Artificial intelligence-based cross-border Internet of Things equipment fault diagnosis and prediction system
The AI-based cross-border IoT device fault diagnosis and prediction system utilizes machine learning and deep learning technologies to achieve efficient and accurate fault identification and prediction, solving the problems of low efficiency and error-proneness in existing technologies, and providing intelligent device optimization and a good user experience.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- GUANGZHOU JINCAIZHILIAN DIGITAL TECH CO LTD
- Filing Date
- 2024-01-31
- Publication Date
- 2026-05-08
AI Technical Summary
Existing methods for diagnosing faults in cross-border IoT devices rely on manual operation, which is inefficient and prone to errors, making it difficult to achieve efficient and accurate fault diagnosis and prediction.
An AI-based cross-border IoT equipment fault diagnosis and prediction system is adopted, including modules such as data acquisition, preprocessing, AI-CIOT fault diagnosis engine, fault diagnosis algorithm library, equipment status monitoring and prediction, equipment fault knowledge graph construction, deep learning capability integration, real-time fault early warning, and equipment maintenance and repair. It utilizes machine learning and deep learning technologies for automated fault identification and prediction.
It improves the accuracy and efficiency of fault diagnosis and prediction, reduces manual intervention, lowers the error rate, and achieves intelligent equipment optimization and a better user experience.
Smart Images
Figure CN121997158A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of cross-border IoT device fault diagnosis and prediction technology, and more specifically, to an artificial intelligence-based cross-border IoT device fault diagnosis and prediction system. Background Technology
[0002] With the development of IoT technology, more and more devices are connecting and exchanging data via the internet. The global distribution of these devices makes fault diagnosis and prediction more complex. Existing cross-border IoT device fault diagnosis processes have some shortcomings that need improvement. Traditional fault diagnosis and prediction methods often rely on manual operation, which is inefficient and prone to errors. Therefore, we propose an AI-based cross-border IoT device fault diagnosis and prediction system to achieve more efficient and accurate fault diagnosis and prediction. Summary of the Invention
[0003] The purpose of this invention is to address the problems raised in the existing background technology. To achieve the above-mentioned objective, this invention provides the following technical solution: a fault diagnosis and prediction system for cross-border IoT devices based on artificial intelligence, including a data acquisition module. The data acquisition module is used to collect operational data from cross-border IoT devices. The data acquisition module is connected to a data preprocessing module. The data collected by the data acquisition module is recorded as: Device = {x1.x2...x...} n};
[0004] The data preprocessing module cleans, standardizes, and extracts features from the collected data for subsequent analysis and prediction. During feature extraction, each device is assigned a feature code: Feature code = {f1, f2, ..., f...} n};
[0005] The data preprocessing module is connected to the AI-CIOT fault diagnosis engine module; the AI-CIOT fault diagnosis engine module processes and analyzes the collected cross-border IoT device data, and then uses pre-stored fault diagnosis algorithms to identify and locate the devices.
[0006] The mathematical formula for fault diagnosis is N = (Ax + By + C) / D;
[0007] Where N: output error (noise); A: numerator of system function (transfer function); B: denominator of system function (transfer function); C: input error (noise); D: common factors of system function (transfer function) (such as gain);
[0008] The AI-CIOT fault diagnosis engine module is connected to the fault diagnosis algorithm library module; the fault diagnosis algorithm library module is equipped with fault diagnosis algorithms, which include traditional statistical methods, machine learning and deep learning algorithms.
[0009] The fault diagnosis algorithm formula is Z=Y*X^(a);
[0010] Where Z: fault characteristic equation; Y: fault characteristic coefficient; X: input signal (such as voltage, current); a: fault characteristic index (reflecting the severity of the fault);
[0011] The AI-CIOT fault diagnosis engine module is connected to the equipment status monitoring and prediction module; the equipment status monitoring and prediction module uses the collected equipment data and methods from the fault diagnosis algorithm library to monitor the status of the equipment in real time.
[0012] The equipment status monitoring and prediction module is connected to the equipment fault knowledge graph construction module. The equipment fault knowledge graph construction module uses the collected equipment data and fault diagnosis results to construct an equipment fault knowledge graph for storing and learning equipment fault modes and patterns.
[0013] The formula for failure mode and pattern is R=Σ(Xi*Xj)^(1 / 2) / (ΣXi^2)^(1 / 2)*ΣXj^(1 / 2); where R: correlation coefficient (reflecting the strength of the linear relationship between two variables); Xi: the observed value of the first variable; Xj: the observed value of the second variable.
[0014] As a preferred technical solution of the present invention, the data collected by the data acquisition module includes equipment parameters, operating status, and environmental parameters. The equipment fault knowledge graph construction module is connected to the deep learning capability integration module. The deep learning capability integration module integrates deep learning algorithms into the fault diagnosis system and uses deep learning models to identify and predict equipment faults.
[0015] As a preferred technical solution of the present invention, the deep learning capability integration module is data-connected to the AI model prediction and optimization module; the AI model prediction and optimization module provides interpretability of the AI model, helping users understand how the AI model performs fault diagnosis and model optimization;
[0016] During model optimization, K = 1 / (1 + e^((a + bX)));
[0017] Where K: optimization function; a: bias term; b: weight; e: base of natural logarithm;
[0018] The AI model prediction and optimization module is connected to the real-time fault warning module; the real-time fault warning module uses the results of equipment status monitoring and prediction to issue fault warnings to users in real time, helping users to take timely measures to prevent equipment failures.
[0019] The prediction formula is (Decision Tree) Y = f(X1,X2,...,Xn);
[0020] Y: Predicted value; X1, X2, ..., Xn: Input features; f: Decision tree model.
[0021] As a preferred technical solution of the present invention, the real-time fault early warning module is data-connected to the equipment maintenance and repair module; the equipment maintenance and repair module provides corresponding equipment maintenance and repair suggestions based on the equipment fault diagnosis results to help users improve equipment operating efficiency and reduce failure rate; the equipment maintenance and repair module is data-connected to the fault diagnosis model optimization and update module; the fault diagnosis model optimization and update module collects new equipment data and fault diagnosis results to optimize and update the fault diagnosis model, so as to improve the accuracy and reliability of the system.
[0022] As a preferred technical solution of the present invention, the fault diagnosis model optimization and update module is data-connected to the system security and privacy protection module; the system security and privacy protection module protects the user's data security and privacy, including data encryption, access control and data desensitization.
[0023] The diagnostic function is Y = f(W1*X1 + W2*X2 + ... + Wn*Xn + b);
[0024] Where Y: predicted value; W1, W2, ..., Wn: weight matrix; X1, X2, ..., Xn: input feature; b: bias term; f: activation function;
[0025] The system security and privacy protection module is connected to the cross-platform device compatibility module; the cross-platform device compatibility module ensures that the system supports various types of cross-border IoT devices, regardless of the communication protocol or data format they use.
[0026] As a preferred technical solution of the present invention, the cross-platform device compatibility module is data-connected to the user interface and interaction module; the user interface and interaction module sets the user interface and interaction mode, so that users can easily use the system to diagnose and predict device faults.
[0027] As a preferred technical solution of the present invention, the user interface and interaction module is connected to the data analysis and visualization module; the data analysis and visualization module uses data analysis and visualization tools to help users understand the operating status of the equipment and the fault diagnosis results more intuitively.
[0028] As a preferred technical solution of the present invention, the data analysis and visualization module and the system performance evaluation and monitoring module are connected in data; the system performance evaluation and monitoring module is responsible for monitoring the performance of the system, including response time, accuracy and other key indicators, and adjusting and optimizing them as needed;
[0029] The adjustment probability formula is P(Y|X)=P(X|Y)*P(Y) / P(X);
[0030] Where P(Y|X): conditional probability given input features; P(X|Y): conditional probability given output; P(Y): prior probability of the class; P(X): prior probability of the input features.
[0031] As a preferred technical solution of the present invention, the system performance evaluation and monitoring module is connected to the equipment fault prediction module; the equipment fault prediction module uses historical equipment data and fault diagnosis results to construct an equipment fault prediction model for predicting future faults.
[0032] As a preferred technical solution of the present invention, the equipment fault prediction module and the data integration and synchronization module are connected; the data integration and synchronization module integrates data from different devices and sources, and synchronizes and updates them to ensure that the system's data and information are up-to-date.
[0033] Compared with existing technologies, the beneficial effects of this invention are as follows: It effectively improves the accuracy of fault diagnosis and prediction: by utilizing machine learning and deep learning technologies, it can more accurately identify equipment fault types and predict the probability of future fault occurrences; it improves the efficiency of fault diagnosis and prediction: by automating data collection, preprocessing, diagnosis, and prediction tasks, it reduces manual intervention and lowers the error rate; it enables intelligent equipment optimization: based on the fault diagnosis and prediction results, it intelligently optimizes the equipment's operating parameters, reducing the probability of fault occurrence and improving the equipment's service life and operating efficiency; and it provides a superior user experience: by displaying fault diagnosis and prediction results, as well as equipment optimization suggestions, through a user-friendly interface, it helps users better understand and grasp the equipment's operating status. Attached Figure Description
[0034] Figure 1 This is a schematic diagram of the logic block of the acquisition and processing module provided by the present invention;
[0035] Figure 2A schematic diagram of the repair and compatibility module logic block provided by this invention;
[0036] Figure 3 This is a schematic diagram of the interaction and synchronization logic provided by the present invention. Detailed Implementation
[0037] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.
[0038] Therefore, the following detailed description of the embodiments of the present invention is not intended to limit the scope of the claimed invention, but merely illustrates some embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention. It should be noted that, in the absence of conflict, the embodiments and features and technical solutions in the embodiments of the present invention can be combined with each other. It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.
[0039] Example 1: Please refer to Figures 1-3 A fault diagnosis and prediction system for cross-border IoT devices based on artificial intelligence includes a data acquisition module for collecting operational data from cross-border IoT devices. The data acquisition module is connected to a data preprocessing module. The data collected by the data acquisition module is recorded as follows: Device = {x1.x2...x...} n};
[0040] The data preprocessing module cleans, standardizes, and extracts features from the collected data for subsequent analysis and prediction. During feature extraction, each device is assigned a feature code: Feature code = {f1, f2, ..., f...} n};
[0041] The data preprocessing module is connected to the AI-CIOT fault diagnosis engine module; the AI-CIOT fault diagnosis engine module processes and analyzes the collected cross-border IoT device data, and then uses pre-stored fault diagnosis algorithms to identify and locate the devices.
[0042] The mathematical formula for fault diagnosis is N = (Ax + By + C) / D;
[0043] Where N: output error (noise); A: numerator of system function (transfer function); B: denominator of system function (transfer function); C: input error (noise); D: common factors of system function (transfer function) (such as gain);
[0044] The AI-CIOT fault diagnosis engine module is connected to the fault diagnosis algorithm library module; the fault diagnosis algorithm library module is equipped with fault diagnosis algorithms, including traditional statistical methods, machine learning and deep learning algorithms.
[0045] The fault diagnosis algorithm formula is Z=Y*X^(a);
[0046] Where Z: fault characteristic equation; Y: fault characteristic coefficient; X: input signal (such as voltage, current); a: fault characteristic index (reflecting the severity of the fault);
[0047] The AI-CIOT fault diagnosis engine module connects with the equipment status monitoring and prediction module; the equipment status monitoring and prediction module uses the collected equipment data and methods from the fault diagnosis algorithm library to monitor the equipment status in real time.
[0048] The equipment condition monitoring and prediction module is connected to the equipment fault knowledge graph construction module. The equipment fault knowledge graph construction module uses the collected equipment data and fault diagnosis results to build an equipment fault knowledge graph for storing and learning equipment fault modes and patterns.
[0049] The formula for failure mode and pattern is R=Σ(Xi*Xj)^(1 / 2) / (ΣXi^2)^(1 / 2)*ΣXj^(1 / 2); where R: correlation coefficient (reflecting the strength of the linear relationship between two variables); Xi: the observed value of the first variable; Xj: the observed value of the second variable.
[0050] The data acquisition module collects data including equipment parameters, operating status, and environmental parameters. The equipment fault knowledge graph construction module and the deep learning capability integration module are connected. The deep learning capability integration module integrates deep learning algorithms into the fault diagnosis system and uses deep learning models to identify and predict equipment faults.
[0051] The deep learning capability integration module is connected to the AI model prediction and optimization module via data connection; the AI model prediction and optimization module provides interpretability of AI models, helping users understand how AI models perform fault diagnosis and model optimization.
[0052] During model optimization, K = 1 / (1 + e^((a + bX)));
[0053] Where K: optimization function; a: bias term; b: weight; e: base of natural logarithm;
[0054] The AI model prediction and optimization module is connected to the real-time fault warning module; the real-time fault warning module uses the results of equipment status monitoring and prediction to issue fault warnings to users in real time, helping users to take timely measures to prevent equipment failures.
[0055] The prediction formula is (Decision Tree) Y = f(X1,X2,...,Xn); Y: predicted value; X1,X2,...,Xn: input features; f: decision tree model.
[0056] The real-time fault early warning module is connected to the equipment maintenance and repair module; based on the diagnostic results of equipment faults, the equipment maintenance and repair module provides corresponding equipment maintenance and repair suggestions to help users improve equipment operating efficiency and reduce failure rate; the equipment maintenance and repair module is connected to the fault diagnosis model optimization and update module; the fault diagnosis model optimization and update module collects new equipment data and fault diagnosis results to optimize and update the fault diagnosis model, thereby improving the accuracy and reliability of the system.
[0057] The fault diagnosis model optimization and update module is data-connected to the system security and privacy protection module. The system security and privacy protection module protects user data security and privacy, including data encryption, access control, and data anonymization. The diagnostic function is Y = f(W1*X1 + W2*X2 + ... + Wn*Xn + b), where Y is the predicted value; W1, W2, ..., Wn are weight matrices; X1, X2, ..., Xn are input features; b is the bias term; and f is the activation function. The system security and privacy protection module is data-connected to the cross-platform device compatibility module. This module ensures the system supports various types of cross-border IoT devices, regardless of their communication protocols or data formats. The cross-platform device compatibility module is also data-connected to the user interface and interaction module. This module sets up the user interface and interaction methods, allowing users to easily use the system for device fault diagnosis and prediction.
[0058] The user interface and interaction module is data-connected to the data analysis and visualization module; the data analysis and visualization module uses data analysis and visualization tools to help users more intuitively understand the equipment's operating status and fault diagnosis results. The data analysis and visualization module is also data-connected to the system performance evaluation and monitoring module; the system performance evaluation and monitoring module is responsible for monitoring system performance, including response time, accuracy, and other key indicators, and adjusting and optimizing it as needed.
[0059] The adjustment probability formula is P(Y|X)=P(X|Y)*P(Y) / P(X);
[0060] Where P(Y|X): conditional probability given input features; P(X|Y): conditional probability given output; P(Y): prior probability of the class; P(X): prior probability of the input features.
[0061] The system performance evaluation and monitoring module is data-connected to the equipment fault prediction module. The equipment fault prediction module uses historical equipment data and fault diagnosis results to build an equipment fault prediction model to predict future faults. The equipment fault prediction module is also data-connected to the data integration and synchronization module. This module integrates data from different devices and sources, synchronizing and updating it to ensure the system's data and information are up-to-date.
[0062] In this invention, data collection involves gathering operational data from cross-border IoT devices in different countries and regions, including device parameters, sensor data, and operation records. Data preprocessing cleans, denoises, and handles missing values to improve data quality. Feature engineering extracts useful features from the preprocessed data, such as device operating status, fault type, and fault occurrence time. Data labeling annotates device fault data to provide a reference for subsequent training and prediction. Model training uses the labeled dataset to train an AI-based fault diagnosis and prediction model. Supervised learning methods, such as support vector machines, neural networks, and decision trees, or unsupervised learning methods, such as clustering and anomaly detection, can be used. Model evaluation assesses model performance, such as accuracy, recall, and F-score, through cross-validation and model scoring methods. Model optimization adjusts and optimizes the model based on the evaluation results, such as adjusting hyperparameters and increasing training data. Fault diagnosis applies the trained model to the operational data of cross-border IoT devices to identify whether faults exist and their types. Fault prediction utilizes the model to analyze historical and real-time data of the devices to predict potential future faults and their occurrence times. Results visualization presents fault diagnosis and prediction results to users in the form of charts and reports, making it easy for them to understand the equipment's operating status and fault trends. Intelligent maintenance recommendations provide users with intelligent maintenance suggestions based on the fault diagnosis and prediction results, such as equipment repair, component replacement, and optimization of operating parameters. Model updates are performed regularly as equipment operating data accumulates to maintain high diagnostic and predictive accuracy.
[0063] The above embodiments are only used to illustrate the present invention and are not intended to limit the technical solutions described in the present invention. Although the present invention has been described in detail with reference to the above embodiments, the present invention is not limited to the specific embodiments described above. Therefore, any modifications or substitutions to the present invention, and all technical solutions and improvements that do not depart from the spirit and scope of the invention, are covered within the scope of the claims of the present invention.
Claims
1. A cross-border IoT device fault diagnosis and prediction system based on artificial intelligence, characterized in that, The system includes a data acquisition module for collecting operational data from cross-border IoT devices. This module is connected to a data preprocessing module. The data collected by the device is recorded as follows: Device = {x1.x2...x...} n }; The data preprocessing module cleans, standardizes, and extracts features from the collected data to facilitate subsequent analysis and prediction. Feature extraction... During the process, each device is assigned a feature code: Feature code = {f1, f2, ..., f n }; The data preprocessing module is connected to the AI-CIOT fault diagnosis engine module; the AI-CIOT fault diagnosis engine module processes and analyzes the collected cross-border IoT device data, and then uses pre-stored fault diagnosis algorithms to identify and locate the devices. The mathematical formula for fault diagnosis is N = (Ax + By + C) / D; Where N: output error (noise); A: numerator of system function (transfer function); B: denominator of system function (transfer function); C: input error (noise); D: common factors of system function (transfer function) (such as gain); The AI-CIOT fault diagnosis engine module is connected to the fault diagnosis algorithm library module; the fault diagnosis algorithm library module is equipped with fault diagnosis algorithms, which include traditional statistical methods, machine learning and deep learning algorithms. The fault diagnosis algorithm formula is Z=Y*X^(a); Where Z: fault characteristic equation; Y: fault characteristic coefficient; X: input signal (such as voltage, current); a: fault characteristic index (reflecting the severity of the fault); The AI-CIOT fault diagnosis engine module is connected to the equipment status monitoring and prediction module; the equipment status monitoring and prediction module uses the collected equipment data and methods from the fault diagnosis algorithm library to monitor the status of the equipment in real time. The equipment status monitoring and prediction module is connected to the equipment fault knowledge graph construction module. The equipment fault knowledge graph construction module uses the collected equipment data and fault diagnosis results to construct an equipment fault knowledge graph for storing and learning equipment fault modes and patterns. The formula for failure mode and pattern is R=Σ(Xi*Xj)^(1 / 2) / (ΣXi^2)^(1 / 2)*ΣXj^(1 / 2); where R: correlation coefficient (reflecting the strength of the linear relationship between two variables); Xi: the observed value of the first variable; Xj: the observed value of the second variable.
2. The cross-border IoT device fault diagnosis and prediction system based on artificial intelligence according to claim 1, characterized in that, The data collected by the data acquisition module includes equipment parameters, operating status, and environmental parameters. The equipment fault knowledge graph construction module is connected to the deep learning capability integration module. The deep learning capability integration module integrates deep learning algorithms into the fault diagnosis system and uses deep learning models to identify and predict equipment faults.
3. The cross-border IoT device fault diagnosis and prediction system based on artificial intelligence according to claim 2, characterized in that, The deep learning capability integration module is connected to the AI model prediction and optimization module; the AI model prediction and optimization module provides interpretability of the AI model, helping users understand how the AI model performs fault diagnosis and model optimization. During model optimization, K = 1 / (1 + e^((a + bX))); Where K: optimization function; a: bias term; b: weight; e: the base of the natural logarithm; The AI model prediction and optimization module is connected to the real-time fault early warning module; The real-time fault warning module uses the results of equipment status monitoring and prediction to issue fault warnings to users in real time, helping users to take timely measures to prevent equipment failures from occurring. The prediction formula is (Decision Tree) Y = f(X1,X2,...,Xn); Y: Predicted value; X1, X2, ..., Xn: Input features; f: Decision tree model.
4. The cross-border IoT device fault diagnosis and prediction system based on artificial intelligence according to claim 3, characterized in that, The real-time fault early warning module is data-connected to the equipment maintenance and repair module. Based on the equipment fault diagnosis results, the equipment maintenance and repair module provides corresponding equipment maintenance and repair suggestions to help users improve equipment operating efficiency and reduce failure rate. The equipment maintenance and repair module is also data-connected to the fault diagnosis model optimization and update module. The fault diagnosis model optimization and update module collects new equipment data and fault diagnosis results to optimize and update the fault diagnosis model, thereby improving the accuracy and reliability of the system.
5. The cross-border IoT device fault diagnosis and prediction system based on artificial intelligence according to claim 4, characterized in that, The fault diagnosis model optimization and update module is connected to the system security and privacy protection module; the system security and privacy protection module protects user data security and privacy, including data encryption, access control and data desensitization. The diagnostic function is Y = f(W1*X1 + W2*X2 + ... + Wn*Xn + b); Where Y: predicted value; W1, W2, ..., Wn: weight matrix; X1, X2, ..., Xn: input features; b: Bias term; f: Activation function; The system security and privacy protection module is connected to the cross-platform device compatibility module; the cross-platform device compatibility module ensures that the system supports various types of cross-border IoT devices, regardless of the communication protocol or data format they use.
6. The cross-border IoT device fault diagnosis and prediction system based on artificial intelligence according to claim 5, characterized in that, The cross-platform device compatibility module is connected to the user interface and interaction module; the user interface and interaction module sets the user interface and interaction methods, enabling users to easily use the system to diagnose and predict device faults.
7. The cross-border IoT device fault diagnosis and prediction system based on artificial intelligence according to claim 6, characterized in that, The user interface and interaction module is connected to the data analysis and visualization module; the data analysis and visualization module uses data analysis and visualization tools to help users understand the operating status of the equipment and the results of fault diagnosis more intuitively.
8. The cross-border IoT device fault diagnosis and prediction system based on artificial intelligence according to claim 7, characterized in that, The data analysis and visualization module and the system performance evaluation and monitoring module are connected; the system performance evaluation and monitoring module is responsible for monitoring the system performance, including response time, accuracy and other key indicators, and adjusting and optimizing it as needed. The adjustment probability formula is P(Y|X)=P(X|Y)*P(Y) / P(X); Where P(Y|X): conditional probability given input features; P(X|Y): conditional probability given output; P(Y): prior probability of the class; P(X): prior probability of the input features.
9. The cross-border IoT device fault diagnosis and prediction system based on artificial intelligence according to claim 8, characterized in that, The system performance evaluation and monitoring module is connected to the equipment fault prediction module; the equipment fault prediction module uses historical equipment data and fault diagnosis results to construct an equipment fault prediction model to predict future faults.
10. The cross-border IoT device fault diagnosis and prediction system based on artificial intelligence according to claim 9, characterized in that, The equipment fault prediction module and the data integration and synchronization module are connected; the data integration and synchronization module integrates data from different devices and sources, and performs synchronization and updates.