Transformer substation early warning method and system based on digital twinborn model

Abstract

The invention discloses a transformer substation early warning method and system based on a digital twinborn model, and relates to the technical field of data analysis, and the method comprises the steps: dividing the range of a target transformer substation into a plurality of power regions, dividing the power regions in consideration of the correlation condition between power equipment, and providing a reliable basis for the adaptation of subsequent spatial resolution. The spatial resolution is determined based on the basic information of the power equipment in each power area and the associated information between the power equipment, and the digital twinborn model of the target substation is constructed, so that the adaptability of the spatial resolution is improved, the change and capability of the substation can be better shown and described, and a more perfect digital twinborn model is established. The abnormal condition of the target substation is monitored through the template of the abnormal event, the abnormal condition is integrated and analyzed, the adaptability and accuracy of substation early warning are improved, normal and safe operation of the substation is ensured, and timely early warning can be performed before the abnormal event occurs.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular, to a substation early warning method and system based on a digital twin model. Background Art

[0002] With the increasing complexity and intelligent requirements of the power system, the substation, as a key node in power transmission, its safe and stable operation is crucial. Digital twin technology realizes all-round, dynamic tracking and simulation prediction of physical entities by constructing a virtual model of the substation. This early warning solution relies on advanced technologies such as the Internet of Things, big data, and cloud computing to collect the operation data of substation equipment in real time and map it to the virtual model. By using big data analysis technology to mine data patterns and combining machine learning algorithms to predict equipment failures and abnormal situations, real-time monitoring and early warning of the substation operation status are achieved. The digital twin model can accurately reflect the actual operation status of the substation, provide scientific decision-making support for operation and maintenance personnel, and effectively improve the safety and reliability of the substation.

[0003] In the prior art, the spatial resolution of the digital twin model of the substation is the same for different regions of the substation, and the association between all power equipment under the substation is not considered, resulting in poor description and capabilities of the digital twin model for the substation, low adaptability and accuracy of substation early warning, and inability to ensure the normal and safe operation of the substation.

[0004] Therefore, how to improve the adaptability and accuracy of substation early warning is a technical problem to be solved at present. Summary of the Invention

[0005] The purpose of the present invention is to solve the problem of low adaptability and accuracy of substation early warning in the prior art, and to propose a substation early warning method based on a digital twin model, which includes:

[0006] Collect the basic information of all power equipment under the target substation and the association information between power equipment, and divide the scope where the target substation is located into multiple power regions;

[0007] Determine the spatial resolution based on the basic information of the power equipment in each power region and the association information between power equipment, and construct a digital twin model of the target substation according to the spatial resolution;

[0008] Establish templates for all abnormal events under the target substation, and monitor the abnormal situations of the target substation based on the digital twin model of the target substation through the templates of abnormal events;

[0009] Integrate and analyze the abnormal situations, and output the abnormal information of the target substation.

[0010] In some embodiments of the present application, the range where the target substation is located is divided into multiple power regions, including,

[0011] Collect the locations of all power equipment under the target substation. The association information between power equipment includes the connectivity relationship and the cooperation relationship between power equipment.

[0012] Taking the locations of power equipment as points, and using the connectivity relationship and the cooperation relationship between power equipment to connect different points and form edges, quantifying the connectivity relationship and the cooperation relationship between power equipment to confirm the lengths of the edges, thus constituting a power equipment association graph.

[0013] Perform the shortest path algorithm and the connectivity algorithm on the power equipment association graph. According to the shortest path algorithm and the connectivity algorithm, divide the range where the target substation is located into multiple power regions, and determine the boundaries of each power region and the power equipment corresponding to each power region.

[0014] In some embodiments of the present application, determine the spatial resolution based on the basic information of the power equipment in each power region and the association information between power equipment, including,

[0015] Establish a matching relationship between the power region and the power equipment according to the power equipment corresponding to each power region.

[0016] The basic information of power equipment includes operation information and safety information. Collect the operation information and safety information of each power equipment, analyze the operation information to determine the representative range of each operation parameter, and analyze the safety information to determine the safety risk of each power equipment.

[0017] Evaluate the capture level of the power region based on the matching relationship between the power region and the power equipment, the representative range of the operation parameters of the power equipment, and the safety risk, and determine the spatial resolution of the power region according to the capture level.

[0018] In some embodiments of the present application, analyze the operation information to determine the representative range of each operation parameter, including,

[0019] Statistical all ranges and characteristic values of each operation parameter. The characteristic values include mean, standard deviation, skewness, and kurtosis.

[0020] Determine the interval length according to the length of all ranges of each operation parameter. Segment all ranges of the operation parameter based on the interval length, count the frequency of the operation parameter appearing in each range segment, draw a frequency distribution diagram of the operation parameter, and determine the first representative range of the operation parameter according to the appearance frequency.

[0021] Determine the symmetry degree of the operation parameter by combining skewness and kurtosis, determine the multiple according to the symmetry degree, and determine the second representative range of the operation parameter based on the mean, standard deviation, and multiple.

[0022] Use the intersection of the first representative range and the second representative range of the operating parameters as the representative range of each operating parameter.

[0023] In some embodiments of the present application, analyzing the security information to determine the security risk of each power device includes,

[0024] The security information includes the past fault records and maintenance records of the power device. Identify all occurred fault modes from the fault records and maintenance records, and determine the unoccurred fault modes according to the type of power device;

[0025] Analyze the trends of faults and maintenance on the fault records and maintenance records, determine the impact degree of the occurred fault modes and predict the impact degree of the unoccurred fault modes. Based on the occurred fault modes, unoccurred fault modes, impact degree of the occurred fault modes, impact degree of the unoccurred fault modes, and the trends of faults and maintenance of the occurred fault modes, determine the risk priority number of each fault mode of the power device, so as to determine the security risk of each power device.

[0026] In some embodiments of the present application, constructing a digital twin model of the target substation according to the spatial resolution includes,

[0027] Obtain the spatial layout data of the target substation, and use 3D modeling software and simulation platform to create the basic framework of the digital twin model;

[0028] Create a virtual model of each power device on the basic framework of the digital twin model, map the location of the power device into the digital twin model, and adjust the corresponding power device model according to the spatial resolution of different power regions. Establish the electrical connection relationship and control logic between devices in the virtual model, and perform logical simulation on the digital twin model to simulate the actual operation process of the substation.

[0029] In some embodiments of the present application, establishing templates for all abnormal events under the target substation includes,

[0030] Determine the matching relationship between abnormal events and fault modes under the target substation. The fault modes include occurred fault modes and unoccurred fault modes. Extract the abnormal characteristics of the occurred fault modes from the fault records and maintenance records, so as to construct the templates of abnormal events corresponding to the occurred fault modes;

[0031] Determine the abnormal characteristics of the unoccurred fault modes, so as to construct the templates of abnormal events corresponding to the unoccurred fault modes, and establish a template library of abnormal events for all fault modes.

[0032] In some embodiments of the present application, monitoring the abnormal conditions of the target substation through the templates of abnormal events includes,

[0033] Collect and extract real-time features according to the digital twin model, calculate the matching degree and evolution probability between the real-time features and the template library of abnormal events of the fault mode, so as to output the abnormal situation of the target substation.

[0034] Correspondingly, the present application also provides a substation warning system based on the digital twin model, including,

[0035] The first module is used to collect the basic information of all power equipment under the target substation and the association information between power equipment, and divide the scope where the target substation is located into multiple power areas;

[0036] The second module is used to determine the spatial resolution based on the basic information of the power equipment in each power area and the association information between power equipment, and construct a digital twin model of the target substation according to the spatial resolution;

[0037] The third module is used to establish templates for all abnormal events under the target substation, and monitor the abnormal situation of the target substation through the templates of abnormal events on the basis of the digital twin model of the target substation;

[0038] The fourth module is used to integrate and analyze the abnormal situation and output the abnormal information of the target substation.

[0039] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0040] 1. Divide the scope where the target substation is located into multiple power areas, consider the association between power equipment and divide the power areas, providing a reliable basis for the subsequent adaptation of spatial resolution. Determine the spatial resolution based on the basic information of the power equipment in each power area and the association information between power equipment, and construct a digital twin model of the target substation, so as to improve the adaptability of the spatial resolution, better describe the changes and capabilities of the substation, and establish a more perfect digital twin model.

[0041] 2. Monitor the abnormal situation of the target substation through the templates of abnormal events, integrate and analyze the abnormal situation, improve the adaptability and accuracy of substation warning, ensure the normal and safe operation of the substation, and be able to give timely warning before the occurrence of abnormal events. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic flow chart of a substation warning method based on a digital twin model proposed by the present invention;

[0043] Figure 2 It is a schematic structural diagram of a substation warning system based on a digital twin model proposed by the present invention. Specific Embodiments

[0044] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0045] Refer to Figure 1 , a substation early warning method based on a digital twin model, including the following steps:

[0046] Step S101, collect the basic information of all power equipment under the target substation and the association information between power equipment, and divide the scope where the target substation is located into multiple power areas.

[0047] In this embodiment, the basic information of power equipment includes operation information, safety information, equipment list, equipment location, etc. Equipment list: Obtain the list of all power equipment in the target substation, including transformers, circuit breakers, disconnectors, instrument transformers, lightning arresters, etc. Equipment operation parameters: Record the basic parameters of each equipment, such as model, specification, rated capacity, rated voltage, rated current, etc. Equipment location: Determine the specific installation location of each equipment in the substation, usually represented in the form of coordinates. The association information includes electrical connection relationships, control logics, and communication protocols, etc. Electrical connection relationship: Draw the electrical main wiring diagram of the substation to clarify the electrical connection relationships between equipment. Control logic: Understand the control logic between equipment, such as protection action sequence, interlock relationship, etc. Communication protocol: Record the protocols used for communication between equipment, such as IEC 61850, Modbus, etc. Data cleaning: Remove duplicate, incorrect, or invalid data. Data standardization: Unify the data format and units for subsequent processing. Data association: Associate the basic information of equipment with the association information to form a complete equipment data set.

[0048] In some embodiments of the present application, the scope where the target substation is located is divided into multiple power areas, including,

[0049] Collect the locations of all power equipment under the target substation. The association information between power equipment includes the connectivity relationship and the cooperation relationship between power equipment;

[0050] Use the locations of power equipment as points, and use the connectivity relationship and the cooperation relationship between power equipment to connect different points and form edges. Quantify the connectivity relationship and the cooperation relationship between power equipment to confirm the length of the edges, and constitute a power equipment association graph;

[0051] Perform the shortest path algorithm and connectivity algorithm on the power equipment association diagram, divide the range where the target substation is located into multiple power regions according to the shortest path algorithm and connectivity algorithm, and determine the boundaries of each power region and the power equipment corresponding to each power region.

[0052] In this embodiment, the connectivity relationship between power equipment is an electrical connection relationship, and the physical connection relationship between power equipment is recorded, such as cable connection, bus connection, etc. The collaborative relationship between power equipment is analyzed to analyze the collaborative effect of power equipment in terms of function, such as the relationship between the protection device and the protected equipment, the collaborative relationship between each device in the automation control system, etc. The position of each power equipment is defined as a node in the diagram, and the node contains the basic information of the equipment (such as equipment type, number, position coordinates, etc.). Connecting edge: According to the connectivity relationship of power equipment, draw an edge between the corresponding nodes. The length of the edge can be quantified as a physical distance (such as cable length) or a logical distance (such as connection complexity). Collaborative edge: According to the collaborative relationship of power equipment, draw an edge between the corresponding nodes. The length of the edge can be quantified as the collaborative intensity (such as the degree of functional dependence, signal transmission frequency, etc.), which usually needs to be determined through expert evaluation or data analysis. Use graph theory tools or programming languages (such as the NetworkX library in Python) to construct the power equipment association diagram and integrate the nodes and edges into the diagram. Regard the power equipment as the nodes in the diagram, and the electrical connection or functional dependence between the equipment as the edges in the diagram. By constructing the equipment association diagram, the association relationship between the equipment can be intuitively displayed. Using graph theory algorithms (such as the shortest path algorithm, connectivity algorithm, etc.) can further analyze the degree of association and path between the equipment.

[0053] In this embodiment, apply the shortest path algorithm such as Dijkstra algorithm or Floyd-Warshall algorithm to calculate the shortest path between any two points in the diagram. This helps to identify the key paths and potential bottlenecks in the substation. Use the connectivity algorithm such as depth-first search (DFS) or breadth-first search (BFS) to determine the connected components in the diagram. Each connected component represents a potential power region. Use the connectivity algorithm to evaluate the connectivity between different devices to ensure good connectivity between the devices within the divided power regions. Combine the position information and functional association information of the devices, and use graph theory algorithms to determine the boundaries of the power regions. The devices on the boundary usually have a weak connectivity or collaborative relationship with the devices in other regions. Through the shortest path algorithm, the shortest connection paths between different regions can be found, and these paths can be used as a reference for the region boundary.

[0054] Step S102, determine the spatial resolution based on the basic information of the power equipment in each power region and the association information between the power equipment, and construct a digital twin model of the target substation according to the spatial resolution.

[0055] In this embodiment, considering the operation information and safety information of each power device, as well as the association between power devices, jointly determine the spatial resolution of each power area to ensure the adaptability of the digital twin model description. By constructing a digital twin model according to the spatial resolution of each power area, the actual layout and equipment distribution of the substation can be more accurately reflected. Different power areas are set with different spatial resolutions according to factors such as equipment density and importance, making the model more refined in complex areas and more concise in simple areas, which not only ensures the accuracy of the model but also improves the calculation efficiency. The model contains the basic information and association information of power devices, enabling the model to simulate the actual operation of the substation and providing an intuitive monitoring and management interface for operation and maintenance personnel. Operation and maintenance personnel can quickly understand the operation status, association relationship, and potential risks of the equipment through the model, so as to make more scientific decisions. Constructing the model according to the spatial resolution can reasonably allocate computing resources and avoid wasting computing power in unimportant areas. High resolution is adopted in key areas to capture more detailed information; low resolution is adopted in non-critical areas to reduce the computational complexity and improve the model operation efficiency.

[0056] In some embodiments of the present application, the spatial resolution is determined based on the basic information of the power devices in each power area and the association information between the power devices, including,

[0057] Establish a matching relationship between the power area and the power device according to the power devices corresponding to each power area;

[0058] The basic information of the power device includes operation information and safety information. Collect the operation information and safety information of each power device, analyze the operation information to determine the representative range of each operation parameter, and analyze the safety information to determine the safety risk of each power device;

[0059] Evaluate the capture level of the power area based on the matching relationship between the power area and the power device, the representative range of the operation parameters of the power device, and the safety risk, and determine the spatial resolution of the power area according to the capture level.

[0060] In this embodiment, the operating parameters of power equipment are collected in real time through the SCADA system, sensors or other monitoring devices, such as voltage, current, temperature, load rate, etc. The operating parameters vary greatly with the substation working conditions and time. It is necessary to determine the representative range of each operating parameter (the range that can represent the relatively common existence of the operating parameter). The safety risk of power equipment is the comprehensive risk situation based on the previous faults and possible faults of the power equipment. According to the representative range of the operating parameters, safety risks and the association of power equipment in the power area, the capture level of the power area is defined. The capture level reflects the degree of demand for monitoring and management in this area. For each power area, considering the fluctuation of the operating parameters of the power equipment therein, the safety risk level and the association relationship between the equipment, the capture level of this area is evaluated. For example, if a power area contains multiple high-risk equipment and the operating parameters fluctuate greatly, the capture level of this area should be higher.

[0061] In this embodiment, the spatial resolution is an index indicating the degree of detail of the power area represented in the digital twin model. High resolution means that the model can more finely reflect the actual situation of this area. According to the capture level of the power area, the spatial resolution of this area is determined. The higher the capture level, the higher the spatial resolution should be to more accurately monitor and manage the power equipment in this area.

[0062] In some embodiments of this application, analyzing the operation information to determine the representative range of each operating parameter includes,

[0063] Statistical all ranges and characteristic values of each operating parameter, and the characteristic values include mean, standard deviation, skewness and kurtosis;

[0064] Determine the interval length according to the length of all ranges of each operating parameter, segment all ranges of the operating parameter based on the interval length, count the frequency of the operating parameter appearing in each range segment, draw the frequency distribution diagram of the operating parameter, and determine the first representative range of the operating parameter according to the appearance frequency;

[0065] Combine skewness and kurtosis to determine the symmetry degree of the operating parameter, determine the multiple according to the symmetry degree, and determine the second representative range of the operating parameter based on the mean, standard deviation and multiple;

[0066] Take the intersection of the first representative range and the second representative range of the operating parameter as the representative range of each operating parameter.

[0067] In this embodiment, the characteristic values include the following:

[0068] Mean:

[0069] The average value of the data, which reflects the central position of the data.

[0070] Standard Deviation:

[0071] An indicator that measures the degree of data dispersion and reflects the dispersion of data points around the mean.

[0072] Skewness:

[0073] An indicator that describes the degree of asymmetry of the data distribution. Positive skewness indicates that the data is skewed to the right, and negative skewness indicates that the data is skewed to the left.

[0074] Kurtosis:

[0075] An indicator that describes the peakedness of the data distribution. A kurtosis greater than 3 indicates that the data distribution is more peaked than the normal distribution, and less than 3 indicates that the data distribution is flatter than the normal distribution.

[0076] In this embodiment, the first representative range of the operating parameters is determined according to the occurrence frequency, and some range segments with higher occurrence frequencies are screened out to form the first representative range (determined by the occurrence frequency). The symmetry degree of the operating parameters is determined by combining skewness and kurtosis (such as weighted average), and according to the statistical results, the mean value plus or minus a certain multiple of the standard deviation can be selected as the representative range. This multiple is usually determined according to the distribution characteristics of the data to obtain the second distribution range (determined by statistical characteristics).

[0077] In some embodiments of the present application, the security information is analyzed to determine the security risk of each power device, including

[0078] The security information includes the previous fault records and maintenance records of the power device. All occurred fault modes are identified from the fault records and maintenance records, and the unoccurred fault modes are determined according to the type of the power device;

[0079] The trends of faults and maintenance are analyzed from the fault records and maintenance records, the influence degree of the occurred fault modes is determined, and the influence degree of the unoccurred fault modes is predicted. Based on the occurred fault modes, unoccurred fault modes, influence degree of the occurred fault modes, influence degree of the unoccurred fault modes, and trends of faults and maintenance of the occurred fault modes, the risk priority number of each fault mode of the power device is determined, thereby determining the security risk of each power device.

[0080] In this embodiment, the past fault records of power equipment are collected, including the time, location, fault type, fault cause, fault impact range, etc. of the fault occurrence. The maintenance records of power equipment are collected, including the maintenance time, maintenance content, replaced components, equipment status after maintenance, etc. The fault records are classified and sorted to identify all occurred fault modes. For example, classification can be carried out according to the fault type (such as short circuit, open circuit, overheating, etc.) or fault cause (such as insulation aging, component damage, etc.). According to the type, structure, working principle, etc. of the power equipment, combined with industry experience and expert knowledge, possible but unoccurred fault modes are determined. For example, for a certain type of transformer, possible but unoccurred fault modes include winding deformation, oil quality deterioration, etc. Time series analysis is carried out on the fault records and maintenance records to identify the trends of faults and maintenance. For example, it can be analyzed whether the frequency of fault occurrence, the interval time of maintenance, etc. show a certain trend over time. The impact degree of the occurred fault modes on aspects such as the performance, safety, and reliability of power equipment is evaluated. Quantitative or qualitative methods can be used for evaluation, such as determining the impact degree according to the power outage time caused by the fault, the degree of equipment damage, etc. The Risk Priority Number (RPN) is an index used to evaluate the risk level of fault modes, usually considering factors such as the occurrence probability, impact degree, and detectability of fault modes. In this embodiment, we can calculate the RPN based on the occurred fault modes, unoccurred fault modes, their impact degrees, and the trends of faults and maintenance.

[0081] In this embodiment, for each fault mode, the RPN is calculated according to its occurrence probability (which can be determined through historical data or expert evaluation), impact degree (as obtained from the above analysis), and detectability (evaluated according to maintenance records and monitoring means).

[0082] Formula example: RPN = occurrence probability × impact degree × detectability (the detectability here can be converted into a coefficient reflecting the maintenance and monitoring effect).

[0083] For unoccurred fault modes, their occurrence probabilities can be estimated by analogy with similar equipment or based on expert experience.

[0084] In this embodiment, based on the matching relationship between power regions and power equipment, the representative range of operating parameters of power equipment, and the safety risk, the capture level of power regions is evaluated. Based on the operating parameter requirements of power equipment, the quality index of the representative range of operating parameters is evaluated. The specific calculation formula is as follows:

[0085]

[0086] Among them, is the capture level of the i1-th power area, β1 and β2 are the conversion coefficients of the operating parameters and safety risks respectively, and n is the number of power equipment under the i1-th power area. are the operation combination weights of the i2-th power equipment and the risk combination weights of the i3-th power equipment respectively. is the comprehensive quality index of all operating parameters of the i2-th power equipment in the i1-th power area. is the comprehensive safety risk of all failure modes of the i3-th power equipment in the i1-th power area. is the strength of the association between power equipment under the i1-th power area (such as the length of the edge). are the first constant and the second constant of the i1-th power area respectively. represents the correction of the sum of operating parameters and safety risks by the strength of the association between power equipment under the power area. The first constant and the second constant are used to balance the size of the correction function and the size of the capture level respectively.

[0087] In some embodiments of the present application, a digital twin model of the target substation is constructed according to the spatial resolution, including

[0088] Obtain the spatial layout data of the target substation, and use 3D modeling software and simulation platform to create the basic framework of the digital twin model;

[0089] Create a virtual model of each power equipment on the basic framework of the digital twin model, map the power equipment location to the digital twin model, and adjust the corresponding power equipment model according to the spatial resolution of different power areas. Establish the electrical connection relationship and control logic between the equipment in the virtual model, and perform logical simulation on the digital twin model to simulate the actual operation process of the substation.

[0090] In this embodiment, the actual spatial layout data of the substation, including buildings, roads, equipment locations, etc., is obtained through measurement, drawings, or GIS systems. Select a suitable 3D modeling software or simulation platform, such as AutoCAD, Revit, Unity3D, etc., according to requirements, for constructing the digital twin model framework of the substation. Using the selected software or platform, create the basic framework of the digital twin model according to the actual spatial layout data of the substation, including buildings, roads, fences, etc. Create virtual models of each device in the model framework according to the collected device information. For critical devices, model them as detailed as possible, including the internal structure and operating components of the devices; for non-critical devices, a simplified modeling method can be adopted. Map the device locations into the digital twin model to ensure that the model is consistent with the spatial layout of the actual substation. Scale and adjust the device models according to the spatial resolution requirements of different power regions. For example, in high-resolution regions, the device models should be more refined to reflect the actual size and details of the devices; in low-resolution regions, the device models can be appropriately simplified to reduce the computational load. Establish the electrical connection relationships and control logics between devices in the virtual model to ensure that the model can accurately reflect the actual operation of the substation. Use the simulation function of the software or platform to verify and test the association relationships between devices to ensure the correctness of the logic. Conduct logical simulation on the digital twin model to simulate the actual operation process of the substation. According to the simulation results, optimize and adjust the model to improve the accuracy and reliability of the model.

[0091] Step S103: Establish templates for all abnormal events under the target substation. Based on the digital twin model of the target substation, monitor the abnormal conditions of the target substation through the templates of abnormal events.

[0092] In some embodiments of the present application, establishing templates for all abnormal events under the target substation includes:

[0093] Determine the matching relationship between abnormal events and fault modes under the target substation. The fault modes include occurred fault modes and unoccurred fault modes. Extract the abnormal characteristics of the occurred fault modes according to the fault records and maintenance records, so as to construct templates for abnormal events corresponding to the occurred fault modes;

[0094] Determine the abnormal characteristics of the unoccurred fault modes, so as to construct templates for abnormal events corresponding to the unoccurred fault modes, and establish a template library for abnormal events of all fault modes.

[0095] In this embodiment, according to the fault records, the occurred fault modes are classified, such as short circuit, open circuit, overheat, insulation aging, etc. Combining the equipment type, operating environment and industry standards, the possible but unoccurred fault modes are determined. Analyze the abnormal events that each fault mode may cause, such as voltage abnormality, current abnormality, temperature abnormality, etc. Establish a matching relationship table between the abnormal events and the fault modes, and clarify the types of abnormal events corresponding to each fault mode. For each occurred fault mode, analyze the fault records and equipment operation data, and extract the abnormal characteristics caused by this fault mode. The abnormal characteristics may include the type of abnormal event, occurrence time, duration, influence range, changes in the operation parameters of related equipment, etc. According to the extracted abnormal characteristics, construct an abnormal event template corresponding to the occurred fault mode. The template should include the basic information of the abnormal event, feature description, possible causes and consequences, etc.

[0096] In this embodiment, for the unoccurred fault modes, combining the equipment type, operating environment and industry standards, predict the abnormal characteristics that this fault mode may cause. According to the predicted abnormal characteristics, construct an abnormal event template corresponding to the unoccurred fault mode. The template should also include the basic information of the abnormal event, feature description, possible causes and consequences, etc., but note that it is marked as a "predicted" or "potential" fault mode.

[0097] In some embodiments of the present application, monitor the abnormal conditions of the target substation through the template of the abnormal event, including

[0098] Collect and extract real-time features according to the digital twin model, calculate the matching degree and evolution probability between the real-time features and the template library of the abnormal events of the fault modes, so as to output the abnormal conditions of the target substation.

[0099] In this embodiment, the real-time features are converted into feature vectors for comparison with the features in the abnormal event template library. For each abnormal event template in the template library, the corresponding feature vectors are also extracted. Similarity algorithms (such as cosine similarity, Euclidean distance, etc.) are used to calculate the similarity between the real-time feature vector and each abnormal event template feature vector. The higher the similarity value, the more matching the real-time feature is with the abnormal feature in the template. According to the similarity calculation results, the matching degree between the real-time feature and each abnormal event template is determined. A matching degree threshold can be set. When the matching degree exceeds the threshold, it is considered that the real-time feature matches the abnormal event template. First, the matching degree is calculated. If the matching degree is high, the abnormal event with a high matching degree is output. If the matching degree is low, it indicates that there is no abnormal event currently. Then, the evolution probability is calculated, which describes the probability of an abnormal event evolving from the current normal state. Based on the historical data analysis results, an evolution probability model is established to predict the evolution probability of the fault mode corresponding to the real-time feature within a certain period in the future. The evolution probability model can consider various factors, such as fault type, device status, operating environment, etc. The real-time feature is input into the evolution probability model to calculate the evolution probability of its corresponding fault mode within a certain period in the future. The higher the evolution probability value, the more likely the fault mode will occur in the future.

[0100] Step S104, integrate and analyze the abnormal situations and output the abnormal information of the target substation.

[0101] In this embodiment, based on the calculation results of the matching degree and the evolution probability, it is judged whether there are abnormal situations in the target substation. When the matching degree exceeds the threshold and the evolution probability is high, it is considered that there are abnormal situations in the target substation. The specific information of the abnormal situation (such as abnormal type, occurrence time, related equipment, evolution probability, etc.) is output to the monitoring interface or the alarm system. At the same time, corresponding processing suggestions or plans can be provided to help the operation and maintenance personnel take measures to deal with the abnormal situation in a timely manner.

[0102] Correspondingly, the present application also provides a substation early warning system based on a digital twin model, as Figure 2 shown, including,

[0103] The first module is used to collect the basic information of all power equipment under the target substation and the association information between power equipment, and divide the scope where the target substation is located into multiple power areas;

[0104] The second module is used to determine the spatial resolution based on the basic information of the power equipment in each power area and the association information between power equipment, and construct a digital twin model of the target substation according to the spatial resolution;

[0105] The third module is used to establish templates for all abnormal events in the target substation. Based on the digital twin model of the target substation, the abnormal conditions of the target substation are monitored through the templates of abnormal events;

[0106] The fourth module is used to integrate and analyze the abnormal conditions and output the abnormal information of the target substation.

[0107] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0108] 1. The range where the target substation is located is divided into multiple power regions. The power regions are divided by considering the association between power equipment, providing a reliable basis for the subsequent adaptation of spatial resolution. Based on the basic information of the power equipment in each power region and the association information between power equipment, the spatial resolution is determined, and the digital twin model of the target substation is constructed, improving the adaptability of the spatial resolution, being able to better display and describe the changes and capabilities of the substation, and establishing a more perfect digital twin model.

[0109] 2. The abnormal conditions of the target substation are monitored through the templates of abnormal events, the abnormal conditions are integrated and analyzed, improving the adaptability and accuracy of substation early warning, ensuring the normal and safe operation of the substation, and being able to give timely early warning before the occurrence of abnormal events.

[0110] Through the description of the above embodiments, those skilled in the art can clearly understand that the present invention can be implemented through hardware or by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention can be embodied in the form of a software product. The software product can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various implementation scenarios of the present invention.

[0111] Those skilled in the art can understand that the drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the drawings are not necessarily essential for implementing the present invention.

[0112] Those skilled in the art can understand that the modules in the system in the implementation scenario can be distributed in the system of the implementation scenario according to the description of the implementation scenario, or can be correspondingly changed and located in one or more systems different from the present implementation scenario. The modules in the above implementation scenario can be combined into one module, or can be further split into multiple sub-modules.

[0113] The above are only the preferred specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention should cover within the protection scope of the present invention by making equivalent substitutions or changes according to the technical solution and inventive concept of the present invention.

Claims

1. A substation early warning method based on a digital twin model, characterized in that: include, Collect the basic information of all power equipment under the target substation and the association information between the power equipment, and divide the area where the target substation is located into multiple power areas; Determine the spatial resolution based on the basic information of the power equipment in each power area and the association information between the power equipment, and build a digital twin model of the target substation according to the spatial resolution; Establish templates for all abnormal events under the target substation. Based on the digital twin model of the target substation, monitor the abnormal conditions of the target substation through the templates of abnormal events. Integrate and analyze abnormal situations and output abnormal information of the target substation.

2. The substation early warning method based on the digital twin model according to claim 1 is characterized in that: The target substation is divided into multiple power areas, including: Collect the locations of all power equipment under the target substation, and the association information between the power equipment includes the connectivity relationship and the coordination relationship between the power equipment; Taking the location of the power equipment as a point, connecting different points and forming edges based on the connectivity relationship and the coordination relationship between the power equipment, quantifying the connectivity relationship and the coordination relationship between the power equipment to confirm the length of the edge, and forming a power equipment association graph; The shortest path algorithm and connectivity algorithm are performed on the power equipment association graph. The range of the target substation is divided into multiple power areas according to the shortest path algorithm and connectivity algorithm, and the boundary of each power area and the power equipment corresponding to each power area are determined.

3. The substation early warning method based on the digital twin model according to claim 2 is characterized in that: The spatial resolution is determined based on the basic information of the power equipment in each power area and the association information between the power equipment, including, Establish a matching relationship between power area and power equipment according to the power equipment corresponding to each power area; The basic information of power equipment includes operation information and safety information. The operation information and safety information of each power equipment are collected, the operation information is analyzed to determine the representative range of each operation parameter, and the safety information is analyzed to determine the safety risk of each power equipment; Based on the matching relationship between power area and power equipment, the representative range of the operating parameters of the power equipment and the capture level of the power area for safety risk assessment, the spatial resolution of the power area is determined according to the capture level.

4. The substation early warning method based on the digital twin model according to claim 3 is characterized in that: Analyze the operating information to determine the representative range for each operating parameter, including, Count all ranges and characteristic values ​​of each running parameter, including mean, standard deviation, skewness and kurtosis; Determine the interval length according to the length of all ranges of each operating parameter, divide all ranges of the operating parameter into segments based on the interval length, count the frequency of occurrence of the operating parameter in each range segment, draw a frequency distribution diagram of the operating parameter, and determine the first representative range of the operating parameter according to the frequency of occurrence; Determine the degree of symmetry of the operating parameter by combining the skewness and the kurtosis, determine the multiple according to the degree of symmetry, and determine the second representative range of the operating parameter based on the mean, the standard deviation and the multiple; The intersection of the first representative range and the second representative range of the operating parameter is taken as the representative range of each operating parameter.

5. The substation early warning method based on the digital twin model according to claim 3 is characterized in that: Analyze safety information to determine the safety risks of each power device, including, Safety information includes past fault records and maintenance records of the power equipment, identifying all the fault modes that have occurred in the fault records and maintenance records, and determining the failure modes that have not occurred based on the type of power equipment; Analyze the failure and maintenance trends based on the failure records and maintenance records, determine the impact of the failure modes that have occurred and predict the impact of the failure modes that have not occurred, and determine the risk priority number of each failure mode of the power equipment based on the failure modes that have occurred, the failure modes that have not occurred, the impact of the failure modes that have occurred, the impact of the failure modes that have not occurred, and the failure and maintenance trends of the failure modes that have occurred, thereby determining the safety risk of each power equipment.

6. The substation early warning method based on the digital twin model according to claim 3 is characterized in that: Build a digital twin model of the target substation based on spatial resolution, including: Obtain the spatial layout data of the target substation and use 3D modeling software and simulation platform to create the basic framework of the digital twin model; Create a virtual model of each power equipment based on the basic framework of the digital twin model, map the location of the power equipment to the digital twin model, and adjust the corresponding power equipment model according to the spatial resolution of different power areas. Establish the electrical connection relationship and control logic between the equipment in the virtual model, perform logical simulation on the digital twin model, and simulate the actual operation process of the substation.

7. The substation early warning method based on the digital twin model according to claim 5 is characterized in that: Create templates for all abnormal events in the target substation, including: Determine the matching relationship between abnormal events and fault modes under the target substation. Fault modes include fault modes that have occurred and fault modes that have not occurred. Extract the abnormal features of the fault modes that have occurred based on fault records and maintenance records, so as to construct a template for abnormal events corresponding to the fault modes that have occurred. Determine the abnormal characteristics of the failure mode that has not occurred, so as to construct the template of the abnormal event corresponding to the failure mode that has not occurred, and establish a template library of abnormal events of all failure modes.

8. The substation early warning method based on the digital twin model according to claim 7 is characterized in that: Monitor abnormal conditions of target substations through abnormal event templates, including: According to the digital twin model, real-time features are collected and extracted, and the matching degree and evolution probability of real-time features and abnormal events in the template library of fault modes are calculated to output the abnormal conditions of the target substation.

9. A substation early warning system based on a digital twin model, characterized in that: include, The first module is used to collect the basic information of all power equipment under the target substation and the association information between the power equipment, and divide the scope of the target substation into multiple power areas; The second module is used to determine the spatial resolution based on the basic information of the power equipment in each power area and the association information between the power equipment, and to construct a digital twin model of the target substation according to the spatial resolution; The third module is used to establish templates for all abnormal events under the target substation. Based on the digital twin model of the target substation, the abnormal conditions of the target substation are monitored through the templates of abnormal events; The fourth module is used to integrate and analyze abnormal situations and output abnormal information of the target substation.

Images