A grid space-time frequency modeling and control method and system based on an inverter type power supply

Abstract

This invention discloses a method and system for spatiotemporal frequency modeling and control of power grids based on inverter-type power supplies, relating to the fields of power engineering and control automation technology. The invention constructs a dynamically balanced inverter model, injects active power into the generator side, regulates the local frequency of the generator side, linearizes the injected frequency and angle to obtain the Jacobian matrix, simplifies it to construct a state-space model, obtains the initial frequency change rate after grid disturbance based on the state-space model, and confirms the design inertia by combining it with a preset frequency change rate limit. The design damping is confirmed based on the steady-state frequency deviation during grid steady-state operation and a preset safety threshold. The grid operation is coordinated by adjusting the design inertia and design damping parameters. This invention solves the problem of spatiotemporal frequency control in low-inertia power grids dominated by inverters, achieves accurate characterization of the dynamic frequency of grid operation, actively suppresses frequency fluctuations during grid disturbances, ensures stable grid operation, and possesses strong engineering practicality.

Description

Technical Field

[0001] This invention relates to the field of power engineering and control automation technology, and in particular to a method and system for spatiotemporal frequency modeling and control of power grids based on inverter-type power supplies. Background Technology

[0002] With the large-scale grid connection of new energy sources such as photovoltaics and wind power, the power system is gradually transforming into a form dominated by inverter-based resources (IBRs), while the proportion of traditional synchronous generators continues to decline. This transformation leads to a significant loss of the system's inherent mechanical inertia, and the differences in inverter control strategies cause the system to exhibit significant heterogeneity, thus triggering prominent spatiotemporal frequency dynamic problems. After a disturbance occurs, the frequency response of different nodes exhibits significant spatial differences and temporal evolution characteristics, seriously threatening system stability and the reliability of protection devices.

[0003] Currently, given the frequent spatiotemporal frequency dynamics of the power grid, power system frequency analysis and control primarily rely on traditional models. These models mainly include analysis methods based on the Center of Inertia (COI) frequency and the Synchronous Reference Frame (SFR) model. These models are constructed under the assumption of uniform system inertia distribution, focusing only on global frequency deviations. Essentially, they are based on the traditional synchronous generator-dominated power grid design approach. Their core logic is to simplify system dynamics by treating the entire network frequency as a unified variable for analysis and control, but they fail to consider the coupling effects between inverter control characteristics and the grid topology.

[0004] Therefore, based on the current state of power grid operation, it can be confirmed that existing technologies have significant limitations in low-inertia power grids dominated by IBR: Insufficient model adaptability: Traditional models ignore the differentiated dynamic characteristics of the two mainstream inverter types, GFM (grid-connected) and GFL (grid-following), and do not integrate grid topology information, making it impossible to accurately capture the spatiotemporal distribution of local frequencies after disturbances, resulting in a large deviation between the analysis results and the actual system.

[0005] Limited analytical depth: Existing research on spatiotemporal frequency phenomena is mostly qualitative, attributing them only to general factors such as control heterogeneity or weak grid coupling. There is a lack of clear models for quantitative analysis of disturbance propagation mechanisms, which makes it difficult to support the design of precise control strategies.

[0006] Poor control performance: Existing control schemes are not optimized for spatiotemporal frequency deviations. They only achieve frequency regulation through single parameter adjustment or local control, which cannot solve problems such as imbalance of the rate of change of frequency (RoCoF) at different nodes and excessive peak deviation. They are difficult to meet the system safety operation requirements under the high penetration rate of IBR.

[0007] Existing technologies have not broken through the "global homogenization" analysis framework and have failed to fully consider the coupling effect of IBR control characteristics, power grid topology, and heterogeneity of inertia and damping distribution. As a result, they cannot match the dynamic behavior of low-inertia power grids and cannot provide effective spatiotemporal frequency control solutions. Summary of the Invention

[0008] In view of this, the present invention provides a grid spatiotemporal frequency modeling and control method based on inverter-type power supply. The purpose is to realize the coupling effect among IBR control characteristics, grid topology, and heterogeneity of inertia and damping distribution, capture the spatiotemporal evolution law of local frequency in the grid dominated by inverter-type power supply, eliminate spatiotemporal frequency deviation in the operation of low-inertia grid, avoid safety hazards such as spatial imbalance and excessive peak deviation, and ensure the stable operation of low-inertia grid.

[0009] To achieve the above objectives, the present invention adopts the following technical solution: A method for spatiotemporal frequency modeling and control of power grids based on inverter-type power supplies includes: S1. Inverter model construction steps: Adjust the actual output power of the grid-connected inverter to be the same as the preset output power to construct a grid-connected inverter model with dynamic frequency balance; Set up grid terminals and inject deviation into the dynamic response power of the grid terminals through droop control to construct a grid-connected inverter model with dynamic frequency balance; Confirm the grid-connected inverter model and the grid-connected inverter model as the inverter model. S2. Local frequency model construction steps: Based on the inverter model, inject active power into the generator side, regulate the local frequency of the generator side, and construct the local frequency model of the generator side. S3. Matrix acquisition steps: Based on the generator-side local frequency model, the active power injected into the system nodes on the generator side and the phase angle are linearized to obtain the Jacobian matrix; S4. State-space model construction steps: Divide the system nodes into generator nodes and load nodes, divide the Jacobian matrix into blocks and eliminate load node variables to obtain a simplified Jacobian matrix, and construct the state-space model based on the simplified Jacobian matrix and the generator-side local frequency model. S5. Design inertia acquisition steps: Obtain the initial values ​​of frequency and angular deviation at the instant of power grid disturbance, and input them into the state space model to obtain the initial frequency change rate. Obtain the design inertia according to the preset frequency change rate limit and the initial frequency change rate. S6. Steps for obtaining design damping: Obtain the steady-state frequency deviation during steady-state operation of the power grid, and obtain the design damping based on the steady-state frequency deviation and the preset safety threshold; the design damping is allocated according to the design inertia ratio. S7. Coordination and stabilization steps: Adjust the parameters of grid-connected inverters and grid-reverse inverters according to the design inertia and design damping to coordinate grid operation.

[0010] Optionally, in the above method, in S1, the deviation is injected into the dynamic response power of the grid terminals through droop control to construct a grid-type inverter model with dynamic frequency balance, specifically including: The active power droop control with low-pass filtering has the following transfer function: ; In the formula, This refers to the frequency deviation at the inverter output port. Current power and rated power P e deviation, R For droop gain, The time constant of the low-pass filter. s For the Laplace operator.

[0011] Optionally, the above method can be further derived to obtain the state-space form of a swing-like equation by further deriving the transfer function: ; in, for x rate of change, x for , P For power setting value, M For equivalent inertia, D Let be the damping constant, in the above formula M Equivalent inertia of grid-connected inverters M GFM = / R , D Damping constant of grid-connected inverter D GFM = 1 / R .

[0012] The above method, optionally, involves the following formula in S2: ; Where, Δ δ G Δ represents the rotor angle deviation of the generator. ω G For generator frequency deviation, , These represent the rates of change of rotor angular deviation and frequency deviation, respectively, Δ P G Injecting incremental power M For equivalent inertia, D Let be the damping constant. ω 0 = 2π f 0, f 0 represents the system's rated frequency.

[0013] The above method, optionally, involves the following formula in S3: ; in, H Let Δ be the Jacobian matrix. P Injection amount of active power for system nodes ， Δ δ Angle difference before and after injecting active power into the system nodes ， element H ij Reflecting nodes i and j The electrical coupling strength is expressed as: ; In the formula ，V , δ These represent the voltage magnitude and phase angle of the system nodes, respectively. g、b These are the conductance and susceptance elements of the nodal admittance matrix.

[0014] The above method, optionally, includes S4 specifically: The simplified formula for calculating the Jacobian matrix is: ; in, G Indicates the generator node. L Indicates the load node. H red To simplify the Jacobian matrix, H red = H GG - ¹ H GL H LL - ¹ H LG , HL = H GL H LL - ¹, H GG This is the coupling submatrix between generator nodes. H GL This is the coupling submatrix between the generator node and the load node. H LL This is the coupling submatrix between load nodes. H LG Δ is the coupling submatrix between the load node and the generator node. P L This is a disturbance to the active power on the load side. The state-space model is as follows: .

[0015] The above method, optionally, involves the following formula in S5: ; in, For load disturbances at nodes i The equivalent transitivity, M i For design inertia, the design inertia must satisfy: ; in, α This is a preset limit for the rate of change of frequency.

[0016] In the above method, optionally, in S6, the stable frequency deviation is: ; in, D i To design damping, the damping must satisfy: ; in, Preset safety threshold A photovoltaic-storage-load coordinated power supply recovery control system for substations, applied to any of the above-mentioned grid spatiotemporal frequency modeling and control methods based on inverter-type power sources, includes an inverter model construction module, a local frequency model construction module, a matrix acquisition module, a state space model construction module, a design inertia acquisition module, a design damping acquisition module, and a coordinated stability module connected in sequence. The inverter model building module is used to adjust the actual output power of the grid-connected inverter to be the same as the preset output power, and build a grid-connected inverter model with dynamic frequency balance; it sets up grid terminals and injects deviation into the dynamic response power of the grid terminals through droop control to build a grid-connected inverter model with dynamic frequency balance; the grid-connected inverter model and the grid-connected inverter model are confirmed as inverter models. The local frequency model construction module is used to inject active power into the generator side based on the inverter model, regulate the local frequency of the generator side, and construct the local frequency model of the generator side. Matrix acquisition module: Based on the generator-side local frequency model, the active power and phase angle injected into the system nodes on the generator side are linearized to obtain the Jacobian matrix; The state-space model construction module is used to divide the system nodes into generator nodes and load nodes, divide the Jacobian matrix into blocks and eliminate load node variables, and construct a state-space model based on the generator-side local frequency model. The inertia acquisition module is designed to acquire the initial values ​​of frequency and angular deviation at the moment of power grid disturbance, and input them into the state space model to obtain the initial frequency change rate. The design inertia is then acquired based on the preset frequency change rate limit and the initial frequency change rate. The design damping acquisition module is used to acquire the steady-state frequency deviation during steady-state operation of the power grid, and to acquire the design damping based on the steady-state frequency deviation and a preset safety threshold; the design damping is allocated according to the design inertia ratio. The coordination and stabilization module is used to adjust the parameters of grid-connected and grid-reverse inverters based on design inertia and design damping, thereby coordinating grid operation.

[0017] As can be seen from the above technical solutions, compared with the prior art, the grid spatiotemporal frequency modeling and control method and system based on inverter-type power supply provided by the present invention has the following beneficial effects: (1) Improved the spatiotemporal frequency control of the power grid: The present invention solves the spatiotemporal frequency control problem of IBR-dominated low-inertia power grid through the integrated solution of "IBR dynamic modeling + topology coupling modeling + inertia-damping coordinated suppression", which improves the control of the power grid and enables the power grid to maintain stable operation under frequent disturbances.

[0018] (2) Improved the accuracy of spatiotemporal frequency control of the power grid: This invention injects active power into the generator side, obtains the Jacobian matrix through linearization, and constructs a simplified state space model. The space model can accurately replicate the frequency change after the disturbance of the IBR-dominated low-inertia power grid, providing clear and reliable data support for the continuous monitoring and data recording of the power grid, and providing a solid theoretical foundation for subsequent power grid operation research. (3) Improved grid operation stability: This invention solves the problems of RoCoF spatial imbalance and steady-state frequency deviation by co-designing the synthetic inertia and damping of IBR, reduces the interference of IBR-dominated low-inertia grid disturbances on the grid, enables the grid to quickly recover stable operation after the disturbance, and ensures reliable power supply of IBR-dominated low-inertia grid. (4) Excellent engineering applicability: This invention addresses the problem of spatiotemporal frequency control in IBR-dominated low-inertia power grids. It can suppress power grid disturbances by adjusting the parameters during power grid operation without adding complex or large equipment. Moreover, the control effect is good and the suppression effect is obvious, which has good engineering applicability. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0020] Figure 1 This invention discloses a flowchart of a grid spatiotemporal frequency modeling and control method based on an inverter-type power supply; Figure 2 This is a schematic diagram of the grid-connected inverter control system disclosed in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of the grid-connected inverter control system disclosed in an embodiment of the present invention; Figure 4 This is a schematic diagram of the improved IEEE-9 node test system structure disclosed in an embodiment of the present invention; Figure 5 The above is a frequency variation curve diagram after the parameter optimization of each node disclosed in the embodiment of the present invention, wherein a, b, c, and d are the frequency-time variation curves of bus 5, bus 7, bus 8, and bus 9, respectively. Figure 6 This is a graph showing the steady-state deviation consistency error under different disturbance positions as disclosed in the embodiments of the present invention; Figure 7 The flowchart for modeling, tuning, and verification of a 60MW load disturbance applied to the IEEE 9-bus test system disclosed in this embodiment of the invention is as follows. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] In this application, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0023] This invention, based on the coupling effect of IBR control characteristics, grid topology, and heterogeneity of inertia and damping distribution, accurately captures the spatiotemporal evolution of local frequencies in inverter-driven power grids. Through parameter co-design, it eliminates spatiotemporal frequency deviations during the operation of low-inertia grids. (See also...) Figure 1 As shown, this invention discloses a grid spatiotemporal frequency modeling and control method based on inverter-type power supplies, comprising: S1. Inverter Model Construction Steps: Adjust the actual output power of the grid-connected inverter to be the same as the preset output power to construct a grid-connected inverter model with dynamic frequency balance; Set up grid terminals and inject deviation into the dynamic response power of the grid terminals through droop control to construct a grid-connected inverter model with dynamic frequency balance; Confirm the grid-connected inverter model and the grid-connected inverter model as the inverter model.

[0024] S2. Local frequency model construction steps: Based on the inverter model, inject active power into the generator side, regulate the local frequency of the generator side, and construct the local frequency model of the generator side.

[0025] S3. Matrix acquisition steps: Based on the generator-side local frequency model, the active power and phase angle injected into the system nodes on the generator side are linearized to obtain the Jacobian matrix.

[0026] S4. State-space model construction steps: Divide the system nodes into generator nodes and load nodes, divide the Jacobian matrix into blocks and eliminate load node variables to obtain a simplified Jacobian matrix, and construct the state-space model based on the simplified Jacobian matrix and the generator-side local frequency model.

[0027] S5. Design inertia acquisition steps: Obtain the initial values ​​of frequency and angular deviation at the instant the power grid disturbance occurs, and input them into the state space model to obtain the initial frequency change rate. Obtain the design inertia according to the preset frequency change rate limit and the initial frequency change rate.

[0028] S6. Steps for obtaining design damping: Obtain the steady-state frequency deviation during steady-state operation of the power grid, and obtain the design damping based on the steady-state frequency deviation and the preset safety threshold; the design damping is allocated according to the design inertia ratio.

[0029] S7. Coordination and stabilization steps: Adjust the parameters of grid-connected inverters and grid-reverse inverters according to the design inertia and design damping to coordinate grid operation.

[0030] Furthermore, in S1, the dynamic response power injection deviation is injected into the grid terminals through droop control to construct a grid-type inverter model with dynamic frequency balance, specifically including: The active power droop control with low-pass filtering has the following transfer function: .

[0031] In the formula, This refers to the frequency deviation at the inverter output port. Current power and rated power P e deviation, R For droop gain, The time constant of the low-pass filter. s For the Laplace operator.

[0032] Furthermore, by further deriving the transfer function, we obtain the state-space form of the swing-like equation: .

[0033] in, for x rate of change, x for , P For power setting value, M For equivalent inertia, D Let be the damping constant, in the above formula M Equivalent inertia of grid-connected inverters M GFM = / R , D Damping constant of grid-connected inverter D GFM = 1 / R .

[0034] This invention addresses the spatiotemporal frequency dynamics of low-inertia power grids dominated by IBR (Inverter-Based Resources) power supplies. It discloses an IBR dynamic modeling method (i.e., an inverter model) in S1. This method forms the basis for grid disturbance suppression and provides a theoretical foundation for subsequent parameter tuning in this invention. The IBR dynamic modeling method specifically includes: 1) Modeling of Grid-Following (GFL) Inverters. GFL inverters track the grid frequency and phase based on a phase-locked loop (PLL). Their core function is to maintain a preset active power (P0). GFL inverters provide both active and reactive power output; however, they lack active frequency regulation capability and only passively respond to grid disturbances. Therefore, their power point tracking characteristics satisfy: .

[0035] In the formula, P For power setting value, P e This represents the actual output power. (Refer to...) Figure 2 When a GFL inverter has no additional frequency control, it is equivalent to a "negative load," passively adjusting its output only according to the grid frequency. Figure 2 In this module, Power Controller is the power controller, Current Controller is the current controller, and abc / dq is the coordinate transformation module, which is the core functional unit used to realize the mutual transformation between the three-phase stationary coordinate system (abc) and the synchronous rotating coordinate system (dq).

[0036] 2) Modeling of Grid-Forming Inverter (GFM). Refer to... Figure 3 The GFM inverter autonomously establishes terminal voltage and frequency, and dynamically responds to power injection deviation through droop control, making it a core component for frequency support. The grid central control system employs an active power droop control GFM inverter with low-pass filtering. The central control system is implemented using a state-space form derived from transfer functions and swing equations; refer to the implementation process in S1 for details.

[0037] Among them, the equivalent inertia of the GFM inverter M GFM = / R With damping constant D GFM = 1 / R The quantitative relationship between control parameters and inertia and damping has been clarified, providing a theoretical basis for further parameter tuning in this invention.

[0038] IBR dynamic modeling provides a component-level foundation for the implementation of the power grid spatiotemporal frequency modeling and control method of this invention. Based on this, this invention provides a detailed description of the construction of a spatiotemporal frequency topology coupling modeling framework from a "local to global" perspective. This framework breaks through the limitation of "global homogenization" in traditional COI / SFR models, deeply coupling IBR dynamics with the power grid topology to achieve accurate quantification of spatiotemporal frequency evolution. The specific steps for constructing the spatiotemporal frequency topology coupling modeling framework are as follows: 1) Local Frequency Dynamic Modeling. For grids dominated by inverter-type power sources, the local frequency dynamic balance on the generator side is adjusted. Further, the formula involved in S2 of this invention is: .

[0039] Where, Δ δ G Δ represents the rotor angle deviation of the generator. ω G For generator frequency deviation, , These represent the rates of change of rotor angular deviation and frequency deviation, respectively, Δ P G Injecting incremental power M For equivalent inertia, D Let be the damping constant. ω 0 = 2π f 0, f 0 represents the system's rated frequency. The above formula is a specific manifestation of the dynamic balance of local frequencies on the generator side.

[0040] 2) Node power-angle linearization coupling. The active power injection at system nodes satisfies the power-angle relationship, and the required Jacobian matrix is ​​obtained after linearization. Furthermore, the formula involved in S3 is: .

[0041] in, H Let Δ be the Jacobian matrix. P Injection amount of active power for system nodes ， Δ δ Angle difference before and after injecting active power into the system nodes ， element H ij Reflecting nodes i andj The electrical coupling strength is expressed as: ; In the formula ，V , δ These represent the voltage magnitude and phase angle of the system nodes, respectively. g、b These are the conductance and susceptance elements of the nodal admittance matrix.

[0042] 3) Topologically coupled state-space modeling. Further, S4 specifically includes: The system nodes are divided into generator nodes ( G ) and load nodes ( L ),right H The matrix is ​​partitioned and load node variables are eliminated to obtain a simplified generator-side relationship. The Jacobian matrix is ​​then derived and calculated based on this simplified relationship. The formula for calculating the simplified Jacobian matrix is ​​as follows: .

[0043] in, G Indicates the generator node. L Indicates the load node. H red To simplify the Jacobian matrix, H red = H GG - ¹ H GL H LL - ¹ H LG , H L = H GL H LL - ¹, H GG This is the coupling submatrix between generator nodes. H GL This is the coupling submatrix between the generator node and the load node. H LL This is the coupling submatrix between load nodes. H LG Δ is the coupling submatrix between the load node and the generator node. P L This refers to the active power disturbance on the load side.

[0044] Combining the local frequency dynamic equations (i.e., the formulas involved in S2), a topologically coupled state-space model is finally constructed. The state-space model is as follows: .

[0045] The model is 2. n G Linear Time-Invariant (LTI) systems ( n G (Number of generator nodes), the state vector includes rotor angle deviation and frequency deviation, the input is load disturbance, and the matrix... A , B Integrates IBR parameters ( M , D ) and power grid topology ( H red , H L This information can accurately quantify the patterns of disturbance propagation.

[0046] This invention is based on the above-mentioned topologically coupled state-space model, and optimizes the configuration. M and D This approach addresses two major issues after grid disturbances: inconsistent frequency changes at different nodes (RoCoF spatial imbalance) and deviations from rated steady-state frequencies. Ultimately, stable operation of the low-inertia grid is achieved by adjusting relevant parameters of the GFL / GFM. The collaborative design of the IBR's synthetic inertia and damping resolves RoCoF spatial imbalance and excessive steady-state frequency deviations, specifically including: 1) Achieve initial RoCoF global homogenization. At the instant the grid disturbance occurs, the initial values ​​of frequency and angular deviation are both 0. Substituting these values ​​into the state-space model, the initial frequency change value is obtained. Furthermore, the formula involved in S5 is: .

[0047] in, For load disturbances at nodes i The equivalent transitivity, M i For design inertia, the design inertia must satisfy: .

[0048] in, α This is a preset limit for the rate of change of frequency.

[0049] 2) Control the steady-state frequency deviation within a safe range. After a disturbance occurs in the power grid, it gradually recovers to a steady state. In steady state (Δ... ω G =0), the system frequency tends to synchronize (Δ =Δ ω∞ 1, where 1 represents an all-1 vector. Furthermore, in S6, the stable frequency deviation is: .

[0050] in, D i To design damping, the damping must satisfy: .

[0051] in, This is a preset safety threshold.

[0052] In the specific implementation process described above for resolving RoCoF spatial imbalance and excessive steady-state frequency deviation, the obtained design inertia and design damping are respectively the nodal points. i The design inertia and design damping of the load nodes are then used, and correspondingly, the design inertia and design damping of other nodes are substituted into the values ​​of the corresponding load nodes. Equivalent transfer quantity is sufficient. To ensure parameter compatibility, damping... D i According to inertia M i Proportional allocation ( D i ∝ M i This ensures spatial matching between inertia and damping.

[0053] The present invention also discloses the following embodiments, which verify the effectiveness of the control method provided by the present invention through an improved IEEE-9 node test system built in PSCAD / EMTDC software. (Refer to...) Figure 4 Test system parameters: reference capacity 100MVA, reference voltage 345kV, including 3 IBR nodes ( G 1- G 3) Initial composite inertia M The damping values ​​are 6s, 3s, and 1s respectively. D The values ​​are 10 pu, 20 pu, and 30 pu, respectively. By applying a 60 MW load disturbance to the test system, the performance difference between the traditional COI / SFR model and the scheme of this invention is compared to verify the feasibility of the control method of this invention.

[0054] Traditional COI / SFR models can only output the global average frequency deviation and cannot identify local frequency risks. However, the topology-coupled state-space model in this embodiment of the invention can accurately quantify disturbance propagation patterns. This state-space model integrates IBR control characteristics and grid topology information to construct a state-space equation containing a Jacobian matrix, which clearly defines the impact of inter-node electrical coupling strength on frequency propagation. Referring to Table 1, when the disturbance occurs at bus 9, GThe peak frequency deviation of the three nodes reached 0.217Hz, with a mean square error of 1.3114e-04 and a standard deviation of 0.0115Hz. Traditional models could not capture this local extreme deviation at all. This invention quantifies the electrical coupling strength between nodes through the Jacobian matrix, predicts high-risk nodes in advance, and reduces the frequency dynamic prediction error to within 5%, providing a precise basis for control strategy design. The spatiotemporal frequency dynamic characterization accuracy during power grid disturbance time is significantly improved.

[0055] Table 1. Quantitative comparison of local frequency deviation indices at different disturbance locations.

[0056] This invention also eliminates the risk of spatial imbalance in power grid disturbances by achieving global homogenization of the initial RoCoF. When a power grid disturbance occurs, the difference in the initial RoCoF stems from the mismatch between the disturbance propagation strength and the node synthesis inertia. Referring to Table 2, without the strategy of this invention, the RoCoF differences between different nodes are significant: when the disturbance occurs at bus 7, Table 2 shows (…). H L Δ P L ) i Values G 1 = -0.0836 G 2 = -0.4079 G 3 = -0.1038, and the calculated RoCoF is -0.0836 / M 1 = -0.0139 Hz / s, -0.4079 / M 2 = -0.1359 Hz / s, -0.1038 / M = 0.1038 Hz / s, the spatial imbalance coefficient is defined as max|RoCoF| / min|RoCoF|, |-0.1359| / |-0.0139|=9.77 times, indicating that the frequency change rate of different nodes varies greatly.

[0057] Table 2. Different interference locations H L Δ P L

[0058] This invention is applied to power grid disturbances. By adjusting the combined inertia of each node, the RoCoF of all nodes is unified to a preset RoCoF limit α = -2 / 60Hz / s (approximately -0.033Hz / s). At this point, the absolute values ​​of RoCoF at each node are equal, and the spatial imbalance coefficient = max|RoCoF| / min|RoCoF| = 0.033 / 0.033 = 1.0, achieving global homogenization. Figure 5As shown, the initial frequency change trajectories of all nodes completely overlap, avoiding the risk of protection malfunction caused by sudden frequency changes in some nodes, and significantly improving the transient stability of the system. Figure 5 In the figure, a, b, c, and d are the frequency-time variation curves of bus 5, bus 7, bus 8, and bus 9, respectively.

[0059] This invention also achieves precise and controllable steady-state frequency deviation in the power grid, ensuring safe grid operation. The steady-state frequency deviation follows a formula, meaning the deviation is determined by the ratio of total disturbance propagation to total damping. A preset safety threshold is also included. =0.5Hz, refer to Table 2, calculate the total disturbance transmission quantity ∑ i ( H L Δ P L ) i =-0.3310-0.1964-0.0891=-0.6165, substituting into the formula, the total damping requirement is calculated to be 1.2332 pu. To ensure parameter compatibility, the combined inertia after design is used. M i Proportional Damping D i ,final G 1. G 2. G The damping values ​​for 3 are 5.55 pu, 2.78 pu, and 0.93 pu, respectively; (Refer to...) Figure 5 After implementation, the steady-state frequency deviation stabilized at -0.5Hz, corresponding to a frequency of 59.5Hz, fully meeting the requirements for low-inertia power grid operation; (Refer to...) Figure 6 Compared to the uncontrolled scenario, the deviation was reduced by 44.4%, and this was unaffected by the location of the disturbance. According to... Figure 6 The results show that the steady-state deviation consistency error is less than 0.01Hz under different disturbance locations.

[0060] This invention also significantly suppresses spatiotemporal frequency differences, improving the synchronization of the power grid system. During the implementation of this invention, different disturbance locations lead to large frequency deviations between nodes: referring to Table 1, for example, in the case of a disturbance at bus 9, as shown in Table 1... G The peak deviation at 3 nodes is 0.2170 Hz. G The peak deviation at node 1 is 0.1287 Hz, and the difference between the two is 0.0883 Hz, with a mean square error of 1.3114e-04, reflecting a significant spatiotemporal frequency difference. After applying the control method of this invention, through inertial-damping coordinated adjustment, the frequency change rate of each node is made consistent, and the steady-state deviation is unified, such as... Figure 6As shown, the frequency trajectories of all nodes almost completely overlap, with peak deviations all less than 0.01Hz, and mean square errors reduced to the order of 1e-06. The spatiotemporal frequency consistency is improved by more than 99%, effectively avoiding equipment damage and system oscillations caused by frequency asynchrony in the power grid system.

[0061] The present invention also discloses an embodiment, with reference to Figure 7 Taking the IEEE 9-bus test system as an example, this paper details the core implementation method of the present invention: the three-step process of "modeling → tuning → verification". The IEEE 9-bus test system has a base capacity of 100MVA and 345kV, and includes 3 IBR nodes ( G 1- G 3) The disturbance scenario is a 60MW load disturbance process applied to bus 7.

[0062] 1. System configuration and modeling implementation.

[0063] Input system parameters: grid topology, node voltage / phase angle (steady-state value), IBR initial inertia ( M =[6,3,1]s) and damping ( D =[10,20,30]pu), and 60MW load disturbance Δ P L .

[0064] Constructing a topological coupling model: Obtain the Jacobian matrix according to S3 in the above embodiment, and simplify to obtain... H red (Generator node coupling matrix) and H L Substituting the disturbance transfer matrix into the state-space model in S4, we obtain the LTI system model.

[0065] Simulation environment: The system is built based on MATLAB / Simulink and integrates the GFM / GFL inverter control module.

[0066] 2. Inertia-damping parameter tuning

[0067] inertia M Design: Based on the design inertia satisfaction formula in S5, and with the preset RoCoFα = -0.033Hz / s as the homogenization target, the following calculations are performed: G 1- M ≈2.53s G 2- M ≈12.36s G 3- M ≈3.14s.

[0068] Damping DDesign: Based on the design damping equation in S6, with a steady-state deviation ≤ 0.5Hz as the target, calculate the total damping ∑ D i ≈1.191 pu, M Proportional allocation G 1- D ≈0.18pu G 2- D ≈0.88pu G 3- D ≈0.13pu.

[0069] Parameter fixing: By adjusting the droop gain R of the GFM inverter and the filter time constant The design required to achieve the steady-state process after power grid disturbance M and D .

[0070] 3. Operation and Effect Verification

[0071] Simulation Startup: After steady-state operation, trigger bus 7 disturbance and record frequency trajectory, RoCoF, steady-state deviation and other indicators.

[0072] Core verification: After control, the RoCoF of all nodes is uniformly -0.033Hz / s (imbalance coefficient 1.0), and the steady-state deviation is -0.5Hz (refer to...). Figure 5 The peak deviation is ≤0.01Hz, which meets the design target.

[0073] 4. Scene adaptation

[0074] Different disturbance locations: Refer to Table 2 and recalculate. H L The vector can be retuned according to the formula in the above embodiment; this implementation method does not require additional hardware, and can be achieved by configuring the inverter control parameters through software. It is compatible with mainstream IBR equipment and has strong engineering operability.

[0075] and Figure 1 Corresponding to the grid spatiotemporal frequency modeling and control method based on inverter-type power supply shown, the present invention also discloses a grid spatiotemporal frequency modeling and control system based on inverter-type power supply, which is applied to the grid spatiotemporal frequency modeling and control method based on inverter-type power supply described in any of the above embodiments. It includes an inverter model construction module, a local frequency model construction module, a matrix acquisition module, a state space model construction module, a design inertia acquisition module, a design damping acquisition module, and a coordinated stability module connected in sequence.

[0076] The inverter model building module is used to adjust the actual output power of the grid-connected inverter to be the same as the preset output power, and build a grid-connected inverter model with dynamic frequency balance; it sets up grid terminals and injects deviation into the dynamic response power of the grid terminals through droop control to build a grid-connected inverter model with dynamic frequency balance; the grid-connected inverter model and the grid-connected inverter model are confirmed as inverter models.

[0077] The local frequency model construction module is used to inject active power into the generator side based on the inverter model, regulate the local frequency of the generator side, and construct the local frequency model of the generator side.

[0078] Matrix acquisition module: Based on the local frequency model on the generator side, the active power and phase angle injected by the system nodes on the generator side are linearized to obtain the Jacobian matrix.

[0079] The state-space model construction module is used to divide the system nodes into generator nodes and load nodes, divide the Jacobian matrix into blocks and eliminate load node variables, and construct a state-space model based on the local frequency model of the generator side.

[0080] The design inertia acquisition module is used to acquire the initial values ​​of frequency and angular deviation at the moment of power grid disturbance, and input them into the state space model to obtain the initial frequency change rate. The design inertia is then acquired based on the preset frequency change rate limit and the initial frequency change rate.

[0081] The design damping acquisition module is used to acquire the steady-state frequency deviation during steady-state operation of the power grid, and to acquire the design damping based on the steady-state frequency deviation and the preset safety threshold; the design damping is allocated according to the design inertia ratio.

[0082] The coordination and stabilization module is used to adjust the parameters of grid-connected and grid-reverse inverters based on design inertia and design damping, thereby coordinating grid operation.

[0083] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for spatiotemporal frequency modeling and control of power grids based on inverter-type power supplies, characterized in that, include: S1. Inverter model construction steps: Adjust the actual output power of the grid-connected inverter to be the same as the preset output power to construct a grid-connected inverter model with dynamic frequency balance; Set up grid terminals and inject deviation into the dynamic response power of the grid terminals through droop control to construct a grid-connected inverter model with dynamic frequency balance; Confirm the grid-connected inverter model and the grid-connected inverter model as the inverter model. S2. Local frequency model construction steps: Based on the inverter model, inject active power into the generator side, regulate the local frequency of the generator side, and construct the local frequency model of the generator side. S3. Matrix acquisition steps: Based on the generator-side local frequency model, the active power injected into the system nodes on the generator side and the phase angle are linearized to obtain the Jacobian matrix; S4. State-space model construction steps: Divide the system nodes into generator nodes and load nodes, divide the Jacobian matrix into blocks and eliminate load node variables to obtain a simplified Jacobian matrix, and construct the state-space model based on the simplified Jacobian matrix and the generator-side local frequency model. S5. Design inertia acquisition steps: Obtain the initial values ​​of frequency and angular deviation at the instant of power grid disturbance, and input them into the state space model to obtain the initial frequency change rate. Obtain the design inertia according to the preset frequency change rate limit and the initial frequency change rate. S6. Steps for obtaining design damping: Obtain the steady-state frequency deviation during steady-state operation of the power grid, and obtain the design damping based on the steady-state frequency deviation and the preset safety threshold; the design damping is allocated according to the design inertia ratio. S7. Coordination and stabilization steps: Adjust the parameters of grid-connected inverters and grid-reverse inverters according to the design inertia and design damping to coordinate grid operation.

2. The grid spatiotemporal frequency modeling and control method based on inverter-type power supply according to claim 1, characterized in that, In S1, a frequency-dynamically balanced grid-type inverter model is constructed by injecting deviation into the dynamic response power of the grid terminals through droop control, specifically including: The active power droop control with low-pass filtering has the following transfer function: ； In the formula, This refers to the frequency deviation at the inverter output port. Current power and rated power P e deviation, R For droop gain, The time constant of the low-pass filter. s For the Laplace operator.

3. The grid spatiotemporal frequency modeling and control method based on inverter-type power supply according to claim 2, characterized in that, Further derivation of the transfer function yields the state-space form of the swing-like equation: ； in, for x rate of change, x for , P For power setting value, M For equivalent inertia, D Let be the damping constant. M Equivalent inertia of grid-connected inverters M GFM = / R , D Damping constant of grid-connected inverter D GFM = 1 / R .

4. The grid spatiotemporal frequency modeling and control method based on inverter-type power supply according to claim 1, characterized in that, The formula involved in S2 is: ； Where, Δ δ G Δ represents the rotor angle deviation of the generator. ω G For generator frequency deviation, , These represent the rates of change of rotor angular deviation and frequency deviation, respectively, Δ P G Injecting incremental power M For equivalent inertia, D Let be the damping constant. ω 0 = 2π f 0, f 0 represents the system's rated frequency.

5. The grid spatiotemporal frequency modeling and control method based on inverter-type power supply according to claim 1, characterized in that, The formula involved in S3 is: ; in, H Let Δ be the Jacobian matrix. P Injection amount of active power for system nodes ， Δ δ Angle difference before and after injecting active power into the system nodes ， element H ij Reflecting nodes i and j The electrical coupling strength is expressed as: ； In the formula ，V , δ These represent the voltage magnitude and phase angle of the system nodes, respectively. g、b These are the conductance and susceptance elements of the nodal admittance matrix.

6. The grid spatiotemporal frequency modeling and control method based on inverter-type power supply according to claim 4, characterized in that, S4 specifically includes: The simplified formula for calculating the Jacobian matrix is: ； in, G Indicates the generator node. L Indicates the load node. H red To simplify the Jacobian matrix, H red = H GG - ¹ H GL H LL - ¹ H LG , H L = H GL H LL - ¹, H GG This is the coupling submatrix between generator nodes. H GL This is the coupling submatrix between the generator node and the load node. H LL This is the coupling submatrix between load nodes. H LG Δ is the coupling submatrix between the load node and the generator node. P L This is a disturbance to the active power on the load side. The state-space model is as follows: 。 7. The grid spatiotemporal frequency modeling and control method based on inverter-type power supply according to claim 6, characterized in that, The formula involved in S5 is: ； in, For load disturbances at nodes i The equivalent transitivity, M i For design inertia, the design inertia must satisfy: ； in, α This is a preset limit for the rate of change of frequency.

8. The grid spatiotemporal frequency modeling and control method based on inverter-type power supply according to claim 7, characterized in that, In S6, the stable frequency deviation is: ； in, D i To design damping, the damping must satisfy: ； in, This is a preset safety threshold.

9. A grid spatiotemporal frequency modeling and control system based on inverter-type power supply, characterized in that, The grid spatiotemporal frequency modeling and control method based on inverter-type power supply according to any one of claims 1-8 includes an inverter model construction module, a local frequency model construction module, a matrix acquisition module, a state space model construction module, a design inertia acquisition module, a design damping acquisition module, and a coordinated stability module connected in sequence. The inverter model building module is used to adjust the actual output power of the grid-connected inverter to be the same as the preset output power, and build a grid-connected inverter model with dynamic frequency balance; it sets up grid terminals and injects deviation into the dynamic response power of the grid terminals through droop control to build a grid-connected inverter model with dynamic frequency balance; the grid-connected inverter model and the grid-connected inverter model are confirmed as inverter models. The local frequency model construction module is used to inject active power into the generator side based on the inverter model, regulate the local frequency of the generator side, and construct the local frequency model of the generator side. Matrix acquisition module: Based on the generator-side local frequency model, the active power and phase angle injected into the system nodes on the generator side are linearized to obtain the Jacobian matrix; The state-space model construction module is used to divide the system nodes into generator nodes and load nodes, divide the Jacobian matrix into blocks and eliminate load node variables, and construct a state-space model based on the generator-side local frequency model. The inertia acquisition module is designed to acquire the initial values ​​of frequency and angular deviation at the moment of power grid disturbance, and input them into the state space model to obtain the initial frequency change rate. The design inertia is then acquired based on the preset frequency change rate limit and the initial frequency change rate. The design damping acquisition module is used to acquire the steady-state frequency deviation during steady-state operation of the power grid, and to acquire the design damping based on the steady-state frequency deviation and a preset safety threshold; the design damping is allocated according to the design inertia ratio. The coordination and stabilization module is used to adjust the parameters of grid-connected and grid-reverse inverters based on design inertia and design damping, thereby coordinating grid operation.

Images