buffer space. Its state can accept external electrical energy for storage, release the stored electrical energy to meet the grid demand, and remain in standby when no action is needed. The action set is defined as charging, discharging and standby.

[0028] S1-1-4. If the wind-storage state of charge value is in the third interval, it indicates that the energy storage capacity of the grid wind-storage system is approaching saturation. Continuing to charge will lead to battery overcharging, which will also damage battery health, increase the risk of thermal runaway, and may cause energy waste. Therefore, the system must prohibit charging behavior, and the action set is defined as discharging.

[0029] S1-2. Calculate the deviation between the current active power of wind power and the required active power of wind power corresponding to the grid dispatch command, and generate the expected power compensation demand based on the direction and magnitude of the deviation.

[0030] Furthermore, the generation of the desired power compensation requirement includes: S1-2-1, subtracting the current wind power active power from the desired wind power active power to obtain the instantaneous deviation value.

[0031] S1-2-2: Obtain the wind power active power sequence within the instantaneous analysis time window before the current moment, and calculate its instantaneous change slope.

[0032] The instantaneous analysis time window refers to a continuous time period preceding the current moment. This is designed to ensure the window length is greater than the minimum time scale of the main control components of the power grid wind-storage system, thereby guaranteeing that the calculated trend information is valuable for the upcoming control cycle. Typically, it should be no less than 1.5 to 2 times the main control cycle of the system.

[0033] S1-2-3. Couple the instantaneous deviation value with the instantaneous change slope to generate the initial demand value.

[0034] It should be added that generating the initial demand value includes: normalizing the instantaneous deviation value and the instantaneous change slope respectively. For example, the instantaneous deviation value is divided by the rated power of the grid wind-storage system to obtain the normalized instantaneous deviation value; the instantaneous change slope is divided by a reference slope, where the reference slope is the ratio of the rated power to a characteristic time constant such as 15 seconds, thereby achieving slope normalization.

[0035] After obtaining the normalized parameters, the initial demand value is calculated using the following weighted fusion formula. , In the formula This refers to the rated power of the power grid's wind and energy storage system. and These are the normalized instantaneous deviation value and the normalized instantaneous change slope, respectively. and These are the weights of the normalized instantaneous deviation value and the normalized instantaneous change slope, respectively.

[0036] Therefore, the above weighted fusion calculation can simultaneously reflect the static compensation demand of real-time power deviation and the dynamic adjustment demand indicated by the power change trend, so that the obtained initial demand value has both the ability to match the current imbalance state and the ability to predict future change trends.

[0037] Among them, weight and The settings can be based on system control objectives and operational experience, or obtained by analyzing historical operational data. As an example, collect historical power deviation sequences, power change slope sequences, and corresponding actual compensation power data. Through correlation analysis or regression modeling, identify the degree of influence of these two factors on the final compensation power decision, and determine their weighting accordingly, while satisfying the following conditions: This enables the accurate quantification and generation of initial demand values, supporting the rational formulation of subsequent power plans.

[0038] S1-2-4. The initial demand value is subjected to limiting processing based on the real-time power capacity of the power grid wind storage system to obtain the desired power compensation demand.

[0039] The limiting process is achieved through the following steps: obtaining the real-time maximum rechargeable power and the real-time maximum dischargeable power of the power grid wind-storage system under the current wind-storage charge state.

[0040] The real-time maximum rechargeable power and real-time maximum dischargeable power are determined by the battery management system and power conversion system within the wind-storage system based on real-time conditions, together constituting the power limit for the safe operation of the energy storage unit at the current moment. Subsequently, the initial demand value is compared with the real-time maximum rechargeable power and real-time maximum dischargeable power, and a limiting is applied according to the following rule: if the initial demand value is negative and its absolute value is greater than the absolute value of the real-time maximum rechargeable power, then the desired power compensation demand is limited to the real-time maximum rechargeable power.

[0041] If the initial demand value is positive and its value is greater than the real-time maximum dischargeable power, then the expected power compensation demand is limited to the real-time maximum dischargeable power.

[0042] If the initial demand value does not exceed the corresponding real-time maximum power limit, then the initial demand value will be directly used as the expected power compensation demand.

[0043] S1-3. Match the desired power compensation requirement with the charging and discharging action set of the grid wind and energy storage system, and obtain the initial working mode of the grid wind and energy storage system according to the preset priority rules.

[0044] Based on the above scheme, the preset priority rule is used to select an alternative mode as the initial operating mode from the set of actions when the operating mode indicated by the expected power compensation demand is inconsistent with the set of charging and discharging actions currently allowed by the grid wind-storage system. Specifically, this includes determining whether the operating mode indicated by the expected power compensation demand belongs to the set of charging and discharging actions.

[0045] If it belongs to the category, then the working mode is determined as the initial working mode; if it does not belong to the category, then a substitute mode is selected from the set of charging and discharging actions as the initial working mode according to the preset priority order, wherein the priority order is: standby mode is selected first; if standby mode is not available, then the charging and discharging mode that has the mildest impact on the state of charge of the wind and energy storage system of the power grid is selected.

[0046] S2. Based on the initial operating mode and the current active power of wind power, a short-term wind power prediction curve is generated by combining historical operating data sequences, and an initial charging and discharging power plan is generated accordingly.

[0047] For example, generating a short-term wind power prediction curve includes: selecting a subset of historical data from the historical operating data sequence that matches the initial operating mode, based on the initial operating mode.

[0048] Based on the current active power of wind power, search for historical moments in the historical data subset that are in the same range as the current active power of wind power.

[0049] Extract the historical wind power sequence within the historical reference time period for each historical moment to form a reference curve for each historical power change.

[0050] Obtain the wind power change curves for the historical reference period prior to the current moment, and calculate their similarity to each historical power change reference curve within the same time period. The similarity is measured by calculating the Pearson correlation coefficient of the power values ​​at each point on the two curves within the same time period, and the historical curve with the highest correlation coefficient is selected as the prediction reference.

[0051] The historical power change curve with the highest similarity is selected, and its power change curve within the historical reference period after the historical moment is used as the short-term wind power prediction curve.

[0052] For example, generating the initial charge and discharge power plan includes: determining the permissible direction of charge and discharge power within the look-ahead optimization time window based on the initial operating mode, and determining the rated charging power or rated discharging power corresponding to the permissible direction as an amplitude limit based on the rated power of the grid wind and storage system.

[0053] It should be added that the determination of the permissible direction and amplitude limit of charging and discharging power includes: if the initial operating mode is charging mode, then the permissible direction of charging and discharging power is charging. If it is discharging mode, then the permissible direction is discharging. If it is standby mode or frequency modulation mode, then the permissible direction is no definite power flow or zero power.

[0054] If the initial operating mode is charging mode, the upper limit of the charging power amplitude is set to the rated charging power of the grid-connected wind and energy storage system. If the initial operating mode is discharging mode, the upper limit of the discharging power amplitude is set to the rated discharging power of the grid-connected wind and energy storage system.

[0055] From the short-term wind power prediction curve, the predicted power values ​​for each future time within the forward optimization time window are extracted, and the corresponding grid dispatch command values ​​for each time are obtained.

[0056] The difference between the grid dispatch command value and the predicted power value at that moment is calculated as the theoretical power demand of wind power and energy storage at that moment.

[0057] Determine whether the direction of the theoretically required power is consistent with the permitted direction. If they are inconsistent, it indicates a fundamental conflict between the grid's dispatching needs at that specific moment and the operational capability of the wind and energy storage equipment based on its own safe state of charge range. For example, when the grid needs wind and energy storage to discharge to supplement the power deficit, but the wind and energy storage is only allowed to charge due to its low state of charge, forced discharge would endanger the equipment's safety. In this case, the charging and discharging power value at that moment is set to 0.

[0058] If they match, the absolute value of the theoretical power demand is compared with the amplitude limit. If the absolute value is less than or equal to the amplitude limit, it indicates that at this moment, the power compensation demand of the power grid is within the safe output capacity range of the rated power of the power grid wind-storage system under the corresponding working mode. In this case, the theoretical power demand is directly used as the charging and discharging power value at this moment.

[0059] If the absolute value is greater than the amplitude limit, it indicates that the power compensation demand of the power grid at that moment exceeds the dynamic bearing limit of the long-term safe operation rating of the power grid wind and energy storage system hardware. In this case, the amplitude limit is taken as the absolute value of the charging and discharging power at that moment.

[0060] The charging and discharging power values ​​at all times within the forward-looking optimization time window are summarized to form an initial charging and discharging power plan.

[0061] S3. Predict the trajectory of wind and energy storage state of charge within the look-ahead optimization time window by simulating the execution of the plan, and verify whether the initial charge and discharge power plan passes. If the verification passes, execute the initial charge and discharge power plan.

[0062] For example, the verification of whether the initial charge-discharge power plan is passed includes: based on the initial charge-discharge power plan, simulating and calculating the evolution trajectory of the wind-storage charge state within the look-ahead optimization time window.

[0063] It should be added that the simulation calculation of the evolution trajectory of the wind and energy storage state of charge within the forward optimization time window includes: Q1, determining the initial state and parameters of the simulation: taking the actual wind and energy storage state of charge value of the grid wind and energy storage system at the current moment as the initial value of the simulation, and taking the initial charging and discharging power plan as the input parameter, wherein the plan includes the planned charging and discharging power value of the grid wind and energy storage system at each moment within the forward optimization time window.

[0064] Q2. Perform iterative calculations: Starting from the beginning of the look-ahead optimization time window, iterate step by step to the end of the window according to the preset calculation step size; for each iteration time, based on the planned charging and discharging power value and the energy conversion efficiency of the grid wind and energy storage system at that time, calculate the net change in the energy stored in the grid wind and energy storage system within that step size, and then obtain the change in the state of charge of wind and energy storage corresponding to that step size. Add this change to the predicted state of charge of wind and energy storage at the previous iteration time to obtain the predicted state of charge of wind and energy storage at the current iteration time.

[0065] Q3. Generate prediction trajectory: Summarize the prediction values ​​of wind and energy storage state of charge at each iteration time within the aforementioned look-ahead optimization time window to form the wind and energy storage state of charge evolution trajectory.

[0066] The predicted values ​​of wind-storage charge state at each moment on the trajectory are compared with the preset safe range of wind-storage charge state.

[0067] If the predicted values ​​of wind and energy storage state of charge at all times are within the safe range of wind and energy storage state of charge, then the verification is successful.

[0068] If the predicted state of charge of wind and energy storage at any given moment is not within the safe range of the state of charge of wind and energy storage, the verification will fail.

[0069] S4. If the verification fails, determine the predicted crossing point of the trajectory crossing the safety boundary based on the wind-storage state of charge prediction trajectory, analyze the changing trend of the short-term wind power prediction curve before and after the predicted crossing point, re-determine the safe working mode based on the trend and generate an optimized charging and discharging power plan.

[0070] Please see Figure 3 As shown, for example, the determination of the safe working mode includes: obtaining the moment corresponding to the first crossing of the safe interval of the wind-storage-charged state from the wind-storage-charged state evolution trajectory as the predicted crossing point.

[0071] From the short-term wind power prediction curve, identify and obtain the local power maxima and local power minima on the curve, and calculate the time distance between the predicted crossing point and each power extremum point.

[0072] If at least one extreme point is less than the preset extreme neighborhood threshold in time distance from the predicted crossing point, the predicted crossing point is determined to be in a sensitive area where the power trend turns, and the corresponding safe working mode is standby mode.

[0073] It should be added that the preset extreme value neighborhood threshold is a time window value used to determine the degree of temporal correlation between the predicted crossing point and the power extreme point in the wind power prediction curve. The preset extreme value neighborhood threshold is obtained through the following steps: Y1. Based on the technical specifications of the converter and battery management system in the grid wind-storage system, obtain the step response time required from the issuance of the command to the power output reaching a steady state, as the power response inertia time.

[0074] Y2. Retrieve the ultra-short-term wind power forecast data and actual output data during the historical operation period, and calculate the forecast error sequence; perform time-domain analysis on the error sequence, and statistically analyze the average time interval between sign reversal or local extrema, as a typical time scale for the forecast error.

[0075] Y3. Compare the power response inertia time with the prediction error typical time scale, and set the maximum value of the two as the preset extreme value neighborhood threshold.

[0076] If the time distance between the predicted crossing point and all extreme points is greater than or equal to the preset extreme value neighborhood threshold, then analyze the slope change characteristics of the short-term wind power prediction curve within the trend analysis window before and after the predicted crossing point.

[0077] If the slope changes from positive to negative, the wind power is determined to be in a deterministic downward trend, and the corresponding safe operating mode is the discharge-dominated mode. If the slope changes from negative to positive, the wind power is determined to be in a deterministic upward trend, and the corresponding safe operating mode is the charging-dominated mode.

[0078] Furthermore, if the slope change characteristics of the short-term wind power prediction curve do not meet the conditions of changing from positive to negative or from negative to positive within the trend analysis window before and after the predicted crossing point, the safe operating mode is determined to be standby mode to ensure that wind and energy storage safety is prioritized when the trend is unclear. When the safe operating mode is standby mode, the value of the optimized charging and discharging power plan in the corresponding time period can be directly set to zero, and the subsequent rolling time-domain optimization calculation steps can be skipped.

[0079] For example, generating an optimized charge and discharge power plan includes: determining the permissible direction of charge and discharge power and its corresponding power amplitude range based on the safe operating mode.

[0080] Within the power amplitude range, with the optimization objective of smoothing the wind-storage charge state trajectory within the look-ahead optimization time window and keeping it away from the safety boundary, the rolling time-domain optimization method is used to calculate and optimize the charging and discharging power plan.

[0081] The rolling time-domain optimization method is achieved by constructing and solving an optimization model. The optimization objective of the model is to minimize the deviation between the predicted wind-storage state of charge (SOC) values ​​at each moment within the look-ahead optimization time window and the pre-set center value of the safe SOC of the wind-storage system. The deviation is quantified by the sum of the squares of the deviations at each moment. To achieve the above optimization objective, the following three types of constraints must be met simultaneously: (1) Power amplitude constraint: The charging and discharging power values ​​at each moment must not exceed the allowable range determined according to the safe operating mode; (2) State safety constraint: The predicted wind-storage SOC values ​​at each moment are within the range defined by the preset safe upper and lower limits; (3) Power change rate constraint: The rate of change of charging and discharging power between any two adjacent moments must not exceed the maximum ramp rate allowed by the technical specifications of the wind-storage converter.

[0082] The model established based on the aforementioned optimization objective and constraints mathematically belongs to a quadratic programming problem with linear constraints. For such problems, mature numerical optimization algorithms such as the interior-point method or the effective set method can be used to solve them, thereby obtaining an optimized charging and discharging power plan that meets all requirements.

[0083] Furthermore, the calculation and optimization of the charge / discharge power plan is performed according to the following steps: Z1, with the predicted crossing point as the center, a preset correction time window is extended forward and backward. Based on the re-determined safe operating mode, the target adjustment direction of the charge / discharge power within this window is determined.

[0084] It is understood that the length of the correction time window must be set to ensure that it covers the time required from the application of power adjustment to the point where it has a significant impact on the state of charge of wind and energy storage at the predicted crossing point, and is generally not less than the sum of the power response inertia time of the grid wind and energy storage system and the control system delay time.

[0085] If the safe operating mode is forced discharge or deceleration charging, the target adjustment direction is to adjust the actual charging and discharging power towards the discharge direction relative to the original plan within a certain period before the predicted crossing point; if the safe operating mode is forced charging or deceleration charging, the target adjustment direction is to adjust the actual charging and discharging power towards the charging direction relative to the original plan within a certain period before the predicted crossing point.

[0086] Z2. Within the correction time window, apply a preset base power correction amount to the initial charge and discharge power plan; based on the power plan after applying the correction amount, re-simulate and calculate the predicted trajectory of wind-storage state of charge within the future time window.

[0087] Z3. Determine whether the new predicted trajectory after applying the correction successfully avoids crossing the safety boundary near the predicted crossing point and does not trigger new boundary crossing risks at other locations.

[0088] If the verification passes, the power plan after applying the base power correction amount is taken as a candidate optimized charge and discharge power plan; if the verification fails, the base power correction amount is amplified in the same direction or reduced in the opposite direction according to the out-of-bounds situation, and the simulation and verification are re-executed in step S4-2 until the minimum effective correction amount that keeps the wind-storage charge state trajectory within the safety boundary throughout the entire process is obtained.

[0089] Z4. Apply the final determined effective correction amount smoothly to the correction time window of the initial charge and discharge power plan in a ramp or step manner, and ensure that the power change rate meets the equipment limit; keep the original value of the power plan outside the correction time window, and finally generate the optimized charge and discharge power plan.

[0090] The generated optimized charge and discharge power plan is validated for power change rate to ensure that its change rate does not exceed the maximum allowable power change rate of the wind-storage converter.

[0091] S5. Based on the predicted values ​​corresponding to the actual wind power and the optimized charging and discharging power plan, determine whether the strategy is mismatched. If there is no mismatch, execute the optimized charging and discharging power plan; otherwise, return to step S1.

[0092] For example, determining whether a strategy is mismatched includes continuously collecting actual wind power sequences within a preset strategy evaluation time window. The strategy evaluation time window is less than or equal to one-third of the look-ahead optimization time window to ensure timely evaluation.

[0093] Obtain the predicted wind power sequence within the strategy evaluation time window for the optimized charge and discharge power plan.

[0094] Calculate the average absolute deviation between the actual wind power sequence and the predicted wind power sequence.

[0095] If the average absolute deviation exceeds the preset mismatch judgment threshold, the strategy is determined to be mismatched; otherwise, the strategy is determined not to be mismatched.

[0096] It should be added that the preset mismatch judgment threshold is used to evaluate the degree of matching between the current energy management strategy and the actual operating state online. The preset mismatch judgment threshold is obtained as follows: P1, obtain the current state of charge of the wind and energy storage system and the corresponding maximum allowable charging and discharging power; based on the currently effective charging and discharging power plan, calculate the remaining real-time power adjustment margin of the wind and energy storage system to track the grid dispatch instructions; subtract a preset safety margin from this adjustment margin to obtain the dynamic boundary value corresponding to the system's real-time compensation capability.

[0097] P2. Retrieve the ultra-short-term wind power prediction sequence and its corresponding actual power sequence from historical operational data; within the sliding time window, calculate the average absolute error between the predicted and actual values ​​at each moment to form an error sequence; calculate the mean of the error sequence. with standard deviation And based on the preset confidence level coefficient Calculate the statistical baseline boundary using the following formula. , .

[0098] P3. Compare the dynamic boundary value with the statistical benchmark boundary, and take the smaller value of the two as the preset mismatch judgment threshold.

[0099] Furthermore, to avoid continuous oscillations in the system control mode between steps S1 and S5 due to frequent strategy mismatches under extreme operating conditions such as continuous and drastic fluctuations in wind power, an anti-oscillation mechanism can be introduced after determining strategy mismatch and before executing the operation of returning to step S1. The anti-oscillation mechanism includes: counting the number of times the strategy is determined to be mismatched within a recent continuous period; if the number does not reach a preset frequent reset number threshold, then executing the operation of returning to step S1; if the number has reached or exceeded the frequent reset number threshold, then maintaining the current optimized charging and discharging power plan unchanged, and generating a system alarm signal requiring manual intervention to ensure system control stability.

[0100] The threshold for the number of frequent resets can be set according to the system's requirements for control stability. As an example, it can be set as the maximum number of policy resets allowed within a monitoring period that is 3 to 5 times the length of the policy evaluation time window, such as 2 to 3 times.

[0101] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0102] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0103] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0104] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0105] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A precise energy management method for wind and energy storage in power grids based on dynamic optimization control, characterized by: The method includes: S1. Based on the set of charging and discharging actions determined by the wind-storage charge state, and combined with the instantaneous deviation direction and magnitude of the current wind power active power and the grid dispatch command, a collaborative decision is made to determine the initial working mode. S2. Based on the initial operating mode and the current active power of wind power, a short-term wind power prediction curve is generated by combining historical operating data sequences, and an initial charging and discharging power plan is generated accordingly. S3. Predict the trajectory of wind and energy storage state of charge within the look-ahead optimization time window by simulating the execution of the plan, and verify whether the initial charge and discharge power plan passes. If the verification passes, execute the initial charge and discharge power plan. S4. If the verification fails, determine the predicted crossing point of the trajectory crossing the safety boundary based on the wind-storage charge state prediction trajectory, analyze the changing trend of the short-term wind power prediction curve before and after the predicted crossing point, re-determine the safe working mode based on the trend and generate an optimized charging and discharging power plan. S5. Based on the predicted values ​​corresponding to the actual wind power and the optimized charging and discharging power plan, determine whether the strategy is mismatched. If there is no mismatch, execute the optimized charging and discharging power plan; otherwise, return to step S1.

2. The method for precise energy management of power grid wind and energy storage based on dynamic optimization control according to claim 1, characterized in that: The determination of the initial working mode includes: Based on the state of charge values ​​of wind and energy storage, determine the set of charging and discharging actions of the grid wind and energy storage system; Calculate the deviation between the current active power of wind power and the required active power of wind power corresponding to the grid dispatch command, and generate the expected power compensation demand based on the direction and magnitude of the deviation. The desired power compensation requirement is matched with the charging and discharging action set of the grid wind and energy storage system, and the initial working mode of the grid wind and energy storage system is obtained according to the preset priority rules.

3. The method for precise energy management of power grid wind and energy storage based on dynamic optimization control according to claim 2, characterized in that: The set of charging and discharging actions for the power grid wind and energy storage system includes: Obtain the discharge prohibition threshold and charging prohibition threshold of the grid wind-storage system. The first interval is defined as the threshold below the discharge prohibition threshold, the second interval is defined as the threshold between the discharge prohibition threshold and the charging prohibition threshold, and the third interval is defined as the threshold above the charging prohibition threshold. If the wind-storage state of charge value is in the first interval, then the action set is defined as charging; If the wind-storage state of charge value is in the second range, then the action set is defined as charging, discharging and standby; If the wind-storage state of charge value is in the third interval, then the action set is defined as discharge.

4. The method for precise energy management of power grid wind and energy storage based on dynamic optimization control according to claim 2, characterized in that: The desired power compensation requirement includes: Subtract the current active power of wind power from the required active power of wind power to obtain the instantaneous deviation value; Obtain the wind power active power sequence within the instantaneous analysis time window prior to the current moment, and calculate its instantaneous change slope; The instantaneous deviation value and the instantaneous change slope are coupled and calculated to generate the initial demand value; The initial demand value is subjected to limiting processing based on the real-time power capacity of the power grid wind-storage system to obtain the desired power compensation demand.

5. The method for precise energy management of power grid wind and energy storage based on dynamic optimization control according to claim 2, characterized in that: The generated short-term wind power prediction curve includes: Based on the initial working mode, a subset of historical data that matches the initial working mode is selected from the historical operating data sequence; Based on the current active power of wind power, search for historical moments in the subset of historical data that are in the same range as the current active power of wind power. Extract the historical wind power sequence within the historical reference time period for each historical moment to form a reference curve for each historical power change; Obtain the wind power change curve within the historical reference time period before the current moment, and calculate its similarity with each historical power change reference curve within the same time period; The historical power change curve with the highest similarity is selected, and its power change curve within the historical reference period after the historical moment is used as the short-term wind power prediction curve.

6. The method for precise energy management of power grid wind and energy storage based on dynamic optimization control according to claim 1, characterized in that: The generation of the initial charge / discharge power plan includes: Based on the initial working mode, the permissible direction of charging and discharging power within the look-ahead optimization time window is determined, and based on the rated power of the grid wind and storage system, the rated charging power or rated discharging power corresponding to the permissible direction is determined as the amplitude limit. From the short-term wind power prediction curve, extract the predicted power values ​​for each future time within the forward optimization time window, and obtain the corresponding grid dispatch command values ​​for each time. Calculate the difference between the grid dispatch command value and the predicted power value at that moment, and use it as the theoretical power demand of wind power storage at that moment; Determine whether the direction of the theoretically required power is consistent with the allowed direction. If they are not consistent, set the charging and discharging power value at that moment to 0. If they match, the absolute value of the theoretical power demand is compared with the amplitude limit. If the absolute value is less than or equal to the amplitude limit, the theoretical power demand is directly used as the charging and discharging power value at that moment. If the absolute value is greater than the amplitude limit, then the amplitude limit is taken as the absolute value of the charging and discharging power at that moment; The charging and discharging power values ​​at all times within the forward-looking optimization time window are summarized to form an initial charging and discharging power plan.

7. The method for precise energy management of power grid wind and energy storage based on dynamic optimization control according to claim 1, characterized in that: The verification of whether the initial charge / discharge power plan passes includes: Based on the initial charge and discharge power plan, the evolution trajectory of the wind-storage charge state within the look-ahead optimization time window is simulated and calculated. The predicted values ​​of wind-storage-charge state at each moment on the trajectory are compared with the preset safe range of wind-storage-charge state. If the predicted values ​​of wind-storage state of charge at all times are within the safe range of wind-storage state of charge, then the verification is successful. If the predicted state of charge of wind and energy storage at any given moment is not within the safe range of the state of charge of wind and energy storage, the verification will fail.

8. The method for precise energy management of power grid wind and energy storage based on dynamic optimization control according to claim 1, characterized in that: The safe operating mode is determined as follows: The moment when the trajectory first crosses the safe interval of the wind-storage-charged state is obtained from the wind-storage-charged state evolution trajectory as the predicted crossing point. From the short-term wind power prediction curve, identify and obtain the local maximum and local minimum power points on the curve, and calculate the time distance between the predicted crossing point and each power extreme point. If at least one extreme point has a time distance from the predicted crossing point that is less than the preset extreme value neighborhood threshold, then the safe working mode is determined to be standby mode. If the time distance between the predicted crossing point and all extreme points is greater than or equal to the preset extreme neighbor threshold, then analyze the slope change characteristics of the short-term wind power prediction curve within the trend analysis window before and after the predicted crossing point. If the slope changes from positive to negative, the safe operating mode is determined to be the discharge-dominated mode; if the slope changes from negative to positive, the safe operating mode is determined to be the charging-dominated mode.

9. The method for precise energy management of power grid wind and energy storage based on dynamic optimization control according to claim 1, characterized in that: The generation of the optimized charge / discharge power plan includes: Based on the aforementioned safe operating mode, determine the permissible direction of charging and discharging power and its corresponding power amplitude range; Within the power amplitude range, with the optimization objective of smoothing the wind-storage charge state trajectory within the look-ahead optimization time window and moving it away from the safety boundary, the rolling time-domain optimization method is used to calculate the optimized charging and discharging power plan. The generated optimized charge and discharge power plan is validated for power change rate to ensure that its change rate does not exceed the maximum allowable power change rate of the wind-storage converter.

10. The method for precise energy management of power grid wind and energy storage based on dynamic optimization control according to claim 1, characterized in that: The determination of whether the strategy is mismatched includes: Within the preset strategy evaluation time window, the actual wind power sequence is continuously collected; Obtain the predicted wind power sequence within the strategy evaluation time window for the optimized charging and discharging power plan; Calculate the average absolute deviation between the actual wind power sequence and the predicted wind power sequence; If the average absolute deviation exceeds the preset mismatch judgment threshold, the strategy is determined to be mismatched; otherwise, the strategy is determined not to be mismatched.

Images