A stereoscopic panoramic monitoring system and method for substation based on digital twin

Abstract

The present invention belongs to the field of transformer monitoring in substations, and in particular, relates to a stereoscopic panoramic monitoring system and method for substations based on digital twins. The system first collects power interaction data inside the substation and status information of the transformer through a data preprocessing unit to form a transformer knowledge graph node database; secondly, a virtual three-dimensional model of each transformer in the substation is constructed according to the data stored in the transformer knowledge graph node database through a three-dimensional modeling unit, and the voltage rise and fall process between the transformers is dynamically adjusted through an interactive control unit; thirdly, a fault state evaluation score is obtained according to the transformer data stored in the transformer knowledge graph node database through a virtual monitoring module, and a state adjustment factor sequence is generated accordingly; finally, the state adjustment factor sequence is fed back to the interactive control unit through a feedback decision module, the working state of the transformer is adjusted, and feedback maintenance is performed on the fault condition.

Description

Technical Field

[0001] The present invention belongs to the field of substation monitoring, and in particular relates to a substation stereoscopic panoramic monitoring system and method based on digital twins. Background Art

[0002] In recent years, with the development of information technology, especially the rise of digital twin technology, more and more research has begun to focus on how to use digital twin technology to build a virtual model of the substation to achieve real-time monitoring and intelligent analysis of the substation operation status. Digital twin technology can simulate the operation status of the substation in a computer environment by creating a virtual model that matches the physical substation, thereby providing strong support for fault diagnosis, preventive maintenance and other aspects.

[0003] For example, although the Chinese patent with authorization announcement number CN113852204B has built a three-dimensional panoramic monitoring system for substations through digital twin technology and realized visual classification of data and abnormal warning, when processing large-scale data and complex fault modes, the correlation efficiency and accuracy of its heterogeneous principle partitioning model and hash function may be limited, and the judgment criteria for abnormal warning are relatively single, which may lead to false alarms or missed alarms; and although the Chinese patent with authorization announcement number CN104852473B uses the IFC information model to achieve precise positioning of faulty equipment, it relies on the accurate input of equipment specification parameters and the accuracy of the fault analysis algorithm, and has weak recognition capabilities for unknown or new fault modes, and the model update and maintenance costs are high.

[0004] The above existing technologies all have the following problems: most of the existing technologies only perform fault monitoring and positioning at the corresponding positions of the substation, but there are obvious deficiencies in the intelligent dynamic adjustment process of the working status of the substation transformer after a fault occurs, resulting in the inability to respond in time even if a problem is discovered. For this reason, the present invention provides a three-dimensional panoramic monitoring system and method for substations based on digital twins. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention proposes a three-dimensional panoramic monitoring system and method for substations based on digital twins. The system first collects the internal power interaction data of the substation and the status information of the transformer through a data preprocessing unit to form a transformer knowledge graph node database; secondly, a virtual three-dimensional model of each transformer in the substation is constructed according to the data stored in the transformer knowledge graph node database through a three-dimensional modeling unit, and the voltage rise-drop process between the transformers is dynamically adjusted through an interactive control unit; thirdly, a virtual monitoring module obtains a fault status assessment score based on the transformer data stored in the transformer knowledge graph node database using a weighted gray cloud assessment model, and generates a state adjustment factor sequence accordingly; finally, the state adjustment factor sequence is fed back to the interactive control unit through a feedback decision module, the working state of the transformer is adjusted, and the fault condition is fed back for maintenance, to ensure that the substation can maintain the best operating state at any time.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A three-dimensional panoramic monitoring system for substations based on digital twins, comprising: a data processing module, a digital twin module, a virtual monitoring module and a feedback decision module;

[0008] The data processing module includes a knowledge database unit; the digital twin module includes a three-dimensional modeling unit and an interactive control unit; the interactive control unit is used to dynamically adjust the voltage increase-decrease process state between the step-up transformer and the step-down transformer based on the power interaction data and status information between all transformers stored in the transformer knowledge graph node database built in the knowledge database unit according to the virtual three-dimensional model of the transformer constructed by the three-dimensional modeling unit and the knowledge database unit The enhanced adjustment model is configured;

[0009] The virtual monitoring module includes a state evaluation unit and an evaluation and discrimination unit; the feedback decision module includes a feedback adjustment unit; the state evaluation unit is used to obtain the fault state evaluation scores of the step-up transformer and the step-down transformer using the state information of the step-up transformer and the step-down transformer stored in the knowledge database unit, and at the same time obtain the corresponding life use state of each transformer using the configured life prediction model, and obtain the state adjustment factor sequence corresponding to the step-up transformer and the step-down transformer using the fault state evaluation scores and life use state of the step-up transformer and the step-down transformer;

[0010] According to the fault state evaluation scores of the step-up transformer and the step-down transformer, the fault states corresponding to the step-up transformer and the step-down transformer are determined by the evaluation and discrimination unit, and the state adjustment factor of the transformer determined to be faulty is adjusted to 0, so as to obtain the state adjustment factor sequence corresponding to the step-up transformer and the step-down transformer after adjustment;

[0011] The state adjustment factor sequence corresponding to the adjusted step-up transformer and step-down transformer is fed back to the enhanced adjustment model through the feedback adjustment unit, and the voltage rise-fall process corresponding to all fault-free transformers is adjusted in real time, and maintenance feedback is provided to the faulty transformers.

[0012] Specifically, the steps of constructing the transformer knowledge graph node database include:

[0013] A1. According to the status information and location information corresponding to the transformer, a virtual three-dimensional model of each transformer in the substation is obtained through a three-dimensional modeling unit, and the virtual three-dimensional model of the transformer is used as a first-level data node in the transformer knowledge graph node database, and the corresponding transformer number and transformer type are marked on each node;

[0014] A2. Configure a data classification model, input the obtained transformer status information into the data classification model for classification, and obtain data corresponding to different types of faults of each transformer;

[0015] A3. According to the data corresponding to different types of faults of each transformer, a corresponding number of secondary sub-nodes are mounted under the corresponding primary data node, and the corresponding type of fault data and data collection timestamp are stored under the secondary sub-node corresponding to each primary data node, and the corresponding data type label is marked on each secondary sub-node;

[0016] A4. According to the power interaction data between all transformers, the interaction relationship data between all transformers is obtained through the entity recognition model, and the transformer power interaction data and the numbers of the transformers with interaction relationships are used to build the relationship connection between all first-level data nodes, and all first-level data nodes are connected.

[0017] Specifically, the steps of obtaining the fault status assessment scores of the step-up transformer and the step-down transformer include:

[0018] B1. Set the evaluation level indicator set and the corresponding scoring interval and first-level evaluation factor pair , and the different transformer fault types stored in the transformer knowledge graph node database are used as secondary evaluation factors , represents the g-th fault type, is the first-level evaluation factor of the operating status of the i-th step-up transformer, is the first-level evaluation factor of the j-th step-down transformer operating status;

[0019] B2. Construct the third-level evaluation factor data corresponding to each second-level evaluation factor according to the corresponding type of fault data saved in each second-level sub-node, and input the constructed second-level evaluation factor, third-level evaluation factor data and evaluation level indicator set into the cloud correlation function in the weighted gray cloud evaluation model to obtain the cloud correlation between the third-level evaluation factor data corresponding to each second-level evaluation factor and each evaluation level in the evaluation level indicator set. and ,in It represents the hth third-level evaluation factor data and the kth evaluation grade index in the gth second-level evaluation factor corresponding to the i-th step-up transformer, Indicates the hth third-level evaluation factor data and the kth evaluation grade index within the gth second-level evaluation factor corresponding to the jth step-down transformer;

[0020] B3. According to and , calculate the cluster cloud correlation corresponding to the i-th step-up transformer The correlation degree of the cluster cloud corresponding to the j-th step-down transformer .

[0021] Specifically, the step of obtaining the fault status assessment scores of the step-up transformer and the step-down transformer also includes:

[0022] B4. According to and , the evaluation score of the i-th step-up transformer corresponding to the g-th fault type is calculated by the weighted formula and the evaluation score of the jth step-down transformer corresponding to the gth fault type ;

[0023] B5. Obtain and set the initial values ​​of the three-level evaluation factor data corresponding to the step-up transformer and the step-down transformer , warning value , and based on the three-level evaluation factors obtained in real time , calculate and obtain the quantitative value of the three-level evaluation factor index corresponding to each fault type;

[0024] B6. According to the quantitative value of the three-level evaluation factor index corresponding to each fault type, the variable weighted value of the corresponding fault type is calculated, and the variable weighted value of each fault type and the corresponding and , calculate the fault status assessment score of each step-up transformer and step-down transformer in turn and .

[0025] Specifically, the step of obtaining the state adjustment factor sequence corresponding to the adjusted step-up transformer and the step-down transformer includes:

[0026] C1. Set the transformer fault status assessment score threshold , the fault status assessment scores of the step-up transformer and step-down transformer and and For comparison, Greater than or Greater than When , it is determined that the corresponding i-th step-up transformer or j-th step-down transformer is in a fault stop state. Less than or equal to or Less than or equal to When , it is determined that the corresponding i-th step-up transformer or j-th step-down transformer is in a normal working state;

[0027] C2. Obtain the life usage status data of the transformer in normal working condition, and input the obtained data into the configured life prediction model to obtain the remaining service life score of each transformer and ,in represents the remaining service life score corresponding to the i-th step-up transformer in the transformer in normal working condition, represents the remaining service life score corresponding to the jth step-down transformer among the transformers in normal working condition;

[0028] C3, according to the transformer in normal working condition or and the corresponding or , calculate the adjustment factor of the step-up transformer in normal working condition and step-down transformer adjustment factor ,in: ,in represents the adjustment factor corresponding to the i-th step-up transformer in normal working condition, The same logic applies.

[0029] Specifically, the step of obtaining the state adjustment factor sequence corresponding to the adjusted step-up transformer and the step-down transformer further includes:

[0030] C4. Set the adjustment factor of the transformer in the fault stop state to 0, and obtain the state adjustment factor sequence corresponding to the step-up transformer and the state adjustment factor sequence corresponding to the step-down transformer after adjustment according to the adjustment factor and the corresponding transformer number of the transformer in the fault stop state and the adjustment factor and the corresponding transformer number of the transformer in the normal working state.

[0031] Specifically, the steps of strengthening the adjustment model construction include:

[0032] D1. Based on the transformer knowledge graph node database, the transformer operation status information and the set target demand data are updated in real time to build the enhanced adjustment model input status at the current moment ,in, Indicates the real-time boost voltage corresponding to the current time t, Indicates the boost voltage target value corresponding to the current boost transformer. Indicates the step-down voltage target value corresponding to the current step-down transformer. represents the input data sequence of the operating status of the i-th step-up transformer, Represents the input data sequence of the j-th step-down transformer operating status.

[0033] Specifically, the steps of strengthening the adjustment model construction also include:

[0034] D2. According to the current input status, set the current action trigger information, including:

[0035] When there is Greater than or Greater than When the first execution action information is triggered, that is, the weighted ash cloud assessment model and the life prediction model are called to obtain the state adjustment factor sequence corresponding to all current transformers, and the state adjustment factor sequence corresponding to the transformer in normal working state in the state adjustment factor sequence is used to assign the step-up or step-down task corresponding to the transformer in the fault stop state to the transformer in normal working state, and at the same time, according to the number corresponding to the adjustment factor of the transformer in the fault stop state, the corresponding transformer fault maintenance feedback is performed;

[0036] When all and Are less than or equal to , but the corresponding boost value is less than Or the corresponding pressure drop value is less than , trigger the second execution action information; that is, use the calculated state adjustment factor sequence corresponding to the step-up transformer or the state adjustment factor sequence corresponding to the step-down transformer to adjust the step-up or step-down task voltage values ​​corresponding to all step-up transformers or step-down transformers, so that the final voltage corresponding to the step-up process and the final voltage corresponding to the step-down process are equal to the corresponding and .

[0037] A method for stereoscopic panoramic monitoring of a substation based on digital twins, comprising the following steps:

[0038] S1. Obtain and preprocess the internal power interaction data of the substation and the location and status information of each transformer, input the preprocessed data into the data classification model and knowledge graph algorithm for classification and graph node construction, and obtain the transformer knowledge graph node database;

[0039] S2. Using the transformer location and status information stored in the transformer knowledge graph node database, a three-dimensional generation algorithm is used to construct a virtual three-dimensional model of the transformer, and according to the status information of all virtual three-dimensional models of transformers and power interaction data, the voltage rise-drop process state between transformers is dynamically adjusted through the configured enhanced adjustment model;

[0040] S3, using the saved status information of the step-up transformer and the step-down transformer, by configuring the weighted ash cloud assessment model, obtain the fault status assessment scores of the step-up transformer and the step-down transformer, and use the configured life prediction model to obtain the corresponding life use status of each transformer;

[0041] S4. Set an evaluation score threshold, identify the transformer fault state according to the corresponding transformer fault state evaluation score and the evaluation score threshold, and use the evaluation score and life usage status of the identified transformer to construct the corresponding transformer state adjustment factor, and feed the corresponding transformer state adjustment factor back to the enhanced adjustment model, make real-time adjustments to the voltage rise and fall processes corresponding to all fault-free transformers, and provide maintenance feedback for faulty transformers.

[0042] A computer-readable storage medium stores computer instructions, which, when executed, execute a stereoscopic panoramic monitoring method for a substation based on digital twins.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] In view of the deficiencies in the prior art, the present invention, through the synergy of the transformer knowledge graph node database and the enhanced adjustment model, can dynamically adjust the voltage state between the step-up and step-down transformers in real time to ensure stable and efficient power transmission; at the same time, the state evaluation and life prediction functions of the virtual monitoring module, through intelligent identification of fault states and immediate adjustment of state adjustment factors, effectively avoid the expansion of faults and imbalance in power grid supply, greatly reduce the downtime caused by faults, and ensure the reliability of power grid operation and the intelligent management efficiency of substations. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a module diagram of a three-dimensional panoramic monitoring system for a substation based on digital twins according to Embodiment 1 of the present invention;

[0046] Figure 2This is a flow chart of a method for stereoscopic panoramic monitoring of substations based on digital twins according to Example 2 of the present invention. DETAILED DESCRIPTION

[0047] Example 1

[0048] See also Figure 1 , an embodiment provided by the present invention: a stereoscopic panoramic monitoring system for a substation based on digital twin, comprising: a data processing module, a digital twin module, a virtual monitoring module and a feedback decision module;

[0049] The data processing module is used for collecting and preprocessing the internal power interaction data of the substation and the location and status information of each transformer; the data processing module includes a data preprocessing unit and a knowledge database unit; further, the transformer in this embodiment includes a step-up transformer and a step-down transformer;

[0050] The data preprocessing unit is used to collect the power interaction data of each transformer in the substation and the location and status information of each transformer, and preprocess the collected data;

[0051] The knowledge database unit classifies the preprocessed data and constructs graph nodes through the configured data classification model and knowledge graph algorithm to obtain a transformer knowledge graph node database.

[0052] Furthermore, the steps of constructing the transformer knowledge graph node database in this embodiment include:

[0053] A1. According to the status information and location information corresponding to the transformer, the three-dimensional modeling unit is used to obtain the transformer virtual three-dimensional model corresponding to each transformer in the substation, and the transformer virtual three-dimensional model is used as the first-level data node in the transformer knowledge graph node database, and the corresponding transformer number and transformer type are marked on each node;

[0054] A2. Configure a data classification model, input the obtained transformer status information into the data classification model for classification, and obtain data corresponding to different types of faults of each transformer;

[0055] A3. According to the data corresponding to different types of faults of each transformer, a corresponding number of secondary sub-nodes are mounted under the corresponding primary data node, and the corresponding type of fault data and data collection timestamp are stored under the secondary sub-node corresponding to each primary data node, and the corresponding data type label is marked on each secondary sub-node;

[0056] A4. According to the power interaction data between all transformers, the interaction relationship data between all transformers is obtained through the entity recognition model, and the transformer power interaction data and the numbers of the transformers with interaction relationships are used to build the relationship connection between all first-level data nodes. All first-level data nodes are connected to obtain the transformer knowledge graph node database.

[0057] In this process, the data preprocessing unit is responsible for collecting and preprocessing the power interaction data and location and status information of each transformer in the substation to ensure the accuracy and availability of the data; then, the knowledge database unit classifies the preprocessed data and constructs the transformer knowledge graph node database through the configured data classification model and knowledge graph algorithm; in this process, firstly, according to the status and location information of the transformer, the three-dimensional modeling unit is used to generate a virtual three-dimensional model of each transformer, and it is used as the first-level data node in the transformer knowledge graph node database, marking the number and type of the transformer; then, the transformer status information is classified through the data classification model to distinguish different types of fault data, and the corresponding second-level sub-nodes are created under the first-level data node to store the fault data and timestamp information, and mark the data type label; finally, the entity recognition model is used to analyze the power interaction relationship between transformers, build the connection between the first-level data nodes, and form a complete transformer knowledge graph node database; this structured approach not only makes data management more orderly, but also intuitively presents the mutual influence relationship between transformers in the form of a graph, which is convenient for quickly locating the source of the fault and predicting potential problems. In addition, the construction of the transformer knowledge graph node database provides a solid foundation for subsequent fault diagnosis, status assessment, and decision support, thereby improving the system's response speed and operation and maintenance efficiency, and reducing downtime and economic losses caused by faults.

[0058] The digital twin module is used to construct a three-dimensional virtual model of the substation, and to embed a corresponding transformer device in the corresponding position of the three-dimensional virtual model and simulate the working state of the transformer; the digital twin module includes a three-dimensional modeling unit and an interactive control unit;

[0059] The three-dimensional modeling unit is used to construct a virtual three-dimensional model of each transformer in the substation by using the location and status information of each transformer stored in the transformer knowledge graph node database through a three-dimensional generation algorithm, and mark the corresponding number and location information on the corresponding transformer virtual three-dimensional model;

[0060] The interactive control unit is used to dynamically adjust the voltage increase-decrease process between transformers through the configured enhanced adjustment model according to the constructed transformer virtual three-dimensional model and the power interaction data and status information between all transformers stored in the transformer knowledge graph node database;

[0061] The virtual monitoring module is used to monitor the operating status of each transformer, and obtain the transformer operating status adjustment factor through the operating status of each transformer, and adjust the transformer voltage increase-reduction working interaction process in real time, so that each transformer is in the optimal interactive state in real time; the virtual monitoring module includes a state evaluation unit and an evaluation and judgment unit;

[0062] The state evaluation unit is used to obtain each step-up transformer fault state evaluation score and each step-down transformer fault state evaluation score by using the stored state information of each step-up transformer and each step-down transformer through the configured weighted ash cloud evaluation model, and simultaneously obtain the corresponding life use state of each transformer by using the configured life prediction model, and construct a step-up transformer state adjustment factor sequence and a step-down transformer state adjustment factor sequence by using the acquired step-up transformer fault state evaluation score and step-down transformer fault state evaluation score and the corresponding transformer life use state;

[0063] Furthermore, in this embodiment, the life prediction model is constructed by a random forest algorithm or a BiLstm algorithm to calculate the remaining service life score corresponding to the corresponding transformer.

[0064] The evaluation and discrimination unit is used to obtain the fault status of the step-up transformer and the step-down transformer according to the step-up transformer fault status evaluation score and the step-down transformer fault status evaluation score by using the configured evaluation score threshold, and adjust the transformer state adjustment factor determined to be faulty in the step-up transformer state adjustment factor sequence and the step-down transformer state adjustment factor sequence to 0;

[0065] Furthermore, in this embodiment, the steps of obtaining the fault status assessment scores of the step-up transformer and the step-down transformer include:

[0066] B1. Set the evaluation level indicator set and the corresponding scoring interval and first-level evaluation factor pair , and the different transformer fault types stored in the transformer knowledge graph node database are used as secondary evaluation factors , represents the g-th fault type, is the first-level evaluation factor of the operating status of the i-th step-up transformer, is the first-level evaluation factor of the j-th step-down transformer operating status;

[0067] Furthermore, in this embodiment, the secondary evaluation factors correspond one-to-one to the fault types;

[0068] Furthermore, in this embodiment, the evaluation level indicator set and the corresponding scoring interval are specifically set by those skilled in the art according to the evaluation requirements and accuracy;

[0069] Furthermore, in this embodiment, the fault type is specifically: ,in, Indicates the core fault in the fault type. Indicates the current circuit overheating in the fault type, Indicates the winding fault in the fault type. Indicates partial discharge in fault type, Indicates the oil flow discharge in the fault type, Indicates arc discharge in fault type, Indicates arc discharge insulation aging among fault types.

[0070] B2. Construct the third-level evaluation factor data corresponding to each second-level evaluation factor according to the corresponding type of fault data saved in each second-level sub-node, and input the constructed second-level evaluation factor, third-level evaluation factor data and evaluation level indicator set into the cloud correlation function in the weighted gray cloud evaluation model to obtain the cloud correlation between the third-level evaluation factor data corresponding to each second-level evaluation factor and each evaluation level in the evaluation level indicator set. and ,in It represents the hth third-level evaluation factor data and the kth evaluation grade index in the gth second-level evaluation factor corresponding to the i-th step-up transformer, Indicates the hth third-level evaluation factor data and the kth evaluation grade index within the gth second-level evaluation factor corresponding to the jth step-down transformer;

[0071] Furthermore, the three-level evaluation factors corresponding to the core fault in this embodiment include: no-load loss: measuring the energy consumption of the transformer when it is no-load, temperature change: monitoring whether the core temperature rises abnormally, and vibration level: detecting whether the core vibration exceeds the normal range; the three-level evaluation factors corresponding to the current loop overheating include: contact resistance: measuring the contact resistance of the connection part, temperature distribution: detecting the temperature distribution of each point in the current loop, and current density: analyzing whether the current density in the current loop exceeds the standard; the three-level evaluation factors corresponding to the winding fault include: DC resistance: measuring the DC resistance value of the winding, interlayer short circuit: detecting the interlayer short circuit of the winding Electrical continuity, inter-turn short circuit: check whether there is a short circuit between winding turns; the three-level evaluation factors corresponding to partial discharge include: discharge amount: measure the amount of charge generated by the discharge activity, discharge frequency: count the number of discharges per unit time; the three-level evaluation factors corresponding to oil flow discharge include oil flow velocity: monitor the speed of oil flow inside the transformer; the three-level evaluation factors corresponding to arc discharge include: discharge duration: record the duration of each discharge, number of discharges: count the number of discharges within a certain period of time; the three-level evaluation factors corresponding to arc discharge insulation aging include insulation paper polymerization degree: test the aging degree of insulation paper.

[0072] B3. According to and , calculate the cluster cloud correlation corresponding to the i-th step-up transformer The correlation degree of the cluster cloud corresponding to the j-th step-down transformer ;

[0073] Specifically, , where H represents the number of third-level evaluation factor data corresponding to the g-th second-level evaluation factor, represents the weight of the hth tertiary evaluation factor within the gth secondary evaluation factor; Obtained through the same calculation process;

[0074] B4. According to and , the evaluation score of the i-th step-up transformer corresponding to the g-th fault type is calculated by the weighted formula and the evaluation score of the jth step-down transformer corresponding to the gth fault type ;

[0075] Furthermore, the evaluation score calculation process of the fault type in this embodiment is specifically as follows: , where K represents the number of evaluation indicators in the evaluation level indicator set, It represents the evaluation score of the g-th fault type for the k-th evaluation level indicator; Acquired through the same process;

[0076] B5. Obtain and set the initial values ​​of the three-level evaluation factor data corresponding to the step-up transformer and the step-down transformer , warning value , and based on the three-level evaluation factors obtained in real time , calculate and obtain the quantitative value of the three-level evaluation factor index corresponding to each fault type;

[0077] Furthermore, in this embodiment, the quantitative value calculation process of the three-level evaluation factor index corresponding to each fault type is as follows:

[0078] ,

[0079] in, It represents the quantitative value of the three-level evaluation factor index corresponding to the g-th fault type;

[0080] B6. According to the quantitative value of the three-level evaluation factor index corresponding to each fault type, the variable weighted value of the corresponding fault type is calculated, and the variable weighted value of each fault type and the corresponding and , calculate the fault status assessment score of each step-up transformer and step-down transformer in turn and .

[0081] Furthermore, in this embodiment, the formula for calculating the fault status evaluation score of each step-up transformer and step-down transformer is:

[0082] ;

[0083] in, represents the fault status assessment score of the i-th step-up transformer, represents the quantitative value of the three-level evaluation factor index corresponding to the g-th fault type on the i-th step-up transformer, G The number of types of all fault types refers to the number of corresponding fault types of the transformer. For example, in this embodiment Corresponding to the 7 types of faults represented, G here is 7; represents the variable weight value corresponding to the g-th fault type on the i-th step-up transformer; is the probability weight of the g-th fault type corresponding to the i-th step-up transformer, specifically: ;in, The historical frequency of the g-th fault type occurring on the i-th step-up transformer is the ratio of the number of times the g-th fault type has occurred in history on the i-th step-up transformer to the total number of times all the above fault types have occurred in history on the i-th step-up transformer, for example: The corresponding fault type has occurred 3 times in the history of the i-th step-up transformer, and the i-th step-up transformer has a total of 20 times in history corresponding to all fault types, then the corresponding ; represents the balancing function. In this embodiment, when the parameter When it is less than 0.5, it represents a serious defect where failure cannot be tolerated; ∈[0.5,1), it means that the requirement for balance is low; when the parameter =1, it is equivalent to the fixed weight mode; similarly The calculation process and same.

[0084] Furthermore, in this embodiment, the step of obtaining the state adjustment factor sequence corresponding to the adjusted step-up transformer and the step-down transformer includes:

[0085] C1. Set the transformer fault status assessment score threshold , the fault status assessment scores of the step-up transformer and step-down transformer and and For comparison, Greater than or Greater than When , it is determined that the corresponding i-th step-up transformer or j-th step-down transformer is in a fault stop state. Less than or equal to or Less than or equal to When , it is determined that the corresponding i-th step-up transformer or j-th step-down transformer is in a normal working state;

[0086] C2. Obtain the life usage status data of the transformer in normal working condition, and input the obtained data into the configured life prediction model to obtain the remaining service life score of each transformer and ,in represents the remaining service life score corresponding to the i-th step-up transformer in the transformer in normal working condition, It represents the remaining service life score corresponding to the jth step-down transformer in the transformer in normal working condition;

[0087] C3, according to the transformer in normal working condition or and the corresponding or , calculate the adjustment factor of the step-up transformer in normal working condition and step-down transformer adjustment factor , specifically: ,in represents the adjustment factor corresponding to the i-th step-up transformer in normal working condition, Similarly, ;

[0088] C4. Set the adjustment factor of the transformer in the fault stop state to 0, and obtain the state adjustment factor sequence corresponding to the step-up transformer and the state adjustment factor sequence corresponding to the step-down transformer after adjustment according to the adjustment factor and the corresponding transformer number of the transformer in the fault stop state and the adjustment factor and the corresponding transformer number of the transformer in the normal working state.

[0089] Furthermore, in this embodiment, the state adjustment factor sequence includes a transformer number and an adjustment factor corresponding to the transformer;

[0090] This process conducts a multi-dimensional analysis of the health status of the transformer by introducing three-level evaluation factors, such as no-load loss, temperature change, and vibration level in core faults. At the same time, the evaluation score of the fault type is obtained through clustering cloud correlation calculation and weighted formula, and then combined with the transformer life prediction model to generate a state adjustment factor sequence. This can not only detect potential faults in a timely manner, but also dynamically optimize the voltage rise-fall process between transformers by adjusting the factor sequence, ensure that the fault-free transformer is in the optimal operating state, and provide timely maintenance feedback for the faulty transformer; this process not only improves the intelligent management level of the substation, reduces the downtime caused by faults, but also enhances the stability and reliability of the system, reduces operation and maintenance costs, and improves the overall operation efficiency of the substation.

[0091] The feedback decision module is used to collect the power interaction data inside the substation and the status information of each transformer in real time, and adjust the working state of the transformer in real time according to the constructed step-up transformer state adjustment factor sequence and step-down transformer state adjustment factor sequence, and perform feedback maintenance on the faulty transformer; the feedback decision module includes a real-time data analysis unit, a feedback adjustment unit and a feedback maintenance unit;

[0092] The real-time data analysis unit is used to obtain the state information data of the step-up transformer and the step-down transformer in real time, and obtain the real-time state adjustment factor sequence of the step-up transformer and the real-time state adjustment factor sequence of the step-down transformer through the state evaluation unit and the evaluation and judgment unit;

[0093] The feedback adjustment unit is used to feed back the real-time state adjustment factor of the step-up transformer and the real-time state adjustment factor of the step-down transformer to the enhanced adjustment model, to make real-time adjustments to the voltage increase-decrease process corresponding to all fault-free transformers, and to feed back the fault state information of the faulty transformer to the feedback maintenance unit, so that the transformer maintains the optimal state in real time during the step-up and step-down process;

[0094] The feedback maintenance unit is used to generate a corresponding maintenance plan according to the transformer fault state information fed back by the feedback adjustment unit, through the configured reasoning information library and the fed back fault information, to assist in the maintenance of the faulty transformer.

[0095] Furthermore, the reasoning information database configured in the present embodiment is constructed through entity relationship extraction algorithm and graph database through historical transformer fault records and transformer maintenance specification manual, and a pre-trained deep reasoning model is configured in the reasoning information database to perform deep reasoning based on the input transformer fault status information; further, the pre-trained deep reasoning model in the present embodiment is an existing model obtained after fine-tuning and training through the transformer fault status information, fault records and transformer maintenance specification manual in the present embodiment.

[0096] Furthermore, the steps of strengthening the adjustment model construction in this embodiment include:

[0097] D1. Based on the transformer knowledge graph node database, the transformer operation status information and the set target demand data are updated in real time to build the enhanced adjustment model input status at the current moment ,in, Indicates the real-time boost voltage corresponding to the current time t, Indicates the boost voltage target value corresponding to the current boost transformer. Indicates the step-down voltage target value corresponding to the current step-down transformer. represents the input data sequence of the operating status of the i-th step-up transformer, represents the input data sequence of the jth step-down transformer operating status;

[0098] D2. According to the current input status, set the current action trigger information, including:

[0099] When there is Greater than or Greater than When the first execution action information is triggered, that is, the weighted ash cloud assessment model and the life prediction model are called to obtain the state adjustment factor sequence corresponding to all current transformers, and the state adjustment factor sequence corresponding to the transformer in normal working state in the state adjustment factor sequence is used to assign the step-up or step-down task corresponding to the transformer in the fault stop state to the transformer in normal working state, and at the same time, according to the number corresponding to the adjustment factor of the transformer in the fault stop state, the corresponding transformer fault maintenance feedback is performed;

[0100] When all and Are less than or equal to , but the corresponding boost value is less than Or the corresponding pressure drop value is less than , trigger the second execution action information; that is, use the calculated state adjustment factor sequence corresponding to the step-up transformer or the state adjustment factor sequence corresponding to the step-down transformer to adjust the step-up or step-down task voltage values ​​corresponding to all step-up transformers or step-down transformers, so that the final voltage corresponding to the step-up process and the final voltage corresponding to the step-down process are equal to the corresponding and ;

[0101] D3. Build the corresponding execution action based on the current execution action trigger information ,in, Indicates the action that triggers the first execution action information to be executed. Indicates the action executed by triggering the second execution action information;

[0102] D4. Construct an enhanced adjustment model based on the SAC algorithm, and input the current input state, the current execution action trigger information and the corresponding execution action into the enhanced adjustment model for training, obtain the trained enhanced adjustment model, and configure the trained enhanced adjustment model to the interactive control unit to perform the corresponding step-up and step-down control between transformers.

[0103] This process achieves intelligent dynamic adjustment of the substation transformer status through the combination of three-dimensional modeling and enhanced adjustment model; specifically, the three-dimensional modeling unit accurately constructs a virtual model of the transformer, providing an intuitive and accurate basis for interactive control; the interactive control unit dynamically adjusts the voltage rise and fall process through the enhanced adjustment model based on the real-time updated transformer operation data and target requirements, ensuring the stability and efficiency of power transmission; in particular, the intelligent triggering of execution actions based on the transformer status not only effectively allocates the normal transformer to take on the task of the faulty transformer, but also provides timely maintenance feedback on the faulty transformer to avoid the expansion of the fault.

[0104] Example 2

[0105] See also Figure 2 Another embodiment provided by the present invention is a method for stereoscopic panoramic monitoring of a substation based on digital twins, comprising the following steps:

[0106] S1. Obtain and preprocess the internal power interaction data of the substation and the location and status information of each transformer, input the preprocessed data into the data classification model and knowledge graph algorithm for classification and graph node construction, and obtain the transformer knowledge graph node database;

[0107] Furthermore, the transformer in this embodiment includes a step-up transformer and a step-down transformer;

[0108] S2. Using the transformer location and status information stored in the transformer knowledge graph node database, a three-dimensional generation algorithm is used to construct a virtual three-dimensional model of the transformer, and according to the status information of all virtual three-dimensional models of transformers and power interaction data, the voltage rise-drop process state between transformers is dynamically adjusted through the configured enhanced adjustment model;

[0109] S3, using the saved status information of the step-up transformer and the step-down transformer, by configuring the weighted ash cloud assessment model, obtain the fault status assessment scores of the step-up transformer and the step-down transformer, and use the configured life prediction model to obtain the corresponding life use status of each transformer;

[0110] S4. Set an evaluation score threshold, identify the transformer fault state according to the corresponding transformer fault state evaluation score and the evaluation score threshold, and use the transformer fault state evaluation score and life usage status after identification to construct the corresponding transformer state adjustment factor, and feed the corresponding transformer state adjustment factor back to the enhanced adjustment model, make real-time adjustments to the voltage rise and fall processes corresponding to all fault-free transformers, and provide maintenance feedback and generate maintenance plans for faulty transformers.

[0111] Example 3

[0112] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, a stereoscopic panoramic monitoring method for a substation based on digital twins is implemented.

[0113] A computer-readable storage medium stores computer instructions, which, when executed, execute a stereoscopic panoramic monitoring method for a substation based on digital twins.

[0114] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in the field may also change, modify, replace and modify the above-mentioned embodiments without departing from the scope of protection of the purpose of the present invention and the claims, and all of these are within the protection of the present invention.

Claims

1. A three-dimensional panoramic monitoring system for substations based on digital twins, characterized in that: include: Data processing module, digital twin module, virtual monitoring module and feedback decision module; The data processing module includes a knowledge database unit; the digital twin module includes a three-dimensional modeling unit and an interactive control unit; the interactive control unit is used to dynamically adjust the voltage increase-decrease process state between the step-up transformer and the step-down transformer according to the power interaction data and status information between all transformers stored in the transformer knowledge graph node database built in the knowledge database unit based on the virtual three-dimensional model of the transformer constructed by the three-dimensional modeling unit and the transformer knowledge graph node database built in the knowledge database unit, through the configured enhanced adjustment model; The virtual monitoring module includes a state evaluation unit and an evaluation and discrimination unit; the feedback decision module includes a feedback adjustment unit; the state evaluation unit is used to obtain the fault state evaluation scores of the step-up transformer and the step-down transformer using the state information of the step-up transformer and the step-down transformer stored in the knowledge database unit, and at the same time obtain the corresponding life use state of each transformer using the configured life prediction model, and obtain the state adjustment factor sequence corresponding to the step-up transformer and the step-down transformer using the fault state evaluation scores and life use state of the step-up transformer and the step-down transformer; According to the fault state evaluation scores of the step-up transformer and the step-down transformer, the fault states corresponding to the step-up transformer and the step-down transformer are determined by the evaluation and discrimination unit, and the state adjustment factors of the transformers determined to be faulty are adjusted to 0, so as to obtain the state adjustment factor sequences corresponding to the step-up transformer and the step-down transformer after adjustment; The state adjustment factor sequence corresponding to the adjusted step-up transformer and step-down transformer is fed back to the enhanced adjustment model through the feedback adjustment unit, and the voltage rise-fall process corresponding to all fault-free transformers is adjusted in real time, and maintenance feedback is provided to the faulty transformers.

2. A three-dimensional panoramic monitoring system for substations based on digital twins as claimed in claim 1, characterized in that: The steps of constructing the transformer knowledge graph node database include: A1. According to the status information and location information corresponding to the transformer, the three-dimensional modeling unit is used to obtain the transformer virtual three-dimensional model corresponding to each transformer in the substation, and the transformer virtual three-dimensional model is used as the first-level data node in the transformer knowledge graph node database, and the corresponding transformer number and transformer type are marked on each node; A2. Configure a data classification model, input the obtained transformer status information into the data classification model for classification, and obtain data corresponding to different types of faults of each transformer; A3. According to the data corresponding to different types of faults of each transformer, a corresponding number of secondary sub-nodes are mounted under the corresponding primary data node, and the corresponding type of fault data and data collection timestamp are stored under the secondary sub-node corresponding to each primary data node, and the corresponding data type label is marked on each secondary sub-node; A4. According to the power interaction data between all transformers, the interaction relationship data between all transformers is obtained through the entity recognition model, and the transformer power interaction data and the numbers of the transformers with interaction relationships are used to build the relationship connection between all first-level data nodes, and all first-level data nodes are connected.

3. A three-dimensional panoramic monitoring system for substations based on digital twins as claimed in claim 2, characterized in that: The step of obtaining the fault status assessment scores of the step-up transformer and the step-down transformer comprises: B1. Set the evaluation level indicator set and the corresponding scoring interval and first-level evaluation factor pair , and the different transformer fault types stored in the transformer knowledge graph node database are used as secondary evaluation factors , represents the g-th fault type, is the first-level evaluation factor of the operating status of the i-th step-up transformer, is the first-level evaluation factor of the j-th step-down transformer operating status; B2. Construct the third-level evaluation factor data corresponding to each second-level evaluation factor according to the corresponding type of fault data saved in each second-level sub-node, and input the constructed second-level evaluation factor, third-level evaluation factor data and evaluation level indicator set into the cloud correlation function in the weighted gray cloud evaluation model to obtain the cloud correlation between the third-level evaluation factor data corresponding to each second-level evaluation factor and each evaluation level in the evaluation level indicator set. and ,in It represents the hth third-level evaluation factor data and the kth evaluation grade index in the gth second-level evaluation factor corresponding to the i-th step-up transformer, Indicates the hth third-level evaluation factor data and the kth evaluation grade index within the gth second-level evaluation factor corresponding to the jth step-down transformer; B3. According to and , calculate the cluster cloud correlation corresponding to the i-th step-up transformer The correlation degree of the cluster cloud corresponding to the j-th step-down transformer .

4. A three-dimensional panoramic monitoring system for substations based on digital twins as claimed in claim 3, characterized in that: The step of obtaining the fault status assessment scores of the step-up transformer and the step-down transformer also includes: B4. According to and , the evaluation score of the i-th step-up transformer corresponding to the g-th fault type is calculated by the weighted formula and the evaluation score of the jth step-down transformer corresponding to the gth fault type ; B5. Obtain and set the initial values ​​of the three-level evaluation factor data corresponding to the step-up transformer and the step-down transformer , warning value , and based on the three-level evaluation factors obtained in real time , calculate and obtain the quantitative value of the three-level evaluation factor index corresponding to each fault type; B6. According to the quantitative value of the three-level evaluation factor index corresponding to each fault type, the variable weighted value of the corresponding fault type is calculated, and the variable weighted value of each fault type and the corresponding and , calculate the fault status assessment score of each step-up transformer and step-down transformer in turn and .

5. A three-dimensional panoramic monitoring system for substations based on digital twins as claimed in claim 4, characterized in that: The step of obtaining the state adjustment factor sequence corresponding to the adjusted step-up transformer and the step-down transformer comprises: C1. Set the transformer fault status assessment score threshold , the fault status assessment scores of the step-up transformer and step-down transformer and and For comparison, Greater than or Greater than When , it is determined that the corresponding i-th step-up transformer or j-th step-down transformer is in a fault stop state. Less than or equal to or Less than or equal to When , it is determined that the corresponding i-th step-up transformer or j-th step-down transformer is in a normal working state; C2. Obtain the life usage status data of the transformer in normal working condition, and input the obtained data into the configured life prediction model to obtain the remaining service life score of each transformer and ,in represents the remaining service life score corresponding to the i-th step-up transformer in the transformer in normal working condition, represents the remaining service life score corresponding to the jth step-down transformer among the transformers in normal working condition; C3, according to the transformer in normal working condition or and the corresponding or , calculate the adjustment factor of the step-up transformer in normal working condition and step-down transformer adjustment factor ,in: ,in represents the adjustment factor corresponding to the i-th step-up transformer in normal working condition, The same logic applies.

6. A three-dimensional panoramic monitoring system for substations based on digital twins as claimed in claim 5, characterized in that: The step of obtaining the state adjustment factor sequence corresponding to the adjusted step-up transformer and the step-down transformer also includes: C4. Set the adjustment factor of the transformer in the fault stop state to 0, and obtain the state adjustment factor sequence corresponding to the step-up transformer and the state adjustment factor sequence corresponding to the step-down transformer after adjustment according to the adjustment factor and the corresponding transformer number of the transformer in the fault stop state and the adjustment factor and the corresponding transformer number of the transformer in the normal working state.

7. A three-dimensional panoramic monitoring system for substations based on digital twins as claimed in claim 6, characterized in that: The steps of building the enhanced adjustment model include: D1. Update the transformer operation status information and the set target demand data in real time according to the transformer knowledge graph node database to build the current input status of the enhanced adjustment model ,in, Indicates the real-time boost voltage corresponding to the current time t, Indicates the boost voltage target value corresponding to the current boost transformer. Indicates the step-down voltage target value corresponding to the current step-down transformer. represents the input data sequence of the operating status of the i-th step-up transformer, Represents the input data sequence of the j-th step-down transformer operating status.

8. A three-dimensional panoramic monitoring system for substations based on digital twins as claimed in claim 7, characterized in that: The step of strengthening the adjustment model construction also includes: D2. According to the current input status, set the current action trigger information, including: When there is Greater than or Greater than When the first execution action information is triggered, that is, the weighted ash cloud assessment model and the life prediction model are called to obtain the state adjustment factor sequence corresponding to all current transformers, and the state adjustment factor sequence corresponding to the transformer in normal working state in the state adjustment factor sequence is used to assign the step-up or step-down task corresponding to the transformer in the fault stop state to the transformer in normal working state, and at the same time, according to the number corresponding to the adjustment factor of the transformer in the fault stop state, the corresponding transformer fault maintenance feedback is performed; When all and Are less than or equal to , but the corresponding boost value is less than Or the corresponding pressure drop value is less than When the second execution action information is triggered; that is, the state adjustment factor sequence corresponding to the step-up transformer or the state adjustment factor sequence corresponding to the step-down transformer is used to adjust the step-up or step-down task voltage values ​​corresponding to all step-up transformers or step-down transformers, so that the final voltage corresponding to the step-up process and the final voltage corresponding to the step-down process are equal to the corresponding and .

9. A method for stereoscopic panoramic monitoring of a substation based on digital twins, which is implemented based on a stereoscopic panoramic monitoring system for a substation based on digital twins as claimed in any one of claims 1 to 7, characterized in that the steps include: S1. Obtain and preprocess the internal power interaction data of the substation and the location and status information of each transformer, input the preprocessed data into the data classification model and knowledge graph algorithm for classification and graph node construction, and obtain the transformer knowledge graph node database; S2. Using the transformer location and status information stored in the transformer knowledge graph node database, a three-dimensional generation algorithm is used to construct a virtual three-dimensional model of the transformer, and according to the status information of all virtual three-dimensional models of transformers and power interaction data, the voltage rise-drop process state between transformers is dynamically adjusted through the configured enhanced adjustment model; S3, using the saved status information of the step-up transformer and the step-down transformer, by configuring the weighted ash cloud assessment model, obtain the fault status assessment scores of the step-up transformer and the step-down transformer, and use the configured life prediction model to obtain the corresponding life use status of each transformer; S4. Set an evaluation score threshold, identify the transformer fault state according to the corresponding transformer fault state evaluation score and the evaluation score threshold, and use the evaluation score and life usage status of the identified transformer to construct the corresponding transformer state adjustment factor, and feed the corresponding transformer state adjustment factor back to the enhanced adjustment model, make real-time adjustments to the voltage rise and fall processes corresponding to all fault-free transformers, and provide maintenance feedback for faulty transformers.

10. A computer-readable storage medium, characterized in that: Computer instructions are stored thereon, and when the computer instructions are executed, a stereoscopic panoramic monitoring method of a substation based on digital twin as described in claim 9 is executed.

Images