An auxiliary service demand calculation method based on dynamic traceability linkage

Abstract

The application provides an auxiliary service demand calculation method based on dynamic tracing linkage, and belongs to the technical field of service demand calculation. By constructing a sending-receiving end joint probability distribution model and a dynamic responsibility tracing mechanism, the application breaks through the limitations of the traditional isolated calculation mode, and realizes the fundamental change of auxiliary service demand from "isolated static" to "coordinated dynamic". By adopting a two-stage coordinated optimization model and a fair allocation mechanism based on cooperative game, the application significantly improves the calculation accuracy and economy, establishes a scientific and reasonable responsibility allocation principle, and effectively solves the cost dispute between the sending and receiving ends. The method can also enhance the adaptability of the system to new energy fluctuations, maintain the continuous applicability of the algorithm through an online rolling update mechanism, and provide reliable technical support for building a safe, economic and fair cross-regional power operation system.

Description

Technical Field

[0001] This invention belongs to the field of service demand calculation technology and also to the field of power system operation and control technology. Specifically, it relates to a method for calculating ancillary service demand based on dynamic source tracing and linkage, and particularly to a method for calculating grid ancillary service demand in the context of inter-regional power transmission. Specifically, it is a method that achieves collaborative calculation and fair allocation of ancillary service demand by quantifying the dynamic coupling relationship between the sending-end grid, the receiving-end grid, and the interconnection channels. Background Technology

[0002] With the continuous expansion of inter-regional and inter-provincial power transmission, the power system has gradually evolved into a complex interconnected system where sending-end and receiving-end power grids are tightly coupled through tie lines. Under this new structure, traditional methods for calculating ancillary service demand are increasingly revealing their inherent limitations. Existing technologies generally adopt isolated balance models based on individual power grids, treating the sending and receiving ends as independent entities for static analysis, completely ignoring the bidirectional dynamic impact of large-scale power transmission. This isolated analysis method results in the system being unable to accurately assess the power deficit risk faced by the receiving end when a sudden drop in renewable energy occurs at the sending end, and also makes it difficult to quantify the back-feeding impact of sudden load changes at the receiving end on the sending end through tie lines. Furthermore, it lacks the ability to model the correlation between sending-end and receiving-end fluctuations, leading to a serious disconnect between ancillary service configuration and actual risks.

[0003] At the mechanism design level, existing methods also have significant shortcomings. When it comes to the cost allocation of ancillary services in cross-regional transactions, existing technologies mostly adopt static mechanisms such as fixed ratios, simple power ratios, or administrative negotiations. These mechanisms neither establish a scientific basis for "responsibility tracing" and violate the basic principle of "whoever causes it, is responsible," nor can they adapt to the dynamic and time-varying characteristics of fluctuating responsibilities at the sending and receiving ends. This unreasonable design not only leads to frequent disputes over the allocation results and inhibits the enthusiasm of market participants to participate in collaboration, but also, due to the disconnect between the calculation model and the market mechanism, makes it difficult to achieve joint optimization of ancillary service procurement costs and system risk costs. Ultimately, the allocation results are either too conservative and wasteful of resources, or too aggressive and out of control of risks.

[0004] Therefore, there is an urgent need in this field for an innovative computational method that can break through the traditional isolated analysis framework, accurately quantify dynamic coupling relationships, and achieve fair economic sharing, so as to support the safe and stable operation of cross-regional power systems and the efficient operation of the market. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention presents a method for calculating ancillary service demand based on dynamic traceability and linkage. By constructing a joint probability distribution model for both the sending and receiving ends and a dynamic responsibility traceability mechanism, it overcomes the limitations of traditional isolated calculation modes, achieving a fundamental shift in ancillary service demand from "isolated and static" to "collaborative and dynamic." Employing a two-stage collaborative optimization model and a fair cost-sharing mechanism based on cooperative game theory, it significantly improves computational accuracy and economy while establishing a scientifically sound responsibility-sharing principle, effectively resolving cost disputes between the sending and receiving ends. This method also enhances the system's adaptability to renewable energy fluctuations, maintaining the algorithm's continuous applicability through an online rolling update mechanism, providing reliable technical support for building a safe, economical, and fair cross-regional power operation system.

[0006] The present invention employs the following technical solution.

[0007] A method for calculating auxiliary service demand based on dynamic source tracing and linkage includes: S1. Collect multi-source data from the sending and receiving end power grids and tie lines, calculate the net fluctuation power, construct a joint probability distribution model based on the Copula function, and generate a set of typical disturbance scenarios through Latin hypercube sampling and cluster analysis; S2. Establish a simplified physical model of the sending-end-tether-receiving end and calculate the power transfer distribution factor. For each disturbance scenario, dynamically track the wave propagation path through sensitivity analysis and accurately calculate the bidirectional responsibility factor of the sending and receiving ends; S3. Construct a two-stage optimization model: The upper layer determines the optimal system demand with the goal of minimizing total cost, while the lower layer uses cooperative game theory and the Shapley value method to fairly distribute the total demand to the sending and receiving ends. S4. Output and receiving end auxiliary service requirements scheme and collaborative invocation strategy.

[0008] Preferably, S1 specifically includes: S1.1 Perform multi-source data acquisition and preprocessing to form a regular data sequence; S1.2 Calculate the net fluctuation power based on the regularized data sequence; S1.3 Construct a joint probability distribution model based on net fluctuation power; S1.4 Generate a set of typical disturbance scenarios based on the joint probability distribution model.

[0009] Preferably, S1.1 specifically includes: Establish a complete power system operation data acquisition system to obtain the following core multi-source data from dispatch automation systems, wide-area measurement systems, and forecasting systems: ultra-short-term forecast data of new energy power generation, actual output data of new energy power generation, load forecast data and actual load data, start-up plan and real-time output of conventional units, ultra-short-term load forecast data, actual load data, local power generation plan and real-time output, planned transmission power of tie lines, real-time operating power of tie lines, rated capacity and operating limits of tie lines; The collected multi-source data is time-aligned, missing values ​​are processed, and outliers are removed to form a regular data sequence with a unified time scale.

[0010] Preferably, S1.2 specifically includes: Based on the regularized data sequence, the net fluctuating power of the sending-end grid and the receiving-end grid are calculated separately to quantify the system imbalance. Net fluctuation power of the sending-end power grid The calculation formula is:

[0011] Net fluctuation power of the receiving-end power grid The calculation formula is:

[0012] in, This represents the predicted output of new energy sources at time t in the sending-end power grid; This indicates the actual output of new energy sources in the sending-end power grid at the previous moment; This represents the predicted value of the load at time t of the sending-end power grid; This represents the actual load of the sending-end power grid at the previous moment; This represents the predicted value of the load at time t of the receiving-end power grid; This represents the actual load of the receiving-end power grid at the previous moment.

[0013] Preferably, S1.3 specifically includes: Based on historical data of net fluctuating power in the sending and receiving power grids, a joint probability distribution model of the two is constructed using the Copula function to accurately characterize the statistical dependence of fluctuations at both ends, as detailed below: (1) Fit the edge distribution of the net fluctuation power at the sending and receiving ends respectively:

[0014]

[0015] (2) Select the optimal Copula function using the AIC criterion and estimate the parameter θ using the maximum likelihood method;

[0016] in, and These are the edge cumulative distribution functions of the net fluctuating power of the sending-end and receiving-end power grids, respectively. Represents the Copula function; This represents the parameters that describe the related structure of the two.

[0017] Preferably, S1.4 specifically includes: Stratified sampling is performed in the joint probability distribution space to ensure that the samples uniformly cover the entire distribution space. The cumulative probability interval [0,1] of each variable is divided into N equally spaced intervals. A sample point is randomly selected in each interval, and the actual power fluctuation value is obtained through inverse transformation.

[0018]

[0019] in, These are points sampled from the Copula function; L-means clustering analysis reduces the massive initial scenarios to a representative set of typical scenarios, and randomly selects M cluster centers. Each scene is assigned to the nearest cluster center. :

[0020] in Let j be the j-th scene point; Recalculate cluster centers, repeat the assignment and update steps until convergence:

[0021] The probability of occurrence of each typical scenario is the proportion of the initial scenarios included in its corresponding cluster to the total number of scenarios. This is the scenario probability, representing the probability of occurrence of a typical scenario. The calculation formula is as follows:

[0022] This ultimately forms a set S of perturbation scenarios, containing typical scenarios and their probabilities of occurrence.

[0023] This represents the fluctuating power value of the sending and receiving power grids in the i-th typical scenario; This indicates the probability of the scenario occurring; This represents the total number of typical scenarios.

[0024] Preferably, S2 specifically includes: S2.1 Construct a simplified model of the sender-receiver joint system; S2.2 Calculate the power transfer distribution factor; S2.3 Calculate dynamic responsibility factors; S2.4 Construct the responsibility factor matrix.

[0025] Preferably, S2.1 specifically includes: An equivalent physical model of the sending-end power grid, tie line, and receiving-end power grid is established, with the sending-end power grid equivalent to a generator node, the receiving-end power grid equivalent to a load node, and the tie line equivalent to a transmission line, forming a two-node system model; the system admittance matrix is ​​constructed based on the actual power grid parameters.

[0026] in, , Represents the self-admittance of the sending and receiving nodes; , Indicates the mutual admittance between the sending and receiving nodes. For the equivalent conductance of the sending node, The equivalent susceptance of the sending node. The equivalent conductance of the receiving node, The equivalent susceptance of the receiving node. and All are equivalent conductances of the tie line. and All are equivalent susceptance of the tie line.

[0027] Preferably, S2.2 specifically includes: Based on the system admittance matrix, the distribution factor of the impact of power injection variations at the sending and receiving ends on the tie line power is calculated: Power transfer distribution factor from sender to tie line :

[0028] Power transfer distribution factor from the receiving end to the tie line :

[0029] in, Indicates the transmission power of the tie line; and Injected power at the sending and receiving nodes.

[0030] Preferably, S2.3 specifically includes: For each typical perturbation scenario generated by S1 Calculate the impact of fluctuations at the sending and receiving ends on the power of the tie line: Wiring power variation caused by sending-end fluctuations :

[0031] Power transfer distribution factor from the receiving end to the tie line :

[0032] in, Indicates the transmission power of the tie line; , This represents the injected power at the sending and receiving nodes.

[0033] Preferably, S2.4 specifically includes: Based on the change in tie line power, construct a dynamic responsibility factor matrix for each scenario:

[0034] The formulas for calculating each element are as follows:

[0035]

[0036]

[0037]

[0038] Where ε is a very small positive number set to avoid division by zero; This indicates the proportion of responsibility that the receiving end must bear for fluctuations at the sending end under scenario i; This indicates the proportion of responsibility that the sending end needs to bear for fluctuations at the receiving end under scenario i; , This indicates the proportion of responsibility each party bears for its own fluctuations; Output: Obtain the set of dynamic responsibility factor matrices corresponding to each typical disturbance scenario. .

[0039] Preferably, S3 specifically includes: S3.1 Constructing and solving the overall demand optimization model of the upper-level joint system; S3.2 Construct a lower-level cooperative game demand sharing model; S3.3 Calculate the Shapley value; S3.4 Perform optimal demand allocation.

[0040] Preferably, S3.1 specifically includes: (1) The comprehensive objective function considering both economy and safety:

[0041] in, and These refer to the ancillary service capacities that should be configured for the sending-end and receiving-end power grids, respectively. Let be the system loss function. Conditional risk value, For the set confidence level, Risk aversion coefficient, ancillary service cost function and Using a quadratic function form:

[0042]

[0043] in and The coefficient of the quadratic term, and The coefficient of the linear term, and For constant terms; The system loss function characterizes the expected power shortage loss caused by insufficient ancillary services:

[0044] in, Indicates the value of load loss; (2) Constraint system System power balance constraints:

[0045] in, This refers to the inherent regulating capability of the system, including natural regulating effects such as load frequency characteristics; Frequency stability safety constraints:

[0046] in, and The equivalent frequency response coefficients of the transmitting and receiving system are given. The maximum allowable frequency deviation; The tie-line transmission capacity safety constraint ensures that the tie-line power does not exceed the limit in various scenarios. The constraint conditions are as follows:

[0047] in For the planned transmission power of the tie line, This represents the maximum permissible transmission capacity of the tie line; Ancillary service technology constraints:

[0048]

[0049] in, and The technical minimums for configuring auxiliary service capacity for both the sending and receiving ends are specified. and Configure the technical limits for auxiliary service capacity for both the sending and receiving ends. This is the minimum backup capacity requirement for the system.

[0050] Preferably, S3.2 specifically includes: Characteristic function of independent operation of the sending-end power grid:

[0051]

[0052]

[0053] Characteristic function of independent operation of the receiving-end power grid:

[0054]

[0055]

[0056] Joint operation characteristic function of sender and receiver: .

[0057] Preferably, S3.3 specifically includes: The Shapeli value of the sending-end power grid:

[0058] The Shapeli value of the receiving-end power grid:

[0059] Where, v( )=0 indicates the characteristic function value of the air alliance.

[0060] Preferably, S3.4 specifically includes: Optimal ancillary service requirements of the sending-end power grid :

[0061] Optimal ancillary service requirements of receiving-end power grid :

[0062] in, The optimal total demand of the joint system is solved for the upper-level model; and These represent the basic ancillary service needs of the sending and receiving power grids, respectively. and These represent the Shapley values ​​at the sending and receiving ends, respectively.

[0063] Preferably, S4 specifically includes: S4.1 Generate auxiliary service requirement scheme; S4.2 Build a cross-regional collaborative invocation strategy library; S4.3 Design an online rolling update mechanism.

[0064] Preferably, S4.1 specifically includes: Quota for ancillary services required by the sending and receiving ends: .

[0065] Preferably, S4.2 specifically includes: Strategy trigger condition definition: For each scenario, set a trigger threshold. :

[0066] in and The trigger threshold for the responsibility factor. and As a responsibility factor; Cooperative Invocation Strategy Matrix :

[0067] Where ψij represents the calling coefficient of resource j in scenario i.

[0068] Preferably, S4.3 specifically includes: (1) Scene set dynamic update:

[0069] in, Forgetting factor, A set of scenarios generated from newly collected data; (2) Online correction of responsibility factors:

[0070]

[0071] in, For learning rate, The loss function; (3) Adaptive cost function parameters:

[0072] Where θ is the cost function parameter and γ is the fitness coefficient.

[0073] The beneficial effects of the present invention are as follows, compared with the prior art: By constructing a joint probability distribution model for both the sending and receiving ends and a dynamic responsibility tracing mechanism, this method overcomes the limitations of traditional isolated calculation models, achieving a fundamental shift in ancillary service demand from "isolated and static" to "collaborative and dynamic." Employing a two-stage collaborative optimization model and a fair cost-sharing mechanism based on cooperative game theory, it significantly improves computational accuracy and economy while establishing a scientifically sound responsibility-sharing principle, effectively resolving cost disputes between the sending and receiving ends. This method also enhances the system's adaptability to renewable energy fluctuations, maintaining the algorithm's continuous applicability through an online rolling update mechanism, providing reliable technical support for building a safe, economical, and fair cross-regional power operation system. Attached Figure Description

[0074] Figure 1 This is a flowchart of the auxiliary service demand calculation method based on dynamic traceability and linkage in this invention. Detailed Implementation

[0075] like Figure 1 As shown, a method for calculating auxiliary service requirements based on dynamic source tracing and linkage includes: S1. Collect multi-source data from the sending and receiving power grids and tie lines, calculate the net fluctuation power, construct a joint probability distribution model based on the Copula function, and generate a set of typical disturbance scenarios through Latin hypercube sampling and cluster analysis to provide a data foundation for subsequent analysis; In a preferred but non-limiting embodiment of the present invention, S1 specifically includes: S1.1 Perform multi-source data acquisition and preprocessing to form a regular data sequence; In a preferred but non-limiting embodiment of the present invention, S1.1 specifically includes: Establish a complete power system operation data acquisition system to obtain the following core multi-source data from dispatch automation systems, wide-area measurement systems, and forecasting systems: ultra-short-term forecast data of new energy power generation, actual output data of new energy power generation, load forecast data and actual load data, start-up plan and real-time output of conventional units, ultra-short-term load forecast data, actual load data, local power generation plan and real-time output, planned transmission power of tie lines, real-time operating power of tie lines, rated capacity and operating limits of tie lines; The collected multi-source data is time-aligned, missing values ​​are processed, and outliers are removed to form a regular data sequence with a unified time scale.

[0076] S1.2 Calculate the net fluctuation power based on the regularized data sequence; In a preferred but non-limiting embodiment of the present invention, S1.2 specifically includes: Based on the regularized data sequence, the net fluctuating power of the sending-end grid and the receiving-end grid are calculated separately to quantify the system imbalance. Net fluctuation power of the sending-end power grid The calculation formula is:

[0077] Net fluctuation power of the receiving-end power grid The calculation formula is:

[0078] in, This represents the predicted output of new energy sources at time t in the sending-end power grid; This indicates the actual output of new energy sources in the sending-end power grid at the previous moment; This represents the predicted value of the load at time t of the sending-end power grid; This represents the actual load of the sending-end power grid at the previous moment; This represents the predicted value of the load at time t of the receiving-end power grid; This represents the actual load of the receiving-end power grid at the previous moment.

[0079] when A value >0 indicates that the actual power supply capacity of the sending-end power grid is lower than expected or the actual load is higher than expected; when When <0, it indicates that the actual power supply capacity of the sending-end power grid is higher than expected or the actual load is lower than expected; when When the value is greater than 0, it indicates that the load at the receiving end is lower than expected, and the demand for power at the sending end is reduced; when When the value is less than 0, it indicates that the load at the receiving end is higher than expected, and the demand for power at the sending end is increasing.

[0080] S1.3 Construct a joint probability distribution model based on net fluctuation power; In a preferred but non-limiting embodiment of the invention, S1.3 specifically includes: Based on historical data of net fluctuating power in the sending and receiving power grids, a joint probability distribution model of the two is constructed using the Copula function to accurately characterize the statistical dependence of fluctuations at both ends, as detailed below: (1) Fit the edge distribution of the net fluctuation power at the sending and receiving ends respectively:

[0081]

[0082] (2) Select the optimal Copula function using the AIC criterion and estimate the parameter θ using the maximum likelihood method;

[0083] in, and These are the edge cumulative distribution functions of the net fluctuating power of the sending-end and receiving-end power grids, respectively. Represents the Copula function; This represents the parameters that describe the related structure of the two.

[0084] S1.4 Generate a set of typical disturbance scenarios based on the joint probability distribution model.

[0085] In a preferred but non-limiting embodiment of the invention, S1.4 specifically includes: Stratified sampling is performed in the joint probability distribution space to ensure that the samples uniformly cover the entire distribution space. The cumulative probability interval [0,1] of each variable is divided into N equally spaced intervals. A sample point is randomly selected in each interval, and the actual power fluctuation value is obtained through inverse transformation.

[0086]

[0087] in, These are points sampled from the Copula function; M-means clustering analysis reduces the massive initial scenarios to a representative set of typical scenarios, and randomly selects M cluster centers. Each scene is assigned to the nearest cluster center. :

[0088] in Let j be the j-th scene point; Recalculate cluster centers, repeat the assignment and update steps until convergence:

[0089] The probability of occurrence of each typical scenario is the proportion of the initial scenarios included in its corresponding cluster to the total number of scenarios. This is the scenario probability, representing the probability of occurrence of a typical scenario. The calculation formula is as follows:

[0090] This ultimately results in a set of disturbance scenarios, S, containing typical scenarios and their probabilities of occurrence. This set of typical disturbance scenarios serves as the input for the "dynamic responsibility attribution" step S2, used to analyze the responsibility allocation relationship under each specific fluctuation scenario. The probabilistic characteristics of the scenario set provide the foundation for risk quantification in step S3, ensuring that the calculation of ancillary service requirements considers both typical operating conditions and extreme disturbance scenarios.

[0091] This represents the fluctuating power value of the sending and receiving power grids in the i-th typical scenario; This indicates the probability of the scenario occurring; This represents the total number of typical scenarios.

[0092] S2. Establish a simplified physical model of the sending-end-connection-receiving end and calculate the power transfer distribution factor. For each disturbance scenario, dynamically track the wave propagation path through sensitivity analysis, accurately calculate the bidirectional responsibility factor of the sending and receiving ends, and achieve scientific quantification of cross-boundary wave responsibility; In a preferred but non-limiting embodiment of the present invention, S2 specifically includes: S2.1 Construct a simplified model of the sender-receiver joint system; In a preferred but non-limiting embodiment of the present invention, S2.1 specifically includes: An equivalent physical model of the sending-end power grid, tie line, and receiving-end power grid is established, with the sending-end power grid equivalent to a generator node, the receiving-end power grid equivalent to a load node, and the tie line equivalent to a transmission line, forming a two-node system model; the system admittance matrix is ​​constructed based on the actual power grid parameters.

[0093] in, , Represents the self-admittance of the sending and receiving nodes; , Indicates the mutual admittance between the sending and receiving nodes. For the equivalent conductance of the sending node, The equivalent susceptance of the sending node. The equivalent conductance of the receiving node, The equivalent susceptance of the receiving node. and All are equivalent conductances of the tie line. and All are equivalent susceptance of the tie line.

[0094] S2.2 Calculate the power transfer distribution factor; In a preferred but non-limiting embodiment of the present invention, S2.2 specifically includes: Based on the system admittance matrix, the distribution factor of the impact of power injection variations at the sending and receiving ends on the tie line power is calculated: Power transfer distribution factor from sender to tie line :

[0095] Power transfer distribution factor from the receiving end to the tie line :

[0096] in, Indicates the transmission power of the tie line; and Injected power at the sending and receiving nodes.

[0097] S2.3 Calculate dynamic responsibility factors; In a preferred but non-limiting embodiment of the present invention, S2.3 specifically includes: For each typical perturbation scenario generated by S1 Calculate the impact of fluctuations at the sending and receiving ends on the power of the tie line: Wiring power variation caused by sending-end fluctuations :

[0098] Power transfer distribution factor from the receiving end to the tie line :

[0099] in, Indicates the transmission power of the tie line; , This represents the injected power at the sending and receiving nodes.

[0100] S2.4 Construct the responsibility factor matrix.

[0101] In a preferred but non-limiting embodiment of the present invention, S2.4 specifically includes: Based on the change in tie line power, construct a dynamic responsibility factor matrix for each scenario:

[0102] The formulas for calculating each element are as follows:

[0103]

[0104]

[0105]

[0106] Where ε is a very small positive number set to avoid division by zero; This indicates the proportion of responsibility that the receiving end must bear for fluctuations at the sending end under scenario i; This indicates the proportion of responsibility that the sending end needs to bear for fluctuations at the receiving end under scenario i; , This indicates the proportion of responsibility each party bears for its own fluctuations; Output: Obtain the set of dynamic responsibility factor matrices corresponding to each typical disturbance scenario. This provides a quantitative basis for optimizing the allocation of subsequent ancillary service needs.

[0107] S3. Construct a two-stage optimization model: The upper layer determines the optimal system demand with the goal of minimizing total cost, while the lower layer uses cooperative game theory and the Shapley value method to fairly distribute the total demand to the sending and receiving ends, thereby achieving a balance between economy, security and fairness. In a preferred but non-limiting embodiment of the present invention, S3 specifically includes: This approach organically combines determining the optimal total demand of the joint system with the issue of fair resource allocation among the participants. Through the synergy of upper-level optimization and lower-level game theory, it achieves a balance between economy, security, and fairness. The specific implementation process includes the following four key stages: S3.1 Constructing and solving the overall demand optimization model of the upper-level joint system; In a preferred but non-limiting embodiment of the present invention, S3.1 specifically includes: The upper-level optimization model is based on the perspective of the entire sending and receiving end integrated system, aiming to minimize the total system operating cost while taking into account both economy and security. This model comprehensively considers the direct procurement costs of ancillary services as well as the system operation risk costs caused by insufficient ancillary services.

[0108] (1) The comprehensive objective function considering both economy and safety:

[0109] in, and These are the ancillary service capacities (unit: MW) that should be configured for the sending-end and receiving-end power grids, respectively. Let be the system loss function. Conditional risk value, For the set confidence level, Risk aversion coefficient, ancillary service cost function and Using a quadratic function form:

[0110]

[0111] in and The coefficient of the quadratic term, and The coefficient of the linear term, and For constant terms; The system loss function characterizes the expected power shortage loss caused by insufficient ancillary services:

[0112] in, Indicates the value of load loss; (2) Constraint system System power balance constraints:

[0113] in, This refers to the inherent regulating capability of the system, including natural regulating effects such as load frequency characteristics; Frequency stability safety constraints:

[0114] in, and The equivalent frequency response coefficients of the transmitting and receiving system are given. The maximum allowable frequency deviation; The tie-line transmission capacity safety constraint ensures that the tie-line power does not exceed the limit in various scenarios. The constraint conditions are as follows:

[0115] in For the planned (or initial) transmission power of the tie line, This represents the maximum permissible transmission capacity of the tie line; Ancillary service technology constraints:

[0116]

[0117] in, and The technical minimums for configuring auxiliary service capacity for both the sending and receiving ends are specified. and Configure the technical limits for auxiliary service capacity for both the sending and receiving ends. This is the minimum backup capacity requirement for the system.

[0118] S3.2 Construct a lower-level cooperative game demand sharing model; In a preferred but non-limiting embodiment of the present invention, S3.2 specifically includes: The lower-level model is based on cooperative game theory. It quantifies the contribution of each participant in the cooperation by defining characteristic functions for different alliance forms.

[0119] Characteristic function of independent operation of the sending-end power grid:

[0120]

[0121]

[0122] Characteristic function of independent operation of the receiving-end power grid:

[0123]

[0124]

[0125] Joint operation characteristic function of sender and receiver: .

[0126] S3.3 Calculate the Shapley value; In a preferred but non-limiting embodiment of the present invention, S3.3 specifically includes: Based on cooperative game theory, the Shapley value method is used for demand allocation, which can ensure the fairness and rationality of the allocation results.

[0127] The Shapeli value of the sending-end power grid:

[0128] The Shapeli value of the receiving-end power grid:

[0129] Where, v( )=0 indicates the characteristic function value of the air alliance.

[0130] S3.4 Perform optimal demand allocation.

[0131] In a preferred but non-limiting embodiment of the present invention, S3.4 specifically includes: The final demand allocation is based on the Shapley value to ensure that the allocation result satisfies both economic efficiency and fairness.

[0132] Optimal ancillary service requirements of the sending-end power grid :

[0133] Optimal ancillary service requirements of receiving-end power grid :

[0134] in, The optimal total demand of the joint system is solved for the upper-level model; and These represent the basic ancillary service needs of the sending and receiving power grids, respectively. and These represent the Shapley values ​​at the sending and receiving ends, respectively.

[0135] S4. Output and receiving end auxiliary service requirements scheme and collaborative invocation strategy.

[0136] In a preferred but non-limiting embodiment of the present invention, S4 specifically includes: Establish an online rolling update mechanism to calibrate model parameters and scenario sets in real time, forming a closed-loop system with self-improvement capabilities to ensure the continuous applicability of the method.

[0137] This step is the final output stage and a guarantee for continuous optimization of the entire method. It aims to transform the computational results of the first three steps into executable scheduling instructions and ensure that the method can adapt to the time-varying characteristics of power grid operation by establishing a dynamic update mechanism. The specific implementation process includes the following key steps: S4.1 Generate auxiliary service requirement scheme; In a preferred but non-limiting embodiment of the present invention, S4.1 specifically includes: Based on the optimal demand allocation results obtained in phase S3, a standardized ancillary service demand scheme is generated. This scheme includes the following core components: Quota for ancillary services required by the sending and receiving ends: .

[0138] S4.2 Build a cross-regional collaborative invocation strategy library; In a preferred but non-limiting embodiment of the present invention, S4.2 specifically includes: For typical disturbance scenarios identified in the S1 phase, corresponding collaborative invocation strategies are formulated to form a complete strategy knowledge base: Strategy trigger condition definition: For each scenario, set a trigger threshold. :

[0139] in and The trigger threshold for the responsibility factor. and As a responsibility factor; Cooperative Invocation Strategy Matrix :

[0140] Where ψij represents the calling coefficient of resource j in scenario i.

[0141] S4.3 Design an online rolling update mechanism.

[0142] In a preferred but non-limiting embodiment of the present invention, S4.3 specifically includes: Establish a time-triggered model parameter update mechanism to ensure that the calculation model always remains consistent with the actual operating state of the power grid: Real-time parameter calibration algorithm: (1) Scene set dynamic update:

[0143] in, Forgetting factor, A set of scenarios generated from newly collected data; (2) Online correction of responsibility factors:

[0144]

[0145] in, For learning rate, The loss function; (3) Adaptive cost function parameters:

[0146] Where θ is the cost function parameter and γ is the fitness coefficient.

[0147] The beneficial effects of the present invention are as follows, compared with the prior art: By constructing a joint probability distribution model for both the sending and receiving ends and a dynamic responsibility tracing mechanism, this method overcomes the limitations of traditional isolated calculation models, achieving a fundamental shift in ancillary service demand from "isolated and static" to "collaborative and dynamic." Employing a two-stage collaborative optimization model and a fair cost-sharing mechanism based on cooperative game theory, it significantly improves computational accuracy and economy while establishing a scientifically sound responsibility-sharing principle, effectively resolving cost disputes between the sending and receiving ends. This method also enhances the system's adaptability to renewable energy fluctuations, maintaining the algorithm's continuous applicability through an online rolling update mechanism, providing reliable technical support for building a safe, economical, and fair cross-regional power operation system.

[0148] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention without departing from the spirit and scope of the present invention. Any modifications or equivalent substitutions should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for calculating auxiliary service demand based on dynamic source tracing and linkage, characterized in that, include: S1. Collect multi-source data from the sending and receiving end power grids and tie lines, calculate the net fluctuation power, construct a joint probability distribution model based on the Copula function, and generate a set of typical disturbance scenarios through Latin hypercube sampling and cluster analysis; S2. Establish a simplified physical model of the sending-end-tether-receiving end and calculate the power transfer distribution factor. For each disturbance scenario, dynamically track the wave propagation path through sensitivity analysis and accurately calculate the bidirectional responsibility factor of the sending and receiving ends; S3. Construct a two-stage optimization model: The upper layer determines the optimal system demand with the goal of minimizing total cost, while the lower layer uses cooperative game theory and the Shapley value method to fairly distribute the total demand to the sending and receiving ends. S4. Output and receiving end auxiliary service requirements scheme and collaborative invocation strategy.

2. The auxiliary service demand calculation method based on dynamic traceability and linkage according to claim 1, characterized in that, S1 specifically includes: S1.1 Perform multi-source data acquisition and preprocessing to form a regular data sequence; S1.2 Calculate the net fluctuation power based on the regularized data sequence; S1.3 Construct a joint probability distribution model based on net fluctuation power; S1.4 Generate a set of typical disturbance scenarios based on the joint probability distribution model.

3. The auxiliary service demand calculation method based on dynamic traceability and linkage according to claim 2, characterized in that, S1.1 specifically includes: Establish a complete power system operation data acquisition system to obtain the following core multi-source data from dispatch automation systems, wide-area measurement systems, and forecasting systems: ultra-short-term forecast data of new energy power generation, actual output data of new energy power generation, load forecast data and actual load data, start-up plan and real-time output of conventional units, ultra-short-term load forecast data, actual load data, local power generation plan and real-time output, planned transmission power of tie lines, real-time operating power of tie lines, rated capacity and operating limits of tie lines; The collected multi-source data is time-aligned, missing values ​​are processed, and outliers are removed to form a regular data sequence with a unified time scale; S1.2 specifically includes: Based on the regularized data sequence, the net fluctuating power of the sending-end grid and the receiving-end grid are calculated separately to quantify the system imbalance. Net fluctuation power of the sending-end power grid The calculation formula is: Net fluctuation power of the receiving-end power grid The calculation formula is: in, This represents the predicted output of new energy sources at time t in the sending-end power grid; This indicates the actual output of new energy sources in the sending-end power grid at the previous moment; This represents the predicted value of the load at time t of the sending-end power grid; This represents the actual load of the sending-end power grid at the previous moment; This represents the predicted value of the load at time t of the receiving-end power grid; This represents the actual load of the receiving-end power grid at the previous moment; S1.3 specifically includes: Based on historical data of net fluctuating power in the sending and receiving power grids, a joint probability distribution model of the two is constructed using the Copula function to accurately characterize the statistical dependence of fluctuations at both ends, as detailed below: (1) Fit the edge distribution of the net fluctuation power at the sending and receiving ends respectively: (2) Select the optimal Copula function using the AIC criterion and estimate the parameter θ using the maximum likelihood method; in, and These are the edge cumulative distribution functions of the net fluctuating power of the sending-end and receiving-end power grids, respectively. Represents the Copula function; This represents the parameters that describe the related structure of the two; S1.4 specifically includes: Stratified sampling is performed in the joint probability distribution space to ensure that the samples uniformly cover the entire distribution space. The cumulative probability interval [0,1] of each variable is divided into N equally spaced intervals. A sample point is randomly selected in each interval, and the actual power fluctuation value is obtained through inverse transformation. in, These are points sampled from the Copula function; K-means clustering analysis reduces the massive initial scenarios to a representative set of typical scenarios, and randomly selects M cluster centers. Each scene is assigned to the nearest cluster center. : in Let j be the j-th scene point; Recalculate cluster centers, repeat the assignment and update steps until convergence: The probability of occurrence of each typical scenario is the proportion of the initial scenarios included in its corresponding cluster to the total number of scenarios. This is the scenario probability, representing the probability of occurrence of a typical scenario. The calculation formula is as follows: This ultimately forms a set S of perturbation scenarios, containing typical scenarios and their probabilities of occurrence. This represents the fluctuating power value of the sending and receiving power grids in the i-th typical scenario; This indicates the probability of the scenario occurring; This represents the total number of typical scenarios.

4. The auxiliary service demand calculation method based on dynamic traceability and linkage according to claim 3, characterized in that, S2 specifically includes: S2.1 Construct a simplified model of the sender-receiver joint system; S2.2 Calculate the power transfer distribution factor; S2.3 Calculate dynamic responsibility factors; S2.4 Construct the responsibility factor matrix.

5. The auxiliary service demand calculation method based on dynamic traceability and linkage according to claim 4, characterized in that, S2.1 specifically includes: An equivalent physical model of the sending-end power grid, tie line, and receiving-end power grid is established, with the sending-end power grid equivalent to a generator node, the receiving-end power grid equivalent to a load node, and the tie line equivalent to a transmission line, forming a two-node system model; the system admittance matrix is ​​constructed based on the actual power grid parameters. in, , Represents the self-admittance of the sending and receiving nodes; , Indicates the mutual admittance between the sending and receiving nodes. For the equivalent conductance of the sending node, The equivalent susceptance of the sending node. The equivalent conductance of the receiving node, The equivalent susceptance of the receiving node. and All are equivalent conductances of the tie line. and All are equivalent susceptance of the tie line; S2.2 specifically includes: Based on the system admittance matrix, the distribution factor of the impact of power injection variations at the sending and receiving ends on the tie line power is calculated: Power transfer distribution factor from sender to tie line : Power transfer distribution factor from the receiving end to the tie line : in, Indicates the transmission power of the tie line; and Injected power at the sending and receiving nodes; S2.3 specifically includes: For each typical perturbation scenario generated by S1 Calculate the impact of fluctuations at the sending and receiving ends on the power of the tie line: Wiring power variation caused by sending-end fluctuations : Power transfer distribution factor from the receiving end to the tie line : in, Indicates the transmission power of the tie line; , This indicates the injected power at the sending and receiving nodes; S2.4 specifically includes: Based on the change in tie line power, construct a dynamic responsibility factor matrix for each scenario: The formulas for calculating each element are as follows: Where ε is a very small positive number set to avoid division by zero; This indicates the proportion of responsibility that the receiving end must bear for fluctuations at the sending end under scenario i; This indicates the proportion of responsibility that the sending end needs to bear for fluctuations at the receiving end under scenario i; , This indicates the proportion of responsibility each party bears for its own fluctuations; Output: Obtain the set of dynamic responsibility factor matrices corresponding to each typical disturbance scenario. .

6. The auxiliary service demand calculation method based on dynamic traceability and linkage according to claim 5, characterized in that, S3 specifically includes: S3.1 Constructing and solving the overall demand optimization model of the upper-level joint system; S3.2 Construct a lower-level cooperative game demand sharing model; S3.3 Calculate the Shapley value; S3.4 Perform optimal demand allocation.

7. The auxiliary service demand calculation method based on dynamic traceability and linkage according to claim 6, characterized in that, S3.1 specifically includes: (1) The comprehensive objective function considering both economy and safety: in, and These refer to the ancillary service capacities that should be configured for the sending-end and receiving-end power grids, respectively. Let be the system loss function. Conditional risk value, For the set confidence level, Risk aversion coefficient, ancillary service cost function and Using a quadratic function form: in and The coefficient of the quadratic term, and The coefficient of the linear term, and For constant terms; The system loss function characterizes the expected power shortage loss caused by insufficient ancillary services: in, Indicates the value of load loss; (2) Constraint system System power balance constraints: in, This refers to the inherent regulating capability of the system, including natural regulating effects such as load frequency characteristics; Frequency stability safety constraints: in, and The equivalent frequency response coefficients of the transmitting and receiving system are given. The maximum allowable frequency deviation; The tie-line transmission capacity safety constraint ensures that the tie-line power does not exceed the limit in various scenarios. The constraint conditions are as follows: in For the planned transmission power of the tie line, This represents the maximum permissible transmission capacity of the tie line; Ancillary service technology constraints: in, and The technical minimums for configuring auxiliary service capacity for both the sending and receiving ends. and Configure the technical limits for auxiliary service capacity for both the sending and receiving ends. This is the minimum backup capacity requirement for the system; S3.2 specifically includes: Characteristic function of independent operation of the sending-end power grid: Characteristic function of independent operation of the receiving-end power grid: Joint operation characteristic function of sender and receiver: ； S3.3 specifically includes: The Shapeli value of the sending-end power grid: The Shapeli value of the receiving-end power grid: Where, v( )=0 indicates the characteristic function value of the air alliance; S3.4 specifically includes: Optimal ancillary service requirements of the sending-end power grid : Optimal ancillary service requirements of receiving-end power grid : in, The optimal total demand of the joint system is solved for the upper-level model; and These represent the basic ancillary service needs of the sending and receiving power grids, respectively. and These represent the Shapley values ​​at the sending and receiving ends, respectively.

8. The auxiliary service demand calculation method based on dynamic traceability and linkage according to claim 7, characterized in that, S4 specifically includes: S4.1 Generate auxiliary service requirement scheme; S4.2 Build a cross-regional collaborative invocation strategy library; S4.3 Design an online rolling update mechanism.

9. The auxiliary service demand calculation method based on dynamic traceability and linkage according to claim 8, characterized in that, S4.1 specifically includes: Quota for ancillary services required by the sending and receiving ends: ； S4.2 specifically includes: Strategy trigger condition definition: For each scenario, set a trigger threshold. : in and The trigger threshold for the responsibility factor. and As a responsibility factor; Cooperative Invocation Strategy Matrix : Where ψij represents the calling coefficient of resource j in scenario i.

10. The auxiliary service demand calculation method based on dynamic traceability and linkage according to claim 9, characterized in that, S4.3 specifically includes: (1) Scene set dynamic update: in, Forgetting factor, A set of scenarios generated from newly collected data; (2) Online correction of responsibility factors: in, For learning rate, The loss function; (3) Adaptive cost function parameters: Where θ is the cost function parameter and γ is the fitness coefficient.

Images