Multi-scene linkage power grid simulation dispatching system and method

Abstract

The application discloses a multi-scene linkage power grid simulation scheduling system and method, relates to the technical field of power system simulation, and comprises the following units: a data unit: used for acquiring power grid data of different regions, analyzing the power grid data to obtain heterogeneous data, wherein the heterogeneous data comprises a differentiated topological structure, standard equipment parameters and rule constraints; a processing unit: used for obtaining a regional parameter set based on the heterogeneous data; a modeling unit: used for constructing a cross-region interconnection topological model based on the regional parameter set; an event unit: used for obtaining a monitoring event based on the power grid data; and a scheduling unit: used for constructing a scene structure body, obtaining a simulation result based on the cross-region interconnection topological model, the monitoring event and the scene structure body, and solving the problems that a traditional power system simulation system cannot dynamically adapt to multi-region heterogeneous structures by using a unified standardized template or a fixed topological template, and that multi-scene independent analysis leads to an increased scheduling misjudgment rate.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system simulation, in particular to a multi-scenario linkage power grid simulation scheduling system and method. BACKGROUND

[0002] Traditional power system simulation systems adopt unified standardized templates or fixed topological templates, but with the expansion of regional power grid interconnection scale, the problem of power grid topology heterogeneity is increasingly prominent: due to differences in historical development, resource distribution and load characteristics, different regions form mixed topological structures such as ring networks and mesh networks (such as the 500kV double ring network in East China and the radial collection network of Northwest wind farms), and there are significant non-standard differences in device parameters (transformer impedance or line capacity) and operating rules (N-1 criterion / voltage deviation limit). The traditional power system simulation system cannot be compatible with such heterogeneity and cannot dynamically adapt to multi-regional heterogeneous structures.

[0003] Moreover, traditional power system simulation systems are usually single simulation models, which run independently for multiple scenarios such as voltage overrun, new energy fluctuation and short circuit fault, lack dynamic linkage mechanisms triggered by events, and are prone to fall into local optimal traps. For example, optimizing a single scenario independently (such as solving voltage overrun) may worsen other scenarios (such as increasing line loss), and the lack of event linkage mechanism may lead to time sequence misalignment. For example, when event A (short circuit fault) and event B (voltage drop) occur successively, independent simulation cannot capture the chain reaction of protection misoperation caused by voltage drop due to fault, thus increasing the dispatching misjudgment rate. SUMMARY

[0004] In order to solve the problem that the traditional power system simulation system adopts unified standardized templates or fixed topological templates and cannot dynamically adapt to multi-regional heterogeneous structures, and the problem of increasing dispatching misjudgment rate caused by multi-scenario independent analysis, the present application provides a multi-scenario linkage power grid simulation scheduling system, which comprises:

[0005] A data unit is configured to obtain power grid data of different regions, analyze the power grid data to obtain heterogeneous data, and the heterogeneous data comprises differential topological structure, standard device parameters and rule constraints.

[0006] A processing unit is configured to obtain a regional parameter set based on the heterogeneous data.

[0007] A modeling unit is configured to construct a cross-regional interconnected topological model based on the regional parameter set.

[0008] An event unit is configured to obtain a monitoring event based on the power grid data.

[0009] The scheduling unit is configured to construct a scene structure, and obtain a simulation result based on the cross-region interconnection topology model, the monitoring event, and the scene structure.

[0010] The system automatically identifies topological differences through topological adaptive analysis, uniformly adapts different regional power grid structures, solves the problem of different topological structures of regional power grids, and realizes seamless integration of heterogeneous topologies. The system converts device parameters of each region into a regional parameter set with a regional label, solves the problem of different device parameters of different regions, and realizes dynamic adaptation of simulation models to multi-region parameters. The system improves the compatibility and fault tolerance of the model. The system distinguishes different event types and their corresponding trigger conditions, realizes standardized docking of events and modeling based on a standardized scene structure, realizes an event-driven multi-scene linkage mechanism, and realizes automatic linkage simulation of complex scenes such as multi-scene, solves the problem of scene isolation and weak fault tolerance, reduces the error rate of dispatching, and improves the accuracy of simulation analysis results.

[0011] Further, the data unit specifically includes:

[0012] The topology unit is configured to obtain a regional ring network topology, a regional radial topology, and a regional mesh network topology based on the power grid data.

[0013] The regional node-branch structure is obtained based on the regional ring network topology, and the regional directed graph is obtained based on the regional node-branch structure. The regional tree structure is obtained based on the regional radial topology, and the regional mesh structure is obtained based on the regional mesh network topology.

[0014] The difference topology structure is obtained based on the regional directed graph, the regional tree structure, and the regional mesh structure.

[0015] The parameter unit is configured to obtain device parameters based on the power grid data, convert the device parameters into a standard unit, and obtain the standard device parameters.

[0016] The rule unit is configured to construct a rule standard library, and the rule standard library includes a plurality of standard constraint rules.

[0017] The rule unit is configured to obtain regional operation rules based on the power grid data, match the regional operation rules with the rule standard library, obtain a standard constraint rule based on the rule standard library if the matching is successful, obtain a rule threshold and a rule variable based on the regional operation rules if the matching fails, and generate a constraint rule expression based on the rule threshold and the rule variable.

[0018] and for matching the region operation rules and the keywords, obtaining special variables and special keywords, generating special rules based on the special variables and the special keywords, and generating an event trigger function based on the special rules;

[0019] and for obtaining the rule constraint based on the standard constraint rule, the constraint rule expression, and the event trigger function.

[0020] Different processing is performed for different topologies, a structure difference deconstruction algorithm is used, non-standard topology types are compatible, the structure is more accurate and clear, and the applicability is stronger; device parameters are normalized to achieve non-standard data consistency; region self-defined text rules are converted into mathematical constraint conditions, natural language actions are converted into mathematical conditional statements, a self-defined constraint function is generated, non-standard operation rules are compatible, a simulation model dynamically adapts to multi-region parameters, and the compatibility and fault tolerance of the model are improved.

[0021] Further, the processing unit is specifically configured to:

[0022] a mapping relationship table of the difference topology structure and the standard device parameters is constructed, and a region identification code of the standard device parameters is injected; based on the difference topology structure, the standard device parameters, the rule constraint, the mapping relationship table, and the region identification code, a labeled region parameter set is obtained.

[0023] Region identification code injection prevents cross-region parameter confusion and reduces model mismatching rate.

[0024] Further, the modeling unit is specifically configured to:

[0025] a parameter compatibility layer is constructed, and the region parameter set is converted based on the parameter compatibility layer to obtain a cross-region parameter set;

[0026] a region boundary node is obtained based on the difference topology structure, a region tie line is obtained based on the region boundary node, and the cross-region interconnection topology model is constructed based on the region tie line and the cross-region parameter set.

[0027] The parameter compatibility layer converts the region parameter set to solve unit system / reference value difference; the boundary node is automatically identified, the concept of virtual tie line is proposed to simulate cross-region interaction, cross-region topology is seamlessly spliced and multi-label region connection detection is performed, cross-region interaction points are accurately positioned, boundary positioning error is reduced, and splicing time consumption is reduced.

[0028] Further, the event unit is specifically configured to:

[0029] a plurality of classification events, first event attributes of the classification events, first trigger conditions of the classification events, and first priorities of the classification events are obtained based on the power grid data;

[0030] setting a custom event, a second event attribute of the custom event, a second trigger condition of the custom event, a trigger action of the custom event and a second priority of the custom event;

[0031] the first event attribute comprises an event position, an event time and a severity, and the second event attribute comprises an added load capacity, an access point position and a time plan;

[0032] obtaining the monitoring event based on the classified event and the custom event.

[0033] generating a modeling requirement according to a real-time monitoring event or a user-defined event.

[0034] Further, the scheduling unit is specifically configured to:

[0035] constructing a model template library and an algorithm library, setting a mapping rule of the classified event and the custom event and the model template library, the model template library containing modeling templates in different scenarios, and the algorithm library containing several power flow calculation algorithms, several reactive power optimization algorithms and several source-grid-load-storage coordination algorithms;

[0036] constructing the scenario structure, the scenario structure including an event type, a simulation mode, a time span, a space range, a data dependency and an output requirement field;

[0037] obtaining several scenario models based on the first trigger condition, the second trigger condition, the mapping rule and the model template library;

[0038] obtaining a simulation result based on the cross-region interconnection topology model, the scenario model, the monitoring event, the scenario structure and the algorithm library.

[0039] constructing a scenario structure, reducing modeling obstacles caused by regional topological / parameter differences, breaking data silos, realizing cross-scene data penetration, supporting automatic association of multi-scene data, generating a composite event response chain, and accurately quantifying calculation requirements through the fields of the structure body, and optimizing resource scheduling. Modularly package power flow calculation, reactive power optimization and other algorithms to support dynamic loading / replacement.

[0040] Further, the scheduling unit is specifically configured to:

[0041] constructing a double-layer optimization model, the double-layer optimization model including an upper-layer model and a lower-layer model;

[0042] constructing an instruction mapping rule library, the instruction mapping rule library containing several parameter variables, regional types corresponding to the parameter variables and instruction templates;

[0043] Obtaining optimization variables based on the simulation results, obtaining a change in the optimization variables, and obtaining a plurality of differentiated instructions based on the change and the instruction mapping rule base; and obtaining a scheduling strategy based on the differentiated instructions;

[0044] The first calculation formula of the upper model is:

[0045] ;

[0046] ;

[0047] in, represents the decision variable vector, Indicates the Periods, Indicates the time period, represents the power generation cost function, Indicates time period The active power output of the unit is represents the spare capacity cost, Indicates time period The spare capacity, represents the cascading risk penalty coefficient, represents the cascading failure risk indicator, Indicates the devices, Indicates the total number of devices. Indicates the degree of failure, Indicates the The interconnection line power of each device, Indicates the Safety margin of each device;

[0048] The second calculation formula of the lower model is:

[0049] ;

[0050] in, represents the constraint matrix of region A, represents the state variable, Represents a preset matrix.

[0051] A two-layer optimization model for adaptive scheduling decisions is constructed. The upper-layer model uses mixed integer programming to process discrete decisions with the goal of minimizing the loss of the entire network. The lower-layer model uses the Lagrange multiplier method to process constraint coupling to meet regional differentiation constraints. The two-layer optimization model of economic upper layer + security lower layer breaks through the limitations of traditional single-layer optimization and realizes the coordinated optimization of economy and safety. The two-layer optimization outputs differentiated instructions according to the regional characteristic mapping rule library based on regional characteristics, realizes the precise adaptation of scheduling instructions, improves the accuracy of instructions, and makes cross-regional scheduling strategies more adaptable to each region in linkage scenarios, thereby improving decision-making accuracy.

[0052] Furthermore, the scheduling unit is further configured to:

[0053] Obtaining a real-time risk assessment value based on the scheduling strategy, adjusting the constraint boundary of the two-level optimization model based on the real-time risk assessment value, and obtaining an optimization decision based on the adjusted two-level optimization model;

[0054] The third calculation formula of the real-time risk assessment value is:

[0055] ;

[0056] in, Represents the real-time risk assessment value, represents the inter-regional tie line power, Indicates the degree of failure, Indicates the The safety margin of each device, Indicates the Safety limit values ​​for each device, Indicates the device number.

[0057] Through sensitivity analysis, risk warnings are converted into constraint boundary adjustments and then into specific scheduling instructions. A risk-driven real-time constraint boundary adjustment technology is proposed to solve the unsolvable optimization problem in high-risk scenarios and improve feasibility.

[0058] Furthermore, the system further includes a reconstruction unit, which is specifically configured to:

[0059] Obtaining a regional root node of the regional radial topology, and obtaining a regional mesh number based on the regional mesh structure;

[0060] Obtaining reconstruction data, and obtaining a reconstructed ring network topology, a reconstructed radial topology, and a reconstructed mesh network topology based on the reconstruction data; obtaining a reconstructed node-branch structure based on the reconstructed ring network topology, obtaining a reconstructed tree structure based on the reconstructed radial topology, marking a reconstructed root node of the reconstructed radial topology, and obtaining a reconstructed mesh structure and a reconstructed number of meshes based on the reconstructed mesh network topology;

[0061] obtaining a first similarity degree based on the reconstructed ring network topology and the regional ring network topology, obtaining a second similarity degree based on the reconstructed tree structure and the regional tree structure, and obtaining a third similarity degree based on the reconstructed mesh number and the regional mesh number;

[0062] determining whether the first similarity degree exceeds a first threshold value, and if so, obtaining a first similar region based on the first similarity degree, and obtaining a first adjustment structure based on the reconstructed node-branch structure, the regional node-branch structure and the first similar region;

[0063] determining whether the second similarity degree exceeds a second threshold value, and if so, obtaining a second similar region based on the second similarity degree, and obtaining a second adjustment structure based on the reconstructed root node, the regional root node and the second similar region;

[0064] determining whether the third similarity degree exceeds a third threshold value, and if so, obtaining a third similar region based on the third similarity degree, obtaining a first node of the reconstructed mesh structure and a first connection edge of the first node, and obtaining a second node of the regional mesh structure and a second connection edge of the second node;

[0065] obtaining a third adjustment structure based on the first node, the first connection edge, the second node and the second connection edge;

[0066] obtaining a reconstructed topology model based on the first adjustment structure, the second adjustment structure, the third adjustment structure and the reconstructed data, and updating the cross-region interconnection topology model to the reconstructed topology model.

[0067] Based on the existing simulation model, the similarity of different regional topological structures is determined, and reconstruction is performed on the basis of the similarity, thereby reducing the reconstruction time and improving the efficiency.

[0068] The application also provides a multi-scene linkage power grid simulation scheduling method, which comprises the following steps:

[0069] obtaining power grid data of different regions, and analyzing the power grid data to obtain heterogeneous data, wherein the heterogeneous data comprises a differential topological structure, standard equipment parameters and rule constraints;

[0070] obtaining a regional parameter set based on the heterogeneous data;

[0071] constructing a cross-region interconnection topology model based on the regional parameter set;

[0072] obtaining a monitoring event based on the power grid data;

[0073] constructing a scene structure, and obtaining a simulation result based on the cross-region interconnection topology model, the monitoring event and the scene structure.

[0074] The principle and effect of the method are similar to the system, and the method will not be described in detail.

[0075] The one or more technical solutions provided by the application have at least the following technical effects or advantages:

[0076] 1. The system automatically identifies topological differences through topological adaptive analysis, uniformly adapts different regional power grid structures, solves the problem of different topological structures of different regional power grids, realizes seamless integration of heterogeneous topologies, converts device parameters of each region into a regional parameter set with a regional label, solves the problem of different device parameters of different regions, internally stores a rule standard library, uniformly converts regional custom rules into calculable constraints, solves the problem of diversified operation rules of different regional power grids, integrates heterogeneous regional data, thereby realizing dynamic adaptation of simulation models to multi-regional parameters, improving the compatibility and fault tolerance of the model, distinguishing different event types and their corresponding trigger conditions, realizing standardized docking of events and modeling based on a standardized scene structure, realizing an event-driven multi-scene linkage mechanism, thereby realizing automatic linkage simulation of complex scenes such as multi-scene, solving the problem of isolated scenes and weak fault tolerance, reducing the misjudgment rate of dispatching, and improving the accuracy of simulation analysis results.

[0077] 2. The scene structure is constructed, the modeling obstacles caused by the topological / parameter differences between regions are reduced, the data islands are broken, the cross-scene data is connected, the multi-scene data is automatically associated, the composite event response chain is generated, and the demand is accurately quantified through the structure field, and the resource scheduling is optimized.

[0078] 3. A double-layer optimization model of adaptive dispatching decision is constructed, the upper model adopts mixed integer programming to process discrete decisions to minimize the total network loss, the lower model adopts the Lagrange multiplier method to process constraint coupling, meets the differentiated constraints of the regions, and the double-layer optimization model of economic upper layer + safety lower layer breaks through the limitations of traditional single-layer optimization, realizes the collaborative optimization of economy and safety, the double-layer optimization outputs differentiated instructions according to the region characteristic mapping rule library based on the region characteristics, realizes accurate adaptation of dispatching instructions, improves the accuracy of instructions, and the cross-regional dispatching strategy is more suitable for each region in the linkage scene, and improves the decision accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0079] The accompanying drawings described herein are used to provide further understanding of the embodiments of the application, form a part of the application, and do not constitute a limitation on the embodiments of the application;

[0080] Figure 1 is a running schematic diagram of the multi-scene linkage power grid simulation dispatching system in the application. DETAILED DESCRIPTION

[0081] In order to enable the above-mentioned objects, features and advantages of the present application to be more clearly understood, the present application will be described in further detail below with reference to the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other in the case of not conflicting with each other.

[0082] In the following description, a large number of specific details are set forth in order to facilitate a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein within the scope of the present application, and therefore the scope of protection of the present application is not limited by the specific embodiments disclosed below.

[0083] Embodiment 1

[0084] Reference Figure 1 The embodiment provides a multi-scenario linkage power grid simulation scheduling system, and the system comprises:

[0085] A data unit is configured to acquire power grid data of different regions, analyze the power grid data to obtain heterogeneous data, and the heterogeneous data comprises differentiated topological structures, standard equipment parameters and rule constraints; for example, a topological type is determined by using methods such as node degree analysis, loop detection and connected component analysis, or the topological type is classified by using topological feature extraction.

[0086] The data unit specifically comprises:

[0087] A topological unit is configured to obtain regional ring network topologies, regional radial topologies and regional mesh network topologies based on the power grid data;

[0088] A regional node-branch structure is obtained based on the regional ring network topologies, a regional directed graph is obtained based on the regional node-branch structure, a regional tree structure is obtained based on the regional radial topologies, and a regional mesh structure is obtained based on the regional mesh network topologies;

[0089] The differentiated topological structures are obtained based on the regional directed graph, the regional tree structure and the regional mesh structure;

[0090] A parameter unit is configured to obtain equipment parameters based on the power grid data, convert the equipment parameters into standard values to obtain the standard equipment parameters; for example, a transformer short-circuit impedance is converted into a standard value, and a calculation formula of the transformer short-circuit impedance can be as follows:

[0091] ;

[0092] wherein, Znorm represents a normalized standard value impedance, which is dimensionless, and realizes parameter comparability of equipment of different capacities, Zpercent represents a transformer short-circuit impedance percentage, which is dimensionless and essentially represents an impedance pressure drop ratio. The division by 100 is to convert the percentage into a standard base, represents the system reference capacity, represents the reference voltage, usually taking the rated value of the voltage level where the device is located.

[0093] Rule unit: for building a rule standard library, which contains several standard constraint rules;

[0094] Based on the power grid data, obtain the regional operation rules, match the regional operation rules with the rule standard library, if the matching is successful, obtain the standard constraint rules based on the rule standard library; if the matching fails, obtain the rule threshold and the rule variable based on the regional operation rules, and generate a constraint rule expression based on the rule threshold and the rule variable; such as using NLP (Natural Language Processing, a professional artificial intelligence for analyzing human language) to extract key thresholds and variables, and constructing inequality constraints, for example, the procedure text: wind power curtailment rate ≤ 15% → convert constraint wind ≤ 0.15.

[0095] Match the regional operation rules with the keywords to obtain special variables and special keywords, generate special rules based on the special variables and the special keywords, and generate an event trigger function based on the special rules; such as a special rule: prefer to cut photovoltaic when new energy fluctuation is out of limit, use NLP to obtain the trigger condition through new energy, fluctuation and out of limit, obtain the trigger action by prefer, cut and photovoltaic, construct the IF-THEN structure (logic control structure), and thus generate the event trigger function.

[0096] Based on the standard constraint rules, the constraint rule expression and the event trigger function, obtain the rule constraint.

[0097] Processing unit: for obtaining a set of regional parameters based on the heterogeneous data;

[0098] The processing unit is specifically configured to:

[0099] Construct a mapping relationship table of the differentiated topology structure and the standard device parameters, as shown in Table 1:

[0100] Table 1 Mapping relationship table of device parameters and topology nodes

[0101] Device ID Belonging area Connection node Parameter type Parameter value T_EC_501 East China Bus501 K_T(transformer ratio) 525 / 230 L_NW_332 Northwest China Bus332 R+jX(line impedance) 0.12+j0.25

[0102] Wherein, R represents resistance, X represents reactance, and j represents imaginary unit.

[0103] Inject the regional identification code of the standard device parameters, such as East China: EC, Northwest: NW.

[0104] Based on the differentiated topology structure, the standard equipment parameters, the rule constraints, the mapping relationship table and the area identification code, a labeled area parameter set is obtained.

[0105] A modeling unit is configured to construct a cross-area interconnection topology model based on the area parameter set.

[0106] The modeling unit is specifically configured to:

[0107] A parameter compatibility layer is constructed, and the area parameter set is converted based on the parameter compatibility layer to obtain a cross-area parameter set. The parameter compatibility layer solves the unit system / reference value difference. For example, the normalized per-unit impedance of a transformer short-circuit impedance is converted. The calculation formula can be:

[0108]

[0109] Among them, represents the dimensionless impedance compatible with the new reference system after conversion, represents the normalized per-unit impedance, represents the reference capacity of the new system, represents the reference capacity of the original system.

[0110] Based on the differentiated topology structure, an area boundary node is obtained, such as automatically identifying the area boundary node using a graph theory algorithm or a breadth-first search (BFS) algorithm. The area identification code of the branch connected by the node is obtained. If the area identification codes of the nodes connected by the branch are two or more, it is determined as a boundary node. Based on the area boundary node, an area tie line is obtained, such as adding a virtual tie line for the boundary nodes of any two adjacent areas. The impedance of the virtual tie line can be the average of the boundary impedances of the two areas. Based on the area tie line and the cross-area parameter set, the cross-area interconnection topology model is constructed. The cross-area interconnection topology model can adopt an existing dynamic equivalence model.

[0111] An event unit is configured to obtain a monitoring event based on the power grid data.

[0112] The event unit is specifically configured to:

[0113] Based on the power grid data, a plurality of classification events, first event attributes of the classification events, first trigger conditions of the classification events and first priorities of the classification events are obtained.

[0114] For example, the classification events include voltage events (detecting voltage out-of-limit), equipment events (identifying load rate) and short-circuit events (capturing circuit breaker tripping or abnormal fault current).

[0115] ​setting a custom event, a second event attribute of the custom event, a second triggering condition of the custom event, a triggering action of the custom event, and a second priority of the custom event;

[0116] The custom event can be: industry expansion (new load / power supply access, such as industrial park expansion), new energy event (photovoltaic fluctuation> threshold, abnormal charging and discharging of energy storage), EV charging event (local load surge> threshold caused by concentrated operation of charging piles), three-phase imbalance event (uneven distribution of three-phase load, uneven distribution of single-phase load, and total load balance coefficient less than threshold), and power factor out-of-limit event (power factor deviates from the design range).

[0117] According to the emergency degree and the influence range of the event, the event is given a priority.

[0118] The first event attribute includes event location, event time, and severity, and the second event attribute includes new load capacity, access point location, and time plan.

[0119] Based on the classified event and the custom event, the monitoring event is obtained.

[0120] The scheduling unit is configured to construct a scenario structure, and obtain a simulation result based on the cross-region interconnection topology model, the monitoring event, and the scenario structure.

[0121] The scheduling unit is specifically configured to:

[0122] A model template library and an algorithm library are constructed, and a mapping rule of the classified event and the custom event and the model template library is set, such as:

[0123] The voltage event corresponds to calling a reactive power compensation optimization model to perform voltage adjustment simulation;

[0124] The device overload event corresponds to calling a load transfer model to analyze a load redistribution scheme;

[0125] The industry expansion event corresponds to calling a power grid expansion model to analyze the impact of the new load on the power grid;

[0126] The new energy access event corresponds to calling a new energy fluctuation analysis model to evaluate the impact on voltage stability;

[0127] The electric vehicle charging event corresponds to calling a load response optimization model to optimize the charging strategy.

[0128] The model template library contains modeling templates under different scenarios, and the algorithm library contains a plurality of power flow calculations, a plurality of reactive power optimization algorithms, and a plurality of source-grid-load-storage coordination algorithms.

[0129] The scene structure is constructed, and the scene structure includes an event type, a simulation mode, a time span, a space range, a data dependency and an output requirement field; for example, the event type is, for example, low voltage, industry access, etc.; the simulation mode is, for example, power flow calculation, time sequence power flow calculation, reactive power optimization, etc.; the time span is determined according to the event nature to determine the time range of the simulation, for example, real-time events are real-time, and predictive events are 24 hours in the future; the space range is the area affected by the event, for example, the area where the 10kV line L12 and the transformer T5 are located; the data dependency specifies the data source required for modeling, such as real-time measurement data, historical load curve, device parameter library, etc.; and the output requirement field specifies the key indicators required to output the simulation results, such as voltage distribution, load rate and optimization scheme, etc.

[0130] Based on the first trigger condition, the second trigger condition, the mapping rule and the model template library, a plurality of scene models are obtained;

[0131] Based on the cross-region interconnection topology model, the scene model, the monitoring event, the scene structure and the algorithm library, a simulation result is obtained. For example, a plurality of scene models and scene structures are obtained according to the monitoring event and the mapping rule, and an algorithm of the algorithm library is called according to the simulation mode, a plurality of scene models are analyzed in combination with the cross-region interconnection topology model to obtain the simulation result.

[0132] Embodiment 2

[0133] In this embodiment, the scheduling unit is specifically further used for:

[0134] A double-layer optimization model is constructed, and the double-layer optimization model includes an upper-layer model and a lower-layer model;

[0135] An instruction mapping rule library is constructed, and the instruction mapping rule library includes a plurality of parameter variables, a region type corresponding to the parameter variables and an instruction template;

[0136] Based on the simulation result, an optimization variable is obtained, a change amount of the optimization variable is obtained, and a plurality of differential instructions are obtained based on the change amount and the instruction mapping rule library; a scheduling strategy is obtained based on the differential instructions; and the instruction mapping rule library is shown in Table 2:

[0137] Table 2 Instruction mapping rule library

[0138] Optimization variable change amount Area type Generate instruction template Active power > 0 Thermal power dominant area Start / stop machine instruction Active power < 0 New energy area Abandon wind / solar instruction Reactive power > 0 Weak power grid area Reactive power compensation instruction

[0139] The first calculation formula of the upper-layer model is:

[0140] ;

[0141] ;

[0142] in, represents the decision variable vector, Indicates the Periods, Indicates the time period, represents the power generation cost function, Indicates time period The active power output of the unit is represents the spare capacity cost, Indicates time period The spare capacity, represents the cascading risk penalty coefficient, represents the cascading failure risk indicator, Indicates the devices, Indicates the total number of devices. Indicates the degree of failure, Indicates the The interconnection line power of each device, Indicates the Safety margin of each device;

[0143] The second calculation formula of the lower model is:

[0144] ;

[0145] in, represents the constraint matrix of region A, represents the state variable, Represents the preset matrix of area A.

[0146] In this embodiment, if there are multiple regions, the constraints of the corresponding regions can be superimposed in the lower model, such as

[0147] ;

[0148] in, represents the constraint matrix of region B, represents the state variable, Represents the preset matrix of area B.

[0149] The scheduling unit is further configured to:

[0150] Obtaining a real-time risk assessment value based on the scheduling strategy, adjusting the constraint boundary of the two-level optimization model based on the real-time risk assessment value, and obtaining an optimization decision based on the adjusted two-level optimization model;

[0151] The third calculation formula of the real-time risk assessment value is:

[0152] ;

[0153] wherein, represents a real-time risk assessment value, represents an inter-area tie-line power, represents a fault degree, represents a safety margin of the device, represents a safety limit of the device, represents a device number.

[0154] The preliminary decision of the upper layer optimization output, such as unit output and tie-line power, is taken as the input of risk assessment, triggering the dynamic adjustment of constraint boundary. For example, when it is detected that the real-time risk assessment value of a certain area > 0.9, the voltage constraint limit value is automatically relaxed (from 10% to 12%), and the adjusted constraint condition directly affects the instruction generation logic after being re-injected into the lower layer constraint for iteration. After the constraint of the East China region is relaxed, the instruction template library preferentially selects the deep peak regulation of thermal power units rather than load shedding, thereby reducing the economic cost.

[0155] Embodiment 3

[0156] On the basis of the above-mentioned embodiments, in this embodiment, the system further comprises a reconstruction unit, and the reconstruction unit is specifically configured to:

[0157] obtain a regional root node of the regional radial topology, and obtain a regional mesh number based on the regional mesh structure;

[0158] obtain reconstruction data, obtain a reconstruction ring network topology, a reconstruction radial topology and a reconstruction mesh network topology based on the reconstruction data, obtain a reconstruction node-branch structure based on the reconstruction ring network topology, obtain a reconstruction tree structure based on the reconstruction radial topology, mark a reconstruction root node of the reconstruction radial topology, and obtain a reconstruction mesh structure and a reconstruction mesh number based on the reconstruction mesh network topology; if the mesh number satisfies a graph theory formula, the mesh number = branch number - node number + connected component number, the connected component number can be obtained according to a depth-first search (DFS) or a Union-Find algorithm.

[0159] obtain a first similarity degree based on the reconstruction ring network topology and the regional ring network topology, for example, the ring network topology is divided into regions, and the similarity is calculated based on a feature vector method or a node / edge centrality measurement comparison; obtain a second similarity degree based on the reconstruction tree structure and the regional tree structure, for example, the tree structure is divided into different regions, and the similarity is calculated by using a tree isomorphism (AHU algorithm), a tree edit distance (TED) or a feature method or a tree kernel method; obtain a third similarity degree based on the reconstruction mesh number and the regional mesh number, for example, the reconstruction mesh number / region mesh number;

[0160] determining whether the first similarity exceeds a first threshold, and if so, obtaining a first similar region based on the first similarity, and obtaining a first adjustment structure based on the reconstructed node-branch structure, the regional node-branch structure, and the first similar region;

[0161] determining whether the second similarity exceeds a second threshold, and if so, obtaining a second similar region based on the second similarity, and obtaining a second adjustment structure based on the reconstructed root node, the regional root node, and the second similar region;

[0162] determining whether the third similarity exceeds a third threshold, and if so, obtaining a third similar region based on the third similarity, and obtaining a first node of the reconstructed mesh structure and a first connection edge of the first node, and obtaining a second node of the regional mesh structure and a second connection edge of the second node;

[0163] obtaining a third adjustment structure based on the first node, the first connection edge, the second node, and the second connection edge;

[0164] obtaining a reconstructed topology model based on the first adjustment structure, the second adjustment structure, the third adjustment structure, and the reconstructed data, and updating the cross-region interconnection topology model to the reconstructed topology model. That is, on the basis of the cross-region interconnection topology model, similar structures are adjusted, and non-similar structures are regenerated according to the reconstructed data.

[0165] Embodiment 4

[0166] On the basis of the above-mentioned embodiments, the present embodiment further provides a multi-scene joint power grid simulation scheduling method, the method comprising:

[0167] obtaining power grid data of different regions, and analyzing the power grid data to obtain heterogeneous data, the heterogeneous data comprising a differentiated topology structure, standard equipment parameters, and rule constraints;

[0168] obtaining a regional parameter set based on the heterogeneous data;

[0169] constructing a cross-region interconnection topology model based on the regional parameter set;

[0170] obtaining a monitoring event based on the power grid data;

[0171] constructing a scene structure, and obtaining a simulation result based on the cross-region interconnection topology model, the monitoring event, and the scene structure.

[0172] While the preferred embodiments of the application have been described, additional variations and modifications can be made to these embodiments by those skilled in the art once they have the benefit of the present disclosure without departing from the spirit and scope of the application. Accordingly, it is intended that the appended claims include all such modifications and variations as fall within the scope of the present application.

[0173] It is apparent that those skilled in the art can make various changes and modifications to the application without departing from the spirit and scope of the application. It is therefore intended that the present application cover all such changes and modifications that are within its scope.

Claims

1. A multi-scenario linkage power grid simulation dispatching system, characterized in that, The system comprises: a data unit configured to acquire power grid data of different regions, parse the power grid data to obtain heterogeneous data, and the heterogeneous data comprising differentiated topological structures, standard equipment parameters and rule constraints; a processing unit configured to obtain a regional parameter set based on the heterogeneous data; a modeling unit configured to construct a cross-region interconnection topological model based on the regional parameter set; an event unit configured to obtain monitoring events based on the power grid data; a dispatching unit configured to construct a scenario structure, and obtain simulation results based on the cross-region interconnection topological model, the monitoring events and the scenario structure; the data unit specifically comprises: a topological unit configured to obtain regional ring network topologies, regional radial topologies and regional mesh network topologies based on the power grid data; obtain regional node-branch structures based on the regional ring network topologies, obtain regional directed graphs based on the regional node-branch structures, obtain regional tree structures based on the regional radial topologies, and obtain regional mesh structures based on the regional mesh network topologies; obtain the differentiated topological structures based on the regional directed graphs, the regional tree structures and the regional mesh structures; a parameter unit configured to obtain equipment parameters based on the power grid data, convert the equipment parameters into standard values, and obtain the standard equipment parameters; a rule unit configured to construct a rule standard library comprising a plurality of standard constraint rules, obtain regional operation rules based on the power grid data, match the regional operation rules with the rule standard library, obtain standard constraint rules based on the rule standard library if the matching is successful, obtain rule thresholds and rule variables based on the regional operation rules if the matching fails, and generate constraint rule expressions based on the rule thresholds and the rule variables; match the regional operation rules with keywords to obtain special variables and special keywords, generate special rules based on the special variables and the special keywords, and generate event trigger functions based on the special rules; and obtain the rule constraints based on the standard constraint rules, the constraint rule expressions and the event trigger functions. The processing unit is specifically configured to: 2.The multi-scenario linkage power grid simulation scheduling system according to claim 1, characterized in that, construct a mapping relationship table of the differentiated topological structures and the standard equipment parameters, inject regional identification codes of the standard equipment parameters, and obtain a labeled regional parameter set based on the differentiated topological structures, the standard equipment parameters, the rule constraints, the mapping relationship table and the regional identification codes. The modeling unit is specifically configured to: 3.The multi-scenario linkage power grid simulation scheduling system according to claim 1, characterized in that, construct a parameter compatibility layer, convert the regional parameter set based on the parameter compatibility layer to obtain a cross-region parameter set; obtain regional boundary nodes based on the differentiated topological structures, obtain regional tie lines based on the regional boundary nodes, and construct the cross-region interconnection topological model based on the regional tie lines and the cross-region parameter set. The event unit is specifically configured to:

4. The multi-scenario linkage power grid simulation scheduling system according to claim 1, characterized in that, obtain a plurality of classification events, first event attributes of the classification events, first trigger conditions of the classification events and first priorities of the classification events based on the power grid data. ​ setting a custom event, a second event attribute of the custom event, a second trigger condition of the custom event, a trigger action of the custom event and a second priority of the custom event; the first event attribute comprises an event location, an event time and a severity, and the second event attribute comprises an added load capacity, an access point location and a time plan; based on the classified event and the custom event, obtaining the monitoring event.

5. The multi-scenario linkage power grid simulation scheduling system according to claim 4, characterized in that, The scheduling unit is specifically used for: constructing a model template library and an algorithm library; setting a mapping rule of the classified event and the custom event and the model template library, the model template library containing modeling templates in different scenarios, and the algorithm library containing several power flow calculations, several reactive power optimization algorithms and several source-grid-load-storage coordination algorithms; constructing the scene structure, the scene structure including an event type, a simulation mode, a time span, a space range, a data dependency and an output requirement field; based on the first trigger condition, the second trigger condition, the mapping rule and the model template library, obtaining several scene models; based on the cross-region interconnection topology model, the scene model, the monitoring event, the scene structure and the algorithm library, obtaining a simulation result. 6.The multi-scenario linkage power grid simulation scheduling system according to claim 5, characterized in that, The scheduling unit is specifically used for: constructing a double-layer optimization model, the double-layer optimization model including an upper-layer model and a lower-layer model; constructing an instruction mapping rule library, the instruction mapping rule library containing several parameter variables, region types corresponding to the parameter variables and instruction templates; based on the simulation result, obtaining an optimization variable, acquiring a change amount of the optimization variable, based on the change amount and the instruction mapping rule library, obtaining several differentiated instructions; based on the differentiated instructions, obtaining a scheduling strategy; a first calculation formula of the upper-layer model is: ； ； wherein, denotes a decision variable vector, denotes the th time period, denotes the time period cycle, denotes the generation cost function, denotes the active power output of the th unit in the time period, denotes the reserve capacity cost, denotes the reserve capacity of the th unit in the time period, denotes the cascading risk penalty coefficient, denotes the cascading failure risk indicator, denotes the th device, denotes the total number of devices, denotes the failure degree, denotes the tie-line power of the th device, denotes the security margin of the th device; a second calculation formula of the lower-layer model is: ； wherein, represents a constraint matrix of region A, represents a state variable, represents a preset matrix. 7.The multi-scenario linkage power grid simulation scheduling system according to claim 6, characterized in that, The scheduling unit is specifically used for: based on the scheduling strategy, obtaining a real-time risk assessment value, adjusting a constraint boundary of the double-layer optimization model based on the real-time risk assessment value, and obtaining an optimization decision based on the adjusted double-layer optimization model; a third calculation formula of the real-time risk assessment value is: ； wherein, represents a real-time risk assessment value, represents an inter-area tie-line power, represents a fault degree, represents a safety margin of the device, represents a safety limit value of the device, represents a device number. 8.The multi-scenario linkage power grid simulation scheduling system according to claim 1, characterized in that, The system further includes a reconstruction unit, which is specifically used for: obtaining a region root node of the region radial topology, and obtaining a region mesh number based on the region mesh structure; obtaining reconstruction data, and obtaining a reconstruction ring network topology, a reconstruction radial topology and a reconstruction mesh network topology based on the reconstruction data; obtaining a reconstruction node-branch structure based on the reconstruction ring network topology, obtaining a reconstruction tree structure based on the reconstruction radial topology, marking a reconstruction root node of the reconstruction radial topology, and obtaining a reconstruction mesh structure and a reconstruction mesh number based on the reconstruction mesh network topology; based on the reconstruction ring network topology and the region ring network topology, obtaining a first similarity, based on the reconstruction tree structure and the region tree structure, obtaining a second similarity, and based on the reconstruction mesh number and the region mesh number, obtaining a third similarity; determining whether the first similarity exceeds a first threshold, and if so, obtaining a first similar region based on the first similarity, and obtaining a first adjustment structure based on the reconstructed node-branch structure, the regional node-branch structure, and the first similar region; determining whether the second similarity exceeds a second threshold, and if so, obtaining a second similar region based on the second similarity, and obtaining a second adjustment structure based on the reconstructed root node, the regional root node, and the second similar region; determining whether the third similarity exceeds a third threshold, and if so, obtaining a third similar region based on the third similarity, and obtaining a first node of the reconstructed mesh structure and a first connection edge of the first node, and obtaining a second node of the regional mesh structure and a second connection edge of the second node; obtaining a third adjustment structure based on the first node, the first connection edge, the second node, and the second connection edge; obtaining a reconstructed topology model based on the first adjustment structure, the second adjustment structure, the third adjustment structure, and the reconstructed data, and updating the cross-region interconnection topology model to the reconstructed topology model.

9. A multi-scenario linkage power grid simulation scheduling method, characterized in that, The method comprises: obtaining power grid data of different regions, and parsing the power grid data to obtain heterogeneous data, the heterogeneous data comprising a differentiated topology structure, standard equipment parameters, and rule constraints; obtaining a regional parameter set based on the heterogeneous data; constructing a cross-region interconnection topology model based on the regional parameter set; obtaining a monitoring event based on the power grid data; constructing a scenario structure, and obtaining a simulation result based on the cross-region interconnection topology model, the monitoring event, and the scenario structure; The specific steps of parsing the power grid data to obtain heterogeneous data comprise: obtaining a regional ring network topology, a regional radial topology, and a regional mesh network topology based on the power grid data; obtaining a regional node-branch structure based on the regional ring network topology, and obtaining a regional directed graph based on the regional node-branch structure; obtaining a regional tree structure based on the regional radial topology; and obtaining a regional mesh structure based on the regional mesh network topology; obtaining the differentiated topology structure based on the regional directed graph, the regional tree structure, and the regional mesh structure; obtaining equipment parameters based on the power grid data, converting the equipment parameters to a per-unit value, and obtaining the standard equipment parameters; constructing a rule standard library, the rule standard library containing a plurality of standard constraint rules; and matching the regional operation rules with the rule standard library, and if the matching is successful, obtaining a standard constraint rule based on the rule standard library; if the matching fails, obtaining a rule threshold and a rule variable based on the regional operation rules, and generating a constraint rule expression based on the rule threshold and the rule variable; and matching the regional operation rules with keywords to obtain a special variable and a special keyword, generating a special rule based on the special variable and the special keyword, and generating an event trigger function based on the special rule; and obtaining the rule constraints based on the standard constraint rule, the constraint rule expression, and the event trigger function.

Images