Power grid frequency modulation real-time analysis method and system

Abstract

The application relates to the technical field of power grid frequency modulation, and discloses a power grid frequency modulation real-time analysis method and system. The method comprises the following steps: a power grid frequency modulation real-time analysis method comprises the following steps: constructing a virtual inertia coefficient and a droop coefficient parallel structure to process a frequency disturbance signal; calculating a frequency change rate threshold according to load disturbance and power change of a thermal power unit; obtaining a frequency control interval identifier through double-interval comparison and discrimination; dynamically adjusting energy storage power output through a proportional-integral control algorithm; and obtaining a frequency modulation analysis result through closed-loop feedback calculation of a frequency deviation root mean square error and energy storage cycle equivalent. The application solves the problems of missing frequency change rate and frequency deviation collaborative discrimination and fixed frequency modulation strategy that cannot be adaptively adjusted in the existing energy storage frequency modulation technology.

Description

Technical Field

[0001] This application relates to the field of power grid frequency regulation technology, and in particular to a real-time analysis method and system for power grid frequency regulation. Background Technology

[0002] Traditional power grid frequency regulation technology mainly relies on the primary, secondary, and tertiary frequency regulation mechanisms of thermal power units, achieving frequency stability control through turbine governors, automatic generation control systems, and economic dispatch programs. With the large-scale grid connection of new energy sources such as wind power and photovoltaics, energy storage systems, with their advantages of fast response speed and high input-output controllability, are widely used in the field of power grid frequency regulation. Currently, energy storage participation in frequency regulation mainly adopts a comprehensive control method combining virtual droop control and virtual inertial control.

[0003] Existing energy storage frequency regulation technologies have significant shortcomings. First, frequency regulation control only focuses on steady-state frequency deviation indicators, ignoring the impact of frequency change rate on grid security. When system inertia decreases, an excessively large frequency change rate can lead to synchronous machine pole slippage, damaging the internal structure of the equipment. Second, existing frequency regulation methods employ fixed-parameter control strategies, failing to dynamically adjust the energy storage response intensity according to the degree of disturbance. This results in over-adjustment during minor disturbances, wasting energy storage lifespan, and insufficient response during severe disturbances, threatening grid stability.

[0004] The problem with existing technologies is the lack of a collaborative discrimination mechanism for frequency change rate and frequency deviation. This makes it impossible to adaptively select energy storage frequency regulation strategies based on different disturbance levels, thus hindering the ability to ensure effective frequency regulation while maintaining the economical use of the energy storage system. When the power grid experiences disturbances of varying severity, fixed control strategies cannot fully leverage the rapid response advantage of energy storage under severe disturbances, nor can they prevent unnecessary energy storage losses under minor disturbances. Summary of the Invention

[0005] This application provides a real-time analysis method and system for power grid frequency regulation, which solves the problems of lack of collaborative discrimination between frequency change rate and frequency deviation in existing energy storage frequency regulation technology, and the inability to adaptively adjust the fixed frequency regulation strategy.

[0006] In a first aspect, this application provides a real-time analysis method for power grid frequency regulation, the method comprising:

[0007] Step S1: By constructing a parallel structure of energy storage virtual inertia coefficient and energy storage virtual droop coefficient, the grid frequency disturbance signal is comprehensively processed to obtain energy storage frequency regulation power output data;

[0008] Step S2: Perform correlation calculation on the system frequency change rate based on the grid load disturbance and the power change of thermal power units to obtain the frequency change rate threshold at the disturbance time;

[0009] Step S3: Perform dual-interval comparison and discrimination processing on the frequency change rate threshold at the disturbance moment and the preset frequency change rate threshold to obtain the frequency control interval identifier;

[0010] Step S4: Based on the frequency control interval identifier, the energy storage frequency modulation power output data is dynamically adjusted using a proportional-integral control algorithm to obtain the optimized energy storage output power;

[0011] Step S5: Perform closed-loop feedback calculation on the optimized energy storage output power, the root mean square error of frequency deviation, and the energy storage cycle equivalent to obtain the real-time analysis results of power grid frequency regulation.

[0012] Secondly, this application provides a real-time analysis system for power grid frequency regulation, the power grid frequency regulation real-time analysis system comprising:

[0013] The response module is used to perform comprehensive response processing on the grid frequency disturbance signal by constructing a parallel structure of energy storage virtual inertia coefficient and energy storage virtual droop coefficient, and obtain energy storage frequency regulation power output data;

[0014] The correlation module is used to perform correlation calculations on the system frequency change rate based on grid load disturbances and thermal power unit power changes, and to obtain the frequency change rate threshold at the disturbance time.

[0015] The discrimination module is used to perform dual-interval comparison and discrimination processing on the frequency change rate threshold at the disturbance time and the preset frequency change rate threshold to obtain the frequency control interval identifier;

[0016] The adjustment module is used to dynamically adjust the energy storage frequency modulation power output data according to the frequency control interval identifier using a proportional-integral control algorithm to obtain optimized energy storage output power.

[0017] The feedback module is used to perform closed-loop feedback calculations on the optimized energy storage output power, the root mean square error of the frequency deviation, and the energy storage cycle equivalent to obtain the real-time analysis results of power grid frequency regulation.

[0018] Thirdly, a real-time power grid frequency regulation analysis device is provided, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor invokes the instructions in the memory to cause the real-time power grid frequency regulation analysis device to execute the aforementioned real-time power grid frequency regulation analysis method.

[0019] Fourthly, a computer-readable storage medium is provided, wherein instructions are stored therein, which, when executed on a computer, cause the computer to perform the aforementioned real-time analysis method for power grid frequency regulation.

[0020] The technical solution provided in this application, by constructing a parallel structure of energy storage virtual inertia coefficient and virtual droop coefficient, simultaneously responds to both frequency change rate and steady-state frequency deviation signals, overcoming the technical limitation of traditional energy storage frequency regulation relying solely on a single frequency deviation index. The rapid response capability of the energy storage virtual inertia coefficient to the frequency change rate and the steady-state adjustment function of the virtual droop coefficient to the frequency deviation are organically combined to form a comprehensive response mechanism to grid disturbances. This mechanism can rapidly suppress the frequency change trend in the early stages of a disturbance through the inertial component, and continuously correct the frequency deviation until a stable state is restored through the droop component. Based on the frequency change rate correlation calculation processing of grid load disturbance and thermal power unit power change, the load disturbance amplitude, system inertia, and thermal power unit delay characteristics are incorporated into a unified analysis framework, establishing an accurate calculation method for the frequency change rate threshold at the disturbance moment, thus solving the problem of ambiguity in the correlation between frequency change rate and system disturbance in existing technologies. The dual-interval comparison and discrimination processing establishes first and second frequency control intervals by setting a preset frequency change rate threshold, corresponding to the conventional frequency regulation mode and the enhanced frequency regulation mode, respectively. This enables intelligent switching of the energy storage frequency regulation strategy. When the disturbance is minor, the conventional mode is used to avoid over-adjustment; when the disturbance is severe, it automatically switches to the enhanced mode to ensure the frequency regulation effect. The proportional-integral control algorithm dynamically adjusts the energy storage virtual parameters according to the frequency control interval identifier. Through the real-time change of the proportional coefficient, it achieves adaptive control of the frequency regulation intensity, breaking away from the rigid mode of traditional fixed parameter control.

[0021] The closed-loop feedback computation process establishes a multi-objective comprehensive evaluation system for frequency regulation performance and energy storage lifetime through sliding time window sampling, root mean square error calculation of frequency deviation, and energy storage cycle equivalent assessment. The weighted summation algorithm organically integrates technical and economic indicators, providing a quantitative basis for the continuous optimization of energy storage frequency regulation strategies. The entire technical solution forms a complete closed-loop control system from disturbance identification, interval discrimination, dynamic adjustment to effect evaluation, significantly enhancing the adaptability of the energy storage frequency regulation system to complex power grid environments and extending the service life of energy storage equipment while ensuring frequency stability. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of one embodiment of the real-time frequency regulation analysis method for power grids in this application.

[0024] Figure 2 This is a schematic diagram of the frequency change rate numerical calibration process in an embodiment of this application;

[0025] Figure 3 This is a schematic diagram of one embodiment of the real-time frequency regulation analysis system for power grids in this application.

[0026] Figure 4 This is a schematic block diagram of the power grid frequency regulation real-time analysis device in an embodiment of the present invention. Detailed Implementation

[0027] This application provides a method and system for real-time analysis of power grid frequency regulation. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0028] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the real-time analysis method for power grid frequency regulation in this application includes:

[0029] Step S1: By constructing a parallel structure of energy storage virtual inertia coefficient and energy storage virtual droop coefficient, the grid frequency disturbance signal is comprehensively processed to obtain energy storage frequency regulation power output data;

[0030] Step S2: Perform correlation calculation on the system frequency change rate based on the grid load disturbance and the power change of thermal power units to obtain the frequency change rate threshold at the disturbance time;

[0031] Step S3: Perform dual-interval comparison and discrimination processing on the frequency change rate threshold at the disturbance time and the preset frequency change rate threshold to obtain the frequency control interval identifier;

[0032] Step S4: Based on the frequency control interval identifier, the energy storage frequency regulation power output data is dynamically adjusted using a proportional-integral control algorithm to obtain the optimized energy storage output power;

[0033] Step S5: Perform closed-loop feedback calculation on the optimized energy storage output power, the root mean square error of frequency deviation, and the energy storage cycle equivalent to obtain the real-time analysis results of power grid frequency regulation.

[0034] It is understood that the executing entity of this application can be a real-time analysis system for power grid frequency regulation, or it can be a terminal or a server; no specific limitation is made here. This application's embodiment uses a server as an example for illustration.

[0035] Specifically, frequency disturbance signals are processed by constructing a parallel structure of energy storage virtual inertia coefficient and energy storage virtual droop coefficient. The energy storage virtual inertia coefficient characterizes the energy storage system's response to the rate of frequency change, while the energy storage virtual droop coefficient characterizes the energy storage system's response to steady-state frequency deviation. First, the grid frequency deviation signal is differentiated to obtain the rate of frequency change data. Then, the rate of frequency change data is multiplied by the energy storage virtual inertia coefficient to obtain the inertial frequency modulation component. Simultaneously, the frequency deviation signal is multiplied by the energy storage virtual droop coefficient to obtain the droop frequency modulation component. Finally, the two components are superimposed in parallel to obtain the energy storage frequency modulation power output data. This parallel structure differs from traditional single frequency modulation methods, simultaneously responding to both the rate of frequency change and frequency deviation signals.

[0036] Subsequently, the frequency change rate threshold is calculated based on grid load disturbances and thermal power unit power changes. First, the load disturbance amplitude is identified to obtain the disturbance amplitude data. Then, this data is combined with the generator rotational inertia parameters to calculate the equivalent system inertia value. Next, the energy storage virtual inertia coefficient is added to the equivalent system inertia value to obtain the enhanced system total inertia parameter. Since thermal power units exhibit frequency regulation delay characteristics at the moment of disturbance, the power change of the thermal power units is constrained to zero, forming a power balance state during the inertial response phase. At this point, the load disturbance amplitude data is divided by the enhanced system total inertia parameter to obtain the instantaneous frequency change rate value during the disturbance. After numerical calibration, the frequency change rate threshold at the disturbance moment is formed. This calculation process solves the problem of unclear correlation between frequency change rate and load disturbance in existing technologies.

[0037] The dual-interval comparison and discrimination process sets a benchmark threshold for the rate of frequency change and compares the absolute value of the threshold at the time of disturbance with the benchmark threshold. When the rate of frequency change is less than the benchmark threshold, it is classified into the first frequency control interval, corresponding to the normal frequency regulation mode; when the rate of frequency change is greater than or equal to the benchmark threshold, it is classified into the second frequency control interval, corresponding to the enhanced frequency regulation mode. The first frequency control interval ranges from zero to the benchmark threshold for the rate of frequency change, corresponding to the frequency regulation intensity parameters in the stable zone and the normal response mode of the energy storage system. The second frequency control interval ranges from the benchmark threshold to the upper limit of the system frequency change rate, corresponding to the frequency regulation intensity parameters in the enhanced adjustment zone and the enhanced response mode of the energy storage system. This dual-interval division mechanism differs from the traditional fixed threshold judgment, dynamically adjusting the frequency regulation strategy according to the system state.

[0038] The proportional-integral (PI) control algorithm dynamically adjusts based on the frequency control interval identifier. The interval identifier is input into the PPI controller to select the corresponding proportional coefficient parameter, which dynamically changes according to different interval identifiers. Then, the proportional coefficient parameter is used to adjust the amplification factor of the energy storage virtual inertia coefficient and the energy storage virtual droop coefficient, resulting in adjusted inertia and droop coefficients. These adjusted coefficients are multiplied by the energy storage frequency regulation power output data to obtain the enhanced energy storage power regulation amount. Finally, this regulation amount is superimposed on the original energy storage frequency regulation power to obtain the optimized energy storage output power. This dynamic adjustment mechanism solves the problem of insufficient adaptability caused by fixed parameters in traditional energy storage frequency regulation control.

[0039] Closed-loop feedback calculations use a sliding time window to sample the optimized energy storage output power in real time, with the window length set to sixty seconds, collecting sample data of the energy storage power output. The square root of the root mean square error of the frequency deviation is calculated based on historical grid frequency deviation data to obtain a frequency regulation performance evaluation index reflecting the level of frequency control accuracy. Simultaneously, the depth of charge and discharge of the energy storage is cumulatively calculated based on the energy storage power output sample data, yielding the energy storage cycle equivalent, which characterizes the degree of consumption of the energy storage system's lifespan. Finally, a weighted summation algorithm is used to comprehensively evaluate the frequency regulation performance evaluation index and the energy storage cycle equivalent, where the weighting coefficients are determined based on frequency regulation effectiveness and economic requirements, resulting in real-time analysis results of grid frequency regulation.

[0040] In one specific embodiment, step S1 includes:

[0041] The frequency deviation signal of the power grid is processed by differentiation to obtain the frequency change rate data;

[0042] Based on the frequency change rate data, the inertial response is calculated using the energy storage virtual inertial coefficient to obtain the inertial frequency modulation component, where the energy storage virtual inertial coefficient characterizes the response strength of the energy storage system to the frequency change rate.

[0043] Based on the grid frequency deviation signal, the droop response is calculated and processed using the energy storage virtual droop coefficient to obtain the droop frequency modulation component, where the energy storage virtual droop coefficient characterizes the response strength of the energy storage system to the steady-state frequency deviation.

[0044] The energy storage frequency modulation power output data is obtained by performing parallel superposition calculations based on the inertial frequency modulation component and the droop frequency modulation component.

[0045] Specifically, by calculating the time derivative of continuously acquired frequency data, the frequency change rate is obtained by dividing the frequency difference between adjacent time points by the time interval. The data processing of the differential operation includes digital filtering of the frequency signal to remove high-frequency noise, and then calculating the frequency change rate using either the forward differential or center differential method. The forward differential method involves subtracting the previous frequency value from the current frequency value and then dividing by the sampling time interval; the center differential method involves subtracting the previous frequency value from the next frequency value and then dividing by twice the sampling time interval. The frequency change rate data obtained from the differential operation directly reflects the dynamic trend of the power grid frequency; positive values ​​indicate a frequency increase, and negative values ​​indicate a frequency decrease. The magnitude of the value reflects the rate of frequency change.

[0046] The inertial response calculation using the virtual inertial coefficient of energy storage involves directly multiplying the frequency change rate data by the virtual inertial coefficient to obtain the inertial frequency modulation component. The virtual inertial coefficient characterizes the energy storage system's response strength to the frequency change rate, and its magnitude determines the power output amplitude of the energy storage system when the frequency changes rapidly. The data processing logic for inertial response calculation is as follows: when the frequency change rate is positive, the energy storage system outputs negative power to suppress frequency rise; when the frequency change rate is negative, the energy storage system outputs positive power to prevent frequency fall. During the calculation, the virtual inertial coefficient acts as a proportionality coefficient to convert the physical quantity of the frequency change rate into the power output of the energy storage system. The value of the inertial frequency modulation component is equal to the frequency change rate data multiplied by the virtual inertial coefficient.

[0047] The calculation of droop response using the virtual droop coefficient of energy storage involves directly multiplying the grid frequency deviation signal by the virtual droop coefficient to obtain the droop frequency modulation component. The virtual droop coefficient characterizes the response strength of the energy storage system to steady-state frequency deviations; its mechanism is similar to the droop characteristics of a traditional generator but with a faster response speed. The data processing for droop response calculation involves directly subtracting the standard frequency value from the current frequency value to obtain the frequency deviation signal, and then multiplying it by the virtual droop coefficient to obtain the droop frequency modulation component. The calculation logic is that a positive frequency deviation indicates that the frequency is too high, and the energy storage system outputs negative power to absorb excess energy; a negative frequency deviation indicates that the frequency is too low, and the energy storage system outputs positive power to supplement the energy gap. The magnitude of the droop frequency modulation component is directly proportional to the frequency deviation amplitude and the virtual droop coefficient of energy storage.

[0048] Parallel superposition processing adds the inertial frequency modulation (IF) component and the droop IF component to obtain the energy storage frequency regulation power output data. The data processing logic of superposition is the algebraic summation of the two components. When the two components have the same sign, they reinforce each other, increasing the total output power; when the two components have opposite signs, they cancel each other out, decreasing the total output power. Parallel superposition differs from series structures in that the two frequency modulation components act independently on the frequency signal simultaneously. The inertial component primarily responds to the dynamic changes in frequency, while the droop component primarily responds to the steady-state deviation of the frequency. The combination of the two forms a comprehensive response to frequency disturbances. The superimposed energy storage frequency regulation power output data includes both a rapid response to the rate of frequency change and a steady-state adjustment to the frequency deviation. For example, when a large load suddenly enters the power grid, the frequency begins to drop, and the frequency signal changes from the standard frequency towards a lower frequency. Differential processing detects this frequency drop trend and calculates a negative rate of frequency change data, reflecting the speed at which the frequency decreases. Multiplying the virtual inertia coefficient of energy storage by the negative rate of frequency change yields a positive inertial frequency modulation component, indicating that the energy storage system needs to output positive power to prevent the frequency from continuing to drop rapidly. Simultaneously, the frequency deviation signal shows that the current frequency is lower than the standard frequency. Multiplying the virtual droop coefficient of energy storage by the negative frequency deviation yields a positive droop frequency modulation component, indicating that the energy storage system needs to output positive power to correct the frequency deviation. These two positive components are added in parallel to obtain a larger positive energy storage frequency modulation power output data. Based on this data, the energy storage system injects power into the grid to compensate for the power gap caused by increased load, thereby suppressing frequency decline and gradually restoring the frequency to near the standard value.

[0049] In one specific embodiment, step S2 includes:

[0050] Disturbance amplitude identification processing is performed on the power grid load disturbance to obtain load disturbance amplitude data;

[0051] The total inertia of the system is calculated based on the load disturbance amplitude data and the generator rotational inertia parameters to obtain the equivalent system inertia value.

[0052] The virtual inertia coefficient of the energy storage system is accumulated with the inertia value of the equivalent system to obtain the total inertia parameter of the enhanced system.

[0053] Based on the frequency regulation delay characteristics of thermal power units at the moment of disturbance, the power change of thermal power units is subjected to zero-value constraint processing to obtain the power balance state during the inertial response stage.

[0054] Based on the power balance state during the inertial response stage, the load disturbance amplitude data and the total inertia parameter of the enhanced system are divided to obtain the instantaneous frequency change rate of the disturbance.

[0055] The instantaneous frequency change rate of the disturbance is numerically calibrated to obtain the frequency change rate threshold at the disturbance moment.

[0056] Specifically, the disturbance amplitude identification and processing for power grid load disturbances involves real-time monitoring of the power grid power balance. When the load suddenly increases or decreases, the absolute value of the power deficit or surplus is calculated to obtain the load disturbance amplitude data. The data logic of the identification and processing involves calculating the difference between the total load power at the current moment and the total load power at the previous moment. The absolute value of the difference is the load disturbance amplitude; a positive value indicates a load increase, and a negative value indicates a load decrease. Disturbance amplitude identification also includes filtering the disturbance signal to remove measurement noise and small-amplitude random fluctuations, retaining only the true load change signal. The identification algorithm uses a threshold judgment method; only power changes exceeding a preset threshold are considered valid disturbances.

[0057] The calculation of total system inertia combines load disturbance amplitude data with generator rotational inertia parameters, and obtains the equivalent system inertia value through the principle of inertia superposition. Generator rotational inertia parameters include the rotational inertia constants of each operating generator. The inertia contribution of each generator is obtained by multiplying its inertia constant by its rated capacity. The sum of the inertia contributions of all operating generators yields the total system rotational inertia. The calculation must consider the operating status of each generator; only generators connected to the grid contribute to the system inertia, while generators that are shut down or under maintenance contribute zero inertia. The equivalent system inertia value reflects the grid's ability to resist frequency changes; a larger value indicates stronger system inertia and slower frequency changes.

[0058] The cumulative calculation adds the virtual inertia coefficient of the energy storage system to the equivalent system inertia value to obtain the enhanced total inertia parameter of the system. The virtual inertia coefficient of the energy storage system is the equivalent inertia value that simulates the inertial response of a traditional generator through the energy storage control strategy, and its value is determined by the power capacity and control parameters of the energy storage system. The data processing logic of the cumulative calculation is to directly add the two inertia values. The addition of the virtual inertia coefficient of the energy storage system is equivalent to adding extra inertial support to the original system, and the enhanced total inertia parameter of the system is greater than the original equivalent system inertia value. The accumulated inertia parameter includes the physical inertia of the traditional generator and the virtual inertia of the energy storage system, which together contribute to the grid frequency stability control.

[0059] Zero-value constraint processing, based on the frequency regulation delay characteristics of thermal power units, sets the power change of the thermal power units to zero at the instant a disturbance occurs. The frequency regulation delay characteristic of thermal power units refers to the time delay between the occurrence of a frequency deviation and the start of the unit's response, including measurement delay, control system processing delay, and actuator action delay; the total delay time is typically several seconds. The data processing logic of zero-value constraint is that at the initial moment of the disturbance, due to the frequency regulation delay characteristic, the thermal power unit has not yet started operating, and its power output remains unchanged from the pre-disturbance value; the power change is zero. Constraint processing creates a power balance state during the inertial response phase, at which point the power imbalance in the grid is entirely borne by the system inertia, with no other frequency regulation resources involved.

[0060] The division calculation process uses the load disturbance amplitude data as the dividend and the enhanced system total inertia parameter as the divisor to perform a division operation, yielding the instantaneous frequency change rate of the disturbance. The physical meaning of this calculation is based on the power balance principle: during the inertial response phase, the power imbalance caused by the load disturbance is absorbed or released by the system inertia. The ratio of the power imbalance to the system inertia determines the rate of frequency change. The data logic of the division calculation is that the larger the load disturbance amplitude or the smaller the system inertia, the larger the resulting frequency change rate, indicating a faster frequency change, and vice versa. The unit of the calculation result is frequency change per second, directly reflecting the drastic degree of frequency change at the instant of the disturbance.

[0061] Numerical calibration involves comparing the instantaneous frequency change rate of a disturbance with the power grid's operational safety standards to establish a reasonable numerical range and obtain a threshold for the frequency change rate at the disturbance moment. Calibration includes numerical verification and range limiting. Verification checks whether the calculated results are within a reasonable physical range, while range limiting truncates values ​​exceeding the safe operating range. The data processing logic for calibration sets safe upper and lower limits for the frequency change rate based on the power grid equipment's capacity and operational experience; values ​​exceeding these limits are treated as boundary values. The calibrated threshold serves as the benchmark data for subsequent frequency control interval determination, directly influencing the selection of energy storage frequency regulation strategies.

[0062] Figure 2 This is a schematic diagram illustrating the frequency change rate numerical calibration process in an embodiment of this application. Figure 2 As shown, the frequency change rate calibration process includes three core steps: numerical verification, range limitation, and truncation. In the figure, the red solid line represents the original instantaneous frequency change rate of the disturbance, the blue solid line represents the frequency change rate threshold after calibration, the black dashed line indicates the upper and lower limits of safe operation (±0.135 Hz / s), and the black dotted line indicates the normal operation threshold (±0.05 Hz / s). When the original calculated value exceeds the safe operating range, the calibration process automatically performs a truncation operation, constraining the excessive value to the safe boundary value, ensuring that the benchmark data for subsequent frequency control interval judgment meets the requirements of power grid safe operation. The dots and squares in the figure indicate the specific location of the truncation process, visually demonstrating how numerical calibration converts unsafe frequency change rate values ​​into threshold data that conforms to operating standards.

[0063] In one specific embodiment, step S3 includes:

[0064] The preset frequency change rate threshold is numerically set to obtain the frequency change rate benchmark threshold.

[0065] The absolute value of the frequency change rate threshold at the disturbance moment is compared with the absolute value of the frequency change rate reference threshold to obtain the frequency change rate comparison result.

[0066] Based on the comparison results of the frequency change rate, the frequency control interval is divided into two categories to obtain the first frequency control interval and the second frequency control interval. The first frequency control interval corresponds to the conventional frequency modulation mode, and the second frequency control interval corresponds to the enhanced frequency modulation mode.

[0067] Logical judgment processing is performed based on the comparison results of frequency change rate to obtain the frequency control interval identifier.

[0068] Specifically, the numerical setting process for the preset frequency change rate threshold involves determining the safe boundary value of the frequency change rate based on power grid operation safety standards and equipment tolerance, thus obtaining a benchmark threshold as the dividing point for interval division. The data processing for numerical setting includes collecting historical power grid operation data, statistically analyzing the frequency change rate range under normal operating conditions, and determining the safe upper limit of the frequency change rate by combining the frequency tolerance capability of power grid equipment and the requirements of synchronous motor sliding pole protection. The setting process also needs to consider the power grid's inertia level and load characteristics; power grids with lower inertia correspond to smaller benchmark thresholds, while power grids with larger load fluctuations correspond to larger benchmark thresholds. The value of the benchmark threshold directly affects the subsequent interval division results. A threshold that is too small will lead to oversensitivity and frequent triggering of the frequency adjustment mode, while a threshold that is too large will lead to slow response and missing the optimal frequency adjustment opportunity.

[0069] The magnitude comparison operation compares the absolute values ​​of both the frequency change rate threshold and the baseline frequency change rate threshold at the time of the disturbance. The comparison result indicates the magnitude relationship between the two values. The data processing logic for the comparison operation first calculates the absolute value of the two input values ​​to remove the influence of the sign, and then judges the magnitude of the values. The comparison result includes three states: greater than, equal to, and less than. Absolute value processing is used because the frequency change rate can be positive (indicating a frequency increase) or negative (indicating a frequency decrease), but regardless of whether it is increasing or decreasing, the absolute value of the rate of change reflects the severity of the disturbance. The comparison operation uses a digital comparator or software algorithm, outputting a Boolean comparison result: a true value indicates that the frequency change rate threshold at the time of the disturbance is greater than the baseline threshold, and a false value indicates that it is less than or equal to the baseline threshold.

[0070] The binary classification process divides the entire frequency change rate numerical domain into two non-overlapping intervals based on the comparison results, resulting in a first frequency control interval and a second frequency control interval, each corresponding to a different frequency regulation mode. The data logic of this classification process uses a frequency change rate benchmark threshold as the dividing point, splitting the numerical domain in two. The portion below the benchmark threshold is assigned to the first frequency control interval, while the portion above or equal to the benchmark threshold is assigned to the second frequency control interval. The first frequency control interval corresponds to the conventional frequency regulation mode, indicating a mild disturbance with gentle grid frequency changes, where normal response strength from energy storage is sufficient to meet the regulation requirements. The second frequency control interval corresponds to the enhanced frequency regulation mode, indicating a severe disturbance with drastic grid frequency changes, requiring stronger response support from energy storage to prevent frequency instability. This classification process solves the problem of a fixed response strength in traditional frequency regulation methods, dynamically selecting an appropriate frequency regulation strategy based on the disturbance level.

[0071] The logical judgment process performs conditional judgments based on the frequency change rate comparison results, outputting the corresponding frequency control interval identifier as the basis for subsequent dynamic adjustments. The data processing of the logical judgment involves taking the comparison result as conditional input and judging it through logic gates or conditional statements. When the comparison result is true, a second frequency control interval identifier is output, indicating that frequency modulation needs to be increased; when the comparison result is false, a first frequency control interval identifier is output, indicating that conventional frequency modulation should be used. The judgment process includes two stages: input verification and output encoding. Input verification checks the validity of the comparison result to avoid incorrect judgments, while output encoding converts the logical judgment result into a standardized interval identifier code for easy recognition by subsequent processing modules. The interval identifier is represented using binary encoding or an enumeration type, and the identifier code contains interval type and frequency modulation mode information, directly guiding the selection and adjustment of energy storage frequency modulation parameters.

[0072] In one specific embodiment, the process of performing binary classification of the frequency control interval based on the comparison results of the frequency change rate can specifically include the following steps:

[0073] The values ​​in the frequency change rate comparison results that are less than the frequency change rate benchmark threshold are classified into intervals to obtain the numerical range of the first frequency control interval, which is the interval between zero and the frequency change rate benchmark threshold.

[0074] The values ​​in the frequency change rate comparison results that are greater than or equal to the frequency change rate benchmark threshold are classified into intervals to obtain the numerical range of the second frequency control interval, which is the interval between the frequency change rate benchmark threshold and the upper limit of the system frequency change rate.

[0075] The frequency modulation intensity of the energy storage system is set according to the numerical range of the first frequency control interval to obtain the frequency modulation intensity parameters of the stable zone, where the frequency modulation intensity parameters of the stable zone correspond to the normal response mode of the energy storage system.

[0076] The frequency regulation intensity of the energy storage system is enhanced by setting the enhancement level based on the numerical range of the second frequency control interval, resulting in the frequency regulation intensity parameter of the enhanced adjustment zone, which corresponds to the enhanced response mode of the energy storage system.

[0077] Specifically, the interval classification process for frequencies less than the benchmark threshold involves collecting and classifying all values ​​in the frequency change rate comparison results that are less than the benchmark threshold through numerical filtering, thus defining the numerical range of the first frequency control interval. The data logic of the classification process is to traverse all frequency change rate values, compare each value with the benchmark threshold, and mark values ​​that meet the condition as members of the first interval, while excluding values ​​that do not meet the condition. The numerical range of the first frequency control interval is a continuous interval from zero to the end of the frequency change rate benchmark threshold, including zero but excluding the benchmark threshold itself. Zero, as the lower limit of the interval, indicates that the frequency is completely stable without any change, while the benchmark threshold, as the upper limit of the interval, indicates that the frequency change is within a safe and controllable range. The interval classification process also includes boundary value processing and validity verification. Boundary value processing determines the openness or closure of the interval endpoints, and validity verification checks whether the classification results meet the logical requirements.

[0078] The interval classification process, where values ​​are greater than or equal to the frequency change rate benchmark threshold, collects and categorizes the comparison results to form the numerical range definition of the second frequency control interval. The data processing for classification involves filtering all input values ​​based on whether they are greater than or equal to the benchmark threshold. Values ​​meeting the criteria are assigned to the second interval set, while those not meeting the criteria are automatically excluded. The numerical range of the second frequency control interval extends from the frequency change rate benchmark threshold to the upper limit of the system frequency change rate, including the benchmark threshold but with the upper limit as an open boundary. The upper limit of the system frequency change rate is the maximum permissible value of the frequency change rate determined based on the protection settings and safe operation requirements of power grid equipment; exceeding this value poses a risk of power grid instability. Interval classification requires handling boundary overlap issues. The benchmark threshold serves as both the upper boundary of the first interval and the lower boundary of the second interval. Open and closed interval definitions are used to avoid numerical assignment conflicts.

[0079] The standard setting process, based on the numerical range characteristics of the first frequency control interval, sets the frequency regulation intensity of the energy storage system to a response level suitable for normal disturbances, thus obtaining the frequency regulation intensity parameters for maintaining stability. The data logic of the setting process selects relatively small frequency regulation intensity parameter values ​​based on the slow frequency change characteristic of the first interval, avoiding over-adjustment that could cause frequency oscillations. The frequency regulation intensity parameters for maintaining stability include the baseline values ​​of the energy storage virtual inertia coefficient and virtual droop coefficient. These parameter values ​​are determined based on the frequency regulation requirements under normal operating conditions, satisfying frequency stability requirements without placing an excessive burden on the energy storage system. The parameter setting process includes three stages: historical data analysis, simulation verification, and field testing. Historical data analysis statistically analyzes the frequency regulation effect under normal disturbances, simulation verifies the rationality of the parameter settings, and field testing verifies the performance of the parameters in actual operation. The normal response mode of the energy storage system corresponds to relatively mild power output changes, with moderate response speed and limited adjustment range.

[0080] The enhanced response level setting process, based on the numerical range characteristics of the second frequency control interval, sets the frequency regulation intensity of the energy storage system to a response level suitable for severe disturbances, resulting in enhanced adjustment zone frequency regulation intensity parameters. The data processing logic selects a relatively large frequency regulation intensity parameter value based on the drastic frequency changes corresponding to the second interval, ensuring frequency stability even under severe disturbances. The enhanced adjustment zone frequency regulation intensity parameter is an amplified version of the stable zone parameter, multiplied by an amplification factor to obtain a larger virtual inertia coefficient and virtual droop coefficient. The amplification factor is determined based on the severity of the disturbance and frequency regulation requirements analysis; a factor that is too small cannot effectively cope with severe disturbances, while a factor that is too large will lead to over-regulation and energy storage lifespan loss. The enhanced response mode of the energy storage system corresponds to a more aggressive power output strategy, with faster response speed and larger adjustment range, providing strong frequency support in a short time.

[0081] In one specific embodiment, step S4 includes:

[0082] The frequency control interval identifier is input into the proportional-integral controller for control parameter selection processing to obtain the proportional coefficient parameter, which dynamically changes according to the different values ​​of the frequency control interval identifier.

[0083] The virtual inertia coefficient and virtual droop coefficient of energy storage are adjusted by magnification based on the proportional coefficient parameter to obtain the adjusted inertia coefficient and adjusted droop coefficient.

[0084] The adjusted inertia coefficient and adjusted droop coefficient are multiplied with the energy storage frequency regulation power output data to obtain the enhanced energy storage power regulation amount.

[0085] The original energy storage frequency regulation power is superimposed and corrected based on the enhanced energy storage power regulation to obtain the optimized energy storage output power.

[0086] Specifically, the control parameter selection process for the proportional-integral (PI) controller, based on the frequency control interval identifier, selects the corresponding proportional coefficient parameter value according to different interval identifiers through table lookup or conditional judgment. The PI controller internally has a parameter mapping table; the first frequency control interval identifier corresponds to a smaller proportional coefficient parameter value, and the second frequency control interval identifier corresponds to a larger proportional coefficient parameter value. The data logic of the selection process uses the input interval identifier as an index or conditional judgment basis, determining the output proportional coefficient parameter by looking up the mapping relationship or executing conditional branch statements. The dynamic change characteristic of the proportional coefficient parameter means that the parameter value is not fixed but automatically adjusts according to the real-time changes of the frequency control interval identifier. When the interval identifier switches from the first interval to the second interval, the proportional coefficient parameter correspondingly changes from a small value to a large value, and vice versa. The parameter selection process also includes numerical validity checks and smooth transition processing. Validity checks ensure that the selected parameter value is within a reasonable range, and smooth transition processing avoids control oscillations caused by sudden parameter changes.

[0087] The amplification factor adjustment process uses the proportional coefficient parameter as the amplification factor, multiplying it by the virtual inertia coefficient and virtual droop coefficient of the energy storage, respectively, to obtain the amplified adjusted inertia coefficient and adjusted droop coefficient. The data calculation process for adjustment involves direct numerical multiplication; the adjusted inertia coefficient equals the original virtual inertia coefficient multiplied by the proportional coefficient parameter, and the adjusted droop coefficient equals the original virtual droop coefficient multiplied by the proportional coefficient parameter. The physical significance of amplification factor adjustment is to enhance the response strength of energy storage frequency regulation. When the proportional coefficient parameter is greater than a unit value, the adjusted coefficient value is greater than the original coefficient value, and the energy storage frequency regulation output is enhanced. When the proportional coefficient parameter is equal to a unit value, the adjusted coefficient value remains unchanged, and the energy storage frequency regulation output maintains its original level. Amplification factor adjustment also needs to consider energy storage capacity limitations and safe operation constraints; the adjusted coefficient value must not exceed the maximum withstand capacity of the energy storage equipment to avoid overload damage.

[0088] The multiplication operation process multiplies the adjusted inertia coefficient and droop coefficient with the corresponding components in the energy storage frequency regulation power output data to obtain the enhanced energy storage power regulation amount. The operation includes two parallel multiplication calculations: one multiplying the adjusted inertia coefficient with the frequency change rate component, and the other multiplying the adjusted droop coefficient with the frequency deviation component. The data processing logic of the multiplication operation follows the basic control equations of energy storage frequency regulation. The enhanced regulation amount of the inertia component equals the adjusted inertia coefficient multiplied by the frequency change rate data, and the enhanced regulation amount of the droop component equals the adjusted droop coefficient multiplied by the frequency deviation data. The resulting enhanced energy storage power regulation amount contains the superposition of the two components, reflecting the total power output capability of energy storage in the enhanced response mode. The multiplication operation also needs to handle numerical symbols and unit conversions to ensure the correct physical meaning of the calculation results and that the numerical accuracy meets the control requirements.

[0089] The superposition correction process algebraically adds the enhanced energy storage power regulation amount to the original energy storage frequency regulation power to obtain the final optimized energy storage output power. The data logic of the correction process uses the original frequency regulation power as the base power and the enhanced power regulation amount as the correction increment; the sum of the two forms the optimized total output power. The superposition correction uses an algebraic summation method: when the enhanced power regulation amount is positive, the total output power increases; when the enhanced power regulation amount is negative, the total output power decreases; and when the enhanced power regulation amount is zero, the total output power remains unchanged. The correction process also includes power limiting and smoothing filtering. Power limiting ensures that the optimized output power does not exceed the rated capacity range of the energy storage device, and smoothing filtering reduces high-frequency jitter in the power output to protect the energy storage device. The optimized energy storage output power serves as the final command for energy storage frequency regulation control, directly controlling the power output of the energy storage converter to achieve grid frequency regulation.

[0090] In one specific embodiment, step S5 includes:

[0091] The energy storage output power input sliding time window is optimized for real-time sampling processing to obtain energy storage power output sample data;

[0092] Based on historical data of power grid frequency deviation, the root mean square error of frequency deviation is calculated by square root processing to obtain frequency regulation performance evaluation index.

[0093] Based on the energy storage power output sample data, the energy storage charge and discharge depth is cumulatively calculated to obtain the energy storage cycle equivalent, which characterizes the degree of consumption of the energy storage system's lifespan.

[0094] The frequency regulation performance evaluation index and energy storage cycle equivalent are comprehensively evaluated and processed by a weighted summation algorithm to obtain the real-time analysis results of power grid frequency regulation.

[0095] Specifically, by setting a data collection window of fixed length, optimized energy storage output power data is continuously collected to form a sample data set of energy storage power output. The sliding time window is a data caching mechanism with a fixed window length of sixty seconds. The window content is continuously updated over time; new data enters from one end of the window while old data is removed from the other, ensuring that the window always contains power data from the most recent sixty seconds. The data processing logic involves periodically collecting optimized energy storage output power data according to a preset sampling frequency, typically set to ten times per second or higher, to ensure the capture of detailed information about power changes. The sample data within the sliding window are arranged chronologically to form a time series. Each sample contains a timestamp and a corresponding power value, and the number of sample data is equal to the window length multiplied by the sampling frequency. Window sampling also includes a data preprocessing stage to remove outliers and noise interference, and to interpolate and supplement missing data to ensure the completeness and validity of the sample data.

[0096] The square root calculation process first collects historical data on grid frequency deviation, calculates the mean square value of the frequency deviation, and then performs a square root operation on the mean square value to obtain the frequency regulation performance evaluation index. Historical grid frequency deviation data refers to the sequence of differences between the actual frequency value and the standard frequency value recorded within a specific time period. The data collection time span is consistent with the sliding time window of sixty seconds. The data processing steps for calculating the root mean square error include three stages: squaring, averaging, and square rooting. First, each frequency deviation value is squared to remove the influence of the positive or negative sign. Then, the arithmetic mean of all squared values ​​is calculated to obtain the mean square value. Finally, the square root of the mean square value is performed to obtain the root mean square error. The magnitude of the frequency regulation performance evaluation index directly reflects the accuracy level of energy storage frequency regulation control. The smaller the value, the more precise the frequency control and the smaller the deviation; the larger the value, the lower the frequency control accuracy and the larger the deviation. The square root calculation also needs to consider data validity and statistical significance. When the sample size is insufficient, the calculation results lack representativeness. A minimum sample size threshold needs to be set to ensure the reliability of the calculation results.

[0097] Cumulative calculations calculate the cumulative power of energy storage charging and discharging based on sample data of energy storage power output. The energy storage cycle equivalent is obtained through power integration and capacity normalization. The depth of energy storage charge and discharge refers to the cumulative absolute value of charging and discharging power during frequency regulation, reflecting the utilization level and cycle count of the energy storage capacity. The data processing for cumulative calculation involves summing the absolute values ​​of all power samples within a sliding window. Positive power represents energy storage discharge, and negative power represents energy storage charging. The total power change is obtained by summing the absolute values. The energy storage cycle equivalent is obtained by dividing the cumulative power by twice the rated energy storage capacity. This division is because a complete charge-discharge cycle includes both charging and discharging processes, each consuming one unit of energy storage capacity. The cycle equivalent characterizes the degree of consumption of the energy storage system's lifespan. A larger value indicates more frequent energy storage cycle use and faster lifespan consumption, while a smaller value indicates relatively mild energy storage use and slower lifespan consumption. Cumulative calculations also need to consider energy storage efficiency loss and capacity decay factors, correcting the calculation results using an efficiency correction coefficient.

[0098] The weighted summation algorithm assigns different weight coefficients to the frequency regulation performance evaluation index and the energy storage cycle equivalent, and then performs a weighted summation to obtain the real-time analysis result of grid frequency regulation. The data processing logic of the weighted summation is to multiply each evaluation index by its corresponding weight coefficient, and then algebraically sum the weighted values ​​to obtain the comprehensive evaluation score. The weight coefficients reflect the importance of different evaluation indicators; a larger weight coefficient for frequency regulation performance indicates a greater emphasis on frequency control accuracy, while a larger weight coefficient for energy storage cycle equivalent indicates a greater focus on energy storage lifespan. The values ​​of the weight coefficients are usually determined using the analytic hierarchy process (AHP) or expert scoring method, and the sum of the two weight coefficients equals a unit value to ensure the normalization of the evaluation results. The comprehensive evaluation process also includes result standardization and grading. Standardization maps the evaluation score to a standard range for easy comparison and analysis, while grading classifies frequency regulation performance into different levels such as excellent, good, average, and poor based on the score. The real-time analysis result of grid frequency regulation serves as the basis for optimizing energy storage frequency regulation strategies, guiding subsequent parameter adjustments and control strategy improvements.

[0099] The above describes the real-time analysis method for power grid frequency regulation in the embodiments of this application. The following describes the real-time analysis system for power grid frequency regulation in the embodiments of this application. Please refer to [link / reference]. Figure 3 One embodiment of the power grid frequency regulation real-time analysis system in this application includes:

[0100] The response module is used to perform comprehensive response processing on the grid frequency disturbance signal by constructing a parallel structure of energy storage virtual inertia coefficient and energy storage virtual droop coefficient, and obtain energy storage frequency regulation power output data;

[0101] The correlation module is used to perform correlation calculations on the system frequency change rate based on grid load disturbances and thermal power unit power changes, and to obtain the frequency change rate threshold at the disturbance time.

[0102] The discrimination module is used to perform dual-interval comparison and discrimination processing on the frequency change rate threshold at the disturbance time and the preset frequency change rate threshold to obtain the frequency control interval identifier;

[0103] The adjustment module is used to dynamically adjust the energy storage frequency modulation power output data according to the frequency control interval identifier using a proportional-integral control algorithm to obtain optimized energy storage output power.

[0104] The feedback module is used to perform closed-loop feedback calculations on the optimized energy storage output power, the root mean square error of the frequency deviation, and the energy storage cycle equivalent to obtain the real-time analysis results of power grid frequency regulation.

[0105] above Figure 3The real-time analysis system for power grid frequency regulation in this embodiment of the invention is described in detail from the perspective of modular functional entities. The real-time analysis device for power grid frequency regulation in this embodiment of the invention is described in detail from the perspective of hardware processing.

[0106] Reference Figure 4 This invention also provides a real-time power grid frequency regulation analysis device, which can be a server, and its internal structure can be as follows: Figure 4 As shown, the real-time power grid frequency regulation analysis device includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor in this computer design provides computing and control capabilities. The memory of the real-time power grid frequency regulation analysis device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the real-time power grid frequency regulation analysis device is used to store the data corresponding to this embodiment. The network interface of the real-time power grid frequency regulation analysis device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.

[0107] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the power grid frequency regulation real-time analysis device to which the present invention is applied.

[0108] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the real-time analysis method for power grid frequency regulation.

[0109] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0110] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a real-time power grid frequency regulation analysis device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0111] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A real-time analysis method for power grid frequency regulation, characterized in that, The method includes: Step S1: By constructing a parallel structure of energy storage virtual inertia coefficient and energy storage virtual droop coefficient, the grid frequency disturbance signal is comprehensively processed to obtain energy storage frequency regulation power output data; Step S2: Perform correlation calculation on the system frequency change rate based on the grid load disturbance and the power change of thermal power units to obtain the frequency change rate threshold at the disturbance time; Step S3: Perform dual-interval comparison and discrimination processing on the frequency change rate threshold at the disturbance moment and the preset frequency change rate threshold to obtain the frequency control interval identifier; Step S4: Based on the frequency control interval identifier, the energy storage frequency modulation power output data is dynamically adjusted using a proportional-integral control algorithm to obtain the optimized energy storage output power; Step S5: Perform closed-loop feedback calculation on the optimized energy storage output power, the root mean square error of frequency deviation, and the energy storage cycle equivalent to obtain the real-time analysis results of power grid frequency regulation.

2. The real-time analysis method for power grid frequency regulation according to claim 1, characterized in that, Step S1 includes: The frequency deviation signal of the power grid is processed by differentiation to obtain the frequency change rate data; Based on the frequency change rate data, inertial response calculation is performed using the energy storage virtual inertial coefficient to obtain the inertial frequency modulation component, wherein the energy storage virtual inertial coefficient characterizes the response strength of the energy storage system to the frequency change rate. The droop response is calculated and processed using the virtual droop coefficient of the energy storage system based on the grid frequency deviation signal to obtain the droop frequency modulation component, wherein the virtual droop coefficient of the energy storage system characterizes the response strength of the energy storage system to the steady-state frequency deviation. The energy storage frequency modulation power output data is obtained by performing parallel superposition operations on the inertial frequency modulation component and the droop frequency modulation component.

3. The real-time analysis method for power grid frequency regulation according to claim 1, characterized in that, Step S2 includes: The power grid load disturbance is processed to identify the disturbance amplitude, and load disturbance amplitude data is obtained. Based on the load disturbance amplitude data and generator rotational inertia parameters, the total system inertia is calculated to obtain the equivalent system inertia value. The energy storage virtual inertia coefficient and the equivalent system inertia value are accumulated to obtain the total inertia parameter of the enhanced system. Based on the frequency modulation delay characteristics of the thermal power unit at the moment of disturbance, the power change of the thermal power unit is subjected to zero-value constraint processing to obtain the power balance state in the inertial response stage. Based on the power balance state of the inertial response stage, the load disturbance amplitude data and the total inertia parameter of the enhanced system are divided to obtain the instantaneous frequency change rate of the disturbance. The instantaneous frequency change rate of the disturbance is calibrated to obtain the frequency change rate threshold at the disturbance moment.

4. The real-time analysis method for power grid frequency regulation according to claim 1, characterized in that, Step S3 includes: The preset frequency change rate threshold is numerically set to obtain the frequency change rate benchmark threshold. The absolute value of the frequency change rate threshold at the disturbance moment is compared with the absolute value of the frequency change rate reference threshold to obtain the frequency change rate comparison result. The frequency control interval is divided into two categories based on the frequency change rate comparison results to obtain a first frequency control interval and a second frequency control interval, wherein the first frequency control interval corresponds to the conventional frequency modulation mode and the second frequency control interval corresponds to the enhanced frequency modulation mode. Logical judgment processing is performed based on the frequency change rate comparison results to obtain the frequency control interval identifier.

5. The real-time analysis method for power grid frequency regulation according to claim 4, characterized in that, The frequency control interval is divided into two categories based on the frequency change rate comparison results to obtain a first frequency control interval and a second frequency control interval, wherein the first frequency control interval corresponds to the conventional frequency modulation mode and the second frequency control interval corresponds to the enhanced frequency modulation mode, including: The values ​​in the frequency change rate comparison results that are less than the frequency change rate benchmark threshold are classified into intervals to obtain the numerical range of the first frequency control interval, wherein the numerical range of the first frequency control interval is the interval between zero and the frequency change rate benchmark threshold. The values ​​in the frequency change rate comparison results that are greater than or equal to the frequency change rate benchmark threshold are classified into intervals to obtain the numerical range of the second frequency control interval, wherein the numerical range of the second frequency control interval is the interval between the frequency change rate benchmark threshold and the upper limit of the system frequency change rate. The frequency modulation intensity of the energy storage system is set to a conventional level based on the numerical range of the first frequency control interval to obtain the frequency modulation intensity parameter for maintaining stability, wherein the frequency modulation intensity parameter for maintaining stability corresponds to the normal response mode of the energy storage system. The frequency modulation intensity of the energy storage system is enhanced by setting the enhancement level according to the numerical range of the second frequency control interval, thereby obtaining the frequency modulation intensity parameter of the enhanced adjustment zone, wherein the frequency modulation intensity parameter of the enhanced adjustment zone corresponds to the enhanced response mode of the energy storage system.

6. The real-time analysis method for power grid frequency regulation according to claim 1, characterized in that, Step S4 includes: The frequency control interval identifier is input into the proportional-integral controller for control parameter selection processing to obtain the proportional coefficient parameter, wherein the proportional coefficient parameter changes dynamically according to different values ​​of the frequency control interval identifier; The energy storage virtual inertia coefficient and the energy storage virtual droop coefficient are adjusted by magnification based on the proportional coefficient parameter to obtain the adjusted inertia coefficient and the adjusted droop coefficient. The adjusted inertia coefficient and the adjusted droop coefficient are multiplied with the energy storage frequency regulation power output data to obtain the enhanced energy storage power regulation amount. The original energy storage frequency modulation power is superimposed and corrected based on the enhanced energy storage power regulation amount to obtain the optimized energy storage output power.

7. The real-time analysis method for power grid frequency regulation according to claim 1, characterized in that, Step S5 includes: The optimized energy storage output power input sliding time window is sampled in real time to obtain energy storage power output sample data. The root mean square error of the frequency deviation is calculated by square root processing based on historical data of power grid frequency deviation to obtain frequency regulation performance evaluation index. Based on the energy storage power output sample data, the energy storage charge and discharge depth is cumulatively calculated to obtain the energy storage cycle equivalent, which characterizes the degree of energy storage system lifespan consumption. The frequency regulation performance evaluation index and the energy storage cycle equivalent are comprehensively evaluated and processed by a weighted summation algorithm to obtain the real-time analysis results of the power grid frequency regulation.

8. A real-time analysis system for power grid frequency regulation, characterized in that, For implementing the real-time analysis method for power grid frequency regulation as described in any one of claims 1-7, the real-time analysis system for power grid frequency regulation includes: The response module is used to perform comprehensive response processing on the grid frequency disturbance signal by constructing a parallel structure of energy storage virtual inertia coefficient and energy storage virtual droop coefficient, and obtain energy storage frequency regulation power output data; The correlation module is used to perform correlation calculations on the system frequency change rate based on grid load disturbances and thermal power unit power changes, and to obtain the frequency change rate threshold at the disturbance time. The discrimination module is used to perform dual-interval comparison and discrimination processing on the frequency change rate threshold at the disturbance time and the preset frequency change rate threshold to obtain the frequency control interval identifier; The adjustment module is used to dynamically adjust the energy storage frequency modulation power output data according to the frequency control interval identifier using a proportional-integral control algorithm to obtain optimized energy storage output power. The feedback module is used to perform closed-loop feedback calculations on the optimized energy storage output power, the root mean square error of the frequency deviation, and the energy storage cycle equivalent to obtain the real-time analysis results of power grid frequency regulation.

9. A real-time analysis device for power grid frequency regulation, characterized in that, The method includes a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the real-time analysis method for power grid frequency regulation as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is run by the processor, it causes the processor to perform the real-time analysis method for power grid frequency regulation as described in any one of claims 1 to 7.

Images