Voice interactive power plant equipment state query system and method

Abstract

The invention relates to a voice interactive power plant equipment state query system and method. The system comprises a voice interaction module, a data processing module, a data communication module and a feedback module. The voice interaction module is electrically connected with the data processing module, and the voice interaction module is used for receiving user voice and converting the user voice into a voice instruction; the data processing module is used for analyzing the voice instruction, is connected with a data center through the data communication module on the basis of an analysis result, and calls real-time data from a DCS (Distributed Control System) and a time sequence database of the data center; the data communication module adopts a private network and is used for realizing communication between the data processing module and a data center; and the feedback module is connected with the data processing module and is used for feeding back the real-time data to a user. Compared with the prior art, the method has the advantages that the requirements of industrial equipment state query on real-time performance, data security and multi-equipment data processing capability are met, and the like.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Things applications, and in particular to a voice interactive power plant equipment state query system and method. BACKGROUND

[0002] In equipment maintenance and daily monitoring, traditional equipment state query methods rely on telephone communication or manual query, which has many limitations.

[0003] On the one hand, the on-site environment is noisy, and the call is easily disturbed, affecting the accurate transmission of information; on the other hand, multiple maintenance personnel querying at the same time may cause the phone to be busy, delaying the maintenance progress.

[0004] In addition, relying on manual data transfer is not only inefficient, but also easy to introduce errors, affecting the timeliness and accuracy of decision-making.

[0005] Although there are some mobile applications in the prior art that support equipment state viewing, these applications usually require manual operation and cannot meet the real-time voice interaction requirements, especially in noisy industrial environments, the inconvenience of manual operation is more prominent.

[0006] Reference patent CN119271168A discloses a nursing board interactive system based on intelligent control, which realizes high-precision voice recognition and instruction analysis and execution by integrating voice recognition and control modules, supports instruction analysis and execution, provides real-time voice and visual feedback, and multiple modal interaction modes, significantly improving user experience.

[0007] However, this system mainly focuses on the medical care field, and there are still deficiencies in the real-time performance, safety and multi-device data processing capability for industrial equipment state query.

[0008] Another reference patent CN111061172A proposes an intelligent voice information control system for the elderly, which realizes emergency call, health monitoring data notification, fall alarm and other functions by using voice recognition technology, mainly applied to the field of health care for the elderly.

[0009] Although it has certain technical features in voice interaction and data processing, it cannot meet the requirements of real-time performance, data security and multi-device data processing capability for industrial equipment state query. SUMMARY

[0010] The purpose of the present application is to overcome the deficiencies of the prior art and provide a voice interactive power plant equipment state query system and method.

[0011] The purpose of the present application can be achieved by the following technical solutions:

[0012] According to one aspect of the present application, a voice interactive power plant equipment state query system is provided, which comprises a voice interaction module, a data processing module, a data communication module and a feedback module; the voice interaction module is electrically connected with the data processing module, and is used for receiving user voice and converting the user voice into voice instructions; the data processing module is used for analyzing the voice instructions and, based on the analysis result, connecting a data center through the data communication module to retrieve real-time data from a DCS and a time series database of the data center; the data communication module adopts a 5G private network and is used for realizing communication between the data processing module and the data center; and the feedback module is connected with the data processing module and is used for feeding back the real-time data to the user.

[0013] As a preferred technical solution, the voice interaction module is internally provided with a microphone and a Bluetooth headset, supports voice input, and is integrated with a voice recognition engine, which is used for converting user voice into voice instructions and internally provided with a hotword wake-up function and a voiceprint recognition technology.

[0014] As a preferred technical solution, the voice interactive power plant equipment state query system further comprises a storage module; the storage module adopts a time series database architecture and is used for storing historical query records.

[0015] As a preferred technical solution, the data processing module is integrated with a multi-round dialogue state machine, which is connected with the storage module; when the multi-round dialogue state machine detects a preset keyword, the corresponding context is retrieved, and the voice instructions are changed according to the corresponding context.

[0016] As a preferred technical solution, the data communication module adopts a TLS1.3 protocol and supports ECDH key exchange.

[0017] As a preferred technical solution, the feedback module is connected with a smart device screen and a Bluetooth headset, and is internally provided with a sound, light or vibration alarm device; the feedback module feeds back the real-time data to the user in the following ways: displaying a numerical curve or a state chart of the real-time data on the smart device screen; playing the real-time data result to the Bluetooth headset through a voice synthesis technology; and triggering a sound, light or vibration alarm through the alarm device in a preset emergency state.

[0018] As a preferred technical solution, the voice interactive power plant equipment state query system further comprises a positioning module; the positioning module adopts a UWB and Bluetooth AoA fusion positioning technology, can automatically associate a device topology map of an area where the user is located, and realizes location-based device data filtering.

[0019] According to another aspect of the present application, a voice interactive power plant equipment state query method is provided, which comprises the following steps:

[0020] S1, receiving user voice and positioning the user, converting the user voice into a voice instruction, and determining whether the instruction conforms to a preset basic format and syntax rule, if yes, executing step S2, if not, feeding back to the user for re-input through voice synthesis technology;

[0021] S2, constructing a power plant equipment knowledge system based on device association relationship and historical fault knowledge base using knowledge graph technology;

[0022] S3, combining the power plant equipment knowledge system and the current positioning information to parse the voice instruction, obtaining the device number and the required parameter type;

[0023] S4, based on the device number and the required parameter type, calling the corresponding real-time data from the data center;

[0024] S5, feeding back the real-time data to the user, and storing the query record to the blockchain log chain.

[0025] As a preferred technical solution, the specific process of S2 based on device association relationship and historical fault knowledge base, and constructing a power plant equipment knowledge system using knowledge graph technology includes:

[0026] S21, collecting device association relationship, account information and operation parameters from the power plant equipment management system and the distributed control system, and collecting fault records from the historical fault knowledge base; cleaning the collected data to remove duplicate and error data, and unifying the data format to obtain a data set;

[0027] S22, using named entity recognition technology of natural language processing to extract device entities, parameter entities and fault entities from the data set; and identifying the relationship between entities through dependency syntax analysis;

[0028] S23, using an ontology modeling tool to construct a device knowledge system based on the relationship between entities.

[0029] As a preferred technical solution, the specific process of S3 combining the power plant equipment knowledge system to parse the voice instruction includes:

[0030] S31, using a word segmentation tool of natural language processing to split the voice instruction into lexical units; and extracting entities from the lexical units through named entity recognition technology;

[0031] S32, calculating the semantic similarity of the extracted entities and the entities in the power plant equipment knowledge system, and labeling the logical relationship between the extracted entities in S31 according to the semantic similarity;

[0032] S33, according to the current positioning information, the device topology graph of the area where the user is located is matched, device data at the current user position is screened out, the device number and the required parameter type are extracted by combining the labeled entity and the device data at the current user position.

[0033] Compared with the prior art, the present application has the following beneficial effects:

[0034] 1、In the present application, the system includes a voice interaction module, a data processing module, a data communication module and a feedback module, realizing a full-process closed loop of "voice input→instruction analysis→data retrieval→feedback output", solving the problems of low efficiency and complex operation of traditional manual query in power plants, and providing a standardized and intelligent interactive solution for equipment state query. Manual operation is avoided through voice interaction, the response time is shortened by more than 60%, and the maintenance efficiency is significantly improved; and the power plant data center is directly connected, avoiding manual transfer errors and ensuring data accuracy; meeting the requirements of real-time performance, data security and multi-device data processing capability for industrial equipment state query.

[0035] 2、In the present application, the voice interaction module is built-in with a microphone and a Bluetooth headset, ensuring the portability of voice interaction; the built-in hotword wake-up function reduces the mis-touch rate, and the voiceprint recognition technology realizes user identity association, solving the problems of low voice recognition accuracy and cumbersome identity authentication in noisy industrial environments, and improving the interaction security and practicality.

[0036] 3、In the present application, by setting a storage module combined with a multi-round dialogue state machine, complex interaction can be understood and context can be automatically associated, solving the limitations of traditional single instruction interaction, fitting the actual needs of continuous questioning and associated analysis in power plant inspection and operation and maintenance, and improving the efficiency of human-machine cooperation.

[0037] 4、In the present application, the data communication module adopts TLS1.3 protocol, supports ECDH key exchange, ensures "end-to-end encryption" of data transmission, resists network monitoring and tampering risks, adopts encrypted communication and permission management to prevent data leakage, and improves the security of power plant data transmission.

[0038] 5、In the present application, the positioning module adopts UWB and Bluetooth AoA fusion positioning technology, which can automatically associate the device topology graph of the area where the user is located, realize device data filtering based on position, accurately associate user position and device topology, solve the low efficiency problem of manual switching of device grouping when querying across areas, and make the interaction more suitable for the field operation route.

[0039] 6、In the present application, the knowledge graph technology is used to construct the power plant equipment knowledge system, the knowledge graph construction process is used, and the voice command is analyzed combined with the power plant equipment knowledge system, so as to solve the problem that the traditional query depends on fixed instructions and cannot associate with equipment knowledge, the NLP technology is used to deeply understand the user's intention, the equipment and parameters in the knowledge graph are accurately matched, the ambiguity of natural language instructions is solved, the accuracy of the query result is ensured, the accuracy of the voice interaction is improved, and the voice interaction is suitable for the complex equipment scene of the power plant. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 FIG. 1 is a schematic diagram of a voice interactive power plant equipment state query system according to the present application.

[0041] Figure 2 FIG. 2 is a system data interaction flowchart according to the embodiment.

[0042] Figure 3 FIG. 3 is a schematic diagram of a voice interactive power plant equipment state query method according to the present application. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should fall within the scope of protection of the present application.

[0044] Nowadays, during the maintenance process of power plant equipment, maintenance personnel need to frequently confirm the equipment state values (such as valve opening degree, flow fluctuation, etc.). The traditional way is to contact the operation personnel of the centralized control room through the telephone to query the DCS (Distributed Control System) system data, but there are the following problems:

[0045] 1) The on-site environment is noisy, and the conversation is easily disturbed;

[0046] 2) Multiple maintenance personnel dialing the phone at the same time can easily cause line occupation and delay the maintenance progress;

[0047] 3) It depends on manual data transfer, which is low in efficiency and may introduce errors.

[0048] In view of the limitations of the prior art, the present application aims to provide a voice interactive power plant equipment state query system and method based on a smart bracelet, so as to solve the problem that the real-time voice interactive equipment state query cannot be efficiently, accurately and safely performed in the prior art.

[0049] Embodiment 1

[0050] In the embodiment, a voice interactive power plant equipment state query system is adopted, and a system architecture is shown in Figure 1 The voice interactive module is electrically connected with the data processing module, the voice interactive module is used for receiving user voice and converting the user voice into voice instructions, the data processing module is used for analyzing the voice instructions, and based on an analysis result, real-time data is called from a DCS and a time sequence database of a data center through the data communication module; the data communication module adopts a 5G private network and is used for realizing communication between the data processing module and the data center; the feedback module is connected with the data processing module, and the feedback module is used for feeding back the real-time data to the user.

[0051] In the embodiment, the system is applied to a smart bracelet, and a system data interaction flowchart is shown in Figure 2 The smart bracelet is an interactive entrance of the user, is responsible for collecting user voice instructions, transmits a request to the outside through a 5G / Wi-Fi network, and receives and presents final feedback (data display, vibration / voice alarm) at the same time. The transmission layer is a 5G / Wi-Fi network and a firewall; the 5G / Wi-Fi network provides a high-speed, low-delay data transmission channel, guarantees real-time performance of voice requests and feedback data, the firewall is used as a network security barrier, filters illegal access, prevents network attacks, and ensures safety of internal systems (such as a voice recognition module, a DCS system, and the like).

[0052] The core modules of the processing layer include a voice recognition module and a data request engine, the voice recognition module receives voice requests that are transmitted through a network and filtered by a firewall, converts voice content into analyzable text instructions (such as “query temperature of No. 1 steam turbine”), and is a key link for realizing voice interaction, the data request engine receives text instructions output by the voice recognition module, initiates a data query request to a DCS system (a distributed control system that stores real-time operation data of equipment) and a time sequence database (a database that stores historical and real-time data according to time sequences, and is convenient for trace analysis), and is a “dispatch center” of data flow.

[0053] The data source is a DCS system and a time series database; the DCS system is a core control system of a power plant, and real-time collection and storage of operation parameters (temperature, pressure, rotating speed, etc.) of equipment (such as a boiler, a generator, etc.) are provided to support real-time data for query; the time series database focuses on storage and management of time series data, and can quickly respond to queries based on a time dimension (such as a "trend of turbine vibration in the past 24 hours"), and supplement data dimensions of the DCS system. The feedback end is a bracelet interface, a vibration or voice alarm; the bracelet interface receives query results (such as equipment parameter values, curve charts) returned by the data request engine, and presents them to the user in a visual manner; and the vibration or voice alarm: when the query result triggers a preset threshold (such as an equipment parameter exceeding a standard, a fault warning), the smart bracelet sends a warning to the user through vibration and voice functions, and actively reminds the user of an abnormal state.

[0054] The voice interaction module is built-in with a microphone and a Bluetooth headset, supports voice input (such as "query the opening degree of valve No. 3"), and integrates a voice recognition engine to convert a question into a query instruction.

[0055] The data communication module connects a power plant big data center through a 5G private network, obtains equipment data in the DCS and the time series database in real time through an OPC server (OLE for Process Control Server, Object Linking and Embedding Server for Process Control), and uses an encryption protocol to ensure data transmission safety.

[0056] The data processing module parses a voice instruction, matches equipment numbers and parameter types (such as a valve and flow), and extracts corresponding real-time data from the data center.

[0057] The feedback module displays a numerical curve or a state chart on the bracelet screen, uses a voice synthesis technology (TTS) to broadcast the result to the Bluetooth headset, and triggers a vibration alarm in an emergency state (such as a valve not being closed).

[0058] In the system, the voice interaction module is built-in with a microphone and a Bluetooth headset, supports voice input, and integrates a voice recognition engine, which is used to convert user voice into a voice instruction, and is built-in with a hotword wake-up function and a voiceprint recognition technology.

[0059] The voice interaction power plant equipment state query system further includes a storage module; the storage module uses a time series database architecture, and is used to store historical query records for subsequent tracing and analysis.

[0060] In the system, the data processing module integrates a multi-round dialogue state machine, which is connected to the storage module; when the multi-round dialogue state machine detects a preset keyword, the corresponding context is called, and the voice instruction is changed according to the corresponding context; multi-round dialogue (such as "check the flow again") is supported, and interaction efficiency is improved.

[0061] In the system, the data communication module adopts the TLS1.3 protocol and supports ECDH key exchange (Elliptic Curve Diffie-Hellman key Exchange).

[0062] In the system, the feedback module is connected to the screen of the intelligent device and the Bluetooth headset, and is internally provided with a sound, light or vibration alarm device. The feedback module feeds back real-time data to the user in the following ways: displaying a numerical curve or a state chart of the real-time data on the screen of the intelligent device; using voice synthesis technology to broadcast the real-time data results to the Bluetooth headset; and triggering a sound, light or vibration alarm through the alarm device in a preset emergency state.

[0063] The voice interactive power plant equipment state query system further includes a positioning module. The positioning module adopts a UWB (Ultra-Wideband) and Bluetooth AoA (Angle of Arrival) fusion positioning technology, can automatically associate the device topology map of the area where the user is located, and realize location-based device data filtering. The integrated positioning function automatically associates the device data of the area where the maintenance personnel are located.

[0064] In summary, the system integrates voice interaction, data communication, data processing and feedback modules, supports multi-round dialogue, positioning function and historical query record storage, not only improves the interaction efficiency and the reliability of data acquisition, but also ensures the security of data transmission through the use of encrypted communication and permission management. In addition, the system is designed flexibly and can be adapted to other industrial scenes (such as chemical plants and substations), and has a wide application prospect.

[0065] Embodiment 2

[0066] In this embodiment, a voice interactive power plant equipment state query method is adopted, and the method steps are as shown in Figure 3 , and specifically include:

[0067] S1, receiving user voice, converting the user voice into a voice instruction, and judging whether the instruction conforms to a preset basic format and syntax rule. If yes, step S2 is executed, and if not, the user is fed back to re-input through voice synthesis technology.

[0068] S2, constructing a power plant equipment knowledge system using knowledge graph technology based on the device association relationship and the historical fault knowledge base.

[0069] S3, combining the power plant equipment knowledge system to analyze the voice instruction, and obtaining the device number and the required parameter type.

[0070] S4, based on the device number and the required call parameter type, call the corresponding real-time data from the data center;

[0071] S5, feed the real-time data to the user, and store the query record to the blockchain log chain.

[0072] The specific process of constructing the power plant equipment knowledge system in S2 based on the device association relationship and the historical fault knowledge base includes:

[0073] S21, collect the device association relationship, account information and operation parameters from the power plant equipment management system and the distributed control system, and collect the fault records from the historical fault knowledge base; clean the collected data to remove duplicate and error data, and unify the data format to obtain a data set;

[0074] S22, use the named entity recognition technology of natural language processing to extract device entities, parameter entities and fault entities from the data set; and identify the relationship between entities through dependency syntax analysis;

[0075] S23, use an ontology modeling tool to construct a device knowledge system based on the relationship between entities.

[0076] The specific process of analyzing the voice command in combination with the power plant equipment knowledge system in S3 includes:

[0077] S31, use a word segmentation tool of natural language processing to split the voice command into lexical units; and extract entities from the lexical units through named entity recognition technology;

[0078] S32, calculate the semantic similarity of the extracted entities and the entities in the power plant equipment knowledge system, and label the logical relationship between the extracted entities in S31 according to the semantic similarity;

[0079] S33, based on the labeled entities, extract the device number and the required parameter type.

[0080] The inspector needs to query the "1# turbine bearing temperature real-time data" in the turbine plant, and the operation process is as follows:

[0081] Voice interaction: the inspector triggers voice interaction through the bracelet wake-up word "power plant assistant" and says the command "check 1# turbine bearing temperature".

[0082] The voice interaction module recognizes the command, performs syntax verification, calls the preloaded power plant equipment knowledge system, and enters the analysis process.

[0083] Analysis: The word segmentation tool splits the instruction into "check" "No. 1 steam turbine" "bearing temperature"; the named entity recognition extracts the equipment entity "No. 1 steam turbine" and the parameter entity "bearing temperature"; the semantic similarity is calculated (matching degree with "steam turbine ontology → bearing → temperature parameter" in the knowledge graph is 98.7%), and the logical relationship is labeled as "equipment-owning component-parameter".

[0084] Retrieval: The data processing module generates a query request and sends it to the data center through the 5G private network; the DCS system retrieves the No. 1 steam turbine bearing temperature sensor data in real time: 82°C.

[0085] Feedback: The feedback module displays "No. 1 steam turbine bearing temperature: 82°C (normal range 75-85°C)" on the bracelet screen and voice broadcasts "the current No. 1 steam turbine bearing temperature is 82 degrees Celsius"; since the data is normal, the audible and light alarms are not triggered. The query record is encrypted and stored in the blockchain log chain, realizing traceability and tamper-proofing of operations.

[0086] In summary, the present scheme uses knowledge graph technology to build a power plant equipment knowledge system, constructs a knowledge graph construction process, and analyzes voice instructions combined with the power plant equipment knowledge system, solving the problem of traditional queries relying on fixed instructions and being unable to associate equipment knowledge. Through NLP (Natural Language Processing) technology, the user's intention is deeply understood, and the equipment and parameters in the knowledge graph are accurately matched, solving the ambiguity of natural language instructions, ensuring the accuracy of the query results, improving the accuracy of voice interaction, and making it suitable for complex equipment scenarios in power plants.

[0087] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any skilled person in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present application, and these modifications or replacements should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A voice-interactive power plant equipment status query system, characterized in that, The system includes a voice interaction module, a data processing module, a data communication module, and a feedback module. The voice interaction module is electrically connected to the data processing module. The voice interaction module receives user voice and converts it into voice commands. The data processing module parses the voice commands and, based on the parsing results, connects to the data center via the data communication module to retrieve real-time data from the data center's DCS and time-series database. The data communication module uses a dedicated network to enable communication between the data processing module and the data center. The feedback module is connected to the data processing module, and the feedback module is used to provide real-time data feedback to the user.

2. The voice-interactive power plant equipment status query system according to claim 1, characterized in that, The voice interaction module has a built-in microphone and Bluetooth headset, supporting voice input. It also integrates a voice recognition engine, which converts user voice into voice commands and includes a hot word wake-up function and voiceprint recognition technology.

3. The voice-interactive power plant equipment status query system according to claim 1, characterized in that, The aforementioned voice-interactive power plant equipment status query system also includes a storage module; the storage module adopts a time-series database architecture and is used to store historical query records.

4. The voice-interactive power plant equipment status query system according to claim 3, characterized in that, The data processing module integrates a multi-turn dialogue state machine, which is connected to the storage module. When the multi-turn dialogue state machine detects a preset keyword, it retrieves the corresponding context and changes the voice command according to the context.

5. The voice-interactive power plant equipment status query system according to claim 1, characterized in that, The data communication module adopts the TLS 1.3 protocol and supports ECDH key exchange.

6. The voice-interactive power plant equipment status query system according to claim 1, characterized in that, The feedback module is connected to the screen of the smart device and Bluetooth headset, and has a built-in sound, light or vibration alarm device; The feedback module provides real-time data to the user in the following ways: displaying numerical curves or status charts of the real-time data on the screen of a smart device; or broadcasting the real-time data results to a Bluetooth headset using speech synthesis technology. In a preset emergency state, an alarm device can trigger an audible, visual, or vibration alarm.

7. The voice-interactive power plant equipment status query system according to claim 1, characterized in that, The aforementioned voice-interactive power plant equipment status query system also includes a positioning module; the positioning module adopts UWB and Bluetooth AoA fusion positioning technology, which can automatically associate with the equipment topology map of the user's area and realize location-based equipment data filtering.

8. A voice-interactive method for querying the status of power plant equipment, characterized in that, The method is applied to a voice-interactive power plant equipment status query system as described in any one of claims 1-7, and the method steps include: S1. Receive user voice and locate the user, convert the user voice into voice command, and determine whether the command conforms to the preset basic format and grammar rules. If it conforms, proceed to step S2. If it does not conform, use speech synthesis technology to provide feedback to the user for re-entry. S2. Based on the equipment association and historical fault knowledge base, construct a power plant equipment knowledge system using knowledge graph technology; S3. Combine the power plant equipment knowledge system and current location information to analyze the voice command and obtain the equipment number and the type of parameter to be retrieved; S4. Retrieve the corresponding real-time data from the data center based on the device number and the type of parameter to be retrieved; S5. Real-time data is fed back to the user, and query records are stored in the blockchain log chain.

9. The voice-interactive power plant equipment status query method according to claim 8, characterized in that, The specific process of constructing a power plant equipment knowledge system based on equipment relationships and historical fault knowledge base, using knowledge graph technology in S2, includes: S21. Collect equipment relationships, ledger information, and operating parameters from the power plant equipment management system and distributed control system, and collect fault records from the historical fault knowledge base; clean the collected data, remove duplicate and erroneous data, and unify the data format to obtain a dataset; S22. Using named entity recognition technology from natural language processing, extract device entities, parameter entities, and fault entities from the dataset; identify the relationships between entities through dependency parsing; S23. Using ontology modeling tools, a device knowledge system is constructed based on the relationships between entities.

10. The voice-interactive power plant equipment status query method according to claim 8, characterized in that, The specific process of parsing voice commands in S3, which combines the knowledge system of power plant equipment, includes: S31. Using natural language processing word segmentation tools, speech commands are broken down into lexical units; and entities are extracted from lexical units using named entity recognition technology. S32. Calculate the semantic similarity between the extracted entities and the entities in the power plant equipment knowledge system, and label the logical relationships between the entities extracted in S31 based on the semantic similarity. S33. Based on the current location information, match the device topology map of the user's area, filter out the device data of the current user location, and combine the labeled entities with the device data of the user's current location to extract the device number and the parameter type to be retrieved.

Images