Equipment fault monitoring system and method based on machine learning and industrial Internet of Things

Abstract

The invention discloses an equipment fault monitoring system and method based on machine learning and industrial Internet of Things, and the method comprises the steps: collecting equipment operation parameter data, including temperature, pressure, vibration and other parameters, in real time through an Internet of Things sensor network, carrying out the data cleaning and normalization processing of the equipment operation parameter data through a data preprocessing module, and carrying out the data cleaning and normalization processing; key features are extracted; and a fault prediction model is constructed, a potential fault mode in equipment operation is intelligently and automatically identified, and an early warning signal is generated. Meanwhile, maintenance suggestions are automatically generated or equipment operation parameters are adjusted according to the prediction result, so that the service life of equipment is prolonged, and energy consumption is reduced. The system provided by the invention has the capabilities of high-precision fault prediction, real-time monitoring and early warning and intelligent adaptive learning, can effectively reduce the downtime of equipment, and improves the efficiency and robustness of industrial production.

Description

Technical Field

[0001] The present invention belongs to the field of industrial automation and intelligent manufacturing technology, and in particular relates to an equipment fault monitoring system and method based on machine learning and industrial Internet of Things. Background Art

[0002] With the rapid development of the Industrial Internet of Things (IIoT) and artificial intelligence (AI) technologies, intelligent management of industrial equipment has become crucial for improving production efficiency and reducing maintenance costs. However, traditional equipment management methods rely on scheduled maintenance or emergency repairs following failures, which is inefficient and ineffective in preventing sudden failures. While existing technologies have incorporated machine learning for fault prediction, they still lack data privacy, model adaptability, and real-time performance. Summary of the Invention

[0003] This paper aims to address existing issues in industrial equipment fault management by proposing a real-time monitoring and intelligent prediction system and method for industrial IoT equipment faults based on machine learning. This system collects real-time equipment operating data, uses machine learning algorithms to predict faults, and generates maintenance recommendations to improve equipment reliability and operational efficiency.

[0004] To achieve the above objectives, the present invention provides a machine learning and industrial Internet of Things-based equipment fault monitoring system and method. Specifically, a machine learning and industrial Internet of Things-based equipment fault monitoring system includes:

[0005] Data acquisition module, used to collect equipment operating parameter data;

[0006] A data preprocessing module, connected to the data acquisition module, for performing data cleaning and normalization on the equipment operating parameter data and extracting key features related to equipment failures;

[0007] A model building module, connected to the data preprocessing module, is used to build a machine learning model and train the machine learning model using the preprocessed data to obtain a fault prediction model;

[0008] A fault prediction module, connected to the model building module, is used to predict faults of the equipment based on the fault prediction model to obtain a fault prediction result; and to generate a fault warning signal according to the fault prediction result;

[0009] The feedback and decision module is connected to the fault prediction module and is used to generate maintenance suggestions based on the fault prediction results and dynamically adjust the operating parameters of the equipment.

[0010] Preferably, the data acquisition module includes a distributed control system and a sensor;

[0011] The distributed control system module is connected to the feedback and decision module to monitor the operating status of the equipment in real time and automatically control the equipment according to the instructions of the feedback and decision module;

[0012] The sensors are distributed in key parts of industrial equipment to monitor the operating status of the equipment in real time and obtain equipment operating parameter data;

[0013] The equipment operating parameter data includes temperature, pressure, vibration, and current.

[0014] Preferably, the machine learning model adopts Transformer-based Anomaly Detection Models and combines Federated Learning with Differential Privacy technology to allow distributed training of models on multiple devices while protecting data privacy; automatically extract abnormal patterns in device operation data and perform adaptive learning based on newly collected data.

[0015] Preferably, the fault prediction module generates a fault warning signal in real time according to the probability threshold of equipment failure and notifies maintenance personnel. At the same time, based on the equipment's operating history data, it predicts the remaining service life of the equipment and automatically generates a maintenance plan based on the actual needs of the equipment and pushes it to the maintenance personnel.

[0016] Preferably, the feedback and decision module is integrated with the distributed control system to convert the fault prediction results into specific control instructions;

[0017] The feedback and decision module is integrated with the user interface to display the operating status of the equipment, fault prediction results and maintenance suggestions in real time.

[0018] Preferably, the adjusting device operating parameters includes:

[0019] Analyze the equipment's operating trends based on the equipment's historical operating data and fault prediction results, and optimize the equipment's operating parameters based on the trends.

[0020] When the predicted probability of equipment overheating exceeds a preset probability threshold, the equipment's operating speed is automatically reduced or the cooling system parameters are adjusted;

[0021] When the equipment is in good operating condition, adjust the equipment's operating parameters or reduce the power of the cooling system to optimize equipment energy consumption.

[0022] Preferably, the system further comprises a data calibration module and a communication module connected to the data acquisition module;

[0023] The data calibration module is used to periodically calibrate sensor data;

[0024] The communication module is used to transmit equipment operation data and fault prediction results to the cloud server for further analysis.

[0025] Preferably, the system further comprises a data storage module, and the data storage module is used to store the collected equipment operation data and fault prediction results.

[0026] Preferably, the system further comprises a user interface for displaying equipment operating status, fault prediction results and maintenance recommendations.

[0027] The present invention also provides a method for monitoring equipment failures based on machine learning and industrial Internet of Things, comprising:

[0028] Collect equipment operating parameter data of industrial equipment in real time through the IoT sensor network;

[0029] Clean and normalize the collected equipment operating parameter data and extract key features related to equipment failures;

[0030] Constructing a machine learning model and training the machine learning model using the preprocessed data to obtain a fault prediction model;

[0031] Performing fault prediction on the equipment based on the fault prediction model to obtain a fault prediction result; and generating a fault warning signal according to the fault prediction result;

[0032] Generate maintenance recommendations based on the fault prediction results and dynamically adjust operating parameters of the equipment;

[0033] Monitor the equipment's operating status in real time and automatically control the equipment according to instructions.

[0034] Compared with the prior art, the present invention has the following advantages and technical effects:

[0035] The system of the present invention has high-precision fault prediction, real-time monitoring and early warning, and intelligent adaptive learning capabilities, which can effectively reduce equipment downtime and improve the efficiency and robustness of industrial production.

[0036] The present invention collects equipment operation data in real time, uses machine learning algorithms to predict faults, and generates maintenance recommendations to improve equipment reliability and operating efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of this application. The exemplary embodiments and descriptions of this application are intended to explain this application and do not constitute an improper limitation on this application. In the accompanying drawings:

[0038] Figure 1 Schematic diagram of the system structure of an embodiment of the present invention. DETAILED DESCRIPTION

[0039] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0040] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0041] Example 1

[0042] like Figure 1 As shown, this embodiment provides an equipment fault monitoring system based on machine learning and industrial Internet of Things, including:

[0043] Data acquisition module, used to collect equipment operating parameter data;

[0044] The data preprocessing module is connected to the data acquisition module and is used to clean and normalize the equipment operating parameter data and extract key features related to equipment failures;

[0045] The model building module is connected to the data preprocessing module to build a machine learning model and train the machine learning model with the preprocessed data to obtain a fault prediction model;

[0046] The fault prediction module is connected to the model building module and is used to predict the fault of the equipment based on the fault prediction model, obtain the fault prediction result, and generate a fault warning signal according to the fault prediction result;

[0047] The feedback and decision module is connected to the fault prediction module and is used to generate maintenance suggestions based on the fault prediction results and dynamically adjust the operating parameters of the equipment.

[0048] The equipment fault monitoring system of this embodiment realizes real-time monitoring and fault prediction of industrial equipment through the collaborative work of the data acquisition module, data preprocessing module, model building module, fault prediction module, feedback and decision module, and distributed control system module.

[0049] Furthermore, the data acquisition module includes a distributed control system and sensors;

[0050] The distributed control system module is connected to the feedback and decision-making module to monitor the equipment operation status in real time and automatically control the equipment according to the instructions of the feedback and decision-making module;

[0051] Sensors are distributed in key locations of industrial equipment to monitor the operating status of the equipment in real time and obtain data on its operating parameters. For example, they can measure furnace flue gas temperature, pipe vibration, air supply duct pressure, and residual oxygen concentration in exhaust ducts during metallurgical processes. These sensors can monitor the operating status of the equipment in real time and support a variety of sensor types to meet the monitoring needs of different industrial equipment.

[0052] Among them, equipment operating parameter data includes temperature, pressure, vibration, and current.

[0053] To further optimize the solution, the data acquisition module supports wireless and wired communication methods and can flexibly adapt to different industrial environments.

[0054] To further optimize the solution, the data acquisition module supports custom sensor configuration, and users can select and configure the sensor type and quantity according to actual needs.

[0055] To further optimize the solution, the data acquisition module can dynamically adjust the sampling frequency of the sensor according to the actual needs of the equipment to optimize data acquisition efficiency. The relevant formula for adjusting the sampling frequency is:

[0056]

[0057] Where f is the original sampling frequency, f' is the adjusted sampling frequency, β is the adjustment coefficient, P(failure) is the probability of equipment failure, P base is the baseline probability of equipment failure.

[0058] To further optimize the solution, the distributed control system module can be integrated with the feedback and decision-making module to automatically adjust the equipment's operating parameters based on fault prediction results, thereby extending equipment life and reducing energy consumption. For example, if the fault prediction module predicts that a device may overheat, the distributed control system module can automatically adjust the device's operating speed or cooling system parameters to prevent the failure.

[0059] The main functions of the distributed control system module include:

[0060] (1) Real-time monitoring of the equipment's operating status, including key parameters such as temperature, pressure, and vibration.

[0061] (2) Automatically adjust the equipment's operating parameters, such as speed and power, according to the instructions of the feedback and decision-making module.

[0062] (3) Work in conjunction with the fault prediction module to achieve seamless integration of fault warning and automated control.

[0063] (4) Provide a visual interface for equipment operation, so that operators can monitor and manage equipment status in real time.

[0064] By integrating the distributed control system module into the fault prediction system, this embodiment can achieve the following advantages:

[0065] (1) The module can automatically adjust equipment parameters according to fault prediction results, reduce manual intervention, and improve the automation level of the system.

[0066] (2) The distributed control system module can monitor the operating status of the equipment in real time, detect potential problems in time and take measures.

[0067] (3) The distributed control system module and the fault prediction module work together to achieve seamless connection between fault warning and automatic control, thereby improving the overall performance and reliability of the system.

[0068] In a specific embodiment, this embodiment can be applied to fault prediction of industrial machine tools. By analyzing machine tool vibration data and current changes, spindle failures can be predicted in advance, thus reducing downtime.

[0069] The system of this embodiment supports multiple types of sensors and can adapt to the monitoring needs of different industrial equipment. Sensor nodes are distributed in key parts of the equipment and can monitor the operating status of the equipment in real time.

[0070] Furthermore, the data preprocessing module cleans and normalizes the collected equipment operating parameter data and extracts key features. This process automatically identifies and removes transient, accidental abnormal data points, while normalizing the data to improve data quality and extract characteristic parameters related to equipment failures, providing reliable data input for subsequent machine learning models. The formula for normalization is:

[0071]

[0072] Among them, x is the original data, x ' is the normalized data, min(x) and max(x) are the minimum and maximum values ​​of the data respectively.

[0073] Furthermore, the machine learning model uses Transformer-based Anomaly Detection Models and combines Federated Learning with Differential Privacy technology to allow distributed model training on multiple devices while protecting data privacy; automatically extract abnormal patterns in device operation data and perform adaptive learning based on newly collected data.

[0074] The machine learning model uses a deep learning algorithm that can automatically extract features from equipment operation data and perform adaptive learning based on newly collected data to optimize the accuracy of fault prediction.

[0075] Machine learning models can automatically adjust model parameters based on the type of device and operating environment to improve prediction performance.

[0076] To further optimize the solution, a machine learning model uses Transformer-based Anomaly Detection Models to model and analyze preprocessed data, automatically extracting abnormal patterns in equipment operation data and adaptively learning based on newly collected data to optimize fault prediction accuracy. This model captures abnormal patterns in time series data through a self-attention mechanism, effectively detecting potential faults during equipment operation. Its core formula is as follows:

[0077] (1) Self-Attention Mechanism:

[0078]

[0079] Among them, Q, K, V are query, key, and value matrices respectively; d k is the dimension of the key vector; softmax is a normalization function used to convert the output into a probability distribution.

[0080] (2) Multi-Head Attention Mechanism:

[0081] MultiHead(Q,K,V)=Concat(head1,…,head h )W O

[0082]

[0083] Where h is the number of attention heads; and is the weight matrix of each attention head; W O is the output linear projection matrix.

[0084] (3) Anomaly detection mechanism: Determine whether there is an anomaly by calculating the reconstruction error of time series data. The reconstruction error formula is:

[0085] Error(x)=||x-Reconstruct(x)||2

[0086] When the reconstruction error exceeds a preset threshold, the system generates a fault warning signal.

[0087] To further optimize the solution and enhance data privacy and distributed model training capabilities, the system incorporates Federated Learning with Differential Privacy technology, allowing for distributed model training across multiple devices while protecting data privacy. Federated learning allows multiple devices to train models locally while using differential privacy to protect data privacy and prevent data leaks.

[0088] Federated Learning updates model parameters locally on multiple devices and aggregates the updated parameters into the global model. The update formula is:

[0089]

[0090] Among them, θ global is the global model parameter; θ i is the local model parameter of the i-th device; α i is the weight coefficient of device i, which is usually proportional to the amount of data on the device; N is the total number of devices participating in the training.

[0091] To protect data privacy, the system uses differential privacy technology to inject noise into local model updates. The specific formula is:

[0092]

[0093] Among them, θ' i are the local model parameters after noise injection; The mean is 0 and the variance is σ 2 σ is the noise standard deviation, which is used to control the degree of privacy protection.

[0094] To further optimize the solution, the machine learning model supports both online and offline learning modes, which can be flexibly adjusted according to actual needs. Online learning can be expressed as follows:

[0095]

[0096] Where W is the weight of the model, W' is the updated weight, η is the learning rate, is the gradient of the loss function.

[0097] To further optimize the solution, the machine learning model supports multiple deep learning frameworks, including TensorFlow and PyTorch, and can choose the appropriate framework for development according to user needs.

[0098] To further optimize the solution, the data preprocessing module supports multiple data formats and can process data from different types of sensors.

[0099] Furthermore, the fault prediction module generates a fault warning signal in real time based on the probability threshold of equipment failure and notifies maintenance personnel. At the same time, based on the equipment's operating history data, it predicts the equipment's remaining service life and automatically generates a maintenance plan based on the equipment's actual needs and pushes it to maintenance personnel.

[0100] The probability of failure can be calculated using the following formula:

[0101] P(fault)=σ(Wx+b)

[0102] Where P(failure) is the probability of a device failure and σ is the sigmoid function:

[0103]

[0104] When P(fault) exceeds a preset threshold (such as 0.8), the system generates a fault warning signal.

[0105] The feedback and decision module dynamically adjusts the device's operating parameters based on the fault prediction results. For example, if the device is predicted to overheat, the device's operating speed can be reduced, as expressed in the following formula:

[0106] v′=v-α·sgn(TT max )

[0107] Where v is the original operating speed of the device, v' is the adjusted operating speed, α is the adjustment coefficient, T is the current temperature of the device, and T max is the maximum allowable temperature of the device, and sgn(x) is the sign function.

[0108] To further optimize the solution, the fault prediction module can predict the remaining useful life (RUL) of the equipment based on the equipment's operating history data, and automatically generate a maintenance plan based on the actual needs of the equipment and push it to the maintenance personnel. The relevant formula is:

[0109]

[0110] To further optimize the solution, the fault prediction module supports multiple warning methods, including SMS, email and system notifications.

[0111] Furthermore, the feedback and decision module is integrated with the distributed control system to convert fault prediction results into specific control instructions; for example, when it is predicted that the equipment may overheat, the feedback and decision module can send instructions to the distributed control system module to automatically adjust the equipment's operating speed or the parameters of the cooling system to prevent the occurrence of faults.

[0112] The Feedback and Decision Module, integrated with the user interface, displays real-time equipment status, fault predictions, and maintenance recommendations. Operators can view the real-time status of equipment through the user interface and, based on the Feedback and Decision Module's recommendations, manually intervene or confirm automatically adjusted parameters.

[0113] To further optimize the solution, the feedback and decision-making module can dynamically adjust the operating parameters of the equipment according to the actual needs of the equipment to extend the equipment life and reduce energy consumption.

[0114] To further optimize the solution, the feedback and decision-making module can dynamically adjust the operating parameters of the equipment based on the fault prediction results to extend the equipment life and reduce energy consumption, and automatically generate maintenance plans and push them to maintenance personnel.

[0115] For example, if the fault prediction module predicts a high probability of equipment overheating, the feedback and decision module can automatically reduce the equipment's operating speed or adjust the cooling system parameters to prevent the failure. The adjustment formula is as follows:

[0116]

[0117] Among them, v new is the adjusted running speed; v old is the original running speed; β is the adjustment coefficient (e.g. 0.1); T current is the current temperature; T threshold is the maximum allowable temperature of the device.

[0118] To further optimize the solution, the feedback and decision-making module can also optimize equipment energy consumption based on fault prediction results. For example, by adjusting equipment operating parameters, unnecessary energy consumption can be reduced, while also extending equipment life. For example, when equipment is operating well, the cooling system power can be appropriately reduced to save energy.

[0119] To further optimize the solution, the Feedback and Decision Module combines historical equipment operating data with fault prediction results to analyze equipment operating trends and optimize equipment operating parameters based on these trends. For example, if a piece of equipment frequently overheats, the Feedback and Decision Module can adjust the equipment's operating parameters to reduce the likelihood of overheating.

[0120] To further optimize the plan, the feedback and decision-making module supports integration with the enterprise resource planning (ERP) system, and can directly push maintenance plans to relevant managers.

[0121] To further optimize the solution, the workflow of the feedback and decision module includes:

[0122] (1) The feedback and decision module receives the probability of equipment failure, remaining useful life (RUL) and related failure mode information from the fault prediction module.

[0123] (2) Based on the fault prediction results, the feedback and decision module evaluates the current status and potential risks of the equipment.

[0124] (3) If the probability of failure exceeds a preset threshold (e.g., 0.8), an emergency maintenance notice is generated and the equipment operating parameters are adjusted to prevent failure. If the equipment is operating well but there are potential risks, a preventive maintenance plan is generated. The generated maintenance recommendations or parameter adjustment instructions are sent to the distributed control system module, which then performs the specific control operations.

[0125] (4) Record the adjusted operating parameters and maintenance recommendations in the data storage unit for subsequent analysis and model optimization.

[0126] (5) The equipment’s operating status, fault prediction results, and maintenance recommendations are displayed in real time on the user interface for operators to monitor and manage.

[0127] Furthermore, adjusting the equipment operating parameters includes:

[0128] Combine the equipment's historical operating data and fault prediction results to analyze the equipment's operating trends and optimize the equipment's operating parameters based on the trends;

[0129] When the predicted probability of equipment overheating exceeds a preset probability threshold, the equipment's operating speed is automatically reduced or the cooling system parameters are adjusted;

[0130] When the equipment is in good operating condition, adjust the equipment's operating parameters or reduce the power of the cooling system to optimize equipment energy consumption.

[0131] Furthermore, the system also includes a data calibration module and a communication module connected to the data acquisition module;

[0132] The data calibration module is used to regularly calibrate sensor data to ensure data accuracy and reliability. Data calibration is expressed by the following formula:

[0133] x'=x+δ

[0134] Where x is the original data, x' is the calibrated data, and δ is the calibration deviation.

[0135] The communication module is used to transmit equipment operation data and fault prediction results to the cloud server for further analysis.

[0136] To further optimize the solution, the communication module supports multiple communication protocols, including industrial Ethernet, Modbus and OPC UA, and can be seamlessly connected with existing industrial systems.

[0137] Furthermore, the system also includes a data storage module, which is used to store the collected equipment operation data and fault prediction results for subsequent analysis and model optimization.

[0138] To further optimize the solution, the data storage module supports local storage and cloud storage, and can be flexibly configured according to user needs.

[0139] Furthermore, the system also includes a user interface for displaying the equipment's operating status, fault prediction results, and maintenance recommendations. This interface facilitates operators to monitor and manage equipment status in real time.

[0140] To further optimize the solution, the user interface supports multiple terminal devices, including computers, tablets and mobile phones, allowing operators to monitor equipment status anytime and anywhere.

[0141] Furthermore, the system also includes a data security module for protecting the privacy and security of equipment operation data and fault prediction results.

[0142] Furthermore, the system also includes a system monitoring module for real-time monitoring of the system's operating status to ensure the system's stability and reliability.

[0143] The system of this embodiment can realize intelligent management of industrial equipment, improve the operating efficiency and reliability of equipment, reduce maintenance costs and downtime, and is particularly suitable for equipment failure prediction and health management in the manufacturing industry, energy industry, transportation and other fields.

[0144] The system of this embodiment can improve equipment reliability in industrial production, reduce maintenance costs and optimize production efficiency through real-time data collection, intelligent analysis and fault prediction.

[0145] This embodiment may also involve the integrated application of DCS, and realize automatic monitoring and control of equipment operation by combining with a fault prediction system.

[0146] This embodiment may also involve equipment fault detection in the cross-application field of artificial intelligence and the Internet of Things, specifically using deep learning algorithms to model and analyze the operating data of industrial Internet of Things devices to achieve early intelligent warning and health management of equipment failures.

[0147] This embodiment is applicable to a variety of industrial scenarios, including but not limited to the manufacturing industry, energy industry, and transportation field, and can effectively improve the intelligent management level of equipment and reduce unexpected downtime.

[0148] Example 2: Industrial Machine Tool Spindle Fault Prediction

[0149] 1. Install a variety of sensors at key locations on the machine tool spindle, including vibration sensors, current sensors, and temperature sensors, to collect operating data in real time.

[0150] 2. The system cleans and normalizes the collected data to remove abnormal data points.

[0151] 3. Extract characteristic parameters related to spindle faults, such as the rate of change of vibration frequency and the fluctuation amplitude of current.

[0152] 4. Use Transformer-based Anomaly Detection Models to model and analyze the preprocessed data. The self-attention mechanism is used to capture abnormal patterns in time series data and calculate the reconstruction error:

[0153] Error(x)=||x-Reconstruct(x)||2

[0154] When the reconstruction error exceeds a preset threshold (such as 0.1), the system generates a fault warning signal.

[0155] 5. The distributed control system module automatically adjusts the machine tool's operating speed based on the fault prediction results. The adjustment formula is as follows:

[0156]

[0157] Among them, v new is the adjusted running speed; v old is the original running speed; β is the adjustment coefficient (such as 0.1); T current is the current temperature; T threshold is the maximum allowable temperature.

[0158] 6. By predicting spindle failures in advance, machine downtime is reduced, production efficiency is improved, and maintenance costs caused by sudden failures are reduced.

[0159] Example 3

[0160] Based on the same inventive concept, this embodiment also provides a device fault monitoring method based on machine learning and industrial Internet of Things, including:

[0161] Collect equipment operating parameter data of industrial equipment in real time through the IoT sensor network;

[0162] Clean and normalize the collected equipment operating parameter data and extract key features related to equipment failures;

[0163] Build a machine learning model and train it using preprocessed data to obtain a fault prediction model;

[0164] Predict equipment failures based on the failure prediction model to obtain failure prediction results; and generate failure warning signals based on the failure prediction results;

[0165] Generate maintenance recommendations based on fault prediction results and dynamically adjust equipment operating parameters;

[0166] Monitor the equipment's operating status in real time and automatically control the equipment according to instructions.

[0167] The above are merely preferred embodiments of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A device fault monitoring system based on machine learning and industrial Internet of Things, characterized by: include: Data acquisition module, used to collect equipment operating parameter data; A data preprocessing module, connected to the data acquisition module, for performing data cleaning and normalization on the equipment operating parameter data and extracting key features related to equipment failures; A model building module, connected to the data preprocessing module, is used to build a machine learning model and train the machine learning model using the preprocessed data to obtain a fault prediction model; A fault prediction module, connected to the model building module, is used to predict faults of the equipment based on the fault prediction model and obtain a fault prediction result; and generating a fault warning signal according to the fault prediction result; The feedback and decision module is connected to the fault prediction module and is used to generate maintenance suggestions based on the fault prediction results and dynamically adjust the operating parameters of the equipment.

2. The system according to claim 1, wherein: The data acquisition module includes a distributed control system and a sensor; The distributed control system module is connected to the feedback and decision module, and is used to monitor the operating status of the equipment in real time and automatically control the equipment according to the instructions of the feedback and decision module; The sensors are distributed in key parts of industrial equipment to monitor the operating status of the equipment in real time and obtain equipment operating parameter data; The equipment operating parameter data includes temperature, pressure, vibration, and current.

3. The system according to claim 1, wherein: The machine learning model uses Transformer-based Anomaly Detection Models and combines Federated Learning with Differential Privacy technology to allow distributed model training on multiple devices while protecting data privacy; automatically extract abnormal patterns in device operation data and perform adaptive learning based on newly collected data.

4. The system according to claim 1, wherein: The fault prediction module generates a fault warning signal in real time based on the probability threshold of equipment failure and notifies maintenance personnel. At the same time, based on the equipment's operating history data, it predicts the equipment's remaining service life and automatically generates a maintenance plan based on the equipment's actual needs and pushes it to maintenance personnel.

5. The system according to claim 1, wherein: The feedback and decision module is integrated with the distributed control system to convert the fault prediction results into specific control instructions; The feedback and decision module is integrated with the user interface to display the operating status of the equipment, fault prediction results and maintenance suggestions in real time.

6. The system according to claim 1, wherein: The adjusting device operating parameters includes: Analyze the equipment's operating trends based on the equipment's historical operating data and fault prediction results, and optimize the equipment's operating parameters based on the trends. When the predicted probability of equipment overheating exceeds a preset probability threshold, the equipment's operating speed is automatically reduced or the cooling system parameters are adjusted; When the equipment is in good operating condition, adjust the equipment's operating parameters or reduce the power of the cooling system to optimize equipment energy consumption.

7. The system according to claim 1, wherein: The system also includes a data calibration module and a communication module connected to the data acquisition module; The data calibration module is used to regularly calibrate sensor data; The communication module is used to transmit equipment operation data and fault prediction results to the cloud server for further analysis.

8. The system according to claim 1, wherein: The system further comprises a data storage module, which is used to store collected equipment operation data and fault prediction results.

9. The system according to claim 1, wherein: The system further includes a user interface for displaying equipment operating status, fault prediction results, and maintenance recommendations.

10. A device fault monitoring method based on machine learning and industrial Internet of Things, characterized in that: include: Collect equipment operating parameter data of industrial equipment in real time through the IoT sensor network; Clean and normalize the collected equipment operating parameter data and extract key features related to equipment failures; Constructing a machine learning model and training the machine learning model using the preprocessed data to obtain a fault prediction model; Performing fault prediction on the equipment based on the fault prediction model to obtain a fault prediction result; and generating a fault warning signal according to the fault prediction result; Generate maintenance recommendations based on the fault prediction results and dynamically adjust operating parameters of the equipment; Monitor the equipment's operating status in real time and automatically control the equipment according to instructions.

Images