Power grid dispatch-oriented voice interaction method and device

Abstract

The application relates to a power grid dispatch-oriented voice interaction method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring voice information; performing voice recognition on the voice information to obtain a dispatch instruction; correcting the dispatch instruction according to an instruction library instruction and an execution object; and generating an operation instruction according to the dispatch instruction for execution by a power grid system if the correction is successful. The method can improve the dispatch efficiency and reduce dispatch errors.

Description

Technical Field

[0001] This application relates to the field of voice interaction technology, and in particular to a voice interaction method and apparatus for power grid dispatching. Background Technology

[0002] With the rapid development of the power system and the advancement of power marketization, the integrated characteristics of the large power grid are becoming increasingly apparent, and the coupling relationships between different elements within the system and between the system and the external environment are constantly strengthening. The types of power grid resources are becoming increasingly diverse, the proportion of new energy sources is continuously increasing, and new control objectives such as improving the absorption rate of new energy sources are constantly evolving, further increasing the uncertainty of the power grid state. A series of profound developments and changes in the power system and dispatching operations have led to an increasing complexity in dispatching control strategies and procedures, placing higher demands on the automation and intelligence of dispatching operations. If dispatchers use traditional interactive methods to interact with the power grid system, the efficiency is significantly low, consuming a lot of time in processing each task, seriously affecting the progress of dispatching operations. Furthermore, in the face of a large workload, it is easy to overlook some important tasks.

[0003] The rapid development of artificial intelligence has changed the way humans interact with machines. Among these technologies, speech recognition and semantic understanding are relatively mature and approach a more natural human conversation experience. If AI technology is applied to power dispatching, dispatchers can give dispatching requests via voice, and AI can automatically recognize the voice as instructions that the power system can understand. This can reduce the workload of dispatchers in editing instructions, thereby improving their efficiency and reducing their workload.

[0004] However, ordinary smart voice devices on the market cannot meet the needs of power grid dispatching operations. Because power grid dispatching involves a large number of technical terms, recognition errors are unavoidable. Summary of the Invention

[0005] Therefore, it is necessary to provide a voice interaction method, device, computer equipment, computer-readable storage medium, and computer program product for power grid dispatch that can improve the accuracy of voice recognition in response to the above-mentioned technical problems.

[0006] Firstly, this application provides a voice interaction method for power grid dispatching. The method includes:

[0007] Acquire voice information; perform speech recognition on the voice information to obtain dispatch instructions;

[0008] The scheduling instructions are checked against the instructions in the instruction library and the execution object;

[0009] If the verification is successful, an operation instruction is generated based on the dispatch instruction and executed by the power grid system.

[0010] In one embodiment, acquiring voice information; performing voice recognition on the voice information to acquire scheduling instructions includes:

[0011] Input the voice information into the speech recognition model to obtain the text information;

[0012] Perform semantic error correction on the text information to obtain the scheduling text;

[0013] Extract keywords from the scheduling text, search the knowledge base based on the keywords, and obtain the scheduling instructions that match the scheduling text;

[0014] The scheduling instructions are displayed visually.

[0015] In one embodiment, acquiring voice information; performing voice recognition on the voice information to acquire scheduling instructions includes:

[0016] If no scheduling instruction is matched with the scheduling text, an alarm is issued and related scheduling instructions are given based on keywords for adjusting the voice information.

[0017] In one embodiment, the method further includes:

[0018] Tag the scheduling text and then enter the tagged scheduling text into the knowledge base.

[0019] In one embodiment, verifying the scheduling instructions based on the instruction library instructions and the execution object includes:

[0020] Check whether the type of the scheduling instruction matches the type of the instruction in the instruction library;

[0021] Verify whether the operation content of the scheduling instructions matches the operation content of the instructions in the instruction library;

[0022] Verify whether the scheduling instruction is associated with the execution object.

[0023] In one embodiment, the method further includes:

[0024] Perform speech recognition on voice information to obtain published information;

[0025] Based on the knowledge base, search and adjust the published information to obtain standardized published information;

[0026] Obtain the bound push list based on standardized release information;

[0027] Standardized information is pushed out based on the push list.

[0028] Secondly, this application provides a voice interaction device for power grid dispatching, the device comprising:

[0029] The information acquisition module is used to acquire voice information, perform voice recognition on the voice information, and obtain scheduling instructions;

[0030] The verification module is used to verify the scheduling instructions based on the instructions in the instruction library and the execution object;

[0031] The generation module is used to generate operation instructions based on the scheduling instructions when the calibration is successful, so that the power grid system can execute them.

[0032] Thirdly, this application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method.

[0033] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0034] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method.

[0035] The aforementioned voice interaction method and device for power grid dispatching involves: 1) Recognizing and understanding the semantics of voice information containing instructions provided by dispatchers; 2) Verifying the dispatch instructions to ensure the use of professional terminology and reduce error rates; and 3) Generating operational instructions for the power grid system to execute the dispatcher's commands if the instructions are successfully verified. Compared to traditional manual input of dispatch instructions, this significantly reduces the workload of dispatchers, greatly assists them in quickly analyzing and guiding power grid dispatching work plans, and accelerates the development of intelligent power grid operations. Furthermore, by adding a verification step to the dispatch instructions based on voice recognition technology, more reasonable and accurate instructions are obtained, reducing operational error rates and improving the reliability of power grid system operation. Attached Figure Description

[0036] Figure 1 This is a diagram illustrating the application environment of a voice interaction method for power grid dispatch in one embodiment.

[0037] Figure 2 This is a flowchart illustrating a voice interaction method for power grid dispatch in one embodiment;

[0038] Figure 3 This is a flowchart illustrating the process of obtaining scheduling instructions in one embodiment;

[0039] Figure 4 This is a flowchart illustrating a speech recognition model in one embodiment;

[0040] Figure 5 This is a flowchart illustrating the semantic correction and matching of scheduling instructions in one embodiment;

[0041] Figure 6 This is a structural block diagram of a voice interaction device for power grid dispatching in one embodiment;

[0042] Figure 7 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0044] The voice interaction method for power grid dispatching provided in this application can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on other network servers. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, etc. Server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0045] In one embodiment, such as Figure 2 As shown, a voice interaction method for power grid dispatching is provided, which is then applied to... Figure 1 Taking terminal 102 as an example, the explanation includes the following steps:

[0046] Step 202: Obtain voice information; perform voice recognition on the voice information to obtain dispatch instructions.

[0047] Voice information can be obtained by direct input from the dispatcher or through voice interaction between the dispatcher and the system.

[0048] The main process of speech recognition for speech information is as follows: 1. Feature extraction: extracting acoustic features from sound waveforms; 2. Converting acoustic features into phonemes for pronunciation; 3. Using decoding technology to convert the phoneme sequence into text.

[0049] Specifically, the system acquires voice information containing dispatching decisions from the dispatcher, including decisions related to load balancing, fault isolation, and relay protection. This voice information with dispatching decisions is then processed through speech recognition and converted into text containing the dispatching decisions; this text is the dispatching instruction.

[0050] Step 204: Verify the scheduling instructions based on the instruction library instructions and the execution object.

[0051] To avoid invalid dispatch instructions due to errors in decision-making, language expression, or speech recognition in the voice information provided by dispatchers, it is necessary to verify these instructions. On one hand, verification standardizes dispatch instructions, using more professional terminology to facilitate the generation of more accurate and executable instructions for the power grid system. On the other hand, verification checks the feasibility of the decisions behind the dispatch instructions. For example, a dispatch instruction might contain obvious contradictions, such as instructing the wind turbines in Unit A to shut down, even though no wind turbines exist in Unit A. In such cases, it is necessary to verify the dispatch decision, determine the accuracy of the dispatch instructions, and provide prompts to prevent errors.

[0052] If the verification is successful, proceed to the next step; if the verification fails, provide feedback to the dispatcher via voice or text, prompting the dispatcher to check if the issued voice information is incorrect, and wait for the dispatcher to provide the voice information again.

[0053] In one embodiment, text is converted into sound using text-to-speech (TTS) technology, so that the results of proofreading failures and other information are fed back to the dispatcher via voice. The dispatcher then provides voice feedback and other information, thereby achieving voice interaction.

[0054] Step 206: If the calibration is successful, generate an operation instruction based on the dispatch instruction and execute it through the power grid system.

[0055] If step 204 verification is successful, an operation instruction is generated based on the dispatch instruction and automatically sent to the power grid system. This operation instruction can be a command that the power grid system can execute directly, or it can be sent to the operational staff via telephone broadcast. After receiving the operation instruction, the operational staff completes the task according to the instructions and then enters the operation result into the system, thereby completing the dispatch corresponding to this dispatch instruction.

[0056] In one embodiment, after successful verification, the dispatch instructions are first visualized for the dispatcher to confirm their accuracy. Only after receiving confirmation from the dispatcher via voice or other means regarding the dispatch instructions will the dispatch instructions be sent to the power grid system to ensure the accuracy of the dispatch actions.

[0057] In the above embodiments, speech information is recognized to understand its semantics and obtain dispatch instructions. These instructions are then verified to ensure the use of specialized terminology and reduce error rates. Finally, if the verification is successful, operational instructions are generated, allowing the power grid system to execute the dispatcher's specified commands. Compared to traditional manual input of dispatch instructions, this significantly alleviates the workload of dispatchers, greatly assisting them in quickly analyzing and guiding power grid dispatching work plans, and accelerating the development of intelligent power grid operations. Furthermore, by adding a verification step to the dispatch instructions based on speech recognition technology, more reasonable and accurate instructions are obtained, reducing operational error rates and improving the reliability of the power grid system.

[0058] In one embodiment, such as Figure 3 As shown, step 202 includes:

[0059] Step 302: Input the voice information into the voice recognition model to obtain the text information.

[0060] Among them, such as Figure 4 As shown, the Automatic Speech Recognition (ASR) model is an acoustic model trained on a large amount of collected audio data using specialized deep learning techniques. It converts natural language into text. The audio data comes from manually recorded speech (recordings) or imported audio files (mp3 files). The speech recognition model consists of an acoustic feature extraction part and a deep learning part. After receiving the audio data, the acoustic feature extraction part first extracts acoustic features, and then the deep learning part converts the extracted features into text. This speech recognition model is trained using a large amount of speech and text data.

[0061] In one embodiment, the deep learning component employs an Artificial Neural Network (ANN), a method that simulates human neural activity and possesses some human characteristics, such as automatic adaptation and autonomous learning. Its strong classification and mapping capabilities are highly valuable in speech recognition technology. Using an ANN allows for the storage of valuable data, followed by data processing methods such as cleaning, filtering, and selection of the feature data. Finally, data training yields different models, and these models are labeled to form feature models.

[0062] Step 304: Perform semantic error correction on the text information to obtain the scheduling text.

[0063] like Figure 5As shown, after processing with a speech recognition model, preliminary text information is obtained. This text information has a certain error rate and needs to be corrected by a semantic correction algorithm. The semantic correction algorithm extracts keywords from the obtained text information and infers the correctness of the statement based on the keywords. If the accuracy rate of the inference result is less than 60%, it is marked as abnormal; otherwise, it is marked as correct. It also incorporates relevant knowledge from a language knowledge base to provide further correction, resulting in a scheduled text. After correction, the scheduled text is output, completing the semantic correction. The semantic correction algorithm employs a traditional semantic information-based correction method.

[0064] Step 306: Extract keywords from the scheduling text, search the knowledge base based on the keywords, and obtain the scheduling instructions that match the scheduling text.

[0065] like Figure 5 As shown, keywords are extracted from the scheduling text using a neural network algorithm, and these keywords are then input into a knowledge base for searching. The knowledge base contains a large number of professional terms related to power grid scheduling, and scheduling instructions related to the scheduling text are obtained through keyword search.

[0066] In this embodiment, if the instructions given in the scheduling text are unclear, all relevant scheduling instructions will be provided based on keywords, and the scheduler will confirm the matching scheduling instructions.

[0067] Step 308: Visualize the scheduling instructions.

[0068] By visualizing dispatch instructions, dispatchers can confirm whether they are correct, thereby improving the accuracy of dispatch instructions.

[0069] This embodiment discloses a data processing procedure from text information to scheduling instructions, including steps for standardizing the scheduling text language and visualizing the scheduling instructions. The purpose of these steps is to make the scheduling instructions more accurate and reliable, so that subsequent operation instructions generated based on the scheduling instructions are more accurate.

[0070] In one embodiment, step 202 includes: if the scheduling text does not match the scheduling instruction, issuing an alarm prompt and providing associated scheduling instructions based on keywords for adjusting the voice information.

[0071] If the analysis results of the dispatch text are ambiguous and cannot match the accurate dispatch instructions, the dispatcher can select from the sub-keywords under the feedback keywords, or receive dispatch instructions related to the feedback keywords. The dispatcher can then adjust the voice information based on the feedback and give the interactive voice instructions again until valid voice information is obtained and the accurate dispatch instructions can be matched.

[0072] In one embodiment, the method further includes: tagging the scheduling text and entering the tagged scheduling text into a knowledge base.

[0073] The knowledge base uses the Elasticsearch search engine to match relevant content with input keywords. Each received dispatch text is analyzed, and the results are tagged to enrich the knowledge base. The knowledge base is also periodically trained and analyzed, and low-value datasets are removed to maintain its health.

[0074] In one embodiment, verifying the scheduling instruction based on the instruction library instruction and the execution object includes: verifying whether the type of the scheduling instruction matches the type of the instruction in the instruction library; verifying whether the operation content of the scheduling instruction matches the operation content of the instruction in the instruction library; and verifying whether the scheduling instruction is associated with the execution object.

[0075] By matching the type, it is determined that the scheduling instruction can be accurately converted into an operation instruction. By matching the operation content, it is determined that the scheduling instruction is compliant and feasible. By judging whether the execution object is related, it is determined that the scheduling instruction has a clear executable object. Through these three verifications, the scheduling error rate can be greatly reduced, thereby reducing operational errors, improving the reliability of the power grid system, ensuring the normal and healthy operation of the power grid, and promoting the intelligent and modern development of the power grid.

[0076] In one embodiment, the method further includes: performing speech recognition on the voice information to obtain the published information; searching and adjusting the published information based on a knowledge base to obtain standardized published information; obtaining a bound push list based on the standardized published information; and pushing the standardized published information based on the push list.

[0077] In addition to issuing dispatch instructions, dispatchers also issue dispatch-related notifications. After a notification is input into a speech recognition model via voice, the model performs speech recognition to obtain the information. The knowledge base retrieves the main information (such as keywords), analyzes it, and displays complete information on technical terms. The system then adjusts the notification based on these terms, for example, replacing non-standard words with professional jargon. Once the dispatcher confirms the message is correct, they can issue a voice command to search for recipients. The system automatically links the notification to a list of recipients, which is then displayed to the dispatcher for confirmation. The dispatcher can further adjust the recipients on the list as needed. After adjustment, the dispatcher sends a message sending time command to the system. At the scheduled time, the notification is pushed to the accounts of the recipients on the list. The message distribution is then complete.

[0078] In one embodiment, the method includes:

[0079] Input the voice information into the speech recognition model to obtain the text information.

[0080] Perform semantic error correction on the text information to obtain the scheduling text.

[0081] Extract keywords from the scheduling text, search the knowledge base based on the keywords, and obtain the scheduling instructions that match the scheduling text.

[0082] If no scheduling instruction is matched with the scheduling text, an alarm is issued and related scheduling instructions are given based on keywords for adjusting the voice information.

[0083] Tag the scheduling text and then enter the tagged scheduling text into the knowledge base.

[0084] The scheduling instructions are displayed visually.

[0085] The scheduling instructions are checked against the instructions in the instruction library and the execution object. The check includes: checking whether the type of the scheduling instruction matches the type of the instruction in the instruction library; checking whether the operation content of the scheduling instruction matches the operation content of the instruction in the instruction library; and checking whether the scheduling instruction is associated with the execution object.

[0086] This method also includes: performing speech recognition on the voice information to obtain the published information; searching and adjusting the published information based on the knowledge base to obtain standardized published information; obtaining the bound push list based on the standardized published information; and pushing the standardized published information based on the push list.

[0087] By implementing the embodiments of this invention, the overall functionality becomes more flexible and adaptable. The dispatching operation mode changes from the traditional static human-machine interaction to a dynamic interaction between voice and the system, making it more practical and in line with the direction of intelligent system development. Furthermore, this embodiment provides dispatchers with certain business content analysis, reducing their workload and minimizing operational errors. This alleviates the workload of dispatchers, improves the reliability of the power grid system, greatly assists dispatchers in guiding work decisions, and accelerates the development of intelligent power grid operations.

[0088] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0089] Based on the same inventive concept, this application also provides a voice interaction device for implementing the aforementioned voice interaction method for power grid dispatching. The solution provided by this device is similar to the implementation described in the above method. Therefore, the specific limitations of the one or more voice interaction device embodiments for power grid dispatching provided below can be found in the limitations of the voice interaction method for power grid dispatching described above, and will not be repeated here.

[0090] In one embodiment, such as Figure 6 As shown, a voice interaction device for power grid dispatching is provided, comprising: an information acquisition module 602, a proofreading module 604, and a generation module 606, wherein:

[0091] The information acquisition module 602 is used to acquire voice information, perform voice recognition on the voice information, and acquire scheduling instructions.

[0092] The verification module 604 is used to verify the scheduling instructions based on the instruction library instructions and the execution object.

[0093] The generation module 606 is used to generate operation instructions based on the scheduling instructions when the calibration is successful, so that the power grid system can execute them.

[0094] The information acquisition module 602 is also used to input voice information into the speech recognition model to acquire text information; perform semantic error correction on the text information to acquire scheduling text; extract keywords from the scheduling text, search in the knowledge base based on the keywords to acquire scheduling instructions that match the scheduling text; and visualize the scheduling instructions.

[0095] The information acquisition module 602 is also used to issue an alarm if the scheduling text does not match the scheduling instruction, and to provide associated scheduling instructions based on keywords for adjusting the voice information.

[0096] The verification module 604 is also used to verify whether the type of the scheduling instruction matches the type of the instruction in the instruction library; to verify whether the operation content of the scheduling instruction matches the operation content of the instruction in the instruction library; and to verify whether the scheduling instruction is associated with the execution object.

[0097] The device also includes an input module, which is used to tag the scheduling text and input the tagged scheduling text into the knowledge base.

[0098] The device also includes an information publishing module. After the information acquisition module 602 performs speech recognition on the voice information and acquires the publishing information, the information publishing module searches and adjusts the publishing information based on the knowledge base to obtain standardized publishing information; it then obtains the bound push list based on the standardized publishing information; and finally, it pushes the standardized publishing information according to the push list.

[0099] The various modules in the aforementioned voice interaction device for power grid dispatch can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0100] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 7 As shown, this computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores voice data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a voice interaction method for power grid dispatching.

[0101] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 7As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a voice interaction method for power grid dispatching.

[0102] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0103] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described method embodiments.

[0104] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the above-described method embodiments.

[0105] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the above-described method embodiments.

[0106] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0107] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0108] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0109] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A voice interaction method for power grid dispatching, characterized in that, The method includes: Input the voice information into the speech recognition model to obtain preliminary text information; Based on a semantic correction algorithm, keywords are extracted from the preliminary text information, and the correctness of the speech information is inferred based on the keywords. If the accuracy rate of the inference result is less than 60%, the preliminary text information is determined to be abnormal, and the preliminary text information is corrected by combining the knowledge related to the keywords in the knowledge base to obtain the scheduling text. If the accuracy rate of the inference result is not less than 60%, the preliminary text information is determined to be correct, and the preliminary text information is determined to be the scheduling text. Extract keywords from the scheduling text, search the knowledge base based on the keywords to obtain scheduling instructions that match the scheduling text; or, if the scheduling instructions corresponding to the scheduling text are unclear, search the knowledge base for all scheduling instructions related to the keywords; or, if the scheduling text does not match the scheduling instructions, issue an alarm and provide associated scheduling instructions based on the keywords for adjusting the voice information. The scheduling instructions are visualized to confirm whether they are correct. The scheduling instructions are checked according to the instructions in the instruction library and the execution object; If the verification is successful, an operation instruction is generated based on the scheduling instruction and executed by the power grid system.

2. The method according to claim 1, characterized in that, The speech recognition model is obtained through specialized deep training on a large amount of collected audio data. The speech recognition model is used to convert natural language into text language. The speech recognition model includes an acoustic feature extraction part and a deep learning part. The acoustic feature extraction part is used to extract acoustic features. The deep learning part is used to convert the extracted acoustic features into text.

3. The method according to claim 1, characterized in that, The step of extracting keywords from the scheduling text and searching the knowledge base based on the keywords to obtain scheduling instructions that match the scheduling text includes: Keywords are extracted from the scheduling text using a neural network algorithm, and the scheduling instructions that match the scheduling text are obtained by searching the knowledge base based on the keywords.

4. The method according to claim 2, characterized in that, The method further includes: The scheduling text is tagged, and the tagged scheduling text is entered into the knowledge base.

5. The method according to claim 1, characterized in that, The step of verifying the scheduling instruction based on the instruction library instruction and the execution object includes: Verify whether the type of the scheduling instruction matches the type of the instruction library instruction; Verify whether the operation content of the scheduling instruction matches the operation content of the instruction library instruction; Verify whether the scheduling instruction is associated with the execution object.

6. The method according to any one of claims 2 to 4, characterized in that, The method further includes: The voice information is subjected to speech recognition to obtain the published information; Based on the knowledge base, the published information is searched and adjusted to obtain standardized published information; The list of push notifications bound to the device is obtained based on the standardized release information. The standardized release information is pushed according to the push list.

7. A voice interaction device for power grid dispatching, characterized in that, The device includes: The information acquisition module is used to input voice information into a speech recognition model to acquire preliminary text information; based on a semantic correction algorithm, it extracts keywords from the preliminary text information and infers the correctness of the voice information based on the keywords. If the accuracy rate of the inference result is less than 60%, the preliminary text information is determined to be abnormal, and the preliminary text information is corrected by combining knowledge related to the keywords in the knowledge base to acquire scheduling text; if the accuracy rate of the inference result is not less than 60%, the preliminary text information is determined to be correct, and the preliminary text information is identified as scheduling text; keywords are extracted from the scheduling text, and the scheduling instructions matching the scheduling text are searched in the knowledge base based on the keywords; or, if the scheduling instructions corresponding to the scheduling text are unclear, all scheduling instructions related to the keywords are searched in the knowledge base; or, if the scheduling text does not match the scheduling instructions, an alarm is issued, and associated scheduling instructions are given based on the keywords for adjusting the voice information; the scheduling instructions are visualized to confirm whether the scheduling instructions are correct. The verification module is used to verify the scheduling instructions based on the instructions in the instruction library and the execution object; The generation module is used to generate operation instructions based on the scheduling instructions when the calibration is successful, so as to enable the power grid system to execute them.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images