A method and system for modeling power grid evolution behavior based on dynamic digital twins

Abstract

This invention provides a method and system for modeling power grid evolution behavior based on dynamic digital twins, belonging to the field of smart grid digital twin technology. The method captures the interaction characteristics of power plant output, user load response, and energy storage dispatch strategies in real time through a digital twin interface, forming behavioral coupling factors with complementary and conflict weights. Implicit tensor fusion technology and nonlinear projection are used to construct power grid response relationships and multi-scale demand elastic surfaces. Further steps, such as virtual behavioral anchor point perturbation, separate offsets to identify imbalance regions, generate evolution trajectories, extract the optimal cooperative path, and dynamically adjust the weights of the behavioral coupling factors to achieve rapid convergence of the global cooperative stability domain. This invention includes modules for generating behavioral coupling factors, supports various operations, effectively solves many problems in power grid resource interaction, and improves the dynamic adaptability and cooperative efficiency of power grid evolution modeling.

Description

Technical Field

[0001] This invention relates to the field of digital twin technology for smart grids, specifically to a method and system for modeling the evolutionary behavior of power grids based on dynamic digital twins. Background Technology

[0002] As energy systems evolve towards intelligence, the application of digital twin technology in power grid modeling is gradually deepening, mainly used for mirror mapping and state extrapolation between physical systems and virtual spaces. Existing technologies mostly construct local simulation models based on single-dimensional data, or approximate the interaction relationships of multiple entities through linear superposition.

[0003] However, with the increasing penetration of distributed resources in the power grid and the growing complexity of interaction scenarios, the spatiotemporal behavioral game among multiple entities exhibits high dynamism and asymmetry, highlighting the limitations of existing technologies. These issues make it difficult for the power grid to achieve efficient coordination and dynamic optimization across the spatiotemporal dimensions when dealing with fluctuations in the high proportion of renewable energy and user-side demand response. Summary of the Invention

[0004] To achieve the above objectives, this invention provides a method and system for modeling power grid evolution behavior based on dynamic digital twins;

[0005] In a first aspect, embodiments of the present invention provide a method for modeling the evolution behavior of power grids based on dynamic digital twins;

[0006] The system captures power plant output characteristics, user load response patterns, and energy storage operator scheduling strategies in real time through a digital twin interface, abstracting them into behavioral coupling factors. These behavioral coupling factors are then used to extract the potential complementarity and conflict weights of resource interactions between entities using implicit tensor fusion technology. Furthermore, the time-series data of power plant output, user load, and energy storage strategies are input into a three-dimensional implicit tensor model to decompose the output fluctuation cycle characteristics, load elasticity distribution patterns, and energy storage regulation rules, and to extract cross-entity interaction features. A dynamic mask matrix is ​​generated based on historical grid conflict events to prune the resource competition path between power plants and energy storage, preserving the supply-demand complementarity path between power plants and users. The difference between power plant regulation delay and user response delay is calculated, and the original matching degree is dynamically weakened using an exponential decay function to generate an effective complementarity coefficient. This effective complementarity coefficient is then fused with the real-time regulation difference between power plants and energy storage through a dual-channel fusion process, with complementary terms weighted and conflict terms threshold filtered to form a coupling matrix. Real-time detection of sudden output changes or load distortions triggers weight adjustments to enhance the weight of complementary paths and suppress conflict terms. The pruning rules and decay parameters are corrected through digital twin feedback.

[0007] Furthermore, based on behavioral coupling factors, a grid response relationship is constructed. Power plant output fluctuations are mapped to a user-side demand elasticity surface via nonlinear projection, forming a grid decision space with energy storage charging and discharging strategies. Specifically, this includes: utilizing output cycle characteristics and load elasticity patterns, mapping the fundamental and high-frequency components of power plant output to user-side price-sensitive and overload-risk areas via non-orthogonal projection, generating a multi-scale demand elasticity surface to capture the distortion diffusion path of output fluctuations on the user side; discretizing charging and discharging rates and capacity margins into spatiotemporal gradient units according to energy storage regulation rules, establishing a buffer correlation matrix between these units and the supply-demand imbalance region in the elasticity surface, where the correlation strength is regulated by pruned complementary paths; dynamically fusing conflict suppression factors and real-time complementarity based on effective complementarity coefficients and historical conflict rates, and weighted reconstructing the initial connection strength; and expanding and fusing the demand elasticity surface and energy storage buffer matrix using implicit tensor basis extension, embedding the corrected dynamic strength as an orthogonal dimension constraint to form a grid decision space containing asymmetric coupling chains. Its orthogonality ensures the resolvability of subsequent virtual behavioral anchor point disturbance trajectories and the stability domain deviation measurement.

[0008] Furthermore, a power grid evolution model is generated by introducing virtual behavioral anchor points to disturb the power grid response relationship. The virtual behavioral anchor points are based on the orthogonal coupling chain of the power grid decision space, extracting the breakpoints of resource interaction in historical evolution, and combining the extreme values ​​of complementary path weight decay and conflict path mutation thresholds to cross-locate spatiotemporal heterogeneous anchor points. The anchor points with the largest deviation from the real-time complementarity are selected, and virtual output or load disturbances are injected into their associated nodes through digital twins to induce local resonance of the coupling chain. The trajectory offset after disturbance is separated by the independence of orthogonal basis vectors to identify new breakpoints not covered by historical conflicts. The new breakpoints are fed back to the complementary path pruning rules to dynamically update the conflict threshold and verify the convergence of the corrected coupling chain. If the threshold is not reached, a secondary disturbance is triggered until the evolution trajectory satisfies the chain resonance constraint.

[0009] Furthermore, by triggering local resonance of the coupled chain through virtual behavioral anchor point perturbation, a model evolution trajectory reflecting the dynamics of resource interaction is generated. The optimal cooperative path is extracted based on the deviation between the trajectory and the target stability domain, where the deviation is defined as a composite function of resource mismatch, cooperative delay cost, and evolutionary inertia resistance. Specifically, this includes: decomposing the resonance response of the coupled chain in the orthogonal decision space; dynamically adjusting the connection weights of the power plant-user-energy storage node according to the trajectory offset; reconstructing the perturbation propagation path; separating the evolution trajectory of each node after perturbation using orthogonal basis vectors; calculating the superposition of power plant output deviation, user load distortion rate, and energy storage regulation lag; extracting the implicit imbalance region; mapping the implicit imbalance region to the coupling matrix; dynamically updating the effective coefficients of complementary paths and conflict term filtering rules; and reversing the elastic surface mapping parameters through the digital twin interface; verifying the convergence of the corrected coupled chain under the stability domain deviation metric; if the threshold is not reached, triggering a secondary perturbation of the new anchor point.

[0010] Furthermore, the optimal cooperative path is injected into the adversarial evolution channel of the digital twin. Cooperative deviation is detected through virtual behavioral chain resonance, and the fusion weights of the behavioral coupling factors are dynamically adjusted until the deviation is lower than a preset tolerance threshold. Specifically, this includes: separating the spatiotemporal coupling components in the trajectory offset caused by virtual behavioral anchor point perturbation based on the independence of orthogonal basis vectors, and constructing a response vector for chain resonance; combining the boundary constraints of the target stable domain, performing a tensor inner product operation on the response vector and the dynamic intensity of the coupling matrix to generate a cooperative deviation index; if the cooperative deviation index exceeds the preset tolerance threshold, the fusion weights of the behavioral coupling factors are dynamically adjusted, and the dynamic mask matrix and complementary coefficient attenuation parameters are corrected in reverse through implicit tensor basis expansion, thereby enhancing the complementary path weights and suppressing conflict terms, triggering a secondary response of local resonance in the coupling chain.

[0011] Secondly, embodiments of the present invention provide a power grid evolution behavior modeling system based on dynamic digital twins;

[0012] The system comprises a behavior coupling factor generation module, a power grid evolution behavior model module, a cooperative path optimization module, and a feedback control module. The behavior coupling factor generation module captures power plant output characteristics, user load response patterns, and energy storage operator scheduling strategies in real time via a digital twin interface, and extracts the complementarity and conflict weights of resource interactions between entities using implicit tensor fusion technology. The power grid evolution behavior model module constructs power grid response relationships based on behavior coupling factors, generates a power grid decision space by fusing a demand elasticity surface and an energy storage buffer matrix through nonlinear projection, and dynamically corrects node connection strength. The cooperative path optimization module triggers local resonance of the coupling chain through virtual behavior anchor point perturbation, generates the model evolution trajectory, and extracts the optimal cooperative path. The feedback control module injects the optimal cooperative path into the adversarial evolution channel, detects cooperative deviations, and dynamically adjusts the fusion weights to ensure that the evolution trajectory converges to the target stability domain.

[0013] Furthermore, the power grid evolution behavior model module captures the distortion diffusion path of power output fluctuations through non-orthogonal projection technology, and establishes a buffer correlation matrix in conjunction with the energy storage spatiotemporal gradient unit to dynamically adjust the correlation strength of the supply-demand imbalance region; the cooperative path optimization module uses orthogonal basis vectors to separate the disturbance trajectory offset, extract the hidden imbalance region, and reversely correct the elastic surface parameters; the feedback control module quantifies the cooperative deviation through tensor inner product operation, triggers the dynamic adjustment of the fusion weight and the secondary response of the chain resonance, and realizes the rapid convergence of the global cooperative stability domain.

[0014] This invention quantifies the complementary and conflicting relationships between power grid entities in real time using dynamic digital twin technology. Combined with virtual behavior anchor point perturbation and chain resonance mechanisms, it overcomes the limitations of traditional static modeling methods in dynamic evolution path optimization and collaborative stability control. Compared to existing technologies, this invention provides a power grid evolution behavior modeling method and system based on dynamic digital twins, offering the following advantages:

[0015] (1) This invention achieves refined extraction of multi-entity resource interaction features and active suppression of conflict paths through implicit tensor fusion and dynamic mask pruning techniques.

[0016] (2) Based on the perturbation mechanism of orthogonal decision space and virtual behavior anchor point, this invention enhances the dynamic identification ability of evolution model for new imbalance regions.

[0017] (3) The present invention ensures the global coordinated stability of the power grid evolution behavior under complex disturbances by using chain resonance constraints and deviation-driven feedback adjustment; the method and system provide operable technical support for the efficient allocation of power grid resources and real-time optimization of evolution paths. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below. 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.

[0020] Example 1: This embodiment of the invention provides a specific implementation method for a power grid evolution behavior modeling method based on dynamic digital twins. Taking a smart grid in a certain region as an example, the power grid includes distributed photovoltaic power stations, smart user load clusters, and distributed energy storage systems.

[0021] Photovoltaic output is affected by weather changes, exhibiting a 24-hour cycle of base frequency fluctuations and high-frequency minute-level disturbances. User loads are divided into price-sensitive areas and overload risk areas. Distributed energy storage systems need to dynamically adjust charging and discharging rates and capacity margins to balance supply and demand. Traditional modeling methods cannot capture the asymmetry of dynamic interactions among multiple entities due to static weight allocation, resulting in low collaborative efficiency.

[0022] This embodiment uses dynamic digital twin technology to achieve closed-loop control from data-driven modeling to global optimization. The specific implementation process is as follows:

[0023] Photovoltaic output time-series data is captured in real time through a digital twin interface and decomposed into a 24-hour periodic component of the base frequency and a 5-minute high-frequency disturbance component. For example, the base frequency fluctuation amplitude is ±20% under sunny conditions, while the instantaneous fluctuation of the high-frequency disturbance can reach ±50% under cloudy or rainy conditions. The user load response curve is divided into a price-sensitive zone and an overload risk zone. The load elasticity coefficient of the price-sensitive zone is 0.8, allowing a response delay of 5 minutes. The load elasticity coefficient of the overload risk zone is 0.2, with a response delay of only 10 seconds. The distributed energy storage system scheduling strategy includes a charging and discharging rate threshold of 2MW / min and dynamic adjustment rules for capacity margin. The upper limit of capacity proportion during off-peak hours is 80%. After inputting the three types of data into a three-dimensional implicit tensor model, the output fluctuation characteristics are decomposed along the time, space, and frequency domain dimensions to extract the load elasticity distribution pattern and analyze the energy storage adjustment rules. Cross-entity interaction residual features are extracted through an implicit tensor fusion algorithm. For example, when the photovoltaic output drops by 50% within 10 minutes, the user's price-sensitive zone load is not adjusted synchronously due to the response delay, causing the energy storage system to overcharge and trigger a protection shutdown event.

[0024] A dynamic mask matrix is ​​generated based on historical conflict events to prune the resource competition paths between photovoltaic (PV) and energy storage. For example, conflict paths where the energy storage charging and discharging rates exceed limits during high-frequency disturbances in PV output are eliminated, while complementary paths for active load adjustment in price-sensitive areas of PV base frequency output are retained. By calculating the difference between a 30-second PV adjustment delay and a 5-minute user response delay, a time window sliding matching mechanism is adopted, and an exponential decay function is defined to dynamically correct the initial matching degree. The decay function time constant is set to 10 minutes, dynamically weakening the initial matching degree of 0.9 to an effective complementary coefficient of 0.75. The energy storage adjustment difference is monitored in real time, and when the deviation between the actual and target values ​​of the charging and discharging rates exceeds 15%, a conflict term filtering mechanism is triggered. The dual-channel fusion superimposes the effective complementary coefficient and conflict terms with a weight of 7:3 to generate a spatiotemporal coupling matrix. When a sudden change in PV output exceeding 40% or a user load distortion rate exceeding 20% ​​is detected, the complementary path weight is automatically increased to 8:2, the conflict filtering threshold is simultaneously tightened to 10%, and the decay function time constant is optimized to 5 minutes through a feedback loop.

[0025] The grid response relationship is constructed based on a spatiotemporal coupling matrix. Nonlinear projection maps the photovoltaic output base frequency to the user's price-sensitive area. For example, the 24-hour periodic fluctuation of the base frequency is correlated with the elastic enhancement of the user's off-peak electricity hours, allowing for delayed load response to match low output. High-frequency 5-minute disturbance components are mapped to the user's overload risk area, triggering protection mechanisms to prevent voltage exceedances. The energy storage system is discretized into spatiotemporal gradient units based on a charge / discharge rate of 2MW / min and a capacity margin of 80%, with each 5-minute period as a time unit, and a buffer correlation matrix is ​​established with the supply-demand imbalance region in the elastic surface.

[0026] The correlation strength in the buffer correlation matrix is ​​regulated by the pruned complementary paths, with an effective complementarity coefficient of 0.75 corresponding to a buffer strength of 0.6. The historical conflict rate of 5 times / day is dynamically integrated with the real-time complementarity, with the conflict suppression factor weight set to 0.4 and the real-time complementarity weight set to 0.6. The initial connection strength of 0.6 is reconstructed to 0.52 through weighted refactoring. The demand elasticity surface and the energy storage buffer matrix are integrated through implicit tensor basis expansion, and orthogonal dimension constraints are embedded to form the power grid decision space. The orthogonal basis vectors independently represent the output fluctuation, load elasticity, and energy storage regulation dimensions, ensuring that the evolution trajectory of each dimension can be separated and analyzed after the virtual behavior anchor point disturbance.

[0027] When locating virtual behavior anchor points within the orthogonal decision space, target anchor points are selected by combining historical evolution breakpoint data and real-time complementarity deviation. The orthogonal decision space is constructed through implicit tensor basis expansion, and its core consists of three independent orthogonal basis vectors. The first basis vector is used to characterize the frequency domain characteristics of the fluctuating energy distribution of power output, the second basis vector maps the user load elastic gradient, and the third basis vector describes the spatiotemporal regulation capability of energy storage. Specifically, the first basis vector is generated by decomposing the fundamental frequency component and high-frequency disturbance component of photovoltaic power output time series data. For example, the energy distribution of the 24-hour periodic fluctuation of the fundamental frequency and the 5-minute high-frequency disturbance is projected onto the spatiotemporal dimension to quantify the spectral characteristics of the dynamic changes in power output. The second basis vector is formed by non-orthogonal projection of the elastic coefficient difference and response delay time between the user price sensitive area and the overload risk area, reflecting the hierarchical characteristics of the load-side elastic response. The third basis vector is generated by orthogonal decomposition based on the discretization rules of the energy storage charge and discharge rate threshold, capacity margin, and spatiotemporal gradient unit, describing the spatiotemporal constraints of energy storage regulation.

[0028] When a certain anchor point deviates from the target value of 0.8 by 6.25% due to a complementarity coefficient of 0.75, a virtual disturbance is injected into the nodes of its buffer correlation matrix; for example, simulating a further 10% reduction in photovoltaic output, or forcing users in price-sensitive areas to extend the load delay response time to 8 minutes, triggering local resonance in the coupled chain; the trajectory offset after the disturbance is analyzed by orthogonal basis vector separation: the first basis vector shows a 12% shift in the power fluctuation energy distribution, reflecting the degree of diffusion of high-frequency disturbance energy in the frequency domain; the second basis vector detects an increase in voltage fluctuation rate to 5.5% in the overload risk area, reflecting the deterioration of voltage stability caused by insufficient load elasticity; the third basis vector reveals that the energy storage regulation lag reaches 7%. The problem stems from the deviation in charge and discharge rates under capacity margin constraints. Residual analysis further identifies new breakpoints, such as a certain energy storage node being unable to respond to high-frequency disturbances due to insufficient capacity margin, leading to a 1.5MW increase in the supply-demand gap. This breakpoint is fed back to the dynamic mask matrix, adding energy storage capacity verification conditions and reconstructing the complementarity coefficient to 0.78. If the evolution trajectory deviates from the target stability domain, resulting in a resource mismatch exceeding 5%, a secondary disturbance is triggered. The elasticity coefficient of the overload risk zone is increased to 0.25 along the second basis vector to enhance voltage stability, and the energy storage charge and discharge rate is optimized to 2.3MW / min along the third basis vector to break through the spatiotemporal gradient constraint until the voltage fluctuation rate falls back to 4.8%.

[0029] After injecting the optimized collaborative path into the adversarial evolution channel of the digital twin, a chain resonance detection model is constructed based on the independence of orthogonal basis vectors. The first chain resonance vector is composed of the tensor product of the power output fluctuation energy distribution represented by the first basis vector and the user voltage fluctuation rate mapped by the second basis vector, quantifying the nonlinear coupling effect of power output disturbance on user-side voltage stability. The second chain vector is generated by time-domain convolution of the energy storage regulation lag described by the third basis vector and the load response delay time in the second basis vector, reflecting the cumulative impact of the collaborative failure of energy storage regulation lag and load response delay on the supply-demand gap. Combined with the boundary conditions of the target stability domain, a composite constraint index is defined: resource mismatch ≤ 5%, voltage fluctuation rate ≤ 5%, and collaborative delay cost ≤ 3%. The current... The coordination deviation index was 5.2%, with the first chain vector contributing 3.1% and the second chain vector contributing 2.1%. After exceeding the preset tolerance threshold, the feedback control engine was activated to dynamically adjust the weights of the behavioral coupling factors: the complementary path weights were increased from 8:2 to 9:1 to enhance the control priority of the output-load complementary path, the conflict deviation rate threshold was tightened from 10% to 8% to suppress the activation probability of the energy storage-output conflict path, and the implicit tensor basis was expanded to refine the energy storage spatiotemporal gradient unit to a 3-minute granularity to improve the regulation time resolution. After two chain resonance iterations, the output fluctuation energy offset was reduced to 10%, the voltage fluctuation rate was stabilized at 4.7%, the energy storage regulation lag was optimized to 4.5%, the coordination deviation index converged to 2.8%, and the global stability domain was achieved.

[0030] Example 2: This embodiment of the invention provides a power grid evolution behavior modeling system based on dynamic digital twins. The system consists of four core modules: a behavior coupling factor generation module, a power grid evolution behavior model module, a cooperative path optimization module, and a feedback control module. Each module interacts with chain-like control logic through a standardized data interface. The specific technical implementation is as follows:

[0031] The behavioral coupling factor generation module collects real-time photovoltaic output time-series data, user load response curves, and energy storage scheduling strategies through a digital twin interface. This module incorporates a three-dimensional implicit tensor processor to perform frequency domain decomposition on the photovoltaic output data, extracting the energy distribution characteristics of the fundamental frequency component and high-frequency disturbance components. It also analyzes the elasticity coefficients of the price-sensitive area and the overload risk area of ​​the user load, and generates dynamic adjustment parameters based on energy storage charge / discharge rate thresholds and capacity margin rules. Cross-entity interaction features are extracted using an implicit tensor fusion algorithm, such as the correlation analysis between photovoltaic output drop events and user response delays. Combined with a historical conflict event database, a dynamic mask matrix is ​​generated to prune the resource competition paths between photovoltaic and energy storage, preserving complementary paths between output troughs and proactive user adjustments. The output of the behavioral coupling factor generation module transmits the spatiotemporal coupling matrix to the grid evolution behavior model module via a standardized JSON interface. The data fields include output spectrum distribution, load elasticity gradient, and energy storage adjustment parameters.

[0032] After receiving the spatiotemporal coupling matrix, the grid evolution behavior model module uses a nonlinear projection engine to map the photovoltaic output base frequency to the user price-sensitive area and associate high-frequency disturbance components with the overload risk area, constructing a multi-scale demand elastic surface. This surface quantifies the transmission effect of output fluctuations on user-side voltage stability and load distortion; for example, high-frequency disturbances cause a 3% increase in voltage fluctuation rate in the overload risk area. The energy storage system constructs a buffer correlation matrix based on spatiotemporal gradient units, and its capacity margin and charge / discharge rate dynamically match the supply-demand imbalance area of ​​the elastic surface. Through orthogonalization constraints, the elastic surface and the buffer matrix are merged into a grid decision space. The three orthogonal basis vectors independently represent the spectral energy distribution of output, the load elastic stratification characteristics, and the spatiotemporal regulation capability of energy storage. The grid decision space data is transmitted to the collaborative path optimization module through a high-speed binary stream to ensure low-latency parsing of multi-dimensional data.

[0033] The collaborative path optimization module locates virtual behavioral anchor points based on the power grid decision space and filters high-risk anchor points with complementary degree deviations exceeding the threshold using a historical evolution breakpoint library. It also simulates scenarios of sudden output drops or load delays through a virtual disturbance injection mechanism. For example, a 10% drop in photovoltaic output may trigger local resonance in the coupling chain. The trajectory offset after the disturbance is analyzed by orthogonal basis vectors as output spectrum shift, voltage fluctuation rate increase, and energy storage regulation lag. After residual analysis identifies new breakpoints, the conflict path filtering rules of the mask matrix are dynamically updated, and the correction parameters are fed back to the preceding behavioral coupling factor generation module. If the resource mismatch exceeds the limit, the collaborative path optimization module optimizes the response capability of the overload risk zone along the load elastic gradient, increases the charge and discharge rate threshold along the energy storage regulation dimension, generates the optimal collaborative path, and transmits it to the feedback control module via a RESTful API.

[0034] The feedback control module injects the cooperative path into the adversarial evolution channel to construct a chain resonance detection model. The first chain resonance vector is generated by the tensor product of the output spectrum energy and voltage fluctuation, quantifying the nonlinear impact of power supply disturbances on user-side stability. The second chain vector reveals the cumulative effect of cooperative failure through the convolution operation of energy storage regulation lag and load response delay. The feedback control module calculates the cooperative deviation index in conjunction with the boundary conditions of the target stability domain, triggering dynamic weight adjustment and conflict threshold optimization. For example, the complementary path weight is increased to 9:1, and the energy storage spatiotemporal gradient granularity is refined to 3 minutes. After two iterations of optimization, the convergence time of the system's global stability domain is shortened, the voltage fluctuation rate suppression efficiency is improved, and the resource mismatch is stabilized.

[0035] The various modules of the system achieve seamless interaction through standardized interfaces. The behavioral coupling factor generation module and the power grid evolution model module encapsulate data in JSON format; decision space data is transmitted through binary streams; feedback control commands are synchronized in reverse through RESTful APIs; the three-dimensional implicit tensor processor supports frequency domain-spatiotemporal joint analysis; the orthogonalization engine achieves multi-dimensional data decoupling; and the chain resonance model quantifies dynamic game effects through tensor operations.

[0036] This invention addresses the problems of traditional power grid modeling methods in dynamic interactive scenarios, such as coarse modeling of multiple single-entity game dynamics, sluggish response to high-frequency disturbances, and insufficient control of supply-demand imbalance risks. It achieves a systemic breakthrough through dynamic digital twin technology. Traditional models, due to static coupling and linearization, struggle to characterize the dynamic asymmetric interaction characteristics of photovoltaic power output, user load, and energy storage regulation, resulting in limited collaborative efficiency. This system extracts multi-entity interaction features through a behavior coupling factor generation module and dynamically prunes conflict paths, significantly reducing resource mismatch risks. Based on an orthogonal decision space, it achieves independent analysis of power output spectrum, load elasticity, and energy storage regulation dimensions, accurately locating high-frequency disturbance propagation paths and improving dynamic response sensitivity. Furthermore, through a modular architecture and standardized interface design, the system realizes end-to-end control from data acquisition to collaborative optimization, effectively improving the power grid's adaptability and operational economy in complex dynamic scenarios, providing reliable technical support for the efficient consumption and safe operation of high-proportion renewable energy.

[0037] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for modeling the evolutionary behavior of power grids based on dynamic digital twins, characterized in that, Includes the following steps: S1. Real-time capture of power plant output characteristics, user load response patterns and energy storage operator scheduling strategies through digital twin interface, and abstract them into behavioral coupling factors, wherein the behavioral coupling factors include potential complementarity and conflict weights of resource interaction between entities. S2. Based on the behavioral coupling factor, the power grid response relationship is constructed. The power output fluctuation of the power plant is mapped to the user-side demand elastic surface through nonlinear projection. It forms a power grid decision space with the energy storage charging and discharging strategy. The connection strength between nodes in the power grid response relationship is inversely corrected by the historical conflict rate and real-time complementarity of the entity behavior. S3. By introducing virtual behavioral anchor points to disturb the power grid response relationship, a power grid evolution model is generated; S4. By disturbing the power grid response relationship through virtual behavioral anchor points, local resonance of the coupling chain is triggered, generating a model evolution trajectory that reflects the dynamics of resource interaction; the optimal cooperative path is extracted based on the deviation between the model evolution trajectory and the target stability domain, wherein the deviation is defined as a composite function of resource mismatch, cooperative delay cost and evolutionary inertia resistance. S5. Inject the optimal collaborative path into the adversarial evolution channel of the digital twin, detect collaborative deviations through virtual behavioral chain resonance, and dynamically adjust the fusion weight of behavioral coupling factors until the deviation is lower than the preset tolerance threshold. Among them, the behavioral coupling factor achieves asymmetric complementary configuration of power plant, user and energy storage operator resources in the spatiotemporal dimension through implicit tensor fusion and coupling-driven optimization of grid response relationship, and the grid decision space is constrained by chain resonance of decision verification to make the behavior of the grid evolution model converge to the global cooperative stability domain. Among them, the virtual behavior anchor point is an orthogonal coupled chain based on the power grid decision space. It extracts the breakpoints of resource interaction in historical evolution and cross-locates spatiotemporally heterogeneous anchor points by combining the extreme value of complementary path weight decay and the threshold of conflict path mutation. It selects the anchor point with the largest deviation from the real-time complementarity and injects virtual output or load disturbances into its associated nodes through digital twins to induce local resonance of the coupled chain. Through the independence of orthogonal basis vectors, it separates the trajectory offset after disturbance and identifies new breakpoints that are not covered by historical conflicts. The new breakpoints are fed back to the complementary path pruning rules to dynamically update the conflict threshold and verify the convergence of the modified coupled chain under the deviation metric of the power grid evolution stability domain.

2. The method for modeling power grid evolution behavior based on dynamic digital twins according to claim 1, characterized in that: The process of abstracting the behavior coupling factor in step S1 is as follows: S11. Input the time series data of power plant output, user load and energy storage strategy into the three-dimensional implicit tensor model, decompose the output fluctuation period characteristics, extract the load elastic distribution pattern, and analyze the energy storage regulation rules. Extract the cross-entity interaction features implicit in the residuals of the three decompositions; S12. Generate a dynamic mask matrix based on historical power grid conflict events, prune the power plant-energy storage resource competition path in the interaction features, and retain the power plant-user supply and demand complementary path. S13. In the complementary path, calculate the difference between the power plant regulation delay and the user response delay, and dynamically weaken the original matching degree through an exponential decay function to generate an effective complementary coefficient. S14. The effective complementarity coefficient and the real-time adjustment difference between the power plant and energy storage are fused in a dual-channel manner, the complementary terms are weighted and superimposed, and the conflict terms are threshold filtered to generate a coupling matrix. S15. Real-time detection of sudden output changes or load distortions triggers weight adjustments, enhances the weights of complementary paths and suppresses conflict terms, and corrects pruning rules and attenuation parameters through digital twin feedback.

3. The method for modeling power grid evolution behavior based on dynamic digital twins according to claim 1, characterized in that: The process of constructing the power grid response relationship in step S2 is as follows: S21. Based on the power output cycle characteristics and load elasticity pattern, the power output fundamental frequency and high frequency components of the power plant are mapped to the price-sensitive area and overload risk area on the user side through non-orthogonal projection, generating a multi-scale demand elasticity surface to capture the distortion diffusion path of power output fluctuations on the user side. S22. According to the energy storage regulation rules, the charging and discharging rate and capacity margin are discretized into spatiotemporal gradient units, and a buffer correlation matrix between them and the supply and demand imbalance region in the elastic surface is established. The correlation strength is regulated by the complementary path after pruning in step S1, and the conflict path is eliminated by threshold filtering. S23. Based on the effective complementarity coefficient and historical conflict rate, dynamically fuse the conflict suppression factor and real-time complementarity to reconstruct the initial connection strength in a weighted manner; S24. The demand elasticity surface and the energy storage buffer matrix are extended and merged according to the implicit tensor basis in step S1. The modified dynamic intensity is embedded as an orthogonal dimension constraint to form a power grid decision space containing asymmetric coupling chains. Its orthogonality ensures the analyzability of the subsequent virtual behavior anchor point disturbance trajectory and the stability domain deviation measurement.

4. The method for modeling power grid evolution behavior based on dynamic digital twins according to claim 1, characterized in that: The construction process of the power grid evolution model is as follows: S31. Based on virtual disturbance, the resonant response of the coupling chain is decomposed in the orthogonal decision space, and the connection weight of the power plant-user-energy storage node is dynamically adjusted according to the trajectory offset to reconstruct the disturbance propagation path; S32. Using orthogonal basis vectors to separate the evolution trajectories of each node after disturbance, calculate the superposition of power plant output deviation, user load distortion rate and energy storage regulation lag, and extract the hidden imbalance area not covered by conflict pruning. S33. Map the latent imbalance region to the coupling matrix, dynamically update the effective coefficients of complementary paths and the conflict term filtering rules, and reverse correct the elastic surface mapping parameters through the digital twin interface; S34. Verify the convergence of the corrected coupled chain under the stability region deviation metric. If the threshold is not reached, trigger a secondary perturbation of the new anchor point.

5. The method for modeling power grid evolution behavior based on dynamic digital twins according to claim 4, characterized in that: The identification of latent imbalance regions in step S32 includes: S321. Perform time-domain convolution between the power plant output deviation and the conflict weight in the behavior coupling factor to generate a conflict correlation vector; at the same time, perform reverse gradient matching between the user load distortion rate and the complementarity coefficient to generate a complementary correlation vector. S322. Based on the trajectory offset direction after the virtual behavior anchor point disturbance, the conflict correlation vector and the complementary correlation vector are dynamically superimposed as basis vectors of the orthogonal power grid decision space to generate a correlation matrix containing spatiotemporal game characteristics, and abnormal paths are filtered through the historical conflict paths in the dynamic mask matrix. S323. If the rate of increase of the conflict weight in the correlation matrix exceeds the dynamic decay threshold of the complementary coefficient, it is determined to be a strategy conflict zone between power plant and energy storage; if the distortion rate of the complementary correlation vector on the user-side basis vector exceeds the overload critical value of the elastic surface, it is determined to be a resource mismatch zone between user and energy storage. S324. The judgment result is spatiotemporally aligned with the pruned complementary path, the region that is repeated with the historical conflict event is removed, a dynamic graph containing only the new imbalance is generated, and the mapping weight and the update step size of the coupling matrix of the supply and demand imbalance region in the elastic surface are corrected in reverse.

6. The method for modeling power grid evolution behavior based on dynamic digital twins according to claim 1, characterized in that: The model evolution trajectory is triggered by virtual behavior anchor point disturbances, representing the dynamic state transition sequence of the power grid evolution model after being disturbed by virtual output or load disturbances in the power grid decision space. Its generation process is as follows: the trajectory offset after disturbance is separated by orthogonal basis vectors, and combined with the local resonance response of the coupling chain, a spatiotemporal evolution path including power plant output deviation, user load distortion and energy storage regulation lag is formed; the extraction of the optimal cooperative path is based on the principle of minimizing deviation, and the evolution trajectory converges to the target stability domain by reverse correction of complementary path weights and conflict thresholds.

7. The method for modeling power grid evolution behavior based on dynamic digital twins according to claim 6, characterized in that: The optimal cooperative path is injected into the adversarial evolution channel of the digital twin. Based on the independence of orthogonal basis vectors, the spatiotemporal coupling component in the trajectory offset caused by virtual behavior anchor point perturbation is separated to construct the response vector of chain resonance. Combining the boundary constraints of the target stability region, the response vector and the dynamic intensity of the coupling matrix are subjected to tensor inner product operation to generate a cooperative deviation index. If the cooperative deviation index exceeds the preset tolerance threshold, the fusion weight of the behavior coupling factor is dynamically adjusted. The dynamic mask matrix and the complementarity coefficient attenuation parameter are corrected in reverse through implicit tensor basis expansion to enhance the complementary path weight and suppress the conflict term, triggering the secondary response of local resonance of the coupling chain.

8. A power grid evolution behavior modeling system based on dynamic digital twins, capable of implementing the method described in any one of claims 1-7, characterized in that, The system includes: Behavioral coupling factor generation module: Real-time capture of power plant output characteristics, user load response patterns and energy storage operator scheduling strategies through digital twin interface, and abstract them into behavioral coupling factors, wherein the behavioral coupling factors include the potential complementarity and conflict weight of resource interaction between entities; The power grid evolution behavior model module constructs the power grid response relationship based on the behavior coupling factor. It maps the power output fluctuation of the power station to the user-side demand elastic surface through nonlinear projection, and forms a power grid decision space with the energy storage charging and discharging strategy. The connection strength between nodes in the power grid response relationship is inversely corrected by the historical conflict rate and real-time complementarity of the entity behavior. Cooperative path optimization module: By introducing virtual behavioral anchor points to disturb the power grid response relationship, a power grid evolution model is generated; the optimal cooperative path is extracted based on the deviation between the model evolution trajectory and the target stability domain, where the deviation is defined as a composite function of resource mismatch, cooperative delay cost and evolutionary inertia resistance; Feedback control module: Injects the optimal collaborative path into the adversarial evolution channel of the digital twin, detects collaborative deviations through virtual behavioral chain resonance, and dynamically adjusts the fusion weight of behavioral coupling factors until the deviation is lower than the preset tolerance threshold; Among them, the virtual behavior anchor point is an orthogonal coupled chain based on the power grid decision space. It extracts the breakpoints of resource interaction in historical evolution and cross-locates spatiotemporally heterogeneous anchor points by combining the extreme value of complementary path weight decay and the threshold of conflict path mutation. It selects the anchor point with the largest deviation from the real-time complementarity and injects virtual output or load disturbances into its associated nodes through digital twins to induce local resonance of the coupled chain. Through the independence of orthogonal basis vectors, it separates the trajectory offset after disturbance and identifies new breakpoints that are not covered by historical conflicts. The new breakpoints are fed back to the complementary path pruning rules to dynamically update the conflict threshold and verify the convergence of the modified coupled chain under the deviation metric of the power grid evolution stability domain.

Images