Operation scheme generation method, apparatus and device, and readable storage medium

Abstract

The invention discloses an operation scheme generation method, device and equipment and a readable storage medium, and the method can respond to the voice input operation of an operator, and converts an input voice instruction into an input text; identifying a target operation task of the operator based on the input text; on the basis, the use threshold can be reduced; a knowledge graph used for indicating environment data of different areas and job execution plans corresponding to different job categories can be obtained; based on the target operation task and the knowledge graph, generating an operation scheme for guiding an operator to complete the target operation task; on the basis, a large amount of historical operation experience and environmental factors can be comprehensively referred to, and the operation scheme is generated in a targeted mode. Visibly, according to the application, a lot of historical experiences and environmental factors can be synthesized through the knowledge graph, the operation scheme conforming to the voice input of the operator is generated, and the comprehensiveness, accuracy and safety of the operation scheme are improved while the application universality is ensured.

Description

Technical Field

[0001] The present application relates to the field of electric power technology, and more specifically, to an operation plan generation method, device, equipment and readable storage medium. Background Art

[0002] Power operations cover the construction, maintenance, overhaul, and upgrade of power systems, and are an important part of maintaining the stable operation of power systems. However, power operations are extremely complex and involve a wide range of professional knowledge and technical fields.

[0003] In the past, the formulation of power operation plans relied heavily on the personal experience of operators. Experienced operators were able to plan different operation steps, required tools and materials, and safety precautions for different operation tasks based on their previous experience in similar operations. However, personal experience has limitations, and the experience levels of different operators vary, which may lead to incomplete or inaccurate operation plans, and even worse, there may be certain safety risks. Summary of the invention

[0004] In view of this, the present application provides a method, device, equipment and readable storage medium for generating an operation plan, which are used to solve the shortcomings of incomplete and inaccurate operation plans in the prior art.

[0005] In order to achieve the above objectives, the proposed solution is as follows:

[0006] A method for generating an operation plan, comprising:

[0007] Responding to the operator's voice input operation, converting the input voice command into input text;

[0008] Based on the input text, identifying the target work task of the operator;

[0009] Obtain a knowledge graph that indicates environmental data for different areas and job execution plans for different job categories;

[0010] Based on the target work task and the knowledge graph, a work plan is generated to guide the operator to complete the target work task.

[0011] Optionally, responding to the voice input operation of the operator and converting the input voice instruction into input text includes:

[0012] Respond to the operator's voice input operation, combine the thought chain technology, and guide the operator to input multiple voice commands;

[0013] Convert and integrate each voice command to form input text.

[0014] Optionally, the identifying the target work task of the operator based on the input text includes:

[0015] Performing entity extraction on the input text and extracting entity relationships from the input text;

[0016] Based on the entities and the entity relationships, a target job task is generated.

[0017] Optionally, performing entity extraction on the input text and extracting entity relationships from the input text includes:

[0018] Obtaining a small language model, where the small language model is trained by a plurality of speech translation texts including a job task;

[0019] Using a sentence corpus and a large language model, adjusting parameters of the small language model, wherein the large language model is larger than the small language model;

[0020] The context semantics perception layer and implicit dependency decoding layer in the latest small language model are used to perform entity extraction and entity relationship extraction on the input text.

[0021] Optionally, the step of obtaining a knowledge graph indicating environmental data of different regions and job execution plans corresponding to different job categories includes:

[0022] Obtain the operation execution plan corresponding to different operation tasks. Each operation execution plan includes the operation steps, operation safety specifications, operation risk categories, operation categories, emergency response measures, required tools and materials, operation intentions, power equipment and operation site types of the corresponding operation tasks;

[0023] Combined with quantitative pruning technology and feature reuse technology, structured knowledge is extracted from each job execution plan to build a job graph;

[0024] Based on all the power equipment contained in different areas and the environmental data of each power equipment, the operation map is updated to form a knowledge map after the update.

[0025] Optionally, generating a work plan for guiding the operator to complete the target work task based on the target work task and the knowledge graph includes:

[0026] Based on the target operation task, determine the operation type, operation power equipment, operation target location, operation key indicators and operation objectives of the operator;

[0027] Based on the knowledge graph, multiple operation execution plans corresponding to the operation type are selected, and based on the knowledge graph and the operation target location, an environment type corresponding to the operation target location is determined;

[0028] Selecting a target operation execution plan that matches the operation power equipment and the environment type from the selected operation execution plans;

[0029] The operation plan is generated based on the target operation execution plan, the operation key indicators and the operation objectives.

[0030] Optionally, generating the operation plan based on the target operation execution plan, the operation key indicators and the operation target includes:

[0031] Based on the key indicators of the operation and the operation objectives, the target operation execution plan is personalized modified to obtain an operation plan after the modification.

[0032] A device for generating an operation plan, comprising:

[0033] A conversion module, used to respond to the operator's voice input operation and convert the input voice command into input text;

[0034] An identification module, used for identifying a target operation task of the operator based on the input text;

[0035] An acquisition module, used to acquire a knowledge graph indicating environmental data of different regions and job execution plans corresponding to different job categories;

[0036] A generation module is used to generate a work plan for guiding the operator to complete the target work task based on the target work task and the knowledge graph.

[0037] A work plan generating device includes a memory and a processor;

[0038] The memory is used to store programs;

[0039] The processor is used to execute the program to implement the various steps of the above-mentioned operation plan generation method.

[0040] A readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, each step of the above-mentioned operation plan generation method is implemented.

[0041] It can be seen from the above technical solutions that the operation plan generation method provided by the present application can respond to the voice input operation of the operator and convert the input voice command into input text; based on the input text, the target operation task of the operator is identified; based on this, the present application can determine the operation task of the operator with the help of the operator's voice command, so that some operators who are not proficient in the operation of electronic equipment or have limited recognition can also easily use the present application, lower the usage threshold, and improve the practicality of the present application; at the same time, voice input can greatly reduce the input time, and has low requirements on the scene in which the operator is located, thereby improving the wide application of the present application; subsequently, a knowledge graph can be obtained to indicate the environmental data of different areas and the operation execution plan corresponding to different operation categories; based on the target operation task and the knowledge graph, an operation plan is generated to guide the operator to complete the target operation task; based on this, the present application can provide personalized operation plans for operators according to the target operation task. At the same time, the knowledge graph can indicate the environmental data of different regions and the operation execution plan corresponding to different operation categories, so that the knowledge graph summarizes a large amount of operation experience and a large number of operation processes. At the same time, the environmental data has a huge impact on the specific process of the operation. For example, the execution process of operations in different terrains is different. Therefore, this application combines the target operation task and the knowledge graph, and can comprehensively refer to a large amount of historical operation experience and environmental factors to generate a targeted operation plan for guiding the operator to complete the target operation task. At the same time, the knowledge graph can also be continuously updated and improved, further ensuring the comprehensiveness, accuracy and security of the reference experience, thereby ensuring the comprehensiveness, accuracy and security of the operation plan produced by this application. It can be seen that this application can integrate a large amount of historical experience and environmental factors through the knowledge graph to generate an operation plan that conforms to the operator's input voice, while ensuring the wide application, while improving the comprehensiveness, accuracy and security of the operation plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present application 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 embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0043] Figure 1 A flow chart of a method for generating a work plan disclosed in an embodiment of the present application;

[0044] Figure 2 This is a structural block diagram of a work plan generating device disclosed in an embodiment of the present application;

[0045] Figure 3This is a hardware structure block diagram of a job plan generating device disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0047] The method for generating a work plan in the present application can be used in many general or special computing device environments or configurations, such as personal computers, server computers, handheld devices or portable devices, tablet devices, multi-processor devices, and distributed computing environments including any of the above devices or equipment.

[0048] Next, combine Figure 1 The operation plan generation method of this application is introduced in detail, including the following steps:

[0049] Step S1, responding to the operator's voice input operation, converting the input voice instruction into input text.

[0050] Specifically, after receiving the wake-up voice of the operator, multiple voice commands input by the operator can be received, and each voice command can be aggregated to form an input voice;

[0051] The wake-up voice may be a variety of voice blocks related to the job solution query, for example, it may be a voice block containing a "job solution generation" tone, or it may be a voice block containing a "job" tone.

[0052] Step S2: Based on the input text, identify the target work task of the operator.

[0053] Specifically, text analysis technology can be used to parse the input text and identify the target work tasks of the operators.

[0054] Text analysis technology can be machine learning methods and dependency syntax analysis technology, etc.

[0055] Step S3: Obtain a knowledge graph indicating environmental data of different areas and job execution plans corresponding to different job categories.

[0056] Specifically, a large amount of historical operation experience and environmental data from different regions can be integrated in advance to build a knowledge graph.

[0057] When executing this application, a pre-built knowledge graph is retrieved.

[0058] The work categories may include scheduled inspections, fault repairs, power equipment replacements, power equipment installations and other tasks related to power work.

[0059] The environmental data of each area may include all the electric power equipment included in the corresponding area and the environmental type of the location of each electric power equipment.

[0060] Step S4: Based on the target work task and the knowledge graph, generate a work plan for guiding the operator to complete the target work task.

[0061] Specifically, based on the target job task, each job execution plan in the knowledge graph can be screened and reorganized to construct a job plan that matches the target job task.

[0062] It can be seen from the above technical solutions that the operation plan generation method provided by the present application can respond to the voice input operation of the operator and convert the input voice command into input text; based on the input text, the target operation task of the operator is identified; based on this, the present application can determine the operation task of the operator with the help of the operator's voice command, so that some operators who are not proficient in the operation of electronic equipment or have limited recognition can also easily use the present application, lower the usage threshold, and improve the practicality of the present application; at the same time, voice input can greatly reduce the input time, and has low requirements on the scene in which the operator is located, thereby improving the wide application of the present application; subsequently, a knowledge graph can be obtained to indicate the environmental data of different areas and the operation execution plan corresponding to different operation categories; based on the target operation task and the knowledge graph, an operation plan is generated to guide the operator to complete the target operation task; based on this, the present application can provide personalized operation plans for operators according to the target operation task. At the same time, the knowledge graph can indicate the environmental data of different regions and the operation execution plan corresponding to different operation categories, so that the knowledge graph summarizes a large amount of operation experience and a large number of operation processes. At the same time, the environmental data has a huge impact on the specific process of the operation. For example, the execution process of operations in different terrains is different. Therefore, this application combines the target operation task and the knowledge graph, and can comprehensively refer to a large amount of historical operation experience and environmental factors to generate a targeted operation plan for guiding the operator to complete the target operation task. At the same time, the knowledge graph can also be continuously updated and improved, further ensuring the comprehensiveness, accuracy and security of the reference experience, thereby ensuring the comprehensiveness, accuracy and security of the operation plan produced by this application. It can be seen that this application can integrate a large amount of historical experience and environmental factors through the knowledge graph to generate an operation plan that conforms to the operator's input voice, while ensuring the wide application, while improving the comprehensiveness, accuracy and security of the operation plan.

[0063] Next, the operation plan generation method provided in this application will be introduced in detail through a specific example.

[0064] In response to the operator's voice input operation, the input voice command is converted into input text, and the content of the input text is "We are going to perform line maintenance and insulator replacement work on this high-voltage tower. The tower is about 30 meters high and the line voltage level is 220kV."

[0065] The input text is parsed to determine that the operator's target task is "to perform line maintenance and insulator replacement on a tower with a height of about 30 meters and a line voltage level of 220kV."

[0066] Get pre-built knowledge graphs;

[0067] The target operation task is compared with the content contained in the knowledge graph, and an operation plan is constructed to guide the operators to complete the "line maintenance and insulator replacement of the electric tower with a height of about 30 meters and a line voltage level of 220kV."

[0068] In some embodiments of the present application, the process of converting the input voice instruction into input text in response to the voice input operation of the operator in step S1 is described in detail, and the steps are as follows:

[0069] S10. Respond to the operator's voice input operation, combine the thought chain technology, and guide the operator to input multiple voice commands.

[0070] Specifically, it can respond to the voice input operation of the operator, analyze the content contained in the voice instructions input by the operator, determine the operation information, and combine the thinking chain technology to guide the operator to input voice instructions containing other operation information.

[0071] For example, it can respond to the operator's voice input operation and, combined with the thought chain technology, guide the operator to input multiple voice commands related to the operation type, operation target, key operation indicators and operation objectives.

[0072] S11, converting and integrating each voice command to form an input text.

[0073] Specifically, each voice command may be converted one by one to obtain a converted text corresponding to each voice command;

[0074] The various conversion texts can be cleaned and integrated to obtain the input text.

[0075] Each voice command may be converted in a variety of ways to obtain a converted text corresponding to each voice command.

[0076] For example, a hidden Markov model (HMM) or a Gaussian mixture model (GMM) may be used to convert each voice command to obtain a converted text corresponding to each voice command.

[0077] It can be seen from the above technical solution that this embodiment provides an optional method of responding to the voice input operation of the operator and converting the input voice command into input text. The above method can be combined with the thinking chain technology to guide the operator to provide key information for generating the operation plan. The operator does not need to learn the input content in advance. All relevant information can be input under the guidance of this application, which further ensures that the present application can be adapted to each operator and improves the wide application of the present application.

[0078] In some embodiments of the present application, the process of step S2, identifying the target work task of the operator based on the input text, is described in detail, and the steps are as follows:

[0079] S20: Perform entity extraction on the input text, and extract entity relationships from the input text.

[0080] Specifically, dependency syntax may be used to perform entity extraction on the input text, and entity relationships may be extracted from the input text.

[0081] S21. Generate a target operation task based on the entity and the entity relationship.

[0082] Specifically, based on each entity and each entity relationship, a subject may be determined, a verb connecting each entity may be selected, an object may be determined, and modifiers of the verb and the object may be determined.

[0083] It can be seen from the above technical solution that this embodiment provides an optional method for identifying the target work tasks of operators based on input text. Through the above method, the target work tasks can be generated by extracting entities and entity relationships, thereby better completing the generation of work plans.

[0084] In some embodiments of the present application, step S20, extracting entities from the input text and extracting entity relationships from the input text is described in detail, and the steps are as follows:

[0085] S200: Obtain a small language model, where the small language model is trained by a plurality of speech translation texts including job tasks.

[0086] Specifically, an initial language model constructed based on dependency syntax can be obtained;

[0087] The initial language model is trained using multiple speech translation texts containing job tasks, and a small language model is obtained after training.

[0088] S201. Utilize a sentence corpus and a large language model to adjust parameters of the small language model, wherein the scale of the large language model is larger than that of the small language model.

[0089] Specifically, a large-scale sentence-based corpus and a large language model can be used to fine-tune the small language model and adjust its parameters.

[0090] The scale of the large language model may be larger than that of the small language model.

[0091] For example, the large language model can be an LLM model, which can be a deep learning model trained with a large amount of text data, and can generate natural texts in multiple languages ​​and understand the meaning of texts in different languages. The small language model can be an ollama model, which can be based on syntax and trained with multiple speech translation texts containing job tasks, and can only understand language texts related to job information.

[0092] S202. Utilize the context semantic perception level and implicit dependency decoding level in the latest small language model to perform entity extraction and entity relationship extraction on the input text.

[0093] Specifically, the context semantic perception level and implicit dependency decoding level in the latest small language model can be used to fuse surface vocabulary features and deep semantic features in the input text to perform entity extraction and entity relationship extraction on the input text.

[0094] It can be seen from the above technical solution that this embodiment provides an optional method for extracting entities from the input text and extracting entity relationships from the input text. Through the above method, a small language model can be used to extract entities and entity relationships. The small language model is trained with multiple speech translation texts containing job tasks and fine-tuned with a large language model. The power information reasoning ability of the small language model is further improved with the help of multiple language translation texts, and the common sense reasoning ability and content recognition ability of the small language model are improved with the help of a large language model. Thus, the extraction reliability of the small language model is improved, and the reliability of the job solution of this application is further guaranteed.

[0095] In some embodiments of the present application, the process of step S3, obtaining a knowledge graph indicating environmental data of different regions and job execution plans corresponding to different job categories, is described in detail, and the steps are as follows:

[0096] S30. Obtain the operation execution plans corresponding to different operation tasks, each operation execution plan includes the operation steps, operation safety specifications, operation risk categories, operation categories, emergency response measures, required tools and materials, operation intentions, power equipment and operation site types of the corresponding operation tasks.

[0097] Specifically, job execution plans corresponding to various job tasks may be obtained.

[0098] Different information can form different work tasks, and different work tasks can correspond to the same work category. Different work tasks can correspond to the same work risk category.

[0099] The operation steps can be the steps that need to be completed in a corresponding order in the process of completing the operation task.

[0100] The job risk category can be used to indicate the safety level required to complete the corresponding job task.

[0101] The job category can be risk investigation, construction, maintenance, overhaul, configuration upgrade and other job-related types.

[0102] The operation safety specifications may be the safety protection methods and operation step specifications that should be paid attention to when performing the corresponding operation steps.

[0103] Emergency response measures can be used to indicate the emergency measures corresponding to each accident, for example, emergency response measures for electric shock accidents and emergency response measures for power equipment failures.

[0104] The required tools and materials may be the tools and materials required to perform the operation steps.

[0105] The work intention may be the goal and purpose of the corresponding work task.

[0106] The power equipment may be power-related equipment that the operator needs to handle.

[0107] The work site type can be used to indicate the type of land on which the corresponding work task is carried out, for example, mountains, plains, etc.

[0108] S31. Combine quantitative pruning technology and feature reuse technology to extract structured knowledge of each job execution plan and build a job graph.

[0109] Specifically, the quantitative pruning technology and feature reuse technology can be combined to extract the operation steps, operation safety specifications, operation risk categories, operation categories, emergency response measures, required tools and materials, operation intentions and work site types from each operation text, and the operation steps, operation safety specifications, operation risk categories, operation categories, emergency response measures, required tools and materials, operation intentions and work site types can be structured and integrated to obtain an operation map.

[0110] S32. Based on all the power equipment contained in different areas and the environmental data of each power equipment, the operation map is updated to form a knowledge map after the update.

[0111] Specifically, the environmental data corresponding to each power equipment in different areas can be cleaned and added to the operation graph to obtain a knowledge graph.

[0112] It can be seen from the above technical solution that this embodiment provides an optional way to obtain a knowledge graph. By constructing a knowledge graph, a large amount of historical job execution plans and environmental data can be summarized, so that a comprehensive and safe job plan can be better planned through the knowledge graph in the future.

[0113] In some embodiments of the present application, step S4, a process of generating a work plan for guiding the operator to complete the target work task based on the target work task and the knowledge graph, is described in detail, and the steps are as follows:

[0114] S40. Based on the target operation task, determine the operation type, operation power equipment, operation target location, operation key indicators and operation objectives of the operator.

[0115] Specifically, key points can be extracted from the target operation task, such as operation type, operation power equipment, operation target location, operation key indicators and operation objectives.

[0116] S41. Based on the knowledge graph, select multiple job execution plans corresponding to the job type, and determine the environment type corresponding to the job target location based on the knowledge graph and the job target location.

[0117] Specifically, target job execution plans that can be used for reference can be selected one by one from the knowledge graph.

[0118] Among them, multiple job execution plans that match the job type can be selected from various job execution plans in the knowledge graph.

[0119] Based on the knowledge graph, the power equipment that matches the target location of the operation is determined, and the environmental data of the matching power equipment is determined. Based on the environmental data of the matching power equipment, all environmental types corresponding to the target location of the operation are determined.

[0120] S42. Filter the target operation execution plan that matches the operation power equipment and the environment type from the selected operation execution plans.

[0121] Specifically, an operation execution plan that matches both the operation power equipment and the environment type may be selected from the selected operation execution plans as the target operation execution plan.

[0122] S43: Generate the operation plan based on the target operation execution plan, the operation key indicators and the operation objectives.

[0123] Specifically, all target job execution plans can be processed with reference to the job key indicators and job objectives to obtain the job plan.

[0124] It can be seen from the above technical solution that this embodiment provides an optional method for generating a work plan for guiding the operator to complete the target work task based on the target work task and the knowledge graph. Through the above method, the target work task is used as a guide to further screen the work execution plans that can be used for reference in the knowledge graph, and the screened target work execution plans are further personalized based on the target work task to ensure the reliability and personalization of the work plan.

[0125] In some embodiments of the present application, step S43, the process of generating the operation plan based on the target operation execution plan, the operation key indicators and the operation target, is described in detail, and the steps are as follows:

[0126] S430: Based on the key indicators of the operation and the operation target, the target operation execution plan is personalized modified to obtain an operation plan after the modification.

[0127] Specifically, each target operation execution plan can be integrated, and the integrated target execution plan can be modified based on the key indicators and operation objectives to generate an operation plan.

[0128] It can be seen from the above technical solution that this embodiment provides an optional method for generating the operation plan based on the target operation execution plan, the key operation indicators and the operation objectives. Through the above method, customized operation plans can be provided for different operation tasks of different operators. While improving the application flexibility of this application, the accuracy of this application is guaranteed by combining the target operation execution plan.

[0129] Next, we will combine Figure 2 The operation plan generation device provided in the present application is introduced in detail. The operation plan generation device provided below can be compared with the operation plan generation method provided above.

[0130] See also Figure 2 It can be found that the operation plan generating device may include:

[0131] The conversion module 10 is used to respond to the voice input operation of the operator and convert the input voice instruction into input text;

[0132] An identification module 20, configured to identify a target operation task of the operator based on the input text;

[0133] An acquisition module 30 is used to acquire a knowledge graph indicating environmental data of different regions and job execution plans corresponding to different job categories;

[0134] The generation module 40 is used to generate a work plan for guiding the operator to complete the target work task based on the target work task and the knowledge graph.

[0135] Transformation modules may include:

[0136] The first conversion unit is used to respond to the voice input operation of the operator and guide the operator to input multiple voice commands in combination with the thought chain technology;

[0137] The second conversion unit is used to convert and integrate various voice instructions to form input text.

[0138] The identification module may include:

[0139] An entity extraction unit, used to perform entity extraction on the input text and extract entity relationships from the input text;

[0140] The target operation task generating unit is used to generate the target operation task based on the entity and the entity relationship.

[0141] The entity extraction unit may include:

[0142] A first entity extraction component is used to obtain a small language model, where the small language model is trained with a plurality of speech translation texts containing job tasks;

[0143] A second entity extraction component is used to adjust parameters of the small language model using a sentence corpus and a large language model, wherein the scale of the large language model is larger than that of the small language model;

[0144] The third entity extraction component is used to utilize the context semantic perception level and implicit dependency decoding level in the latest small language model to perform entity extraction and entity relationship extraction on the input text.

[0145] Acquisition modules may include:

[0146] The operation execution plan acquisition unit is used to acquire the operation execution plans corresponding to different operation tasks. Each operation execution plan includes the operation steps, operation safety specifications, operation risk categories, operation categories, emergency response measures, required tools and materials, operation intentions, power equipment and operation site types of the corresponding operation tasks;

[0147] The job graph construction unit is used to extract structured knowledge from each job execution plan and construct a job graph by combining quantitative pruning technology and feature reuse technology;

[0148] The operation map updating unit is used to update the operation map based on all the power equipment contained in different areas and the environmental data of each power equipment, and form a knowledge map after the update.

[0149] Generate modules can include:

[0150] An operation target determination unit, used to determine the operation type, operation power equipment, operation target location, operation key indicators and operation target of the operator based on the target operation task;

[0151] An environment type determination unit, configured to select a plurality of operation execution schemes corresponding to the operation type based on the knowledge graph, and determine the environment type corresponding to the operation target location based on the knowledge graph and the operation target location;

[0152] A target operation execution plan selection unit, used to select a target operation execution plan that matches the operation power equipment and the environment type from the selected operation execution plans;

[0153] The operation key indicator utilizing unit is used to generate the operation plan based on the target operation execution plan, the operation key indicator and the operation target.

[0154] Operational key indicator utilization units may include:

[0155] The target job execution plan enters a modification subunit, which is used to perform personalized modification on the target job execution plan based on the key indicators of the job and the job objectives, and obtains the job plan after the modification.

[0156] The operation plan generation device provided in the embodiment of the present application can be applied to operation plan generation equipment, such as PC terminals, cloud platforms, servers and server clusters. Figure 3 The hardware structure diagram of the operation plan generation device is shown in FIG. Figure 3 ,The hardware structure of the operation plan generation device may include: at least one processor 1, at least one communication interface 2, at least one memory 3 and at least one communication bus 4;

[0157] In the embodiment of the present application, the number of the processor 1, the communication interface 2, the memory 3, and the communication bus 4 is at least one, and the processor 1, the communication interface 2, and the memory 3 communicate with each other through the communication bus 4;

[0158] The processor 1 may be a central processing unit CPU, or an application-specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present invention, etc.;

[0159] The memory 3 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), etc., such as at least one disk memory;

[0160] The memory stores a program, and the processor can call the program stored in the memory, wherein the program is used to:

[0161] Responding to the operator's voice input operation, converting the input voice command into input text;

[0162] Based on the input text, identifying the target work task of the operator;

[0163] Obtain a knowledge graph that indicates environmental data for different areas and job execution plans for different job categories;

[0164] Based on the target work task and the knowledge graph, a work plan is generated to guide the operator to complete the target work task.

[0165] Optionally, the detailed functions and extended functions of the program may refer to the above description.

[0166] The embodiment of the present application further provides a readable storage medium, which may store a program suitable for execution by a processor, wherein the program is used to:

[0167] Responding to the operator's voice input operation, converting the input voice command into input text;

[0168] Based on the input text, identifying the target work task of the operator;

[0169] Obtain a knowledge graph that indicates environmental data for different areas and job execution plans for different job categories;

[0170] Based on the target work task and the knowledge graph, a work plan is generated to guide the operator to complete the target work task.

[0171] Optionally, the detailed functions and extended functions of the program may refer to the above description.

[0172] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0173] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0174] The above description of the disclosed embodiments enables professionals and technicians in the field to implement or use the present application. Various modifications to these embodiments will be apparent to professionals and technicians in the field, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. The various embodiments of the present application can be combined with each other. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for generating a work plan, characterized in that: include: Responding to the operator's voice input operation, converting the input voice command into input text; Based on the input text, identifying the target work task of the operator; Obtain a knowledge graph that indicates environmental data for different areas and job execution plans for different job categories; Based on the target work task and the knowledge graph, a work plan is generated to guide the operator to complete the target work task.

2. The method for generating a work plan according to claim 1, characterized in that: The step of responding to the voice input operation of the operator and converting the input voice instruction into input text includes: Respond to the operator's voice input operation, combine the thought chain technology, and guide the operator to input multiple voice commands; Convert and integrate each voice command to form input text.

3. The method for generating a work plan according to claim 1, characterized in that: The step of identifying the target work task of the operator based on the input text includes: Performing entity extraction on the input text and extracting entity relationships from the input text; Based on the entities and the entity relationships, a target job task is generated.

4. The method for generating a work plan according to claim 3, characterized in that: Performing entity extraction on the input text and extracting entity relationships from the input text, including: Obtaining a small language model, where the small language model is trained by a plurality of speech translation texts including a job task; Using a sentence corpus and a large language model, adjusting parameters of the small language model, wherein the large language model is larger than the small language model; The context semantics perception layer and implicit dependency decoding layer in the latest small language model are used to perform entity extraction and entity relationship extraction on the input text.

5. The method for generating a work plan according to claim 1, characterized in that: The step of obtaining a knowledge graph indicating environmental data of different regions and job execution plans corresponding to different job categories includes: Obtain the operation execution plan corresponding to different operation tasks. Each operation execution plan includes the operation steps, operation safety specifications, operation risk categories, operation categories, emergency response measures, required tools and materials, operation intentions, power equipment and operation site types of the corresponding operation tasks; Combined with quantitative pruning technology and feature reuse technology, structured knowledge is extracted from each job execution plan to build a job graph; Based on all the power equipment contained in different areas and the environmental data of each power equipment, the operation map is updated to form a knowledge map after the update.

6. The method for generating a work plan according to claim 5, characterized in that: The generating, based on the target operation task and the knowledge graph, an operation plan for guiding the operator to complete the target operation task includes: Based on the target operation task, determine the operation type, operation power equipment, operation target location, operation key indicators and operation objectives of the operator; Based on the knowledge graph, multiple operation execution plans corresponding to the operation type are selected, and based on the knowledge graph and the operation target location, an environment type corresponding to the operation target location is determined; Selecting a target operation execution plan that matches the operation power equipment and the environment type from the selected operation execution plans; The operation plan is generated based on the target operation execution plan, the operation key indicators and the operation objectives.

7. The method for generating a work plan according to claim 6, characterized in that: The generating the operation plan based on the target operation execution plan, the operation key indicators and the operation target includes: Based on the key indicators of the operation and the operation objectives, the target operation execution plan is personalized modified to obtain an operation plan after the modification.

8. A device for generating a work plan, characterized in that: include: A conversion module, used to respond to the operator's voice input operation and convert the input voice command into input text; An identification module, used for identifying a target operation task of the operator based on the input text; An acquisition module, used to acquire a knowledge graph indicating environmental data of different regions and job execution plans corresponding to different job categories; A generation module is used to generate a work plan for guiding the operator to complete the target work task based on the target work task and the knowledge graph.

9. A device for generating a work plan, characterized in that: including memory and processor; The memory is used to store programs; The processor is used to execute the program to implement each step of the operation plan generation method as described in any one of claims 1-7.

10. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, each step of the method for generating a work plan as described in any one of claims 1 to 7 is implemented.

Images