Maintenance scheme generation system and method based on historical data and Internet of Things

Abstract

The invention discloses a maintenance scheme generation system and method based on historical data and the Internet of Things, and relates to the field of maintenance systems. The system architecture of the maintenance scheme generation system based on the historical data and the Internet of Things comprises a data acquisition layer, a data storage layer, a data processing layer, a scheme generation layer and a user interaction layer, and the data processing layer comprises the following implementation steps: step 1, data cleaning and preprocessing: cleaning the acquired Internet of Things data; and 2, a data analysis engine: performing deep analysis on the processed data by using a data mining algorithm. According to the historical data and Internet of Things-based maintenance scheme generation system, data analysis is carried out in a mode of combining a deep learning algorithm and a machine learning algorithm, the advantages of deep learning in the aspect of processing complex data can be played, and the advantages of the machine learning algorithm in the aspects of interpretability and small sample learning can also be utilized.

Description

Technical Field

[0001] The present invention relates to the technical field of maintenance systems, and in particular to a maintenance plan generation system and method based on historical data and the Internet of Things. Background Art

[0002] Before the existing maintenance plan generation system generates a plan, it first needs to use the Internet of Things sensor network to collect the operating parameter data of the equipment in real time, and then transmit this data to the data processing center through wireless and wired communication methods.

[0003] However, the amount of data collected is very large, and the existing conventional data processing method is traditional machine learning algorithm. Traditional machine learning algorithm is generally ineffective when processing large-scale and complex data structures. Although it can discover patterns in the data, the accuracy of the algorithm may not be high enough for some complex and nonlinear problems, resulting in the accuracy of fault prediction and maintenance plans needs to be improved. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a maintenance plan generation system and method based on historical data and the Internet of Things, which solves the problem that the existing conventional data processing method is the traditional machine learning algorithm. The traditional machine learning algorithm is generally not effective when processing large-scale and complex data structures. Although it can discover patterns in the data, the accuracy of the algorithm may not be high enough for some complex and nonlinear problems, resulting in the problem that the accuracy of fault prediction and maintenance plans needs to be improved.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a maintenance plan generation system based on historical data and the Internet of Things, whose system architecture includes: A data collection layer, wherein the data collection layer is provided with an Internet of Things sensor network and a historical data interface, and the Internet of Things sensor network and the historical data interface perform data collection; A data storage layer, wherein the data storage layer adopts a distributed database architecture; Data processing layer: The data processing layer includes the following implementation steps: Step 1: Data cleaning and preprocessing: Clean the collected IoT data to remove noise, outliers and duplicate data; organize and standardize historical data to ensure data consistency and availability.

[0006] Step 2: Data analysis engine: Use data mining algorithms to conduct in-depth analysis on the processed data; for example, identify different operating status modes of equipment through cluster analysis; use association rule mining to find out the potential relationship between fault type and equipment operating parameters; build a fault prediction model with the help of decision tree algorithm Solution generation layer: Based on the data analysis results and combined with the equipment maintenance knowledge base, an intelligent decision-making model is built; the model automatically generates maintenance solution recommendations based on the current operating status, fault characteristics, and historical maintenance of the equipment; User interaction layer: The user interaction layer includes maintenance personnel terminals and management personnel consoles.

[0007] Preferably, the IoT sensor network is specifically a sensor deployed at various parts of the equipment, such as a temperature sensor, a pressure sensor, a vibration sensor, and a current sensor, which collects the operating parameter data of the equipment in real time and transmits the data to a data processing center through wireless and wired communication methods; The historical data interface connects to the company's existing equipment management database and historical maintenance record system to obtain historical data information on past equipment maintenance records, fault types, maintenance times, and replacement parts.

[0008] Preferably, the data storage layer uses a Hadoop distributed file system and a cloud storage service to store massive amounts of IoT real-time data and historical data, and classify and store the data to facilitate subsequent data query, analysis, and management.

[0009] Preferably, the equipment maintenance knowledge base includes information on maintenance manuals and expert experience, and the maintenance plan recommendations include maintenance steps, required tools and materials, and maintenance personnel skill requirements.

[0010] Preferably, the maintenance personnel terminal provides maintenance personnel with a convenient mobile application and computer client interface, displays equipment fault information, maintenance plan details, and maintenance progress tracking functions, and maintenance personnel can receive maintenance task notifications, feedback the actual situation during the maintenance process, and adjust the maintenance plan according to the actual situation on the terminal; The administrator console is for use by equipment managers, and provides functions such as overall equipment operating status monitoring, maintenance plan formulation and scheduling, maintenance resource management, and maintenance performance evaluation; managers can view various statistical reports and data analysis charts through the console to make scientific decisions.

[0011] Preferably, a deep learning algorithm is used in the data cleaning and preprocessing step, and a machine learning algorithm is used in the data analysis engine step; Combine deep learning algorithms (such as convolutional neural networks and recurrent neural networks) with traditional machine learning algorithms (such as decision trees and support vector machines). Deep learning algorithms are good at processing large-scale and complex data structures and can automatically extract deep features from data, such as feature learning of complex vibration waveforms and multivariate temperature and pressure change curves during equipment operation. Machine learning algorithms have advantages in interpretability and small sample learning. Through fusion, deep learning is used to preprocess and extract features from the massive amount of raw data collected by the Internet of Things, and then the extracted features are input into the machine learning algorithm for fault classification, prediction, and maintenance plan generation. This can not only improve the system's ability to process complex data, but also ensure the interpretability and accuracy of the maintenance plan.

[0012] Preferably, the deep learning algorithm includes: Convolutional neural network: It has unique advantages in processing images, audio and sensor data. For example, for image data of equipment operation (such as crack detection images on the surface of industrial equipment) and two-dimensional matrices after sensor data conversion (such as arranging vibration sensor data over a period of time into a two-dimensional matrix), the convolution layer of CNN can automatically learn local features in the data, such as edges and textures in the image and specific frequency patterns in vibration data. The pooling layer further reduces the dimensionality of the features and extracts more representative feature information.

[0013] Recurrent neural networks and their variants: Suitable for processing sequence data, such as the sequence of equipment operating parameters changing over time. Taking the time series of equipment temperature and pressure parameters as an example, RNN can learn the dependency between parameters at different time steps and capture the time series features in the data, such as the trend and periodicity of temperature changes. These deep learning models are trained with a large amount of historical data and real-time IoT data to learn deep, abstract feature representations in the data.

[0014] Preferably, the machine learning algorithm comprises: Decision tree: Taking the features extracted by deep learning as input, the decision tree can build a classification model based on the extracted features; for example, in the classification of equipment failures, the decision tree can classify the fault types according to the equipment appearance image features extracted by the convolutional neural network and the operating parameter time series features extracted by the recurrent neural network. The advantage of the decision tree is that it is highly interpretable and can intuitively display the decision logic between features and fault types, such as "if there is a crack of a specific shape in the equipment appearance image and the temperature has continued to rise above the threshold in the past 24 hours, the fault type is mechanical structure damage."

[0015] Support vector machine: Constructs a hyperplane in a high-dimensional feature space for classification and regression prediction. Using the features extracted by deep learning, the support vector machine can find the optimal classification hyperplane to separate the equipment data of different fault types. For example, for multivariate equipment operating parameter features, SVM can determine a hyperplane that maximizes the distance from sample points of different fault categories to the hyperplane, thereby improving the accuracy and generalization ability of classification. In terms of regression prediction, such as predicting the remaining service life of equipment, the support vector machine can establish a prediction model based on the features extracted by deep learning and output an estimated value of the time the equipment can still operate normally.

[0016] A maintenance plan generation method based on historical data and the Internet of Things, characterized in that the method comprises the following steps: Step 1: Data collection phase: Use the IoT sensor network to collect equipment operating parameter data in real time, and then transmit this data to the data processing center through wireless and wired communication methods; use the historical data interface to obtain past equipment maintenance records, fault types, maintenance time, and historical data information on replaced parts; Step 2: Data storage stage: The massive amount of IoT real-time data and historical data collected are transmitted to the data storage layer of the distributed database architecture using the Hadoop distributed file system and cloud storage service, where they are classified and stored for easy subsequent data query, analysis and management. Step 3: Data processing stage: Use deep learning algorithms to clean the collected IoT data, remove noise, outliers and duplicate data, organize and standardize historical data to ensure data consistency and availability; and use data mining algorithms to conduct in-depth analysis of the processed data; Step 4: Solution generation phase: Based on the data analysis results and combined with the equipment maintenance knowledge base, an intelligent decision-making model is constructed; the model automatically generates maintenance plan recommendations based on the current operating status of the equipment, fault characteristics, and historical maintenance conditions, covering maintenance steps, required tools and materials, and maintenance personnel skill requirements; Step 5: User interaction stage: Maintenance personnel receive equipment fault information and maintenance plan details through convenient mobile applications and computer client interfaces. Equipment managers use the management console to monitor the overall operation status of the equipment, formulate and schedule maintenance plans, manage maintenance resources, and evaluate maintenance performance.

[0017] The present invention discloses a maintenance plan generation system and method based on historical data and the Internet of Things, which has the following beneficial effects: 1. This maintenance plan generation system based on historical data and the Internet of Things uses deep learning algorithms for data cleaning and preprocessing, and can automatically extract deep-level features from the data. It is particularly suitable for processing large-scale, complex data structures, such as complex vibration waveforms during equipment operation and multi-variable temperature and pressure change curves, which effectively improves data quality and lays a good foundation for subsequent analysis. By combining deep learning algorithms with machine learning algorithms for data analysis, we can not only give play to the advantages of deep learning in processing complex data, but also take advantage of the advantages of machine learning algorithms in interpretability and small sample learning, so as to more accurately classify faults, predict and generate maintenance plans, and ensure the interpretability and accuracy of maintenance plans.

[0018] 2. The maintenance plan generation system based on historical data and the Internet of Things provides maintenance personnel with convenient mobile applications and computer client interfaces, allowing them to obtain equipment failure information, maintenance plan details, and maintenance progress tracking functions in a timely manner. At the same time, they can easily receive maintenance task notifications and feedback on actual conditions, which improves the work efficiency and participation of maintenance personnel. It provides managers with a fully functional management console, enabling them to fully monitor the overall operation status of equipment, formulate and schedule maintenance plans, manage maintenance resources, evaluate maintenance performance, and view various statistical reports and data analysis charts, which facilitates scientific decision-making and improves managers' control over equipment maintenance management. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 It is a schematic diagram of the overall structure of the present invention; Figure 2 Schematic diagram of the architecture of the data processing layer of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] The embodiment of the present application provides a maintenance plan generation system and method based on historical data and the Internet of Things, which solves the problem that the existing conventional data processing method is a machine learning algorithm. The effect of machine learning algorithms is relatively general when processing large-scale and complex data structures. Although it can discover patterns in the data, the accuracy of the algorithm may not be high enough for some complex and nonlinear problems, resulting in the problem that the accuracy of fault prediction and maintenance plans needs to be improved. By combining deep learning algorithms with machine learning algorithms for data analysis, it can not only give play to the advantages of deep learning in processing complex data, but also take advantage of the advantages of machine learning algorithms in interpretability and small sample learning, so as to more accurately classify faults, predict and generate maintenance plans, and ensure the interpretability and accuracy of maintenance plans.

[0023] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0024] The embodiment of the present invention discloses a maintenance plan generation system based on historical data and the Internet of Things. Figure 1-2 As shown, its system architecture includes: The data collection layer is provided with an IoT sensor network and a historical data interface, and the IoT sensor network and the historical data interface are used for data collection; Data storage layer, the data storage layer adopts a distributed database architecture; Data processing layer: The data processing layer includes the following implementation steps: Step 1: Data cleaning and preprocessing: Clean the collected IoT data to remove noise, outliers and duplicate data; organize and standardize historical data to ensure data consistency and availability.

[0025] Step 2: Data analysis engine: Use data mining algorithms to conduct in-depth analysis on the processed data; for example, identify different operating status modes of equipment through cluster analysis; use association rule mining to find out the potential relationship between fault type and equipment operating parameters; build a fault prediction model with the help of decision tree algorithm Solution generation layer: Based on the data analysis results and combined with the equipment maintenance knowledge base, an intelligent decision-making model is built; the model automatically generates maintenance solution recommendations based on the current operating status, fault characteristics, and historical maintenance of the equipment; User interaction layer: The user interaction layer includes maintenance personnel terminals and management personnel consoles.

[0026] Using deep learning algorithms for data cleaning and preprocessing can automatically extract deep features from the data, and is particularly suitable for processing large-scale, complex data structures, such as complex vibration waveforms during equipment operation and multivariable temperature and pressure change curves, effectively improving data quality and laying a good foundation for subsequent analysis. By combining deep learning algorithms with machine learning algorithms for data analysis, we can not only leverage the advantages of deep learning in processing complex data, but also take advantage of the advantages of machine learning algorithms in interpretability and small sample learning, thereby more accurately classifying faults, predicting, and generating maintenance plans, ensuring the interpretability and accuracy of maintenance plans.

[0027] Furthermore, the IoT sensor network specifically refers to sensors deployed at various parts of the equipment, such as temperature sensors, pressure sensors, vibration sensors, and current sensors, which collect the operating parameter data of the equipment in real time and transmit the data to the data processing center through wireless and wired communication methods; The historical data interface connects to the company's existing equipment management database and historical maintenance record system to obtain historical data information on past equipment maintenance records, fault types, maintenance times, and replacement parts.

[0028] Furthermore, the data storage layer uses the Hadoop distributed file system and cloud storage service to store massive amounts of IoT real-time and historical data, and classify and store the data to facilitate subsequent data query, analysis, and management.

[0029] Furthermore, the equipment maintenance knowledge base contains information on maintenance manuals and expert experience, and maintenance solution recommendations include maintenance steps, required tools and materials, and maintenance personnel skill requirements.

[0030] Furthermore, the maintenance personnel terminal provides maintenance personnel with a convenient mobile application and computer client interface, which displays equipment fault information, maintenance plan details, and maintenance progress tracking functions. Maintenance personnel can receive maintenance task notifications, feedback on the actual situation during the maintenance process, and adjust the maintenance plan according to the actual situation on the terminal; The management console is for use by equipment managers, providing functions such as overall equipment operation status monitoring, maintenance plan formulation and scheduling, maintenance resource management, and maintenance performance evaluation. Managers can view various statistical reports and data analysis charts through the console to make scientific decisions.

[0031] Furthermore, deep learning algorithms are used in the data cleaning and preprocessing steps, and machine learning algorithms are used in the data analysis engine step; Combine deep learning algorithms (such as convolutional neural networks and recurrent neural networks) with traditional machine learning algorithms (such as decision trees and support vector machines). Deep learning algorithms are good at processing large-scale and complex data structures and can automatically extract deep features from data, such as feature learning of complex vibration waveforms and multivariate temperature and pressure change curves during equipment operation. Machine learning algorithms have advantages in interpretability and small sample learning. Through fusion, deep learning is used to preprocess and extract features from the massive amount of raw data collected by the Internet of Things, and then the extracted features are input into the machine learning algorithm for fault classification, prediction, and maintenance plan generation. This can not only improve the system's ability to process complex data, but also ensure the interpretability and accuracy of the maintenance plan.

[0032] Specifically disclosed, deep learning algorithms include: Convolutional neural network: It has unique advantages in processing images, audio and sensor data. For example, for image data of equipment operation (such as crack detection images on the surface of industrial equipment) and two-dimensional matrices after sensor data conversion (such as arranging vibration sensor data over a period of time into a two-dimensional matrix), the convolution layer of CNN can automatically learn local features in the data, such as edges and textures in the image and specific frequency patterns in vibration data. The pooling layer further reduces the dimensionality of the features and extracts more representative feature information.

[0033] Recurrent neural networks and their variants: Suitable for processing sequence data, such as the sequence of equipment operating parameters changing over time. Taking the time series of equipment temperature and pressure parameters as an example, RNN can learn the dependency between parameters at different time steps and capture the time series features in the data, such as the trend and periodicity of temperature changes. These deep learning models are trained with a large amount of historical data and real-time IoT data to learn deep, abstract feature representations in the data.

[0034] Specifically disclosed, machine learning algorithms include: Decision tree: Taking the features extracted by deep learning as input, the decision tree can build a classification model based on the extracted features; for example, in the classification of equipment failures, the decision tree can classify the fault types according to the equipment appearance image features extracted by the convolutional neural network and the operating parameter time series features extracted by the recurrent neural network. The advantage of the decision tree is that it is highly interpretable and can intuitively display the decision logic between features and fault types, such as "if there is a crack of a specific shape in the equipment appearance image and the temperature has continued to rise above the threshold in the past 24 hours, the fault type is mechanical structure damage."

[0035] Support vector machine: Constructs a hyperplane in a high-dimensional feature space for classification and regression prediction. Using the features extracted by deep learning, the support vector machine can find the optimal classification hyperplane to separate the equipment data of different fault types. For example, for multivariate equipment operating parameter features, SVM can determine a hyperplane that maximizes the distance from sample points of different fault categories to the hyperplane, thereby improving the accuracy and generalization ability of classification. In terms of regression prediction, such as predicting the remaining service life of equipment, the support vector machine can establish a prediction model based on the features extracted by deep learning and output an estimated value of the time the equipment can still operate normally.

[0036] A maintenance plan generation method based on historical data and the Internet of Things, the method comprising the following steps: Step 1: Data collection phase: Use the IoT sensor network to collect equipment operating parameter data in real time, and then transmit this data to the data processing center through wireless and wired communication methods; use the historical data interface to obtain past equipment maintenance records, fault types, maintenance time, and historical data information on replaced parts; Step 2: Data storage stage: The massive amount of IoT real-time data and historical data collected are transmitted to the data storage layer of the distributed database architecture using the Hadoop distributed file system and cloud storage service, where they are classified and stored for easy subsequent data query, analysis and management. Step 3: Data processing stage: Use deep learning algorithms to clean the collected IoT data, remove noise, outliers and duplicate data, organize and standardize historical data to ensure data consistency and availability; and use data mining algorithms to conduct in-depth analysis of the processed data; Step 4: Solution generation phase: Based on the data analysis results and combined with the equipment maintenance knowledge base, an intelligent decision-making model is constructed; the model automatically generates maintenance plan recommendations based on the current operating status of the equipment, fault characteristics, and historical maintenance conditions, covering maintenance steps, required tools and materials, and maintenance personnel skill requirements; Step 5: User interaction stage: Maintenance personnel receive equipment fault information and maintenance plan details through convenient mobile applications and computer client interfaces. Equipment managers use the management console to monitor the overall operation status of the equipment, formulate and schedule maintenance plans, manage maintenance resources, and evaluate maintenance performance.

[0037] Using deep learning algorithms for data cleaning and preprocessing can automatically extract deep features from the data, and is particularly suitable for processing large-scale, complex data structures, such as complex vibration waveforms during equipment operation and multivariable temperature and pressure change curves, effectively improving data quality and laying a good foundation for subsequent analysis. By combining deep learning algorithms with machine learning algorithms for data analysis, we can not only leverage the advantages of deep learning in processing complex data, but also take advantage of the advantages of machine learning algorithms in interpretability and small sample learning, thereby more accurately classifying faults, predicting, and generating maintenance plans, ensuring the interpretability and accuracy of maintenance plans.

[0038] It provides maintenance personnel with convenient mobile applications and computer client interfaces, allowing them to obtain equipment failure information, maintenance plan details, and maintenance progress tracking functions in a timely manner. At the same time, they can easily receive maintenance task notifications and feedback on actual conditions, which improves the work efficiency and participation of maintenance personnel. It provides managers with a fully functional management console, allowing them to fully monitor the overall operation status of equipment, formulate and schedule maintenance plans, manage maintenance resources, evaluate maintenance performance, and view various statistical reports and data analysis charts, which facilitates scientific decision-making and improves managers' control over equipment maintenance management.

[0039] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A maintenance plan generation system based on historical data and the Internet of Things, characterized in that: The system architecture includes: A data collection layer, wherein the data collection layer is provided with an Internet of Things sensor network and a historical data interface, and the Internet of Things sensor network and the historical data interface perform data collection; A data storage layer, wherein the data storage layer adopts a distributed database architecture; Data processing layer: The data processing layer includes the following implementation steps: Step 1: Data cleaning and preprocessing: Clean the collected IoT data to remove noise, outliers and duplicate data; organize and standardize historical data to ensure data consistency and availability; Step 2: Data analysis engine: Use data mining algorithms to conduct in-depth analysis of processed data; Solution generation layer: Based on the data analysis results and combined with the equipment maintenance knowledge base, an intelligent decision-making model is built; the model automatically generates maintenance solution recommendations based on the current operating status, fault characteristics, and historical maintenance of the equipment; User interaction layer: The user interaction layer includes maintenance personnel terminals and management personnel consoles.

2. A maintenance plan generation system based on historical data and the Internet of Things according to claim 1, characterized in that: The IoT sensor network is specifically a sensor deployed at various parts of the equipment, such as a temperature sensor, a pressure sensor, a vibration sensor, and a current sensor, which collects the operating parameter data of the equipment in real time and transmits the data to a data processing center through wireless and wired communication methods; The historical data interface connects to the company's existing equipment management database and historical maintenance record system to obtain historical data information on past equipment maintenance records, fault types, maintenance times, and replacement parts.

3. A maintenance plan generation system based on historical data and the Internet of Things according to claim 1, characterized in that: The data storage layer uses a Hadoop distributed file system and a cloud storage service to store massive amounts of real-time and historical data of the Internet of Things, and to classify and store the data to facilitate subsequent data query, analysis, and management.

4. A maintenance plan generation system based on historical data and the Internet of Things according to claim 1, characterized in that: The equipment maintenance knowledge base includes information on maintenance manuals and expert experience, and the maintenance solution recommendations include maintenance steps, required tools and materials, and maintenance personnel skill requirements.

5. A maintenance plan generation system based on historical data and the Internet of Things according to claim 1, characterized in that: The maintenance personnel terminal provides maintenance personnel with a convenient mobile application and computer client interface, which displays equipment fault information, maintenance plan details, and maintenance progress tracking functions. Maintenance personnel can receive maintenance task notifications, feedback the actual situation during the maintenance process, and adjust the maintenance plan according to the actual situation on the terminal; The administrator console is for use by equipment administrators, and provides functions such as overall equipment operating status monitoring, maintenance plan formulation and scheduling, maintenance resource management, and maintenance performance evaluation; administrators can view various statistical reports and data analysis charts through the console.

6. A maintenance plan generation system based on historical data and the Internet of Things according to claim 1, characterized in that: A deep learning algorithm is used in the data cleaning and preprocessing steps, and a machine learning algorithm is used in the data analysis engine step.

7. A maintenance plan generation system based on historical data and the Internet of Things according to claim 6, characterized in that: The deep learning algorithm includes: Convolutional neural networks: have unique advantages in processing images, audio, and sensor data; Recurrent neural networks and their variants: suitable for processing sequence data, such as the sequence of equipment operating parameters changing over time.

8. A maintenance plan generation system based on historical data and the Internet of Things according to claim 6, characterized in that: The machine learning algorithm includes: Decision tree: Taking the features extracted by deep learning as input, the decision tree can build a classification model based on the extracted features; Support vector machine: A hyperplane is constructed in a high-dimensional feature space for classification and regression prediction. Using the features extracted by deep learning, the support vector machine can find the optimal classification hyperplane to separate the equipment data of different fault types.

9. A maintenance plan generation method based on historical data and the Internet of Things according to any one of claims 1 to 8, characterized in that: The method comprises the following steps: Step 1: Data collection phase: Use the IoT sensor network to collect equipment operating parameter data in real time, and then transmit this data to the data processing center through wireless and wired communication methods; use the historical data interface to obtain historical data information on past equipment maintenance records, fault types, maintenance time, and replacement parts; Step 2: Data storage stage: The massive amount of IoT real-time data and historical data collected are transmitted to the data storage layer of the distributed database architecture using the Hadoop distributed file system and cloud storage service, where they are classified and stored for easy subsequent data query, analysis and management. Step 3: Data processing stage: Use deep learning algorithms to clean the collected IoT data, remove noise, outliers and duplicate data, organize and standardize historical data to ensure data consistency and availability; and use data mining algorithms to conduct in-depth analysis of the processed data; Step 4: Solution generation phase: Based on the data analysis results and combined with the equipment maintenance knowledge base, an intelligent decision-making model is constructed; the model automatically generates maintenance plan recommendations based on the current operating status of the equipment, fault characteristics, and historical maintenance conditions, covering maintenance steps, required tools and materials, and maintenance personnel skill requirements; Step 5: User interaction stage: Maintenance personnel receive equipment fault information and maintenance plan details through convenient mobile applications and computer client interfaces. Equipment managers use the management console to monitor the overall operation status of the equipment, formulate and schedule maintenance plans, manage maintenance resources, and evaluate maintenance performance.

Images