Power grid frequency safety optimization method, device, equipment and medium

Abstract

The application discloses a power grid frequency safety optimization method, device, equipment and medium. By quantifying physical inertia and virtual inertia uniformly and realizing dynamic partitioning based on electrical distance, accurate identification and quantitative evaluation of inertia void in the power grid are realized. Further, the minimum inertia demand of the partition is introduced into the power market clearing model as a rigid safety constraint, allowing new energy to participate in market bidding as a virtual inertia service provider, and through the optimization of joint decision of power and inertia resources, the resource allocation scheme with the lowest total social cost is automatically generated on the premise of ensuring frequency safety, realizing the coordinated optimization of safety and economy. Finally, the marginal price of inertia resources is extracted from the optimization model, and a node inertia scarcity index is further constructed, which is converted into an intuitive economic signal and frequency safety warning, thereby systematically improving the stability and power supply reliability of power grid operation.

Description

Technical Field

[0001] This invention relates to the field of power system operation control technology, and in particular to a method, apparatus, equipment and medium for power grid frequency security optimization. Background Technology

[0002] With the acceleration of the global energy transition and the deepening of the "dual carbon" goal, the installed capacity and power generation share of clean energy, represented by wind power and photovoltaics, in the power system continue to rise. The structure and operation mode of the power system are undergoing profound reconstruction, gradually evolving from the traditional "source follows load" unidirectional balance mode to a new type of power system with multi-dimensional interaction and deep synergy of "source, grid, load and storage".

[0003] Currently, to ensure grid frequency security, system operators are often forced to adopt planned measures, such as mandating the online operation of high-cost synchronous generator units through mandatory start-up, or implementing measures like wind and solar power curtailment to maintain system inertia levels. These practices not only hinder the efficient absorption of clean energy but also significantly increase system operating costs, resulting in resource waste and economic losses. Summary of the Invention

[0004] In view of this, the present invention provides a power grid frequency security optimization method, apparatus, electronic device and medium to solve the technical problems of high operating cost and insufficient economic efficiency of traditional power grid frequency security assurance methods.

[0005] Firstly, a method for optimizing power grid frequency security is provided, the method comprising: The first operating parameters of each generator set in the power system and the second operating parameters of the power grid are obtained. The first operating parameters include the physical parameters of the synchronous generator sets at each node, the operating parameters of the new energy generator sets, and the control parameters. Based on the first and second operating parameters, the equivalent virtual inertial constant of the new energy generator set is determined; Based on the second operating parameters, the power grid is divided into multiple inertia support regions by a preset clustering algorithm, and the minimum inertia requirement value for each inertia support region is determined to maintain frequency stability. Based on the minimum inertial support requirements of each inertial support region, a joint optimization clearing model is constructed with the goal of minimizing the total system operating cost. Solving the joint optimization clearing model yields the start-stop state variables and active power output variables of the synchronous generator set that meet frequency security requirements, as well as the virtual inertia service participation variables and equivalent virtual inertia constant variables of the new energy generator set. Based on the solution results of the joint optimization clearing model, an inertia scarcity index is determined to reflect the scarcity of inertia support resources in different regions, and a power grid frequency security early warning is carried out based on the inertia scarcity index.

[0006] Secondly, a power grid frequency security optimization device is provided, the device comprising: The acquisition module is used to acquire the first operating parameters of each generator set in the power system and the second operating parameters of the power grid. The first operating parameters include the physical parameters of the synchronous generator sets at each node, the operating parameters of the new energy generator sets, and the control parameters. The first determining module is used to determine the equivalent virtual inertial constant of the new energy generator set based on the first operating parameters and the second operating parameters. The second determining module is used to divide the power grid into multiple inertia support regions based on the second operating parameters and through a preset clustering algorithm, and to determine the minimum inertia requirement value required to maintain frequency stability for each inertia support region. The model building module is used to construct a joint optimization clearing model with the goal of minimizing the total operating cost of the system, based on the minimum inertial support requirements of each inertial support region. The generation module is used to solve the joint optimization clearing model to obtain the start-stop state variables and active power output variables of the synchronous generator set that meet the frequency security requirements, as well as the virtual inertia service participation variables and equivalent virtual inertia constant variables of the new energy generator set. The third determination module is used to determine the inertia scarcity index, which reflects the scarcity of inertial support resources in different regions, based on the solution results of the joint optimization clearing model, and to conduct power grid frequency security early warning based on the inertia scarcity index.

[0007] Thirdly, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described power grid frequency security optimization method.

[0008] Fourthly, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the above-described power grid frequency security optimization method.

[0009] The aforementioned power grid frequency security optimization methods, devices, electronic equipment, and storage media achieve accurate identification and quantitative assessment of inertia voids in the power grid by unifying and quantifying physical and virtual inertia and implementing dynamic partitioning based on electrical distance. Furthermore, the minimum inertia requirement for each partition is introduced as a rigid security constraint into the electricity market clearing model, allowing renewable energy sources to participate in market bidding as virtual inertia service providers. Through optimized joint decision-making on power and inertia resources, and under the premise of ensuring frequency security, the resource allocation scheme with the lowest total social cost is automatically generated based on market competition mechanisms, achieving optimal synergy between security and economy. Finally, the marginal price of inertia resources is extracted from the optimization model, and a node inertia scarcity index is further constructed, transforming it into an intuitive economic signal and frequency security early warning, thereby systematically improving the stability of power grid operation and the reliability of power supply. Attached Figure Description

[0010] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a flowchart illustrating a power grid frequency security optimization method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the power grid frequency security optimization device in one embodiment of the present invention. Detailed Implementation

[0011] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the accompanying drawings in the present invention are only for illustrative and descriptive purposes and are not intended to limit the scope of protection of the present invention.

[0012] Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this invention illustrate operations implemented according to some embodiments of the invention. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or performed simultaneously. Moreover, those skilled in the art, guided by the content of this invention, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.

[0013] Furthermore, the embodiments described herein are merely some, not all, of the embodiments of the invention. The components of the embodiments of the invention described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0014] It should be noted that the term "comprising" will be used in the embodiments of the present invention to indicate the presence of a feature subsequently declared, but does not exclude the addition of other features. It should also be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0015] With the advancement of global energy transition and the "dual carbon" goal, the penetration rate of new energy sources such as wind power and photovoltaics in the power system continues to increase. The power system is gradually shifting from the traditional "source follows load" operation mode to a new power system architecture that deeply integrates and coordinates "source, grid, load and storage".

[0016] However, the large-scale and high-proportion integration of new energy sources has also brought new and severe challenges to the safe and stable operation of the power grid frequency, specifically in the following aspects: 1. The system's physical inertia level has decreased significantly, weakening its frequency disturbance immunity: Traditional power systems mainly rely on the physical rotational inertia of synchronous generator sets (such as thermal and hydropower) to maintain frequency stability. The kinetic energy stored in their rotors can be released instantaneously when the system experiences a power deficit, effectively suppressing frequency fluctuations. However, wind and solar power units are typically connected to the grid through power electronic converters, and their operation is physically decoupled from the grid frequency, lacking the inertial response characteristics of traditional synchronous generator sets. As new energy sources gradually replace synchronous generator sets, the system's equivalent inertia level has decreased significantly, leading to a significant increase in the grid's frequency change rate when subjected to disturbances. This makes it extremely easy to trigger low-frequency load shedding protection and even cause grid frequency collapse.

[0017] 2. Uneven spatial and temporal distribution of inertia resources renders traditional "total inertia assessment" methods ineffective: Existing frequency security assessment methods are mostly based on "infinite inertia for a single unit" or "equivalent inertia of the system" models, assuming synchronous frequency changes across the entire network and focusing only on the overall inertia level of the system. In reality, due to the limitations of power grid topology and transmission capacity, new energy power generation exhibits a significant regional concentration characteristic, with large wind and solar power bases mostly located at the end of the grid, resulting in significant spatial differences in inertia support capacity across different regions. When local disturbances occur, if the electrical connections in that region are weak and local inertia is insufficient, even if the overall inertia of the entire network is sufficient, the region may still experience frequency instability before external support arrives, forming an "inertia hole." Existing global assessment methods struggle to identify such local risks.

[0018] 3. Lack of market mechanisms to incentivize renewable energy to provide frequency support: Currently, while technologies such as doubly-fed induction generators (DFIGs) and virtual synchronous generators can achieve "virtual inertia," the lack of corresponding value quantification and cost compensation mechanisms means that renewable energy power plants generally operate in maximum power point tracking (MPPT) mode, neither reserving reserve capacity nor participating in system frequency regulation. Existing electricity market clearing models primarily focus on energy balance, failing to incorporate "regional minimum inertia demand" into their optimization objectives. This leads dispatching agencies to often rely on administrative measures, such as forcibly activating synchronous generators or implementing wind and solar curtailment, to maintain system inertia levels in order to ensure frequency security. Such practices not only waste clean energy but also significantly increase system operating costs.

[0019] In summary, there is currently a lack of a comprehensive solution that can uniformly quantify physical and virtual inertia, accurately identify weak links in the grid's inertia space, and optimize resource allocation through market mechanisms. Based on this, this application proposes a collaborative optimization method for grid frequency security assessment and market clearing adapted to high-proportion renewable energy integration. By optimizing joint decision-making on power and inertia resources, and relying on market competition mechanisms while ensuring frequency security, it automatically generates the resource allocation scheme with the lowest total social cost, achieving optimal synergy between security and economy.

[0020] The following is a detailed description of this case, in conjunction with the relevant accompanying drawings in the instruction manual.

[0021] Please see Figure 1 This description and embodiment provide a method for optimizing power grid frequency security, specifically including the following steps: S10: Obtain the first operating parameters of each generator set in the power system and the second operating parameters of the power grid; The first operating parameters include the physical parameters of the synchronous generator sets at each node, the operating parameters of the new energy generator sets, and the control parameters.

[0022] It is understood that the executing entity of this invention can be a power grid frequency security optimization device, a terminal, or a server; no specific limitation is made here. This embodiment of the invention will be described using a server as an example.

[0023] In this step, the first operating parameters of each generator unit at each node in the power system, as well as the second operating parameters characterizing the grid connection relationship and operating status, are collected in real time through the Energy Management System (EMS) and Wide Area Measurement System (WAMS). The first operating parameters characterize the physical characteristics, real-time status, and controllability of each generator unit, specifically including: physical parameters of synchronous generator units (including the rated capacity and inherent inertia time constant of the synchronous generator unit at node i); operating parameters of new energy generator units (including the real-time active power and maximum available power of the new energy generator unit at node i); and control parameters of new energy generator units (including the virtual inertia control coefficient and droop control coefficient of the new energy generator unit at node i).

[0024] Furthermore, the second operating parameter is used to define the connection structure, physical characteristics, and safe operating boundaries of the power network, specifically including: grid topology and branch parameters: including the connection relationship of grid nodes and the resistance and reactance parameters of each branch; system-level parameters and safety thresholds: including system base capacity, system rated frequency, maximum allowable frequency change rate, and operating constraint parameters such as transmission section stability limit and node voltage safety upper and lower limits.

[0025] Alternatively, synchronous generator sets refer to traditional generators in the system, such as coal-fired, gas-fired, and hydroelectric generator sets, which can provide physical inertia; new energy generator sets refer to wind turbines, photovoltaic units, and other units in the system, which can provide virtual inertia through power electronic converters.

[0026] S20: Based on the first operating parameter and the second operating parameter, determine the equivalent virtual inertial constant of the new energy generator set.

[0027] In this step, by collecting the first and second operating parameters, the equivalent virtual inertia constant of the new energy generator unit under its current operating state can be calculated, thereby achieving a standardized quantitative characterization of the virtual inertia support capability of the new energy generator unit. This method enables the frequency support capabilities provided by traditional synchronous generator units and power electronic interface-type new energy generator units to be objectively evaluated, quantitatively compared, and systematically aggregated under a unified physical dimension (unit: "second"), thus establishing a reliable and consistent technical benchmark for subsequent grid zonal frequency security assessments and market-based ancillary service transactions.

[0028] In one embodiment of this application, a specific scheme for calculating the equivalent virtual inertia constant is provided. In S20, that is, based on the first operating parameter and the second operating parameter, the equivalent virtual inertia constant of the new energy generator set is determined, specifically including the following steps S21-S26: S21: Construct an active power response model for new energy generator sets with virtual inertia control function; The active power response model includes at least one differential control term related to the rate of change of frequency.

[0029] In this step, a standardized active power response model is defined to describe the dynamic output characteristics that the renewable energy converter should follow when sensing changes in system frequency. This model is identified and adopted from the control system architecture or technical specifications of the renewable energy generator set as its standardized control law for participating in system frequency regulation, and must contain at least one differential control term directly related to the rate of change of system frequency.

[0030] Specifically, when the system detects a frequency disturbance (i.e., the rate of frequency change df / dt ≠ 0), the additional active power output of the new energy generator set g with virtual inertia control function can be expressed as:

[0031] in, The additional active power output of the new energy generator set g; df / dt is the virtual inertia control coefficient of the new energy generator unit g; df / dt is the system frequency change rate. For the droop control coefficient of the new energy generator set g, this coefficient characterizes the unit's ability to adjust the magnitude of frequency deviation (analogous to primary frequency regulation). This represents the system frequency deviation.

[0032] Furthermore, in the active power response model This term is the key differential control term. Mathematically, this term simulates the physical inertial response of a synchronous generator (whose output change is proportional to the rotor angular acceleration), enabling the new energy generator set to make an instantaneous power response to the rate of frequency change, thereby providing crucial dynamic damping in the initial stage of disturbance and effectively suppressing frequency abrupt changes.

[0033] The above approach unifies the dynamic characterization methods of various new energy devices with rapid frequency response capabilities in principle, providing an indispensable mathematical and physical framework for transforming their adjustment capabilities into measurable, comparable, and tradable virtual inertia commodities.

[0034] S22: The theoretical inertia constant is calculated based on the virtual inertia control coefficient of the new energy generator set in the active power response model, as well as the system rated frequency and system reference capacity.

[0035] In this step, the new energy virtual inertia power response model represents a power response process, which needs to be converted into a standard inertia constant with "seconds" as the unit of measurement. Based on this, the core control parameter in the active power response model of the new energy generator set, namely the virtual inertia control coefficient, is extracted. This coefficient represents the unit's ability to suppress the rate of frequency change (analogous to physical inertia). Combining the unified benchmark required for standard power system analysis, namely the system rated frequency (usually 50Hz) and the system benchmark capacity, the theoretical inertia constant of the new energy generator set under ideal conditions is calculated through theoretical derivation and dimensional analysis.

[0036] Specifically, the formula for calculating the theoretical inertial constant is:

[0037] in, It is the theoretical inertial constant; This represents the system's baseline capacity. The control coefficient is then calculated using this formula. It is converted into a theoretical inertial constant with time dimension, realizing the conversion from control behavior parameters to system physical parameters. This allows the inertial support potential of different new energy generator sets and even traditional synchronous generator sets to be initially compared and superimposed under a unified physical dimension.

[0038] S23: Obtain the minimum reserve capacity commitment value of the new energy generator set, and based on the maximum available power, real-time active power and minimum reserve capacity commitment value of the new energy generator set, determine the power reserve adequacy factor of the new energy generator set used to provide virtual inertia support.

[0039] In this step, the maximum available power and real-time active power of the new energy generator set are obtained. The difference between the two represents the maximum power space that the unit can instantly generate under the current natural conditions (wind speed, sunlight), i.e., instantaneous reserve capacity. Combined with the unit's minimum reserve capacity commitment value (which represents the power margin that the unit promises to reserve for providing frequency regulation services), the power reserve adequacy factor is calculated. This allows the ability of the new energy generator set to provide virtual inertia to be realistically adjusted from its theoretical maximum value based on the current actual available power reserves.

[0040] Specifically, the formula for calculating the power reserve adequacy factor is as follows:

[0041] in, This refers to the power reserve margin factor. Let g be the maximum available power of the new energy generator unit at node i. Let g be the real-time active power of the new energy generator unit g at node i; Let g be the minimum reserve capacity commitment value of the new energy generator unit g at node i; This constitutes an energy reserve constraint term, used to characterize whether the unit currently has sufficient "power reserve" to support the effective release of its virtual inertia.

[0042] S24: Obtain the converter response delay parameters and the time constant threshold of the system inertia response of the new energy generator set, and determine the delay attenuation factor when the new energy generator set provides virtual inertia support based on the converter response delay parameters and the time constant threshold.

[0043] In this step, the converter response delay parameter is obtained. This parameter characterizes the total time delay (usually in milliseconds or seconds) required for the new energy generator set to adjust its power output from sensing a frequency disturbance in the system. It integrates the time consumed by multiple stages, including frequency measurement filtering, control algorithm calculation, communication transmission, and power device operation. Simultaneously, the time constant threshold of the system inertial response is obtained to measure the effective time window of inertial support. Based on these two parameters, the delay decay factor is calculated using an exponential decay form. The longer the delay, the smaller the effect on suppressing RoCoF, and the more severe the discount on the equivalent inertial value. This allows for the correction of the virtual inertial theoretical capability on a dynamic time scale, ensuring that the final evaluated equivalent inertial constant not only reflects the magnitude of the support capability but also accurately reflects the key engineering reality of its response speed.

[0044] Specifically, the formula for calculating the delay attenuation factor is as follows:

[0045] in, This is the delay decay factor; This refers to the converter response delay parameter; The threshold value is the time constant.

[0046] Optionally, the converter response delay parameter can be obtained from the technical specifications provided by the equipment manufacturer, field test data, or theoretical estimation based on the control architecture. The time constant threshold of the system inertial response is set by the system operating mechanism based on the overall dynamic characteristics of the power grid and safe operation standards.

[0047] S25: Multiply the power reserve adequacy factor by the delay decay factor to obtain the response confidence factor of the new energy generator set when providing virtual inertia support.

[0048] In this step, two independent correction factors characterizing power reserve and response speed are multiplied to form a comprehensive confidence index that fully reflects the effective support capability under current actual operating conditions, namely the response confidence factor. This response confidence factor is a dimensionless number between 0 and 1, used to unify the impact of power availability and time efficiency into a single scalar. By introducing this response confidence factor to characterize the reliability of the unit's inertia support, the weakening effects of "insufficient reserve capacity" and "communication and execution delays" on the inertia support effect can be explicitly considered, thereby enhancing the model's robustness, avoiding errors caused by the randomness of new energy output and the response delay of power electronic devices, and preventing discrepancies between calculated results and actual support capabilities due to directly applying ideal physical formulas.

[0049] S26: Multiply the theoretical inertial constant by the response confidence factor to obtain the equivalent virtual inertial constant of the new energy generator set.

[0050] In this step, the calculated theoretical inertia constant is multiplied by the response confidence factor to calculate the equivalent virtual inertia constant of the new energy generator set. This constant has the same dimensions as the inherent inertia time constant of a synchronous generator. The equivalent virtual inertia constant of the new energy generator set is determined by its virtual inertia control coefficient, which will serve as a core decision variable in the subsequent market optimization model, thereby achieving an effective connection between virtual inertia capacity quantification and market clearing optimization. For example, a wind turbine might be evaluated as H... vir =2.5 seconds, which means that under the current operating conditions, the frequency support effect it can provide is equivalent to that of a synchronous generator with an inertial time constant of 2.5 seconds.

[0051] Specifically, the formula for calculating the equivalent virtual inertial constant is as follows:

[0052]

[0053] in, Let g be the equivalent virtual inertial constant of the new energy generator unit g at node i; Let be the confidence factor of the response of the new energy generator unit g at node i.

[0054] By using the above method, the unit of the physical inertia constant of thermal power units and the equivalent virtual inertia constant of new energy power generation units is unified to "second", so that they can be directly compared and aggregated under the same evaluation framework.

[0055] S30: Based on the second operating parameters, the power grid is divided into multiple inertia support regions by a preset clustering algorithm, and the minimum inertia requirement value required to maintain frequency stability is determined for each inertia support region.

[0056] In this step, based on acquiring and integrating the second operating parameters of the power grid, and according to the actual electrical structure and real-time operating status of the power grid, the entire power grid is divided into multiple inertia support regions through dynamic partitioning, thereby identifying naturally formed inertia cooperative units in the power grid. Subsequently, for each divided inertia support region, based on the power deficit under the anticipated fault scenario and the system safety threshold, the minimum inertia requirement value that must be met to limit the initial frequency change rate of the disturbance within the allowable range is calculated for that region.

[0057] In one embodiment of this application, a specific power grid area division scheme is provided. In S30, based on the second operating parameters, a preset clustering algorithm is used to divide the power grid into multiple inertia support regions, and each inertia support region is determined as the minimum inertia requirement value required to maintain frequency stability. Specifically, this includes the following steps S31-S34: S31: Calculate the electrical distance between any two nodes (using adjacent nodes) based on the power grid topology parameters and equipment impedance parameters.

[0058] In this step, based on the power grid topology parameters (including the connection relationships of all buses, lines, and transformers) and the equipment impedance parameters (resistance R and reactance X) of each branch obtained from the energy management system, a node admittance matrix describing the steady-state electrical characteristics of the entire network is constructed. This matrix is ​​the basic model for power system network analysis, and its matrix elements are filled with the admittances of each branch according to the topological connection relationships. Subsequently, the node admittance matrix is ​​inverted to obtain the node impedance matrix. Each element in the node impedance matrix has a clear physical meaning: its diagonal element Z... ii The self-impedance of a node reflects the rise in its voltage when a unit current is injected from it, characterizing the node's electrical stiffness; the off-diagonal element Z ij This is called mutual impedance, which reflects the voltage change at node i when a unit current is injected from node j, and directly measures the electrical coupling strength between the two nodes. Based on this node impedance matrix, the electrical distance d between any two nodes i and j can be calculated. ij .

[0059] Specifically, the formula for calculating electrical distance is:

[0060] in, The electrical distance between node i and node j; , These are the self-impedances in the nodal impedance matrix; represents the mutual impedance in the node impedance matrix.

[0061] Optionally, self-impedance and mutual impedance can be obtained through the node admittance relationship of the power grid topology and the power flow sensitivity matrix of the power flow calculation, which is not specifically limited in this application.

[0062] S32: Based on the electrical distance between all nodes, the power grid is divided into multiple inertia support regions using a pre-defined clustering algorithm.

[0063] In this step, based on the electrical coupling relationships between nodes, the entire network is dynamically divided into several inertia support regions that are internally closely connected and externally relatively decoupled. These regions can automatically identify naturally formed, physically meaningful dynamic cooperative units within the power grid, thereby decomposing the complex issue of overall network frequency stability into a series of more manageable and assessable regional sub-problems. This lays a crucial structural identification foundation for achieving precise regional inertia control.

[0064] In one embodiment of this application, a specific inertia support region division scheme is provided. In S32, based on the electrical distance between all nodes, the power grid is divided into multiple inertia support regions using a preset clustering algorithm, specifically including the following steps S321-S326: S321: Construct an electrical distance matrix based on the electrical distances between all nodes.

[0065] In this step, the calculated electrical distance d between all node pairs is used as a basis. ij Construct an N×N symmetric matrix D, where N is the total number of power grid nodes, and the matrix elements D ij =d ij This matrix provides a complete description of the electrical affinity relationships between all nodes in the network.

[0066] S322: Determine the similarity weight between any two nodes based on the electrical distance matrix.

[0067] In this step, a Gaussian kernel function is used to transform the electrical distance, and the similarity weight w between any two nodes i and j is calculated. ij :

[0068] in, This represents the variance of the electrical distance between all nodes.

[0069] S323: Construct a similarity weight matrix based on similarity weights.

[0070] In this step, the similarity weights w of all node pairs are used. ij Construct an N×N symmetric similarity weight matrix W, where W ij =w ij This matrix defines the weighted connectivity graph of the entire network.

[0071] S324: Obtain the degree matrix and calculate the Lapses matrix based on the degree matrix and the similarity weight matrix.

[0072] In this step, the degree matrix is ​​obtained from the directed graph relationships between all nodes in the network. It is a diagonal matrix. Then, the normalized Laplacian matrix L is calculated:

[0073] S325: Calculate the eigenvectors corresponding to the first k smallest eigenvalues ​​in the Lapses matrix, and construct the feature space based on the eigenvectors.

[0074] In this step, the eigenvectors {v1, v2, ..., v3} corresponding to the first k smallest eigenvalues ​​of the Laplacian matrix L are calculated. k Arrange these k eigenvectors column-wise to form an N×k matrix. Each row of this matrix (a k-dimensional vector) represents the coordinates of the corresponding node in the original power grid within this newly constructed spectral feature space. In this feature space, nodes with close electrical connections have coordinate vectors that are geometrically close to each other.

[0075] S326: By using a pre-defined clustering algorithm, the nodes in the feature space are divided into K clusters, which serve as multiple inertia support regions.

[0076] In this step, the N points (i.e., the coordinates of the N nodes) in the aforementioned feature space are used as input, and the K-means clustering algorithm is employed to divide them into K clusters. Each cluster corresponds to an inertial support region in the power grid that is electrically interconnected and internally dynamically coordinated, denoted as {Ω1, Ω2, ..., Ω...}. K The parameter K can be preset according to actual needs, or automatically determined by criteria such as eigenvalue gaps; this application does not impose specific limitations on it.

[0077] The above method enables data-driven dynamic partitioning based entirely on the real-time electrical structure of the power grid, thus providing a scientific, objective, and reproducible basis for network structure partitioning to facilitate precise partition inertia control in the future.

[0078] S33: Obtain the power loss value generated by each inertia support region under the preset maximum power deficit fault scenario, as well as the maximum allowable frequency change rate threshold of the system.

[0079] In this step, for each inertia support region, a preset maximum power deficit fault scenario is determined, and the corresponding power loss value under this scenario is obtained, serving as the benchmark disturbance intensity for evaluating the frequency stability of that region. Furthermore, the maximum permissible frequency change rate threshold is obtained from the system's safe operation procedures. This threshold, as a global safety limit, defines the limit of frequency drop rate that the system can withstand in the initial stage of a disturbance, and is a direct safety target for calculating the required inertial buffering capacity of the calculated region. By obtaining these two key parameters, complete and explicit input conditions are provided for calculating the minimum inertia required for each region to withstand the most severe power impact and control the frequency change rate within the safe allowable range using the rotor motion equations.

[0080] Optionally, the power loss value is typically set according to the "N-1" criterion or the more stringent "N-1-1" criterion for power system security analysis. For example, it can be assumed that the most important external power supply channel to the area (such as a critical tie line or transformer group) is blocked. The active power deficit (in MW) caused by this anticipated fault, which needs to be instantaneously balanced by the remaining power sources in the area, is the power loss value, reflecting the scale of the most severe power surge the area may suffer. Further, the maximum permissible rate of frequency change threshold is set by the grid dispatching agency based on system protection configurations (especially the settings of low-frequency load shedding devices), generator withstand capabilities, and relevant technical standards, for example, 0.5 Hz / s or 1.0 Hz / s.

[0081] S34: Based on the power loss value, the system rated frequency, the maximum frequency change rate threshold, and the system reference capacity, the minimum inertia requirement value that each inertia support region must meet to maintain frequency stability is calculated.

[0082] In this step, the initial response form of the classical rotor motion equations of the power system is applied. In the instant immediately after the disturbance occurs (approximately within the first cycle), neglecting the operation of the frequency controller and the load frequency characteristics, the system frequency change rate is mainly determined by the power deficit and the total system inertia. For the designated inertia support region, to ensure its frequency safety, the frequency change rate in this region cannot exceed the maximum tolerance threshold set by the system. Therefore, the minimum inertia requirement for region k is... It can be represented as:

[0083] in, The minimum inertia requirement for region k; This represents the maximum anticipated power deficit for region k. The system's rated frequency; This is the maximum allowable rate of frequency change for the system. It is the sum of the rated capacities of all generator sets within region k.

[0084] S40: Based on the minimum inertial support requirements of each inertial support region, construct a joint optimization clearing model with the goal of minimizing the total operating cost of the system.

[0085] In this step, an optimization function is established with the goal of minimizing the total system operating cost. In addition to incorporating conventional power balance constraints, line transmission capacity constraints, and unit operating characteristic constraints into the model, the minimum inertia requirements of each region obtained from zonal assessments are integrated as rigid safety constraints, thereby constructing a joint optimization clearing model that deeply integrates safety and economy. This model ensures that the clearing scheme meets the frequency stability requirements of the entire system's spatiotemporal distribution from the source. This allows the system operator to automatically discover the most cost-effective resource combination through market competition mechanisms while strictly ensuring frequency safety. This fundamentally abandons the traditional safety-oriented model that relies on administrative orders to force startup or restrict renewable energy output, achieving a balance between system safety and operational economy.

[0086] In one embodiment of this application, a specific model construction scheme is provided. In S40, a joint optimization clearing model is constructed based on the minimum inertial support requirements of each inertial support region, with the goal of minimizing the total operating cost of the system. This specifically includes the following steps S41-S43: S41: Construct an optimization function with the objective of minimizing the total operating cost of the system; The total system operating cost includes: fuel costs and start-up and shutdown costs of the synchronous generator sets, as well as the auxiliary service costs of the new energy generator sets providing virtual inertial support.

[0087] In this step, the total system operating cost C within the scheduling cycle is determined. total The optimization function with minimization as its core objective. This total cost represents a significant expansion of the traditional range of power generation costs, and its complete structure is as follows: (1) Traditional power generation cost: including fuel cost C of all synchronous generator sets fuel (Pi,t) (usually a quadratic function of output) and start-up / shutdown cost C start (ui,t). When the system experiences frequency disturbances, in the past, for safety reasons, dispatchers would force the operation of thermal power plants (no matter how expensive) or force the curtailment of wind power (no matter how regrettable). Now, under the premise of meeting safety constraints, we can weigh the benefits of operating an additional thermal power unit against paying for new energy power plants to increase their virtual inertia parameters, transforming "safety" into an "economic" issue, thereby improving the allocation efficiency of resources across society.

[0088] (2) Virtual inertia auxiliary service cost: The service cost C for introducing new energy generator sets to provide virtual inertia support vir(Hj,tvir). This cost reflects the opportunity cost or resource depletion incurred by renewable energy power plants for reserving backup power, changing operating points, and providing rapid frequency response. In the traditional model, renewable energy is simply viewed as a troublemaker; in this model, it becomes a provider of safety services. The virtual inertia of renewable energy constructed in the previous step not only allows for the assessment of its regulation capabilities but also enables its inclusion in market mechanisms, giving it real value and reducing the resource waste caused by simply increasing the number of thermal power units in the traditional model.

[0089] Specifically, the mathematical expression of the optimization function is as follows:

[0090] in, To minimize the total system operating cost; T is the total number of scheduling periods; A collection of synchronous generator sets; A collection of new energy generator sets; Let i be the fuel cost of synchronous generator set i during time period t; The active power output of synchronous generator unit i during time period t; The start-up and shutdown cost of synchronous generator set i during time period t; This represents the start-up and shutdown status of synchronous generator set i during time period t. The value is 1 when synchronous generator set i is started during time period t, and 0 otherwise. Ancillary service costs for providing virtual inertia services to new energy generator set j during time period t; The equivalent virtual inertial constant provided by the new energy generator unit j during time period t.

[0091] Optionally, the model's decision variables include the start-stop state variables of the synchronous generator set. Active output variable And virtual inertia service participation variables and equivalent virtual inertia constant variables of new energy generator sets. By optimizing and adjusting the values ​​of these decision variables, the total operating cost of the system can be minimized while satisfying all operational and safety constraints.

[0092] Furthermore, ancillary service costs typically include equipment wear and tear costs or opportunity costs (such as power generation losses due to deviations from maximum power point tracking operating modes).

[0093] S42: Based on the minimum inertia requirement value of each inertia support region, generate regional inertia safety constraints to ensure the frequency security of the power grid; The regional inertia safety constraint is that the sum of the physical inertia provided by all synchronous generator sets and the equivalent virtual inertia provided by all new energy generator sets within each inertia support area shall not be less than the minimum inertia requirement value of the area.

[0094] In this step, the minimum inertia requirement value for each region k obtained from the partition evaluation is determined. This is transformed into a set of rigid inequality constraints:

[0095] in, Let be the total equivalent inertia at node i. Based on the calculated equivalent virtual inertial constant of the new energy source, it can be expressed as:

[0096] in, Let be the inherent inertial time constant of synchronous generator set c in node i.

[0097] For each defined inertia partition During time period t, the following must be satisfied:

[0098] in, This is an indicator variable for whether the new energy generator set j provides inertia service during time period t. If it participates in providing inertia service, the value is 1; otherwise, it is 0.

[0099] By employing the above method, and through a zonal assessment of the power grid inertia distribution, a set of refined regional security constraints is generated. This refines the control granularity from the entire network to the zonal level, enabling spatiotemporal early warning and multidimensional assessment of the entire network's inertia. If only a single network-wide inertia constraint is set, a "free-riding" phenomenon of inertia resources may occur spatially. For example, a thermal power plant in area A might provide inertia support free of charge to area B, which has weak electrical connections, without ensuring the local frequency stability of area B. By introducing the aforementioned zonal constraints, each region is required to have sufficient local inertia support capacity, thereby effectively solving the frequency stability problem of local power grids.

[0100] S43: Multiple preset power system operation constraints and regional inertia safety constraints are integrated into the optimization function framework to generate a joint optimization clearing model. Among them, the preset power system operation constraints include at least one of the following: power balance constraints, line transmission limit constraints, and unit ramping constraints.

[0101] In this step, regional inertia safety constraints and classic power market operating constraints (including power balance constraints, line power flow transmission limit constraints, unit output upper and lower limits, and ramp rate constraints) are integrated into an optimization framework aimed at minimizing the total system operating cost. This ultimately forms a complete Safety-Constrained Unit Combination (SCUC) or Safety-Constrained Economic Dispatch (SCED) model, i.e., a joint optimization clearing model. Mathematically, this model is represented as a large-scale mixed-integer linear programming problem.

[0102] By employing the above methods, based on traditional safety-constrained unit combination or safety-constrained economic dispatch models, a minimum inertia demand constraint for each zone is added, and a virtual inertia ancillary service pricing mechanism is introduced, thereby achieving joint market clearing of electricity and inertia support services. Under the premise of satisfying all physical and safety constraints, this model can automatically weigh the economics of increasing the number of high-priced thermal power units to provide physical inertia against paying fees to incentivize new energy sources to provide virtual inertia (and potentially requiring additional power generation to compensate for power losses due to reserve requirements), thus automatically generating a frequency security guarantee scheme with the lowest total social cost.

[0103] S50: Solve the joint optimization clearing model to obtain the start-stop state variables and active power output variables of the synchronous generator set that meet the frequency security requirements, as well as the virtual inertia service participation variables and equivalent virtual inertia constant variables of the new energy generator set.

[0104] In this step, a mathematical programming solver is used to solve the joint optimization clearing model. During the solution process, while simultaneously considering generation costs, virtual inertia service costs, and all safety conditions including partitioned inertia constraints, the optimization algorithm automatically searches for the global optimal solution or a high-quality feasible solution that meets engineering accuracy requirements. Subsequently, based on the model's solution results, an executable scheduling plan and market procurement scheme are generated, specifically including: a generator start-up and shutdown plan, specifying the start-up and shutdown status of each synchronous generator unit in each time period of the future scheduling cycle; a generator output plan, specifying the baseline active power output values ​​of each synchronous generator unit and new energy generator unit in each time period; and a virtual inertia auxiliary service procurement scheme, specifying the amount of virtual inertia service provided by each new energy generator unit in each time period as determined by market clearing (or the corresponding control parameter settings) and its settlement price. This scheme fully specifies who will provide inertial support services, when, and how much, not only meeting the power balance requirements but also, through market-based contracts, pre-purchasing and deploying the spatiotemporally distributed inertial resources necessary to maintain system frequency stability.

[0105] S60: Based on the solution results of the joint optimization clearing model, determine the inertia scarcity index to reflect the scarcity of inertial support resources in different regions, and conduct frequency security early warning based on the inertia scarcity index.

[0106] In this step, after obtaining the solution results of the joint optimization clearing model, the economic signals contained therein are extracted and an inertial scarcity index is formed. The inertial scarcity index of each region is compared with the preset safety threshold, and a corresponding frequency security warning signal is generated accordingly.

[0107] In this way, the complex security constraints that were originally hidden within the optimization model are transformed into clear, intuitive and operable economic signals and risk indicators, realizing a leap from implicit model conditions to explicit operational guidance, thereby supporting the power grid to achieve more proactive and precise frequency stability risk management.

[0108] In one embodiment of this application, a specific nodal inertia pricing and spatiotemporal early warning scheme is provided. In S60, based on the solution results of the joint optimization clearing model, an inertia scarcity index is determined to reflect the scarcity of inertia support resources in different regions, and a power grid frequency security early warning is performed based on the inertia scarcity index. Specifically, this includes the following steps S61-S63: S61: Based on the envelope theorem and duality theory, the Lagrange multipliers corresponding to the inertia safety constraints of each partition are extracted from the solved joint clearing optimization model and used as the shadow price of the inertia support resources of each inertia support region.

[0109] S62: Based on shadow prices and combined with the transmission congestion of power grid lines, the inertia scarcity index of each inertia support region in the power grid is calculated.

[0110] S63: Compare the inertia scarcity index with a preset safety threshold, and issue a power grid frequency safety warning based on the comparison result.

[0111] For steps S61-S63, the joint clearing optimization model is a mixed integer linear programming (MILP) model. Based on the solution results of the joint clearing optimization model, the start-up and shutdown states of the generator set are obtained. When these parameters are substituted into the model as known parameters, MILP is transformed into a continuous linear programming problem (LP).

[0112] Construct the following Lagrange function:

[0113] in, For Lagrange multipliers; This refers to the left-hand side of the partition inertia safety constraint inequality.

[0114] Based on the envelope theorem and duality theory in mathematical programming, Lagrange multipliers corresponding to the inertia safety constraints of each partition can be extracted from the solved joint optimization model. These multipliers are called shadow prices in economics, and their physical meaning is: at the optimal solution, the minimum increase in the total system cost caused by each unit increase (e.g., 1 MW·s) in the minimum inertia demand of region k. Therefore, the Lagrange multiplier directly and quantitatively reflects the marginal economic value and scarcity of inertia support resources in that region. A higher price indicates a higher marginal cost of ensuring frequency safety in that region, and a more scarce resource. When the minimum value is taken, it is relative to the variable on the right-hand side of the constraint condition. The rate of change is the shadow price of inertia services in that region, i.e.:

[0115] in, The shadow price of inertia service for partition k in time period t reflects the marginal economic cost incurred by the region in maintaining frequency security.

[0116] Furthermore, while regional shadow prices reflect the overall scarcity of resources, they do not consider the impact of grid congestion on the effective allocation of resources. Therefore, corrections are needed by incorporating real-time grid power flow distribution and line congestion conditions. Based on shadow prices and considering the impact of line congestion, the inertia scarcity index of node n in time period t is defined as:

[0117] in, Let n be the inertia scarcity index for node n in time period t. The shadow price of partition k; This represents the congestion correction component for node n during time period t. The value of this component depends on the node's location within the transmission congestion section (e.g., sender or receiver) and the severity of the congestion. If node n is located at the receiver end of a severely congested section, the actual availability of its inertia resources is limited by transmission capacity. A value greater than 0 increases the scarcity index of the node. This index comprehensively reflects the dual impact of whether resources are scarce or not, and whether resources can be delivered, thus achieving a refined spatiotemporal measurement of inertia scarcity.

[0118] Furthermore, the calculated inertia scarcity index of each region or node is compared with a preset safety threshold, and a graded frequency safety early warning signal is generated based on the comparison results to provide early warning of power grid frequency safety.

[0119] By employing the above methods, the complex dual variables hidden within the optimization model are transformed into clear, intuitive, and actionable economic signals (shadow prices) and risk signals (scarcity index and early warning). This not only provides a pricing basis for the settlement of virtual inertia auxiliary services, but more importantly, it establishes a closed-loop feedback mechanism from market clearing results to the identification of weak links in the power grid, and then to proactive risk prevention and control. This achieves a leap from "implicit model constraints" to "explicit operational guidance," supporting the power grid to achieve more proactive and precise frequency stability risk management.

[0120] In one embodiment of this application, a specific frequency security early warning scheme is provided. In S63, the inertia scarcity index is compared with a preset security threshold, and a power grid frequency security early warning is given based on the comparison result. The scheme specifically includes the following steps: For any inertia support region, if the inertia scarcity index is higher than the preset safety threshold, an early warning signal is generated and sent indicating that the inertia support region is severely lacking and has a high frequency of safety risks. If the inertia scarcity index is equal to 0, then the inertia resources in the inertia support area are considered to be abundant.

[0121] In this embodiment, the calculated inertia scarcity index for each region is compared with a preset safety threshold. For any inertia support region, if its inertia scarcity index is higher than the set high-risk threshold, it indicates that the region's inertia resources are in a state of extreme shortage, and the system is using resources with extremely high marginal costs (such as cutting off some new energy sources or starting peak thermal power) to ensure frequency security. At this time, a high-level warning should be triggered, and it is recommended to restrict the access of new fluctuating power sources (such as wind power and photovoltaics) in the region during operation, or to prioritize the configuration of flexible support resources such as energy storage and synchronous condensers in the region in the grid planning. If the inertia scarcity index of the region is close to or equal to zero, it indicates that the region's inertia resources are abundant, the supply and demand relationship is relaxed, new energy sources can maintain maximum power point tracking operation, and the system is in a frequency-safe state.

[0122] As can be seen, the above scheme achieves accurate identification and quantitative assessment of inertia voids in the power grid by unifying and quantifying physical and virtual inertia and implementing dynamic partitioning based on electrical distance. Furthermore, by introducing the minimum inertia requirement for each partition as a rigid security constraint into the electricity market clearing model, renewable energy sources are allowed to participate in market bidding as virtual inertia service providers. Through optimized joint decision-making on power and inertia resources, and under the premise of ensuring frequency security, the scheme automatically generates the resource allocation plan with the lowest total social cost based on market competition, achieving optimal synergy between security and economy. Finally, the marginal price of inertia resources is extracted from the optimization model, and a node inertia scarcity index is further constructed, transforming it into an intuitive economic signal and frequency security early warning, thereby systematically improving the stability of power grid operation and the reliability of power supply.

[0123] In one embodiment, a power grid frequency security optimization device is provided, which corresponds one-to-one with the power grid frequency security optimization method described in the above embodiments. For example... Figure 2 As shown, the power grid frequency security optimization device 100 includes: an acquisition module 101, a first determination module 102, a second determination module 103, a model construction module 104, a generation module 105, and a third determination module 106. Detailed descriptions of each functional module are as follows: The acquisition module 101 is used to acquire the first operating parameters of each generator set in the power system and the second operating parameters of the power grid. The first operating parameters include the physical parameters of the synchronous generator sets at each node, the operating parameters of the new energy generator sets, and the control parameters. The first determining module 102 is used to determine the equivalent virtual inertial constant of the new energy generator set based on the first operating parameters and the second operating parameters; The second determining module 103 is used to divide the power grid into multiple inertia support regions based on the second operating parameters and through a preset clustering algorithm, and to determine the minimum inertia requirement value required to maintain frequency stability for each inertia support region. The model building module 104 is used to build a joint optimization clearing model with the goal of minimizing the total operating cost of the system, based on the minimum inertial support requirements of each inertial support region. The generation module 105 is used to solve the joint optimization clearing model to obtain the start-stop state variables and active power output variables of the synchronous generator set that meet the frequency security requirements, as well as the virtual inertia service participation variables and equivalent virtual inertia constant variables of the new energy generator set. The third determination module 106 is used to determine the inertia scarcity index, which reflects the scarcity of inertia support resources in different regions, based on the solution results of the joint optimization clearing model, and to conduct power grid frequency security early warning based on the inertia scarcity index.

[0124] In one embodiment, the first determining module 102 is specifically used for: Construct an active power response model for a new energy generator set with virtual inertia control function, wherein the active power response model contains at least one differential control term related to the rate of frequency change; Based on the virtual inertia control coefficient of the new energy generator set in the active power response model, as well as the system rated frequency and system reference capacity, the theoretical inertia constant is calculated. Obtain the minimum reserve capacity commitment value of the new energy generator set, and determine the power reserve adequacy factor of the new energy generator set to provide virtual inertia support based on the maximum available power, real-time active power and minimum reserve capacity commitment value of the new energy generator set. The converter response delay parameters and the time constant threshold of the system inertia response of the new energy generator set are obtained. Based on the converter response delay parameters and the time constant threshold, the delay attenuation factor when the new energy generator set provides virtual inertia support is determined. Multiplying the power reserve adequacy factor by the delay decay factor yields the response confidence factor of the new energy generator set when providing virtual inertia support; Multiplying the theoretical inertial constant by the response confidence factor yields the equivalent virtual inertial constant of the new energy generator set.

[0125] In one embodiment, the second determining module 103 is specifically used for: Calculate the electrical distance between any two nodes based on power grid topology parameters and equipment impedance parameters; Based on the electrical distance between all nodes, the power grid is divided into multiple inertia support regions using a pre-defined clustering algorithm; Obtain the power loss value generated by each inertia support region under the preset maximum power deficit fault scenario, as well as the maximum allowable frequency change rate threshold of the system; Based on the power loss value, the system rated frequency, the maximum frequency change rate threshold, and the system reference capacity, the minimum inertia requirement value that each inertia support region must meet to maintain frequency stability is calculated.

[0126] In one embodiment, the second determining module 103 is further configured to: Construct an electrical distance matrix based on the electrical distances between all nodes; Based on the electrical distance matrix, determine the similarity weight between any two nodes; Construct a similarity weight matrix based on similarity weights; Obtain the degree matrix, and calculate the Lapses matrix based on the degree matrix and the similarity weight matrix; Calculate the eigenvectors corresponding to the first k smallest eigenvalues ​​in the Lapses matrix, and construct the feature space based on the eigenvectors; By using a pre-defined clustering algorithm, the nodes in the feature space are divided into K clusters, which serve as multiple inertia support regions.

[0127] In one embodiment, the model building module 104 is specifically used for: Construct an optimization function with the objective of minimizing the total system operating cost, which includes: the fuel cost and start-up and shutdown cost of the synchronous generator set, as well as the auxiliary service cost of the new energy generator set providing virtual inertial support; Based on the minimum inertia requirement value of each inertia support region, regional inertia safety constraints are generated to ensure the frequency security of the power grid. The regional inertia safety constraints are: the sum of the physical inertia provided by all synchronous generator sets and the equivalent virtual inertia provided by all new energy generator sets in each inertia support region is not less than the minimum inertia requirement value of the region. Multiple preset power system operation constraints and regional inertia safety constraints are integrated into the optimization function framework to generate a joint optimization clearing model. The multiple preset power system operation constraints include at least one of the following: power balance constraints, line transmission limit constraints, and unit ramping constraints.

[0128] In one embodiment, the third determining module 106 is specifically used for: Based on the envelope theorem and duality theory, the Lagrange multipliers corresponding to the inertia safety constraints of each partition are extracted from the solved joint clearing optimization model and used as the shadow price of the inertia support resources of each inertia support region. Based on shadow prices and combined with the transmission congestion of power grid lines, the inertia scarcity index of each inertia support region in the power grid is calculated. The inertia scarcity index is compared with a preset safety threshold, and a power grid frequency safety warning is issued based on the comparison results.

[0129] In one embodiment, the third determining module 106 is further configured to: For any inertia support region, if the inertia scarcity index is higher than the preset safety threshold, an early warning signal is generated and sent indicating that the inertia support region is severely lacking and has a high frequency of safety risks. If the inertia scarcity index is equal to 0, then the inertia resources in the inertia support area are considered to be abundant.

[0130] This invention provides a power grid frequency security optimization device 100. By unifying and quantifying physical and virtual inertia and implementing dynamic partitioning based on electrical distance, it achieves accurate identification and quantitative assessment of inertia voids in the power grid. Furthermore, the minimum inertia requirement for each partition is introduced as a rigid security constraint into the electricity market clearing model, allowing renewable energy sources to participate in market bidding as virtual inertia service providers. Through optimized joint decision-making on power and inertia resources, and relying on market competition mechanisms, the device automatically generates the resource allocation scheme with the lowest total social cost while ensuring frequency security, achieving optimal synergy between security and economy. Finally, the marginal price of inertia resources is extracted from the optimization model, and a node inertia scarcity index is further constructed, transforming it into an intuitive economic signal and frequency security early warning, thereby systematically improving the stability of power grid operation and the reliability of power supply.

[0131] Specific limitations regarding the power grid frequency security optimization device can be found in the limitations of the power grid frequency security optimization method described above, and will not be repeated here. Each module in the aforementioned power grid frequency security optimization device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in the electronic device, or stored in the memory of the electronic device in software form, so that the processor can call and execute the corresponding operations of each module.

[0132] In one embodiment, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described power grid frequency security optimization method.

[0133] In one embodiment, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the above-described power grid frequency security optimization method.

[0134] It should be noted that the functions or steps that can be implemented by the computer-readable storage medium or electronic device described above can be referred to the relevant descriptions on the server side and client side in the foregoing method embodiments. To avoid repetition, they will not be described one by one here.

[0135] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0136] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0137] The above-described 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, and should all be included within the protection scope of the present invention.

Claims

1. A method for optimizing power grid frequency security, characterized in that, include: The first operating parameters of each generator set in the power system and the second operating parameters of the power grid are obtained. The first operating parameters include the physical parameters of the synchronous generator sets at each node, the operating parameters of the new energy generator sets, and the control parameters. Based on the first operating parameters and the second operating parameters, the equivalent virtual inertial constant of the new energy generator set is determined; Based on the second operating parameters, the power grid is divided into multiple inertia support regions by a preset clustering algorithm, and each inertia support region is determined as the minimum inertia requirement value required to maintain frequency stability. Based on the minimum inertial support requirements of each inertial support region, a joint optimization clearing model is constructed with the goal of minimizing the total system operating cost. Solving the joint optimization clearing model yields the start-stop state variables and active power output variables of the synchronous generator set that meet frequency security requirements, as well as the virtual inertia service participation variables and equivalent virtual inertia constant variables of the new energy generator set. Based on the solution results of the joint optimization clearing model, an inertia scarcity index is determined to reflect the scarcity of inertia support resources in different regions, and a power grid frequency security early warning is carried out based on the inertia scarcity index.

2. The power grid frequency security optimization method according to claim 1, characterized in that, The operating parameters of the new energy generator set include maximum available power and real-time active power. The second operating parameter includes the system rated frequency and system reference capacity. The step of determining the equivalent virtual inertial constant of the new energy generator set based on the first and second operating parameters specifically includes: Construct an active power response model for a new energy generator set with virtual inertia control function, wherein the active power response model contains at least one differential control term related to the rate of frequency change; Based on the virtual inertia control coefficient of the new energy generator set in the active power response model, as well as the rated frequency of the system and the reference capacity of the system, the theoretical inertia constant is calculated. Obtain the minimum reserve capacity commitment value of the new energy generator set, and based on the maximum available power, the real-time active power and the minimum reserve capacity commitment value of the new energy generator set, determine the power reserve adequacy factor of the new energy generator set for providing virtual inertia support; The converter response delay parameter and the time constant threshold of the system inertia response of the new energy generator set are obtained, and the delay attenuation factor when the new energy generator set provides virtual inertia support is determined based on the converter response delay parameter and the time constant threshold. Multiply the power reserve adequacy factor by the delay attenuation factor to obtain the response confidence factor of the new energy generator set when providing virtual inertia support; Multiplying the theoretical inertia constant by the response confidence factor yields the equivalent virtual inertia constant of the new energy generator set.

3. The power grid frequency security optimization method according to claim 1, characterized in that, The second operating parameters also include grid topology parameters and equipment impedance parameters. The step of dividing the grid into multiple inertia support regions based on the second operating parameters using a preset clustering algorithm, and determining the minimum inertia requirement value for each inertia support region to maintain frequency stability, specifically includes: Based on the power grid topology parameters and the equipment impedance parameters, calculate the electrical distance between any two nodes; Based on the electrical distance between all nodes, the power grid is divided into multiple inertia support regions using a pre-defined clustering algorithm; Obtain the power loss value generated by each inertia support region under the preset maximum power deficit fault scenario, as well as the maximum allowable frequency change rate threshold of the system; Based on the power loss value, the system rated frequency, the maximum frequency change rate threshold, and the system reference capacity, the minimum inertia requirement value that each inertia support region must meet to maintain frequency stability is calculated.

4. The power grid frequency security optimization method according to claim 3, characterized in that, The step of dividing the power grid into multiple inertia support regions based on the electrical distance between all nodes using a preset clustering algorithm specifically includes: Construct an electrical distance matrix based on the electrical distances between all nodes; Based on the electrical distance matrix, the similarity weight between any two nodes is determined; Based on the aforementioned similarity weights, a similarity weight matrix is ​​constructed; Obtain the degree matrix, and calculate the Lapses matrix based on the degree matrix and the similarity weight matrix; Calculate the eigenvectors corresponding to the first k smallest eigenvalues ​​in the Lapsian matrix, and construct a feature space based on the eigenvectors; By using a pre-defined clustering algorithm, the nodes in the feature space are divided into K clusters, which serve as multiple inertia support regions.

5. The power grid frequency security optimization method according to claim 1, characterized in that, The steps of constructing a joint optimization clearing model with the objective of minimizing the total system operating cost, based on the minimum inertial support requirements of each inertial support region, specifically include: An optimization function is constructed with the goal of minimizing the total system operating cost, which includes: the fuel cost and start-up and shutdown cost of the synchronous generator set, as well as the auxiliary service cost of the new energy generator set providing virtual inertial support. Based on the minimum inertia requirement value of each inertia support region, regional inertia safety constraints are generated to ensure the frequency security of the power grid. The regional inertia safety constraints are as follows: the sum of the physical inertia provided by all synchronous generator sets and the equivalent virtual inertia provided by all new energy generator sets in each inertia support region is not less than the minimum inertia requirement value of that region. Multiple preset power system operation constraints and regional inertia safety constraints are integrated into the optimization function framework to generate the joint optimization clearing model. The multiple preset power system operation constraints include at least one of the following: power balance constraints, line transmission limit constraints, and unit ramping constraints.

6. The power grid frequency security optimization method according to claim 1, characterized in that, The steps of determining an inertia scarcity index based on the solution results of the joint optimization clearing model, reflecting the scarcity of inertia support resources in different regions, and conducting power grid frequency security early warning based on the inertia scarcity index, specifically include: Based on the envelope theorem and duality theory, the Lagrange multipliers corresponding to the inertia safety constraints of each partition are extracted from the solved joint clearing optimization model and used as the shadow price of the inertia support resources of each inertia support region. Based on the shadow price and combined with the transmission congestion of the power grid lines, the inertia scarcity index of each inertia support region in the power grid is calculated. The inertia scarcity index is compared with a preset safety threshold, and a power grid frequency safety warning is issued based on the comparison result.

7. The power grid frequency security optimization method according to claim 6, characterized in that, The step of comparing the inertia scarcity index with a preset safety threshold and issuing a power grid frequency security warning based on the comparison result specifically includes: For any inertia support region, if the inertia scarcity index is higher than the preset safety threshold, a warning signal is generated and sent indicating that the inertia support region is severely lacking and has a high frequency of safety risks. If the inertia scarcity index is equal to 0, then the inertia resources in the inertia support region are considered to be sufficient.

8. A power grid frequency security optimization device, characterized in that, include: The acquisition module is used to acquire the first operating parameters of each generator set in the power system and the second operating parameters of the power grid. The first operating parameters include the physical parameters of the synchronous generator sets at each node, the operating parameters of the new energy generator sets, and the control parameters. The first determining module is used to determine the equivalent virtual inertial constant of the new energy generator set based on the first operating parameters and the second operating parameters; The second determining module is used to divide the power grid into multiple inertia support regions based on the second operating parameters and through a preset clustering algorithm, and to determine the minimum inertia requirement value required to maintain frequency stability for each inertia support region. The model building module is used to construct a joint optimization clearing model with the goal of minimizing the total operating cost of the system, based on the minimum inertial support requirements of each inertial support region. The generation module is used to solve the joint optimization clearing model to obtain the start-stop state variables and active power output variables of the synchronous generator set that meet the frequency security requirements, as well as the virtual inertia service participation variables and equivalent virtual inertia constant variables of the new energy generator set. The third determining module is used to determine the inertia scarcity index, which reflects the scarcity of inertia support resources in different regions, based on the solution results of the joint optimization clearing model, and to conduct power grid frequency security early warning based on the inertia scarcity index.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the power grid frequency security optimization method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the power grid frequency security optimization method as described in any one of claims 1 to 7.

Images