A full life cycle and multi-service domain micro-grid digital twin modeling method

Abstract

The application provides a full-life-cycle and multi-service-domain micro-grid digital twin modeling method, and relates to the field of micro-grid modeling.The application determines an input measurement point set, an acting electrical object set and a strategy parameter set, avoids cross-domain fragmentation caused by inconsistent asset identification, non-unique measurement point binding and unclear strategy reference object in traditional engineering.The application constructs a system state vector according to a fixed measurement point sequence, ensures that the digital twin model still has a usable state input interface under the condition of missing and abnormal measurement points.The application gives a deterministic mapping rule for the strategy output control quantity, combines with the device parameter constraint and cross-service-domain consistency check, ensures that the control output is uniquely determined under the same input condition and does not violate the operation constraint.The application generates three service domain differences in the expansion and reconstruction stage, and then forms a new version of the model package and re-executes the consistency check, so that the reconstruction change can be differentially recorded, the version can be traced back, and the model after reconstruction can still be effective.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of micro-grid modeling, in particular to a full life cycle and multi-business domain micro-grid digital twin modeling method. BACKGROUND

[0002] In the application of micro-grid digital twin, domain modeling is used to support the computability, verifiability and traceability management of the model. In view of the continuous state perception, operation and maintenance disposal and expansion and reconstruction change requirements after the micro-grid is put into operation, the present application limits the life cycle stage to the operation control stage, the operation and maintenance stage and the expansion and reconstruction stage, and limits the business domain to the electrical topology and equipment model business domain, the measurement point state business domain and the energy management and control strategy business domain. In existing engineering, the data of the above three types of business domains are respectively derived from equipment account and electrical connection relationship data, measurement point table and real-time measurement data, strategy configuration and control instruction data. There are problems such as inconsistent asset identification, lack of unique binding relationship between measurement points and equipment objects, missing measurement point effectiveness and missing value processing rules, unstructured definition of strategy input measurement point and action object relationship, lack of difference record and version trace of reconstruction and change, etc., which make it difficult to realize the closed loop of operation analysis, operation positioning and reconstruction influence evaluation on the same model main line.

[0003] The Chinese patent "CN202511279705.6 A micro-grid digital twin modeling method" aims to provide high-quality modeling data support for micro-grid operation state monitoring, simulation prediction and optimal scheduling. Its processing flow can be summarized as follows: (1) directly collecting micro-grid operation data through sensors, including meteorological data, equipment state data and load data; (2) for data that cannot be directly collected, first use decision tree as a data type classifier, then use state estimator to obtain the corresponding data, and form the micro-grid operation state monitoring result; (3) for missing data required for modeling, use generative adversarial network for data completion; (4) further obtain high-quality modeling data through a multi-modal fusion method based on deep clustering, so as to support real-time monitoring and optimal scheduling, operation condition simulation prediction, system fluctuation suppression and fault and maintenance demand prediction, and improve the stability and energy utilization efficiency of micro-grid.

[0004] "CN202511279705.6 A micro-grid digital twin modeling method" mainly focuses on the classification and state estimation of operation data collection, missing data completion, and multi-modal fusion to generate modeling data. However, this type of solution still has obvious defects in the engineering-level digital twin domain modeling level: first, the method focuses on improving the quality of modeling data, but does not include key business information such as electrical topology and equipment model, measurement point state, energy management and control strategy in the same "computable, verifiable, traceable" domain model main line, which can easily lead to cross-business domain fragmentation and difficulty in closed-loop supporting operation analysis and control reproduction; second, it lacks a unified correlation mechanism centered on unique asset identification, and does not systematically constrain the unique binding of measurement points-asset objects, the structured mapping of strategy input measurement points and action objects, making it difficult to ensure consistent expression and reproducible calculation of the model under multi-system and multi-source data; third, it does not cover the full life cycle evolution requirements after commissioning, especially the lack of systematic support for change impact assessment, differential record and version traceability in operation and maintenance and expansion and reconstruction scenarios. SUMMARY

[0005] To overcome the shortcomings of the prior art, the present application aims to provide a micro-grid digital twin modeling method for the full life cycle and multiple business domains, comprising: Step 1: Import electrical equipment account data and electrical connection relationship data, and perform field mapping and verification on the electrical objects contained therein to obtain a plurality of electrical objects that pass the verification; the electrical objects are primary equipment or electrical units in the micro-grid; Step 2: According to the electrical equipment account data and the electrical connection relationship data, construct the electrical topology and equipment model business domain , the measurement point state business domain , and the energy management and control strategy business domain ; Step 3: Verify the electrical topology and equipment model business domain , the measurement point state business domain , and the energy management and control strategy business domain , and obtain consistency verification conclusion ; When , , pass the verification, and when , the verification , , does not pass, and output the defects of the verification process record; Step 4: Integrate the electrical topology and equipment model business domain , the measurement point state business domain , and the energy management and control strategy business domain and consistency check conclusion Packaging as a running control stage digital twin model package iff , marked as effective when , marked as ineffective; Step 5: In the operation and maintenance stage, build an operation and maintenance event set, and according to the operation and maintenance event set, verify to generate an operation and maintenance stage digital twin model package ; Step 6: In the expansion and reconstruction stage, the version evolution is carried out on and , and the new version model package generated is verified according to step 3, when , mark as effective and take it as an expansion and reconstruction stage digital twin model package ; when , mark as ineffective.

[0006] Optionally, step 1 specifically includes: Step 1.1: Import the electrical equipment account data and electrical connection relationship data and map them to the preset standard field to obtain a standardized data structure; Step 1.2: Set the basic field and the key field combination, the key field combination containing one or more key fields, the key field representing the electrical connection relationship of the electrical object; Step 1.3: For each electrical object, according to the basic field and the key field combination, the field verification is carried out on the standardized data structure, when the basic field is missing or the key field combination cannot uniquely determine the electrical object or its electrical connection relationship, it is recorded as a defect item and the defect list is output; when the basic field is not missing and the key field combination can uniquely determine the electrical object and its electrical connection relationship, it is characterized by passing the verification, and then a plurality of electrical objects passing the verification are obtained.

[0007] Optionally, the electrical topology and device model business domain in step 2 is constructed by the following steps: Step A1: Map the electrical objects passing the verification to nodes to obtain a node set , the number of electrical object nodes, and a unique asset identifier is established for each node ; Step A2: According to the electrical connection relationship data, generate a connection edge set between electrical objects ,​​ is the number of connection edges; Step A3: for each node , extract the node from the electrical equipment account data according to the unique asset identifier , obtain the device parameter set , wherein represents the device parameter item of the node ; set the standard device parameter item, when the device parameter item in the standard device parameter item is missing in the device parameter set, the missing device parameter item is taken as the missing parameter item, when the parameter value of the device parameter item exceeds the preset allowed range, the physical reasonable range or the standard rated range of the electrical object, the device parameter item is taken as the out-of-range item; the missing parameter item is taken as the parameter defect list, or the out-of-range item is taken as the parameter defect list, or the missing parameter item and the out-of-range item are taken together to form the parameter defect list; Step A4: according to the node set and the connection edge set, construct the electrical topology graph model ; the electrical topology graph model, the unique asset identifier of the node set, the device parameter set constitute the electrical topology and device model business domain , and output the parameter defect list.

[0008] Optionally, the measurement point state business domain in step 2 is constructed by the following steps: Step B1: obtain the measurement point table and generate the measurement point set S according to the measurement point table, each measurement point in the measurement point set S contains the measurement point identifier, the unit, the sampling period, the value range and the target identifier, the target identifier contains the electrical object identifier to which the measurement point belongs, or the loop identifier, or the electrical object identifier to which the measurement point belongs and the loop identifier; Step B2: determine the unique asset identifier of the measurement point according to the target identifier; Specifically, when the target identifier contains the electrical object identifier to which the measurement point belongs, the unique asset identifier is matched according to the electrical object identifier to which the measurement point belongs, when the target identifier contains only the loop identifier, a plurality of candidate electrical objects belonging to the same loop identifier are retrieved in the electrical connection relationship data according to the loop identifier, and the unique asset identifier is determined by further determining the measurement point name, the measurement point type, the endpoint information in the electrical connection relationship data or the device type in the electrical equipment account data; Step B3: for each measurement point At each time point Extract observation value ; Step B4: According to the observation value , generate quality label ; Specifically, when the observation value satisfies unit consistency, and satisfies the value range constraint, and satisfies the timestamp validity, , otherwise ; Wherein, the observation value satisfies unit consistency means that the unit of the observation value is consistent with the unit corresponding to the measuring point, the observation value satisfies the value range constraint means that the observation value is within the value range corresponding to the measuring point, and the observation value satisfies the timestamp validity means that the timestamp information of the observation value is complete and the timestamp is corresponding to the sampling period of the measuring point; Step B5: According to the observation value and the quality label , generate the measuring point available value , specifically, when , let ; When , let ; Step B6: According to the fixed measuring point sequence, construct the state vector , wherein Q represents the total number of measuring points, represents the measuring point available value of the measuring point ; The measuring point set S, the unique asset identifier corresponding to the measuring point , and the state vector constitute the measuring point state business domain .

[0009] Optionally, the energy management and control strategy business domain is constructed by the following steps: Step C1: Obtain strategy configuration data in preconfigured strategy description information, wherein the strategy description information at least includes strategy name, strategy type, input measuring point definition, action electrical object definition, strategy parameter and control output rule; The strategy configuration data is parsed, standardized and structured to obtain a strategy parameter set and a plurality of strategy groups to form a strategy set ; Step C2: parsing each policy to obtain multiple input measurement points and the arrangement order of the input measurement points in the policy, and then generating an ordered input measurement point sequence ; The input measurement points in the ordered input measurement point sequence belong to the measurement point set, and when the input measurement points in the ordered input measurement point sequence do not belong to the measurement point set, a policy input defect is recorded; Step C3: parsing each policy to obtain the electrical objects acted on by the policy and mapping them to a unique asset identifier set , obtaining a corresponding device parameter set ; Among them, the unique asset identifier of the electrical object acted on in is consistent with the unique asset identifier of the corresponding node, and when the unique asset identifier of the electrical object acted on in is inconsistent with the unique asset identifier of the corresponding node, a policy electrical object defect is recorded; Step C4: extracting in the state vector according to the ordered input measurement point sequence to obtain a policy input vector ; Step C5: generating a policy output control quantity according to , which is represented as: ; Among them, is the deterministic control mapping function corresponding to the i-th policy; The , , , , , and constitute an energy management and control policy business domain .

[0010] Optionally, step 3 specifically includes: Step 3.1: verifying whether the unique asset identifiers of the node set are unique and complete, verifying whether the unique asset identifiers bound by all measurement points can be uniquely located in the unique asset identifiers of the node set, and verifying whether the unique asset identifiers of all electrical objects acted on in can be uniquely located in the unique asset identifiers of the node set; ​If the unique asset identifiers of the node set are unique and complete, and the unique asset identifiers bound by all the measurement points can be uniquely located in the unique asset identifiers of the node set, and all the unique asset identifiers of the acting electrical objects in the node set can be uniquely located in the unique asset identifiers of the node set, then the asset identifier consistency is 1, otherwise the asset identifier consistency is 0, and the content that fails the verification is recorded as a defect; Step 3.2: Verify whether there is and only one bound unique asset identifier for each measurement point, and the node of the electrical object corresponding to the unique asset identifier belongs to the node set; If there is and only one bound unique asset identifier for each measurement point, and the node of the electrical object corresponding to the unique asset identifier belongs to the node set, then the measurement point binding legality is 1, otherwise the measurement point binding legality is 0, and the content that fails the verification is recorded as a defect; Step 3.3: Verify whether the ordered input measurement point sequence of each strategy is a subset of the measurement point set, if the ordered input measurement point sequence of each strategy is a subset of the measurement point set, then the strategy input legality is 1, otherwise the strategy input legality is 0, and the content that fails the verification is recorded as a defect; Step 3.4: Verify whether the unique asset identifiers of the acting electrical objects in the of each strategy are consistent with the unique asset identifiers of the nodes corresponding thereto, if the unique asset identifiers of the acting electrical objects in the of each strategy are consistent with the unique asset identifiers of the nodes corresponding thereto, then the acting object legality is 1, otherwise the acting object legality is 0, and the content that fails the verification is recorded as a defect; Step 3.5: Verify whether the strategy input vector of each strategy is constructed according to the ordered input measurement point sequence in the state vector , and verify whether the strategy output control quantity can be calculated according to , and , and verify whether the strategy output control quantity passes the constraint check by taking as a constraint source; In the strategy input vector of each strategy, the ordered input measurement point sequence is constructed in the state vector , and the policy output control quantity can be calculated according to , and , and the policy output control quantity is checked by taking as a constraint source, so that the policy computability is 1, otherwise the policy computability is 0, and the contents of the failed check are recorded as defects; Step 3.6: Calculate the consistency check conclusion ; when , it represents , , the check passes, when , then it represents the check , , does not pass, and outputs the recorded defects.

[0011] Optionally, step 5 specifically includes: In the operation and maintenance stage, an operation and maintenance event set is constructed, for the operation and maintenance event in the operation and maintenance event set, the device identifier of the operation and maintenance event is mapped to obtain a unique asset identifier, and it is checked whether the unique asset identifier can be uniquely located in the node set, if yes, it is represented that the operation and maintenance event is associated with the node in the node set, and the operation and maintenance event, the event timestamp, the unique asset identifier of the operation and maintenance event, and the disposal action and disposal result contained in the operation and maintenance event are written into the operation and maintenance stage digital twin model package . If not, it is represented that the operation and maintenance event is abnormally associated with the node in the node set, and the abnormal operation and maintenance event is recorded.

[0012] Optionally, step 6 specifically includes: Take as the first version model package , obtain the electrical equipment account data, electrical connection relationship data, measurement point table and strategy configuration data after expansion and reconstruction, and then generate the first version model package . Compare and item by item to generate electrical topology and device model business domain difference , measurement point state business domain difference and energy management and control strategy business domain difference , and then generate a new version of the model package , wherein, is a version update operator; a new version model package According to step 3, when is marked as effective and is taken as the expansion and reconstruction phase digital twin model package , while , , and , , ; when is marked as not effective.

[0013] The beneficial effects produced by the above technical solutions are: In terms of unified modeling across business domains, the application organizes electrical topology and device parameters, measurement point tables and real-time measurements, energy management and control strategy configuration in a unified asset main line, clearly defines input measurement point sets, acting electrical object sets and strategy parameter sets, and avoids the cross-domain fragmentation caused by inconsistent asset identification, non-unique measurement point binding and unclear strategy reference objects in traditional engineering.

[0014] In terms of state synchronization and availability, the application generates measurement point availability values through measurement point data quality markers and the last valid observation value, and constructs a system state vector in a fixed measurement point order, ensuring that the digital twin model still has a usable state input interface in the case of missing and abnormal measurement points.

[0015] In terms of control behavior reproducibility and verifiability, the application gives a deterministic mapping rule for strategy output control quantities, and combines device parameter constraints and cross-business domain consistency verification to ensure that control output is uniquely determined under the same input conditions and does not violate operating constraints, thereby improving the reproducibility of operation analysis and strategy execution.

[0016] In terms of expansion and reconstruction traceability, the application generates electrical topology and device model business domain differences, measurement point state business domain differences and energy management and control strategy business domain differences in the expansion and reconstruction phase, applies the differences in a deterministic order to form a new version model package and re-performs consistency verification, so that the reconstruction changes can be recorded in differences, the version is traceable, and the post-reconstruction model can still be effective. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is a flowchart of a full life cycle and multi-business domain microgrid digital twin modeling method in an embodiment of the application. DETAILED DESCRIPTION

[0018] ​The specific embodiments of the present application are described in further detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application, but are not used to limit the scope of the present application.

[0019] To solve the above problems, the present application provides a full life cycle and multi-business domain micro-grid digital twin modeling method: an electrical topology graph model and a device parameter set are constructed with a unique asset identifier as the main line, a unique binding relationship between the measurement point and the asset object is established, and a system state vector generation rule is formed, an input measurement point set, an action object set, and a deterministic input-output mapping of the energy management and control strategy are defined, a cross-business domain consistency check rule is set, and when the expansion and reconstruction event is triggered, the electrical topology and device model business domain difference, the measurement point state business domain difference, and the energy management and control strategy business domain difference are generated, and the model version is updated, thereby forming a unified modeling and governance process of the operation control stage digital twin model package, the operation and maintenance stage digital twin model package, and the expansion and reconstruction stage digital twin model package.

[0020] The present application provides a full life cycle and multi-business domain micro-grid digital twin modeling method, which is used to realize the unified construction and versioned governance of the digital twin model in the operation control stage , the operation and maintenance stage and the expansion and reconstruction stage of the micro-grid after commissioning, wherein the full life cycle of the present application is limited to the above three stages, and the multi-business domain of the present application is limited to the electrical topology and device model business domain , the measurement point state business domain and the energy management and control strategy business domain . The method constructs an electrical topology graph model and a device parameter set with a unique asset identifier as the main line, establishes a unique binding relationship between the measurement point and the asset object, defines the measurement point data quality mark and the system state vector generation rule, further defines the input measurement point set, the action object set of the energy management and control strategy, and the deterministic mapping rule for calculating the strategy output control amount from the strategy input vector, thereby realizing the closed-loop association of the three types of business domains in the same model system and the cross-business domain consistency check.

[0021] Specifically, the method first generates the operation control stage digital twin model package, then uniquely associates the operation and maintenance events and asset identifiers to generate the operation and maintenance stage digital twin model package, and when the expansion and reconstruction event occurs, generates the electrical topology and device model business domain difference, the measurement point state business domain difference, and the energy management and control strategy business domain difference, generates a new version of the model according to the version update rule, and re-executes the cross-business domain consistency check, thereby ensuring the consistent expression, calculation and traceability of the model in the above three stages. The application provides a micro-grid digital twin modeling method for the whole life cycle and multiple business domains, which realizes consistent expression, calculation and traceable management of the model in the operation, operation and maintenance and reconstruction stages of the micro-grid after commissioning by uniformly binding the electrical topology and equipment model, the measurement point state and the energy management and control strategy mapping, and combining cross-domain consistency checking and reconstruction differential version evolution. is the boundary, wherein is the operation control stage, is the operation and maintenance stage, is the expansion and reconstruction stage; the business domain set is the range, wherein is the electrical topology and equipment model business domain, is the measurement point state business domain, is the energy management and control strategy business domain. The order of "first modeling, then checking, then packaging, then evolution" is executed, and the output of each step is uniquely associated with the unique asset identifier as a unified association key, in combination with Figure 1 , specifically comprising the following steps: Step 1: Import electrical equipment account data and electrical connection relationship data, and perform field mapping and checking on the electrical objects contained therein to obtain a plurality of electrical objects that pass the checking; The electrical object is a primary equipment or electrical unit in the micro-grid, including but not limited to transformers, circuit breakers, disconnectors, busbars, feeder lines, cables, energy storage converters, photovoltaic inverters, load units and grid-connected point equipment; the electrical connection relationship refers to the actual connection relationship between the above-mentioned electrical objects based on endpoints, loops or lines.

[0022] Step 1.1: Import electrical equipment account data and electrical connection relationship data and map them to a pre-set standard field to obtain a standardized data structure; Step 1.2: Set the basic field and the key field combination, the key field combination contains one or more key fields, and the key field represents the electrical connection relationship of the electrical object; The basic field includes at least: device name, device type, device identifier, model specification and endpoint information related to connection relationship identification; for devices with rated parameters, the basic field also includes rated parameter fields. The key field combination includes one or more of device identifier, endpoint identifier and loop identifier; in different data sources, it can also include connection object identifier, connection end number, line identifier, interval identifier and other fields related to unique determination of the connection relationship.

[0023] The electrical connection relationship refers to an actual electrical connection or conduction relationship between two electrical objects in the microgrid based on a line, a bus, a switch, a terminal, or a grid-connected interface. Each connection edge is determined by its two end nodes and a connection attribute, wherein the two end nodes correspond to two connected electrical objects respectively. The connection attribute is used to represent specific characteristics of the connection relationship, including but not limited to connection type, starting endpoint identifier, ending endpoint identifier, loop identifier, line / bus identifier, connection direction, and connection state. Each connection edge is in one-to-one correspondence with the endpoint information of the corresponding nodes in the node set .

[0024] Step 1.3: For each electrical object, perform field verification on the normalized data structure according to the basic fields and key field combinations. If the basic fields are missing or the key field combinations cannot uniquely determine the electrical object or its electrical connection relationship, record it as a defect item and output a defect list. If the basic fields are not missing and the key field combinations can uniquely determine the electrical object and its electrical connection relationship, it is considered that the verification is passed, and then a plurality of electrical objects that pass the verification are obtained. Step 2: Construct electrical topology and device model business domains , measurement point state business domains , and energy management and control strategy business domains according to the electrical equipment account data and the electrical connection relationship data. The electrical topology and device model business domains include an electrical topology graph model, a unique asset identifier of each node in the node set of the electrical topology graph model , and a device parameter set . The measurement point state business domains include a measurement point set S, a unique asset identifier corresponding to the measurement point , and a state vector . The energy management and control strategy business domains include a strategy set , a unique asset identifier set of electrical objects affected by the strategy , a strategy parameter set , a deterministic control mapping function corresponding to the strategy, and a strategy output control quantity . The electrical topology and device model business domains are constructed by the following steps: Step A1: Map the electrical objects that pass the verification to nodes to obtain a node set , where N is the number of electrical object nodes, and a unique asset identifier is established for each node . ; Step A2: generating a connection edge set between electrical objects according to electrical connection relationship data , is the number of connection edges; Step A3: for each node , extracting the node corresponding electrical object from the electrical equipment account data according to the unique asset identifier , obtaining a device parameter set , wherein represents the device parameter item of the node ; setting standard device parameter items, when a device parameter item in the standard device parameter item is missing in the device parameter set, the missing device parameter item is taken as a missing parameter item, when a parameter value of a device parameter item exceeds the preset allowed range, physical reasonable range or standard rated range of the electrical object, the device parameter item is taken as an out-of-range item; the missing parameter item is taken as a parameter defect list, or the out-of-range item is taken as a parameter defect list, or the missing parameter item and the out-of-range item are taken together to form a parameter defect list; , the rated parameter is the standardized upper field, which is used to uniformly connect the rated parameter information in the equipment account of different sources; rated voltage, rated capacity, switch rated current, energy storage rated energy capacity, etc. are specific parameter items after the rated parameter is expanded according to the type of electrical object, and are bound to the corresponding node. Wherein, binding means establishing a corresponding relationship between the extracted specific device parameter item and the node

[0025] Step A4: constructing an electrical topology graph model according to the node set and the connection edge set ; the electrical topology graph model, the unique asset identifier of the node set , the device parameter set constitute the electrical topology and device model business domain , and output the parameter defect list at the same time; Wherein, the measurement point state business domain is constructed by the following steps: Step B1: Obtain the measurement point table and generate a measurement point set S according to the measurement point table, each measurement point in the measurement point set S contains a measurement point identifier, a unit, a sampling period, a value range, and a target identifier, the target identifier contains an electrical object identifier to which the measurement point belongs, or a loop identifier, or both the electrical object identifier and the loop identifier to which the measurement point belongs; wherein when the loop identifier field is not directly provided in the measurement point table, the loop information in the electrical connection relationship data can be associated and determined according to the electrical object and the electrical connection relationship data.

[0026] Step B2: Determine the measurement point according to the target identifier Corresponding unique asset identifier ; Specifically, when the target identifier contains the electrical object identifier to which the measurement point belongs, the unique asset identifier is matched according to the electrical object identifier to which the measurement point belongs When the target identifier only contains the loop identifier, a plurality of candidate electrical objects belonging to the same loop identifier are retrieved in the electrical connection relationship data according to the loop identifier, and the unique asset identifier is determined by further determining the measurement point name, the measurement point type, the endpoint information in the electrical connection relationship data, or the device type in the electrical equipment account data ; the endpoint information is used to represent the connection position of the candidate electrical object in the connection relationship, such as the access end, the outgoing end, the bus side, the load side, the grid side, or the terminal number; the device type is used to represent the category to which the candidate electrical object belongs, such as circuit breaker, transformer, energy storage converter, photovoltaic inverter, bus, or feeder; finally forming a unique binding function , and the same measurement point cannot be bound to multiple .

[0027] Step B3: For each measurement point At each time Extract the observation value ; Step B4: Generate a quality label according to the observation value ; ; Specifically, when the observation value satisfies the unit consistency, satisfies the value range constraint, and satisfies the timestamp validity, , otherwise ; Wherein, the observation value satisfies the unit consistency, which means that the unit of the observation value is consistent with the unit corresponding to the measurement point, the observation value satisfies the value range constraint, which means that the observation value is within the value range corresponding to the measurement point, and the observation value satisfies the timestamp validity, which means that the observation value The timestamp information is complete and corresponds to the sampling period of the measurement point; Step B5: Based on the observed values and quality marking Generate usable values ​​for measurement points Specifically, when season ;when season ; Step B6: Construct the state vector according to the fixed measurement point sequence. Where Q represents the total number of measuring points, Indicates the measuring point Available values ​​of the measuring points; the set of measuring points S, measuring points Corresponding unique asset identifier State vector Constituting the measurement point status service domain The fixed measurement point sequence remains unchanged within the same model version, and its changes are recorded through differential recording during version evolution.

[0028] Among them, the energy management and control strategy business domain Build it using the following steps: Step C1: Obtain strategy configuration data from the pre-configured strategy description information, which includes at least the strategy name, strategy type, input measurement point definition, active electrical object definition, strategy parameters, and control output rules. The strategy description information is stored in the energy management system, control system, or strategy management module; each strategy is assigned a unique strategy identifier to ensure traceability and referencing during version evolution. Establishing a strategy set refers to representing multiple discrete strategy configuration records as multiple uniquely identifiable strategy objects using a unified strategy data structure.

[0029] The strategy configuration data is parsed, standardized, and structured to obtain a set of strategy parameters. A strategy set is composed of multiple strategies. ; To illustrate the strategy and strategy parameter set, if a certain strategy is to charge energy during off-peak hours and discharge during peak hours, then its strategy parameter set can include: upper limit of charging power, upper limit of discharging power, upper and lower thresholds of SOC, time period division parameters, target power value, etc.

[0030] Step C2: Analyze each policy to obtain multiple input measurement points and their order in the policy, thereby generating an ordered sequence of input measurement points. ; The ordered input measurement point sequence The input measurement points belong to the measurement point set, and are in an ordered input measurement point sequence. If the input measurement point does not belong to the measurement point set, it is recorded as a strategy input defect. Step C3: Parse each strategy to obtain the electrical object affected by the strategy, and map it to a set of unique asset identifiers. , obtain Corresponding set of equipment parameters ; in, The unique asset identifier of the electrical object in the middle is consistent with the unique asset identifier of its corresponding node. When the unique asset identifier of a policy-driven electrical object is inconsistent with the unique asset identifier of its corresponding node, it is recorded as a defect in the policy-driven electrical object. Step C4: Input the measurement point sequence in an ordered manner In the state vector Extracting from the data yields the policy input vector. ; Step C5: According to , and Generate strategy output control quantity , is represented as: ; in, For the first The deterministic control mapping function corresponding to each strategy; The , , , , and The business domain of energy management and control strategy .

[0031] Step 3: Business domain of electrical topology and equipment model Measurement point status business domain Energy Management and Control Strategy Business Domain The consistency check was performed, and the results were obtained. ,when Time, characterization , , The verification passed, when When, then the characterization verification , , If it fails, output the defects recorded during the verification process; Step 3.1: Check whether the unique asset identifiers of the node set are unique and complete, and check whether the unique asset identifiers bound by all measurement points can be uniquely located in the unique asset identifiers of the node set, and check whether the unique asset identifiers of all acting electrical objects in the node set can be uniquely located in the unique asset identifiers of the node set. If the unique asset identifiers of the node set are unique and complete, and the unique asset identifiers bound by all measurement points can be uniquely located in the unique asset identifiers of the node set, and the unique asset identifiers of all acting electrical objects in the node set can be uniquely located in the unique asset identifiers of the node set, then the asset identifier consistency is 1, otherwise the asset identifier consistency is 0, and the content that fails to pass the check is recorded as a defect. Step 3.2: Check whether each measurement point has and only has one bound unique asset identifier, and the node of the electrical object corresponding to the unique asset identifier belongs to the node set. If each measurement point has and only has one bound unique asset identifier, and the node of the electrical object corresponding to the unique asset identifier belongs to the node set, then the measurement point binding legality is 1, otherwise the measurement point binding legality is 0, and the content that fails to pass the check is recorded as a defect. Step 3.3: Check whether the ordered input measurement point sequence of each strategy in the strategy input measurement point sequence is a subset of the measurement point set. If the ordered input measurement point sequence of each strategy is a subset of the measurement point set, then the strategy input legality is 1, otherwise the strategy input legality is 0, and the content that fails to pass the check is recorded as a defect. Step 3.4: Check whether the unique asset identifiers of the acting electrical objects in the strategy action object sequence of each strategy are consistent with the unique asset identifiers of the nodes corresponding thereto. If the unique asset identifiers of the acting electrical objects in the strategy action object sequence of each strategy are consistent with the unique asset identifiers of the nodes corresponding thereto, then the action object legality is 1, otherwise the action object legality is 0, and the content that fails to pass the check is recorded as a defect. It needs to be explained that whether the strategy input measurement point sequence can be constructed by the measurement point set according to the state vector is checked, that is, whether all the referenced measurement points have entered the measurement point set and can be located in the state vector.

[0032] ​​​​​​​​​​​​​​​​Step 3.5: Verify the policy input vector for each policy. Whether to follow the ordered input measurement point sequence In the state vector The system is constructed and the policy output control quantity is verified. Can it be based on , and The calculated and verified output control quantity of the strategy is obtained. Whether to pass Constraint checks as a source of constraints; Policy input vector for each policy It is based on an ordered input sequence of measurement points. In the state vector The structure is constructed within the framework, and the policy outputs control variables. Able to be based on , and The calculated value is the output control quantity of the strategy. By Constraint checks, as a source of constraints, enable policy computability. Set it to 1, otherwise make the strategy computable. The value is 0, and any content that fails the verification is recorded as a defect; It should be noted that the verification Can it be calculated and passed according to formula (4)? Constraint checking involves verifying whether the input vector, policy parameters, and device parameters are complete, and whether the calculation results meet the operational constraints.

[0033] Step 3.6: Calculate the consistency check conclusion ;when Time, characterization , , The verification passed, when When, then the characterization verification , , If it fails, output the defect record. Step 4: Integrate electrical topology with equipment model business domain Measurement point status business domain Energy Management and Control Strategy Business Domain Consistency verification conclusions Encapsulated as a digital twin model package for the operation control phase If and only if hour, Marked as effective when hour, Marked as ineffective; Step 5: constructing an operation and maintenance event set in the operation and maintenance stage, verifying the operation and maintenance event set according to the operation and maintenance event set, and generating an operation and maintenance stage digital twin model package ; ; In the operation and maintenance stage, the operation and maintenance event set is constructed, for the operation and maintenance event in the operation and maintenance event set, the device identifier of the operation and maintenance event is mapped to obtain a unique asset identifier, it is verified whether the unique asset identifier can be uniquely located in the node set, if yes, it is represented that the operation and maintenance event is associated with the node in the node set, and the operation and maintenance event, the event timestamp, the unique asset identifier of the operation and maintenance event, and the disposal action and disposal result contained in the operation and maintenance event are written into the operation and maintenance stage digital twin model package ; , The operation and maintenance disposal can be associated and inquired on the same asset main line; Wherein, the disposal action and disposal result are derived from the operation and maintenance event record, the maintenance work order record or the operation and maintenance disposal feedback record, and are respectively used to represent the processing measures and the processing conclusion for the operation and maintenance event If not, it is represented that the operation and maintenance event is abnormally associated with the node in the node set, and the abnormal operation and maintenance event is recorded; Step 6: in the expansion and reconstruction stage, the version evolution is performed on and the new version model package generated is verified according to step 3, when , the is marked as effective and is taken as an expansion and reconstruction stage digital twin model package ; when , the is marked as ineffective; The is taken as the first version model package , the electrical equipment account data, the electrical connection relationship data, the measuring point table and the strategy configuration data after the expansion and reconstruction are obtained, and then the first version model package is generated; It should be noted that the operation and maintenance stage digital twin model package which has been effective is taken as an old version baseline by the application; if there is no operation and maintenance stage digital twin model package which has been effective, the running control stage digital twin model package which has been effective is taken as an old version baseline, and the old version is taken as the first version model package ;

[0034] The and Perform item-by-item comparisons to identify new, deleted, and modified items, record the values ​​before and after the difference and the associated asset identifiers, and generate the electrical topology and equipment model business domain difference. Measurement point status service domain differential Differentiated from the energy management and control strategy business domain This will generate a new version of the model package. ,in, For version update operators; For the new version of the model package Perform the verification as per step 3. At that time, Marked as effective and included in the digital twin model package for the capacity expansion and upgrade phase. At the same time, retain , and , , ;when At that time, Marked as ineffective.

[0035] The key points of this invention are mainly in the following aspects: 1. This invention uses a unique asset identifier as a unified thread to achieve closed-loop association and consistent expression of three types of business information: electrical topology and equipment model, measurement point status, and energy management and control strategy, in the operation and control phase, maintenance and repair phase, and expansion and renovation phase.

[0036] 2. This invention establishes a unique binding relationship between measurement points and electrical objects corresponding to unique asset identifiers, and constructs a system state vector through measurement point data quality marking and measurement point available value generation rules, forming a unified state input interface for digital twins.

[0037] 3. This invention defines the set of input measurement points, the set of electrical objects acting on the strategy, and the set of strategy parameters for each energy management and control strategy, and defines a deterministic mapping rule from the strategy input vector to the strategy output control quantity, thereby realizing the reproducible calculation of digital twin control behavior.

[0038] 4. During the capacity expansion and renovation phase, this invention generates differentials for the electrical topology and equipment model, measurement point status, and energy management and control strategies, and performs version evolution. At the same time, it re-performs cross-business domain consistency verification to achieve traceable governance of model versions and ensure computability after the renovation.

[0039] The present application proposes a micro-grid digital twin domain modeling method facing the whole life cycle and multiple business domains around the three stages of operation control stage, operation and maintenance stage and expansion and reconstruction stage and the three business domains of electrical topology and equipment model business domain, measurement point state business domain and energy management and control strategy business domain. The electrical topology node, measurement point binding relationship and strategy action object are closed loop associated in the same model system through the unique asset identification as the main line, and the measurement point data quality mark, measurement point available value and system state vector construction rule are used to realize the synchronization of the twin state, and the deterministic mapping rule of the strategy input vector to the strategy output control quantity is used to realize the twin behavior reproduction, so as to form a computable, verifiable and traceable digital twin modeling and governance mechanism. The specific effects are as follows: In the aspect of unified modeling across business domains, the present application organizes the electrical topology and equipment parameters, the measurement point table and real-time measurement, and the energy management and control strategy configuration with the unified asset main line, clearly defines the input measurement point set, the electrical object set and the strategy parameter set, and avoids the cross-domain fragmentation caused by the inconsistent asset identification, the non-unique measurement point binding and the unclear strategy reference object in the traditional engineering.

[0040] In the aspect of state synchronization and availability, the present application generates the measurement point available value through the measurement point data quality mark and the last valid observation value, and constructs the system state vector according to the fixed measurement point order, so as to ensure that the digital twin model still has a usable state input interface under the condition of missing and abnormal measurement points.

[0041] In the aspect of controllable behavior reproducible and verifiable, the present application gives a deterministic mapping rule for the strategy output control quantity, and combines the device parameter constraint and the cross-business domain consistency check, so as to ensure that the control output is uniquely determined under the same input condition and does not violate the operation constraint, thereby improving the reproducibility of operation analysis and strategy execution.

[0042] In the aspect of expansion and reconstruction traceability, the present application generates the electrical topology and equipment model business domain difference, the measurement point state business domain difference and the energy management and control strategy business domain difference in the expansion and reconstruction stage respectively, applies the difference in a deterministic order to form a new version of the model package and re-executes the consistency check, so that the reconstruction change can be recorded by difference, the version is traceable, and the model after reconstruction is still effective.

[0043] In summary, the present application realizes the consistent expression, continuous synchronization, deterministic calculation, cross-domain verification and version traceability of the micro-grid digital twin model in the three stages and three business domains without relying on additional manual repeated alignment, and has clear engineering application value.

[0044] The above description is merely that of the preferred embodiments of the present disclosure and a description of the technical principles of the present disclosure. It should be understood by those skilled in the art that the inventive scope of the embodiments of the present disclosure is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by the combinations of the above technical features or equivalent features without departing from the above inventive concept. For example, the technical solutions formed by the mutual replacement of the above features and the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) form the technical solutions.

Claims

1. A microgrid digital twin modeling method covering the entire lifecycle and multiple service domains, characterized in that, include: Step 1: Import electrical equipment ledger data and electrical connection relationship data, and perform field mapping and validation on the electrical objects contained therein to obtain multiple electrical objects that pass the validation; the electrical objects are primary equipment or electrical units in the microgrid; Step 2: Based on the electrical equipment ledger data and electrical connection relationship data, construct the electrical topology and equipment model business domain. Measurement point status business domain Energy Management and Control Strategy Business Domain ; Step 3: Business domain of electrical topology and equipment model Measurement point status business domain Energy Management and Control Strategy Business Domain The consistency check was performed, and the results were obtained. ,when Time, characterization , , The verification passed, when When, then the characterization verification , , If it fails, output the defects recorded during the verification process; Step 4: Integrate electrical topology with equipment model business domain Measurement point status business domain Energy Management and Control Strategy Business Domain Consistency verification conclusions Encapsulated as a digital twin model package for the operation control phase If and only if hour, Marked as effective when hour, Marked as ineffective; Step 5: During the operation and maintenance phase, construct an operation and maintenance event set, and then perform operations based on the operation and maintenance event set. Verification is performed to generate a digital twin model package for the operation and maintenance phase. ; Step 6: During the expansion and renovation phase, for And to carry out version evolution, and to generate new version model packages. Perform the verification according to step 3. At that time, Marked as effective and included in the digital twin model package for the capacity expansion and upgrade phase. ;when At that time, Marked as ineffective.

2. The microgrid digital twin modeling method with full lifecycle and multiple service domains as described in claim 1, characterized in that, Step 1 specifically includes: Step 1.1: Import electrical equipment ledger data and electrical connection relationship data and map them to preset standard fields to obtain a standardized data structure; Step 1.2: Set up a combination of basic fields and key fields, wherein the combination of key fields includes one or more key fields, and the key fields represent the electrical connection relationships of electrical objects; Step 1.3: For each electrical object, perform field validation on the normalized data structure based on the combination of basic fields and key fields. If a basic field is missing or the combination of key fields cannot uniquely identify the electrical object or its electrical connection relationship, record it as a defect and output a defect list. If no basic field is missing and the combination of key fields can uniquely identify the electrical object and its electrical connection relationship, the validation is considered successful, and multiple valid electrical objects are obtained.

3. The microgrid digital twin modeling method with full lifecycle and multiple service domains as described in claim 1, characterized in that, Step 2 involves constructing the electrical topology and equipment model business domain. This can be achieved through the following steps: Step A1: Map the verified electrical objects to nodes to obtain a node set. , For the number of nodes in the electrical object, and for each node Establish unique asset identifier ; Step A2: Generate a set of connection edges between electrical objects based on the electrical connection relationship data. , This represents the number of connected edges; Step A3: For each node Extract nodes from electrical equipment ledger data based on unique asset identifiers. The corresponding electrical object's equipment parameter items yield the equipment parameter set. ,in, Represents a node The Each device parameter item; set standard device parameter items; when a device parameter item is missing from the standard device parameter items in the device parameter set, the missing device parameter item is treated as a missing parameter item; when the parameter value of a device parameter item exceeds the preset allowable range, physical reasonable range, or standard rated range of the electrical object, the device parameter item is treated as an out-of-bounds item; the missing parameter item is treated as a parameter defect list, or the out-of-bounds item is treated as a parameter defect list, or the missing parameter item and the out-of-bounds item are combined to form a parameter defect list; Step A4: Construct the electrical topology graph model based on the set of nodes and the set of connecting edges. The electrical topology model and the unique asset identifier of the node set. Equipment parameter set Business domain constituting electrical topology and equipment model It also outputs a list of parameter defects.

4. The microgrid digital twin modeling method with full lifecycle and multiple service domains as described in claim 3, characterized in that, Step 2 Measurement point status service domain Build it using the following steps: Step B1: Obtain the measurement point table and generate a measurement point set S based on the measurement point table. Each measurement point in the measurement point set S includes a measurement point identifier, unit, sampling period, value range, and target identifier. The target identifier includes the electrical object identifier to which the measurement point belongs, or the circuit identifier, or both the electrical object identifier and the circuit identifier to which the measurement point belongs. Step B2: Determine the measuring points based on the target markings. Corresponding unique asset identifier ; Specifically, if the target identifier includes the identifier of the electrical object to which the measuring point belongs, then a unique asset identifier is matched based on the identifier of the electrical object to which the measuring point belongs. When the target identifier only contains a loop identifier, multiple candidate electrical objects belonging to the same loop identifier are retrieved from the electrical connection relationship data based on the loop identifier. Further determination is made by combining the measurement point name, measurement point type, endpoint information in the electrical connection relationship data, or equipment type in the electrical equipment ledger data to determine the unique asset identifier bound to it. ; The endpoint information is used to characterize the connection position of the candidate electrical object in the connection relationship; the device type is used to characterize the category to which the candidate electrical object belongs; Step B3: For each measuring point At every moment Extract observations ; Step B4: Based on the observed values Generate quality tags ; Specifically, in the observed values When the unit consistency is satisfied, the value range constraint is satisfied, and the timestamp validity is satisfied, ,otherwise ; Among them, the observed values Unit consistency represents the observation values The units are consistent with the units corresponding to the measuring points, and the observed values Satisfying the range of values ​​indicates that the observed values Within the range of values ​​corresponding to the measuring point, the observed values The timestamp validity indicates the observation value The timestamp information is complete and corresponds to the sampling period of the measurement point; Step B5: Based on the observed values and quality marking Generate usable values ​​for measurement points Specifically, when season ;when season ; Step B6: Construct the state vector according to the fixed measurement point sequence. Where Q represents the total number of measuring points, Indicates the measuring point Available values ​​of the measuring points; the set of measuring points S, measuring points Corresponding unique asset identifier State vector Constituting the measurement point status service domain .

5. The microgrid digital twin modeling method with full lifecycle and multiple service domains as described in claim 4, characterized in that, The energy management and control strategy business domain described in step 2 Build it using the following steps: Step C1: Obtain strategy configuration data from the pre-configured strategy description information, which includes at least the strategy name, strategy type, input measurement point definition, active electrical object definition, strategy parameters, and control output rules. The strategy configuration data is parsed, standardized, and structured to obtain a set of strategy parameters. A strategy set is composed of multiple strategies. ; Step C2: Analyze each policy to obtain multiple input measurement points and their order in the policy, thereby generating an ordered sequence of input measurement points. ; The ordered input measurement point sequence The input measurement points belong to the measurement point set, and are in an ordered input measurement point sequence. If the input measurement point does not belong to the measurement point set, it is recorded as a strategy input defect. Step C3: Parse each strategy to obtain the electrical object affected by the strategy, and map it to a set of unique asset identifiers. , obtain Corresponding set of equipment parameters ; in, The unique asset identifier of the electrical object in the middle is consistent with the unique asset identifier of its corresponding node. When the unique asset identifier of a policy-driven electrical object is inconsistent with the unique asset identifier of its corresponding node, it is recorded as a defect in the policy-driven electrical object. Step C4: Input the measurement point sequence in an ordered manner In the state vector Extracting from the data yields the policy input vector. ; Step C5: According to , and Generate strategy output control quantity , is represented as: ； in, For the first The deterministic control mapping function corresponding to each strategy; The , , , , and The business domain of energy management and control strategy .

6. The microgrid digital twin modeling method with full lifecycle and multiple service domains according to claim 5, characterized in that, Step 3 specifically includes: Step 3.1: Verify whether the unique asset identifiers of the node set are unique and complete, and verify whether the unique asset identifiers bound to all test points can be uniquely located within the unique asset identifiers of the node set. Can the unique asset identifier of all electrical objects in the node set be uniquely located among the unique asset identifiers of the node set? If the unique asset identifiers of the node set are unique and complete, and the unique asset identifiers bound to all measurement points can be uniquely located within the unique asset identifiers of the node set, and If the unique asset identifier of all electrical objects can be uniquely located within the unique asset identifier of the node set, then the asset identifier is consistent. Set to 1; otherwise, ensure asset identification consistency. The value is 0, and any content that fails the verification is recorded as a defect; Step 3.2: Verify that each measurement point has one and only one uniquely bound asset identifier, and that the node of the electrical object corresponding to the unique asset identifier belongs to the node set; If each measurement point has one and only one uniquely bound asset identifier, and the node of the electrical object corresponding to that unique asset identifier belongs to the node set, then the measurement point binding is valid. Set the value to 1; otherwise, verify the validity of the measurement point binding. The value is 0, and any content that fails the verification is recorded as a defect; Step 3.3: Verify the ordered input measurement point sequence for each strategy Is the set of measurement points a subset of the set of measurement points? If the ordered input measurement point sequence for each strategy is... If is a subset of the set of measurement points, then let the policy input be valid. Set the value to 1; otherwise, ensure the policy input is valid. The value is 0, and any content that fails the verification is recorded as a defect; Step 3.4: Verify each strategy Whether the unique asset identifier of the electrical object in the middle is consistent with the unique asset identifier of its corresponding node, if each strategy If the unique asset identifier of the electrical object being acted upon matches the unique asset identifier of its corresponding node, then the object's legitimacy is confirmed. Set the value to 1; otherwise, determine the validity of the target object. The value is 0, and any content that fails the verification is recorded as a defect; Step 3.5: Verify the policy input vector for each policy. Whether to follow the ordered input measurement point sequence In the state vector The system is constructed and the policy output control quantity is verified. Can it be based on , and The calculated and verified output control quantity of the strategy is obtained. Whether to pass Constraint checks as a source of constraints; Policy input vector for each policy It is based on an ordered input sequence of measurement points. In the state vector The structure is constructed within the framework, and the policy outputs control variables. Able to be based on , and The calculated value is the output control quantity of the strategy. By Constraint checks, as a source of constraints, enable policy computability. Set it to 1, otherwise make the strategy computable. The value is 0, and any content that fails the verification is recorded as a defect; Step 3.6: Calculate the consistency check conclusion ;when Time, characterization , , The verification passed, when When, then the characterization verification , , If it fails, output the defective record.

7. The microgrid digital twin modeling method with full lifecycle and multiple service domains as described in claim 6, characterized in that, Step 5 specifically includes: During the operation and maintenance (O&M) phase, an O&M event set is constructed. For each O&M event in the O&M event set, the device identifier of the O&M event is mapped to a unique asset identifier. The unique asset identifier is verified to ensure it can be uniquely located within the node set. If so, it indicates that the O&M event is associated with a node in the node set. The O&M event, event timestamp, unique asset identifier of the O&M event, and the handling actions and results contained in the O&M event are written into the digital twin model package of the O&M phase. ; If not, it indicates that the operation and maintenance event is not associated with the nodes in the node set, and the abnormal operation and maintenance event is recorded.

8. The microgrid digital twin modeling method with full lifecycle and multiple service domains according to claim 7, characterized in that, Step 6 specifically includes: Will As the first Version model package Obtain the electrical equipment ledger data, electrical connection relationship data, measurement point table, and strategy configuration data after the expansion and renovation, and then generate the first... Version model package ; right and Perform item-by-item comparisons to generate business domain differences between electrical topology and equipment model. Measurement point status service domain differential Differentiated from the energy management and control strategy business domain This will generate a new version of the model package. ,in, For version update operators; For the new version of the model package Perform the verification as per step 3. At that time, Marked as effective and included in the digital twin model package for the capacity expansion and upgrade phase. At the same time, retain , and , , ;when At that time, Marked as ineffective.

Images