Work ticket generation method and device and storage medium

Abstract

The invention discloses a work ticket generation method and device and a storage medium, and the method comprises the steps: constructing a plurality of preset structural bodies according to an intelligent substation configuration description file, and enabling the plurality of preset structural bodies to be used for representing the related information of conductive equipment, a transformer and a bus; determining an equipment adjacency matrix according to the plurality of preset structural bodies, wherein the adjacency matrix is used for representing a connection relationship between every two primary equipment; constructing a feature vector of each node according to a preset dimension; determining a feature matrix according to the adjacent matrix and the feature vector; the feature matrix is input to a preset network model, a full connection layer of a preset network is connected with a plurality of classifiers, and the classifiers are used for outputting work ticket related information; and determining the work ticket according to the work ticket related information output by the preset network model. The equipment adjacency matrix can be obtained according to the intelligent substation configuration description file, and the work ticket related information can be accurately determined by using the preset network model, so that the work ticket is accurately generated, and the generation efficiency of the work ticket is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power management, and in particular relates to a work ticket generation method, device and storage medium. Background Art

[0002] Substations are key hubs in the modern power grid system, and regular maintenance of equipment is an effective guarantee for the stable operation of the power grid. The electrical work ticket system is an important measure to ensure safe production and power operation safety. Maintenance and testing personnel carry out corresponding work in the station based on the plant and station work tickets. An accurate and complete work ticket is a reliable prerequisite for safe and orderly work.

[0003] Currently, work tickets are generated by manually filling in the form. However, this method requires the on-duty personnel to have certain operating experience, otherwise the problem of inappropriate content of the work ticket is likely to occur. How to quickly and accurately generate work tickets has become an urgent problem to be solved. Summary of the invention

[0004] The purpose of the present invention is to provide a work ticket generation method, device and storage medium, which can improve the generation efficiency of work tickets.

[0005] According to a first aspect of the present invention, there is provided a method for generating a work ticket, comprising:

[0006] Constructing a plurality of preset structures according to the smart substation configuration description file, wherein the plurality of preset structures are used to represent relevant information of the conductive equipment, the transformer and the busbar;

[0007] Determine a device adjacency matrix according to the plurality of preset structures;

[0008] Construct the feature vector of each node according to the preset dimension;

[0009] Determine a feature matrix according to the adjacency matrix and the feature vector;

[0010] The feature matrix is ​​input into a preset network model, wherein the fully connected layer of the preset network is connected to a plurality of classifiers, and the classifiers are used to output work ticket related information;

[0011] The work ticket is determined according to the work ticket related information output by the preset network model.

[0012] According to a second aspect of the present invention, there is provided a work ticket generating device, comprising:

[0013] A parsing module, used to construct a plurality of preset structures according to the smart substation configuration description file, wherein the plurality of preset structures are used to represent relevant information of the conductive equipment, the transformer and the busbar;

[0014] An adjacency matrix determination module, used to determine a device adjacency matrix according to the plurality of preset structures;

[0015] A feature vector determination module, used to construct a feature vector for each node according to a preset dimension;

[0016] A feature matrix determination module, used to determine a feature matrix according to the adjacency matrix and the feature vector;

[0017] A model application module, used for inputting the feature matrix into a preset network model, wherein the fully connected layer of the preset network is connected to a plurality of classifiers, and the classifiers are used for outputting work ticket related information;

[0018] A work ticket determination module is used to determine the work ticket according to the work ticket related information output by the preset network model.

[0019] According to a third aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0020] at least one processor; and

[0021] a memory communicatively connected to the at least one processor; wherein,

[0022] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the work ticket generation method described in any embodiment.

[0023] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to implement the work ticket generation method described in any embodiment of the present invention when executed by a processor.

[0024] The technical solution provided by the present invention is to construct a plurality of preset structures according to the configuration description file of the smart substation, and the plurality of preset structures are used to represent the relevant information of the conductive equipment, transformer and busbar; determine the equipment adjacency matrix according to the plurality of preset structures; construct the feature vector of each node according to the preset dimension; determine the feature matrix according to the adjacency matrix and the feature vector; input the feature matrix into a preset network model, the fully connected layer of the preset network is connected with a plurality of classifiers, and the classifiers are used to output the relevant information of the work ticket; determine the work ticket according to the relevant information of the work ticket output by the preset network model. Compared with the current manual filling of the work ticket, the present invention can obtain the equipment adjacency matrix according to the configuration description file of the smart substation, and then determine the adjacency matrix between the primary equipment in the three-dimensional space that can reflect the three-dimensional space structure; on this basis, the preset network model can be used to accurately determine the relevant information of the work ticket, and then accurately generate the work ticket, thereby improving the efficiency of generating the work ticket.

[0025] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0027] Figure 1 It is a flowchart of a method for generating a work ticket provided by an embodiment of the present invention;

[0028] Figure 2 It is a schematic diagram of the file structure of a smart substation configuration description file provided by an embodiment of the present invention;

[0029] Figure 3 is a schematic diagram of a preset structure provided by an embodiment of the present invention;

[0030] Figure 4 is a schematic diagram of dimensions included in a feature vector provided by an embodiment of the present invention;

[0031] Figure 5 It is a schematic diagram of the graph convolutional network structure provided by an embodiment of the present invention;

[0032] Figure 6 It is a structural schematic diagram of a work ticket generating device provided by an embodiment of the present invention;

[0033] Figure 7 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0034] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0035] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0036] Substations are key hubs in the modern power grid system, and regular maintenance of equipment is an effective guarantee for the stable operation of the power grid. The electrical work ticket system is an important measure to ensure safe production and power operation safety. Maintenance and testing personnel carry out corresponding work in the station based on the plant and station work tickets. An accurate and complete work ticket is a reliable prerequisite for safe and orderly work.

[0037] From the perspective of substation operation, work tickets are generally generated by manually drafting tickets or importing historical tickets. The former requires high operating experience of the operator, while the latter lacks flexibility and has poor portability. In general, these two ticketing methods rely heavily on the professional quality and technical level of the operator, and both have a certain risk of mis-filling.

[0038] Safety measures (safety measures) are part of the work ticket, and work permits, i.e., live parts retained at the work site, are also part of the work ticket. At present, the methods for generating work tickets are mostly based on the perspective of maintenance or relay protection, and are centered around the automatic generation of secondary safety measures, ignoring the work permit and the part of primary safety measures. The inventor found that compared with the secondary part, the actual wiring on site may be different from the drawings, the voltage level is higher, and the risk of accidental contact by personnel is greater, so it is necessary to conduct a detailed analysis. Maintenance of primary equipment such as knife switches and switches is a common maintenance work, and analysis of them has high practical value. It should be noted that due to the numerous and complex work tasks, there may be multiple feasible safety measures at the same time under the company's current work ticket management and execution specifications. The flexibility and coordination of the generation of primary safety measures are large, and it is difficult to describe them with unified rules, so it is necessary to conduct a detailed exploration. In addition, the work permit is inseparable from the control of the current system operation mode, and the previous generation method does not consider the operating status of the system, and the results obtained are relatively one-sided and the accuracy is not high. Therefore, the present invention proposes a method for generating a plant station work ticket, especially for the generation of the first type of plant station work ticket. The specific scheme is as follows.

[0039] Figure 1 1 is a flow chart of a method for generating a work ticket provided by an embodiment of the present invention. This embodiment is applicable to the case of generating a work ticket for a plant or station. The method can be executed by a work ticket generating device, and the work ticket generating device can be implemented in the form of hardware and / or software. The method includes:

[0040] S101. Construct a plurality of preset structures according to a configuration description file of a smart substation, wherein the plurality of preset structures are used to represent relevant information of conductive equipment, transformers and busbars.

[0041] The smart substation configuration description file (system specification description, SSD) describes the substation primary system topology, primary equipment model information, primary equipment and secondary equipment association information, substation application system function configuration, etc. The topology model of the SSD file is stored in the configuration file in the form of a structure tree, such as Figure 2 As shown, the SSD file is described using the substation configuration description language, namely Structured Control Language (SCL).

[0042] The embodiment of the present invention extracts part of the information from the smart substation configuration description file to form multiple preset structures. Optionally, constructing multiple preset structures according to the smart substation configuration description file can be implemented in the following manner:

[0043] The smart substation configuration description file is parsed according to the DOM method to obtain a conductive device structure, a transformer structure and a busbar structure; the conductive device structure includes basic information of the conductive device and the name of the bay to which it belongs, and the transformer structure includes basic information of the transformer and the name of the bay to which it belongs; the busbar structure includes connection points contained in the busbar, relevant logical node information and endpoint information.

[0044] SSD files include header files and substation information. Substation information includes transformers and voltage levels. From the perspective of the structure tree hierarchy, the parsing of SSD files can be divided into three types of bay parsing: bus bay related information parsing, main transformer bay related information parsing, and circuit breaker (outgoing line) bay related information parsing.

[0045] The SSD file data is in an unstructured XML format and has a huge amount of data. If parsed directly, it will consume a lot of memory and time. For the problem of generating work tickets for power plants, the topological structure of the equipment, the correlation information of primary and secondary equipment, and the necessary primary and secondary equipment information are the key contents that the present invention hopes to obtain by parsing the SSD file. Therefore, in order to simplify the information acquisition process of the equipment, reduce the acquisition of weakly related equipment and redundant information, and ensure the input of strongly related equipment and key information, the present invention respectively constructs the transformer PowerTransformer structure, the conducting equipment ConductingEquipment structure, and the bus Bus_Bay structure. The above is the pseudo code of that structure as follows Figure 3 As shown, it is used to save the configuration information of the bus bay, main transformer bay, circuit breaker (outgoing line) bay and their related secondary parts.

[0046] For example, the conductive device structure includes name, description, type; voltage level (such as 500kV, 220kV, 110kV or 10kV, etc.); bay name, bay description; several logical nodes contained in the device; endpoint information contained in the device; conductive device return type. According to the endpoint type contained in the device, it can be determined whether the device contains the end point of a certain logical branch. The logical branch extends from the busbar and ends at the endpoint. The logical branch consists of multiple logical points.

[0047] The DOM (Document Object Model) solution is used to parse the SSD file, and the principle of DOM solution reading and parsing is explained by taking the interval of a conductive device (such as a circuit breaker) as an example.

[0048] First, use the elementsByTagName() interface provided by QDomDocument to read and parse the configuration information of SCL, Substation, VoltageLevel, Voltage, Bay, Connectivity Node, ConductingEquipment, LNode and Terminal objects at each level of the SSD file layer by layer to generate QTreeWidgetItem at each level.

[0049] Secondly, these nodes are parsed into a linked list of QDomElement, and the attribute() interface of QDomElement is used to read the configuration information of each level of the conductive equipment interval and save it into the structure ConductingEquipment.

[0050] Finally, the QTreeWidgetItem corresponding to the ConductingEquipment hierarchy and the ConductingEquipment structure are stored in the mapping table QHash according to the corresponding relationship, thereby establishing a mapping relationship between QTreeWidgetItem and a primary device.

[0051] S102: Determine a device adjacency matrix according to the multiple preset structures.

[0052] The adjacency matrix is ​​used to represent the connection relationship between two primary devices.

[0053] After constructing the above structure, the device adjacency matrix can be determined according to the relevant logical nodes in the structure and the node information contained therein.

[0054] Optionally, determining the device adjacency matrix according to the multiple preset structures may be implemented in the following manner:

[0055] A mapping table for representing device connection relationships is determined based on the bus structure, including connection points, related logical node information and endpoint information included in the bus; and a device adjacency matrix is ​​determined based on the mapping table.

[0056] In the work ticket, when it comes to switch and protection inspection, it is necessary to fill in the information of low voltage and secondary circuit, especially the connector, soft pressure plate, transfer switch, etc. Generally speaking, the above information is strongly related to the equipment to be powered off. When the equipment to be powered off is powered off, some interlocking devices will exit the corresponding function. If manual input is used when establishing the automatic generation feature of the work ticket, the workload is large and it is difficult to ensure the accuracy.

[0057] In the SSD file, the logical node LNode is a node under the primary device object used to associate the primary device with the secondary device and its signals. Therefore, the present invention parses the logical node of the IED device in the SSD file, associates the logical node physical device of the protection device IED with the main device of the interval, obtains the device information such as protection, measurement and control, and merging unit, and forms a complete mapping list QHash containing the association information of the primary and secondary devices of the substation.

[0058] According to the integrated mapping table QHash, the Connectivity Node and Terminal attribute information between each device are obtained for topological modeling and identification. Since grounding wire measures may appear in the subsequent security arrangement, grounding wire device nodes are added on the low-voltage side of the main transformer bay and between the CT and the switch in the 10kV outgoing line bay, which belong to the ES device. In addition, in order to accurately generate a work permit, it is necessary to verify its spatial correspondence after obtaining the topological structure, that is, to verify the accuracy of the adjacent bay information. For the 500kV part, due to its unique 3 / 2 wiring, the accuracy of the adjacent bay information is guaranteed. For the 220kV, 110kV, and 10kV parts, since the Connectivity Node of the busbar side switch points to the same point, it is difficult to clarify its adjacent information. Therefore, when parsing, a virtual identifier is added to the busbar side switch of the outgoing line bay. If there is a third virtual identifier between the two busbar side switches, it is considered that the two outgoing line bays are not adjacent. On the contrary, if there is no other virtual identifier, it is considered adjacent. Based on this, the adjacent intervals are checked according to the actual line laying conditions and the 10kV screen layout diagram. After the above steps, the present invention obtains the connection relationship between each device and forms an adjacency matrix A.

[0059] If the Terminals endpoint information of device i is consistent with the Connectivity Node node information of device j, it means that the two devices have a connection relationship in topology, that is, the two points are adjacent. If the Terminals endpoint information is crossed, there is a connection relationship. For the main transformer, the connection relationship is determined based on its winding information and circuit breaker interval. It is expressed in the device connection relationship matrix as:

[0060]

[0061] S103: construct a feature vector for each node according to a preset dimension.

[0062] Optionally, constructing a feature vector for each point according to a preset dimension can be implemented in the following way:

[0063] The feature vector of each point is determined according to a combination of one or more preset dimensions, wherein the dimensions include: the type of work, the working range, the voltage level, the interval type, whether it is a device to be repaired and maintained, the device type, the main device, whether there is equipment within the safety distance, the device number, the number of protection devices, the type of protection pressure plate, or the type of interactive protection pressure plate.

[0064] Features are the abstraction and understanding of the actual problem by the GCN model. In this problem, the primary wiring diagram is crucial for the automatic generation of work tickets. However, since it cannot reflect three-dimensional spatial information, the effect of machine learning is not ideal. Therefore, the present invention sets the feature quantity of whether there is equipment within the safe distance to characterize the spatial layout of electrical equipment.

[0065] Since the work contents of different work tickets vary greatly, the number of equipment they contain is also different. In order to solve the problem of inconsistent data input dimensions, the present invention designs a feature preprocessing step, and divides the work tickets into three categories based on expert experience: "equipment maintenance", "equipment regular inspection", and "equipment expansion". At the same time, it is determined whether the work involves secondary work, that is, it is divided into three categories: work that only involves one time, work that only involves two times, and work that involves both one and two times, and refines the work ticket type.

[0066] Then, after the SSD file is parsed, the devices in the same interval are classified into a set. If any device in the set appears in the work task, it means that the device in the set needs to be powered off, so the main device in the set is selected, and the main device is used as one of the features of the node when constructing the feature. For the relevant secondary equipment information, three feature vectors are constructed: the number of protection device sets, the type of protection device pressure plate, and the interactive protection device information. Specifically, it is refined into the number of protection device sets, generally 1-5 sets, the type of protection device pressure plate, generally its own failure protection pressure plate, protection jump outlet pressure plate and backup protection jump outlet pressure plate, the type of interactive protection device pressure plate, generally interactive failure joint jump pressure plate, backup protection locking standby self-throwing pressure plate, and the above pressure plates are instantiated and solidified according to the secondary equipment content obtained by parsing.

[0067] Three labels are set based on the input question. Label 1 is the power status of the equipment. Label 2 is the equipment status, specifically the disconnected and closed states. Label 3 is the protection pressure plate information related to the investment and withdrawal, specifically the protection outlet pressure plate that exits itself and interacts, and the pressure plate that does not need to be changed.

[0068] In summary, the present invention combines the input work location, work content, and the topological structure and device information obtained by SSD analysis to construct Figure 4 The feature quantity shown is used to obtain the feature vector X of the point, with a dimension of n×12, where n is the total number of device nodes.

[0069] S104: Determine a feature matrix according to the adjacency matrix and the feature vector.

[0070] Optionally, determining the characteristic matrix according to the adjacency matrix and the characteristic vector can be implemented in the following manner:

[0071] The adjacency matrix is ​​corrected; the eigenvector is multiplied on the left and the weight parameter is multiplied on the right according to the corrected adjacency matrix, the operation result is added to the bias parameter, and the extraction result is obtained through the activation function; the weight parameter and the bias parameter are optimized through back propagation; the eigenvector is multiplied on the left and the weight parameter is multiplied on the right according to the extraction result, the operation result is added to the bias parameter, and the activation function is passed through two layers of graph convolution layers to obtain the feature matrix.

[0072] After obtaining the feature vector of each point, data preprocessing is performed on it. The present invention performs the following data preprocessing, re-encoding the text-type data and converting it into a numerical type.

[0073] The feature vectors of the type of work, the scope of work, the interval type, the equipment category, the type of protective pressure plate, and the interactive type of protective pressure plate are coded with dummy variables. The feature vectors of whether the equipment is working and whether there is equipment within the safety distance are processed using unique hot coding. The voltage level feature vector is coded with serial numbers. The feature vectors of the main equipment and equipment number are standardized, and the distance between the data is maintained while scaling to between 0 and 1.

[0074]

[0075] Among them, x is the original data, μ is the mean of the data set, and δ is the standard deviation of the data set.

[0076] For the adjacency matrix A and feature vector X obtained in the above steps, GCN is used to extract data features.

[0077] First, the adjacency matrix A is modified as follows:

[0078] A′=A+I

[0079]

[0080] Where: I is the identity matrix; D is the degree matrix, which is a diagonal matrix composed of the degree d(vi) of each node, and its value is equal to the sum of the corresponding rows or columns of the node in the adjacency matrix A.

[0081]

[0082] Further, according to the following steps, the modified adjacency matrix A″ is multiplied by the input matrix X on the left, and then multiplied by the weight parameter W on the right and added to the bias parameter b, and the feature extraction result H2 is obtained through the activation function. The specific formula is as follows:

[0083] H (2) =f Relu (A″H (1) W (1) +b (1) )

[0084] Among them, H (1) =X

[0085] The weight parameter W and bias parameter b in the above formula are model parameters that need to be obtained by general back propagation. The subscripts (1) and (2) in the above formula are used to distinguish the values ​​of H, W, and b after each step.

[0086] H (3) =f ReLU (A″H (2) W (2) +b (2) )

[0087] Finally, the feature matrix H is obtained through the second layer of graph convolution (3) The graph convolution structure is as follows Figure 5 As shown, the above formula corresponds to the two GCN layers and ReLu layers of the graph convolutional network.

[0088] S105, inputting the feature matrix into a preset network model, wherein the fully connected layer of the preset network is connected to a plurality of classifiers, and the classifiers are used to output work ticket related information.

[0089] Optionally, inputting the feature matrix into a preset network model may be implemented in the following manner:

[0090] The feature matrix is ​​input into a graph convolutional neural network, which includes two graph convolutional layers, a graph pooling layer, a fully connected layer and a classifier layer; the classifier layer includes a first classifier, a second classifier and a third classifier, the first classifier is used to determine whether the device is disconnected or closed according to the device status; the second classifier is used to determine whether the device is energized according to the power supply condition of the device; the third classifier is used to classify the corresponding exited pressure plate.

[0091] After the feature extraction of the GCN network, the graph pooling layer and the fully connected layer are input, and finally the classifier is connected to obtain the final output result. The GCN structure is as follows: Figure 5 shown.

[0092] In order to accurately obtain the work ticket and permission, the model of the present invention sets three classifiers. The first classifier (classifier 1) determines the state of the equipment and classifies whether the equipment should be disconnected or closed. The second classifier (classifier 2) determines the power supply of the equipment and classifies whether the equipment is powered. The third classifier (classifier 3) classifies the corresponding exiting press plate.

[0093] Furthermore, before inputting the feature matrix into the graph convolutional neural network, it also includes:

[0094] Get the precision and total regression rate of the graph convolutional neural network;

[0095] Determine a micro function result according to the precision rate and the total regression rate;

[0096] Evaluating the graph convolutional neural network according to the microscopic function result;

[0097] Accordingly, the feature matrix is ​​input into the graph convolutional neural network, including:

[0098] The feature matrix is ​​input into the graph convolutional neural network for evaluation success.

[0099] Classifiers 1, 2, and 3 use the sigmoid function to map the intermediate vector to the interval (0, 1), and finally output the prediction vector Y. The micro-F1 function is used to evaluate the work ticket generation model, and the calculation formula is as follows:

[0100]

[0101] In the formula, p c is the overall accuracy, r c is the total recall rate. TP i The positive sample of the i-th class is predicted as a positive sample, FP i The i-th class of negative samples is mistakenly predicted as positive samples. i The positive sample of the i-th class is predicted as a negative sample.

[0102] S106. Determine the work ticket according to the work ticket related information output by the preset network model.

[0103] Use the conditional judgment IF-THEN logic to perform safety checks. During operation, the maintenance equipment has an obvious disconnection point. If multiple related equipment are maintained at the same time, there should also be corresponding disconnection points. This disconnection point is called an island, and it is also necessary to ensure that there must be a grounding point inside the physical disconnection point of each branch in the island.

[0104] If (if): the maintenance equipment is a primary equipment and the maintenance equipment branch is grounded or the maintenance equipment branch is grounded. Then (Then): the grounding safety check is passed

[0105] If (if): the number of the knife switches pulled open is consistent with the number of the knife switch motor power supply and control power supply disconnected, and the number of the disconnected switches is consistent with the number of the disconnected switch energy storage motor power supply air switches. Then (Then): the disconnected air switch verification is passed.

[0106] If (if): Line bay and line TYD secondary voltage air switch should be disconnected. Then (Then): Anti-reverse charging check passed.

[0107] The above is the verification of safety measures. For the circuit breaker (switch) number that should be disconnected, the isolating switch (knife switch) number that should be opened, the grounding knife switch number that should be closed or the installed grounding wire that is obtained through the verification, the corresponding AC and DC power disconnection conditions are supplemented. For the knife switch, disconnect its knife switch control power supply and motor power supply air switch. For the switch, switch the switch control mode conversion switch to the "local" position, and disconnect the switch control power supply and energy storage power supply air switch.

[0108] Hang signboards at the switch and knife switch operating handles, and set signboards at the back machine switch and knife switch respectively. If the ground knife on the line side is in the closed state, hang a signboard that says "Do not close the switch, someone is working on the line!" At the same time, hang an additional signboard at the line TYD secondary side voltage air switch. Otherwise, hang a signboard that says "Do not close the switch, someone is working!"

[0109] Hang a "Enter and Exit Here!" sign at the entrance and exit of the fence, and hang a "Work Here" sign at the work site.

[0110] For the work permit part, based on the GCN network's classification of the equipment's energized status, the energized busbars, conductors and switches at the work site are output. In addition, based on the verification of adjacent intervals in step one, the energized status of adjacent intervals is output.

[0111] The work ticket generation method provided by the present invention constructs multiple preset structures according to the configuration description file of the smart substation, and the multiple preset structures are used to represent the relevant information of the conductive equipment, transformer and busbar; determines the device adjacency matrix according to the multiple preset structures, and the adjacency matrix is ​​used to represent the connection relationship between the primary equipment in pairs; constructs the feature vector of each node according to the preset dimension; determines the feature matrix according to the adjacency matrix and the feature vector; inputs the feature matrix into a preset network model, and the fully connected layer of the preset network is connected with multiple classifiers, and the classifier is used to output the relevant information of the work ticket; determines the work ticket according to the relevant information of the work ticket output by the preset network model. Compared with the current manual filling of the work ticket, the present invention can obtain the device adjacency matrix according to the configuration description file of the smart substation, and then determine the adjacency matrix between the primary equipment in pairs in the three-dimensional space that can reflect the three-dimensional space structure; on this basis, the preset network model can be used to accurately determine the relevant information of the work ticket, and then accurately generate the work ticket, thereby improving the generation efficiency of the work ticket.

[0112] The present invention obtains the equipment information and topological structure of the current substation by parsing the smart substation configuration description (SSD) file. When expansion, equipment update, or transformation occurs, the latest topological connection relationship can be obtained by reloading the SSD file. The previous SAX parsing method has problems such as excessive parsing burden and equipment redundancy. For the problem of automatically generating work tickets, the present invention uses the DOM parsing method and merges the equipotential points and nodes that do not affect the topological connection relationship to improve the parsing efficiency.

[0113] The three-dimensional spatial information cannot be reflected by the two-dimensional topological connection relationship. The present invention uses the spatial information of each node as a feature quantity, records and transmits the spatial information of each node and adjacent nodes through a graph convolutional neural network (GCN), and obtains a three-dimensional spatial information input.

[0114] A single primary wiring diagram cannot reflect the diverse actual situations. The present invention uses the work type, working scope, whether there is equipment within the safe distance, etc. as the feature quantities of the graph convolutional neural network, designs a GCN suitable for this problem, and obtains the work ticket arrangement under different working conditions by inputting the feature quantities of the above different samples.

[0115] For the pressure plate insertion and withdrawal problem, based on the logical node LNode, the logical node physical device of the IED is associated with the main device of the interval, and a complete primary and secondary device mapping linked list QHash is formed to obtain the pressure plate information.

[0116] Based on the on-site work experience, several safety measures generation and verification logics that are not considered by the existing generation methods are supplemented to further generate a complete work ticket.

[0117] Figure 6A structural schematic diagram of a work ticket generating device provided in an embodiment of the present invention, the work ticket generating device is used to implement the above-mentioned work ticket generating method, including: a parsing module 21, an adjacency matrix determination module 22, a eigenvector determination module 23, a eigenmatrix determination module 24, a model application module 25 and a work ticket determination module 26.

[0118] The parsing module 21 is used to construct a plurality of preset structures according to the configuration description file of the smart substation, wherein the plurality of preset structures are used to represent relevant information of the conductive equipment, the transformer and the busbar;

[0119] An adjacency matrix determination module 22, configured to determine a device adjacency matrix according to the plurality of preset structures;

[0120] A feature vector determination module 23, used to construct a feature vector for each node according to a preset dimension;

[0121] A feature matrix determination module 24, configured to determine a feature matrix according to the adjacency matrix and the feature vector;

[0122] A model application module 25, used for inputting the feature matrix into a preset network model, wherein the fully connected layer of the preset network is connected to a plurality of classifiers, and the classifiers are used for outputting work ticket related information;

[0123] The work ticket determination module 26 is used to determine the work ticket according to the work ticket related information output by the preset network model.

[0124] Based on the above implementation, optionally, the parsing module 21 is used to:

[0125] The smart substation configuration description file is parsed according to the DOM method to obtain a conductive device structure, a transformer structure and a busbar structure; the conductive device structure includes basic information of the conductive device and the name of the bay to which it belongs, and the transformer structure includes basic information of the transformer and the name of the bay to which it belongs; the busbar structure includes connection points contained in the busbar, relevant logical node information and endpoint information.

[0126] Based on the above implementation, optionally, the adjacency matrix determination module 22 is used to:

[0127] Determine a mapping table for representing a connection relationship of devices according to the bus structure including connection points, related logical node information and endpoint information included in the bus;

[0128] A device adjacency matrix is ​​determined according to the mapping table.

[0129] Based on the above implementation, optionally, the feature vector determination module 23 is used to:

[0130] The feature vector of each point is determined according to a combination of one or more preset dimensions, wherein the dimensions include: the type of work, the working range, the voltage level, the interval type, whether it is a device to be repaired and maintained, the device type, the main device, whether there is equipment within the safety distance, the device number, the number of protection devices, the type of protection pressure plate, or the type of interactive protection pressure plate.

[0131] Based on the above implementation, optionally, the model application module 25 is used to:

[0132] The feature matrix is ​​input into a graph convolutional neural network, which includes two graph convolutional layers, a graph pooling layer, a fully connected layer and a classifier layer; the classifier layer includes a first classifier, a second classifier and a third classifier, the first classifier is used to determine whether the device is disconnected or closed according to the device status; the second classifier is used to determine whether the device is energized according to the power supply condition of the device; the third classifier is used to classify the corresponding exited pressure plate.

[0133] Based on the above implementation, optionally, the feature matrix determination module 24 is used to:

[0134] Modify the adjacency matrix;

[0135] According to the modified adjacency matrix, the eigenvector is multiplied on the left and the weight parameter is multiplied on the right, the operation result is added to the bias parameter, and the extraction result is obtained through the activation function;

[0136] The weight parameters and bias parameters are optimized through back propagation; according to the extraction result, the feature vector is multiplied on the left and the weight parameter is multiplied on the right, the operation result is added to the bias parameter, and the feature matrix is ​​obtained through the activation function and two layers of graph convolution layers.

[0137] Based on the above implementation, optionally, a model verification module is further included, which is used to:

[0138] Get the precision and total regression rate of the graph convolutional neural network;

[0139] Determine a micro function result according to the precision rate and the total regression rate;

[0140] Evaluating the graph convolutional neural network according to the microscopic function result;

[0141] Accordingly, the feature matrix is ​​input into the graph convolutional neural network, including:

[0142] The feature matrix is ​​input into the graph convolutional neural network for evaluation success.

[0143] The work ticket generating device provided by the present invention comprises a parsing module 21, which is used to construct a plurality of preset structures according to a configuration description file of an intelligent substation, wherein the plurality of preset structures are used to represent relevant information of conductive equipment, transformers and busbars; an adjacency matrix determining module 22, which is used to determine an equipment adjacency matrix according to the plurality of preset structures, wherein the adjacency matrix is ​​used to represent the connection relationship between each pair of primary equipment; a feature vector determining module 23, which is used to construct a feature vector of each node according to a preset dimension; a feature matrix determining module 24, which is used to determine a feature matrix according to the adjacency matrix and the feature vector; a model application module 25, which is used to input the feature matrix into a preset network model, wherein the fully connected layer of the preset network is connected to a plurality of classifiers, and the classifiers are used to output relevant information of the work ticket; and a work ticket determining module 26, which is used to determine the work ticket according to the relevant information of the work ticket output by the preset network model. Compared with the current manual filling out of work tickets, the present invention can obtain the equipment adjacency matrix based on the smart substation configuration description file, and then determine the adjacency matrix between the primary devices in the three-dimensional space that can reflect the three-dimensional spatial structure; on this basis, the preset network model can be used to accurately determine the relevant information of the work ticket, and then accurately generate the work ticket, thereby improving the efficiency of work ticket generation.

[0144] The work ticket generating device provided in the embodiment of the present invention can execute the work ticket generating method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0145] Figure 7 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0146] like Figure 7As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0147] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0148] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a work ticket generation method.

[0149] In some embodiments, the work ticket generation method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the work ticket generation method described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to execute the work ticket generation method by any other appropriate means (for example, by means of firmware).

[0150] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0151] The computer program for implementing the work ticket generation method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0152] The embodiment of the present invention further provides a computer-readable storage medium, which stores computer instructions. The computer instructions can be used to enable a processor to execute a work ticket generation method, the method comprising:

[0153] Constructing a plurality of preset structures according to the smart substation configuration description file, wherein the plurality of preset structures are used to represent relevant information of the conductive equipment, the transformer and the busbar;

[0154] Determine a device adjacency matrix according to the plurality of preset structures;

[0155] Construct the feature vector of each node according to the preset dimension;

[0156] Determine a feature matrix according to the adjacency matrix and the feature vector;

[0157] The feature matrix is ​​input into a preset network model, wherein the fully connected layer of the preset network is connected to a plurality of classifiers, and the classifiers are used to output work ticket related information;

[0158] The work ticket is determined according to the work ticket related information output by the preset network model.

[0159] On the basis of the above implementation, optionally, multiple preset structures are constructed according to the smart substation configuration description file, including:

[0160] The smart substation configuration description file is parsed according to the DOM method to obtain a conductive device structure, a transformer structure and a busbar structure; the conductive device structure includes basic information of the conductive device and the name of the bay to which it belongs, and the transformer structure includes basic information of the transformer and the name of the bay to which it belongs; the busbar structure includes connection points contained in the busbar, relevant logical node information and endpoint information.

[0161] On the basis of the above implementation, optionally, determining a device adjacency matrix according to the plurality of preset structures includes:

[0162] Determine a mapping table for representing a connection relationship of devices according to the bus structure including connection points, related logical node information and endpoint information included in the bus;

[0163] A device adjacency matrix is ​​determined according to the mapping table.

[0164] On the basis of the above implementation, optionally, constructing a feature vector of each point according to a preset dimension includes:

[0165] The feature vector of each point is determined according to a combination of one or more preset dimensions, wherein the dimensions include: the type of work, the working range, the voltage level, the interval type, whether it is a device to be repaired and maintained, the device type, the main device, whether there is equipment within the safety distance, the device number, the number of protection devices, the type of protection pressure plate, or the type of interactive protection pressure plate.

[0166] On the basis of the above implementation, optionally, inputting the feature matrix into a preset network model includes:

[0167] The feature matrix is ​​input into a graph convolutional neural network, which includes two graph convolutional layers, a graph pooling layer, a fully connected layer and a classifier layer; the classifier layer includes a first classifier, a second classifier and a third classifier, the first classifier is used to determine whether the device is disconnected or closed according to the device status; the second classifier is used to determine whether the device is energized according to the power supply condition of the device; the third classifier is used to classify the corresponding exited pressure plate.

[0168] On the basis of the above implementation, optionally, determining a feature matrix according to the adjacency matrix and the feature vector includes:

[0169] Modify the adjacency matrix;

[0170] According to the modified adjacency matrix, the eigenvector is multiplied on the left and the weight parameter is multiplied on the right, the operation result is added to the bias parameter, and the extraction result is obtained through the activation function;

[0171] The weight parameters and bias parameters are optimized through back propagation; according to the extraction result, the feature vector is multiplied on the left and the weight parameter is multiplied on the right, the operation result is added to the bias parameter, and the feature matrix is ​​obtained through the activation function and two layers of graph convolution layers.

[0172] Based on the above implementation, optionally, before the feature matrix is ​​input into the graph convolutional neural network, the method further includes:

[0173] Get the precision and total regression rate of the graph convolutional neural network;

[0174] Determine a micro function result according to the precision rate and the total regression rate;

[0175] Evaluating the graph convolutional neural network according to the microscopic function result;

[0176] Accordingly, the feature matrix is ​​input into the graph convolutional neural network, including:

[0177] The feature matrix is ​​input into the graph convolutional neural network for evaluation success.

[0178] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0179] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0180] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0181] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0182] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0183] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for generating a work ticket, characterized in that: include: Constructing a plurality of preset structures according to the smart substation configuration description file, wherein the plurality of preset structures are used to represent relevant information of the conductive equipment, the transformer and the busbar; Determine a device adjacency matrix according to the plurality of preset structures; Construct the feature vector of each node according to the preset dimension; Determine a feature matrix according to the adjacency matrix and the feature vector; The feature matrix is ​​input into a preset network model, wherein the fully connected layer of the preset network is connected to a plurality of classifiers, and the classifiers are used to output work ticket related information; The work ticket is determined according to the work ticket related information output by the preset network model.

2. The method according to claim 1, characterized in that: Construct multiple preset structures according to the smart substation configuration description file, including: The smart substation configuration description file is parsed according to the DOM method to obtain a conductive device structure, a transformer structure and a busbar structure; the conductive device structure includes basic information of the conductive device and the name of the bay to which it belongs, and the transformer structure includes basic information of the transformer and the name of the bay to which it belongs; the busbar structure includes connection points, related logical node information and endpoint information contained in the busbar.

3. The method according to claim 2, characterized in that Determining a device adjacency matrix according to the plurality of preset structures includes: Determine a mapping table for representing a connection relationship of devices according to the bus structure including connection points, related logical node information and endpoint information included in the bus; A device adjacency matrix is ​​determined according to the mapping table.

4. The method according to claim 1, characterized in that: Construct the feature vector of each point according to the preset dimensions, including: The feature vector of each point is determined according to a combination of one or more preset dimensions, wherein the dimensions include: the type of work, the working range, the voltage level, the interval type, whether it is a device to be repaired and maintained, the device type, the main device, whether there is equipment within the safety distance, the device number, the number of protection devices, the type of protection pressure plate, or the type of interactive protection pressure plate.

5. The method according to claim 1, characterized in that Inputting the feature matrix into a preset network model includes: The feature matrix is ​​input into a graph convolutional neural network, which includes two graph convolutional layers, a graph pooling layer, a fully connected layer and a classifier layer; the classifier layer includes a first classifier, a second classifier and a third classifier, the first classifier is used to determine whether the device is disconnected or closed according to the device status; the second classifier is used to determine whether the device is energized according to the power supply condition of the device; the third classifier is used to classify the corresponding exited pressure plate.

6. The method according to claim 5, characterized in that Determining a feature matrix according to the adjacency matrix and the feature vector includes: Modify the adjacency matrix; According to the modified adjacency matrix, the eigenvector is multiplied on the left and the weight parameter is multiplied on the right, the operation result is added to the bias parameter, and the extraction result is obtained through the activation function; The weight parameters and bias parameters are optimized through back propagation; according to the extraction result, the feature vector is multiplied on the left and the weight parameter is multiplied on the right, the operation result is added to the bias parameter, and the feature matrix is ​​obtained through the activation function and two layers of graph convolution layers.

7. The method according to claim 5, characterized in that Before inputting the feature matrix into the graph convolutional neural network, it also includes: Get the precision and total regression rate of the graph convolutional neural network; Determine a micro function result according to the precision rate and the total regression rate; Evaluating the graph convolutional neural network according to the microscopic function result; Accordingly, the feature matrix is ​​input into the graph convolutional neural network, including: The feature matrix is ​​input into the graph convolutional neural network for evaluation success.

8. A work ticket generating device, characterized in that: include: A parsing module, used to construct a plurality of preset structures according to the smart substation configuration description file, wherein the plurality of preset structures are used to represent relevant information of the conductive equipment, the transformer and the busbar; An adjacency matrix determination module, used to determine a device adjacency matrix according to the plurality of preset structures; A feature vector determination module, used to construct a feature vector for each node according to a preset dimension; A feature matrix determination module, used to determine a feature matrix according to the adjacency matrix and the feature vector; A model application module, used for inputting the feature matrix into a preset network model, wherein the fully connected layer of the preset network is connected to a plurality of classifiers, and the classifiers are used for outputting work ticket related information; A work ticket determination module is used to determine the work ticket according to the work ticket related information output by the preset network model.

9. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the work ticket generating method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the work ticket generation method according to any one of claims 1 to 7 when executed.

Images