Industrial control configuration software system and method based on intelligent voice interaction

Abstract

The invention relates to the technical field of industrial automation control, and discloses an industrial control configuration software system based on intelligent voice interaction, and the system comprises a voice recognition module which is based on a hybrid architecture of a deep neural network and a hidden Markov model and is used for receiving a voice signal of an operator and converting the voice signal into text data; and the natural language processing module is used for analyzing an operation command in the text data and extracting a device name, an operation instruction and related parameters through a bidirectional long-short-term memory network model. Through combination of augmented reality and virtual reality technologies, visual display and operation feedback of industrial equipment states are realized so as to enhance the perception capability of field operators on the equipment states, remote operators are allowed to perform remote fault diagnosis and operation simulation through VR, more intuitive and efficient operation support is provided, and the operation experience of the field operators is improved. The accuracy and decision-making efficiency of equipment operation are effectively improved, and the risk of misoperation is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial automation control, and in particular to an industrial control configuration software system and method based on intelligent voice interaction. Background Art

[0002] In the field of industrial automation, industrial control configuration software systems, as the core platform of human-computer interaction, are widely used in production process monitoring, data acquisition and equipment control. However, existing industrial control configuration systems mainly rely on traditional graphical user interfaces. Operators usually use a mouse, keyboard or touch screen to operate and control the equipment. The interaction mode has significant problems as follows: Traditional graphical interfaces contain buttons and menus, and operators need to switch between multiple interfaces. Especially when dealing with emergencies, the complex operation process will lead to slow response and reduce production efficiency; Most traditional configuration systems rely on local control, and the remote control and management functions are not perfect. Especially in industrial environments where equipment is widely distributed, there is a lack of effective remote monitoring and fault diagnosis methods, which affects the intelligent management of production. Traditional configuration software can display equipment status in the form of charts, etc., but the static display method cannot satisfy the operator's quick understanding and judgment of equipment status, especially in complex industrial scenarios. Operators need real-time, dynamic and visual operation feedback.

[0003] Therefore, those skilled in the art provide an industrial control configuration software system and method based on intelligent voice interaction to solve the above problems. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides an industrial control configuration software system and method based on intelligent voice interaction to solve the problems raised in the above background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: an industrial control configuration software system based on intelligent voice interaction, comprising: The speech recognition module, based on a hybrid architecture of deep neural networks and hidden Markov models, is used to receive the operator's voice signals and convert them into text data; The natural language processing module is used to parse the operation commands in the text data and extract the device name, operation instructions, device status parameters, operation settings and execution conditions through the bidirectional long short-term memory network model; Augmented reality interaction module, which is used to display the 3D virtual model of industrial equipment and its real-time status through AR devices to provide operation guidance and feedback; Multimodal interaction module, used to combine voice, gesture and touch screen to realize various forms of interactive operations; Industrial Internet of Things platform integration module, used to collect operating data of industrial equipment and synchronize it to the cloud for analysis and processing; The remote control module is used to generate remote control commands through voice and send commands to remote devices through the industrial Internet of Things platform to achieve remote operation.

[0006] Preferably, the speech recognition module comprises: The speech feature extraction unit extracts the features of the speech signal through the Mel frequency cepstral coefficients. The formula is: , in, is the speech signal, is the fast Fourier transform, is the discrete cosine transform; The speech recognition unit performs time series modeling based on the hidden Markov model and classifies the features of speech frames through a deep neural network.

[0007] Preferably, the natural language processing module extracts the device name, action and parameters in the operation command based on a bidirectional long short-term memory network structure, which is specifically implemented by the following formula: , in, represents the hidden layer state at time t, is the input word sequence, Represents the concatenation operation of the forward and backward LSTM network outputs.

[0008] Preferably, the augmented reality interaction module includes: The device status mapping unit converts the device operation status parameters The three-dimensional virtual model mapped to the AR device, the device status is visualized through color, size or animation changes, and the mapping function is: , in, is the mapping function, is the state parameter vector of the device; Operation guidance unit, used to dynamically display operation steps through AR interface , the formula is: , in, For step content, The real-time status of the device.

[0009] Preferably, the multimodal interaction module includes: The gesture recognition unit captures gesture information through a 3D depth camera, uses a convolutional neural network to recognize gestures, and outputs control commands , where g represents the gesture feature vector; The touch interaction unit determines and controls the device by combining the touch screen with voice input.

[0010] Preferably, the industrial Internet of Things platform integration module includes: A data acquisition unit is used to collect the operating status of the equipment through equipment sensors and ensure data synchronization through a time synchronization protocol; The data analysis unit analyzes and predicts the equipment operation data based on the machine learning model. The prediction formula is: , in, is the machine learning prediction function, is the current device status, For the prediction time.

[0011] Preferably, the remote control module includes: The remote command generation unit generates remote control commands through voice, and transmits the commands to the remote device for execution through the industrial Internet of Things platform. The remote control commands are generated by the following formula: , in, Indicates remote control instructions. For operation actions, is the device name, is the operation parameter.

[0012] Preferably, in the speech recognition unit, time series modeling is performed based on a hidden Markov model, wherein the hidden Markov model is implemented by the following formula: The Hidden Markov Model defines the generation process of speech signals as follows: State Collection : Implicit speech state set, each Indicates the voice status; Observation Collection : The characteristic sequence of observations, each Represents the speech features at time t; State transition probability matrix : Defined by state Transfer to state The probability is expressed as: , Observation probability matrix : Defined in state Generate observations The probability is expressed as: , Initial state probability vector :Define the system to be in state at the initial moment The probability is expressed as: .

[0013] Preferably, the gesture recognition unit is combined with voice input, so that the operator can determine the device by voice and confirm the operation or adjust the device parameters by gesture.

[0014] An industrial control configuration software method based on intelligent voice interaction comprises the following steps: Step 1: Receive the operator's voice signal through the voice recognition module and convert it into text data, and extract voice features using Mel frequency cepstral coefficients; Step 2: Parse the text data through the natural language processing module and use the Bi-LSTM model to extract the device name, operation instructions and parameters in the operation command; Step 3: The augmented reality interaction module displays the 3D model and real-time status of the equipment in the operator's AR device, and provides operation guidance and feedback according to the operator's voice instructions; Step 4: Use the multimodal interaction module to combine voice, gestures and touch screen to operate the device, where gestures are recognized by convolutional neural networks and control commands are output ; Step 5: Collect equipment operation data through the industrial Internet of Things platform and use machine learning algorithms to analyze and predict the data; Step 6: Generate a remote control command through the remote control module and send it to the remote device. The remote command is generated and executed through the remote control instruction.

[0015] The present invention provides an industrial control configuration software system and method based on intelligent voice interaction. It has the following beneficial effects: 1. The present invention combines augmented reality with virtual reality technology to achieve visual display of industrial equipment status and operational feedback, so as to enhance the on-site operator's perception of equipment status, and allow remote operators to perform remote fault diagnosis and operation simulation through VR, providing more intuitive and efficient operation support, effectively improving the accuracy of equipment operation and decision-making efficiency, and reducing the risk of misoperation.

[0016] 2. The present invention realizes a highly flexible multimodal operating system by combining multiple interaction methods such as voice, gestures, and touch screens, thereby solving the inconvenience of the traditional single interaction mode in complex or noisy environments. The operator determines the most suitable input method according to the specific scenario, and the combination of touch screen and voice allows the operator to adjust parameters and select equipment more efficiently in a multi-tasking environment. Its multimodal interaction significantly improves the operational flexibility and response speed of the system, reduces operational complexity, and improves the work efficiency of operators.

[0017] 3. The present invention realizes the real-time collection, analysis and remote control functions of equipment operation data through the integration of industrial Internet of Things. The application of industrial Internet of Things platform enables the equipment operation data to be transmitted to the cloud in real time. Combined with machine learning and big data analysis algorithms, the system can intelligently predict equipment failures and provide optimization operation suggestions. In addition, the remote control function allows operators to remotely operate equipment through voice commands in different locations, thereby improving the flexibility and real-time nature of management, reducing the cost of on-site maintenance and operation, and significantly improving the stability and safety of equipment operation. At the same time, intelligent scheduling can be used to improve production efficiency and reduce equipment downtime. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a system framework diagram of the present invention; Figure 2 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0019] In order to make the technical personnel in the technical field understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in combination with the drawings in the embodiment of the present invention. Obviously, the described embodiment is a partial embodiment of the present invention, not a complete embodiment. Based on the embodiment of the present invention, other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present invention.

[0020] The present invention is described in detail below in conjunction with the accompanying drawings: Example: Please see attached Figure 1 and attached Figure 2 , an embodiment of the present invention provides an industrial control configuration software system based on intelligent voice interaction, comprising: The speech recognition module, based on a hybrid architecture of deep neural networks and hidden Markov models, is used to receive the operator's voice signals and convert them into text data; The natural language processing module is used to parse the operation commands in the text data and extract the device name, operation instructions, device status parameters, operation settings and execution conditions through the bidirectional long short-term memory network model; Augmented reality interaction module, which is used to display the 3D virtual model of industrial equipment and its real-time status through AR devices to provide operation guidance and feedback; Multimodal interaction module, used to combine voice, gesture and touch screen to realize various forms of interactive operations; Industrial Internet of Things platform integration module, used to collect operating data of industrial equipment and synchronize it to the cloud for analysis and processing; The remote control module is used to generate remote control commands through voice and send commands to remote devices through the industrial Internet of Things platform to achieve remote operation.

[0021] The combined structure of the speech recognition module combines the advantages of time series modeling and feature classification. It can maintain a high recognition accuracy in noisy industrial environments, greatly improving the input flexibility of the system, allowing operators to complete complex operations without relying on manual input devices, thereby improving operational efficiency; The natural language processing module realizes in-depth understanding of natural language, can parse complex instructions and ambiguous sentences of operators, support multiple operation requirements, improve the system's operation convenience and the accuracy of instruction recognition, and reduce the learning cost of operators; The augmented reality interaction module greatly enhances the operator's perception of equipment status and reduces operational errors. Especially in remote maintenance and fault diagnosis, it realizes real-time visualization of equipment status, making complex operations more convenient and accurate. The multimodal interaction module solves the limitations of the traditional single operation mode. Especially in noisy environments, gesture or touch screen interaction can replace voice input, significantly improving the system's adaptability and operational flexibility, and improving the overall interaction efficiency and user experience. The integrated module of the industrial Internet of Things platform realizes remote monitoring and intelligent analysis of the equipment operation status. The system can proactively predict potential faults, optimize equipment operation parameters, realize intelligent equipment management, significantly improve equipment management efficiency and operation safety, support efficient use of remote data, and reduce maintenance and management costs; The remote control module significantly improves the flexibility of equipment management, allowing operators to complete operations without having to be on site. This is especially true in emergency situations and remote maintenance scenarios, providing effective support for improving production efficiency and reducing operating costs.

[0022] The speech recognition module includes: The speech feature extraction unit extracts the features of the speech signal through the Mel frequency cepstral coefficients. The formula is: , in, is the speech signal, is the fast Fourier transform, is the discrete cosine transform; The speech recognition unit performs time series modeling based on the hidden Markov model and classifies the features of speech frames through a deep neural network.

[0023] The features extracted by the speech feature extraction unit through MFCC are closer to the human ear's auditory perception of speech, so that the spectral information of the speech signal can be more effectively represented. The feature extraction of speech has strong robustness and can still maintain high accuracy under environmental noise. At the same time, MFCC reduces the dimension of the data, making the subsequent recognition processing more efficient. The combination of hidden Markov model and deep neural network in the speech recognition unit can realize the temporal pattern recognition and high-precision classification of speech, and solve the accuracy and timing limitations of traditional models when processing continuous speech signals. The hidden Markov model is suitable for processing the time-dependent characteristics of speech signals, while the deep neural network can enhance the feature classification ability, so that it can maintain a high recognition rate in complex and noisy environments, realize accurate extraction of voice commands, significantly improve the system's recognition efficiency and adaptability to complex speech environments, and provide the system with a more flexible and convenient voice interaction method.

[0024] The natural language processing module extracts the device name, action and parameters in the operation command based on the bidirectional long short-term memory network structure, which is implemented by the following formula: , in, represents the hidden layer state at time t, is the input word sequence, Represents the concatenation operation of the forward and backward LSTM network outputs.

[0025] The natural language processing module accurately extracts key elements from sentences and adapts to the complex semantics and contextual changes in natural language, so that the system can recognize a variety of voice commands and generate accurate control instructions. The bidirectional LSTM takes into account the context in sentence parsing, so that the system can understand the user's intention more accurately, especially in ambiguous or complex instructions. It can identify implicit meanings and improve the accuracy and robustness of natural language understanding. In addition, by capturing contextual relationships, Bi-LSTM is suitable for processing lengthy and complex sentences, so that the system can process complex instructions without losing semantic accuracy, and dynamically adjust the understanding of polysemous words according to the context, effectively improving the system's adaptability to natural language input.

[0026] The augmented reality interaction module includes: The device status mapping unit converts the device operation status parameters The three-dimensional virtual model mapped to the AR device, the device status is visualized through color, size or animation changes, and the mapping function is: , in, is the mapping function, is the state parameter vector of the device; Operation guidance unit, used to dynamically display operation steps through AR interface , the formula is: , in, For step content, The real-time status of the device.

[0027] The equipment status mapping unit displays the operating status of the equipment in real time, so that the operator can more intuitively understand the real-time status changes of the equipment. Visual prompts such as color, size and animation can promptly reflect equipment abnormalities or changes in operating conditions, and quickly discover and deal with problems. The real-time mapping of equipment status makes it easier for operators to understand and judge equipment status without relying on traditional numbers or charts, thereby improving the intuitiveness and efficiency of operations. The operation guidance unit effectively reduces the risk of operator error by dynamically displaying the operation steps in the AR interface, especially in complex operations and high-risk environments. Operators can accurately complete various tasks according to the step-by-step prompts and can effectively perform them in emergency or unfamiliar scenarios. Dynamic operation guidance improves the efficiency of task completion and operation accuracy, making the operation process safer and more controllable, and is suitable for industrial scenarios with high precision requirements.

[0028] The multimodal interaction module includes: The gesture recognition unit captures gesture information through a 3D depth camera, uses a convolutional neural network to recognize gestures, and outputs control commands , where g represents the gesture feature vector.

[0029] The gesture recognition unit provides operators with a convenient non-contact control method, which is particularly suitable for industrial environments with high noise or inconvenient voice operation. Adding gesture recognition on the basis of traditional operation mode helps to improve the flexibility of operation, reduce dependence on physical input devices, and reduce the risk of misoperation. In addition, gesture recognition supports a variety of gesture commands to adapt to different operation requirements, helping operators to complete complex control tasks more intuitively, making the system highly adaptable and user-friendly. In emergency or manual input situations, gesture recognition provides an efficient and fast means of interaction, which helps to improve work efficiency and safety.

[0030] The integrated modules of the Industrial Internet of Things platform include: A data acquisition unit is used to collect the operating status of the equipment through equipment sensors and ensure data synchronization through a time synchronization protocol; The data analysis unit analyzes and predicts the equipment operation data based on the machine learning model. The prediction formula is: , in, is the machine learning prediction function, is the current device status, For the prediction time.

[0031] The data acquisition unit enables the system to monitor the status of equipment in real time, providing accurate data support for remote operation and analysis. The time synchronization protocol ensures the time sequence consistency of various types of data, thereby improving the accuracy and integrity of the data, laying the foundation for subsequent intelligent analysis and fault prediction. Through the collection and synchronization of real-time data, the system can better adapt to the dynamically changing production environment, support cross-device collaborative management, and effectively improve the operation stability and management efficiency of the equipment; The data analysis unit uses machine learning algorithms to perform intelligent analysis and trend prediction of equipment status, and can proactively identify potential equipment failure risks. The system can issue early warnings before equipment failures occur, provide maintenance recommendations, and significantly reduce downtime risks and maintenance costs. At the same time, the prediction function of the data analysis unit can optimize operating parameters based on the historical operating data of the equipment, thereby improving equipment efficiency and production reliability. Overall, the data analysis unit gives the system intelligent decision-making capabilities, supports predictive maintenance and proactive optimization of equipment, and greatly improves the stability and safety of industrial processes.

[0032] The remote control module includes: The remote command generation unit generates remote control commands through voice, and transmits the commands to the remote device for execution through the industrial Internet of Things platform. The remote control commands are generated by the following formula: , in, Indicates remote control instructions. For operation actions, is the device name, is the operation parameter.

[0033] The remote command generation unit provides operators with a convenient remote control method, allowing operators to directly generate control instructions through voice and execute them remotely. In production scenarios, operators can control equipment in real time without arriving at the site, greatly improving management flexibility and response speed. Especially in emergency situations or scenarios where equipment is widely distributed, the unit provides a more efficient control method. At the same time, instructions are transmitted through the IIoT platform to ensure the real-time and accuracy of instruction transmission, which helps to achieve unified management and collaborative operation across regions and devices, thereby improving the efficiency and reliability of the production process.

[0034] In the speech recognition unit, time series modeling is performed based on the hidden Markov model, where the hidden Markov model is implemented by the following formula: The Hidden Markov Model defines the generation process of speech signals as follows: State Collection : Implicit speech state set, each Indicates the voice status; Observation Collection : The characteristic sequence of observations, each Represents the speech features at time t; State transition probability matrix : Defined by state Transfer to state The probability is expressed as: , Observation probability matrix : Defined in state Generate observations The probability is expressed as: , Initial state probability vector :Define the system to be in state at the initial moment The probability is expressed as: .

[0035] The gesture recognition unit is combined with voice input, and the operator can determine the equipment by voice and confirm the operation or adjust the equipment parameters by gesture.

[0036] In the speech recognition unit, the hidden Markov model is used for time series modeling, so that the speech recognition system has the ability to handle time dependence and can stably recognize speech sequences in a noisy environment. The state transition and observation probability matrix of the hidden Markov model enable the system to adapt to the dynamic changes of continuous speech and effectively improve the recognition accuracy. In addition, the modeling method of the hidden Markov model can convert complex speech input into an easy-to-process probability sequence, making the system more robust in actual operation and able to more accurately parse the operator's voice instructions; In the gesture recognition unit, the combination of gesture recognition and voice input provides operators with a flexible and adaptable control method, which is particularly suitable for industrial environments with high noise or limited manual operation convenience. Voice input can quickly identify equipment and preliminary instructions, and gestures provide intuitive, contactless operation confirmation and adjustment methods, reducing the risk of misoperation. In emergency scenarios or complex operations, multimodal interaction can improve the operator's work efficiency and accuracy, significantly improve the system's ease of use and interactive experience, and make the industrial control process smoother and more efficient.

[0037] An industrial control configuration software method based on intelligent voice interaction comprises the following steps: Step 1: Receive the operator's voice signal through the voice recognition module and convert it into text data, and extract voice features using Mel frequency cepstral coefficients; Step 2: Parse the text data through the natural language processing module and use the Bi-LSTM model to extract the device name, operation instructions and parameters in the operation command; Step 3: The augmented reality interaction module displays the 3D model and real-time status of the equipment in the operator's AR device, and provides operation guidance and feedback according to the operator's voice instructions; Step 4: Use the multimodal interaction module to combine voice, gestures and touch screen to operate the device, where gestures are recognized by convolutional neural networks and control commands are output ; Step 5: Collect equipment operation data through the industrial Internet of Things platform and use machine learning algorithms to analyze and predict the data; Step 6: Generate a remote control command through the remote control module and send it to the remote device. The remote command is generated and executed through the remote control instruction.

[0038] MFCC feature extraction technology enables the system to effectively extract core information from complex speech signals, improving the accuracy and robustness of speech recognition. Especially in noisy industrial environments, the system can accurately identify instructions, laying a solid foundation for subsequent operation analysis; The bidirectional structure of Bi-LSTM can capture the contextual information of a sentence, which helps the system understand complex semantic structures and accurately identify the key content of operation instructions. Its parsing method effectively improves the system's understanding of voice instructions, supports the diverse needs of operators, and makes the system more adaptable. AR technology provides operators with intuitive equipment status display and operation feedback, allowing operators to monitor equipment in real time and obtain dynamic operation guidance. Through the three-dimensional visualization effect of AR, the system significantly improves operators' understanding and control perception of equipment status, especially in complex or remote environments, greatly reducing the risk of misoperation; Multimodal interaction provides flexible control methods, allowing operators to determine the most suitable operation mode in different environments. In noisy or visually limited environments, gesture recognition provides non-contact operation methods, making the system more adaptable, improving user experience and efficiency, and ensuring operation accuracy in emergency or special situations. The combination of IIoT platform and machine learning enables the system to intelligently monitor the status of equipment and perform predictive maintenance, detect potential failures in advance and provide optimization suggestions, improve the operational safety and stability of equipment, reduce maintenance costs, ensure efficient and continuous operation of equipment, and significantly improve the management intelligence level of the system; The remote control module allows operators to generate control commands through voice at a remote location and send them to the device for execution, realizing real-time cross-regional equipment management. Its remote control function provides flexibility and convenience for industrial field operations, and is particularly suitable for distributed equipment management and emergency response scenarios, greatly improving the efficiency and response speed of equipment management.

[0039] The industrial control configuration software method of the present invention realizes a highly intelligent and adaptable industrial control process from voice input to AR interaction, from multimodal control to remote equipment management. Through the deep integration of voice recognition and natural language processing, augmented reality, IIoT and machine learning technologies, the method improves the convenience of operation and the intelligence level of the system, and significantly enhances the efficiency, safety and flexibility of equipment management.

[0040] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An industrial control configuration software system based on intelligent voice interaction, characterized in that: include: The speech recognition module, based on a hybrid architecture of deep neural networks and hidden Markov models, is used to receive the operator's voice signals and convert them into text data; The natural language processing module is used to parse the operation commands in the text data and extract the device name, operation instructions, device status parameters, operation settings and execution conditions through the bidirectional long short-term memory network model; Augmented reality interaction module, which is used to display the 3D virtual model of industrial equipment and its real-time status through AR devices to provide operation guidance and feedback; Multimodal interaction module, used to combine voice, gesture and touch screen to realize various forms of interactive operations; Industrial Internet of Things platform integration module, used to collect operating data of industrial equipment and synchronize it to the cloud for analysis and processing; The remote control module is used to generate remote control commands through voice and send commands to remote devices through the industrial Internet of Things platform to achieve remote operation.

2. The industrial control configuration software system based on intelligent voice interaction according to claim 1 is characterized in that: The speech recognition module comprises: The speech feature extraction unit extracts the features of the speech signal through the Mel frequency cepstral coefficients. The formula is: , in, is the speech signal, is the fast Fourier transform, is the discrete cosine transform; The speech recognition unit performs time series modeling based on the hidden Markov model and classifies the features of speech frames through a deep neural network.

3. The industrial control configuration software system based on intelligent voice interaction according to claim 1 is characterized in that: The natural language processing module extracts the device name, action and parameters in the operation command based on the bidirectional long short-term memory network structure, which is specifically implemented by the following formula: ， in, represents the hidden layer state at time t, is the input word sequence, Represents the concatenation operation of the forward and backward LSTM network outputs.

4. The industrial control configuration software system based on intelligent voice interaction according to claim 1 is characterized in that: The augmented reality interaction module includes: The device status mapping unit converts the device operation status parameters The three-dimensional virtual model mapped to the AR device, the device status is visualized through color, size or animation changes, and the mapping function is: , in, is the mapping function, is the state parameter vector of the device; Operation guidance unit, used to dynamically display operation steps through AR interface , the formula is: ， in, For step content, The real-time status of the device.

5. The industrial control configuration software system based on intelligent voice interaction according to claim 1 is characterized in that: The multimodal interaction module includes: The gesture recognition unit captures gesture information through a 3D depth camera, uses a convolutional neural network to recognize gestures, and outputs control commands , where g represents the gesture feature vector; The touch interaction unit determines and controls the device by combining the touch screen with voice input.

6. The industrial control configuration software system based on intelligent voice interaction according to claim 1 is characterized in that: The industrial Internet of Things platform integration module includes: A data acquisition unit is used to collect the operating status of the equipment through equipment sensors and ensure data synchronization through a time synchronization protocol; The data analysis unit analyzes and predicts the equipment operation data based on the machine learning model. The prediction formula is: , in, is the machine learning prediction function, is the current device status, For the prediction time.

7. The industrial control configuration software system based on intelligent voice interaction according to claim 1 is characterized in that: The remote control module comprises: The remote command generation unit generates remote control commands through voice, and transmits the commands to the remote device for execution through the industrial Internet of Things platform. The remote control commands are generated by the following formula: ， in, Indicates remote control instructions. For operation actions, is the device name, is the operation parameter.

8. The industrial control configuration software system based on intelligent voice interaction according to claim 2 is characterized in that: In the speech recognition unit, time series modeling is performed based on a hidden Markov model, wherein the hidden Markov model is implemented by the following formula: The Hidden Markov Model defines the generation process of speech signals as follows: State Collection : Implicit speech state set, each Indicates the voice status; Observation Collection : The characteristic sequence of observations, each Represents the speech features at time t; State transition probability matrix : Defined by state Transfer to state The probability is expressed as: ， Observation probability matrix : Defined in state Generate observations The probability is expressed as: ， Initial state probability vector :Define the system to be in state at the initial moment The probability is expressed as: .

9. The industrial control configuration software system based on intelligent voice interaction according to claim 5 is characterized in that: The gesture recognition unit is combined with voice input, and the operator can determine the device by voice and confirm the operation or adjust the device parameters by gesture.

10. An industrial control configuration software method based on intelligent voice interaction, according to an industrial control configuration software system based on intelligent voice interaction according to any one of claims 1 to 9, characterized in that: The steps include: Step 1: Receive the operator's voice signal through the voice recognition module and convert it into text data, and extract voice features using Mel frequency cepstral coefficients; Step 2: Parse the text data through the natural language processing module and use the Bi-LSTM model to extract the device name, operation instructions and parameters in the operation command; Step 3: The augmented reality interaction module displays the 3D model and real-time status of the equipment in the operator's AR device, and provides operation guidance and feedback according to the operator's voice instructions; Step 4: Use the multimodal interaction module to combine voice, gestures and touch screen to operate the device, where gestures are recognized by convolutional neural networks and control commands are output ; Step 5: Collect equipment operation data through the industrial Internet of Things platform and use machine learning algorithms to analyze and predict the data; Step 6: Generate a remote control command through the remote control module and send it to the remote device. The remote command is generated and executed through the remote control instruction.

Images