A grid-connected energy storage state of charge recovery method and system based on joint opportunity score
Patent Information
- Application Number
- CN202611153156.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-31
- Publication Date
- 2026-10-30
AI Technical Summary
[0007]发明目的:本发明提供一种基于联合机会评分的并网储能荷电状态恢复方法及系统,旨在解决现有技术存在的储能参与一次调频时SOC恢复经济性差、与一次调频义务冲突、极端工况兜底能力不足等问题
[0069]1、通过将SOC恢复时机与电力市场信号联合决策,使储能在SOC偏低时优先选择低电价时段或AGC向下调节时段充电、在SOC偏高时优先选择高电价时段或AGC向上调节时段放电。
Smart Images

Figure CN123018972A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of power system energy storage control technology, and particularly relates to a method and system for restoring the state of charge of grid-connected energy storage based on joint chance scoring. Background Technology
[0002] When grid-connected energy storage participates in primary frequency regulation, its output power command is determined by the frequency deviation through a droop control law. Under conditions of long-term unidirectional frequency deviation in the system, such as continuous over-frequency operation during peak renewable energy periods and continuous under-frequency operation during evening peak periods, the energy storage will continuously charge and discharge in one direction, causing the state of charge (SOC) to monotonically deviate from the set target value in one direction. When the SOC reaches the upper or lower threshold (such as 10% or 90%), the energy storage is forced to withdraw from frequency regulation or significantly reduce its frequency regulation capability, thereby triggering a series of problems such as decreased frequency regulation availability, loss of frequency regulation compensation revenue, and deterioration of performance indicators.
[0003] Existing energy storage SOC recovery and scheduling methods attempt to address the aforementioned problems. However, these methods suffer from the following main shortcomings:
[0004] First, existing SOC recovery methods based on frequency signals, which correct SOC by modifying droop characteristics or response power, only work during the frequency modulation response process and cannot actively recover SOC within the frequency dead zone, resulting in limited recovery speed. On the other hand, fixed-rate self-recovery methods based on the frequency dead zone do not consider external market signals in terms of recovery timing. They may passively charge during periods of high electricity prices, thus increasing electricity purchase costs, or passively discharge during periods of low electricity prices, thus losing electricity sales revenue, resulting in poor economic efficiency.
[0005] Second, the electricity market arbitrage dispatch model is not coupled with the primary frequency regulation obligation. Existing methods that plan day-ahead / intra-day charge and discharge curves with spot market arbitrage or frequency regulation mileage compensation as the goal typically have a dispatch scale of 5 minutes to hours. They are not coupled with the second-level primary frequency regulation power command at the same control level and cannot directly serve the primary frequency regulation task with a second-level response. Furthermore, they do not explicitly reserve the power required for primary frequency regulation as a hard constraint, which may lead to insufficient frequency regulation response capability when large frequency events occur.
[0006] Third, there is a lack of mandatory recovery and dispatch feedback mechanisms under the SOC warning zone. When the SOC has deviated significantly from the target and the direction of the Automatic Generation Control (AGC) dispatch command is continuously opposite to the direction of SOC recovery, the existing methods cannot take effective protective actions without violating the mandatory AGC command. In the end, they can only rely on the hard protection at the energy storage device level, such as when the SOC reaches its bottom and stops operating. Summary of the Invention
[0007] Purpose of the invention: This invention provides a method and system for restoring the state of charge (SOC) of grid-connected energy storage based on joint opportunity scoring, aiming to solve the problems of poor economic efficiency of SOC restoration when energy storage participates in primary frequency regulation, conflict with primary frequency regulation obligations, and insufficient backup capability under extreme operating conditions in existing technologies.
[0008] Technical solution: The present invention provides a method for restoring the state of charge of grid-connected energy storage based on joint opportunity scoring, comprising:
[0009] S1. Obtain the state of charge (SOC), grid frequency, and rate of change of frequency of the energy storage system, and receive electricity market signals;
[0010] S2. Compile the grid frequency difference over a historical preset period into a grid historical frequency fluctuation statistical data set; obtain the primary frequency regulation parameters of the energy storage system; based on the grid historical frequency fluctuation statistical data set and the primary frequency regulation parameters of the energy storage system, predict the maximum power to be called for primary frequency regulation in the future preset period; and calculate the upper limit of power available for state of charge recovery by combining the rated power of the energy storage system.
[0011] S3. Calculate the SOC deviation direction and SOC deviation degree score based on the State of Charge (SOC) and the preset target SOC; calculate the market deviation direction and market deviation degree score based on the electricity market signal and the preset electricity market signal reference value; construct a joint opportunity score based on the SOC deviation direction, SOC deviation degree score, market deviation direction, and market deviation degree score.
[0012] S4. Compare the joint opportunity score with a preset trigger threshold. When the joint opportunity score is greater than the preset trigger threshold and the upper limit of power available for state of charge recovery is greater than 0, issue a state of charge recovery power command.
[0013] S5. During the execution of the state-of-charge recovery power command, monitor the grid frequency deviation, frequency change rate, and primary frequency regulation response power of the energy storage system. When any preset zeroing condition is met, reset the state-of-charge recovery power to zero, so that the energy storage system returns to an operating mode with primary frequency regulation response as the sole objective. The zeroing conditions include:
[0014] (a) The power grid frequency deviation exceeds the preset frequency emergency threshold;
[0015] (b) The rate of frequency change exceeds the preset emergency threshold for the rate of frequency change;
[0016] (c) The primary frequency regulation response power of the energy storage system is not zero.
[0017] Furthermore, the electricity market signals include AGC dispatch command signals and spot market price signals; the primary frequency regulation parameters of the energy storage system include the droop coefficient of the energy storage system. Frequency dead zone width Rated power .
[0018] Furthermore, the prediction of the maximum power to be activated during a preset frequency modulation period in the future includes:
[0019] Based on historical frequency fluctuation statistics of the power grid, frequency deviation within a predetermined future time period is estimated using empirical distribution or normal fitting methods. The probability distribution is taken as the extreme value corresponding to the 95% confidence level. ;
[0020] According to the extreme value droop coefficient of energy storage system Frequency dead zone width Rated power Predict the maximum power to be used by a frequency modulation within a preset time period in the future. The formula is:
[0021] .
[0022] Furthermore, the power limit for state-of-charge recovery can be calculated using the rated power of the combined energy storage system, as shown in the formula:
[0023] ;
[0024] in, Rated power, .
[0025] Furthermore, the step of calculating the SOC deviation direction and SOC deviation degree score based on the State of Charge (SOC) and the preset target SOC includes: based on the SOC and the preset target SOC... The formula for calculating SOC deviation is:
[0026] ;
[0027] The symbol represents the SOC deviation direction, denoted as the SOC recovery direction symbol. The formula is:
[0028] ;
[0029] ,but , ,but , At that time, ;
[0030] SOC Deviation Score Through SOC deviation absolute value With the preset normalization constant The calculation yields the following formula:
[0031] .
[0032] Furthermore, the calculation of the market deviation direction and degree score based on the electricity market signal and the preset electricity market signal reference value includes: when the electricity market signal is an AGC dispatch command signal available, the score is based on the energy storage power corresponding to the AGC dispatch command. Symbol of SOC recovery direction Calculate the deviation direction of the AGC market. AGC Market Deviation Score ;
[0033] Deviating from the AGC market As a market deviation direction The AGC market is deviating from its direction. The formula is: ;
[0034] when and When the numbers are the same, ,when and When the signs are different, ,when hour, ;
[0035] AGC market deviation score As a score for market deviation The formula is:
[0036] ;
[0037] in, The energy storage power corresponding to the AGC scheduling command; This is the preset AGC command normalization constant, i.e., the preset electricity market signal reference value;
[0038] When electricity market signals are available as spot market price signals, based on the current spot market price... Compared with the preset reference price That is, the preset electricity market signal reference value, to calculate the direction of price market deviation. Price market deviation score ;
[0039] The price market deviates from its direction As a market deviation direction Price market deviation The formula is: ;
[0040] when and hour, ;when and hour, Other cases It can be -1 or 0;
[0041] Scoring the degree of price market deviation As a score for market deviation The formula is:
[0042] ;
[0043] in, This refers to the current spot market price; For preset reference price, This represents the standard deviation of prices during the same period.
[0044] When both the AGC scheduling command signal and the spot market price signal are available, the market deviation score is the larger of the AGC market deviation score and the price market deviation score, and the market deviation direction is the market deviation direction corresponding to the larger value.
[0045] Furthermore, the joint opportunity scoring formula is as follows:
[0046] ;
[0047] in, Scoring for joint opportunities, Score the degree of market deviation. Score the degree of deviation from SOC. , The weighting coefficients and .
[0048] Furthermore, the trigger threshold The base value is used when the deviation from the State of Charge (SOC) does not exceed the preset warning threshold. When the SOC deviation exceeds the warning threshold, the value is reduced to... ;when and At that time, a power restoration command for the state of charge is issued. ;otherwise ;
[0049] The formula for the state-of-charge recovery power command is:
[0050] , ;
[0051] The state-of-charge recovery power command The amplitude satisfies the following formula:
[0052] , ;
[0053] in, The preset recovery strength coefficient; It is a pre-defined monotonically non-decreasing function;
[0054] When the energy storage system executes the AGC scheduling command, it restores the power command. Power corresponding to AGC command and primary frequency response power The output is superimposed at the PCS output terminal, and the total output power is:
[0055] ;
[0056] and .
[0057] Furthermore, this also includes S6, and the degree of deviation from SOC. Exceeding the preset warning threshold and detecting that the AGC scheduling command direction is opposite to the SOC recovery direction The opposite continues, and the duration exceeds the preset time. At that time, the energy storage system feeds back the reduced adjustable capacity to the upper-level AGC scheduling system. ,satisfy
[0058] .
[0059] The present invention discloses a grid-connected energy storage state-of-charge recovery system based on joint opportunity scoring, comprising:
[0060] The data acquisition module is used to acquire the state of charge (SOC), grid frequency, and rate of change of the energy storage system, and to receive electricity market signals.
[0061] The power upper limit prediction module is used to assemble the grid frequency difference over a historical preset period into historical grid frequency fluctuation statistics; obtain the primary frequency regulation parameters of the energy storage system; based on the historical grid frequency fluctuation statistics and the primary frequency regulation parameters of the energy storage system, predict the maximum power to be called by primary frequency regulation in the future preset period; and calculate the power upper limit that can be used for state of charge recovery in combination with the rated power of the energy storage system.
[0062] The joint opportunity score construction module is used to calculate the SOC deviation direction and SOC deviation degree score based on the State of Charge (SOC) and the preset target SOC; to calculate the market deviation direction and market deviation degree score based on the electricity market signal and the preset electricity market signal reference value; and to construct the joint opportunity score based on the SOC deviation direction, SOC deviation degree score, market deviation direction, and market deviation degree score.
[0063] The triggering module is used to compare the joint opportunity score with a preset triggering threshold. When the joint opportunity score is greater than the preset triggering threshold and the upper limit of power available for state of charge recovery is greater than 0, a state of charge recovery power command is issued.
[0064] The execution and zeroing module is used to monitor the grid frequency deviation, frequency change rate, and primary frequency regulation response power of the energy storage system during the execution of the state-of-charge recovery power command. When any preset zeroing condition is met, the state-of-charge recovery power is zeroed, so that the energy storage system returns to an operating mode with primary frequency regulation response as the sole objective. The zeroing conditions include:
[0065] (a) The power grid frequency deviation exceeds the preset frequency emergency threshold;
[0066] (b) The rate of frequency change exceeds the preset emergency threshold for the rate of frequency change;
[0067] (c) The primary frequency regulation response power of the energy storage system is not zero.
[0068] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages:
[0069] 1. By combining the timing of SOC recovery with electricity market signals, energy storage can prioritize charging during low-price periods or periods of downward AGC adjustment when SOC is low, and prioritize discharging during high-price periods or periods of upward AGC adjustment when SOC is high.
[0070] 2. Through power reservation calculation and emergency switching mechanism, the primary frequency regulation power is regarded as the highest priority hard constraint to ensure that the PCS retains sufficient capacity to respond to frequency events when any SOC recovery action occurs.
[0071] 3. Through the warning zone adaptive threshold mechanism and scheduling feedback mechanism, when the SOC deviates significantly, it is forced to recover and actively feeds back the limited capacity to the upper-level scheduling, so as to avoid the SOC hitting the hard upper and lower limits and being forced to exit frequency regulation, thereby improving the availability of energy storage resources at all times. Attached Figure Description
[0072] Figure 1 This is a flowchart of the method of the present invention.
[0073] Figure 2 This is a schematic diagram of the topology and controller location of an energy storage power station system.
[0074] Figure 3 The droop characteristic curve of primary frequency modulation is labeled with dead zone and emergency threshold.
[0075] Figure 4 This is a graph showing the relationship between the SOC warning zone and the trigger threshold.
[0076] Figure 5 This is a schematic diagram of the dispatch and feedback mechanism for the warning zone. Detailed Implementation
[0077] like Figure 1 As shown, the present invention provides a method for restoring the state of charge of grid-connected energy storage based on joint opportunity scoring, comprising:
[0078] S1. Obtain the energy storage system's state of charge (SOC), grid frequency (f), and rate of frequency change. It also receives electricity market signals, including automatic generation control (AGC) dispatch command signals and spot market price signals.
[0079] S2. Compile the grid frequency differences over a preset historical time period into historical grid frequency fluctuation statistics; obtain the primary frequency regulation parameters of the energy storage system; and based on the historical grid frequency fluctuation statistics and the primary frequency regulation parameters of the energy storage system, predict the maximum power to be activated by primary frequency regulation within a preset future time period. And combined with the rated power of the energy storage system Calculate the upper limit of power available for state-of-charge recovery. ;
[0080] Specifically, the power grid frequency differences over a preset historical time period are used to compile historical power grid frequency fluctuation statistics; the primary frequency regulation parameters of the energy storage system are obtained, including the droop coefficient of the energy storage system. Frequency dead zone width Rated power ;
[0081] Based on historical frequency fluctuation statistics of the power grid, frequency deviation within a predetermined future time period is estimated using empirical distribution or normal fitting methods. The probability distribution is taken as the extreme value corresponding to the 95% confidence level. ;
[0082] According to the extreme value droop coefficient of energy storage system Frequency dead zone width Rated power Predict the maximum power to be used by a frequency modulation within a preset time period in the future. The formula is:
[0083] ;
[0084] Based on the rated power of the energy storage system The maximum power to be activated by a single frequency modulation within a preset time period in the future. Calculate the upper limit of power available for state-of-charge (SOC) recovery. As a hard constraint for subsequent SOC recovery actions, the power limit The formula is:
[0085] ;
[0086] in, ;
[0087] S3. Calculate the SOC deviation direction and SOC deviation degree score based on the State of Charge (SOC) and the preset target SOC; calculate the market deviation direction and market deviation degree score based on the electricity market signal and the preset electricity market signal reference value; construct a joint opportunity score S based on the SOC deviation direction, SOC deviation degree score, market deviation direction, and market deviation degree score.
[0088] Specifically, based on the State of Charge (SOC) and the preset target State of Charge... The formula for calculating SOC deviation is:
[0089] ;
[0090] The symbol represents the SOC deviation direction, denoted as the SOC recovery direction symbol. The formula is:
[0091] ;
[0092] This means that the SOC is too high. This indicates that discharge is required for recovery;
[0093] This means that the SOC is low. This indicates that charging is required to restore functionality;
[0094] hour, This indicates that no recovery is needed.
[0095] SOC Deviation Score Through SOC deviation absolute value With the preset normalization constant The calculation yields the following formula:
[0096] ;
[0097] Based on the type of received electricity market signal, scores for the direction and degree of market deviation are calculated respectively.
[0098] When the electricity market signal is an Automatic Generation Control (AGC) dispatch command signal, the energy storage power is based on the AGC dispatch command. Symbol of SOC recovery direction Calculate the deviation direction of the AGC market. AGC Market Deviation Score ; deviating the direction of the AGC market As a market deviation direction The AGC market is deviating from its direction. The formula reflecting the consistency between the AGC adjustment direction and the SOC recovery direction is:
[0099] ;
[0100] when and When the signs are the same, the AGC adjustment direction is consistent with the SOC recovery direction, and the market deviates from the direction. ,when and When the signs are opposite, the AGC adjustment direction is opposite to the SOC recovery direction, and the market deviates from the direction. ,when At that time, without AGC adjustment instructions, the market deviated from its direction. ;
[0101] AGC Market Deviation Score The relative strength of AGC command power is reflected in the score of AGC market deviation. As a score for market deviation The formula is:
[0102] ;
[0103] in, This represents the energy storage power corresponding to the AGC scheduling command, with positive for discharging and negative for charging. The preset AGC command normalization constant (i.e., the preset electricity market signal reference value); when and Same number It is positive; and When the signs are different, Negative; hour, It is zero.
[0104] When electricity market signals are available as spot market price signals, based on the current spot market price... Compared with the preset reference price (i.e., the preset electricity market signal reference value), calculate the direction of price market deviation. Price market deviation score ; deviate from the direction of the price market As a market deviation direction Price market deviation The formula reflecting the synergy between the direction of price deviation and the direction of SOC recovery is:
[0105] ;
[0106] When the SOC is low, it needs to be charged to restore it. And the current price is lower than the reference price. )hour, When the SOC is too high, it needs to be discharged to recover. And the current price is higher than the reference price. )hour, Other cases It can be -1 or 0;
[0107] Price Market Deviation Score Reflecting the relative strength of price deviation, a score is assigned to the degree of price market deviation. As a score for market deviation The formula is:
[0108] ;
[0109] in, The price is the current spot market price, expressed in yuan / MWh. For preset reference price, This represents the standard deviation of prices during the same period. When the SOC is low, charging is required to restore it. And the current price is lower than the reference price. )hour Positive; when SOC is too high, discharge is required to restore it. And the current price is higher than the reference price. )hour Positive; other cases It can be negative or zero.
[0110] When both AGC scheduling command signals and spot market price signals are available, the market deviation score is the greater of the weighted sum or the price market deviation score, and the market deviation direction is the direction corresponding to the greater of the weighted sum or the price market deviation score.
[0111] The formula for taking the larger value between the AGC market deviation score and the price market deviation score is as follows:
[0112] ;
[0113] ;
[0114] Based on the SOC deviation direction, SOC deviation degree score, market deviation direction, and market deviation degree score, a joint opportunity score S is constructed, which integrates market economy and the necessity of SOC restoration into a unified decision indicator.
[0115] The formula for the joint opportunity score S is:
[0116] ;
[0117] in, , The weighting coefficients and ;
[0118] When SOC deviation degree When the preset warning threshold is exceeded, Automatically enlarge The corresponding reduction is made to increase the priority of SOC recovery in the scoring.
[0119] S4. The joint opportunity score S is compared with a preset trigger threshold. Compare; when and At that time, a power restoration command for the state of charge is issued. ;
[0120] Specifically, the joint opportunity score S is compared with a preset trigger threshold. Comparison, the trigger threshold The base value is used when the deviation from the State of Charge (SOC) does not exceed the preset warning threshold. When the SOC deviation exceeds the warning threshold, the value is reduced to... (satisfy );
[0121] when and At that time, a power restoration command for the state of charge is issued. ;otherwise ;
[0122] The formula for the state-of-charge recovery power command is:
[0123] , ;
[0124] The state-of-charge recovery power command The amplitude satisfies the following formula:
[0125] , ;
[0126] in, The preset recovery strength coefficient; For definition in The monotonically non-decreasing function on the SOC causes the larger the deviation from the target value, the greater the recovery amplitude. The symbol is composed of Decide;
[0127] When the energy storage system executes the AGC scheduling command, it restores the power command. Power corresponding to AGC command and primary frequency response power The output is superimposed at the PCS output terminal, and the total output power is:
[0128] ;
[0129] And require .
[0130] When the total power exceeds the limit, priority is given to ensuring and The execution, for Limit the amplitude or set it to zero.
[0131] S5, Power recovery command in the state of charge During the execution, the power grid frequency deviation was monitored in real time. Rate of change of frequency Primary frequency regulation response power of energy storage system When any preset zeroing condition is met, the state of charge is restored to power. Zeroing the system restores it to an operating mode where primary frequency regulation response is the sole objective; the zeroing conditions include:
[0132] (a) Exceeding the preset frequency emergency threshold;
[0133] (b) The preset frequency change rate emergency threshold has been exceeded.
[0134] (c) The primary frequency regulation response power of the energy storage system If the value is not zero, it means the frequency deviation exceeds the dead zone, and frequency modulation has been activated.
[0135] S6, Warning Zone Dispatch Feedback: When SOC Deviation Degree Exceeding the preset warning threshold and detecting that the AGC scheduling command direction is opposite to the SOC recovery direction The opposite continues, and the duration exceeds the preset time. At that time, the energy storage system feeds back the reduced adjustable capacity to the upper-level AGC scheduling system. This reduces or avoids instructions that are opposite to the SOC recovery direction during subsequent AGC instruction allocation; among which satisfy And its direction is restricted to being the same as... A consistent direction.
[0136] This invention's method is compatible with various market environments, employing abstract descriptions of market signals (AGC commands, spot prices) and not being bound to specific market rules. This makes the method applicable to spot markets in different provinces across China, as well as in scenarios involving future policy adjustments. Furthermore, this invention's method can be smoothly integrated with existing BMS and EMS systems. Implemented as a station-level coordination controller, it uses standard power measurement quantities and market signal interfaces as inputs and outputs standard PCS power commands, ensuring compatibility with existing energy storage power station secondary systems and facilitating application in engineering retrofits and new projects.
[0137] The present invention discloses a grid-connected energy storage state-of-charge recovery system based on joint opportunity scoring, comprising:
[0138] The data acquisition module is used to acquire the state of charge (SOC), grid frequency, and rate of change of the energy storage system, and to receive electricity market signals.
[0139] The power upper limit prediction module is used to assemble the grid frequency difference over a historical preset period into historical grid frequency fluctuation statistics; obtain the primary frequency regulation parameters of the energy storage system; based on the historical grid frequency fluctuation statistics and the primary frequency regulation parameters of the energy storage system, predict the maximum power to be called by primary frequency regulation in the future preset period; and calculate the power upper limit that can be used for state of charge recovery in combination with the rated power of the energy storage system.
[0140] The joint opportunity score construction module is used to calculate the SOC deviation direction and SOC deviation degree score based on the State of Charge (SOC) and the preset target SOC; to calculate the market deviation direction and market deviation degree score based on the electricity market signal and the preset electricity market signal reference value; and to construct the joint opportunity score based on the SOC deviation direction, SOC deviation degree score, market deviation direction, and market deviation degree score.
[0141] The triggering module is used to compare the joint opportunity score with a preset triggering threshold. When the joint opportunity score is greater than the preset triggering threshold and the upper limit of power available for state of charge recovery is greater than 0, a state of charge recovery power command is issued.
[0142] The execution and zeroing module is used to monitor the grid frequency deviation, frequency change rate, and primary frequency regulation response power of the energy storage system during the execution of the state-of-charge recovery power command. When any preset zeroing condition is met, the state-of-charge recovery power is zeroed, so that the energy storage system returns to an operating mode with primary frequency regulation response as the sole objective. The zeroing conditions include:
[0143] (a) The power grid frequency deviation exceeds the preset frequency emergency threshold;
[0144] (b) The rate of frequency change exceeds the preset emergency threshold for the rate of frequency change;
[0145] (c) The primary frequency regulation response power of the energy storage system is not zero.
[0146] Regarding the technical solution of the present invention, the following is a detailed description... Figures 2 to 5 This will detail the experimental process of applying the technical solution to a specific experiment and the technical effects of the technical solution.
[0147] This experiment uses a 10MW / 20MWh grid-connected independent energy storage power station connected to a 110kV point of common coupling. The energy storage power station participates in the primary frequency regulation of the local power grid and receives AGC commands from the power grid dispatch center; the region already has an operational electricity spot market, and the energy storage can obtain real-time spot price signals. Figure 2 As shown, the energy storage power station is equipped with a station-level coordination controller (which executes the method described in this invention), which is connected to a battery management system (BMS) and a power conversion system (PCS) below, and to the power grid dispatching master station and the electricity spot market system above. It communicates with the energy storage device through IEC 61850 and with the dispatching master station through IEC 60870-5-104 telecontrol protocol.
[0148] To avoid ambiguity in directional descriptions, all directional concepts in the claims and specification of this invention are expressed using uniform algebraic notation, without using natural language directional terms such as "up" or "down." The specific conventions are as follows:
[0149] (1) Energy storage power P: positive for discharging and negative for charging;
[0150] (2) Frequency deviation , The rated frequency is 50Hz.
[0151] (3) SOC deviation ; This indicates that the SOC is too high and needs to be restored by discharging. This indicates that the SOC is too low and needs to be charged to recover.
[0152] (4) SOC recovery direction sign ; For discharge, For charging, No recovery is required;
[0153] (5) Energy storage power corresponding to AGC command : This is an upward adjustment command (requiring discharge). This is a downward adjustment command (requiring charging).
[0154] Algebraic expression of directional consistency: when and When the signs are the same, the AGC adjustment direction is consistent with the SOC recovery direction; when the signs are opposite, the opposite is true.
[0155] Step S1, Real-time Data Acquisition: The station-level coordination controller acquires data in each control cycle. (Example: 1 second) Data acquisition begins. The current state of charge (SOC), grid frequency f, and its rate of change are acquired from the measuring device. (in (Estimation using sliding window differential filtering); Receive the energy storage power corresponding to the current AGC command from the scheduling system communication interface. Receive the current spot market price from the market system communication interface. .
[0156] Step S2, Primary frequency modulation power reserve calculation: the maximum power that may be called up for primary frequency modulation. Based on historical frequency fluctuation statistics of the power grid, current renewable energy output forecasts, and current system frequency regulation reserve level estimates, the following formula is used for calculation:
[0157] ;
[0158] In the formula: Reserved value for primary frequency regulation power, in MW; The droop factor is expressed in MW / Hz; in this experiment, it is set to 200MW / Hz. To estimate the extreme values of frequency deviation under a predefined confidence level, historical frequency data from the past 24 hours are used to estimate the deviation within the next 15 minutes using empirical distribution or normal fitting methods. The probability distribution is taken, and the extreme values are taken at the 95% confidence level, with a typical value of 0.04Hz; The frequency dead zone width is 0.03Hz in this experiment (upper limit according to GB / T 40595). The rated power of the energy storage is 10MW in this experiment.
[0159] Calculated .
[0160] Based on this, the upper limit of power available for SOC recovery can be calculated:
[0161] ;
[0162] In the formula: The upper limit of power available for SOC recovery, in MW. =10−2=8 MW, this value serves as the hard upper limit for the amplitude of subsequent SOC recovery commands, ensuring that the PCS is always reserved for primary frequency modulation. The capacity.
[0163] S3, Construction of Joint Opportunity Score: The joint opportunity score S is constructed using the following weighting method:
[0164] ;
[0165] In the formula: S represents the joint chance score; The market signal score reflects the direction of the electricity market signal and the direction of SOC recovery. The degree of consistency; SOC deviation score; , These are the weighting coefficients. .
[0166] SOC deviation score is calculated using the following formula:
[0167] ;
[0168] In the formula: For SOC deviation, take =0.5; To predetermine the normalization constant, we set it to 0.5 in this experiment.
[0169] When the electricity market signal is an AGC dispatch command signal, the market signal score is determined by the following formula:
[0170] ;
[0171] In the formula: This represents the energy storage power corresponding to the AGC scheduling command, with positive for discharging and negative for charging. The normalization constant for the AGC command is 5MW, which is used in this experiment; Restore the direction sign for SOC. When and When the signs are the same (AGC adjustment direction is consistent with SOC recovery direction) Positive; negative when signs are opposite. hour It is zero.
[0172] When spot market price signals are available for the electricity market, the market signal score is determined by the following formula:
[0173] ;
[0174] In the formula: The price is the current spot market price, expressed in yuan / MWh. To set a reference price, this experiment uses the moving average of 400 yuan / MWh over the past 24 hours. The standard deviation of prices during the same period is taken as 100 yuan / MWh. When the SOC is low, charging is required to restore it. And the current price is lower than the reference price. )hour Positive; when SOC is too high, discharge is required to restore it. And when the current price is higher than the reference price Positive; other cases It can be negative or zero.
[0175] When both types of market signals are available simultaneously, the market signal score calculated when the AGC scheduling instruction signal is available and the market signal score calculated when the spot market price signal is available can be weighted and summed or the larger value can be taken. In this experiment, the larger value is taken to make it easier to trigger.
[0176] S4. Resumption command issuance and adaptive threshold: (e.g.) Figure 4 As shown, the trigger threshold and weighting coefficients Adaptive switching based on SOC deviation:
[0177] ;
[0178] ;
[0179] In the formula: The trigger threshold for the base region is set to 0.5 in this experiment; The threshold for triggering the alert zone is set to 0.2 in this experiment, which satisfies the condition. ; The SOC warning threshold is set to 0.2 in this experiment; For the base region weights, this experiment uses (0.5, 0.5); The warning zone weight is set to (0.2, 0.8) in this experiment. That is, when the SOC enters the warning zone, the SOC deviation weight automatically increases and the market signal weight decreases accordingly, so as to improve the priority of SOC recovery in the scoring.
[0180] when and At that time, the amplitude of the recovery command is dynamically determined by the following formula:
[0181] ;
[0182] In the formula: To set the recovery strength coefficient, a value of 0.6 was used in this experiment; For definition in A monotonically non-decreasing function on the t such that the greater the deviation of the SOC from the target value, the greater the recovery amplitude. This experiment uses t . And limit the result to .
[0183] The final power restoration command is:
[0184] ;
[0185] otherwise When the energy storage system executes the AGC scheduling command, it restores the power command. Power corresponding to AGC command and primary frequency response power The output is superimposed at the PCS output terminal, and the total output power is:
[0186] ;
[0187] And require .
[0188] When the total power exceeds the limit, priority is given to ensuring and The execution, for Limit the amplitude or set it to zero.
[0189] S5, Frequency Priority Emergency Switching: In the recovery command During execution, the controller monitors the grid frequency deviation in real time at millisecond intervals. rate of change Primary frequency regulation response power of energy storage system The primary frequency modulation droop characteristic curve with dead zone and emergency threshold labeled is as follows: Figure 3 As shown, immediately when any of the following conditions are met... Resetting the system to zero restores the energy storage system to an operating mode where primary frequency regulation response is the sole objective.
[0190] ;
[0191] ;
[0192] ;
[0193] In the formula: The emergency frequency threshold is set to 0.10Hz in this example. The emergency threshold for the rate of frequency change is set to 0.20 Hz / s in this example. The primary frequency modulation response power is determined by the frequency deviation via a droop control law. This indicates that the frequency deviation has exceeded the dead zone and frequency modulation has been activated. (Conditional expression) ,Mode Coverage of severe frequency event conditions. The coverage frequency has just left the dead zone but has not yet reached the emergency threshold. All three conditions must be met to ensure the integrity of the frequency modulation obligation. None of them can be omitted.
[0194] S6, Warning Zone Dispatch Feedback: (e.g., ...) Figure 5 As shown, the controller maintains a rolling counter. Record the continuous duration during which the SOC is in the alert zone and the AGC command direction is opposite to the SOC recovery direction. Scheduling feedback is triggered when the following conditions are met:
[0195] ;
[0196] In the formula: The feedback duration threshold is set to 5 minutes in this example. This continuous feature differs from typical instantaneous feedback, avoiding frequent false feedback caused by short-term reverse responses in AGC commands.
[0197] Upon triggering, the controller sends an adjustable capacity-limited signal to the scheduling system communication interface. This reduces or avoids the allocation of subsequent AGC instructions that are opposite to the direction of SOC recovery. satisfy:
[0198] ;
[0199] Right now The direction is restricted to being with This consistency guides the upper-level scheduler to avoid the allocation of instructions for continuously emptying the SOC, without violating the mandatory AGC instructions.
[0200] Rated power of energy storage in the experiment =10MW, rated capacity =20MWh (2-hour energy storage); droop coefficient =200MW / Hz; Frequency dead zone =±0.03Hz; Frequency emergency threshold =0.10Hz, emergency threshold for frequency change rate =0.20Hz / s; target =50%, warning threshold =20%, normalization constant =0.5; Trigger threshold =0.5、 =0.2; Basic region weight =(0.5,0.5), warning zone weight (0.2,0.8); recovery strength coefficient =0.6; Spot price reference value =40 yuan / MWh, standard deviation =100 yuan / MWh; AGC normalization constant =5MW; Feedback duration threshold =5min; Control cycle =1s. The above values are exemplary parameters and can be adjusted in actual projects according to specific circumstances such as grid guidelines, energy storage power station design, and market rules.
[0201] Compared to methods without SOC recovery and traditional dead-zone fixed-rate recovery methods, the method of this invention is expected to proactively initiate forced recovery before the SOC approaches the boundary under long-term unidirectional frequency deviation or continuous reverse AGC conditions. This avoids energy storage being forced to exit frequency regulation due to SOC exceeding the limit, and is expected to significantly improve the all-time availability of energy storage resources. Through power reservation and emergency switching, it is expected that any SOC recovery action will not crowd out the primary frequency regulation response capability, and the availability rate of the primary frequency regulation response is expected to be significantly improved compared to the solution without recovery. By triggering SOC recovery through market signals, it is expected that the recovery process will be concentrated during low electricity price periods or AGC downward adjustment periods (charging recovery scenario), or concentrated during high electricity price periods and AGC upward adjustment periods (discharging recovery scenario), thereby transforming the recovery cost in traditional methods into recovery benefits.
[0202] In extreme conditions where the State of Charge (SOC) enters the warning zone and the Automatic Control Controller (AGC) continues to reverse, the method of this invention proactively feeds back the limited capacity to the upper-level scheduling system through a scheduling feedback mechanism. This ensures that subsequent AGC command allocations do not run in the opposite direction to the SOC recovery, thereby improving the sustainable availability of energy storage resources without violating mandatory AGC commands. This fallback mechanism is not present in existing methods.
[0203] The method described in this invention is not limited to the above embodiments, and can also be applied to the following scenarios:
[0204] (a) Energy storage supporting new energy power stations: Integrate this method into the EMS of new energy power stations, and extend the market signal to the settlement price of new energy power stations or the green electricity trading price;
[0205] (b) Independent energy storage leasing: The market-based approach to SOC recovery in this method can be combined with a capacity allocation mechanism for multi-user leasing;
[0206] (c) Hybrid energy storage system: The frequency regulation power reservation and market signal triggering framework of this method can be extended to other energy storage forms that participate in primary frequency regulation.
[0207] During implementation, weight , Threshold , Parameters such as warning thresholds and emergency thresholds can be optimized offline using historical data or adaptively tuned online. Market signal scoring function. Not limited to linear normalization, nonlinear forms can also be used. The controller can be implemented by an industrial-grade embedded controller or a programmable logic controller (PLC), and the control software includes a data acquisition module, a scoring calculation module, a limiting and emergency judgment module, and a scheduling and communication module.
[0208] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the invention by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the invention should be included within the scope of the claims.
Claims
1. A method for restoring the state of charge of grid-connected energy storage based on joint chance scoring, characterized in that, include: S1. Obtain the state of charge (SOC), grid frequency, and rate of change of frequency of the energy storage system, and receive electricity market signals; S2. Compile the grid frequency difference over a historical preset period into a grid historical frequency fluctuation statistical data set; obtain the primary frequency regulation parameters of the energy storage system; based on the grid historical frequency fluctuation statistical data set and the primary frequency regulation parameters of the energy storage system, predict the maximum power to be called for primary frequency regulation in the future preset period; and calculate the upper limit of power available for state of charge recovery by combining the rated power of the energy storage system. S3. Calculate the SOC deviation direction and SOC deviation degree score based on the SOC and the preset target SOC. The market deviation direction and degree score are calculated based on the electricity market signal and the preset electricity market signal reference value; A joint opportunity score is constructed based on SOC deviation direction, SOC deviation degree score, market deviation direction, and market deviation degree score. S4. Compare the joint opportunity score with a preset trigger threshold. When the joint opportunity score is greater than the preset trigger threshold and the upper limit of power available for state of charge recovery is greater than 0, issue a state of charge recovery power command. S5. During the execution of the state-of-charge recovery power command, monitor the grid frequency deviation, frequency change rate, and primary frequency regulation response power of the energy storage system. When any preset zeroing condition is met, reset the state-of-charge recovery power to zero, so that the energy storage system returns to an operating mode with primary frequency regulation response as the sole objective. The zeroing conditions include: (a) The power grid frequency deviation exceeds the preset frequency emergency threshold; (b) The rate of frequency change exceeds the preset emergency threshold for the rate of frequency change; (c) The primary frequency regulation response power of the energy storage system is not zero.
2. The method for restoring the state of charge of grid-connected energy storage based on joint opportunity scoring according to claim 1, characterized in that, The electricity market signals include AGC dispatch command signals and spot market price signals; the primary frequency regulation parameters of the energy storage system include the droop coefficient of the energy storage system. Frequency dead zone width Rated power .
3. The method for restoring the state of charge of grid-connected energy storage based on joint opportunity scoring according to claim 2, characterized in that, The prediction of the maximum power to be activated by primary frequency modulation within a future preset time period includes: Based on historical frequency fluctuation statistics of the power grid, frequency deviation within a predetermined future time period is estimated using empirical distribution or normal fitting methods. The probability distribution is taken as the extreme value corresponding to the 95% confidence level. ; According to the extreme value droop coefficient of energy storage system Frequency dead zone width Rated power Predict the maximum power to be used by a frequency modulation within a preset time period in the future. The formula is: 。 4. The grid-connected energy storage state-of-charge recovery method based on joint opportunity scoring according to claim 3, characterized in that, The rated power calculation of the combined energy storage system can be used as the upper limit of the power for state-of-charge recovery, and the formula is: ; in, Rated power, .
5. The grid-connected energy storage state-of-charge recovery method based on joint opportunity scoring according to claim 4, characterized in that, The step of calculating the SOC deviation direction and degree score based on the State of Charge (SOC) and the preset target SOC includes: based on the SOC and the preset target SOC... The formula for calculating SOC deviation is: ; The symbol represents the SOC deviation direction, denoted as the SOC recovery direction symbol. The formula is: ; ,but , ,but , At that time, ; SOC Deviation Score Through SOC deviation absolute value With the preset normalization constant The calculation yields the following formula: 。 6. The grid-connected energy storage state-of-charge recovery method based on joint opportunity scoring according to claim 5, characterized in that, The calculation of market deviation direction and degree scores based on the electricity market signal and the preset electricity market signal reference value includes: when the electricity market signal is an AGC dispatch command signal, the energy storage power corresponding to the AGC dispatch command is used. Symbol of SOC recovery direction Calculate the deviation direction of the AGC market. AGC Market Deviation Score ; Deviating from the AGC market As a market deviation direction The AGC market is deviating from its direction. The formula is: ; when and When the numbers are the same, ,when and When the signs are different, ,when hour, ; AGC market deviation score As a score for market deviation The formula is: ; in, The energy storage power corresponding to the AGC scheduling command; This is the preset AGC command normalization constant, i.e., the preset electricity market signal reference value; When electricity market signals are available as spot market price signals, based on the current spot market price... Compared with the preset reference price That is, the preset electricity market signal reference value, to calculate the direction of price market deviation. Price market deviation score ; The price market deviates from its direction As a market deviation direction Price market deviation The formula is: ; when and hour, ;when and hour, Other cases It can be -1 or 0; Scoring the degree of price market deviation As a score for market deviation The formula is: ; in, This refers to the current spot market price; For preset reference price, This represents the standard deviation of prices during the same period. When both the AGC scheduling command signal and the spot market price signal are available, the market deviation score is the larger of the AGC market deviation score and the price market deviation score, and the market deviation direction is the market deviation direction corresponding to the larger value.
7. The grid-connected energy storage state-of-charge recovery method based on joint opportunity scoring according to claim 6, characterized in that, The formula for joint opportunity scoring is: ; in, Scoring for joint opportunities, Score the degree of market deviation. Score the degree of deviation from SOC. , The weighting coefficients and .
8. The method for restoring the state of charge of grid-connected energy storage based on joint opportunity scoring according to claim 7, characterized in that, The trigger threshold The base value is used when the deviation from the State of Charge (SOC) does not exceed the preset warning threshold. When the SOC deviation exceeds the warning threshold, the value is reduced to... ;when and At that time, a power restoration command for state of charge is issued. ;otherwise ; The formula for the state-of-charge recovery power command is: , ; The state-of-charge recovery power command The amplitude satisfies the following formula: , ; in, The preset recovery strength coefficient; It is a pre-defined monotonically non-decreasing function; When the energy storage system executes the AGC scheduling command, it restores the power command. Power corresponding to AGC command and primary frequency response power The output is superimposed at the PCS output terminal, and the total output power is: ; and .
9. The method for restoring the state of charge of grid-connected energy storage based on joint opportunity scoring according to claim 8, characterized in that, This also includes S6, and the degree of deviation from SOC. Exceeding the preset warning threshold and detecting that the AGC scheduling command direction is opposite to the SOC recovery direction The opposite continues, and the duration exceeds the preset time. At that time, the energy storage system feeds back the reduced adjustable capacity to the upper-level AGC scheduling system. ,satisfy 。 10. A grid-connected energy storage state-of-charge recovery system based on joint chance scoring, characterized in that, include: The data acquisition module is used to acquire the state of charge (SOC), grid frequency, and rate of change of the energy storage system, and to receive electricity market signals. The restored power upper limit prediction module is used to compose the power grid frequency difference over a historical preset period into historical power grid frequency fluctuation statistics. Obtain the primary frequency regulation parameters of the energy storage system. Based on the historical frequency fluctuation statistics of the power grid and the primary frequency regulation parameters of the energy storage system, predict the maximum power to be called by the primary frequency regulation in the future preset period. Combine the rated power of the energy storage system to calculate the upper limit of the power that can be used for state of charge recovery. The joint opportunity scoring module is used to calculate the SOC deviation direction and SOC deviation degree score based on the SOC and the preset target SOC. The market deviation direction and degree score are calculated based on the electricity market signal and the preset electricity market signal reference value; A joint opportunity score is constructed based on SOC deviation direction, SOC deviation degree score, market deviation direction, and market deviation degree score. The triggering module is used to compare the joint opportunity score with a preset triggering threshold. When the joint opportunity score is greater than the preset triggering threshold and the upper limit of power available for state of charge recovery is greater than 0, a state of charge recovery power command is issued. The execution and zeroing module is used to monitor the grid frequency deviation, frequency change rate, and primary frequency regulation response power of the energy storage system during the execution of the state-of-charge recovery power command. When any preset zeroing condition is met, the state-of-charge recovery power is zeroed, so that the energy storage system returns to an operating mode with primary frequency regulation response as the sole objective. The zeroing conditions include: (a) The power grid frequency deviation exceeds the preset frequency emergency threshold; (b) The rate of frequency change exceeds the preset emergency threshold for the rate of frequency change; (c) The primary frequency regulation response power of the energy storage system is not zero.