A power transformation asset management method and system

Abstract

The embodiment of the specification provides a power transformation asset management method and system, comprising: acquiring basic asset information of a power transformation device; the basic asset information comprises at least one of state monitoring information, device information, device maintenance information and power supply quantity of the power transformation device; determining asset management cost based on the basic asset information; updating asset management information of the power transformation device based on the basic asset information and the asset management cost; and determining a key monitoring device and issuing early warning information to an associated user terminal based on the updated asset management information of the power transformation device.

Description

Technical Field

[0001] This manual pertains to the field of power management, and in particular to methods and systems for substation asset management. Background Technology

[0002] Substations are used to transform voltage, transmit, and distribute electrical energy between power plants and users, connecting power grids of different voltage levels. The main electrical equipment in a substation includes primary equipment such as transformers and high-voltage circuit breakers that directly generate, transmit, distribute, and utilize electrical energy, as well as secondary equipment such as relay protection devices and automatic devices that measure, monitor, control, and protect the operating conditions of primary equipment and systems. Therefore, during equipment use, a method is needed to monitor the long-term costs of equipment use, the potential risks caused by changes in equipment condition, and the costs associated with these risks. Based on the monitored information, a method can be used to identify substation equipment and its sub-equipment that require focused monitoring in the future, thereby improving the efficiency and accuracy of subsequent equipment management. Summary of the Invention

[0003] One embodiment of this specification provides a substation asset management method, comprising: acquiring basic asset information of substation equipment; the basic asset information including at least one of: status monitoring information, equipment information, equipment maintenance information, and substation power supply; determining asset management costs based on the basic asset information; updating the asset management information of the substation equipment based on the basic asset information and the asset management costs; and identifying key monitoring equipment and issuing early warning information to associated user terminals based on the updated asset management information of the substation equipment.

[0004] One embodiment of this specification provides a substation asset management system, including an acquisition module, a determination module, an update module, and an early warning module. The acquisition module acquires basic asset information of substation equipment. The basic asset information includes at least one of the following: status monitoring information, equipment information, equipment maintenance information, and substation power supply. The determination module determines asset management costs based on the basic asset information. The update module updates the asset management information of the substation equipment based on the basic asset information and the asset management costs. The early warning module, based on the updated asset management information of the substation equipment, identifies key monitoring equipment and sends early warning information to associated user terminals.

[0005] One embodiment of this specification provides a substation asset management device, including a processor, which is used to execute the substation asset management method described in any of the above embodiments.

[0006] One embodiment of this specification provides a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions from the storage medium, the computer executes the substation asset management method described in any of the above embodiments. Attached Figure Description

[0007] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0008] Figure 1 This is a system block diagram illustrating a substation asset management method according to some embodiments of this specification;

[0009] Figure 2 This is a flowchart illustrating a method for substation asset management according to some embodiments of this specification;

[0010] Figure 3 This is an exemplary schematic diagram of a substation asset knowledge graph according to some embodiments of this specification;

[0011] Figure 4 This is an exemplary flowchart illustrating the determination of end-of-life costs according to some embodiments of this specification;

[0012] Figure 5 This is an exemplary flowchart illustrating the prediction of loss costs according to some embodiments of this specification. Detailed Implementation

[0013] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0014] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0015] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0016] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0017] Figure 1 This is a system module diagram of a substation asset management system according to some embodiments of this specification. For example... Figure 1 As shown, the substation asset management system 100 may include an acquisition module 110, a determination module 120, an update module 130, and an early warning module 140.

[0018] Module 110 is used to acquire basic asset information of substation equipment. For more details on acquiring basic asset information, please refer to [link / reference needed]. Figure 2 And its related descriptions.

[0019] Module 120 is used to determine asset management costs. More details on determining asset management costs can be found in [link to relevant documentation]. Figure 2 , Figure 4 , Figure 5 And its related descriptions.

[0020] Update module 130 is used to update the asset management information of substation equipment. More details regarding updating asset management information can be found in [link to relevant documentation]. Figure 2 , Figure 3 And its related descriptions.

[0021] The early warning module 140 is used to identify key monitoring equipment and send early warning information to associated user terminals. More details regarding the issuance of early warning information can be found in [link to relevant documentation]. Figure 2 And its related descriptions.

[0022] Figure 2 This is a flowchart illustrating a substation asset management method according to some embodiments of this specification. For example... Figure 2 As shown, process 200 includes the following steps. In some embodiments, process 200 is executed by acquisition module 110.

[0023] Step 210: Obtain basic asset information of the substation equipment; the basic asset information includes at least one of the following: condition monitoring information, equipment information, equipment maintenance information, and power supply of the substation equipment.

[0024] Transformer equipment refers to equipment that directly produces, transmits, distributes, and uses electrical energy, and can consist of multiple sub-equipment. For example, transformer equipment may include transformers, high-voltage circuit breakers, disconnect switches, busbars, surge arresters, capacitors, reactors, and other sub-equipment. Basic asset information for transformer equipment refers to information related to the basic assets of the transformer equipment. For example, basic asset information may include at least one of the following: condition monitoring information, equipment information, equipment maintenance information, and the power supply quantity of the transformer equipment.

[0025] Condition monitoring information refers to relevant information obtained from monitoring the status of equipment. For example, condition monitoring information may include equipment usage time, operating status, standby status, outage status, fault repair status, and degree of damage—information related to the equipment's status. The degree of damage can be determined based on the number of fault repairs and usage time. For instance, the more fault repairs, the greater the degree of damage; the longer the equipment's usage time, the greater the degree of damage.

[0026] Equipment information refers to information inherent to the equipment. For example, equipment information may include information related to the equipment itself, such as equipment type, quantity, and procurement cost.

[0027] Equipment maintenance information refers to relevant information generated during equipment maintenance. For example, equipment maintenance information may include maintenance and repair records, maintenance and repair costs, and equipment status monitoring information during maintenance and repair.

[0028] The power supply capacity of a power transformer refers to the amount of electricity supplied by the power transformer within a certain period of time. For example, supplying 300,000 kilowatts of electricity per hour.

[0029] In some embodiments, the acquisition module 110 can acquire basic asset information of the power equipment in a variety of ways.

[0030] In some embodiments, the acquisition module 110 can acquire status monitoring information through a monitoring system. The monitoring system may include cameras, sensing devices, handheld terminals, electronic tags, etc. For example, a handheld terminal can be used to manually detect whether the device has been deactivated, and a timer can be used to obtain the usage time of the device.

[0031] In some embodiments, the acquisition module 110 can obtain equipment information through manual input. For example, the equipment type, quantity, etc. can be manually input; or the equipment information can be obtained by querying the procurement and installation records of each substation.

[0032] In some embodiments, the acquisition module 110 can acquire equipment maintenance information through manual input. In some embodiments, the acquisition module 110 can acquire equipment maintenance information by retrieving historical monitoring information from the monitoring system. The historical monitoring information includes equipment maintenance information, which can be stored in a storage device or a database.

[0033] In some embodiments, the acquisition module 110 can determine the power supply of the power equipment based on the historical power supply data of the power equipment. For example, if a power equipment has supplied an average of 300,000 kilowatts of electricity per hour over a period of time, then the power supply of the power equipment is determined to be 300,000 kilowatts per hour.

[0034] Step 220: Determine the asset management cost based on the basic asset information.

[0035] Asset management costs refer to the costs incurred in managing and using substation equipment and related assets. For example, the costs incurred in maintaining substation equipment during its use. In some embodiments, asset management costs may include replacement costs, maintenance costs, scrapping costs, and loss costs. Replacement costs and maintenance costs are described in detail below. More information on scrapping costs can be found in [link to relevant documentation]. Figure 4 And related descriptions. For more information on loss costs, please refer to [link / reference]. Figure 5 And related descriptions.

[0036] In some embodiments, the determining module 120 can determine the asset management cost based on historical asset management costs. For example, if the current underlying asset information is similar to a certain historical underlying asset information, the determining module 120 can retrieve the historical asset management cost corresponding to that historical underlying asset information and determine it as the current asset management cost. Both the historical underlying asset information and the corresponding historical asset management cost can be stored in a storage device or a database.

[0037] In some embodiments, a substation asset knowledge graph can be constructed based on basic asset information, and asset management costs can be determined based on the substation asset knowledge graph through a cost prediction model. The cost prediction model is a machine learning model.

[0038] A substation asset knowledge graph refers to a graph that reflects the asset status of substation equipment, consisting of nodes and edges. For more information on substation asset knowledge graphs, please see [link to relevant documentation]. Figure 3 And related content.

[0039] In some embodiments, the asset management cost of substation equipment can be determined based on a substation asset knowledge graph and a cost prediction model. In some embodiments, the asset management cost of each sub-equipment can be determined based on a subgraph of the substation asset knowledge graph and a cost prediction model, and the sum of the asset management costs of all sub-equipment is taken as the asset management cost of the substation equipment. In some embodiments, based on the subdivision of the subgraph of the substation asset knowledge graph, the specific asset management costs of each sub-equipment, such as replacement costs, maintenance costs, scrapping costs, and loss costs, can be determined using a cost prediction model, thereby determining the asset management cost of each sub-equipment. For detailed explanations, please refer to other sections of this specification, such as... Figure 4 , Figure 5 Descriptions, etc.

[0040] The cost prediction model is a machine learning model. In some embodiments, the cost prediction model may include a graph neural network model.

[0041] The input to the cost forecasting model is a knowledge graph of substation assets. The output of the cost forecasting model is the asset management cost of the substation equipment.

[0042] An initial cost prediction model can be trained based on training samples and their labels. The initial cost prediction model can be a cost prediction model without set parameters. Training samples can include historical substation asset knowledge graphs, and labels can include historically generated actual asset management costs. Training samples and labels can be determined based on historical data stored in storage devices or databases; labels can be obtained through manual annotation. Training samples can be input into the initial cost prediction model, and a loss function can be constructed based on the output of the initial cost prediction model and the labels. The parameters of the initial cost prediction model are then iteratively updated based on the loss function until preset conditions are met, completing the training and obtaining a trained cost prediction model. These preset conditions can include the loss function being less than a threshold, convergence, or the training period reaching a threshold.

[0043] In some embodiments, the input to the cost prediction model can also be a subgraph of the substation asset knowledge graph. The output of the cost prediction model can also be the asset management cost of each sub-device. Correspondingly, when training the cost prediction model, the training samples can also include a subgraph of the substation asset knowledge graph. For the remaining training components, please refer to the description in the preceding sections of this specification.

[0044] In some embodiments, the cost prediction model includes a first prediction model, a second prediction model, a third prediction model, and a fourth prediction model. The first prediction model is used to determine scrapping costs; more information about the first prediction model can be found at [link to relevant documentation]. Figure 4 And related content. The second prediction model is used to determine the loss cost; for more information on the second prediction model, see [link to relevant documentation]. Figure 5And related content. The third and fourth prediction models are used to determine replacement costs and maintenance costs, respectively. For more information on the third and fourth prediction models, please refer to the relevant content below.

[0045] Based on the knowledge graph of substation assets, the asset management cost can be determined by the cost prediction model, which can make the asset management cost more accurate. The use of the cost prediction model can also greatly shorten the determination time of asset management cost and effectively improve the calculation efficiency.

[0046] Step 230: Update the asset management information of the substation equipment based on the basic asset information and asset management costs.

[0047] Asset management information refers to information related to the management of power substation equipment assets. In some embodiments, asset management information may include equipment type and quantity, equipment status, equipment management strategy, and corresponding asset management costs. The equipment management strategy may include whether the equipment should be replaced, whether it should continue to be used, or whether it should be scrapped.

[0048] In some embodiments, asset management information can be updated via electronic tags. For example, electronic tags can be linked and stored to equipment type and quantity, equipment status, equipment management policies, and corresponding asset management costs. Based on basic asset information and asset management costs, it can be determined whether equipment should be replaced, continued to be used, or scrapped, thereby updating the equipment management policy. For example, if the replacement cost of substation equipment exceeds a replacement cost threshold, the equipment is determined to be scrapped, and the equipment management policy is updated to determine that the equipment is scrapped. Since equipment type and quantity, equipment status, equipment management policies, and corresponding costs are linked and stored via electronic tags, updates to the equipment management policy will cause corresponding updates to the asset management information.

[0049] In some embodiments, when the asset management information of the power equipment is updated, the asset management cost will also be updated accordingly. The updated asset management cost may include the replacement cost of the power equipment, the maintenance cost of the power equipment, the scrapping cost of the power equipment, the loss cost of the power equipment, and the asset management cost of each sub-equipment, etc.

[0050] Step 240: Based on the updated asset management information of the substation equipment, identify key monitoring equipment and send early warning information to the relevant user terminals.

[0051] Key monitoring equipment refers to equipment that requires focused monitoring due to potential problems or excessively high maintenance costs. In some embodiments, substation equipment or its sub-equipment whose updated asset management information meets preset conditions can be designated as key monitoring equipment, such as equipment whose replacement or maintenance costs exceed preset cost values.

[0052] Associated user terminals refer to user terminals that are associated with the power equipment for monitoring the equipment. These terminals can include mobile devices, desktop computers, etc., with no specific limitations. Users of the associated user terminals can be maintenance personnel, managers, etc., of the power equipment.

[0053] Warning information refers to information that alerts associated user terminals. For example, warning information may include asset management information for key monitored equipment and subsequent equipment management strategies. In some embodiments, the warning module 140 may send warning information to associated user terminals via wireless communication.

[0054] Based on basic asset information and asset management costs, the asset management information of substation equipment is updated, thereby identifying key monitoring equipment and issuing early warning information to related user terminals. This can prompt related user terminals to focus on monitoring the equipment, enabling them to promptly confirm equipment management strategies and achieve efficient and accurate management of substation equipment.

[0055] Figure 3 This is an exemplary schematic diagram of a substation asset knowledge graph according to some embodiments of this specification.

[0056] like Figure 3 The diagram shows a substation asset knowledge graph 300, which consists of several nodes and edges connecting the nodes. The nodes in the substation asset knowledge graph can include substation equipment nodes, sub-equipment nodes, procurement nodes, installation nodes, maintenance nodes, and scrapping nodes. Each node can include corresponding node attribute information.

[0057] A transformer equipment node can refer to a node corresponding to a transformer. For example, such as... Figure 3 In this context, substation node A can be a corresponding substation device A. In some embodiments, the node attribute information of a substation node may include the component information of its corresponding substation, the overall status of the substation, the power supply of the substation, and historical data of each sub-device. The component information may refer to the usage status of each sub-device of the substation. The overall status of the substation may include the commissioning time of the substation. The historical data of each sub-device may include data related to the procurement, installation, maintenance, and scrapping status and corresponding costs. The historical data of each sub-device may also be time-series data composed of historical data collected at multiple time points. Since at least some of the costs associated with the procurement, installation, maintenance, and scrapping of sub-devices change over time, the procurement, installation, maintenance, and scrapping of sub-devices can each correspond to a separate time-series data point.

[0058] A sub-device node can refer to a node corresponding to a specific sub-device. For example, sub-device nodes can correspond to transformers, high-voltage circuit breakers, disconnect switches, busbars, surge arresters, capacitors, reactors, etc. Figure 3The sub-device nodes C and B shown can represent the transformer C and high-voltage circuit breaker B that constitute the power equipment A. It should be noted that... Figure 3 The nodes and edges shown are for illustrative purposes only; actual graph data structures are more complex. In some embodiments, the node attribute information of a sub-device node may include device type, device model, device number, status monitoring information, etc. For details on status monitoring information, please refer to [link to relevant documentation]. Figure 2 And related descriptions.

[0059] A procurement node can refer to a node related to the procurement information of a sub-equipment. In some embodiments, the node attribute information of a procurement node may include procurement costs, suppliers, and supplier quality parameter requirements. Supplier quality parameter requirements refer to parameters set to detect whether the quality of the equipment provided by the supplier meets the standards.

[0060] In some embodiments, a substation asset knowledge graph can be categorized based on the procurement information of the included sub-equipment, and multiple procurement nodes with different attribute information can be set. Then, the sub-equipment node corresponding to each sub-equipment can be connected to the corresponding procurement node through corresponding edges based on the type of its procurement information. For example... Figure 3 The sub-device node C and sub-device node B shown are both connected to the procurement node D through edges, indicating that the procurement information of the sub-devices corresponding to sub-device node C and sub-device node B belongs to the same type, such as the procurement cost being within the same range, the suppliers being in the same region, and the suppliers' quality parameter requirements being at the same level.

[0061] An installation node can refer to a node related to the device installation information of a sub-device. In some embodiments, the node attribute information of an installation node may include installation acceptance status, installation and dismantling labor costs, installation environment, etc.

[0062] A maintenance node can refer to a node related to the equipment maintenance information of a sub-device. In some embodiments, the node attribute information of a maintenance node may include maintenance and repair information. Maintenance and repair information may include maintenance items, maintenance costs, maintenance time, repair items, repair reasons, and repair time, etc.

[0063] A scrapping node can refer to a node related to the scrapping information of a sub-equipment. In some embodiments, the node attribute information of a scrapping node may include scrapping transportation distance, scrapping environment, dismantling labor costs, etc.

[0064] Similar to the setting of procurement nodes, a substation asset knowledge graph can set up multiple installation nodes, multiple maintenance nodes, and multiple scrapping nodes based on the specific circumstances of the included substation equipment and sub-equipment. Furthermore, the node attribute information corresponding to different installation nodes, different maintenance nodes, and different scrapping nodes can be different.

[0065] The substation asset knowledge graph has five types of edges: type 1, type 2, type 3, type 4, and type 5. Each type of edge includes attribute information corresponding to its type.

[0066] Type 1 edges can be used to connect substation nodes and sub-device nodes, such as... Figure 3 As shown, the first type of edge can be represented as "edge 1", where sub-device node B is connected to substation equipment node A through edge 1a, and sub-device node C is connected to substation equipment node A through edge 1b, indicating that substation equipment node A includes two sub-devices. Figure 3 The substation asset knowledge graph shown is for illustrative purposes only. A substation asset knowledge graph can include several substation equipment nodes, and each substation equipment node can be connected to its corresponding several sub-equipment nodes through several Type 1 edges. The attribute information corresponding to the Type 1 edges includes sub-equipment type, function, etc.

[0067] The second type of edge can be used to connect each sub-device node to the corresponding procurement node, such as... Figure 3 As shown, the second type of edge can be represented as "edge 2". In this edge, sub-device node B is connected to procurement node D through edge 2b, and sub-device node C is connected to procurement node D through edge 2a, indicating that the procurement information of the sub-devices corresponding to sub-device node C and sub-device node B belong to the same type. Figure 3 The substation asset knowledge graph shown is for illustrative purposes only. A substation asset knowledge graph can include several procurement nodes, and each sub-equipment node can be connected to its corresponding procurement node via a Type 2 edge. The attribute information corresponding to the Type 2 edge is the procurement time.

[0068] The third type of edge can be used to connect each sub-device node to its corresponding installation node, such as... Figure 3 As shown, the edge of type 3 can be represented as "edge 3".

[0069] Among them, sub-device node B is connected to installation node E through edge 3b, and sub-device node C is connected to installation node E through edge 3a, indicating that the installation information of the sub-devices corresponding to sub-device node C and sub-device node B belong to the same type. Figure 3 The substation asset knowledge graph shown is for illustrative purposes only. A substation asset knowledge graph can include several installation nodes, and each sub-device node can be connected to its corresponding installation node via a Type 3 edge. The attribute information corresponding to the Type 3 edge is the installation time.

[0070] The fourth type of edge can be used to connect each sub-device node to its corresponding maintenance node, such as... Figure 3As shown, the fourth type of edge can be represented as "edge 4". In this edge, sub-device node B is connected to maintenance node F through edge 4b, and sub-device node C is connected to maintenance node F through edge 4a, indicating that the maintenance information of the sub-devices corresponding to sub-device node C and sub-device node B belong to the same type. Figure 3 The substation asset knowledge graph shown is for illustrative purposes only. A substation asset knowledge graph can include several maintenance nodes, and each sub-device node can be connected to its corresponding maintenance node via a Type 4 edge. The attribute information corresponding to the Type 4 edge is the number of maintenance visits and the number of repair visits.

[0071] The fifth type of edge can be used to connect each sub-device node to its corresponding decommissioned node, such as... Figure 3 As shown, the fifth type of edge can be represented as "edge 5". In this edge, sub-device node B is connected to scrapped node G through edge 5b, and sub-device node C is connected to scrapped node G through edge 5a, indicating that the scrapping information of the sub-devices corresponding to sub-device node C and sub-device node B can belong to the same type. Figure 3 The substation asset knowledge graph shown is for illustrative purposes only. A substation asset knowledge graph can include several scrapped nodes. Each sub-device node can be connected to its corresponding scrapped node via a Type 5 edge. The attribute information corresponding to the Type 5 edge is whether it is scrapped and the scrapping time. Specifically, if the sub-device corresponding to a certain sub-device node does not need to be scrapped, then that sub-device node can be connected to any scrapped node in the substation asset knowledge graph, and the attribute information corresponding to the Type 5 edge is "not scrapped".

[0072] In some embodiments, the node attribute information of a sub-device node may further include the failure probability. In some embodiments, the substation asset knowledge graph also includes a type 6 edge. When a failure in one sub-device affects a failure in another sub-device, the two sub-device nodes corresponding to these two sub-devices are connected by a type 6 edge. Figure 3 In this context, edge type 6 can be represented as "edge 6". In this example, sub-device node B and sub-device node C are connected by edge 6a, indicating that a failure in either sub-device (sub-device node C or sub-device node B) will affect the failure of the other sub-device. The attribute information of edge type 6 includes the degree of fault correlation.

[0073] Failure probability refers to the probability that a sub-device will fail. In some embodiments, failure probability can be expressed as a percentage. The higher the percentage, the greater the failure probability. For example, failure probability can be expressed as 20%, 50%, or 80%, where 80% represents a failure probability greater than 20%. Failure probability can be statistically analyzed or predicted based on historical usage data of the corresponding sub-device. In some embodiments, the failure probability of a sub-device can be positively correlated with the usage time of the sub-device; for example, the longer the usage time, the higher the corresponding failure probability.

[0074] Fault correlation refers to the degree to which faults between sub-equipment affect each other. For example, fault correlation can be represented numerically (e.g., any integer between 1 and 10). The greater the degree of mutual influence between faults in sub-equipment, the greater the fault correlation. Fault correlation can be preset based on empirical values ​​and can be a fixed value after being preset. For example, the fault correlation between two sub-equipment can be determined based on historical operation and configuration data of the substation equipment and remains unchanged after being set.

[0075] Using the failure probability as the node attribute information of the sub-device node and the failure correlation degree as the attribute information of the edge connecting the sub-device node can reflect the mutual influence of failures between sub-devices. This allows the constructed substation asset knowledge graph to reflect the interrelationship between sub-devices, thereby making the asset management cost obtained based on the substation asset knowledge graph more accurate.

[0076] In some embodiments, the determining module 120 can update the fault probabilities in the substation asset knowledge graph. For example, when a sub-equipment experiences problems, damage, repairs, maintenance, or other situations that affect its quality, a fault probability update is performed.

[0077] In some embodiments, when updating the fault probability of the substation asset knowledge graph, the update can start from the sub-device node whose quality has changed (e.g., when a problem occurs, damage occurs, repairs occur, maintenance occurs, etc.), and then gradually update from the adjacent nodes of that node until each sub-device node has been updated. The fault probability of the sub-device node whose quality has changed can be updated using formula (1). Formula (1) s′=s+a*T

[0078] Where s′ represents the updated failure probability of the sub-device node whose quality has changed, s represents the unupdated failure probability of the sub-device node whose quality has changed, a represents the quality change factor, and T represents the conversion baseline value.

[0079] In some embodiments, the conversion reference value T for different sub-devices can be the same or different, and the initial value of the conversion reference value T for each sub-device is preset to 0. T can be updated as the sub-device is used; the longer the sub-device is used, the larger T becomes. When a sub-device is scrapped, the failure probability of the sub-device node that has changed can be directly adjusted to the maximum value, such as 100%. The determination of the quality change factor a is described in the relevant description below.

[0080] The failure probability of other sub-device nodes can be updated using formula (2). Formula (2) is: si″=si+r1*s1+r2*s2+…+rn*sn

[0081] Where si″ represents the updated failure probability of the neighboring sub-device node i (hereinafter referred to as sub-device node i) of the sub-device node whose quality has changed, si represents the failure probability of sub-device node i before the update, and s1 to sn represent the failure probabilities corresponding to the n neighboring sub-device nodes of sub-device node i, respectively. For the n neighboring sub-device nodes whose failure probabilities have been updated, their corresponding failure probabilities are taken as the updated failure probabilities, and those that have not been updated are taken as the failure probabilities before the update. For example, if s1 is the aforementioned sub-device node whose quality has changed, then the value of s1 can be the aforementioned s′. r1 to rn represent the failure correlation degree between the n neighboring sub-device nodes of sub-device node i and sub-device node i, respectively. For example, if s1 is the aforementioned sub-device node whose quality has changed, then r1 represents the failure correlation degree between the aforementioned sub-device node whose quality has changed and sub-device node i. The failure correlation degree can be determined based on the edge attribute of the sixth type edge between the two sub-device nodes. If there is no sixth type edge between the two sub-device nodes, the failure correlation degree is 0.

[0082] Here, the neighboring child device nodes of child device node i can refer to child device nodes that are directly connected to child device node i via an edge. For example... Figure 3 Sub-device node C and sub-device node B are neighboring sub-device nodes.

[0083] By updating the failure probabilities of sub-devices whose quality has changed, and eventually updating the failure probabilities of each sub-device, the failure probabilities of each sub-device can be made more accurate, thereby making the constructed substation asset knowledge graph more accurate.

[0084] In some embodiments, the quality change factor 'a' can be predicted using a factor model. The factor model is a machine learning model.

[0085] The inputs to the factor model are the mode of quality change (e.g., problem occurrence, damage, repair, maintenance, etc.), the type of sub-equipment, historical change data of the sub-equipment, and the performance parameters of the sub-equipment at the time of manufacture. The mode of change can represent different changes based on different values. For example, mode 1 represents a problem, mode 2 represents damage, mode 3 represents repair, and mode 4 represents maintenance, etc. The output of the factor model is the quality change factor.

[0086] An initial factor model can be trained based on training samples and their labels. The initial factor model can be a factor model without set parameters. Training samples can include historical change patterns (problems, damage, repairs, maintenance, etc.), historical sub-device types, historical change data of sub-devices, and performance parameters of sub-devices at the time of manufacture. Labels can be determined based on the ratio between a preset usage duration and the actual usage duration, i.e., the label is the preset usage duration / actual usage duration, which can be represented numerically and theoretically should be a number ≥ 1. Labels can be obtained through manual annotation. The preset usage duration is the usage duration given at the time of manufacture. Training samples and labels can be determined based on historical data stored in storage devices or databases. Training samples can be input into the initial factor model, and a loss function can be constructed based on the output of the initial factor model and the labels. The parameters of the initial factor model are iteratively updated based on the loss function until preset conditions are met, completing the training and obtaining a trained factor model. Preset conditions can include the loss function being less than a threshold, convergence, or the training period reaching a threshold.

[0087] By determining the quality change factor through a factor model and using this quality change factor to update the failure probability of sub-devices whose quality has changed, the update of the failure probability can be made more accurate.

[0088] Understandable Figure 3 The substation asset knowledge graph in the image is for illustrative purposes only. The actual substation asset knowledge graph is more complex and can include more nodes and more edges.

[0089] Figure 4 This is a schematic diagram illustrating the determination of scrap costs based on some embodiments of this specification. For example... Figure 4 As shown, process 400 includes the following steps. In some embodiments, process 400 is executed by determination module 120.

[0090] Step 410: Determine whether the sub-equipment of the power equipment is scrapped.

[0091] In some embodiments, the scrapping status of sub-equipment can be confirmed manually. For example, based on experience, it can be determined whether the sub-equipment should continue to be used. If it is determined that the sub-equipment should no longer be used, then the sub-equipment of the power transformer can be scrapped.

[0092] In some embodiments, the determination of whether a sub-device is scrapped can also be based on its failure probability, number of replacements, and number of maintenances. For example, if the failure probability of a sub-device node in the substation asset knowledge graph is greater than a probability threshold, and the number of replacements and maintenances of the sub-device node are greater than the corresponding number thresholds, then the sub-device is scrapped.

[0093] For more information on the failure probability of sub-device nodes, please refer to [link / reference]. Figure 3This includes related content. The number of times a sub-device node is replaced and maintained refers to the number of times the sub-device is replaced and the number of times it is maintained, respectively, and can be expressed numerically. For example, the number of replacements and maintenance could be 3 times and 5 times, respectively. The number of times a sub-device is replaced and maintained can be obtained based on the sub-device's historical maintenance records.

[0094] For determining the failure probability of sub-device nodes, please refer to [link / reference]. Figure 3 The number of replacements and maintenance performed on sub-device nodes can be retrieved from storage devices or databases storing historical maintenance records. In some embodiments, a sub-device may be designated as scrapped if any one of its failure probability, replacement count, or maintenance count exceeds a corresponding threshold. For example, if the failure probability threshold is set to 80%, the sub-device is deemed scrapped when its failure probability exceeds 80%. If the replacement count threshold is set to 5, the sub-device is deemed scrapped when its replacement count exceeds 5 (e.g., 6). If the maintenance count threshold is set to 10, the sub-device is deemed scrapped when its replacement count exceeds 10 (e.g., 11).

[0095] In some embodiments, the determination of whether a sub-device is obsolete can be based on predicted replacement and maintenance costs. For example, if both the predicted replacement and maintenance costs are higher than corresponding preset cost thresholds, the sub-device is determined to be obsolete. As another example, if the predicted maintenance cost is greater than the replacement cost, the sub-device is determined to be obsolete. More information on predicted replacement and maintenance costs can be found in the relevant descriptions below.

[0096] The determination of whether a sub-device is scrapped can be made based on the failure probability, replacement frequency, and maintenance frequency of the sub-device node. At the same time, the determination of whether a sub-device is scrapped can be made based on maintenance cost and replacement cost. The above determination methods are diversified and can achieve accurate determination of whether a sub-device is scrapped.

[0097] Step 420: In response to the sub-equipment being scrapped, determine the cost type of asset management cost as scrap cost, and extract the scrap cost subgraph from the substation asset knowledge graph.

[0098] Scrapping costs refer to the costs incurred when equipment is scrapped. For example, scrapping costs may include the cost of manual dismantling, transportation costs, etc.

[0099] The scrap cost subgraph refers to a subgraph extracted from the substation asset knowledge graph to reflect the scrapping status of sub-equipment. For example... Figure 4 The following is based on Figure 3The scrap cost subgraph 430-1 extracted from the substation asset knowledge graph 300 includes substation equipment node A, sub-equipment node B, and scrapped node G. Edges include type 1 edge "edge 1a" and type 5 edge "edge 5b". Substation equipment node A is connected to sub-equipment node B via edge 1a, and sub-equipment node B is connected to scrapped node G via edge 5b. Since scrap cost subgraph 430-1 indicates that sub-equipment node B in the substation asset knowledge graph 300 needs to be scrapped, the attribute information of "edge 5b" includes scrapping. For explanations of the attributes of other nodes and edges in scrap cost subgraph 430-1, please refer to [link to relevant documentation]. Figure 3 The above-mentioned scrap cost sub-diagram 430-1 is only an example. The scrap cost sub-diagram may also include other substation equipment nodes, other sub-equipment nodes that need to be scrapped, and scrapping nodes corresponding to the corresponding sub-equipment that need to be scrapped.

[0100] In some embodiments, a corresponding scrap cost subgraph can be extracted based on each sub-device. For example, in a substation asset knowledge graph, a scrap cost subgraph can be obtained by extracting a sub-device node, and the substation equipment nodes and scrapped nodes connected to the sub-device node through edges. If the sub-device corresponding to a certain sub-device node does not need to be scrapped, then the edge attribute of the fifth type of edge in its corresponding scrap cost subgraph can include "not scrapped," and its corresponding scrap cost is 0.

[0101] Step 430: Process the scrap cost subgraph based on the first prediction model to determine the scrap cost; the first prediction model is a graph neural network model.

[0102] The first predictive model is a graph neural network model used to determine scrap costs, which can be obtained by training a machine learning model.

[0103] like Figure 4 As shown, the input of the first prediction model 430-2 is the scrap cost subgraph 430-1, and the output of the first prediction model 430-2 is the scrap cost 430-3. The scrap cost 430-3 can be obtained based on the output of the fifth type of edge in the scrap cost subgraph 430-1, and can be used to represent the scrap cost generated when the sub-device corresponding to the sub-device node of the fifth type of edge is scrapped.

[0104] The first prediction model can be trained based on training samples and labels. The training samples for the first prediction model can include a historical scrap cost submap, and the labels can include the historical actual scrap cost of the corresponding sub-equipment in the historical scrap cost submap. The training samples and labels can be determined based on historical data stored in storage devices or databases. The training process for the first prediction model is similar to that for the cost prediction model; see [link to documentation] for details. Figure 1 And its related descriptions.

[0105] Determining scrap costs based on the first prediction model can yield more accurate scrap costs and shorten the time required to determine scrap costs, thereby improving the management efficiency of substation assets.

[0106] Figure 5 This is a schematic diagram illustrating the process of predicting loss costs according to some embodiments of this specification. For example... Figure 5 As shown, process 500 includes the following steps. In some embodiments, process 500 is executed by determination module 120.

[0107] Step 510: Determine the risk level of the sub-equipment of the power equipment.

[0108] The risk level of a sub-equipment reflects the degree of risk that may occur to it. The risk level can be represented by a positive integer, such as 1, 2, etc., with higher numbers indicating higher risk levels. A maximum risk level (e.g., 10) can be set. In some embodiments, the risk level of a sub-equipment can be determined based on the probability of a sub-equipment malfunctioning, the impact of the malfunction on the power distribution equipment, etc. For example, the higher the probability of a sub-equipment malfunctioning, the higher its risk level. Similarly, the greater the impact of the malfunction on the power distribution equipment, the higher its risk level.

[0109] In some embodiments, the risk level of a sub-equipment of a power equipment can be determined manually. For example, the risk level of a sub-equipment can be determined based on experience, assessing the degree of impact of the sub-equipment on the power equipment. If the manual assessment indicates that the sub-equipment has a small impact on the power equipment, the risk level of the sub-equipment is determined to be 1; if the manual assessment indicates that the sub-equipment has a large impact on the power equipment, the risk level of the sub-equipment is determined to be 5, etc.

[0110] The risk level of a sub-equipment is positively correlated with the failure probability of each sub-equipment node in the substation asset knowledge graph. For more information on the failure probability of sub-equipment nodes, please refer to [link to relevant documentation]. Figure 3 And related content.

[0111] In some embodiments, the risk level of a sub-device can be determined by the positive correlation between its risk level and the failure probability of each sub-device node in the substation asset knowledge graph. For example, a failure probability of 10% for a sub-device node corresponds to a risk level of 1, and a failure probability of 20% for a sub-device node corresponds to a risk level of 2, and so on.

[0112] Since the failure probability of sub-device nodes is updated in real time, determining the risk level of a sub-device based on the failure probability of its sub-device nodes will result in a more accurate risk level, which in turn will make the predicted loss cost more accurate.

[0113] Step 520: In response to the risk level being greater than the preset risk threshold, the cost type of asset management cost is determined to be loss cost, and the loss cost subgraph is extracted from the substation asset knowledge graph.

[0114] The preset risk threshold can be determined based on the failure probability of sub-devices that may cause power outages. The higher the failure probability, the higher the risk of power outage, and the higher the probability of power loss and equipment damage. In some embodiments, the preset risk threshold can be manually set, such as setting the preset risk threshold to 8.

[0115] Loss costs refer to the costs incurred due to equipment damage. For example, loss costs can include electricity loss costs and equipment damage loss costs. Specifically, electricity loss costs refer to the costs incurred due to power outages.

[0116] The loss cost subgraph refers to a subgraph extracted from the substation asset knowledge graph to reflect the loss situation of sub-equipment. For example... Figure 5 The following is based on Figure 3 The loss cost subgraph 530-1 extracted from the substation asset knowledge graph 300 includes nodes such as substation equipment node A, sub-equipment node B, and procurement node D. Edges include type 1 edge "edge 1a" and type 2 edge "edge 2b". Substation equipment node A is connected to sub-equipment node B via edge 1a, and sub-equipment node B is connected to procurement node D via edge 2b. Loss cost subgraph 530-1 indicates that the risk level of sub-equipment node B in the substation asset knowledge graph 300 exceeds a preset risk threshold, resulting in loss costs. For a description of the attributes of the nodes and edges in loss cost subgraph 530-1, please refer to [link to relevant documentation]. Figure 3 The description is as follows. The above loss cost sub-graph 530-1 is only an example. The loss cost sub-graph may also include other substation equipment nodes, sub-equipment nodes corresponding to other sub-equipment with risk levels greater than the preset risk threshold, and procurement nodes corresponding to the corresponding sub-equipment, etc.

[0117] In some embodiments, a corresponding loss cost subgraph can be extracted based on each sub-device. For example, in a substation asset knowledge graph, a loss cost subgraph can be obtained by extracting a sub-device node, and the substation equipment nodes and procurement nodes connected to the sub-device node through edges.

[0118] Step 530: Process the loss cost subgraph based on the second prediction model to determine the loss cost; the second prediction model is a graph neural network model.

[0119] The second predictive model is a graph neural network model used to determine the loss cost, which can be obtained based on training a machine learning model.

[0120] like Figure 5As shown, the input to the second prediction model 530-2 is the loss cost subgraph 530-1, and the output of the second prediction model 530-2 is the loss cost 530-3. The loss cost 530-3 can be obtained based on the output of the second type of edge in the loss cost subgraph 530-1, and can be used to represent the loss cost generated by the sub-device corresponding to the sub-device node of the second type of edge.

[0121] The second prediction model can be trained based on training samples and labels. The training samples for the second prediction model can include a historical loss cost submap, and the labels can include the historical actual loss cost of the corresponding sub-device in the historical loss cost submap. The training samples and labels can be determined based on historical data stored in storage devices or databases. The training process for the second prediction model is similar to that for the cost prediction model; see [link to documentation] for details. Figure 1 And its related descriptions.

[0122] Determining loss costs based on the second prediction model results in more accurate loss costs and shortens the time required to determine loss costs, thereby improving the management efficiency of substation assets.

[0123] In some embodiments, asset management costs include replacement costs. Replacement costs refer to the costs incurred when replacing equipment. For example, replacement costs may include expenses related to repurchasing, installation, dismantling, etc.

[0124] In some embodiments, a replacement cost subgraph is extracted from the substation asset knowledge graph, and the replacement cost subgraph is processed based on a third prediction model to determine the replacement cost.

[0125] The replacement cost subgraph is a subgraph extracted from the substation asset knowledge graph to reflect the replacement status of sub-equipment. The nodes in the replacement cost subgraph include substation equipment nodes, sub-equipment nodes, procurement nodes, and installation nodes. Edges include Type 1 edges, Type 2 edges, and Type 3 edges, along with corresponding node and edge attributes. Specifically, substation equipment nodes are connected to sub-equipment nodes via Type 1 edges, sub-equipment nodes are connected to their corresponding procurement nodes via Type 2 edges, and sub-equipment nodes are connected to their corresponding installation nodes via Type 3 edges. For more information on nodes and edges, please refer to [link to relevant documentation]. Figure 3 And related descriptions.

[0126] In some embodiments, a corresponding replacement cost subgraph can be extracted based on each sub-device. For example, in a substation asset knowledge graph, a replacement cost subgraph can be obtained by extracting a sub-device node and the substation equipment node, procurement node, and installation node connected to the sub-device node via edges.

[0127] The third predictive model is a graphical network model used to determine replacement costs, which can be obtained based on training a machine learning model.

[0128] The input to the third prediction model is the replacement cost subgraph, and the output of the third prediction model is the replacement cost. The replacement cost can be obtained based on the output of the second-type and third-type edges in the replacement cost subgraph. It can be used to represent the replacement cost generated by the sub-device corresponding to the sub-device node of the second-type and third-type edges. The total replacement cost corresponding to a sub-device can be the sum of the outputs of its corresponding second-type and third-type edges in the replacement cost subgraph.

[0129] The third prediction model can be trained based on training samples and labels. The training samples for the third prediction model can include a historical replacement cost submap, and the labels can include the historical actual replacement costs (such as actual purchase costs, installation costs, etc.) of the corresponding sub-equipment in the historical replacement cost submap. Training samples and labels can be determined based on historical data stored in storage devices or databases. The training process for the third prediction model is similar to that for the cost prediction model; see [link to documentation] for details. Figure 1 And its related descriptions.

[0130] Determining replacement costs based on the third prediction model can yield more accurate replacement costs and shorten the time required to determine replacement costs, thereby improving the management efficiency of substation assets.

[0131] In some embodiments, asset management costs include maintenance costs. Maintenance costs refer to the costs incurred in maintaining equipment. For example, maintenance costs may include the cost of manual maintenance.

[0132] In some embodiments, a maintenance cost subgraph is extracted from the substation asset knowledge graph, and the maintenance cost subgraph is processed based on a fourth prediction model to determine the maintenance cost.

[0133] The maintenance cost subgraph is a subgraph extracted from the substation asset knowledge graph to reflect the maintenance status of sub-equipment. The nodes of the maintenance cost subgraph include substation equipment nodes, sub-equipment nodes, and maintenance nodes. Edges include Type 1 edges and Type 4 edges, along with corresponding node and edge attributes. Substation equipment nodes and sub-equipment nodes are connected via Type 1 edges, and sub-equipment nodes and maintenance nodes are connected via Type 4 edges. For more information on nodes and edges, please refer to [link to relevant documentation]. Figure 3 And related descriptions.

[0134] In some embodiments, a corresponding maintenance cost subgraph can be extracted based on each sub-device. For example, in a substation asset knowledge graph, a maintenance cost subgraph can be obtained by extracting a sub-device node, and the substation equipment nodes and maintenance nodes connected to the sub-device node through edges.

[0135] The fourth predictive model is a graph neural network model used to determine maintenance costs, which can be obtained by training a machine learning model.

[0136] The input to the fourth prediction model is the maintenance cost subgraph, and the output is the maintenance cost. The maintenance cost can be obtained based on the output of the fourth type of edge in the maintenance cost subgraph, and can be used to represent the maintenance cost generated by the sub-device corresponding to the sub-device node of the fourth type of edge.

[0137] The fourth prediction model can be trained based on training samples and labels. The training samples for the fourth prediction model can include a historical maintenance cost submap, and the labels can include the historical actual maintenance costs of the corresponding sub-devices within the historical maintenance cost submap. The training samples and labels can be determined based on historical data stored in storage devices or databases. The training process for the fourth prediction model is similar to that for the cost prediction model; see [link to documentation] for details. Figure 1 And its related descriptions.

[0138] Determining maintenance costs based on the fourth prediction model can yield more accurate results and shorten the time required for cost determination, thereby improving the management efficiency of substation assets.

[0139] In some embodiments, a substation asset management device includes a processor that can be used to perform a substation asset management method.

[0140] In some embodiments, a computer-readable storage medium stores computer instructions. When a computer reads the computer instructions from the storage medium, the computer can execute a substation asset management method. It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects may be any one or a combination of the above-mentioned effects, or any other possible beneficial effects.

[0141] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0142] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0143] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of embodiments that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the substance and scope of the embodiments described herein. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.

[0144] Similarly, it should be noted that, in order to simplify the descriptions disclosed herein and thus aid in the understanding of one or more embodiments, the foregoing description of embodiments in this specification sometimes combines multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of the single embodiments disclosed above.

[0145] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0146] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0147] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A method for managing substation assets, comprising: Obtain basic asset information for power equipment; The basic asset information includes at least one of the following: status monitoring information, equipment information, equipment maintenance information, and power supply of substation equipment; Based on the aforementioned basic asset information, a substation asset knowledge graph is constructed. The nodes of this knowledge graph include substation equipment nodes, sub-equipment nodes, procurement nodes, installation nodes, maintenance nodes, and scrapping nodes. Each node includes corresponding node attribute information. The node attribute information of the sub-equipment nodes includes the failure probability. The edges of the substation asset knowledge graph include six types of edges. Type 1 edges connect the substation equipment nodes to the sub-equipment nodes. Type 2 edges connect each sub-equipment node to its corresponding procurement node. Type 3 edges connect each sub-equipment node to its corresponding installation node. Type 4 edges connect each sub-equipment node to its corresponding maintenance node. Type 5 edges connect each sub-equipment node to its corresponding scrapping node. When a failure of one sub-equipment affects a failure of another sub-equipment, the two corresponding sub-equipment nodes are connected by a Type 6 edge. Each type of edge includes attribute information corresponding to its edge type. The attribute information of the Type 6 edge includes the failure correlation degree. Based on the substation asset knowledge graph, asset management costs are determined through a cost prediction model; the cost prediction model is a machine learning model. Based on the basic asset information and the asset management cost, update the asset management information of the substation equipment; wherein, the asset management information refers to information on the assets related to the management of the substation equipment, including equipment type and quantity, equipment status, equipment management strategy, and corresponding asset management cost; Based on the updated asset management information of the substation equipment, key monitoring equipment is identified and early warning information is sent to the associated user terminals; The method further includes updating the failure probability, wherein: When a situation arises during the use of the sub-device that affects its quality, a fault probability update is performed. This fault probability update includes: The update begins with the failure probability of the sub-device node whose quality has changed, and then progressively updates the failure probabilities of adjacent nodes; wherein the formula for updating the failure probability of the sub-device node whose quality has changed is: , The updated failure probability of the sub-device node whose quality has changed is represented by s, the unupdated failure probability of the sub-device node whose quality has changed is represented by a, and the conversion baseline value is represented by T. The quality change factor a is determined by a factor model, which is a machine learning model. The input of the factor model is the quality change mode, the sub-device type, the historical change data of the sub-device, and the performance parameters of the sub-device at the time of manufacture. The change mode includes problems, damage, repair, and maintenance. The formula for updating the failure probability of the adjacent nodes is: , The fault probability of the neighboring sub-device node i after the quality change is represented by si, the fault probability of the neighboring sub-device node i before the update is represented by s1 to sn, the fault probabilities corresponding to the n neighboring sub-device nodes of the neighboring sub-device node i are respectively represented by s1 to sn, and the fault correlation degree between the n neighboring sub-device nodes of the neighboring sub-device node i and the neighboring sub-device node i is respectively represented by r1 to rn.

2. The method according to claim 1, wherein the asset management cost includes scrapping cost; and the cost prediction model includes a first prediction model; The determination of asset management costs based on the aforementioned basic asset information includes: Determine whether the sub-equipment of the power equipment is scrapped; In response to the fact that the sub-equipment is scrapped, the cost type of the asset management cost is determined to be the scrap cost, and the scrap cost sub-graph is extracted from the substation asset knowledge graph; The scrap cost is determined by processing the scrap cost subgraph based on the first prediction model; the first prediction model is a graph neural network model.

3. The method according to claim 1, wherein the asset management cost includes loss cost; and the cost prediction model includes a second prediction model; The determination of asset management costs based on the aforementioned basic asset information includes: Determine the risk level of the sub-equipment of the power equipment; In response to the risk level being greater than a preset risk threshold, the cost type of the asset management cost is determined to be the loss cost, and a loss cost subgraph is extracted from the substation asset knowledge graph. The loss cost subgraph is processed based on the second prediction model to determine the loss cost; the second prediction model is a graph neural network model.

4. A substation asset management system, comprising an acquisition module, a determination module, an update module, and an early warning module; The acquisition module is used to acquire basic asset information of the power equipment. The basic asset information includes at least one of the following: status monitoring information, equipment information, equipment maintenance information, and power supply of substation equipment; The determining module is used to construct a substation asset knowledge graph based on the basic asset information. The nodes of the substation asset knowledge graph include substation equipment nodes, sub-equipment nodes, procurement nodes, installation nodes, maintenance nodes, and scrapping nodes. Each node includes corresponding node attribute information, and the node attribute information of the sub-equipment nodes includes the failure probability. The edges of the substation asset knowledge graph include six types of edges: type 1 edges connect the substation equipment nodes and the sub-equipment nodes; type 2 edges connect each sub-equipment node and its corresponding procurement node; type 3 edges connect each sub-equipment node and its corresponding procurement node; and type 4 edges connect each sub-equipment node and its corresponding procurement node. The sub-device nodes and their corresponding installation nodes are described. Edges of type 4 connect each sub-device node to its corresponding maintenance node. Edges of type 5 connect each sub-device node to its corresponding scrap node. When a fault in one sub-device affects a fault in another sub-device, the two corresponding sub-device nodes are connected by edges of type 6. Each type of edge includes attribute information corresponding to its type. The attribute information of the edge of type 6 includes fault correlation. Based on the substation asset knowledge graph, asset management costs are determined using a cost prediction model, which is a machine learning model. The update module is used to update the asset management information of the substation equipment based on the basic asset information and the asset management cost; wherein, the asset management information refers to information on the management of assets related to the substation equipment, including equipment type and quantity, equipment status, equipment management strategy, and corresponding asset management cost; The early warning module is used to identify key monitoring equipment and send early warning information to associated user terminals based on the updated asset management information of the substation equipment. The determining module is further configured to update the fault probability, wherein: When a situation arises during the use of the sub-device that affects its quality, a fault probability update is performed. This fault probability update includes: The update begins with the failure probability of the sub-device node whose quality has changed, and then progressively updates the failure probabilities of adjacent nodes; wherein the formula for updating the failure probability of the sub-device node whose quality has changed is: , The updated failure probability of the sub-device node whose quality has changed is represented by s, the unupdated failure probability of the sub-device node whose quality has changed is represented by a, and the conversion baseline value is represented by T. The quality change factor a is determined by a factor model, which is a machine learning model. The input of the factor model is the quality change mode, the sub-device type, the historical change data of the sub-device, and the performance parameters of the sub-device at the time of manufacture. The change mode includes problems, damage, repair, and maintenance. The formula for updating the failure probability of the adjacent nodes is: , The fault probability of the neighboring sub-device node i after the quality change is represented by si, the fault probability of the neighboring sub-device node i before the update is represented by s1 to sn, the fault probabilities corresponding to the n neighboring sub-device nodes of the neighboring sub-device node i are respectively represented by s1 to sn, and the fault correlation degree between the n neighboring sub-device nodes of the neighboring sub-device node i and the neighboring sub-device node i is respectively represented by r1 to rn.

5. The system according to claim 4, wherein the asset management cost includes scrapping cost; and the cost prediction model includes a first prediction model; The determination of asset management costs based on the aforementioned basic asset information includes: Determine whether the sub-equipment of the power equipment is scrapped; In response to the fact that the sub-equipment is scrapped, the cost type of the asset management cost is determined to be the scrap cost, and the scrap cost sub-graph is extracted from the substation asset knowledge graph; The scrap cost is determined by processing the scrap cost subgraph based on the first prediction model; the first prediction model is a graph neural network model.

6. The system according to claim 4, wherein the asset management cost includes loss cost; and the cost prediction model includes a second prediction model; The determination of asset management costs based on the aforementioned basic asset information includes: Determine the risk level of the sub-equipment of the power equipment; In response to the risk level being greater than a preset risk threshold, the cost type of the asset management cost is determined to be the loss cost, and a loss cost subgraph is extracted from the substation asset knowledge graph. The loss cost subgraph is processed based on the second prediction model to determine the loss cost; the second prediction model is a graph neural network model.

7. A substation asset management device, comprising a processor, characterized in that, The processor is used to execute the substation asset management method as described in any one of claims 1 to 3.

8. A computer-readable storage medium storing computer instructions, wherein when a computer reads the computer instructions in the storage medium, the computer executes the substation asset management method as described in any one of claims 1 to 3.

Images