Master integration multi-level interactive optimization operation method and system

Abstract

The application discloses a main-distribution integrated multi-level interactive optimal operation method and system, relates to the technical field of main-distribution collaborative optimal operation of a power system, and comprises the following steps: obtaining main layer operation requirements and distribution layer physical operation states, identifying a distribution layer structure sensitive unit and an operation state switching boundary thereof; the distribution layer divides the physical constraint space into modes and intervals according to the operation state switching boundary, and generates a structure stable executable solution domain with an operation mode identifier; the structure stable executable solution domain is uploaded to the main layer, and an optimization space with limited variable values is constructed; the main layer solves a global objective function in the limited optimization space, determines an operation scheme meeting the structure stable constraint of the distribution layer, and issues the operation scheme for execution. The application can reduce the problems that the main layer optimization scheme is not executable on the distribution layer side and is repeatedly corrected.

Description

Technical Field

[0001] This invention relates to the field of power system main and distribution coordinated optimization operation technology, and more specifically, to a method and system for integrated main and distribution multi-level interactive optimization operation. Background Technology

[0002] In a multi-level collaborative operation scenario integrating main and distribution layers, the main layer typically establishes an optimization model based on macroscopic state parameters such as overall network load demand, resource allocation targets, and regional power balance, and issues adjustment commands such as power regulation, voltage support, load transfer, or resource allocation to the distribution layers. The distribution layers, in turn, need to combine real-time physical constraints such as node voltage, branch power flow, equipment capacity, protection settings, voltage regulating equipment status, and local topology status to ensure that the main layer's adjustment commands can be implemented under the current operating conditions. However, the executable range of the distribution layers is not always a continuous and stable ordinary constraint space. Under the influence of the main layer's adjustment requirements, some distribution layer units may approach the operational state switching boundaries such as protection actions, power flow direction switching, voltage regulating equipment actions, or topology reconfiguration. Once these boundaries are crossed, the constraint structure, control mode, or topology status of the distribution layers will change, making it impossible for the adjustment scheme obtained in the main layer's continuous optimization model to be directly executed on the distribution layer side.

[0003] Existing master-slave collaborative optimization methods often employ an interactive process of "master layer optimizes first, slave layer verifies or corrects later." This means the master layer first solves for the adjustment scheme based on the global objective function, and then the slave layer verifies its feasibility based on real-time physical constraints. When the master layer's output scheme falls into the slave layer's inexecutable region or crosses the slave layer's operational state switching boundary, the slave layer typically needs to perform rollback, limit, secondary correction, or feedback to the master layer for recalculation. This can easily lead to master-slave feasibility mismatch, increased interaction iterations, and delayed scheduling response.

[0004] The above-disclosed technical solutions have at least the following technical problems: existing master-slave collaborative optimization usually involves solving the master layer and then verifying it by the slave layer. It fails to incorporate the operating state switching boundary of the sensitive unit of the slave layer structure before the master layer optimization, which makes it easy for the optimal solution of the master layer to cross the slave layer protection action, operating mode switching or topology switching boundary, resulting in problems such as slave layer not being executable, repeated correction and scheduling response lag. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a multi-level interactive optimization operation method integrating main and auxiliary layers. By identifying the operation state switching boundary of sensitive units in the auxiliary layer structure and incorporating the structurally stable and executable solution domain into the main layer optimization space in advance, this method solves the problems in the existing main-auxiliary collaborative optimization where the main layer scheme is not executable on the auxiliary layer side, requires repeated correction, and has a delayed response.

[0006] To achieve the above objectives, the present invention provides the following technical solution: On the one hand, the integrated multi-level interactive optimization operation method of the main layer includes the following steps: obtaining the main layer's operation requirements and the physical operation status of the supporting layer, identifying the structurally sensitive units of the supporting layer and their operation status switching boundaries; the supporting layer uses the switching boundaries of the structurally sensitive units as boundaries to divide the physical constraint space into modes and intervals, generating a structurally stable executable solution domain with operation mode identifiers; uploading the structurally stable executable solution domain to the main layer, and constructing an optimization space with restricted variable values ​​on the main layer side; the main layer solves the global objective function within the restricted optimization space, determines the operation scheme that satisfies the structural stability constraints of the supporting layer, and issues it for execution.

[0007] In a preferred embodiment, identifying sensitive units of the layered structure and their operational state switching boundaries includes: mapping the main layer's operational requirements to the layered physical variable space to determine the direction of their effect on local units of the layered structure; filtering candidate units based on the direction of effect that meet the conditions of proximity protection triggering, operational mode switching, or topology switching; performing disturbance evolution judgment on the candidate units, identifying candidate units that undergo structural changes within a preset disturbance range as structurally sensitive units, and determining the critical conditions that trigger structural changes as operational state switching boundaries.

[0008] In a preferred embodiment, mapping the main layer operational requirements to the supporting layer physical variable space includes: constructing a directed physical association graph based on the current main-supporting network topology and power flow direction; using the boundary nodes corresponding to the adjustment demand components as source nodes, searching for candidate propagation paths for the adjustment demand components under directional consistency constraints; verifying the capacity carrying capacity and operational constraint continuity of the candidate propagation paths, eliminating propagation paths that cannot carry the adjustment requirements or will cause path interruption, and obtaining a set of physical propagation paths for the adjustment demand components in the supporting layer network.

[0009] In a preferred embodiment, determining candidate units that will undergo structural changes within a preset disturbance range as structurally sensitive units includes: screening candidate units with protection triggering, operation mode switching, or topology switching mechanisms, and constructing a critical state interval based on the operation state threshold interval of the candidate units; when the current operation state of a candidate unit enters the critical state interval, performing state evolution simulation on the candidate unit in conjunction with the direction of action of the main layer operation requirements in the physical variable space of the supporting layer; if a candidate unit triggers structural changes within a preset adjustment disturbance range, it is determined as a structurally sensitive unit, and the corresponding triggering condition is determined as the operation state switching boundary.

[0010] In a preferred embodiment, generating a structurally stable executable solution domain with an operating mode identifier includes: dividing the physical constraint space of the layered structure into operating modes based on the operating state switching boundary of the structurally sensitive unit, and fixing the corresponding state in each operating mode; solving the initial feasible solution domain within the mode using the main layer adjustment variable as the parameter variable, and identifying the structural trigger point that triggers the operating state switching of the structurally sensitive unit; dividing the initial feasible solution domain into intervals using the structural trigger point as the boundary, determining the sub-feasible solution domain that satisfies the preset physical operating constraints and maintains the unchanged operating mode after the division as the structurally stable executable solution domain, and configuring the operating mode identifier.

[0011] In a preferred embodiment, the step of constructing an optimization space with restricted variable values ​​on the main layer side includes: the auxiliary layer identifies the structurally stable executable solution domain according to the running mode state, and establishes corresponding running mode identifiers for continuous executable intervals, discrete executable intervals, or discrete candidate points; after receiving the structurally stable executable solution domain with running mode identifiers, the main layer constructs an adjustment subspace according to the running mode identifiers, so that the variable value range of each adjustment subspace is limited by the structurally stable executable solution domain under the corresponding running mode identifier.

[0012] In a preferred embodiment, constructing an optimization space with restricted variable values ​​further includes: establishing a mapping relationship between the principal-level optimization variables and the structurally stable executable solution domain within each adjustment subspace; determining the value boundaries, candidate value sets, or feasible domains of variable combinations for the principal-level optimization variables when the structurally stable executable solution domain is a continuous interval, a discrete point set, or a multidimensional region; performing a consistency check on the candidate values ​​or variable combinations to eliminate variable combinations that cross different operating mode identifiers and do not meet the mode switching conditions, thereby obtaining an optimization space with restricted variable values.

[0013] In a preferred embodiment, determining and executing an operation scheme that satisfies the stability constraints of the layered structure includes: generating candidate operation schemes within a constrained optimization space; eliminating candidate operation schemes that do not meet the consistency verification of the operation mode; calculating the global objective function value for the candidate operation schemes that pass the verification; and selecting an operation scheme that satisfies the main layer operation objective and the stability constraints of the layered structure and executing it in the layered structure.

[0014] On the other hand, the integrated multi-level interactive optimization operation system includes: a unit and boundary identification module: used to obtain the main layer's operation requirements and the physical operation status of the supporting layer, and identify the structurally sensitive units of the supporting layer and their operation status switching boundaries; an executable solution domain generation module: used by the supporting layer to divide the physical constraint space into modes and intervals based on the switching boundaries of the structurally sensitive units, and generate a structurally stable executable solution domain with operation mode identifiers; a constrained optimization space construction module: used to upload the structurally stable executable solution domain to the main layer, and construct an optimization space with constrained variable values ​​on the main layer side; and a global objective optimization module: used by the main layer to solve the global objective function within the constrained optimization space, determine the operation scheme that satisfies the structural stability constraints of the supporting layer, and issue it for execution.

[0015] The technical effects and advantages of the main and auxiliary integrated multi-level interactive optimization operation method and system of this invention are as follows: This invention generates an executable solution domain based on real-time physical operating status and local constraints in the auxiliary layer, and uploads this executable solution domain to the main layer to construct a constrained optimization space. This ensures that the optimization variables of the main layer are always limited to the executable range of the auxiliary layer, thereby avoiding unexecutable instructions and repeated verifications caused by inconsistencies between the main layer's decisions and the auxiliary layer's physical constraints. At the same time, the main layer performs global objective optimization and optimal solution selection within the constrained optimization space, and issues the optimal running plan for execution. This can improve the stability and convergence efficiency of the running schedule while ensuring executability, and reduce the running risks and scheduling rollbacks caused by constraint mismatch. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the integrated main and auxiliary multi-level interactive optimization operation method of the present invention; Figure 2 This is a schematic diagram of the main and auxiliary integrated multi-level interactive optimization operation system of the present invention; Figure 3 Main layer-complementary layer layered architecture and interaction data flow diagram; Figure 4 This is a schematic diagram of a two-way boundary expansion search. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1, Figure 1 The present invention provides a multi-level interactive optimization operation method for main and auxiliary systems, comprising the following steps: S1, obtain the main layer's operational requirements and the physical operational status of the auxiliary layer, and identify the sensitive units of the auxiliary layer structure and their operational status switching boundaries; In this embodiment, the main layer refers to the upper-level control layer used for global operation coordination, power balance regulation, or regional resource allocation, and the auxiliary layer refers to the distribution-side control layer used for local power flow control, voltage regulation, protection coordination, or topology adjustment. The main layer's operational requirements characterize the regulation targets that the main layer needs to transmit to the auxiliary layer during the current operating cycle, including at least one of the following: power regulation requirements, voltage support requirements, load transfer requirements, resource allocation requirements, or operational margin improvement requirements.

[0019] The physical operating status of the layer is used to characterize the actual electrical operating status of the layer's local units within the current operating cycle, including at least one of node voltage, branch current, power flow direction, equipment capacity occupancy status, protection operation status, voltage regulating equipment operating status, switch status, and local operating margin. The layer's local units include at least one of feeders, distribution areas, distribution transformers, sectionalizing switches, tie switches, reactive power compensation devices, distributed power supply access units, energy storage access units, or adjustable load access units.

[0020] In this embodiment, the structure-sensitive unit refers to a layered physical unit whose operating state is prone to protection actions, topology switching, control mode switching, or equipment level switching under the operational requirements of the main layer. Compared with ordinary constraint units, the characteristic of a structure-sensitive unit is that when its operating variables cross the corresponding operating state switching boundary, the constraint structure, control mode, or topology state of the layered network will change, thereby rendering the optimization scheme calculated by the main layer according to the original continuous feasible space no longer practically executable.

[0021] In this embodiment, obtaining the main layer's operational requirements and the physical operational status of the supporting layers, and identifying sensitive units of the supporting layer structure and their operational status switching boundaries, includes: Obtain the main layer system operation status parameters and the real-time physical operation status parameters of each local unit in the auxiliary layer; The operating state parameters of the main layer system are decomposed into basic demand components and adjustment demand components. The basic demand components are used to characterize the rigid demand required to maintain the global operating balance of the main layer, and the adjustment demand components are used to characterize the flexible adjustment demand that can be redistributed among the local units of the auxiliary layer. Based on the current primary and secondary network topology, path reachability analysis is performed on the regulation demand components to determine the set of physical propagation paths of the regulation demand components in the secondary network. The sensitivity contribution relationship of the adjustment demand component to the physical variables of the stratification is calculated along the physical propagation path set, and the demand influence coefficient matrix is ​​obtained. Based on the demand influence coefficient matrix, the main layer operation demand is mapped to the supporting layer physical variable space to obtain the scope and direction of the main layer operation demand on the supporting layer local units; Based on the real-time physical operating status parameters of each local unit of the layer, constraint sensitivity screening is performed on the local units of the layer to identify structurally sensitive units, and the operating status switching boundary is determined according to the protection triggering conditions, operating mode switching conditions or topology switching conditions corresponding to the structurally sensitive units.

[0022] It should be noted that the main layer system operation status parameters include at least one of the following: global load demand, resource allocation target, regional power balance deviation, and system operation margin index; the real-time physical operation status parameters include at least one of the following: node voltage amplitude and phase angle, branch current and power flow direction, local equipment capacity occupancy rate, and local safety margin coefficient.

[0023] Furthermore, based on the current primary and secondary network topology, the path reachability analysis of the adjustment demand components is performed to determine the set of physical propagation paths of the adjustment demand components in the secondary network, including: A directed physical association graph is constructed based on the current primary and secondary network topology; wherein, the nodes in the directed physical association graph represent secondary physical nodes, the edges represent branch connection relationships, and the edge direction is determined by the current power flow direction or energy transmission direction; Using the boundary node corresponding to the adjustment demand component as the source node, a path expansion search under the direction consistency constraint is performed in the directed physical association graph to generate an initial propagation path set; Perform a capacity continuity check on each propagation path in the initial propagation path set to determine whether the remaining capacity of each branch on the path meets the transmission requirements of the adjustment demand component. The set of paths after passing through direction consistency verification, capacity continuity verification, closed loop verification, and node saturation constraint verification is determined as the set of physical propagation paths of the adjustment demand component in the layered network.

[0024] The sensitivity contribution relationship of the adjustment demand component to the stratified physical variables is calculated along the physical propagation path to obtain the demand influence coefficient matrix, which includes: The path transmission weights for each propagation path are determined based on the remaining capacity, path impedance, power flow direction, and operational margin of key nodes along the path. Based on the change in the physical variables of the stratification layer under unit demand disturbance, the local sensitivity of the regulating demand component on the propagation path is determined. The local sensitivity of the same ligand physical variable on different propagation paths is weighted and superimposed to obtain the comprehensive sensitivity contribution value of the regulation demand component to the ligand physical variable. When a node or branch on the propagation path approaches capacity saturation, the comprehensive sensitivity contribution value is corrected for capacity saturation to obtain the corrected sensitivity contribution value. Based on the corrected sensitivity contribution values ​​between each regulation demand component and each stratified physical variable, a demand influence coefficient matrix is ​​formed.

[0025] The path transmission weight coefficient is calculated using the following formula:

[0026] The specific formula for calculating the local sensitivity is as follows:

[0027] The specific formula for calculating the comprehensive sensitivity contribution value is as follows:

[0028] The specific formula for calculating the corrected comprehensive sensitivity contribution value is as follows:

[0029]

[0030] in, The path transmission weight coefficient for each path is the k-th adjustment demand component of the main layer. For path The minimum remaining capacity is the minimum difference between the rated capacity of each branch on the path and the current load. The number of possible propagation paths, For unit demand disturbance in the path The local sensitivity coefficient on, For physical variables, For along the path The allocated demand flow, Obtained through small perturbation simulation of the current physical model. Let be the overall sensitivity contribution value of the k-th demand component to the j-th physical variable. The suppression coefficient, The maximum load rate of the path. This is the preset load rate threshold.

[0031] Furthermore, the constraint sensitivity screening of the local units of the composite layer to identify structurally sensitive units includes: Identify physical units in the layered network that possess protection triggering mechanisms, operating mode switching mechanisms, or topology switching mechanisms, and form a candidate unit set; For each candidate unit, extract its operating state threshold range, and construct a critical state range based on the operating state threshold range; Determine whether the current operating state of the candidate unit has entered the critical state interval. If it has entered the critical state interval, mark the candidate unit as a structurally sensitive unit. Based on the direction of action after mapping the main layer's operational requirements to the physical variable space of the secondary layer, state evolution simulation is performed on the structurally sensitive unit to determine whether the structurally sensitive unit will trigger protection actions, operation mode switching, or topology switching during the main layer's adjustment process. If a structurally sensitive unit triggers a protection action, operation mode switch, or topology switch within a preset adjustment disturbance range, the triggering condition corresponding to that structurally sensitive unit is determined as the operation state switching boundary.

[0032] The protection triggering mechanism includes at least one of branch overload protection, voltage over-limit protection, or protection linkage action; the operation mode switching mechanism includes at least one of power flow direction switching, reactive power compensation device activation or deactivation, or transformer tap position switching; the topology switching mechanism includes at least one of sectionalizing switch action, tie switch action, or local topology reconfiguration.

[0033] Through the above steps, the main layer's operational requirements are first decomposed and mapped to the auxiliary layer's physical variable space, enabling the auxiliary layer to identify the propagation path and direction of influence of these operational requirements in the local network. Simultaneously, the auxiliary layer identifies structurally sensitive units and their operational state switching boundaries that could trigger protection actions, topology switching, or operational mode switching based on real-time physical operational states. Therefore, when generating a structurally stable and executable solution domain, the auxiliary layer considers not only common physical constraints such as voltage, current, and capacity, but also the impact of operational state switching boundaries on executability, thus preventing structural changes from being triggered when the main layer's optimization results are executed on the auxiliary layer side.

[0034] S2, the layering is divided into modes and intervals based on the switching boundary of the structurally sensitive unit, generating a structurally stable and executable solution domain with operation mode identifier; In this embodiment, the physical constraint space refers to the space of adjustment variables that are jointly defined by node voltage constraints, branch current constraints, equipment capacity constraints, power flow direction constraints, protection action constraints, voltage regulating equipment action constraints, and topology switching constraints under the current operating state of the stratum and the operating requirements of the main layer.

[0035] The structurally stable executable solution domain refers to the executable area where, based on preset physical operating constraints, its internal adjustment values ​​will not trigger protection actions, topology switching, control mode switching, or equipment level switching in the corresponding operating mode. The operating mode identifier is used to mark the layered operating structure corresponding to the structurally stable executable solution domain. The layered operating structure includes at least one of the following: network topology state, protection action state, power flow direction state, automatic adjustment device operating state, or control mode state.

[0036] The layering, using the switching boundary of the structurally sensitive unit as the demarcation, performs mode partitioning and interval splitting of the physical constraint space to generate a structurally stable and executable solution domain with operating mode identifiers, including: Based on the operation state switching boundary corresponding to the structurally sensitive unit, extract the trigger boundary information of the structurally sensitive unit; Based on the trigger boundary information, the current operating state of the layered network is divided into modes to form a set of candidate operating modes; wherein, each candidate operating mode corresponds to a determined network topology state, protection action state, power flow direction state, automatic adjustment device working state, or control mode state. For each candidate operating mode, the network topology, protection action logic state, power flow direction state, and automatic adjustment device working state corresponding to the candidate operating mode are fixed to form a physical constraint structure within the mode corresponding to the candidate operating mode. Using the main layer adjustment variable as the parameter variable, a boundary search is performed along each adjustment dimension under the physical constraint structure within the mode to determine the initial feasible solution domain that satisfies the preset physical operation constraints. The structural stability of the initial feasible solution domain is verified to determine whether there is a structural trigger point within the initial feasible solution domain that can trigger the switching of the operating state of the structurally sensitive unit. When a structural trigger point exists within the initial feasible domain, the initial feasible domain is divided into multiple sub-feasible domains by using the structural trigger point as the boundary. Structural stability verification is repeatedly performed on each sub-feasible domain until each sub-feasible domain no longer triggers new running state switching within the preset control resolution range; The sub-feasible solution domain that satisfies the preset physical operation constraints and maintains structural stability under the corresponding operation mode is determined as the structurally stable executable solution domain, and an operation mode identifier is configured for the structurally stable executable solution domain.

[0037] The step of extracting the trigger boundary information of the structurally sensitive unit based on the operating state switching boundary corresponding to the structurally sensitive unit includes: Let the operating state variable of the m-th structurally sensitive unit be... Its allowed operating range is:

[0038] in, Main layer adjustment variables, For running state variables In regulating variables The allowed operating range below, , These are the lower and upper switching boundaries of the m-th structurally sensitive unit, respectively.

[0039] Define the m-th structurally sensitive unit in the adjustment variable The following switching margin is:

[0040] When the following conditions are met: When this happens, the sensitive unit of the structure is determined to have entered the critical switching range; where, For the m-th structurally sensitive unit in the adjustment variable Switching margin below, This is the preset switching margin threshold.

[0041] It should be noted that the operating state switching boundary of the structurally sensitive unit can be determined according to the operating state switching conditions of the corresponding structurally sensitive unit. The operating state switching conditions include: a branch disconnection condition triggered when the branch current or power flow value reaches the overload protection setting value and remains there for more than a preset time; a reactive power compensation device activation or deactivation condition triggered when the node voltage reaches the voltage segmentation control trigger threshold; a tap position switching condition triggered when the transformer load rate reaches the tap changer automatic adjustment threshold; a power flow direction switching condition triggered when the power direction reverses and remains for more than a preset direction holding time; and a local topology reconfiguration condition triggered when the protection coordination logic determines that adjacent units meet the linkage action conditions. The satisfaction of any of the above operating state switching conditions can be considered an operating state switching event.

[0042] The process of splitting the initial feasible solution domain into intervals specifically involves: In the r-th candidate operating mode, if there is a value for the adjustment variable... satisfy:

[0043] and

[0044] Then It was identified as the structural trigger point.

[0045] For a one-dimensional adjustment interval, if the initial feasible solution domain is:

[0046] Furthermore, a set of structural trigger points was identified within this interval:

[0047] Then, by splitting the initial feasible solution domain into intervals according to the structural trigger point, we obtain:

[0048] in, To adjust variables, Let r be the set of structure trigger points in the r-th candidate running mode. The number of structural trigger points, This represents the initial feasible solution domain.

[0049] The step of determining the sub-feasible solution domain that satisfies the preset physical operation constraints and maintains structural stability under the corresponding operation mode as the structurally stable executable solution domain includes: The feasible solution domain for any sub-sub ... If the following conditions are met: Furthermore, for any structurally sensitive element m, the following conditions are met within this sub-feasible domain:

[0050] Then the sub-feasible domain is determined as the structurally stable sub-feasible domain.

[0051] The structurally stable executable domain in the r-th running mode is:

[0052] in, For the structurally stable executable solution domain in the r-th running mode, Let l be the feasible solution domain of the l-th structurally stable sub-mode under the r-th operating mode. This represents the number of feasible solution domains for structurally stable subsystems.

[0053] Furthermore, the structurally stable executable solution domain with runtime mode identifier can be represented as:

[0054] in, The set of structurally stable, executable solutions uploaded to the main layer. This is the identifier for the r-th operating mode. This represents the number of operating modes.

[0055] Further, the step of performing boundary search along each adjustment dimension under the physical constraint structure within the mode to determine the initial feasible solution domain that satisfies the preset physical operating constraints includes: The current value of the main layer adjustment variable in the zero-disturbance state is used as the initial reference point; Under the physical constraint structure within the mode corresponding to the fixed candidate operation mode, an adjustment dimension is constructed for each principal layer adjustment variable; Keep other adjustment variables unchanged in a single adjustment dimension, change the adjustment variable in the positive and negative directions respectively, and resolve the running state corresponding to the physical constraint structure in the mode after each adjustment; Based on the obtained operating status, check whether the node voltage, branch current, equipment capacity, power flow direction, protection action status, voltage regulating equipment status, and topology status meet the preset physical operating constraints. When any preset physical operation constraint is detected to have reached the constraint boundary, the boundary value of the adjustment variable in the corresponding direction is recorded. Based on the boundary values ​​of each adjustment dimension, the initial feasible solution domain for this candidate operating mode is determined; When the principal layer adjustment variable is a multidimensional variable, cross-validation is performed on the feasible intervals corresponding to each adjustment dimension to eliminate regions that trigger physical constraint conflicts or operation state switching under multidimensional combinations, thus forming a multidimensional initial feasible solution domain.

[0056] The initial feasible solution domain is solved using the following method: Let the r-th candidate running mode be The physical constraint function in this candidate operating mode is Its constraint boundary is:

[0057] in, This refers to the i-th layer physical operating quantity caused by the primary layer regulation variable q under the r-th candidate operating mode. The layer physical operating quantity includes node voltage, branch current, equipment capacity utilization, power flow direction, or voltage regulating equipment status. Let i be the lower limit constraint value of the i-th physical quantity in the r-th operating mode. This represents the upper limit constraint value for the i-th physical quantity in the r-th operating mode.

[0058] Then the initial feasible solution domain for the r-th candidate running mode is:

[0059] in, Let be the initial feasible solution domain that satisfies the preset physical operation constraints under the r-th candidate operation mode. This represents the number of physical quantities involved in constraint verification under this candidate operating mode.

[0060] The physical operating quantities of the layer include at least one of the following: node voltage, branch current, equipment capacity occupancy rate, power flow direction, transformer tap status, reactive power compensation device activation status, or topology switch status.

[0061] For example, when When representing node voltage, and These represent the lower and upper limits of the node voltage allowed in this operating mode, respectively; when When representing branch current, The allowable current boundary can be zero or reversed. This indicates the upper limit of current for that branch in the corresponding operating mode; when When indicating equipment capacity utilization, Zero can be taken. This indicates the maximum capacity utilization rate allowed for the device in the corresponding operating mode.

[0062] The structural stability verification of the initial feasible solution domain includes: Representative adjustment points are selected within the initial feasible solution domain; the selection interval of the representative adjustment points is determined based on the minimum control resolution, measurement error range, or adjustment execution accuracy of the layered adjustment variable. While keeping the candidate operating modes fixed, the values ​​of the principal layer adjustment variables corresponding to the representative adjustment points are substituted into the physical constraint structure within the mode to obtain the corresponding co-layer physical operating states. The physical operating state of the layer is compared with the operating state switching boundary of the structural sensitive unit to determine whether the representative adjustment point triggers the operating state switching of the structural sensitive unit. When a representative adjustment point triggers a change in the operating state of a structurally sensitive unit, that representative adjustment point is determined as the structural trigger point. Using the structural trigger point as the boundary, the original initial feasible solution domain is split into multiple sub-feasible solution domains, and the structural stability verification is repeatedly performed on each sub-feasible solution domain. When representative adjustment points within a certain sub-feasible domain do not trigger the switching of the operating state of the structurally sensitive unit, and the sub-feasible domain maintains the same operating mode within the preset control resolution range, the sub-feasible domain is determined as a structurally stable sub-feasible domain.

[0063] In this embodiment, structural stability does not require verification of every mathematical point in the continuous space. Instead, it means that within the allowable range of actual control resolution, measurement accuracy, and execution error, the structurally stable executable solution domain will not trigger a new operational state switch. This avoids directly deriving mathematically stable full-range stability from only a small number of sampling points, thereby improving the rigor of the logic.

[0064] Furthermore, after obtaining the structurally stable executable domain, the auxiliary layer can undergo structural compression processing before being uploaded to the main layer. This structural compression processing includes: parameterizing the structurally stable executable domain based on the mapping relationship between the main layer's adjustment variables and the auxiliary layer's key physical variables; representing continuous executable intervals as upper and lower bounds of variables, discrete executable intervals as discrete candidate sets, and multidimensional executable regions as low-dimensional boundary parameter sets; and binding each parameterized expression result with a corresponding running mode identifier.

[0065] Through the above steps, the supplementary layer does not simply upload ordinary feasible intervals to the main layer. Instead, it divides the physical constraint space into modes based on the operational state switching boundaries of structurally sensitive units, and further splits regions that may trigger structural changes. This ensures that the final generated structurally stable and executable solution domain possesses both physical feasibility and operational mode stability. In this way, subsequent optimization by the main layer can avoid adjustment value regions that might trigger protection actions, topology changes, or control mode switching.

[0066] S3 uploads the structurally stable and executable solution domain to the main layer and constructs an optimization space with restricted variable values ​​on the main layer side; In this embodiment, the optimization space with restricted variable values ​​refers to the optimization search space formed by restricting the range of values ​​of the main layer's optimization variables based on the structurally stable and executable solution domain uploaded by the auxiliary layer. The main layer's optimization variables include at least one of the following: power regulation, voltage support, reactive power regulation, load transfer, energy storage charging and discharging commands, distributed power output regulation, or local resource call-up.

[0067] In this embodiment, the step of uploading the structurally stable and executable solution domain to the main layer and constructing an optimization space with restricted variable values ​​on the main layer side includes: The layer identifies the structurally stable executable solution domain according to the running mode state, and establishes a corresponding running mode identifier for each continuous executable interval, discrete executable interval, or discrete candidate point. Transmit the structurally stable and executable domain with the operating mode identifier to the main layer control unit; After receiving the structurally stable executable solution domain, the main layer control unit constructs an adjustment subspace according to the operation mode identifier; wherein, each adjustment subspace corresponds to an operation mode identifier, and the range of variable values ​​in the adjustment subspace is limited by the structurally stable executable solution domain under the corresponding operation mode identifier; Establish a mapping relationship between the principal optimization variables and the structurally stable executable solution domain in each adjustment subspace, and determine the value constraints of the principal optimization variables according to the expression form of the structurally stable executable solution domain; When the structurally stable executable solution domain is a continuous interval, this continuous interval is used as the boundary for the continuous values ​​of the principal layer optimization variables in the corresponding adjustment subspace; When the structurally stable executable solution domain is a discrete interval or a set of discrete points, the discrete interval or set of discrete points is used as the candidate value set of the principal layer optimization variable in the corresponding adjustment subspace. When the structurally stable executable solution domain is a multi-dimensional region, a feasible domain for variable combinations corresponding to the running mode identifier is constructed based on the combination constraint relationship between the optimization variables of each principal layer. Perform a consistency check on the candidate values ​​or combinations of candidate variables for the main layer optimization variables, and eliminate variable combinations that span different operating mode identifiers and do not meet the mode switching conditions. Candidate values ​​or combinations of candidate variables that pass the consistency check of the running mode are determined as the optimization space with restricted variable values, so that the main layer optimization variables can only take values ​​in the structurally stable and executable solution domain under the corresponding running mode identifier.

[0068] Furthermore, when the main layer optimization variables are multidimensional variables, the main layer can construct candidate variable combinations using a progressive dimension-by-dimensional approach. This progressive dimension-by-dimensional approach includes: first generating candidate values ​​within one adjustment dimension, then progressively adding candidate values ​​from other adjustment dimensions; after each adjustment dimension is added, a feasibility check is performed based on the structurally stable executable solution domain and the operating mode identifier, eliminating variable combinations that do not belong to the corresponding structurally stable executable solution domain. This method reduces the number of multidimensional variable combinations while ensuring that the retained candidate variable combinations satisfy the layered structural stability constraints.

[0069] In this embodiment, when constructing an optimization space with restricted variable values, the main layer no longer uses the physical constraints of the ligand layer as a posterior verification condition after solving the global objective function. Instead, it uses the structurally stable and executable solution domain uploaded by the ligand layer as the prior value boundary of the main layer's optimization variables. In other words, the search space of the main layer's objective function is already restricted to the range where the ligand layer has confirmed execution and structural stability before solving.

[0070] Through the above steps, before performing global optimization, the main layer has limited the range of values ​​of the optimization variables to the structurally stable and executable solution domain generated by the auxiliary layer. This can prevent the theoretical optimal solution obtained by the main layer based on the global objective function from triggering protection actions, topology switching, or changes in operating mode when executed on the auxiliary layer side, thereby improving the executability of the main layer optimization scheme.

[0071] S4, the main layer solves the global objective function within the constrained optimization space, determines the operating scheme that satisfies the stability constraints of the layered structure, and issues it for execution.

[0072] In this embodiment, the global objective function is used to evaluate the overall operational effectiveness of different operational schemes under the main layer's operational requirements. The overall operational effectiveness includes at least one of the following: system power balance, network loss level, critical node voltage deviation, system stability margin, resource allocation cost, adjustment smoothness, or safety margin. The layer structure stability constraint refers to the actual executable constraint of the layer, jointly defined by the structurally stable executable solution domain and its operational mode identifier.

[0073] In this embodiment, the main layer solves the global objective function within a constrained optimization space, determines an operating scheme that satisfies the stability constraints of the layered structure, and issues it for execution, including: Generate candidate running schemes within an optimization space where variable values ​​are limited; The candidate running schemes are subjected to a running mode consistency check to determine whether the values ​​of the main layer optimization variables in the candidate running schemes fall into the structurally stable and executable solution domain under the corresponding running mode identifier. Eliminate candidate running schemes that do not meet the consistency verification of the running mode; For candidate running schemes that pass the running mode consistency check, substitute them into the global objective function to perform performance evaluation and obtain the corresponding global objective function value; Based on the global objective function value, select an operation scheme from the candidate operation schemes that satisfies the main layer operation objective and the stability constraint of the layer structure; The operation plan is sent to the auxiliary layer for execution, so that the auxiliary layer adjusts according to the operation mode identifier corresponding to the operation plan.

[0074] The main layer solves the global objective function within a constrained optimization space, including: The optimization space with limited values ​​for variables in the main layer can be represented as:

[0075] in, For optimization space where variable values ​​are restricted, This is the operating mode identifier corresponding to the adjustment variable q.

[0076] The main layer solves the global objective function within the restricted optimization space of the variables:

[0077]

[0078] in, The operational plan determined by the master layer. The global objective function is... , , These are the weight coefficients for the corresponding evaluation items. For network loss evaluation items, For the evaluation item of voltage deviation at key nodes, This is a runtime margin evaluation item.

[0079] To ensure that a larger runtime margin results in a smaller objective function value, the runtime margin evaluation term is set as follows:

[0080] in, For positive smoothing terms, For the m-th structurally sensitive unit in the adjustment variable Switching margin below.

[0081] Furthermore, the global objective function can be constructed based on the main layer's operational objectives. When the main layer's operational objective is to reduce operational losses, the global objective function includes a network loss term; when the main layer's operational objective is to improve voltage stability, the global objective function includes a critical node voltage deviation term and a voltage margin term; when the main layer's operational objective is to improve resource utilization efficiency, the global objective function includes a resource allocation cost term and a smoothness adjustment term. Different objective terms can be assigned corresponding weights according to the current scheduling strategy.

[0082] In one implementation, the main layer sorts candidate operating schemes within a constrained optimization space and determines the candidate operating scheme with the optimal global objective function value as the final operating scheme. In another implementation, when the difference in the global objective function values ​​of multiple candidate operating schemes is less than a preset difference, the main layer preferentially selects the candidate operating scheme with a larger solution domain margin that corresponds to a stable and executable structure, thereby enhancing the adaptability of the operating scheme to load fluctuations, distributed power output fluctuations, or measurement errors.

[0083] Furthermore, the step of distributing the execution plan to the configuration layer for execution includes: Determine the corresponding operating mode identifier based on the operating plan; The values ​​of the primary layer optimization variables in the aforementioned operating scheme are converted into control instructions for the co-layer local units; The control command is issued to the corresponding layered local unit; During execution, monitor the operating status of structurally sensitive units to determine whether they remain within the structurally stable executable solution domain under the corresponding operating mode identifier; When the actual operating state is detected to deviate from the corresponding structurally stable and executable solution domain during execution, the main layer operating requirements and the auxiliary layer physical operating state are reacquired, and the next round of main-auxiliary integrated multi-level interactive optimization operation is entered.

[0084] In this embodiment, instead of independently solving the global objective function first and then having the auxiliary layer perform post-validation, the main layer uses the pre-generated structurally stable and executable solution domain as the prior constraint condition for the main layer's optimization variables after the auxiliary layer has pre-generated the structurally stable and executable solution domain. This reduces the problem of the main layer's optimization results becoming unexecutable on the auxiliary layer side due to protection actions, topology switching, control mode switching, or equipment gear switching, thus improving the executability, stability, and closed-loop response efficiency of the integrated main and auxiliary multi-level interactive optimization operation.

[0085] Example 2, Figure 2 The present invention provides a multi-level interactive optimization operation system integrating main and auxiliary components, comprising the following modules: Unit and boundary identification module: used to obtain the main layer operation requirements and the physical operation status of the auxiliary layer, and to identify the sensitive units of the auxiliary layer structure and their operation status switching boundaries; Executable solution domain generation module: used to divide the physical constraint space into modes and intervals based on the switching boundary of the structurally sensitive unit, and generate a structurally stable executable solution domain with running mode identifier; Constrained optimization space construction module: used to upload the structurally stable and executable solution domain to the main layer, and to build an optimization space with restricted variable values ​​on the main layer side; Global Objective Optimization Module: Used by the main layer to solve the global objective function within a constrained optimization space, determine the operating scheme that satisfies the stability constraints of the layered structure, and issue it for execution.

[0086] Figure 3 The diagram illustrates the hierarchical collaboration and data flow between the main layer and the auxiliary layers. The upper part of the diagram represents the main layer, and the lower part represents the auxiliary layers. The black area on the left of the auxiliary layer indicates boundary identification, used to identify structurally sensitive units and their operational state switching boundaries. The black area on the right indicates executable solution domain generation, used to form a structurally stable executable solution domain with operational mode identifiers based on the operational state switching boundaries. The auxiliary layer uploads the executable solution domain to the main layer. The main layer then constructs a constrained optimization space based on this domain and optimizes within this space according to the global objective. After determining the optimal operating scheme, it sends it down to the auxiliary layer for execution, thereby achieving collaboration between the main layer's optimization and the auxiliary layer's physical executable range.

[0087] Figure 4 The diagram illustrates that under a fixed candidate operating mode and its physical constraint structure, the zero-disturbance value is used as the reference point. The single adjustment dimension is stepped up step by step in the forward / reverse direction with the minimum step size. After each step, the constraint equation is solved to detect the constraint satisfaction state. When any constraint limit boundary is reached, the forward boundary point and the reverse boundary point are recorded respectively, thereby determining the initial feasible solution interval corresponding to the adjustment dimension.

[0088] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0089] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0090] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0091] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0092] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0093] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multi-level interactive optimization operation method integrating main and auxiliary systems, characterized in that: Includes the following steps: Obtain the main layer's operational requirements and the physical operational status of the secondary layers, and identify the sensitive units of the secondary layer structure and their operational status switching boundaries; The layering uses the switching boundary of the structurally sensitive unit as the boundary to divide the physical constraint space into modes and split the intervals, generating a structurally stable and executable solution domain with operating mode identifiers; Upload the structurally stable and executable solution domain to the main layer, and build an optimization space with restricted variable values ​​on the main layer side; The main layer solves the global objective function within a constrained optimization space, determines the operating scheme that satisfies the stability constraints of the layered structure, and issues it for execution.

2. The integrated multi-level interactive optimization operation method for main and auxiliary systems according to claim 1, characterized in that, The identification of sensitive units of the layered structure and their operating state switching boundaries includes: Map the main layer's operational requirements to the supporting layer's physical variable space to determine their direction of action on the supporting layer's local units; Candidate units are selected based on the direction of action, proximity protection triggering, operation mode switching, or topology switching conditions. The candidate units are subjected to disturbance evolution judgment. Candidate units that undergo structural changes within the preset disturbance range are identified as structurally sensitive units, and the critical conditions that trigger structural changes are identified as the operating state switching boundary.

3. The integrated multi-level interactive optimization operation method for main and auxiliary systems according to claim 2, characterized in that, The process of mapping the main layer's operational requirements to the supporting layer's physical variable space includes: Construct a directed physical relationship graph based on the current primary and secondary network topology and power flow direction; Using the boundary node corresponding to the adjustment demand component as the source node, candidate propagation paths for the adjustment demand component are searched under the constraint of directional consistency. The candidate propagation paths are verified for capacity carrying capacity and operational constraint continuity. Propagation paths that cannot carry the adjustment demand or will cause path interruption are eliminated, and the set of physical propagation paths of the adjustment demand component in the layered network is obtained.

4. The integrated multi-level interactive optimization operation method for main and auxiliary systems according to claim 3, characterized in that, The candidate elements that will undergo structural changes within a preset disturbance range are identified as structurally sensitive elements, including: Candidate units with protection triggering, operation mode switching or topology switching mechanisms are selected, and critical state intervals are constructed based on the operation state threshold intervals of the candidate units. When the current operating state of a candidate unit enters the critical state interval, the state evolution simulation of the candidate unit is carried out in combination with the direction of action of the main layer's operating requirements in the physical variable space of the auxiliary layer. If a candidate unit triggers a structural change within a preset adjustment disturbance range, it is identified as a structurally sensitive unit, and the corresponding triggering condition is identified as the operating state switching boundary.

5. The integrated multi-level interactive optimization operation method for main and auxiliary systems according to claim 4, characterized in that, The generation of a structurally stable executable solution domain with a runtime mode identifier includes: Based on the operational state switching boundary of the structurally sensitive unit, the operational mode is divided into the physical constraint space of the layer, and the corresponding state is fixed in each operational mode. The initial feasible solution domain within the mode is solved using the main layer adjustment variable as the parameter variable, and the structural trigger points that trigger the switching of the operating state of the structurally sensitive unit are identified. The initial feasible solution domain is divided into intervals based on the structural trigger point. The sub-feasible solution domain that satisfies the preset physical operation constraints and maintains the unchanged operation mode after the division is determined as the structurally stable executable solution domain, and an operation mode identifier is configured.

6. The integrated multi-level interactive optimization operation method for main and auxiliary systems according to claim 5, characterized in that, The initial feasible solution domain within the solution mode includes: Using the zero-disturbance value of the current main layer adjustment variable as the initial reference point, and under the physical constraint structure within the mode corresponding to the fixed candidate operation mode, the boundary values ​​of the preset physical operation constraints are searched along each adjustment dimension. Candidate feasible intervals are formed based on the boundary values ​​of each adjustment dimension. The candidate feasible intervals of the multidimensional adjustment variables are combined and verified to eliminate regions that trigger physical constraint conflicts or operation state switching, thus obtaining the initial feasible solution domain.

7. The integrated multi-level interactive optimization operation method for main and auxiliary systems according to claim 6, characterized in that, The construction of an optimization space with restricted variable values ​​on the main layer side includes: The layer identifies the structurally stable executable solution domain according to the running mode state, and establishes corresponding running mode identifiers for continuous executable intervals, discrete executable intervals, or discrete candidate points. After receiving the structurally stable executable solution domain with the operation mode identifier, the main layer constructs an adjustment subspace according to the operation mode identifier, so that the range of variable values ​​in each adjustment subspace is limited by the structurally stable executable solution domain under the corresponding operation mode identifier.

8. The integrated multi-level interactive optimization operation method for main and auxiliary systems according to claim 7, characterized in that, The optimization space with restricted variable values ​​also includes: Establish a mapping relationship between the principal-level optimization variables and the structurally stable and executable solution domain within each adjustment subspace; When the structurally stable executable solution domain is a continuous interval, a discrete point set, or a multidimensional region, the value boundaries, candidate value sets, or feasible regions of variable combinations of the principal layer optimization variables are determined respectively. Perform a consistency check on the candidate values ​​or variable combinations to eliminate variable combinations that span different operating mode identifiers and do not meet the mode switching conditions, thus obtaining an optimization space with limited variable values.

9. The integrated multi-level interactive optimization operation method for main and auxiliary systems according to claim 8, characterized in that, The process of determining and executing an operational scheme that satisfies the stability constraints of the layered structure includes: Generate candidate execution plans within a limited optimization space; Eliminate candidate running schemes that do not meet the consistency verification of the running mode; The global objective function value is calculated for the candidate running schemes that pass the verification, and the running scheme that satisfies the main layer running objective and the stability constraints of the secondary layer structure is selected and sent to the secondary layer for execution.

10. A main-supplier integrated multi-level interactive optimization operation system, used to execute the main-supplier integrated multi-level interactive optimization operation method according to any one of claims 1-9, characterized in that, include: Unit and boundary identification module: used to obtain the main layer operation requirements and the physical operation status of the auxiliary layer, and to identify the sensitive units of the auxiliary layer structure and their operation status switching boundaries; Executable solution domain generation module: used to divide the physical constraint space into modes and intervals based on the switching boundary of the structurally sensitive unit, and generate a structurally stable executable solution domain with running mode identifier; Constrained optimization space construction module: used to upload the structurally stable and executable solution domain to the main layer, and to build an optimization space with restricted variable values ​​on the main layer side; Global Objective Optimization Module: Used by the main layer to solve the global objective function within a constrained optimization space, determine the operating scheme that satisfies the stability constraints of the layered structure, and issue it for execution.

Images