Operation ticket reasoning system based on graphic topology

Abstract

The invention discloses a graphic topology-based operation order inference system, which comprises a database, a knowledge base, a topological graph platform and an inference engine, and is characterized in that the topological graph platform is connected with a real-time state of a power grid graph and equipment from the database, and the inference engine uses topological analysis and anti-error verification of the topological graph platform to infer the real-time state of the equipment; reasoning is carried out according to rules called from the knowledge base according to the operation tasks, and an operation order is automatically generated and displayed through a man-machine interface end. According to the method, a dispatcher does not need to manually input the operation ticket, the burden of the dispatcher is reduced to a great extent, the correctness of the operation ticket is also improved, and by introducing a graphic topology analysis function, the expression ability of the rule and the universality of the model are improved, and the reliability and efficiency of invoicing are improved.

Description

Technical Field

[0001] The present invention relates to the field of power systems, and particularly to an operation ticket inference system based on graphic topology. Background Art

[0002] As a basic industry of the national economy and a guarantee for the country's economic development, the safe and stable operation of the power system is becoming increasingly important. Ensuring the safe operation of the power grid has become the top priority of power grid dispatching. With the rapid development of power construction and the improvement of automation level, dispatching operations are becoming increasingly complex, the number of operation steps in dispatching operations is increasing continuously, and the number of operation tickets that dispatchers need to formulate daily is increasing day by day. Considering the factors of safe operation, the work intensity of dispatchers has been increased. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide an operation ticket inference system based on graphic topology that can automatically generate operation tickets according to a graphic topology diagram, reducing the work intensity of dispatchers.

[0004] To solve the above technical problem, a technical solution provided by the present invention is: An operation ticket inference system based on graphic topology, including a database, a knowledge base, a topology graphic platform, and an inference engine. The topology graphic platform docks the power grid graphics and the real-time status of devices from the database. The inference engine uses the topology analysis and error prevention verification of the topology graphic platform to perform inference according to the rules retrieved from the knowledge base based on the operation task, automatically generates an operation ticket, and displays it through a human-machine interface terminal.

[0005] Further, the database includes a power grid graphic database, a device and device parameter database, an operation ticket database, and a typical ticket database. The database is connected to the ESM system in the power grid to synchronize its internal storage. The database serves as the data storage center of the system, used to store knowledge, initial data, and intermediate information generated during the inference process. The database provides the necessary data for the inference engine, including raw data, process data, and final conclusions. Further, the knowledge representation in the knowledge base, that is, the rule, as an important module for the operation ticket system to achieve automatic generation, mainly uses the most reasonable form to describe the problem knowledge. The knowledge representation includes production representation method, predicate logic representation method, frame representation method, and object-oriented representation method.

[0006] Production representation: Production rules are mainly based on symbolic factual knowledge to form a production system. Its characteristics include a fixed expression format, a relatively simple formation method, and no connection between rules. This makes it easier to establish a knowledge base, more convenient to process, with a simple reasoning method and no complex calculations. According to the composition structure of the knowledge base and the inference engine, this structure makes it more convenient to modify the knowledge base without modifying the source program and easier to explain the inference path of the system.

[0007] Predicate logic representation: Predicate logic is an important method for knowledge representation. By introducing predicate logic, the knowledge described in natural language is expressed in the form of function formulas.

[0008] Frame representation: A frame is a data structure that represents fixed object knowledge. Combining the experience in real life, using the similar knowledge in memory as the basic frame, and comparing it with new things, filling in or modifying the details of each other according to the actual situation, thus forming the understanding of the observed things and also supplementing the frame in memory. Its characteristic is that it forms a hierarchy, can integrate descriptive knowledge and procedural knowledge to form a frame network, and is more suitable for representing typical concepts and more complex knowledge content.

[0009] Object - oriented representation: This method systematically introduces the information hiding and data type abstraction of object - oriented programming languages. It takes objects as the basic carriers and regards such complex carriers as the collection of one or more simple carriers. Since an object contains an identifier, a data structure, a set of steps, and a message interface, it has the characteristics of inheritance and encapsulation.

[0010] The topological graph platform is used for the topological analysis of graphs, can realize the drawing of the main wiring diagram of the power system, and provides perfect interfaces to dock with the graphs and real - time status in the database. It is a power grid graph drawing and management platform built based on the integrated technology of graph - database and topological modeling technology, provides strong support for the generation of operation tickets. At the same time, each electrical device on the graph platform fully simulates the physical properties of the devices in reality. Drawing is equivalent to modeling, automatically processes the five - prevention rules, and then automatically adds five - prevention verification when issuing tickets, further improving the intelligence and security of the operation tickets. The inference engine can, based on the currently known facts, utilize the knowledge in the knowledge base, according to certain inference bases, by calling the rules and knowledge in the knowledge base, execute the corresponding operation tasks to obtain the answer to the problem or prove the correctness of a certain hypothesis.

[0011] Furthermore, the knowledge base is only associated with the topological graph through the inference engine. The knowledge base contains some term bases and ticket issuing logic sequences. After receiving an operation task, the inference engine analyzes the wiring method, voltage level, etc. to obtain an operation sequence, and finally decomposes it into operation terms, thereby generating an operation ticket. In this process, the inference engine interacts with the knowledge base and the topological graph platform according to the description of the task, infers the correct operation logic, and generates an operation ticket that meets the user's requirements.

[0012] The basic idea of forward reasoning of the inference engine is: starting from the existing information, searching for available knowledge, selecting and enabling knowledge through conflict resolution, executing the knowledge, changing the solution state, and gradually solving until the problem is solved. Its working process is: the user first stores the information related to the inference in the database. The inference engine selects appropriate rules from the knowledge base according to this information. For a certain rule to hold, all the premise facts of the rule must match successfully. Once it is found that as long as one fact premise of a certain rule does not hold, this rule can be skipped and the next rule can be checked for matching. If all the fact premises of a certain rule match successfully, that is, the rule holds, then the conclusion of the rule is stored in the database, and at the same time, the "usage flag" field of the rule is set to the used state, and this rule does not need to be tested again in the subsequent inference loop process.

[0013] There are two termination conditions for forward reasoning: one is to terminate when a solution that meets the conditions is found; the other is to terminate after all solutions are found.

[0014] The inference algorithm in the inference engine is the Rete algorithm, and its analysis steps are as follows: (1) Input the original data into the working memory; (2) Compare the data with the rules and select the rules that match successfully; (3) If there are conflicts during the rule execution process, multiple rules are activated simultaneously, and the conflicting rules are stored in the conflict set; (4) Resolve the conflicts and put the activated rules into the agenda in the established order; (5) Execute the rules stored in the agenda; (6) Repeat steps 2 to 5 until all the rules in the agenda are executed.

[0015] According to the objects existing in the working memory, the rule engine mechanism in the inference engine uniformly allocates these objects according to the requirements of each rule, puts the conflicting rules into the conflict set for centralized processing, and at the same time activates other rules that meet the conditions; in the conflict set, the rules are sorted according to the priority to generate an agenda, so that the rules with higher priority are triggered first; since the execution part of the rule will affect the object information loaded in the memory, this will change the preconditions of the triggered rules in the agenda, and may withdraw the triggered rules from the agenda; the rule engine executes the rules completely in the order of the queue during the execution process. Once the execution part of the rule affects the data objects in the working memory, then some rule execution instances in the queue will be withdrawn from the queue because the conditions have changed. At the same time, this part of the executed rules may also activate some rules that do not meet the conditions, generating new rule execution instances to enter the queue. Therefore, a dynamic rule execution chain will be formed, and finally a rule inference engine will be formed.

[0016] The beneficial effects of the present invention are as follows: The present invention does not require dispatchers to manually input operation tickets, which greatly reduces the burden on dispatchers, improves the correctness of operation tickets, and improves the expressiveness of rules and the generality of models by introducing graphic topology analysis functions, and improves the reliability and efficiency of ticket writing.

[0017] In order to make the above and other objects, features and advantages of the present invention more obvious and understandable, the following specific preferred embodiments are given, and in conjunction with the accompanying drawings, the detailed description is as follows. Brief Description of the Drawings

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only partial schematic diagrams of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained according to these drawings.

[0019] Figure 1 It is a connection block diagram of an operation ticket inference system based on graphic topology of the present invention; Figure 2 The double busbar plus bypass busbar diagram in the present invention; Figure 3 It is a conversion diagram of four operating states of the switch in the present invention. Detailed Embodiments

[0020] Embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not used to limit the protection scope of the present invention.

[0021] In the present invention, unless otherwise clearly defined and limited, terms such as "installation", "setting", "connection", "fixation", "swivel connection", etc. shall be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or integrated; it can be a mechanical connection, a direct connection, or an indirect connection through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0022] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0023] Embodiment: As Figures 1-3 shown, An operation ticket reasoning system based on graphic topology includes a database, a knowledge base, a topology graphic platform, and an inference engine.

[0024] The database includes a power grid graphic database, a device and device parameter database, an operation ticket database, and a typical ticket database. The database is connected to the ESM system in the power grid to synchronize its internal storage. The database serves as the data storage center of the system, used to store knowledge, initial data, and intermediate information generated during the reasoning process. The database provides the necessary data for the inference engine, including raw data, process data, and final conclusions.

[0025] The knowledge representation in the knowledge base is the rule, which is an important module for the operation ticket system to achieve automatic generation. It mainly uses a most reasonable form to describe the problem knowledge.

[0026] The topology graphic platform is used for graphic topology analysis, can realize the drawing of the main wiring diagram of the power system, and provides perfect interfaces to dock with the graphics and real-time status in the database. It is a power grid graphic drawing and management platform built based on the technology of integrating graphic and data and topology modeling technology, providing strong support for the generation of operation tickets. At the same time, each electrical device on the graphic platform fully simulates the physical properties of the devices in reality. Drawing is modeling, automatically processing the five-prevention rules, so that the five-prevention verification is automatically added when issuing tickets, further improving the intelligence and security of the operation tickets.

[0027] The inference engine can, based on the currently known facts, utilize the knowledge in the knowledge base, according to certain inference bases, by invoking the rules and knowledge in the knowledge base, execute the corresponding operation tasks, obtain the answers to the questions or prove the correctness of a certain hypothesis.

[0028] Only the inference engine is associated between the knowledge base and the topological graph. The knowledge base contains some term bases and ticket issuing logic sequences. After receiving the operation tasks, the inference engine analyzes the wiring mode, voltage level, etc., obtains the operation sequence, and finally decomposes it into operation terms, thereby generating an operation ticket. In this process, the inference engine will interact with the knowledge base and the topological graph platform according to the description of the task, infer the correct operation logic, and generate an operation ticket that meets the user's needs.

[0029] The wiring mode, that is, the electrical main wiring, refers to the circuit for transmitting electric energy that is designed to meet the requirements of predetermined power transmission and operation, etc. in power plants, substations, and power systems, and shows the interconnection relationship between high-voltage electrical equipment.

[0030] The basic idea of the forward inference of the inference engine is: starting from the existing information, looking for available knowledge, selecting and enabling knowledge through conflict resolution, executing the knowledge, changing the solution state, and gradually solving until the problem is solved. Its working process is: the user first stores the information related to the inference in the database. The inference engine selects appropriate rules from the knowledge base according to this information. If a certain rule is to hold, all the premise facts of this rule must match successfully. Once it is found that as long as one fact premise of a certain rule does not hold, this rule can be skipped and the next rule can be checked for matching. If all the fact premises of a certain rule match successfully, that is, when this rule holds, the conclusion of this rule is stored in the database, and at the same time, the "usage flag" field of this rule is set to the used state, and this rule does not need to be tested again in the subsequent inference loop process.

[0031] The inference algorithm in the inference engine is the Rete algorithm, and its analysis steps are as follows: (1) Input the original data into the working memory; (2) Compare the data with the rules and select the rules that match successfully; (3) If there are conflicts during the rule execution process, multiple rules are activated simultaneously, and the conflicting rules are stored in the conflict set; (4) Resolve the conflicts and put the activated rules into the agenda in a predetermined order; (5) Execute the rules stored in the agenda; (6) Repeat steps 2 to 5 until all the rules in the agenda are executed.

[0032] Based on the objects existing in the working memory, the rule engine mechanism in the inference engine uniformly allocates these objects according to the requirements of each rule, puts the conflicting rules into the conflict set for centralized processing, and at the same time triggers other rules that meet the conditions; in the conflict set, the rules are sorted according to the priority to generate an agenda, so that the rules with higher priority are triggered first; since the execution part of the rule will affect the object information loaded in the memory, this will change the preconditions of the triggered rules in the agenda, and may withdraw the triggered rules from the agenda; the rule engine executes the rules completely in the order of the queue. Once the execution part of the rule affects the data objects in the working memory, then some rule execution instances in the queue will be withdrawn from the queue because the conditions have changed. At the same time, this part of the executed rules may also activate some rules that do not meet the conditions, generating new rule execution instances to enter the queue. Therefore, a dynamic rule execution chain will be formed, and finally a rule inference engine will be formed.

[0033] "Running", "hot standby", "cold standby" and "maintenance" are the four most commonly used operating states of equipment; among them, "running" means that the equipment or electrical system is energized and has its own functions; "hot standby" means that the equipment, under the condition of meeting the operating requirements, can be converted to the running state as long as it is switched on once; "cold standby" means that both sides connected to the equipment are in a state without protection measures and have obvious disconnection points on both sides connected to the equipment; "maintenance" means that both sides connected to the equipment have obvious disconnection points, and at the same time, the equipment has grounding wires installed at the disconnection points according to the work requirements. The conversion of the four operating states is Figure 3 as shown.

[0034] For example, assume that the wiring method of a certain line of the invoicing equipment at a certain voltage level is double busbar plus bypass busbar, and the operation task is to transfer the XXI line from running to maintenance. The XXI line exists on both the A substation side and the B substation side, one is an outgoing line and the other is an incoming line; as Figure 3 shown: In the first step, the system first searches in the database according to the wiring method to find a matching wiring diagram.

[0035] In the second step, among the found diagrams, for the XXI line to be invoiced with the operation task of transferring from running to maintenance, the corresponding rule information is searched in the knowledge base to obtain the rule number.

[0036] In the third step, according to the rule number, the invoicing rule operation item table is searched to obtain a set of knowledge records of operation items and save them in the cache.

[0037] In the fourth step, it is judged whether the knowledge has been completely parsed. If so, it indicates that the inference process has been completed and the inference ends; if not, the inference continues.

[0038] Step 5: Parse the knowledge based on the known device information, obtain the solution of the current knowledge, and store it in the result queue.

[0039] Step 6: Solve all the knowledge iteratively until the solution in Step 4 is completed.

[0040] Step 7: Display the data in the result queue on the display screen of the human-machine interface terminal to complete the automatic invoicing process.

[0041] When the system performs error prevention verification, it uses the same reasoning mechanism. The difference is that when an error occurs during the solution of knowledge, the explanation module searches for corresponding explanation information in the system expert library and displays it to the user, and then terminates the reasoning process.

[0042] The process of the above embodiment is as follows: The topology graph platform docks the power grid graph and the real-time status of devices from the database. The inference engine uses the topology analysis and error prevention verification of the topology graph platform, and performs reasoning according to the rules retrieved from the knowledge base based on the operation task, automatically generates an operation ticket, and displays it through the human-machine interface terminal. This reduces the burden on dispatchers, improves the correctness of the operation ticket, and enhances the reliability and efficiency of invoicing.

[0043] Note that the above is only a preferred embodiment of the present invention and the applied technical principles. Those skilled in the art will understand that the present invention is not limited to the specific embodiments here. Various obvious changes, re-adjustments, and substitutions can be made by those skilled in the art without departing from the protection scope of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments. Without departing from the concept of the present invention, more other equivalent embodiments can be included, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. An operation ticket reasoning system based on graph topology, characterized by: It includes a database, a knowledge base, a topology graphics platform and an inference engine. The topology graphics platform is connected to the real-time status of the power grid graphics and the equipment from the database. The inference engine uses the topology analysis and anti-error verification of the topology graphics platform to perform inference according to the rules retrieved from the knowledge base according to the operation task, automatically generates an operation ticket, and displays it through the human-machine interface. The database includes a power grid graphic database, a device and device parameter database, an operation ticket database and a typical ticket database. The database is connected to the ESM system in the power grid and synchronizes its internal storage. The database serves as the data storage center of the system and is used to store knowledge, initial data and intermediate information generated during the reasoning process. The database provides the necessary data for the reasoning engine, including original data, process data and final conclusions. The knowledge representation in the knowledge base is rules, which are important modules for the automatic generation of the operation ticket system. They mainly describe the problem knowledge in the most reasonable form. The knowledge representation includes production representation, predicate logic representation, framework representation and object-oriented representation. The topological graphics platform is used for topological analysis of graphics, can realize the drawing of the main wiring diagram of the power system, and provides a complete interface so that it can be connected with the graphics in the database and the real-time status. It is a power grid graphics drawing and management platform built based on the library data integration technology and topological modeling technology, which provides strong support for the generation of operation tickets. At the same time, each electrical equipment on the graphics platform completely simulates the physical properties of the equipment in reality. Drawing is modeling, and the five-prevention rules are automatically processed. Then, the five-prevention check is automatically added when the ticket is issued, which once again improves the intelligence and security of the operation ticket; The reasoning machine can use the knowledge in the knowledge base according to the currently known facts, and perform corresponding operation tasks by calling the rules and knowledge in the knowledge base according to certain reasoning basis to obtain the answer to the problem or prove the correctness of a certain hypothesis.

2. The operation ticket reasoning system based on graph topology according to claim 1 is characterized in that: The knowledge base and the topology graph are only linked through the inference engine. The knowledge base contains some terminology libraries and ticket issuance logic sequences. After receiving the operation task, the inference engine obtains the operation sequence by analyzing the wiring method, voltage level, etc., and finally decomposes it into operation terms to generate an operation ticket. In this process, the inference engine will interact with the knowledge base and the topology graph platform according to the description of the task, infer the correct operation logic, and generate an operation ticket that meets user needs.

3. The operation ticket reasoning system based on graph topology according to claim 2 is characterized in that: The basic idea of ​​the forward reasoning of the inference engine is: starting from the existing information, looking for available knowledge, selecting and enabling knowledge through conflict resolution, executing knowledge, changing the solution state, and gradually solving the problem until the problem is solved; its working process is: the user first stores the information related to the reasoning into the database, and the inference engine selects appropriate rules from the knowledge base based on this information. If a rule is to be established, all the premise facts of the rule must be matched successfully; once it is found that as long as one fact premise of a rule is not established, the rule can be skipped and the next rule can be checked to see if it can be matched; if all the fact premises of a rule are matched successfully, that is, the rule is established, the conclusion of the rule is stored in the database, and the "use flag" field of the rule is set to the used state, so that the rule does not need to be tested again in the subsequent reasoning cycle.

4. The operation ticket reasoning system based on graph topology according to claim 3 is characterized by: There are two termination conditions for forward reasoning: one is to end when a solution that meets the conditions is found; the other is to end when all solutions are found.

5. The operation ticket reasoning system based on graph topology according to claim 1 or 4 is characterized in that: The inference algorithm in the inference engine is the Rete algorithm, and its analysis steps are as follows: (1) Input the original data into the working memory; (2) Compare the data and the rules and select the rules that match successfully; (3) If there is a conflict in the execution of the rules, activate multiple rules at the same time and store the conflicting rules in the conflict set; (4) Resolve the conflict and put the activated rules into the agenda in the predetermined order; (5) Execute the rules stored in the agenda; (6) Repeat steps 2 to 5 until all the rules in the agenda are executed.

Images