An intelligent collaborative decision-making method for a power distribution system

Abstract

The application relates to the technical field of power distribution systems, in particular to a power distribution system intelligent collaborative decision-making method, which comprises the following steps: S1, collecting the sampling frequency, uploading delay and data integrity identification of each sub-region sensing node, analyzing the consistency of operation data structure, identifying time alignment defects or logical coupling faults, and generating a heterogeneous drift feature map; S2, combining historical data, topological mechanisms and boundary rules, reconstructing control input, and generating a data consistency control input set; S3, based on the input set, scheduling targets and system constraints, constructing and distributing a collaborative strategy, and realizing intelligent collaborative regulation and control. According to the application, the heterogeneous drift feature map is constructed, the spatio-temporal consistency operation data is reconstructed, and the collaborative strategy set with adaptive execution capability is generated, so that the accurate identification, effective repair and intelligent collaborative regulation and control of the power distribution system under the condition of inconsistent data are realized.

Description

Technical Field

[0001] This invention relates to the field of power distribution system technology, and in particular to an intelligent collaborative decision-making method for power distribution systems. Background Technology

[0002] With the rapid development of new power systems, distributed power sources, energy storage devices and multi-level loads are constantly being connected to the distribution network. The traditional distribution system operation mode, which is mainly based on central dispatch, is facing the need for intelligent and collaborative transformation. In order to improve the system's adaptability to multi-source heterogeneous operating environments, more and more intelligent sensing nodes are being deployed in various sub-regions of the distribution system to achieve refined monitoring of operating status and inter-regional linkage control, and to assist in building a flexible and efficient collaborative control mechanism.

[0003] However, existing distributed sensing devices in power distribution systems suffer from problems such as inconsistent sampling frequencies, unstable data upload delays, and packet loss. This leads to drift deviations in the time dimension and logical structure of sensing data from different sub-regions, making it difficult to directly use them as effective inputs for control strategies. Traditional strategy generation methods often rely on idealized synchronous data assumptions and lack the ability to identify and repair heterogeneous data drift states, which can easily lead to strategy mismatch and control failure. This is especially true at the boundaries of multiple regions, where coupling breaks are more likely to occur, seriously affecting the global coordination and robustness of the system. Summary of the Invention

[0004] This invention provides an intelligent collaborative decision-making method for power distribution systems. By constructing a heterogeneous drift feature map, reconstructing spatiotemporal consistency control inputs, and generating a set of collaborative strategies with adaptability and autonomy based on scheduling objectives and system constraints, the method achieves highly reliable intelligent collaborative control of power distribution systems in scenarios with inconsistent data, significantly improving the accuracy of strategy generation and the stability of execution.

[0005] A method for intelligent collaborative decision-making in a power distribution system includes the following steps:

[0006] S1 collects operational data from multiple sub-regions in the power distribution system, including the sampling frequency, upload delay, and data integrity identifier of the sensing nodes. By performing structural consistency analysis on the operational data of each sub-region, heterogeneous drift segments with time alignment defects or logical coupling faults are identified, and heterogeneous drift feature maps characterizing the data drift features of each region are generated.

[0007] S2, based on the generated heterogeneous drift feature map, combined with historical redundant data, topology completion mechanism and physical boundary prior rules, dynamically generates a reconstructed consistent version of the control input for each sub-region, forming a data consistency control input set for policy generation;

[0008] S3, using the data consistency control input set as the basis for strategy generation, and combining the current scheduling target and system constraints, construct a collaborative strategy set, and distribute the collaborative strategy set to each sub-region control unit to realize intelligent collaborative control of the power distribution system under data inconsistency.

[0009] Optionally, S1 includes:

[0010] S11 collects operational data from multiple sub-regions in the power distribution system. The operational data includes the sampling frequency, upload delay, and data integrity identifier of the sensing nodes, and performs standardization processing on the operational data from different sub-regions.

[0011] S12. After completing the standardized collection of operational data, based on the data organization structure, time window alignment degree and logical topological connectivity between sub-regions, structural consistency analysis is performed on the operational data of each sub-region. Heterogeneous drift segments with time alignment defects or logical coupling faults are identified, and the spatial distribution, drift type and influence range of heterogeneous drift segments are extracted to construct a heterogeneous drift feature map reflecting the data consistency status of sub-regions.

[0012] Optionally, S11 includes:

[0013] S111, for each sub-area R in the power distribution system k (where k = 1, 2, ..., K, and K is the total number of regions), collecting data from N sensing nodes deployed in each region. k,m The running data output by (the m-th node in the k-th region) includes the sampling frequency f. k,m Upload delay Δ k,m Data integrity identifier γ k,m (γ k,m =1 indicates that the data within the current time window is complete, γ k,m =0 indicates packet loss or incomplete upload), and for each sub-region R k The original running dataset D k ;

[0014] S112, for all sub-regions R k With a unified observation time window T obs and sampling reference frequency f ref As a structural alignment reference, a node-aware state matrix S is constructed. k ;

[0015] S113, standardize the operational data of each sensing node to obtain a standard operational data vector, and combine the standard operational data vectors of all sensing nodes to form a standard operational dataset for the sub-region.

[0016] Optionally, S12 includes:

[0017] S121, after completing the standardized collection of operational data, for each sub-region R k Temporal features are extracted from the upload behavior of sensing nodes. By calculating the upload time centroid and overall missing rate of sensing nodes, the time offset and data loss within the standard time window are analyzed. Simultaneously, logically adjacent region pairs (R... k ,R u The temporal centroid difference and alignment score between the regions are calculated to identify temporally misaligned pairs, forming a temporal drift candidate set Ω. time ;

[0018] S122 analyzes the cooperative response of sub-regions in terms of logical topology, assesses whether there is a coupling failure, calculates the consistency between the expected cooperative level and the actual level by sensing the average upload behavior of nodes and combining the power topology coupling weights between sub-regions, and evaluates whether the residual exceeds the residual judgment threshold, identifies logical fault region pairs, and constructs a logical drift candidate set Ω. logic ;

[0019] S123, the identified time-type drift candidate set Ω time And logic class drift candidate set Ω logic After merging, the spatial affected range, number of drift types, and number of adjacent faults of each region are extracted, and finally a heterogeneous drift feature map Ψ is constructed.

[0020] Optionally, S2 includes:

[0021] S21, for sub-regions with time alignment defects in heterogeneous drift feature maps, identify upload delays, missing segments and sampling misalignment issues in the running data, use historical redundant data for interpolation to complete, and unify the sampling rhythm by referring to the time step of adjacent regions to achieve time consistency processing of running data.

[0022] S22. After completing the time consistency processing, for the sub-regions with logical coupling faults in the heterogeneous drift feature map, analyze the relationship between the power flow and control response between them and the adjacent regions, infer the missing control information, and combine the physical boundary conditions (load capacity and equipment limitations) to correct the operating data after time consistency processing, and generate a consistent control input set with logical closure.

[0023] Optionally, S21 includes:

[0024] S211, for the sub-regions marked with time alignment defects in the heterogeneous drift feature map, based on the upload behavior matrix of each sensing node, features of upload delay, data missingness, and sampling misalignment are extracted to identify the set of data segments to be corrected.

[0025] S212, for the identified set of data segments to be corrected By utilizing historical redundant data and employing interpolation methods to fill in missing points, the continuous observation sequence of the sensing node within the time window T is reconstructed.

[0026] S213, after completing the interpolation, remap the sensing nodes with inconsistent sampling frequencies in each sub-region to the reference time step Δt. ref .

[0027] Optionally, S22 includes:

[0028] S221, after completing the time consistency processing, based on the regions marked as having logical coupling faults in the heterogeneous drift feature map (R... k ,R u This study analyzes the power exchange behavior and control response state of sensing nodes within a standard observation time window, identifies pairs of sensing nodes with discontinuous power direction, lost response information, or untransmitted boundary signals, and constructs a missing control information index set C. k,u ;

[0029] S222, for the missing control information index set C k,u To fill the missing operational data in the data, proxy operational data is generated based on the operational behavior of similar nodes in adjacent regions within the same time window, using a response similarity supplementation method.

[0030] S223, for the supplementary proxy operation data, based on the physical boundary conditions of the maximum or minimum operating capacity, load adjustment range, and power response upper limit of the equipment to which the sensing node belongs, upper and lower limit constraints are corrected, and finally a consistent control input set with logical closure is generated.

[0031] Optionally, S3 includes:

[0032] S31. After obtaining the consistency control input set of each sub-region, construct a set of cooperative strategies based on the current scheduling objectives and system constraints (equipment capacity limits, network topology status, tie-line power boundaries).

[0033] S32 decomposes and distributes the generated collaborative strategy set at the regional granularity. Based on the current operating status, boundary conditions and execution capabilities of each sub-region, it automatically selects the matching sub-strategy version to achieve local autonomous execution.

[0034] Optionally, S31 includes:

[0035] S311 After obtaining the consistency control input set of each sub-region, the comprehensive scheduling objective function is set according to the current operating scenario. The objectives include minimizing total active power loss, minimizing node voltage deviation, and optimizing load adjustable response.

[0036] S312 introduces system constraints based on the set comprehensive scheduling objective function, including equipment capacity limits, network topology status (power balance), and tie-line power flow boundaries.

[0037] S313, based on the scheduling objective function and system constraints, outputs a set of cooperative strategies.

[0038] Optionally, S32 includes:

[0039] S321, based on the generated set of cooperative strategies According to sub-region numbering The collaborative strategy set is divided into multiple regional sub-strategy sets, and each sub-strategy set corresponds to the control instructions for all sensing nodes in the sub-region.

[0040] S322, After receiving the sub-policy set, the sub-region control unit checks the current operating state. Boundary conditions Γ k Available control set Ω k The analysis is performed, and the most suitable strategy sub-version is selected within the feasible domain.

[0041] S323, the control unit, based on the selected strategy sub-version Perform equipment control operations within the designated area.

[0042] The beneficial effects of this invention are:

[0043] This invention, by constructing a heterogeneous drift feature map, can accurately identify data inconsistency problems in power distribution systems caused by uploading delays, sampling misalignments, or logical breaks in sensing nodes. It enables systematic diagnosis of time alignment defects and logical coupling breaks. This multi-dimensional structural consistency analysis mechanism breaks through the limitations of traditional methods that rely on a single data indicator for judgment, and improves the system's accuracy in identifying data drift states and its response timeliness.

[0044] This invention, by introducing a historical redundant data interpolation and completion mechanism and a logical compensation method, and combining physical boundary conditions for upper and lower limit constraint correction, realizes the spatiotemporal consistency reconstruction of sensing data, generates a consistent control input set with logical closure, effectively compensates for structural deficiencies and response gaps in the operating data, and improves the data recovery capability and operational stability of the power distribution system under heterogeneous drift background.

[0045] This invention establishes a cooperative strategy generation function based on a consistent control input set, combines a set of scheduling objective functions and system operation constraints, outputs an optimal cooperative control strategy set that can be deployed in different regions, and supports each sub-region to autonomously select a sub-strategy version based on its execution capability. This achieves closed-loop adaptive cooperative control between strategy generation and execution, improving the overall flexibility, robustness and autonomous response capability of the system. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a schematic diagram of the decision-making method according to an embodiment of the present invention;

[0048] Figure 2 This is a schematic diagram of the heterogeneous drift feature map construction process according to an embodiment of the present invention. Detailed Implementation

[0049] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. Those skilled in the art may employ other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0050] like Figures 1-2 As shown, a smart collaborative decision-making method for a power distribution system includes the following steps:

[0051] S1 collects operational data from multiple sub-regions in the power distribution system, including the sampling frequency, upload delay, and data integrity identifier of the sensing nodes. By performing structural consistency analysis on the operational data of each sub-region, heterogeneous drift segments with time alignment defects or logical coupling faults are identified, and heterogeneous drift feature maps characterizing the data drift features of each region are generated.

[0052] S2, based on the generated heterogeneous drift feature map, combined with historical redundant data, topology completion mechanism and physical boundary prior rules, dynamically generates a reconstructed consistent version of the control input for each sub-region, forming a data consistency control input set for policy generation;

[0053] S3 uses the data consistency control input set as the basis for strategy generation, combines the current scheduling objectives and system constraints to construct a collaborative strategy set, and distributes the collaborative strategy set to each sub-region control unit to realize intelligent collaborative control of the power distribution system under data inconsistency.

[0054] S1 includes:

[0055] S11 collects operational data from multiple sub-regions in the power distribution system. The operational data includes the sampling frequency, upload delay, and data integrity identifier of the sensing nodes, and performs standardization processing on the operational data from different sub-regions.

[0056] S12. After completing the standardized collection of operational data, based on the data organization structure, time window alignment degree and logical topological connectivity between sub-regions, structural consistency analysis is performed on the operational data of each sub-region. Heterogeneous drift segments with time alignment defects or logical coupling faults are identified, and the spatial distribution, drift type and influence range of heterogeneous drift segments are extracted to construct a heterogeneous drift feature map reflecting the data consistency status of sub-regions.

[0057] S11 includes:

[0058] S111, for each sub-area R in the power distribution system k (where k = 1, 2, ..., K, and K is the total number of regions), collecting data from N sensing nodes deployed in each region. k,m The running data output by (the m-th node in the k-th region) includes the sampling frequency f. k,m Upload delay Δ k,m Data integrity identifier γ k,m (γ k,m =1 indicates that the data within the current time window is complete, γ k,m =0 indicates packet loss or incomplete upload), and for each sub-region R k The original running dataset D k , is represented as:

[0059] D k ={(f k,m ,Δ k,m ,γ k,m |m=1,2,...,M k};

[0060] Among them, M k Let K be the number of sensing nodes in the k-th region.

[0061] S112, for all sub-regions R k With a unified observation time window T obs and sampling reference frequency f ref As a structural alignment reference, a node-aware state matrix S is constructed. k , is represented as:

[0062]

[0063] Where T = Tobs ·f ref This indicates the number of data points that should be present within a standard time window. Let be the upload behavior matrix of the m-th sensing node in the k-th distribution region at time t;

[0064] S113, standardize the operational data of each sensing node to obtain a standard operational data vector, and combine the standard operational data vectors of all sensing nodes to form a standard operational dataset for the sub-region. Represented as:

[0065]

[0066] in, Let Δ be the standard operating data vector of the m-th sensing node in the k-th electron distribution region. max To maximize the allowable upload latency.

[0067] S12 includes:

[0068] S121, after completing the standardized collection of operational data, for each sub-region R k Temporal features are extracted from the upload behavior of sensing nodes. By calculating the upload time centroid and overall missing rate of sensing nodes, the time offset and data loss within the standard time window are analyzed. Simultaneously, logically adjacent region pairs (R... k ,R u The temporal centroid difference and alignment score between the regions are calculated to identify temporally misaligned pairs, forming a temporal drift candidate set Ω. time , is represented as:

[0069]

[0070] If A is satisfied k,u thr or p k >p thr or p u >p thr Then, the region pair (k,u) is marked as a time-aligned drift pair and added to the set Ω. time ;

[0071] Where, μ k,m The node uploads the activity time focus. p represents the average of the time center of gravity of the regional upload activities. k θ represents the missing data upload rate for the region. k,u A represents the time centroid difference between region pairs. k,u For region alignment scores, the closer to 1, the more synchronized. (A) thr p is the threshold for time alignment determination. thr ​Set a threshold for determining missing data;

[0072] The threshold for time alignment is based on the maximum allowable time offset step size θ of the system. max To configure, it is represented as:

[0073]

[0074] The threshold for determining missing data is expressed as follows:

[0075] p thr =1-α cover ;

[0076] Where, α cover This represents the minimum sensing coverage requirement within the area.

[0077] S122 analyzes the cooperative response of sub-regions in terms of logical topology, assesses whether there is a coupling failure, calculates the consistency between the expected cooperative level and the actual level by sensing the average upload behavior of nodes and combining the power topology coupling weights between sub-regions, and evaluates whether the residual exceeds the residual judgment threshold, identifies logical fault region pairs, and constructs a logical drift candidate set Ω. logic , is represented as:

[0078]

[0079]

[0080] ρ k,u =λ k,u -β k,u ;

[0081] If g k,u =1 and ρ k,u >ρ thr Then, the region (k,u) is denoted as a logical coupling fault and added to the set Ω. logic ;

[0082] Among them, g k,u w is a topological connectivity marker. k,u λ represents the coupling strength between region pairs. k,u To achieve the desired level of synergy, For region R k The average upload behavior at time t, β k,u For actual collaborative consistency, ρ k,u For the cooperative residuals of the region pairs, ρ thr The threshold for residual judgment;

[0083] Residual judgment threshold ρ thr Represented as:

[0084] ρ thr=μ ρ +δ ρ ;

[0085] Where, μ ρ δ is the mean residual of all normal region pairs in the network. ρ This is the residual safety buffer coefficient;

[0086] S123, the identified time-type drift candidate set Ω time And logic class drift candidate set Ω logic After merging, the spatial affected area, number of drift types, and number of adjacent faults in each region are extracted, and a heterogeneous drift feature map Ψ is finally constructed, represented as:

[0087]

[0088] η k =∑ u g k,u ·1((k,u)∈Ω time ∪Ω logic );

[0089] Ψ k,u,1 =1-A k,u ,Ψ k,u,2 =max(ρ k,u ,0),Ψ k,0,3 =σ k ,Ψ k,0,4 =η k ;

[0090] in, For region R k The set of anomalous time segments, σ k η represents the proportion of failed nodes within the region. k Ψ represents the number of connections in a region affected by adjacency drift. k,u,1 Ψ represents the time-aligned defect intensity of region pair (k,u) in the heterogeneous drift pattern. k,u,2 Ψ represents the topological logical coupling residual strength of region pairs (k,u) in the heterogeneous drift map. k,0,3 For region R k Internal spatial drift coverage ratio, Ψ k,0,4 For region R k Adjacency drift affects the count.

[0091] S2 includes:

[0092] S21, for sub-regions with time alignment defects in heterogeneous drift feature maps, identify upload delays, missing segments and sampling misalignment issues in the running data, use historical redundant data for interpolation to complete, and unify the sampling rhythm by referring to the time step of adjacent regions to achieve time consistency processing of running data.

[0093] S22. After completing the time consistency processing, for the sub-regions with logical coupling faults in the heterogeneous drift feature map, analyze the relationship between the power flow and control response between them and the adjacent regions, infer the missing control information, and combine the physical boundary conditions (load capacity and equipment limitations) to correct the operating data after time consistency processing, and generate a consistent control input set with logical closure.

[0094] S21 includes:

[0095] S211, for the sub-regions marked with time alignment defects in the heterogeneous drift feature map, based on the upload behavior matrix of each sensing node, features of upload delay, data missingness, and sampling misalignment are extracted to identify the set of data segments to be corrected. Specifically, it includes:

[0096] (1) Average upload latency of nodes:

[0097] in, Let T be the actual upload time step of the node at time t, and T be the total time window step size. Average upload latency for nodes;

[0098] (2) Node upload missing rate:

[0099] Where, q k,m The missing rate of node uploads;

[0100] (3) Determine sampling misalignment: in, A Boolean flag indicating whether there is a sampling frequency misalignment at the node. f is the actual effective sampling frequency of the node. ref The set reference sampling frequency, if the sampling frequency deviation exceeds ε f If ε is true, then it is True; otherwise, it is False. f The allowable deviation threshold for the sampling frequency, if it meets the following conditions... or q k,m >q thr or Then mark the sensing node N. k,m There is a time alignment defect. Record the time period and add it to the set of data segments to be corrected. μ thr q is the threshold for average upload latency. thr The threshold for the upload missing rate;

[0101] The allowable deviation threshold for sampling frequency is expressed as:

[0102] εf =δ f ·f ref ;

[0103] Where, δ f This is the frequency tolerance factor;

[0104] The threshold for average upload latency is expressed as:

[0105]

[0106] Where, Δt max Δt is the maximum allowable delay time. ref For reference sampling period;

[0107] The threshold for upload missing rate is expressed as:

[0108] q thr =1-α cov ;

[0109] Where, α cov The minimum desired perceived data coverage;

[0110] S212, for the identified set of data segments to be corrected By utilizing historical redundant data and employing interpolation methods to fill in missing points, the continuous observation sequence of the sensing node within the time window T is reconstructed, specifically including:

[0111] (1) Linear interpolation (for missing points) ):

[0112]

[0113] in, The padded value obtained by interpolation. These are the original running data of the node at times t1 and t2, respectively, where t1 and t2 are the effective time points before and after the interpolation reference.

[0114] (2) Update the node data sequence after interpolation:

[0115]

[0116] S213, after completing the interpolation, remap the sensing nodes with inconsistent sampling frequencies in each sub-region to the reference time step Δt. ref , is represented as:

[0117]

[0118] in, T is the time-aligned running data sequence after resampling, and T′ is the total number of time steps after resampling.

[0119] S22 includes:

[0120] S221, after completing the time consistency processing, based on the regions marked as having logical coupling faults in the heterogeneous drift feature map (R... k ,R u This study analyzes the power exchange behavior and control response state of sensing nodes within a standard observation time window, identifies pairs of sensing nodes with discontinuous power direction, lost response information, or untransmitted boundary signals, and constructs a missing control information index set C. k,u , is represented as:

[0121]

[0122] like Then add the boundary node pair (i,j) of the region to the missing control information index set C. k,u ,in, For region R k To R u Power injection, For region R u To R k Power feedback, For power balance residuals, The average power inconsistency within the time window, The threshold for determining power residuals;

[0123] Power residual judgment threshold Represented as:

[0124]

[0125] Where, μ ∈ λ represents the historical average power residual of the entire system. P For the safety factor, σ ∈ The standard deviation of the historical residuals for the entire system;

[0126] S222, for the missing control information index set C k,u For missing runtime data, based on the runtime behavior of similar nodes in adjacent regions within the same time window, a response similarity-based supplementation method is used to generate proxy runtime data to fill in boundary logical gaps, represented as follows:

[0127]

[0128] in, This is the set of proxy nodes used for push notifications. This refers to the actual operational data of proxy node n in the adjacent region. To supplement the running data of node i, ω i,nThe weights represent the behavioral similarity between node i and node n.

[0129] S223, for the supplementary proxy operation data, based on the physical boundary conditions of the maximum or minimum operating capacity, load adjustment range, and power response upper limit of the equipment to which the sensing node belongs, upper and lower limit constraints are adjusted to ensure its physical feasibility, and finally a consistent control input set with logical closure is generated. Represented as:

[0130]

[0131] in, This represents the final running data of node i at time t. These are the upper and lower limits of operation for the device corresponding to node i, respectively.

[0132] S3 includes:

[0133] S31. After obtaining the consistency control input set of each sub-region, construct a set of cooperative strategies based on the current scheduling objectives and system constraints (equipment capacity limits, network topology status, tie-line power boundaries).

[0134] S32 decomposes and distributes the generated collaborative strategy set at the regional granularity. Based on the current operating status, boundary conditions and execution capabilities of each sub-region, it automatically selects the matching sub-strategy version to achieve local autonomous execution.

[0135] S31 includes:

[0136] S311, after obtaining the consistency control input set of each sub-region, a comprehensive scheduling objective function is set according to the current operating scenario. The objectives include minimizing total active power loss, minimizing node voltage deviation, and optimizing load adjustable response. The comprehensive scheduling objective function is expressed as follows:

[0137]

[0138] in, This represents the total active power loss of the line. The sum of squares of the node voltage offsets. The load response deviation is represented by α1, α2, and α3, which are the corresponding weighting coefficients, and u is the set of system control variables.

[0139] S312, based on the set comprehensive scheduling objective function, introduces system constraints, including equipment capacity limits, network topology status (power balance), and tie-line power flow boundaries, expressed as:

[0140] Device capacity limit (based on injected power):

[0141] Tie line power boundary (based on power flow constraints):

[0142] Network topology power flow balance relationship:

[0143] Among them, P i Inject active power into node i. P represents the upper and lower limits of the device capacity of node i, respectively. i,j (u) represents the power flow from node i to j, defined by the system control variable set u. Let i,j be the power carrying capacity limit of the tie line. The set of adjacent nodes that are topologically connected to node i;

[0144] S313, based on the scheduling objective function and system constraints, outputs a set of cooperative strategies. Represented as:

[0145]

[0146] Where, Φ coord For the policy generation function, Θ target To synthesize the parameters of the scheduling objective function, Θ bound These are system constraints.

[0147] S32 includes:

[0148] S321, based on the generated set of cooperative strategies According to sub-region numbering The collaborative strategy set is divided into multiple regional sub-strategy sets. Each sub-strategy set corresponds to the control instructions for all sensing nodes within its sub-region, as follows:

[0149]

[0150] in, For sub-region k, it is the set of sub-policies.

[0151] S322, After receiving the sub-policy set, the sub-region control unit checks the current operating state. Boundary conditions Γ k Available control set Ω k The analysis is performed, and the most suitable strategy sub-version is selected within the feasible domain. Specifically, it includes:

[0152] (1) Constructing the feasible domain constraint set: The control unit analyzes the current state, boundary conditions, and control reachability to determine the feasible control space at the current moment, represented as:

[0153]

[0154] in, This is the set of currently feasible policy candidates;

[0155] (2) Calculate the policy fit matching degree: for each candidate policy in the sub-policy set Assess its fit with the current operating state and select the policy version that best suits the control requirements, denoted as:

[0156]

[0157] Where η(u) is the policy matching deviation function value, φ is the state matching deviation weight coefficient, ζ is the control execution cost weight coefficient, and Cost(u) is the expected control cost of the system control variable set u. The predicted value of the target operating state corresponding to the system control variable set u;

[0158] (3) Select the optimal sub-strategy version: Select the final execution strategy version for the current sub-region based on the minimum matching deviation criterion, expressed as:

[0159]

[0160] S323, the control unit, based on the selected strategy sub-version Perform equipment control operations within the designated area.

[0161] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0162] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for intelligent collaborative decision-making in a power distribution system, characterized in that, Includes the following steps: S1 collects operational data from multiple sub-regions in the power distribution system, including the sampling frequency, upload delay, and data integrity identifier of the sensing nodes. By performing structural consistency analysis on the operational data of each sub-region, heterogeneous drift segments with time alignment defects or logical coupling faults are identified, and heterogeneous drift feature maps characterizing the data drift features of each region are generated. S2, based on the generated heterogeneous drift feature map, combined with historical redundant data, topology completion mechanism and physical boundary prior rules, dynamically generates a reconstructed consistent version of the control input for each sub-region, generating a data consistency control input set, specifically including: S21, for sub-regions with time alignment defects in heterogeneous drift feature maps, identify upload delays, missing segments and sampling misalignment issues in the running data, use historical redundant data for interpolation to complete, and unify the sampling rhythm by referring to the time step of adjacent regions to achieve time consistency processing of running data. S22, after completing the time consistency processing, for the sub-regions with logical coupling faults in the heterogeneous drift feature map, analyze the relationship between the power flow and control response between them and the adjacent regions, infer the missing control information, and combine the physical boundary conditions to correct the running data after the time consistency processing, and generate a consistent control input set with logical closure. S22 includes: S221, after completing the time consistency processing, based on the regions marked as having logical coupling faults in the heterogeneous drift feature map... This study analyzes the power exchange behavior and control response state of sensing nodes within a standard observation time window, identifies pairs of sensing nodes with discontinuous power direction, lost response information, or untransmitted boundary signals, and constructs an index set of missing control information. ; S222, Index set of missing control information To fill the missing operational data in the data, proxy operational data is generated based on the operational behavior of similar nodes in adjacent regions within the same time window, using a response similarity supplementation method. S223, for the supplementary proxy operation data, based on the physical boundary conditions of the maximum or minimum operating capacity, load adjustment range, and power response upper limit of the equipment to which the sensing node belongs, upper and lower limit constraints are corrected, and finally a consistent control input set with logical closure is generated. ; S3, using the data consistency control input set as the basis for strategy generation, and combining the current scheduling target and system constraints, construct a collaborative strategy set, and distribute the collaborative strategy set to each sub-region control unit to realize intelligent collaborative control of the power distribution system under data inconsistency.

2. The intelligent collaborative decision-making method for a power distribution system according to claim 1, characterized in that, S1 includes: S11 collects operational data from multiple sub-regions in the power distribution system. The operational data includes the sampling frequency, upload delay, and data integrity identifier of the sensing nodes, and performs standardization processing on the operational data from different sub-regions. S12. After completing the standardization of the running data, based on the data organization structure, time window alignment and logical topological connectivity between sub-regions, structural consistency analysis is performed on the running data of each sub-region. Heterogeneous drift segments with time alignment defects or logical coupling faults are identified, and the spatial distribution, drift type and influence range of heterogeneous drift segments are extracted to construct a heterogeneous drift feature map that reflects the consistency status of the data in the sub-region.

3. The intelligent collaborative decision-making method for a power distribution system according to claim 2, characterized in that, S11 includes: S111, for each sub-area in the power distribution system. Data is collected from sensing nodes deployed in various regions. The output runtime data includes the sampling frequency. Upload delay Data integrity identifier and for each sub-region Constructing the original running dataset ; S112, for all sub-regions With a unified observation time window and sampling reference frequency As a structural alignment reference, a node-aware state matrix is ​​constructed. ; S113, standardize the operational data of each sensing node to obtain a standard operational data vector, and use the standard operational data vectors of all sensing nodes to form the standard operational dataset of the sub-region. .

4. The intelligent collaborative decision-making method for a power distribution system according to claim 3, characterized in that, S12 includes: S121, after completing the standardization processing of the operational data, for each sub-region Temporal features are extracted from the upload behavior of sensing nodes. By calculating the upload time centroid and overall missing rate of sensing nodes, the time offset and data loss within the standard time window are analyzed. Simultaneously, logically adjacent regions are analyzed. The temporal centroid difference and alignment score are calculated to identify temporally misaligned region pairs, forming a candidate set of temporal drift. ; S122, analyze the cooperative response of sub-regions in terms of logical topology, assess whether there is a coupling failure phenomenon, calculate the consistency between the expected cooperative level and the actual level by sensing the average upload behavior of the nodes and combining the power topology coupling weights between sub-regions, and evaluate whether the cooperative residual between the expected cooperative level and the actual consistency exceeds the residual judgment threshold, identify logical fault region pairs, and construct a logical drift candidate set. ; S123, the identified time-type drift candidate set And logic class drift candidate set After merging, the spatial extent of the affected area, the number of drift types, and the number of adjacent faults affecting each region are extracted, ultimately constructing a heterogeneous drift feature map. .

5. The intelligent collaborative decision-making method for a power distribution system according to claim 1, characterized in that, S21 includes: S211, for the sub-regions marked with time alignment defects in the heterogeneous drift feature map, based on the upload behavior matrix of each sensing node, features of upload delay, data missingness, and sampling misalignment are extracted to identify the set of data segments to be corrected. ; S212, for the identified set of data segments to be corrected By utilizing historical redundant data and employing interpolation methods to fill in missing points, the sensing nodes are reconstructed within the time window. A continuous observation sequence within; S213, after completing the interpolation, remap the sensing nodes with inconsistent sampling frequencies in each sub-region to the reference time step. .

6. The intelligent collaborative decision-making method for a power distribution system according to claim 5, characterized in that, S3 includes: S31, After obtaining the consistency control input set of each sub-region, construct a set of cooperative strategies based on the set comprehensive scheduling objective function and system constraints; S32 decomposes and distributes the generated collaborative strategy set at the regional granularity. Based on the current operating status, boundary conditions and execution capabilities of each sub-region, it automatically selects the matching sub-strategy version to achieve local autonomous execution.

7. The intelligent collaborative decision-making method for a power distribution system according to claim 6, characterized in that, S31 includes: S311 After obtaining the consistency control input set of each sub-region, the comprehensive scheduling objective function is set according to the current operating scenario. The comprehensive scheduling objective function includes minimizing total active power loss, minimizing node voltage deviation, and optimizing load adjustable response. S312, based on the set comprehensive scheduling objective function, introduces system constraints, including equipment capacity limits, network topology status and tie-line power flow boundaries; S313, based on the comprehensive scheduling objective function and system constraints, outputs a set of cooperative strategies. .

8. The intelligent collaborative decision-making method for a power distribution system according to claim 7, characterized in that, S32 includes: S321, based on the generated set of cooperative strategies According to sub-region numbering The collaborative strategy set is divided into multiple regional sub-strategy sets, and each sub-strategy set corresponds to the control instructions for all sensing nodes in the sub-region. S322, After receiving the sub-policy set, the sub-region control unit checks the current operating state. Boundary conditions Available control sets The analysis is performed, and the most suitable strategy sub-version is selected within the feasible domain. ; S323, the control unit, based on the selected strategy sub-version Perform equipment control operations within the designated area.

Images