Power failure handling method, device and system based on natural language large model and computer program

Abstract

The invention provides a power failure handling method based on a natural language large model, which comprises the following steps of: connecting a power monitoring system and the large model through an intermediate agent, detecting a failure event by utilizing a sliding time window, converting binary data generated by the monitoring system into a natural language which can be understood by the large model, and automatically submitting the natural language to the large model. A large model gives a solution of a related problem, and an intermediate agent analyzes and classifies the solution and then provides the solution to a dispatcher for auxiliary decision making. According to the technical scheme, the accuracy of fault analysis and fault processing can be greatly improved, manual judgment errors are reduced, and the reliability and stability of a power supply system are improved.

Description

Technical Field

[0001] The present disclosure relates to the field of power monitoring technology, and in particular to a power fault handling method, device, system and computer program based on a natural language large model. Background Art

[0002] The power supply system is the basic guarantee system for the production of industrial and mining enterprises, and the power monitoring system is an important means to ensure the safe operation of the power supply system. The typical feature of the power monitoring system is that the monitoring objects are densely populated. At present, it mainly relies on the business ability and professional skills of the dispatchers to make analysis and decisions based on the monitoring data. When a power supply failure occurs, the fault problem must be solved and the power supply must be restored in a short time. Once the dispatcher makes a mistake in judgment, serious consequences will occur. Not only will the power supply not be restored quickly, but the accident may also be expanded, causing greater losses.

[0003] With the advancement of artificial intelligence industry applications, artificial intelligence technology has been introduced into the power sector. At present, there is a natural language big model (hereinafter referred to as the "big model") that can provide power failure handling solutions. When in use, the dispatcher asks questions based on the fault information. The big model, based on the pre-learned knowledge and the contextual question-and-answer environment, gives a fault handling solution to assist the dispatcher in decision-making and execution.

[0004] However, the interaction process with the large natural language model is achieved through natural language question and answer. This process not only requires a long interaction time, but also has relatively high professional requirements for dispatchers. The efficiency and reliability of fault handling still have room for improvement. Summary of the invention

[0005] The present invention provides a method, device, system and computer program for handling power faults based on a natural language large model, which are used to solve the problems of low efficiency and reliability of existing power fault handling methods.

[0006] To solve the above technical problems, the present invention is achieved as follows:

[0007] On the one hand, the present invention provides a method for handling power faults based on a natural language large model, comprising:

[0008] Obtain binary real-time event data generated by the power monitoring system;

[0009] Converting the real-time event data within a preset time window into textual materials;

[0010] Submit the text material, corresponding questions and answer requirements to the big model, and receive a text report returned by the big model;

[0011] The content of the text report is formatted into a code adapted to the power monitoring system so as to be read by the power monitoring system.

[0012] On the other hand, the present invention provides a power fault handling device based on a natural language large model, comprising:

[0013] A monitoring data acquisition module is used to acquire binary real-time event data generated by the power monitoring system;

[0014] A text conversion module, used to convert the real-time event data within a preset time window into text material;

[0015] A large model interaction module, used to submit the text material, corresponding questions and answer requirements to the large model, and receive a text report returned by the large model;

[0016] The text parsing module is used to format the content of the text report into a code adapted to the power monitoring system so as to be read by the power monitoring system.

[0017] On the other hand, the present invention also provides a power fault handling system based on a natural language large model, characterized in that it includes:

[0018] Power monitoring system, used to monitor the power supply network and generate real-time event data;

[0019] The intermediate agent is used to obtain the binary real-time event data generated by the power monitoring system; convert the real-time event data within a preset time window into text materials; submit the text materials, corresponding questions and answer requirements to the large model, and receive the text report returned by the large model; format the content of the text report into a code adapted to the power monitoring system for reading by the power monitoring system

[0020] The technical solution of the present invention can interact with the natural language big model through the intermediate agent, and can convert the real-time fault data of the on-site operation of the power monitoring system into natural language that the big model can understand by combining the real-time data, the power grid structure and other related information. After automatically submitting to the big model, the big model will provide solutions to related problems, and the intermediate agent will parse and classify the solutions and provide them to the dispatcher for auxiliary decision-making. The application of the technical solution of the present invention can greatly improve the accuracy of fault analysis and fault handling, reduce manual judgment errors, and improve the reliability and stability of the power supply system. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0022] Figure 1 A schematic diagram of an application scenario of the power failure handling solution provided by an embodiment of the present disclosure;

[0023] Figure 2 A flowchart of a method for handling power failures provided in an embodiment of the present disclosure;

[0024] Figure 3 This is a schematic diagram of data flow in an embodiment of the present disclosure;

[0025] Figure 4 A structural block diagram of a power failure handling device provided in an embodiment of the present disclosure;

[0026] Figure 5 A structural block diagram of a power fault handling system provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the technical solutions in the present disclosure, the technical solutions of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present disclosure. In addition, for the sake of clarity, parts that are not related to the description of the exemplary embodiments are omitted in the drawings.

[0028] In this specification, it should be understood that terms such as "including" or "having" are intended to indicate the existence of the features, numbers, steps, behaviors, components, parts or combinations thereof disclosed in the present disclosure, and are not intended to exclude the possibility that one or more other features, numbers, steps, behaviors, components, parts or combinations thereof exist or are added. It should also be noted that the embodiments in the present disclosure and the features in the embodiments may be combined with each other if there is no conflict.

[0029] Figure 1 A schematic diagram of an application scenario of the power failure handling solution provided in an embodiment of the present disclosure.

[0030] like Figure 1As shown in the figure, the application scenario is the coal mine power supply system. The ground monitoring platform communicates with the power distribution device and power equipment in the underground substation through the coal mine power supply system ring network to form a power monitoring system. The ground monitoring platform grasps the real-time status of the power grid through the status of each switch in the power grid, telemetry data, and telesignaling data, and performs remote monitoring. When the power supply fails, the large model analyzes the fault through the monitoring information provided by the intermediate agent and gives the corresponding fault handling plan.

[0031] Figure 1 The intermediate agent in the embodiment of the present disclosure carries the power failure handling method, device or computer program, which runs on the ground and is connected to the ground monitoring platform and the large model. Preferably, the intermediate agent and the operation and maintenance program of the monitoring platform run in the same application terminal or server. The large model can run in an independent server as an independent unit and allow multiple such intermediate agents to interact with it.

[0032] Combine the following Figure 2 An embodiment of the power failure handling method provided by the present disclosure is described.

[0033] Figure 2 A flowchart of a power failure handling method provided by an embodiment of the present disclosure. From a program perspective, the execution subject of the process can be a program installed on an application server or an application terminal. It can be understood that the method can be executed by any device, equipment, platform, or device cluster with computing and processing capabilities.

[0034] like Figure 2 As shown, the method of this embodiment includes operations S210 to S240.

[0035] S210: Obtain binary real-time event data generated by the power monitoring system.

[0036] The existing power monitoring system generates binary data for monitoring the power supply network, including switch status, remote signaling, remote measurement, electricity data, etc. The information displayed on the visual operation and maintenance program is obtained after corresponding analysis and conversion.

[0037] The intermediate agent can read the binary real-time event data generated by the power monitoring system through shared memory. The intermediate agent quickly interacts with the power monitoring system through shared memory, reducing the processing delay of real-time data. Once a power supply failure occurs, the intermediate agent can read the corresponding real-time fault data in the first place.

[0038] S220: Convert the real-time event data within a preset time window into text materials.

[0039] For real-time event data, a sliding time window algorithm is used, and only the data within the time window is processed each time. The specific steps may include: filtering and normalizing the real-time event data for preprocessing; starting the sliding time window algorithm for the preprocessed real-time event data; and text-processing the real-time event data within the time window to obtain the corresponding text material.

[0040] The sliding time window size can be set to 6 to 8 seconds, with a maximum of 15 seconds, and the sliding step is one window. The setting of the sliding window can ensure that fault events are detected in time.

[0041] The real-time event data is filtered and normalized for preprocessing, including: removing unnecessary information in the real-time event data, retaining switch protection startup information and switch tripping information, and unnecessary information including recording startup, setting value modification, etc. The filtered real-time event data is normalized to shield the differences between different protectors. Since a large number of protectors in the power grid come from different manufacturers, there are differences in the output and transmission data of different protectors. After normalization, the problems caused by these differences can be eliminated.

[0042] The data text processing of the real-time event data within the time window may include: clearing the sliding time window that does not contain events, converting the corresponding binary event data into text for each event in the sliding time window that contains events, and adding auxiliary data analysis, the auxiliary data includes the substation name, switch name, switch nature, event name, event nature, event classification, occurrence time, etc. Among them, the switch nature includes: incoming switch, outgoing switch, contact switch, bus tie switch, etc.; the event nature includes: short circuit protection, overcurrent protection, low voltage protection, differential protection, intelligent backup, etc.; event classification includes: protection warning signal, fault protection tripping, etc. That is, according to the preset data specifications of the power system, the data at the corresponding position is extracted, and the field of the auxiliary data is assigned a value. In the text process, the known network topology relationship, substation structure information, etc. can also be used to package the text content obtained through text processing into text materials.

[0043] Furthermore, text processing of real-time event data within the time window may include: text processing of basic information of each event within the time window, the processed text content including substation name, switch name, switch nature, event name, event nature, event classification, and occurrence time.

[0044] Furthermore, the data text processing of the real-time event data within the time window may also include: text processing of the superior-subordinate relationship of each switch involved in each event, and the processed text content includes the substation name, the switch name, the switch's superior switch name, and the subordinate switch name. The intermediate agent forms a text of the superior-subordinate relationship of each switch involved and provides it to the large model, for example: the superior of the central substation 3# high switch is the central substation 1# high switch; the superior of the central substation 1# high switch is the ground substation 12# high switch, etc.

[0045] Furthermore, the data text processing of the real-time event data within the time window may also include: text processing of the structural information of each substation involved in each event, and the processed text content includes the substation name, the number and name of the incoming line switches, and the name of the bus tie switch. In order to provide more detailed information for the large model analysis, the intermediate agent will provide the structural information of each substation involved in the form of text to the large model, for example: the central substation has two incoming lines, 1# high-opening and 11# high-opening, and a bus tie switch, 10# high-opening, etc.

[0046] S230: Submit the text material, corresponding questions and answer requirements to the big model, and receive the text report returned by the big model.

[0047] Provide the above text materials to the big model, and on this basis, raise corresponding questions, give the expected answers, and wait for the big model to respond. For example, you can ask: "Please combine the above information and give a fault analysis report for this fault, including the cause of the fault, fault location, fault isolation plan, and power supply restoration plan." After the big model completes the fault analysis, it will give a text report in the form of a fault analysis report.

[0048] S240: Format the content of the text report into a code adapted to the power monitoring system so as to be read by the power monitoring system.

[0049] The intermediate agent analyzes the text report of the fault analysis returned by the large model and formats and categorizes the report, including:

[0050] Based on the text report, analyze the fault cause and give the cause of the fault, for example: short circuit fault A phase tripping;

[0051] Based on the text report, analyze the fault location and provide the fault point information that caused the fault, for example: the following lines of the 3# high-voltage switch of the central substation;

[0052] Based on the text report, analyze the fault isolation plan and give which switches cannot be closed in order to isolate the fault, for example: the 3# high-voltage switch of the central substation is opened;

[0053] Based on the text report, the power supply restoration plan is analyzed and the switches that should be closed in order to restore power supply are given, for example: the central substation has 1# high-voltage switch closed; the central substation has 2# high-voltage switch closed; the central substation has 4# high-voltage switch closed.

[0054] Therefore, the intermediate agent extracts the text content of the fault cause, fault location, fault isolation plan and power supply restoration plan based on the text report; formats the extracted text content into a code adapted to the power monitoring system and writes it into the shared memory.

[0055] The results provided to the power monitoring system include the following: fault cause, which gives the cause of the fault; fault location, which gives the fault point information that caused the fault; fault isolation plan, which gives which switches cannot be closed in order to isolate the fault; power supply restoration plan, which gives which switches should be closed in order to restore power supply.

[0056] After the intermediate agent writes the above formatting and classification results into the shared memory, the power monitoring system automatically reads them and displays them in categories on the operation and maintenance platform, providing auxiliary decision support for dispatchers to determine the fault point, formulate fault isolation plan and power supply restoration plan.

[0057] Figure 3 Schematic diagram of data flow in the above embodiment method.

[0058] like Figure 3 As shown in the figure, the middle agent obtains binary real-time event data from the power monitoring system. After filtering and normalization, the real-time event data enters the sliding time window. The middle agent processes the real-time event data in each time window into text, and combines the topological relationship information and substation structure information to package it into text materials. The text materials and the questions and answers are submitted to the big model, requiring the big model to make a fault analysis. The big model responds and returns a text report of the fault analysis. The middle agent parses and classifies the text report and provides it to the power monitoring system. The whole process is executed automatically without human intervention. When a power supply failure occurs, the middle agent can detect the fault event through the sliding time window, and automatically use the big model to output the fault handling plan in time.

[0059] According to the power fault handling method of this embodiment, the intermediate agent interacts with the natural language big model, and the real-time fault data of the power monitoring system on-site operation can be converted into natural language and automatically submitted to the natural language big model in combination with the grid topology relationship, substation structure and other related information. The big model provides solutions to related problems, and the intermediate agent parses the solutions and provides them to the dispatcher for auxiliary decision-making. The application of the technical solution of the present invention can greatly improve the accuracy of fault analysis and fault handling, reduce the occurrence of manual judgment errors, and improve the reliability and stability of the power supply system.

[0060] The above is an embodiment of the fault handling method provided in the specification of this disclosure. Correspondingly, the specification of this disclosure also provides an embodiment of an electric power fault handling device 400 based on a natural language large model.

[0061] Figure 4 A structural block diagram of a power fault handling device provided in an embodiment of the present disclosure.

[0062] like Figure 4 As shown, the power fault handling device 400 includes a monitoring data acquisition module 410, a text conversion module 420, a large model interaction module 430 and a text parsing module 440. The power fault handling device 400 can be implemented by software, hardware or a combination of both.

[0063] The monitoring data acquisition module 410 is used to acquire binary real-time event data generated by the power monitoring system.

[0064] The text conversion module 420 is used to convert the real-time event data within a preset time window into text materials.

[0065] The big model interaction module 430 is used to submit text materials, corresponding questions and answer requirements to the big model, and receive a text report returned by the big model.

[0066] The text parsing module 440 is used to format the content of the text report into a code adapted to the power monitoring system so as to be read by the power monitoring system.

[0067] According to the fault handling device of this embodiment, the intermediate agent interacts with the natural language large model, which reduces the dependence on manual work and improves the efficiency and accuracy of fault handling.

[0068] The specification of the present disclosure also provides an embodiment of a power fault handling system 500 based on a natural language big model. Figure 5 This is the structural block diagram of the power fault handling system.

[0069] like Figure 5 As shown, the power fault handling system 500 includes a power monitoring system 510 and an intermediate agent 520 .

[0070] The power monitoring system 510 is used to monitor the power supply network and generate real-time event data;

[0071] The intermediate agent 520 is used to: obtain the binary real-time event data generated by the power monitoring system; convert the real-time event data within a preset time window into text materials; submit the text materials, corresponding questions and answer requirements to the big model, and receive the text report returned by the big model; format the content of the text report into a code that is adapted to the power monitoring system so that it can be read by the power monitoring system 510.

[0072] The power monitoring system 510 and the intermediate agent 520 can be installed on the same terminal device or server and can share memory data. Exchanging data through shared memory can improve data access efficiency and improve fault handling efficiency.

[0073] Another aspect of the present disclosure also provides a computer program, which, when executed by a processor, enables the processor to implement the various methods described above.

[0074] The above describes specific embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0075] The units or modules involved in the embodiments described in the present disclosure may be implemented by software or programmable hardware. The units or modules described may also be set in a processor, and the names of these units or modules do not constitute limitations on the units or modules themselves in some cases.

[0076] Each embodiment in the present disclosure is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, system, and computer program embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0077] The device, system, and computer program provided in the embodiments of the present disclosure correspond to the method. Therefore, the device, system, and computer program also have beneficial technical effects similar to those of the corresponding method. Since the beneficial technical effects of the method have been described in detail above, they will not be repeated here.

[0078] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present disclosure is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other.

Claims

1. A method for handling power failures based on a natural language large model, characterized in that: include: Obtain binary real-time event data generated by the power monitoring system; Converting the real-time event data within a preset time window into textual materials; Submit the text material, corresponding questions and answer requirements to the big model, and receive a text report returned by the big model; The content of the text report is formatted into a code adapted to the power monitoring system so as to be read by the power monitoring system.

2. The method according to claim 1, characterized in that The step of obtaining binary real-time event data generated by the power monitoring system includes: Read binary real-time event data generated by the power monitoring system through shared memory.

3. The method according to claim 1, characterized in that The converting the real-time event data within the preset time window into text material includes: Filtering and normalizing the real-time event data for preprocessing; Starting a sliding time window algorithm for the pre-processed real-time event data; The real-time event data within the time window is processed into text form to obtain the corresponding text material.

4. The method according to claim 3, characterized in that The data text processing of the real-time event data within the time window includes: The basic information of each event within the time window is converted into text, and the processed text content includes substation name, switch name, switch nature, event name, event nature, event classification, and occurrence time.

5. The method according to claim 4, characterized in that Also includes: The superior-subordinate relationship of each switch involved in each of the events is processed in text form, and the processed text content includes the substation name, the switch name, the superior switch name, and the subordinate switch name.

6. The method according to claim 4, characterized in that Also includes: The structural information of each substation involved in each of the events is processed into text form, and the processed text content includes the substation name, the number and name of incoming line switches, and the name of the busbar switch.

7. The method according to claim 1, characterized in that The step of formatting the content of the text report into a code adapted to the power monitoring system comprises: Based on the text report, extract the text content of the fault cause, fault location, fault isolation plan and power supply restoration plan; The extracted text content is formatted into a code adapted to the power monitoring system and written into a shared memory.

8. The method according to claim 3, characterized in that The filtering and normalization preprocessing of the real-time event data includes: The information of unnecessary events in the real-time event data is removed, the switch protection start-up information and the switch tripping information are retained, and the filtered real-time event data is normalized to shield the differences between different protectors.

9. A power fault handling device based on a natural language large model, characterized in that: include: A monitoring data acquisition module is used to acquire binary real-time event data generated by the power monitoring system; A text conversion module, used to convert the real-time event data within a preset time window into text material; A large model interaction module, used to submit the text material, corresponding questions and answer requirements to the large model, and receive a text report returned by the large model; The text parsing module is used to format the content of the text report into a code adapted to the power monitoring system so as to be read by the power monitoring system.

10. A power fault handling system based on a natural language large model, characterized in that: include: Power monitoring system, used to monitor the power supply network and generate real-time event data; An intermediate agent is used to obtain binary real-time event data generated by the power monitoring system; Convert the real-time event data within a preset time window into text material; submit the text material, corresponding questions and answer requirements to the big model, and receive a text report returned by the big model; format the content of the text report into a code adapted to the power monitoring system for reading by the power monitoring system.

11. A computer program, which, when executed by a processor, enables the processor to implement the method according to any one of claims 1 to 8.

Images