A simulation method, system, device and medium for a production operation support system side device

Abstract

The application discloses a kind of production operation support system side equipment simulation method, system, equipment and medium, belong to electric power system equipment simulation technical field, including: first graph structure is constructed, based on first graph structure screening history data extracts historical violation propagation relationship, based on historical violation propagation relationship to first graph structure executes structure change operation, based on the first graph structure after structure change second graph structure is constructed and establishes structure enhancement edge, in first graph structure when executing derivation chain search, along the structure enhancement edge in second graph structure expansion obtains cross graph constraint set, based on cross graph constraint set executes constraint solving and obtains simulation prediction result.The application is realized by establishing structure enhancement edge derivation relationship reconstruction and violation propagation structure enhancement synchronization, derivation chain search expands along structure enhancement edge and improves evaluation coverage, cross graph interlocking verification excludes the derivation chain of triggering self feedback loop, interlocking release operation adjusts derivation strategy according to simulation result.

Description

Technical Field

[0001] This invention relates to the field of power system equipment simulation technology, specifically to a simulation method, system, equipment, and medium for side-side equipment of a production operation support system. Background Technology

[0002] In power system operation, risk assessment of side-side equipment relies on real-time monitoring of various physical quantities, including parameters such as power, voltage, and current. These parameters are constrained by physical laws and equipment characteristics, such as power conservation, Ohm's law, and equipment capacity limitations. When sensor failures or communication interruptions render some parameters unobservable, existing technologies derive calculations based on the physical relationships between parameters to recover the missing data. However, power systems exhibit complex dynamic evolution characteristics under abnormal operating conditions. An anomaly in a single parameter can trigger a chain reaction over time through physical coupling and control response, forming a fault propagation chain. Existing derivation methods only focus on the static physical relationships between parameters, neglecting the temporal patterns of anomaly propagation. This can lead to derivation results that contradict historical anomaly propagation patterns, deviating from the actual evolutionary trajectory of the system. Summary of the Invention

[0003] In view of the above-mentioned problems, the present invention provides a simulation method, system, equipment and medium for side-side equipment of a production operation support system.

[0004] Therefore, the technical problem solved by this invention is: when sensor failure leads to missing observation data, how to integrate the temporal patterns of historical anomaly propagation in the derivation calculation process, so that the derivation strategy is adapted to the dynamic evolution characteristics of the system, and avoid the derivation results triggering a chain reaction that is inconsistent with the real system.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a simulation method for side-side equipment of a production operation support system, comprising,

[0006] Construct a first graph structure, with the first constraint relation expression as the first node and the directed edge representing the derivation relationship between the first nodes as the first directed edge;

[0007] Based on the structure of the first graph, historical data is filtered to extract the historical violation propagation relationships between the second constraint relationship expressions;

[0008] Based on the historical violation of the propagation relationship, a structural change operation is performed on the first graph structure, which alters the first directed edge in the first graph structure.

[0009] The second graph structure is constructed based on the first graph structure after structural changes. The second graph structure uses the second constraint relationship expression as the second node and the directed edge representing the historical violation propagation relationship between the second nodes as the second directed edge. During the construction process, the second graph structure establishes structural enhancement edges based on the structural changes of the first graph structure.

[0010] The derivation chain search is performed in the first graph structure. When searching for derivation paths, the derivation chain search is extended along the structural enhancement edges in the second graph structure to obtain the cross-graph constraint set.

[0011] A simulation constraint set is constructed based on the cross-graph constraint set, and the simulation prediction results are obtained by performing constraint solving.

[0012] As a preferred embodiment of the simulation method for side-side equipment of a production operation support system according to the present invention, the structural change operation includes:

[0013] Identify the first directed edge in the first graph structure that is inconsistent with the historical violation propagation relationship;

[0014] In response to the identification of a first directed edge that is inconsistent with the historical violation propagation relationship, the identified first directed edge is deleted. A structural association pointer is established between the starting node and the ending node of the deleted first directed edge. The structural association pointer records the path formed by other first directed edges in the first graph structure from the starting node to the ending node.

[0015] As a preferred embodiment of the simulation method for side-side equipment of a production operation support system according to the present invention, wherein: the second graph structure establishes structural enhancement edges based on the structural changes of the first graph structure during the construction process, including: identifying second directed edges between start-point nodes and end-point nodes in the second graph structure that represent the same constraint relationship expression in the second graph structure;

[0016] Identify other second directed edges starting from the second node, the endpoint of the second directed edge;

[0017] In response to the structural association pointing between the starting second node and the ending second node of other second directed edges, which represent the same constraint relationship expression in the first graph structure, a structural reinforcement edge is established from the second directed edge to other second directed edges.

[0018] As a preferred embodiment of the simulation method for side-side equipment of a production operation support system according to the present invention, wherein: the derivation chain search, when searching for the derivation path, expands along the structural enhancement edge in the second graph structure to obtain the cross-graph constraint set, including: searching for the derivation path in the first graph structure;

[0019] In response to the existence of a structural association pointing to the current first node and the predecessor first node in the derivation path, check whether there is a structural reinforcement edge between the second directed edge between the starting node and the ending node of the structural association pointing to the second node in the second graph structure that represents the same constraint relationship expression.

[0020] In response to the existence of a structurally reinforcing edge, a second reachable node is searched along the structurally reinforcing edge, and the constraint relation expression represented by the searched second node is added to the cross-graph constraint set of the derivation chain.

[0021] As a preferred embodiment of the simulation method for side-side equipment of a production operation support system according to the present invention, the method further includes, before constructing the simulation constraint set based on the cross-graph constraint set, performing cross-graph interlock verification on the searched derivation chain, wherein the cross-graph interlock verification includes:

[0022] Substitute the derivation result of the derivation chain into the constraint relationship expression represented by the second node of the second graph structure to identify the violated second node;

[0023] Propagate traversal from the violated second node along the second directed edge and the structural reinforcement edge;

[0024] Check whether the second node of the starting point of the structural enhancement edge traversed by the propagation traversal belongs to the cross-graph constraint set of the derivation chain;

[0025] In response to the set of cross-graph constraints where the second node of the starting point belongs to the derivation chain, it is determined that a cross-graph interlocking loop has been formed and the derivation chain is deleted.

[0026] As a preferred embodiment of the simulation method for side-side equipment of a production operation support system according to the present invention, the method further includes: identifying multiple derivation chains with the same first node at the endpoint before constructing the simulation constraint set based on the cross-graph constraint set;

[0027] Calculate the intersection of the cross-graph constraint sets of multiple derivation chains;

[0028] Starting from the second node represented by the constraint relation expression in the intersection in the second graph structure, search for the reachable second node along the structure-enhancing edges;

[0029] Count the number of overlapping second nodes found in multiple derivation chains;

[0030] In response to the fact that the ratio of the number of overlapping nodes to the total number of second nodes found is within a preset range, the cross-graph constraint sets of multiple derivation chains are merged to construct a simulation constraint set.

[0031] As a preferred embodiment of the simulation method for side equipment of a production operation support system of the present invention, wherein: in response to the simulation prediction result obtained by performing constraint solving based on the simulation constraint set constructed based on the cross-graph constraint set not meeting the preset conditions, an interlock release operation is performed on the first graph structure and the second graph structure;

[0032] The interlock release operation includes: restoring the deleted first directed edge in the first graph structure and deleting the structure association pointer;

[0033] Remove the structural reinforcement edges from the structure in the second diagram.

[0034] Based on the first graph structure after the interlock is released, the historical data is re-filtered, and structural change operations, second graph structure construction, and derivation chain search are performed to obtain the cross-graph constraint set.

[0035] This invention provides a simulation system for side-side equipment of a production operation support system.

[0036] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a simulation system for side-side equipment of a production operation support system, comprising:

[0037] The graph structure construction module is used to construct the first graph structure and the second graph structure. The second graph structure is constructed based on the first graph structure by filtering historical data.

[0038] The structural interlock module is used to perform structural interlock operations on the first graph structure and the second graph structure. The structural interlock operation causes structural changes in the first graph structure to trigger the establishment of structural reinforcement edges in the second graph structure.

[0039] The derivation chain search module is used to perform derivation chain search based on the first graph structure and the second graph structure after structural interlocking. When searching for derivation paths in the first graph structure, the derivation chain search extends along the structural enhancement edges in the second graph structure to obtain the cross-graph constraint set.

[0040] The simulation solution module is used to construct a set of simulation constraints based on the cross-graph constraint set, and to perform constraint solving to obtain simulation prediction results.

[0041] This invention provides a computer device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of a simulation method for a side device of a production operation support system.

[0042] The present invention provides a computer-readable storage medium storing a computer program thereon, wherein the computer program, when executed by a processor, implements the steps of a simulation method for a side device of a production operation support system.

[0043] The beneficial effects of this invention are as follows: This invention establishes a propagation enhancement association by adjusting the derivation association of the constraint derivation relationship graph to trigger the constraint violation propagation graph. This enables the reconstruction of the constraint derivation relationship and the enhancement of the constraint violation propagation structure to be completed simultaneously. This allows the construction of the derivation chain to incorporate the temporal propagation characteristics of historical constraint violations, thereby improving the degree of conformity between the derivation chain and the actual operating law of the power system.

[0044] When searching for derivation paths in the constraint derivation graph, the derivation chain search enhances the correlation extension along the constraint violation propagation graph to obtain the derivation coverage constraint set. This expands the constraint range covered by the derivation chain from constraints directly associated with the derivation path to constraints reachable by constraint violation propagation, thereby improving the coverage of the simulation constraint set constructed based on the derivation chain for evaluating the power system operating state.

[0045] Cross-graph interlock verification simulates constraint violation propagation by substituting the derivation results of the derivation chain into the constraint violation propagation graph. It detects whether the constraint violation propagation caused by the derivation results forms a loop through propagation enhancement association and returns to the constraint range covered by the derivation chain. It identifies and eliminates derivation chains that trigger self-feedback propagation loops, reducing the risk that the derivation chain will cause the power system operation status assessment to deviate from the actual evolution path.

[0046] When the simulation prediction results do not meet the preset conditions, the interlock release operation restores the original derivation association of the constraint derivation relationship graph and removes the propagation enhancement association of the constraint violation propagation graph. By re-screening historical data and reconstructing the constraint derivation relationship graph and the constraint violation propagation graph, the derivation chain selection strategy is adjusted according to the feedback of the simulation evaluation results, which improves the adaptability of the derivation chain construction method to changes in the operating conditions of the power system. Attached Figure Description

[0047] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a general flowchart of a simulation method for a side-side device of a production operation support system, provided as an embodiment of the present invention.

[0049] Figure 2 This is a flowchart illustrating the structural change operation of a simulation method for side-side equipment in a production operation support system, as provided in one embodiment of the present invention.

[0050] Figure 3 The flowchart illustrates a simulation method for side-side equipment in a production operation support system, as provided in one embodiment of the present invention, for establishing structural enhancement. Detailed Implementation

[0051] To make the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0052] Example 1, referring to Figures 1-3 This is one embodiment of the present invention, which provides a simulation method for side-side equipment of a production operation support system, including:

[0053] S1: Construct the first graph structure, with the first constraint relationship expression as the first node and the directed edge representing the derivation relationship between the first nodes as the first directed edge.

[0054] In some embodiments, the edge-side monitoring device configures a first constraint relationship expression to assess the operational risk of the wind farm's edge-side energy storage system. The edge-side monitoring device acquires the operating parameters of the wind farm's edge-side energy storage system, substitutes these parameters into the first constraint relationship expression for a satisfaction check, and identifies any violated first constraint relationship expressions. The edge-side monitoring device determines a risk level based on the number and severity of the violated first constraint relationship expressions; this risk level characterizes the operational risk of the wind farm's edge-side energy storage system.

[0055] Understandably, operating parameters include physical quantities reflecting the operating status of the energy storage system, such as wind turbine power generation, energy storage charging and discharging power, inverter output voltage, and inverter output current. By substituting the operating parameters into the first constraint expression for a satisfaction check, it is possible to identify which constraint expressions are violated. The more constraint expressions are violated, or the more severe the violations, the higher the determined risk level.

[0056] For example, the first constraint relationship expression includes a power balance constraint describing the balance relationship between wind turbine power generation, energy storage charging and discharging power, and grid-connected output power; an inverter power constraint describing the calculation relationship between energy storage discharging power and inverter output voltage and output current; an inverter voltage constraint describing the safe range of inverter output voltage; an inverter current constraint describing the safe range of inverter output current; and an energy storage power constraint describing the safe range of energy storage charging and discharging power.

[0057] Furthermore, the side-side monitoring device identifies the constraint physical type identifier for each first constraint relationship expression. For example, the constraint physical type identifier for the power balance constraint and the inverter power constraint is power conservation type, and the constraint physical type identifier for the inverter voltage constraint is voltage limit type.

[0058] Furthermore, the edge monitoring device traverses the first constraint relationship expressions and identifies the derivation relationships between the first constraint relationship expressions based on the parameter association relationships of the first constraint relationship expressions. In response to two first constraint relationship expressions having common parameters and the parameters of the endpoint first constraint relationship expression being derived from the parameters of the starting first constraint relationship expression through algebraic transformation, the edge monitoring device identifies a derivation relationship between the starting first constraint relationship expression and the endpoint first constraint relationship expression.

[0059] In some embodiments, the edge monitoring device labels the identified derivation relationships with operation nesting feature identifiers. The edge monitoring device analyzes the dependencies of each algebraic transformation operation during the derivation process and sets operation nesting feature identifiers for the derivation relationships according to the execution order and nesting level of the algebraic transformation operations.

[0060] In response to the sequential execution of algebraic transformation operations and the fact that the input of each operation depends only on the output of the previous operation, the side-side monitoring device sets the operation nesting feature identifier of the derivation relationship to sequential type. For example, when deriving the energy storage charging power from the power balance constraint, the energy storage charging power is calculated sequentially through algebraic transformations of the power balance constraint, and the side-side monitoring device sets the operation nesting feature identifier of the derivation relationship to sequential type.

[0061] In response to the fact that the input of a certain algebraic transformation operation depends on the output of multiple operations, the side-side monitoring device sets the operation nesting feature identifier of the derivation relationship to nested type. For example, when deriving the inverter output current boundary from power balance constraints and inverter voltage constraints, the calculation of the inverter output current boundary depends simultaneously on the energy storage discharge power derived from the power balance constraints and the inverter output voltage range defined by the inverter voltage constraints. In this case, the side-side monitoring device sets the operation nesting feature identifier of the derivation relationship to nested type.

[0062] Furthermore, the edge monitoring device constructs a first graph structure based on the first constraint relation expression and the derivation relation. The edge monitoring device constructs each first constraint relation expression as a first node in the first graph structure, and the first node contains the first constraint relation expression and its attached constraint physical type identifier. The edge monitoring device constructs the derivation relation as a directed edge connecting the first nodes, which serves as the first directed edge. The first directed edge connects the starting first node to the ending first node, and the first directed edge contains the operation nesting feature identifier of the derivation relation.

[0063] S2: Based on the structure of the first graph, filter the historical data and extract the historical violation propagation relationship between the second constraint relationship expressions.

[0064] In some embodiments, the edge monitoring device extracts the historical sequence of output levels within a historical runtime period. Within the historical runtime period, the edge monitoring device performs a satisfaction check on the first constraint expression based on the operating parameters, and outputs a risk level based on the number and severity of violations of the first constraint expression. The risk levels are arranged into a four-level sequence from low to high severity, including level one, level two, level three, and level four. The edge monitoring device extracts the risk level sequence output within the historical runtime period from the historical operating data as the historical sequence of output levels.

[0065] Furthermore, the edge monitoring equipment identifies level transition moments in the historical output level sequence. The edge monitoring equipment traverses the historical output level sequence, and in response to the difference in risk levels between two adjacent moments, marks the subsequent moment as a level transition moment.

[0066] Furthermore, the edge monitoring equipment extracts the constraint violation sequence corresponding to the level transition time. The edge monitoring equipment extracts the operating parameters of the level transition time from historical operating data, substitutes the operating parameters into the first constraint relationship expression for satisfaction check, identifies the violated first constraint relationship expression and its violation time, and forms the constraint violation sequence.

[0067] In some embodiments, the edge monitoring device filters constraint violation sequences based on a first graph structure. The edge monitoring device traverses the violated first constraint relationship expressions recorded in the constraint violation sequence and searches for the corresponding first node in the first graph structure. If a violated first constraint relationship expression has a corresponding first node in the first graph structure, the edge monitoring device retains the constraint violation record. If a violated first constraint relationship expression does not have a corresponding first node in the first graph structure, the edge monitoring device deletes the constraint violation record.

[0068] Furthermore, the edge monitoring device extracts historical violation propagation relationships from the filtered constraint violation time series. The edge monitoring device iterates through the filtered constraint violation time series, identifying constraint violation sequence patterns that repeatedly occur across multiple level transition events. In response to the violation time of the starting first constraint relationship expression being earlier than the violation time of the ending first constraint relationship expression, and the sequence pattern repeating across multiple level transition events, the edge monitoring device identifies a historical violation propagation relationship between the starting first constraint relationship expression and the ending first constraint relationship expression.

[0069] For example, fluctuations in wind turbine output power cause a violation of power balance constraints at time t1, and changes in energy storage discharge power cause a violation of inverter power constraints at time t2. Time t2 is later than time t1. The sequential pattern repeats in multiple level transition events. Side-side monitoring equipment identifies a historical violation propagation relationship between power balance constraints and inverter power constraints.

[0070] In some embodiments, the edge-side monitoring device labels the derivation type of the extracted historical violation propagation relationship based on the first directed edge pair in the first graph structure. The edge-side monitoring device traverses the extracted historical violation propagation relationship, and for the starting point first constraint relationship expression and the ending point first constraint relationship expression of the historical violation propagation relationship, it searches in the first graph structure whether there exists a first directed edge from the first node corresponding to the starting point first constraint relationship expression to the first node corresponding to the ending point first constraint relationship expression.

[0071] In response to the existence of a first directed edge in the first graph structure, from the first node corresponding to the first constraint expression at the starting point to the first node corresponding to the first constraint expression at the ending point, the edge-side monitoring device reads the operation nesting feature identifier contained in the first directed edge. If the operation nesting feature identifier is sequential, the edge-side monitoring device labels the derivation type of the historical violation propagation relationship as sequential derivation type. If the operation nesting feature identifier is nested, the edge-side monitoring device labels the derivation type of the historical violation propagation relationship as nested derivation type.

[0072] Furthermore, the side-side monitoring equipment defines the first constraint relationship expression involved in the historical violation propagation relationship as the second constraint relationship expression. The second constraint relationship expression is a subset of the first constraint relationship expression, and it includes the first constraint relationship expression that appears as the start or end point in the historical violation propagation relationship.

[0073] S3: Perform a structural change operation on the first graph structure based on the historical violation propagation relationship. The structural change operation changes the first directed edge in the first graph structure.

[0074] In some embodiments, the edge monitoring device performs a structural change operation on the first graph structure based on the historical violation propagation relationship extracted in step S2. The structural change operation adjusts the first directed edges in the first graph structure to ensure that the derivation relationship in the first graph structure is consistent with the historical violation propagation relationship.

[0075] In some embodiments, such as Figure 2 As shown, the structural change operation in step S3 includes the following steps:

[0076] S31: Identify the first directed edge in the first graph structure that is inconsistent with the historical violation propagation relationship.

[0077] In some embodiments, the specific implementation of step S31, which identifies the first directed edge in the first graph structure that is inconsistent with the historical violation propagation relationship, includes the following steps:

[0078] The edge monitoring device traverses the first directed edges in the first graph structure. For each first directed edge, the edge monitoring device extracts the first constraint relationship expression corresponding to the first node of the starting point and the first constraint relationship expression corresponding to the first node of the ending point of the first directed edge.

[0079] Furthermore, the edge monitoring device searches for historical violation propagation relationships from the starting point first constraint relationship expression to the ending point first constraint relationship expression in the historical violation propagation relationships extracted in step S2. In response to the existence of a corresponding historical violation propagation relationship, the edge monitoring device reads the derivation type label of the historical violation propagation relationship.

[0080] Furthermore, the edge monitoring device reads the operation nesting feature identifier contained in the first directed edge and compares the consistency between the operation nesting feature identifier and the derivation type annotation. If the operation nesting feature identifier is sequential and the derivation type annotation is sequential derivation type, the edge monitoring device marks the first directed edge as having passed type verification. If the operation nesting feature identifier is nested and the derivation type annotation is nested derivation type, the edge monitoring device marks the first directed edge as having passed type verification.

[0081] In response to the inconsistency between the operation nesting feature identifier and the derivation type label, the edge monitoring device marks the first directed edge as a type verification conflict. In response to the fact that the first directed edge does not have a corresponding historical violation propagation relationship in the historical violation propagation relationship, the edge monitoring device marks the first directed edge as historical missing.

[0082] Furthermore, the edge monitoring device will identify the first directed edge marked as a type verification conflict as the first directed edge that is inconsistent with the historical violation propagation relationship.

[0083] For example, the first directed edge from the power balance constraint to the inverter power constraint is identified as sequential in its operation nesting feature. If a historical violation propagation relationship exists between the power balance constraint and the inverter power constraint, the derivation type is labeled as sequential derivation. Since the operation nesting feature identifier matches the derivation type label, the edge-side monitoring device marks this first directed edge as having passed type verification.

[0084] S32: In response to the identification of a first directed edge that is inconsistent with the historical violation propagation relationship, delete the identified first directed edge, establish a structural association pointer between the starting node and the ending node of the deleted first directed edge, and record the path formed by other first directed edges in the first graph structure from the starting node to the ending node.

[0085] In some embodiments, in step S32, in response to identifying a first directed edge inconsistent with the historical violation propagation relationship, the identified first directed edge is deleted, and a structural association pointer is established between the start node and the end node of the deleted first directed edge. The structural association pointer records the path formed by other first directed edges in the first graph structure from the start node to the end node. The specific implementation includes the following steps:

[0086] The edge monitoring device performs path decomposition on the first directed edge marked as having a type verification conflict in step S31. The edge monitoring device extracts the first node of the starting point and the first node of the ending point of the first directed edge marked as having a type verification conflict, and searches for the path formed by other first directed edges in the first graph structure from the first node of the starting point to the first node of the ending point.

[0087] Furthermore, the edge monitoring device requires that all first directed edges on the path formed by other first directed edges be marked as having passed type verification in step S31. The edge monitoring device extracts the operation nesting feature identifiers contained in all first directed edges on the path formed by other first directed edges. In response to the fact that all operation nesting feature identifiers contained in the path formed by other first directed edges are sequential, the edge monitoring device determines that the derivation type of the path formed by other first directed edges is sequential. In response to the fact that there are first directed edges on the path formed by other first directed edges containing nested operation nesting feature identifiers, the edge monitoring device determines that the derivation type of the path formed by other first directed edges is nested.

[0088] Furthermore, the edge-side monitoring device reads the starting first constraint expression corresponding to the starting first node and the ending first constraint expression corresponding to the ending first node of the first directed edge marked as having a type verification conflict. It then searches for historical violation propagation relationships from the starting first constraint expression to the ending first constraint expression in the historical violation propagation relationships extracted in step S2, and reads the derivation type annotation of these historical violation propagation relationships. The edge-side monitoring device compares the consistency of the derivation type of the paths formed by other first directed edges with the derivation type annotation. In response to the consistency between the derivation type of the paths formed by other first directed edges and the derivation type annotation, the edge-side monitoring device confirms that the paths formed by the other first directed edges are paths that meet the requirements.

[0089] Upon finding a path that meets the requirements, the edge monitoring device deletes the first directed edge marked as a type verification conflict. The edge monitoring device establishes a structural association pointer between the starting and ending nodes of the deleted first directed edge. This structural association pointer records the sequence of first nodes and the sequence of first directed edges traversed by the path formed by the other first directed edges in the first graph structure from the starting node to the ending node.

[0090] For example, the first directed edge from the power balance constraint to the inverter voltage constraint is marked as a type verification conflict in step S31. The edge-side monitoring device searches for paths formed by other first directed edges in the first graph structure from the first node corresponding to the power balance constraint to the first node corresponding to the inverter voltage constraint. The edge-side monitoring device finds the following other paths formed by first directed edges: the first directed edge from the first node corresponding to the power balance constraint to the first node corresponding to the inverter power constraint, and the first directed edge from the first node corresponding to the inverter power constraint to the first node corresponding to the inverter voltage constraint. Both first directed edges on these other paths are marked as type verification passed in step S31, and the derivation type of these other paths is consistent with the derivation type marking of the historical violation propagation relationship from the power balance constraint to the inverter voltage constraint in step S2. The side-side monitoring device deletes the first directed edge from the power balance constraint to the inverter voltage constraint, establishes a structural association pointer between the first node corresponding to the power balance constraint and the first node corresponding to the inverter voltage constraint, and records the sequence of first nodes and first directed edges traversed by the path formed by other first directed edges.

[0091] In response to the absence of a path that meets the requirements in the first graph structure, the edge monitoring device retains the first directed edge marked as a type verification conflict and marks it as pending verification.

[0092] In step S3, the present invention identifies the first directed edge in the first graph structure that is inconsistent with the historical violation propagation relationship and performs derivation path decomposition. By searching for paths formed by other first directed edges in the first graph structure and establishing structural associations, the derivation relationship in the first graph structure is kept consistent with the historical violation propagation relationship. This avoids the derivation chain built based on inconsistent derivation relationships from triggering constraint violation propagation that is inconsistent with the actual operating rules of the wind farm side-side energy storage system.

[0093] S4: Construct a second graph structure based on the first graph structure after structural changes. The second graph structure uses the second constraint relationship expression as the second node and the directed edges representing the historical violation propagation relationship between the second nodes as the second directed edges. During the construction process, the second graph structure establishes structural enhancement edges based on the structural changes of the first graph structure.

[0094] In some embodiments, the edge monitoring device constructs a second graph structure based on the first graph structure after the structural change in step S3. The edge monitoring device then re-filters the constraint violation sequences extracted in step S2 based on the first graph structure after the structural change, constructs a set of confirmatory violation propagation associations, and constructs the second graph structure based on this set of confirmatory violation propagation associations.

[0095] Furthermore, the side monitoring device traverses the constraint violation sequence extracted in step S2. For two adjacent violated first constraint relationship expressions in the constraint violation sequence, it checks whether there is a first directed edge or structural association pointing between the first nodes corresponding to the two violated first constraint relationship expressions in the first graph structure after structural change.

[0096] In response to the existence of a first directed edge in the first graph structure after structural changes, from the first node corresponding to the first constraint relationship expression at the starting point to the first node corresponding to the first constraint relationship expression at the ending point, and the first directed edge being marked as having passed type verification in step S31, the edge-side monitoring device retains records of the two violated first constraint relationship expressions in the constraint violation sequence. In response to the existence of a structural association pointing from the first node corresponding to the first constraint relationship expression at the starting point to the first node corresponding to the first constraint relationship expression at the ending point, the edge-side monitoring device retains records of the two violated first constraint relationship expressions in the constraint violation sequence.

[0097] In response to the absence of a first directed edge and no structural association pointing in the first graph structure after the structural change, the edge monitoring device deletes subsequent records of the two violated first constraint relationship expressions.

[0098] Furthermore, the edge monitoring device performs violation propagation association extraction on the re-filtered constraint violation time series to obtain a set of confirmatory violation propagation associations. From the re-filtered constraint violation time series, the edge monitoring device identifies pairs of first constraint relationship expressions where the violation time of the starting first constraint relationship expression is earlier than the violation time of the ending first constraint relationship expression, and uses these pairs of first constraint relationship expressions as historical violation propagation relationships in the set of confirmatory violation propagation associations.

[0099] Furthermore, the edge monitoring device constructs a second graph structure based on the second constraint relation expression defined in step S2 and the set of confirmatory violation propagation associations. The edge monitoring device constructs each second constraint relation expression as a second node in the second graph structure. The edge monitoring device constructs the historical violation propagation relations in the set of confirmatory violation propagation associations as directed edges connecting the second nodes, which are then used as second directed edges. These second directed edges connect the starting second node to the ending second node. The starting second node of the second directed edge corresponds to the starting first constraint relation expression of the historical violation propagation relation, and the ending second node of the second directed edge corresponds to the ending first constraint relation expression of the historical violation propagation relation.

[0100] In some embodiments, such as Figure 3 As shown, the process of constructing the second graph structure in step S4, based on the structural changes of the first graph structure, to establish structurally enhanced edges includes the following steps:

[0101] S41: Identify the start node and end node in the second graph structure. The second directed edge between the second nodes with the same constraint relationship expression in the second graph structure.

[0102] The edge monitoring device traverses the structural association pointers established in step S32, extracting the starting first constraint relation expression corresponding to the starting first node and the ending first constraint relation expression corresponding to the ending first node of the structural association pointer. The edge monitoring device then searches for the second directed edge in the second graph structure, from the second node corresponding to the starting first constraint relation expression to the second node corresponding to the ending first constraint relation expression.

[0103] In response to the existence of a second directed edge in the second graph structure, from the second node corresponding to the first constraint relation expression at the starting point to the second node corresponding to the first constraint relation expression at the ending point, the edge monitoring device marks the second directed edge as a path decomposition source violation.

[0104] For example, in step S32, a structural association is established between the first node corresponding to the power balance constraint and the first node corresponding to the inverter voltage constraint. The side-side monitoring device searches for a second directed edge in the second graph structure from the second node corresponding to the power balance constraint to the second node corresponding to the inverter voltage constraint, and marks the second directed edge as a path decomposition source violation pair.

[0105] S42: Identify other second directed edges starting from the second node, the endpoint of the second directed edge.

[0106] Furthermore, the edge-side monitoring device identifies other second directed edges starting from the endpoint of the second directed edge marked as a path decomposition source violation pair. In the second graph structure, the edge-side monitoring device identifies other second directed edges originating from the endpoint of the second directed edge marked as a path decomposition source violation pair.

[0107] S43: In response to the structural association pointing between the starting second node and the ending second node of other second directed edges, which represent the same constraint relationship expression in the first graph structure, a structural reinforcement edge is established from the second directed edge to other second directed edges.

[0108] Furthermore, the edge monitoring device checks whether the other identified second directed edges meet the conditions for establishing structural reinforcement edges. For each other second directed edge, the edge monitoring device extracts the starting first constraint relationship expression corresponding to the starting second node and the ending first constraint relationship expression corresponding to the ending second node of the other second directed edge.

[0109] The edge monitoring device checks whether there is a structural association pointing from the first node corresponding to the first constraint expression at the starting point to the first node corresponding to the first constraint expression at the ending point in the first graph structure after the structural change. In response to the existence of a structural association pointing from the first node corresponding to the first constraint expression at the starting point to the first node corresponding to the first constraint expression at the ending point in the first graph structure after the structural change, the edge monitoring device establishes a structural reinforcement edge from the path decomposition source violation pair to other second directed edges.

[0110] A structural reinforcement edge connects two second directed edges. The origin of the structural reinforcement edge is the path decomposition source violation pair, and the destination of the structural reinforcement edge is another second directed edge. A structural reinforcement edge indicates that the violation propagation from the path decomposition source violation pair can extend along the structural reinforcement edge to other second directed edges.

[0111] For example, the second directed edge from the second node corresponding to the power balance constraint to the second node corresponding to the inverter voltage constraint is labeled as a path decomposition source violation pair. The edge-side monitoring device, starting from the second node corresponding to the inverter voltage constraint, identifies other second directed edges from the second node corresponding to the inverter voltage constraint to the second node corresponding to the inverter current constraint. In the first graph structure after structural changes, the edge-side monitoring device checks whether there are structural association pointers from the first node corresponding to the inverter voltage constraint to the first node corresponding to the inverter current constraint. In response to the existence of structural association pointers, the edge-side monitoring device establishes structural reinforcement edges from the path decomposition source violation pair to other second directed edges.

[0112] Through step S4, the present invention re-screens the constraint violation timing based on the first graph structure after structural changes and constructs a second graph structure. By identifying the second directed edge corresponding to the structure associated with the first graph structure and establishing a structure enhancement edge, the constraint violation propagation path in the second graph structure is kept synchronized with the derivation path of the first graph structure. This avoids the chain reaction triggered by the derivation chain constructed based on the violation propagation relationship before adjustment, which is inconsistent with the actual fault propagation law of the wind farm side-side energy storage system.

[0113] S5: Perform a derivation chain search in the first graph structure. When searching for derivation paths, the derivation chain search extends along the structural enhancement edges in the second graph structure to obtain the cross-graph constraint set.

[0114] In some embodiments, the edge-side monitoring device identifies a set of key obstructed constraints before performing a derivation chain search in the first graph structure. The edge-side monitoring device extracts data reception anomaly events from the historical operation logs of the edge-side energy storage system, recording missing observations of operating parameters due to sensor malfunctions or communication interruptions.

[0115] Furthermore, the edge monitoring equipment identifies observation-impeded constraints. The edge monitoring equipment iterates through the first constraint relationship expression, and in response to the recording of observation missing events in data reception anomaly events on which the operating parameters upon which the first constraint relationship expression depends, the edge monitoring equipment marks the first constraint relationship expression as an observation-impeded constraint.

[0116] Furthermore, the edge monitoring equipment identifies and derives hindered constraints. Starting from the observed hindered constraints, the edge monitoring equipment traverses the first graph structure in the reverse direction along the first directed edge and the structure association, marking the first constraint relationship expression corresponding to the first node encountered as a hindered constraint.

[0117] Furthermore, the edge monitoring equipment selects first constraint relationship expressions that appear as nodes in both the first and second graph structures from the observed and derived obstruction constraints, forming a set of key obstruction constraints. The edge monitoring equipment then excludes the first constraint relationship expressions from the set of key obstruction constraints, and the remaining first constraint relationship expressions constitute the initial usable constraint set.

[0118] In some embodiments, the edge monitoring device performs constraint satisfaction solving based on an initial set of available constraints. The edge monitoring device constructs a constraint solving problem from the first constraint relation expression in the initial set of available constraints, and obtains an initial feasible value space through a constraint solving algorithm. The edge monitoring device performs a risk assessment on the initial feasible value space and obtains a risk level distribution based on the risk assessment results. The edge monitoring device calculates the ordinal difference between the highest and lowest levels in the four-level sequence of the risk level distribution, using this difference as an initial level span parameter.

[0119] In some embodiments, the edge monitoring device performs a derivation chain search in the first graph structure starting from each first constraint relation expression in the set of critical obstructed constraints. The edge monitoring device takes the first node corresponding to the first constraint relation expression in the set of critical obstructed constraints as the search target node, and searches for a derivation path in the first graph structure along the reverse direction pointed to by the first directed edge and the structure association, starting from the search target node.

[0120] In some embodiments, the derivation chain search in step S5, which expands the cross-graph constraint set along the structural enhancement edges in the second graph structure when searching for the derivation path, includes the following steps:

[0121] S51: Search for the derivation path in the first graph structure.

[0122] The edge monitoring device starts from the target node and follows the reverse search path pointed to by the first directed edge and the structural association. When selecting the next node at each step of the search path, the edge monitoring device selects nodes from the predecessor nodes of the current first node that simultaneously satisfy the conditions for the existence of the path in the derivation graph and the conditions for the existence of the path in the propagation graph as candidate predecessor nodes.

[0123] Furthermore, the edge monitoring device checks the existence conditions of the derivation graph path. The edge monitoring device checks in the first graph structure whether there is a first directed edge or a structural association pointing from the candidate predecessor node to the current first node. If there is a first directed edge from the candidate predecessor node to the current first node and the first directed edge is marked as having passed type verification in step S31, the edge monitoring device determines that the candidate predecessor node satisfies the existence conditions of the derivation graph path. If there is a structural association pointing from the candidate predecessor node to the current first node, the edge monitoring device determines that the candidate predecessor node satisfies the existence conditions of the derivation graph path.

[0124] Furthermore, the edge-side monitoring device checks the propagation graph path existence condition. The edge-side monitoring device checks within the second graph structure whether there is a violation of the propagation path from the second node corresponding to the candidate predecessor node to the second node corresponding to the search target node. In response to the existence of a path connecting the second node corresponding to the candidate predecessor node to the second node corresponding to the search target node along the second directed edge and the structural reinforcement edge within the second graph structure, the edge-side monitoring device determines that the candidate predecessor node satisfies the propagation graph path existence condition.

[0125] Furthermore, the edge monitoring device selects a predecessor node from candidate predecessor nodes that simultaneously satisfy the existence conditions of the derivation graph path and the propagation graph path. In response to the fact that the current first node of the derivation path belongs to the initial available constraint set, the edge monitoring device marks the derivation path as a candidate derivation chain.

[0126] For example, the side-side monitoring device starts its reverse search from the first node corresponding to the inverter current constraint as the target node. The side-side monitoring device identifies the first node corresponding to the inverter voltage constraint as a candidate predecessor node. In the first graph structure, there exists a structural association pointing from the first node corresponding to the inverter voltage constraint to the first node corresponding to the inverter current constraint, satisfying the path existence condition of the derivation graph. In the second graph structure, there exists a path from the second node corresponding to the inverter voltage constraint to the second node corresponding to the inverter current constraint along the second directed edge, satisfying the path existence condition of the propagation graph. The side-side monitoring device selects the first node corresponding to the inverter voltage constraint as the predecessor node of the derivation path.

[0127] S52: In response to the existence of a structural association pointing to the current first node and the predecessor first node in the derivation path, check in the second graph structure whether there is a structural reinforcement edge between the second directed edge between the starting node and the ending node of the structural association pointing to the second node that represents the same constraint relationship expression in the second graph structure.

[0128] Furthermore, the edge monitoring device performs cross-graph expansion on node pairs with structural associations in the derivation path. The edge monitoring device traverses the first node traversed by the derivation path. In response to the existence of structural associations between the current first node and the predecessor first node in the derivation path, the edge monitoring device extracts the starting first constraint relationship expression corresponding to the starting first node and the ending first constraint relationship expression corresponding to the ending first node of the structural association.

[0129] The edge monitoring device searches for a second directed edge in the second graph structure, from the second node corresponding to the first constraint expression at the starting point to the second node corresponding to the first constraint expression at the ending point. The edge monitoring device then checks whether this second directed edge contains a structural reinforcement edge.

[0130] For example, the derivation path shows a structural association pointing from the first node corresponding to the power balance constraint to the first node corresponding to the inverter voltage constraint. The side-side monitoring device searches for a second directed edge in the second graph structure, from the second node corresponding to the power balance constraint to the second node corresponding to the inverter voltage constraint, and checks whether there is a structural reinforcement edge on the second directed edge.

[0131] S53: In response to the existence of a structurally reinforcing edge, search for a second reachable node along the structurally reinforcing edge, and add the constraint relation expression represented by the searched second node to the cross-graph constraint set of the derivation chain.

[0132] Furthermore, in response to the existence of a structurally enhanced edge on the second directed edge, the edge-side monitoring device searches for a reachable second node in the second graph structure along the structurally enhanced edge. The edge-side monitoring device traverses the second graph structure from the second directed edge, along the structurally enhanced edge and the second directed edge, and records the second node encountered.

[0133] The edge monitoring device adds the first constraint expression corresponding to the traversed second node to the cross-graph constraint set of the derivation chain. In response to the absence of a structural reinforcement edge as the starting point of the traversed second node, the edge monitoring device marks the second node as a propagation dependency closure node and terminates further traversal from the second node.

[0134] For example, a structural reinforcement edge exists on the second directed edge from the second node corresponding to the power balance constraint to the second node corresponding to the inverter voltage constraint. The edge-side monitoring device searches along the structural reinforcement edge starting from the second directed edge, traversing to the second directed edge from the second node corresponding to the inverter voltage constraint to the second node corresponding to the inverter current constraint. The edge-side monitoring device adds the inverter current constraint to the cross-graph constraint set of the derivation chain. In response to the absence of a structural reinforcement edge starting from the second node corresponding to the inverter current constraint, the edge-side monitoring device marks the second node corresponding to the inverter current constraint as a propagation dependency closed node.

[0135] Through step S5, the present invention identifies a set of key obstructed constraints caused by abnormal data reception events, searches for derivation paths in the first graph structure starting from the key obstructed constraints, checks whether there are structural reinforcement edges on the second directed edges corresponding to the structural associations in the derivation paths, and expands along the structural reinforcement edges in the second graph structure to obtain a set of cross-graph constraints. This expands the constraint range covered by the derivation chain from constraints directly associated with the derivation path to constraints that violate propagation reachability, thereby improving the consistency between the derivation chain and the actual fault propagation law of the wind farm side-side energy storage system.

[0136] S6: Construct a simulation constraint set based on the cross-graph constraint set, and perform constraint solving to obtain simulation prediction results.

[0137] In some embodiments, before step S6, cross-graph interlock verification is performed on the searched derivation chains, multiple derivation chains with the same first endpoint node are identified and collaboratively analyzed, and an interlock release operation is performed in response to the simulation prediction result not meeting the preset conditions, including the following steps:

[0138] S61: Substitute the derivation result of the derivation chain into the constraint relationship expression represented by the second node of the second graph structure to identify the violated second node.

[0139] The edge monitoring device performs derivation chain probing for each candidate derivation chain. The edge monitoring device adds the first constraint relation expression corresponding to the first node of the endpoint of the candidate derivation chain and the first constraint relation expression in the cross-graph constraint set of the candidate derivation chain to the initial available constraint set constructed in step S5, forming the probing constraint set.

[0140] Furthermore, the edge monitoring device performs constraint satisfaction solving based on the trial constraint set to obtain the trial feasible value space. The edge monitoring device samples parameter combinations from the trial feasible value space and performs derivation calculations along the derivation path of the candidate derivation chain in the first graph structure. The edge monitoring device substitutes the sampled parameter combinations into the first constraint relation expression corresponding to the first node of the candidate derivation chain's starting point, and performs derivation along the first directed edge and structural association direction in the derivation path to obtain the derivation values ​​of the constraint variables of the first constraint relation expression corresponding to the first node of the candidate derivation chain's ending point.

[0141] Furthermore, the edge monitoring device constructs an extended parameter combination containing the derived values ​​of constraint variables. The edge monitoring device substitutes the extended parameter combination into the first constraint relationship expression represented by the second node in the second graph structure to perform a satisfaction check, and identifies the second node corresponding to the first constraint relationship expression violated by the extended parameter combination.

[0142] For example, the first node at the end of the candidate derivation chain is the first node corresponding to the inverter current constraint, and the cross-graph constraint set includes the inverter current constraint. The edge-side monitoring device samples parameter combinations from the trial feasible value space and derives the derivation values ​​of the constraint variables of the inverter current constraint along the derivation path. The edge-side monitoring device substitutes the extended parameter combination containing the derivation values ​​of the inverter current constraint variables into the first constraint relationship expression in the second graph structure to identify that the second node corresponding to the inverter voltage constraint is violated.

[0143] Understandably, the specific process by which the edge monitoring equipment performs risk assessment on the trial feasible value space is as follows: The edge monitoring equipment samples parameter combinations from the trial feasible value space, substitutes the sampled parameter combinations into the first constraint relationship expression in the trial constraint set for satisfaction checks, and identifies the violated first constraint relationship expressions. The edge monitoring equipment determines the risk level corresponding to the sampled parameter combinations based on the number and severity of the violated first constraint relationship expressions. The edge monitoring equipment determines the risk level for multiple sampled parameter combinations in the trial feasible value space, counts the number of times each risk level appears in the four-level sequence, and obtains the risk level distribution.

[0144] S62: Propagate traversal from the violated second node along the second directed edge and the structural reinforcement edge.

[0145] Furthermore, the edge-side monitoring device starts from the violated second node and performs a propagation traversal in the second graph structure. Starting from the violated second node, the edge-side monitoring device traverses the second graph structure in the forward direction along the second directed edge and the structural reinforcement edge, recording the second directed edge and structural reinforcement edge traversed.

[0146] For example, the side-side monitoring device starts from the second node corresponding to the inverter voltage constraint, traverses along the second directed edge to the second node corresponding to the inverter current constraint, and continues to traverse along the structural reinforcement edge.

[0147] S63: Check whether the second node of the starting point of the structure-enhancing edge traversed by the propagation traversal belongs to the cross-graph constraint set of the derivation chain.

[0148] Furthermore, the edge-side monitoring device checks whether the second node of the starting point of the structural enhancement edge traversed during the propagation traversal belongs to the cross-graph constraint set of the derivation chain. The edge-side monitoring device traverses the structural enhancement edges recorded during the propagation traversal. For each structural enhancement edge, the edge-side monitoring device extracts the first constraint relation expression corresponding to the second node of the starting point of the structural enhancement edge and checks whether the first constraint relation expression belongs to the cross-graph constraint set of the candidate derivation chain.

[0149] S64: In response to the set of cross-graph constraints where the second node of the starting point belongs to the derivation chain, determine that a cross-graph interlocking loop is formed and delete the derivation chain.

[0150] In response to the first constraint relation expression corresponding to the second node at the starting point of the structural reinforcement edge, which belongs to the cross-graph constraint set of the candidate derivation chain, the edge-side monitoring device marks this propagation traversal as a propagation interlocking loop.

[0151] Furthermore, the edge monitoring device statistically analyzes the number of samples that form propagation interlocked loops and the number of samples that do not. If the number of samples forming propagation interlocked loops exceeds the number of samples that do not, the edge monitoring device determines that the candidate derivation chain forms a cross-graph interlocked loop and removes the candidate derivation chain from the candidate derivation chain set.

[0152] For example, the propagation traversal passes through the structural reinforcement edge originating from the second node corresponding to the inverter current constraint. The edge-side monitoring device checks whether the inverter current constraint belongs to the cross-graph constraint set of the candidate derivation chain. In response to the inverter current constraint belonging to the cross-graph constraint set of the candidate derivation chain, the edge-side monitoring device marks this propagation traversal as a propagation interlock loop.

[0153] S65: Identify multiple derivation chains with the same first node at the end point.

[0154] The edge monitoring device performs collaborative analysis on the candidate derivation chains that have not been deleted from the candidate derivation chain set. The edge monitoring device traverses the candidate derivation chain set that has not been deleted, groups the candidate derivation chains according to the first node of the endpoint of the candidate derivation chain, and identifies multiple derivation chains with the same first node of the endpoint as a subset of candidate derivation chains with the same endpoint.

[0155] S66: Calculate the intersection of the cross-graph constraint sets of multiple derivation chains.

[0156] Furthermore, the edge monitoring device extracts the cross-graph constraint set of each candidate derivation chain in the subset of candidate derivation chains with the same endpoint. The edge monitoring device then calculates the intersection of the cross-graph constraint sets of the candidate derivation chains in the subset of candidate derivation chains with the same endpoint.

[0157] Understandably, the intersection of the cross-graph constraint sets of multiple candidate derivation chains within the subset of candidate derivation chains with the same endpoint reflects the constraint range commonly covered by the multiple derivation chains. The fact that the intersection is non-empty indicates that the multiple derivation chains overlap in constraint coverage. Furthermore, by statistically analyzing the number of overlapping second nodes found along the structural enhancement edges from the constraints in the intersection of the multiple derivation chains, the synergy of the multiple derivation chains in constraint violation propagation paths can be evaluated.

[0158] S67: Starting from the second node represented by the constraint relation expression in the intersection set in the second graph structure, search for a reachable second node along the structure-enhancing edges.

[0159] Furthermore, in response to the non-empty intersection of the cross-graph constraint sets, the edge-side monitoring device searches for a reachable second node in the second graph structure by starting from the second node corresponding to the first constraint expression in the intersection and traversing the structural reinforcement edges. For each candidate derivation chain in the subset of candidate derivation chains with the same endpoint, the edge-side monitoring device searches for a reachable second node in the second graph structure by starting from the second node corresponding to the first constraint expression in the intersection and traversing the structural reinforcement edges.

[0160] S68: Count the number of overlapping second nodes found in multiple derivation chains.

[0161] Furthermore, the edge monitoring device counts the number of overlapping second nodes found by multiple candidate derivation chains in the subset of candidate derivation chains with the same endpoint. The edge monitoring device identifies second nodes simultaneously found by multiple candidate derivation chains and counts the number of these second nodes as the overlap count. The edge monitoring device also counts the total number of second nodes found by all candidate derivation chains in the subset of candidate derivation chains with the same endpoint and calculates the ratio of the overlap count to the total number of second nodes found.

[0162] It should be noted that the side-side monitoring equipment pre-sets a preset range for the ratio of overlapping numbers to the total number of second nodes found. This preset range includes a preset lower limit and a preset upper limit for interlocking coordination. The preset lower limit represents the minimum overlap ratio threshold for multiple derivation chains to exhibit coordination on the constraint violation propagation path, while the preset upper limit represents the maximum overlap ratio threshold for multiple derivation chains to exhibit coordination on the constraint violation propagation path.

[0163] For example, the preset interlocking coordination lower limit is set to 0.3, and the preset interlocking coordination upper limit is set to 0.7. If the ratio of the number of overlapping nodes to the total number of second nodes found is less than 0.3, it indicates that the overlap of multiple derivation chains on the constraint violation propagation path is too low, resulting in insufficient coordination, and they are not suitable for merging cross-graph constraint sets. If the ratio of the number of overlapping nodes to the total number of second nodes found is higher than 0.7, it indicates that the overlap of multiple derivation chains on the constraint violation propagation path is too high, potentially indicating redundancy, and they are not suitable for merging cross-graph constraint sets. If the ratio of the number of overlapping nodes to the total number of second nodes found is between 0.3 and 0.7, it indicates that multiple derivation chains have moderate coordination on the constraint violation propagation path, and they are suitable for merging cross-graph constraint sets.

[0164] S69: In response to the ratio of the number of overlapping nodes to the total number of second nodes found being within a preset range, the cross-graph constraint sets of multiple derivation chains are merged to construct a simulation constraint set.

[0165] Furthermore, in response to the ratio of overlapping nodes to the total number of second nodes found being within a preset range, the edge monitoring device merges the cross-graph constraint sets of each candidate derivation chain in the candidate derivation chain subset with the same endpoint. In response to the ratio of overlapping nodes to the total number of second nodes found being outside the preset range, the edge monitoring device selects a single candidate derivation chain from the candidate derivation chain subset with the same endpoint to retain.

[0166] The edge monitoring device constructs a final candidate derivation chain set from the retained candidate derivation chains. The edge monitoring device merges the first constraint relation expression corresponding to the first node of the endpoint of the candidate derivation chain in the initial available constraint set and the final candidate derivation chain set, as well as the first constraint relation expression in the cross-graph constraint set, to form the final constraint set.

[0167] Furthermore, the edge monitoring equipment performs constraint satisfaction solving based on the final constraint set to obtain the final feasible value space. The edge monitoring equipment performs risk assessment on the final feasible value space to obtain the risk level distribution, and uses the ordinal difference between the highest and lowest levels in the four-level sequence as the final level span parameter.

[0168] It should be noted that the edge monitoring equipment is pre-set with a preset risk threshold to determine whether the simulation prediction results require adjustment of the energy storage system's protection strategy. The preset risk threshold is a risk level within a four-level sequence. If the highest risk level in the simulation prediction results exceeds the preset risk threshold, it indicates that the energy storage system has a high operational risk, requiring adjustment of the protection strategy to mitigate that risk.

[0169] For example, the preset risk threshold is set to level two. If the highest level in the simulation prediction result is level one or two, the edge monitoring equipment determines that the operational risk of the energy storage system is within an acceptable range and does not trigger protection strategy adjustments. If the highest level in the simulation prediction result is level three or four, the edge monitoring equipment determines that the operational risk of the energy storage system exceeds the preset risk threshold and triggers protection strategy adjustments.

[0170] Furthermore, the side-side monitoring device compares the final level span parameter with the initial level span parameter obtained in step S5. In response to the final level span parameter being lower than the initial level span parameter, the side-side monitoring device outputs the highest and lowest levels in the final feasible value space as simulation prediction results.

[0171] Furthermore, the edge-side monitoring equipment adjusts the operation strategy of the wind farm's edge-side energy storage system based on simulation prediction results. In response to the highest risk level in the simulation prediction results exceeding a preset risk threshold, the edge-side monitoring equipment generates a risk warning signal and triggers adjustments to the energy storage system's protection strategy. The edge-side monitoring equipment determines the power regulation margin of the energy storage system based on the difference between the highest and lowest risk levels, adjusting the upper limit of the energy storage charging and discharging power to avoid triggering violations of critical obstruction constraints.

[0172] For example, the preset risk threshold is set to level two, and the simulation prediction results show that the highest risk level is level three, exceeding the preset risk threshold. The edge monitoring device generates a risk warning signal and triggers the energy storage system's protection strategy adjustment. The edge monitoring device reduces the upper limit of the energy storage charging power control from the rated power to 80% of the rated power, and narrows the inverter output voltage regulation range from 90% to 110% of the rated voltage to 95% to 105% of the rated voltage, in order to reduce the risk of constraint violation propagation causing level jumps.

[0173] S610: In response to the simulation prediction result obtained by performing constraint solving based on the simulation constraint set constructed based on the cross-graph constraint set not meeting the preset conditions, restore the deleted first directed edge in the first graph structure and delete the structure association pointer.

[0174] If the side-side monitoring equipment responds to the condition that the final level span parameter is not lower than the initial level span parameter, it determines that the simulation prediction result does not meet the preset conditions. The side-side monitoring equipment performs an interlock release operation on the first and second diagram structures.

[0175] The edge monitoring device identifies the structural associations that candidate derivation chains in the final candidate derivation chain set traverse in the first graph structure. The edge monitoring device restores the first directed edge between the starting and ending first nodes of the structural associations in the first graph structure, and adds the first directed edge deleted in step S32 back to the first graph structure. The edge monitoring device then deletes the structural associations.

[0176] S611: Delete the structural enhancement edge in the second graph structure.

[0177] Furthermore, the edge monitoring device identifies the structural reinforcement edges established in step S4 within the second diagram structure. The edge monitoring device then deletes all structural reinforcement edges within the second diagram structure.

[0178] S612: Based on the first graph structure after the interlock is released, the historical data is re-filtered, and the structure change operation, second graph structure construction and derivation chain search are performed to obtain the cross-graph constraint set.

[0179] Furthermore, the edge monitoring device re-executes steps S2, S3, S4, and S5 based on the first graph structure after the interlock is released. The edge monitoring device re-filters the constraint violation sequence extracted in step S2 based on the first graph structure after the interlock is released, extracting historical violation propagation relationships. The edge monitoring device performs a structural change operation on the first graph structure after the interlock is released, identifies the first directed edge inconsistent with the historical violation propagation relationship, and performs path decomposition. The edge monitoring device constructs a second graph structure based on the first graph structure after the structural change operation is re-executed and establishes structurally enhanced edges. The edge monitoring device performs a derivation chain search in the reconstructed first graph structure to obtain a new set of cross-graph constraints.

[0180] Furthermore, the edge monitoring device re-executes step S6 based on the new cross-graph constraint set. In response that the final level span parameter after re-execution is still not lower than the initial level span parameter, the edge monitoring device outputs the highest and lowest levels in the initial feasible value space in step S5 as simulation prediction results.

[0181] Understandably, the specific process by which the edge monitoring equipment performs risk assessment on the final feasible value space is as follows: The edge monitoring equipment samples parameter combinations from the final feasible value space, substitutes these sampled parameter combinations into the first constraint relationship expression in the final constraint set for satisfaction checks, and identifies the violated first constraint relationship expressions. The edge monitoring equipment determines the risk level corresponding to the sampled parameter combinations based on the number and severity of the violated first constraint relationship expressions. The edge monitoring equipment determines the risk level for each sampled parameter combination in the final feasible value space, counts the frequency of each risk level in the four-level sequence, and obtains the risk level distribution.

[0182] Understandably, the specific process by which the edge monitoring device performs risk assessment on the initial feasible value space is as follows: The edge monitoring device samples parameter combinations from the initial feasible value space, substitutes the sampled parameter combinations into the first constraint relationship expression in the initial available constraint set for satisfaction checks, and identifies the violated first constraint relationship expressions. The edge monitoring device determines the risk level corresponding to the sampled parameter combinations based on the number and severity of the violated first constraint relationship expressions. The edge monitoring device determines the risk level for each sampled parameter combination in the initial feasible value space, counts the number of times each risk level appears in the four-level sequence, and obtains the risk level distribution.

[0183] In step S6, this invention verifies whether the constraint violation propagation caused by the derivation result forms a loop that returns to the constraint range covered by the derivation chain through cross-graph interlocking verification, thus avoiding the derivation chain triggering a self-feedback propagation loop that could cause the risk assessment to deviate from the actual evolution path of the wind farm side-side energy storage system. Through derivation chain collaborative analysis, the synergy of multiple derivation chains with the same endpoint is identified and the cross-graph constraint set is merged, improving the coverage of the simulation constraint set in assessing the operating state of the energy storage system. Based on the simulation prediction results, the power control strategy of the energy storage system is adjusted to maintain the safe operation of the energy storage system even in the event of abnormal data reception.

[0184] Example 2 is an embodiment of the present invention, which provides a simulation system for side-side equipment of a production operation support system, including:

[0185] The graph structure construction module is used to construct the first graph structure and the second graph structure. The second graph structure is constructed based on the first graph structure by filtering historical data.

[0186] The structural interlock module is used to perform structural interlock operations on the first graph structure and the second graph structure. The structural interlock operation causes structural changes in the first graph structure to trigger the establishment of structural reinforcement edges in the second graph structure.

[0187] The derivation chain search module is used to perform derivation chain search based on the first graph structure and the second graph structure after structural interlocking. When searching for derivation paths in the first graph structure, the derivation chain search extends along the structural enhancement edges in the second graph structure to obtain the cross-graph constraint set.

[0188] The simulation solution module is used to construct a set of simulation constraints based on the cross-graph constraint set, and to perform constraint solving to obtain simulation prediction results.

[0189] This embodiment also provides an electronic device applicable to a simulation method for a side-side device of a production operation support system, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the simulation method for a side-side device of a production operation support system as proposed in the above embodiment.

[0190] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, it implements a simulation method for a side device of a production operation support system as proposed in the above embodiments.

[0191] The storage medium proposed in this embodiment and the simulation method for implementing a side device of a production operation support system proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0192] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0193] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A simulation method for side-side equipment in a production operation support system, characterized in that, include: Construct a first graph structure, with the first constraint relation expression as the first node and the directed edge representing the derivation relationship between the first nodes as the first directed edge; Based on the structure of the first graph, historical data is filtered to extract the historical violation propagation relationships between the second constraint relationship expressions; Based on the historical violation of the propagation relationship, a structural change operation is performed on the first graph structure, which alters the first directed edge in the first graph structure. The second graph structure is constructed based on the first graph structure after structural changes. The second graph structure uses the second constraint relationship expression as the second node and the directed edge representing the historical violation propagation relationship between the second nodes as the second directed edge. During the construction process, the second graph structure establishes structural enhancement edges based on the structural changes of the first graph structure. The derivation chain search is performed in the first graph structure. When searching for derivation paths, the derivation chain search is extended along the structural enhancement edges in the second graph structure to obtain the cross-graph constraint set. A simulation constraint set is constructed based on a cross-graph constraint set, and the simulation prediction results are obtained by performing constraint solving. The structural change operations include: Identify the first directed edge in the first graph structure that is inconsistent with the historical violation propagation relationship; In response to the identification of a first directed edge that is inconsistent with the historical violation propagation relationship, the identified first directed edge is deleted. A structural association pointer is established between the start node and the end node of the deleted first directed edge. The structural association pointer records the path formed by other first directed edges in the first graph structure from the start node to the end node. The second graph structure establishes structural enhancement edges based on structural changes in the first graph structure during its construction process, including: In the second graph structure, identify the start node and end node; in the second graph structure, represent the second directed edge between second nodes with the same constraint relationship expression. Identify other second directed edges starting from the second node, the endpoint of the second directed edge; In response to the structural association pointing between the starting second node and the ending second node of other second directed edges, which represent the same constraint relationship expression in the first graph structure, a structural reinforcement edge is established from the second directed edge to other second directed edges.

2. The simulation method for side-side equipment of a production operation support system as described in claim 1, characterized in that, The derivation chain search expands along the structural enhancement edges in the second graph structure when searching for the derivation path to obtain a cross-graph constraint set, including: Search for the derivation path within the structure of the first graph; In response to the existence of a structural association pointing to the current first node and the predecessor first node in the derivation path, check whether there is a structural reinforcement edge between the second directed edge between the starting node and the ending node of the structural association pointing to the second node in the second graph structure that represents the same constraint relationship expression. In response to the existence of a structurally reinforcing edge, a second reachable node is searched along the structurally reinforcing edge, and the constraint relation expression represented by the searched second node is added to the cross-graph constraint set of the derivation chain.

3. The simulation method for side-side equipment of a production operation support system as described in claim 2, characterized in that, Before constructing the simulation constraint set based on the cross-graph constraint set, the following is also included: Perform cross-graph interlock verification on the found derivation chain. Cross-graph interlock verification includes: Substitute the derivation result of the derivation chain into the constraint relationship expression represented by the second node of the second graph structure to identify the violated second node; Propagate traversal from the violated second node along the second directed edge and the structural reinforcement edge; Check whether the second node of the starting point of the structural enhancement edge traversed by the propagation traversal belongs to the cross-graph constraint set of the derivation chain; In response to the set of cross-graph constraints where the second node of the starting point belongs to the derivation chain, it is determined that a cross-graph interlocking loop has been formed and the derivation chain is deleted.

4. The simulation method for side-side equipment of a production operation support system as described in claim 3, characterized in that, Before constructing the simulation constraint set based on the cross-graph constraint set, the following is also included: Identify multiple derivation chains with the same first node at the endpoint; Calculate the intersection of the cross-graph constraint sets of multiple derivation chains; Starting from the second node represented by the constraint relation expression in the intersection in the second graph structure, search for the reachable second node along the structure-enhancing edges; Count the number of overlapping second nodes found in multiple derivation chains; In response to the fact that the ratio of the number of overlapping nodes to the total number of second nodes found is within a preset range, the cross-graph constraint sets of multiple derivation chains are merged to construct a simulation constraint set.

5. The simulation method for side-side equipment of a production operation support system as described in claim 4, characterized in that, In response to the fact that the simulation prediction results obtained by solving the simulation constraint set based on the cross-graph constraint set do not meet the preset conditions, an interlock release operation is performed on the first graph structure and the second graph structure. The interlock release operation includes: Restore the first directed edge that was deleted in the first graph structure, and delete the associated pointing edge of the structure; Remove the structural reinforcement edges from the structure in the second diagram. Based on the first graph structure after the interlock is released, the historical data is re-filtered, and structural change operations, second graph structure construction, and derivation chain search are performed to obtain the cross-graph constraint set.

6. A simulation system for side-side equipment of a production operation support system, employing the simulation method for side-side equipment of a production operation support system as described in any one of claims 1 to 5, characterized in that, include: The graph structure construction module is used to construct the first graph structure and the second graph structure. The second graph structure is constructed based on the first graph structure by filtering historical data. The structural interlock module is used to perform structural interlock operations on the first graph structure and the second graph structure. The structural interlock operation causes structural changes in the first graph structure to trigger the establishment of structural reinforcement edges in the second graph structure. The derivation chain search module is used to perform derivation chain search based on the first graph structure and the second graph structure after structural interlocking. When searching for derivation paths in the first graph structure, the derivation chain search extends along the structural enhancement edges in the second graph structure to obtain the cross-graph constraint set. The simulation solution module is used to construct a set of simulation constraints based on the cross-graph constraint set, and to perform constraint solving to obtain simulation prediction results.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the simulation method for side-side equipment of a production operation support system according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the simulation method for the side-side equipment of the production operation support system according to any one of claims 1 to 5.

Images