Internet of Things equipment operation monitoring method and system based on artificial intelligence

Abstract

The invention is applied to the technical field of Internet of Things equipment monitoring, and particularly discloses an Internet of Things equipment operation monitoring method and system based on artificial intelligence, and the method comprises the following steps: S1, data collection; s2, data preprocessing; s3, constructing a model; s4, monitoring and predicting; and S5, early warning and decision making. According to the Internet of Things equipment operation monitoring method and system based on artificial intelligence, the monitoring model based on the attention mechanism is constructed, so that the model can automatically focus key parameters which have relatively great influence on the operation state of the Internet of Things equipment, and thus the monitoring model filters invalid data interference; real-time monitoring data of the Internet of Things equipment are input into the monitoring model, the monitoring model is used for analyzing operation parameters of the Internet of Things equipment, accurate identification of fault features of the Internet of Things equipment and advanced pre-judgment of potential hazards are achieved, and output results of three states are output according to parameter data of the equipment according to output results of the monitoring model.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things device monitoring, and in particular to an Internet of Things device operation monitoring method and system based on artificial intelligence. Background Art

[0002] The Internet of Things refers to the use of information sensing devices to connect any object to the Internet in accordance with agreed protocols to exchange information and communicate, in order to achieve intelligent identification, positioning, tracking, monitoring and management. IoT devices refer to smart hardware that access the Internet through components such as communication modules and have the ability to collect, transmit, process and interact with data. They are the basic units of the IoT system, realizing seamless connection between the physical world and the digital world, and are widely used in industry, agriculture, home, medical care, transportation and other fields. In order to ensure the stable operation of IoT devices and reduce the risk of failure, IoT devices need to be monitored during their operation.

[0003] Traditional IoT device monitoring usually uses a single computer to monitor multiple IoT devices in real time. However, with the increasing number of IoT devices, the traditional monitoring method places heavy pressure on computers to process data, making it difficult to meet the monitoring needs of efficient and stable operation of large-scale IoT devices. Summary of the Invention

[0004] The purpose of the present invention is to provide an artificial intelligence-based IoT device operation monitoring method and system to solve the problem of high data processing pressure and low efficiency in the process of traditionally connecting one computer to monitor multiple IoT devices at the same time as proposed in the background technology.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an artificial intelligence-based IoT device operation monitoring method and system, comprising the following steps: S1. Data collection: Collect parameter data sets during the operation of the equipment to form an original data set; S2, data preprocessing: process the original data set and extract features; S3, model construction: using the processed original data set to build a monitoring model; S4. Monitoring and prediction: Use monitoring models to monitor IoT device operating parameters in real time and predict device failures. S5. Early warning and decision-making: Output graded early warning information based on model prediction results to assist equipment operation and maintenance; Preferably, during the S1 data collection process, sensors are deployed on the IoT devices for data collection. The sensors collect parameter data of the IoT devices in real time during operation. Device identification information is added to the original parameter data collected by the sensors to form an original data set.

[0006] By adopting the above technical solution, sensors can be used to realize real-time collection of IoT device operating parameters, which facilitates the subsequent prediction of the operating status of IoT devices.

[0007] Preferably, the S2 data preprocessing process receives the collected original data set and performs preprocessing and feature extraction on the original data set. The preprocessing includes data cleaning and data normalization. The data cleaning removes noise and outliers in the data. The outliers are defined as data points that deviate from the mean by more than 3 times the standard deviation. The outlier data are removed. The data normalization transforms the cleaned original data set. The formula for the data normalization is as follows: in, is the original data; is the minimum value of the parameter; is the maximum value of the parameter; is the normalized data; The normalized data is subjected to feature extraction using a convolutional neural network, and feature parameters that can reflect the operating status of the device are extracted during the feature extraction process.

[0008] By adopting the above technical solution, preprocessing the collected raw data can increase the speed of subsequent data processing.

[0009] Preferably, during the construction of the S3 model, an artificial intelligence monitoring model based on the attention mechanism is constructed. The monitoring model takes the operating parameters of the Internet of Things device as input variables and the operating status of the Internet of Things device as output variables. The attention mechanism is used to enable the monitoring model to automatically focus on key parameters that have a greater impact on the operating status of the device. The monitoring model is trained using the preprocessed original data set to optimize the monitoring model parameters.

[0010] By adopting the above technical solution, the monitoring model can use the operating parameters of the IoT device as input data to output the operating status of the IoT device.

[0011] Preferably, during the S4 monitoring and prediction process, the trained monitoring model is used to perform real-time monitoring and prediction on the IoT devices. The monitoring model inputs the real-time data collected by the IoT devices and outputs the real-time operating status of the IoT devices. The monitoring model analyzes historical fault data, establishes a mapping relationship between fault modes and features, and classifies equipment fault types and predicts fault times.

[0012] By adopting the above technical solution and training the monitoring model with a large amount of historical data, the output credibility of the monitoring model can be increased.

[0013] Preferably, in the S5 warning and decision-making process, the operating status of the Internet of Things device is divided into three types according to the prediction output results of the monitoring model, the first-level status is the normal status, the second-level status is the warning status, and the third-level status is the fault status. The normal status indicates that all parameters of the Internet of Things device are within the normal range, the warning status indicates that there are small-scale abnormalities in the parameters of the Internet of Things device, and there are potential hidden dangers of failure, and the fault status indicates that there are large-scale abnormalities in the parameters of the Internet of Things device, and the Internet of Things device has failed. In the first-level status, the device does not issue a warning message, and in the second-level status, a text warning message is sent to the user mobile terminal of the manager to remind the manager to pay attention to the operating status of the device. In the third-level status, an emergency alarm message is sent to the user mobile terminal of the manager, and maintenance suggestions are provided to the manager based on the prediction output results of the monitoring model. The maintenance suggestions include prediction of the cause of the failure, maintenance means and required parts.

[0014] By adopting the above technical solution and utilizing the three different output results of the monitoring model, different operation and maintenance methods can be implemented for different operating states of IoT devices.

[0015] Preferably, it includes a central processing unit and a terminal processing unit, one of the central processing units is connected to no less than two terminal processing units at the same time, the central processing unit includes a central computer, the central computer has a built-in model construction module and an early warning decision module, the terminal processing unit includes a monitoring terminal computer, the monitoring terminal computer has a built-in data acquisition module, a data preprocessing module and a real-time monitoring module, the model construction module is used for building a monitoring model, the early warning decision module receives the real-time monitoring data of the Internet of Things device and uses the monitoring model to output the operating status of the device, the data acquisition module is connected to the sensor through a wireless signal to collect the real-time operating data of the Internet of Things device, the data preprocessing module preprocesses the collected data, and the real-time monitoring module inputs the preprocessed equipment operating data into the early warning decision module after abnormal data detection to monitor and warn the equipment operating status.

[0016] By adopting the above technical solution and utilizing a distributed structure composed of a central processing unit and terminal processing units, the data processing speed of the system can be increased.

[0017] Preferably, the central computer in the central processing unit is remotely connected to the user mobile terminal via a wireless signal, the central computer is communicatively connected to the monitoring terminal computer in the terminal processing unit, the monitoring terminal computer is connected to the Internet of Things device, and a data anomaly detection algorithm is provided between the monitoring terminal computer and the central computer. The data anomaly detection algorithm judges the real-time data input by the monitoring terminal computer, and the data anomaly detection algorithm performs an anomaly judgment on the real-time data. If there is no anomaly in the real-time data, the real-time data is input into the central computer for subsequent processing; if there is an anomaly in the real-time data, the real-time data is returned to the data preprocessing module in the monitoring terminal computer for further processing.

[0018] By adopting the above technical solution and utilizing the anomaly detection algorithm between the monitoring terminal computer and the central computer, the influence of abnormal data on the error of the monitoring result can be reduced.

[0019] Compared with the prior art, the present invention has the following beneficial effects: the method and system for monitoring the operation of IoT devices based on artificial intelligence: 1. This invention constructs a monitoring model based on an attention mechanism, enabling the model to automatically focus on key parameters that have a significant impact on the operating status of IoT devices, thereby enabling the monitoring model to filter out invalid data interference. By inputting real-time monitoring data of IoT devices into the monitoring model, the monitoring model is used to analyze the operating parameters of IoT devices, enabling accurate identification of IoT device fault characteristics and early prediction of potential hidden dangers. The output results of the monitoring model are output in three states based on the device parameter data. The output results are matched to the different states of IoT devices through normal state, warning state, and fault state, respectively, to assist IoT device managers in implementing different operation and maintenance measures for the devices based on the output results. 2. The present invention adopts a layered design of central processing units and terminal processing units, and uses a central processing unit to connect multiple terminal processing units at the same time to form a distributed monitoring network. The terminal processing units are responsible for data collection, cleaning and preliminary anomaly detection, and only transmit key feature parameters to the central unit, reducing network transmission pressure and cloud computing load. The central unit focuses on real-time decision-making during model training and monitoring, improving system response speed, thereby supporting real-time monitoring of large-scale Internet of Things devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 Schematic diagram of the process structure of the method of the present invention; Figure 2 This is a schematic diagram of the structure of the system modules of the present invention. DETAILED DESCRIPTION

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only 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 making creative efforts are within the scope of protection of the present invention.

[0022] See also Figure 1-Figure 2 The present invention provides a technical solution: an artificial intelligence-based method and system for monitoring the operation of Internet of Things devices.

[0023] S1. Data collection: Collect parameter data sets during device operation to form an original data set; S2. Data preprocessing: Process and extract features from the original data set; S3. Model construction: Use the processed original data set to build a monitoring model; S4. Monitoring and prediction: Use the monitoring model to monitor the operating parameters of IoT devices in real time and predict device failures; S5. Early warning and decision-making: Output graded early warning information based on the model prediction results to assist equipment operation and maintenance; like Figure 1 As shown in the figure, in the process of operating monitoring of IoT devices, the parameters of IoT devices during operation are first monitored in real time, the collected real-time data is preprocessed, and the preprocessed data is input into a pre-built monitoring model. The monitoring model is used to output the real-time operating status of IoT devices, thereby assisting IoT device managers in equipment operation and maintenance.

[0024] A central processing unit is connected to no less than two terminal processing units at the same time. The central processing unit includes a central computer with a built-in model construction module and an early warning decision module. The terminal processing unit includes a monitoring terminal computer with a built-in data acquisition module, a data preprocessing module and a real-time monitoring module. The model construction module is used to build the monitoring model. The early warning decision module receives the real-time monitoring data of the Internet of Things device and uses the monitoring model to output the operating status of the device. The data acquisition module is connected to the sensor through a wireless signal to collect the real-time operating data of the Internet of Things device. The data preprocessing module preprocesses the collected data. The real-time monitoring module processes the preprocessed device operating data through an abnormal data detection After the test, the early warning decision module is input to monitor and issue early warnings on the equipment operation status. The central computer in the central processing unit is remotely connected to the user's mobile terminal via wireless signals. The central computer is communicated with the monitoring terminal computer in the terminal processing unit. The monitoring terminal computer is connected to the Internet of Things device. A data anomaly detection algorithm is set between the monitoring terminal computer and the central computer. The data anomaly detection algorithm judges the real-time data input by the monitoring terminal computer. The data anomaly detection algorithm judges the anomaly of the real-time data. If there is no anomaly in the real-time data, the real-time data is input into the central computer for subsequent processing; if there is an anomaly in the real-time data, the real-time data is returned to the data preprocessing module in the monitoring terminal computer for further processing; like Figure 2 As shown, before monitoring the IoT devices, a monitoring system is first constructed, using a central processing unit and multiple terminal processing units connected at the same time to form a distributed monitoring system, and a terminal processing unit is connected to an IoT device to be monitored, wherein a central computer is set in the central processing unit, and a monitoring terminal computer is set in the terminal processing unit. The model construction module and early warning decision module built into the central computer are used to realize the construction of the monitoring model and the centralized decision-making of the system, and the data acquisition module, data preprocessing module and real-time monitoring module built into the monitoring terminal computer realize the real-time collection of IoT device data and the preprocessing steps after collection. The preprocessed IoT device data is detected by the data anomaly detection algorithm to reduce the abnormal data input into the central processing unit, thereby ensuring the reliability of the output results of the final monitoring model.

[0025] During the S1 data collection process, sensors are deployed on IoT devices for data collection. The sensors collect parameter data of IoT devices in real time during operation. The original parameter data collected by the sensors are added with device identification information to form an original data set. During the S2 data preprocessing process, the collected original data set is received and preprocessed and feature extracted from the original data set. The preprocessing includes data cleaning and data normalization. Data cleaning removes noise and outliers in the data. Outliers are defined as data points that deviate from the mean by more than 3 times the standard deviation. The outlier data is removed, and data normalization transforms the cleaned original data set. The formula for data normalization is as follows: in, is the original data; is the minimum value of the parameter; is the maximum value of the parameter; The normalized data is used to extract features using a convolutional neural network. During the feature extraction process, characteristic parameters that can reflect the operating status of the equipment are extracted. like Figure 1 and Figure 2 As shown in the figure, the sensors deployed on the IoT devices collect parameters of the IoT devices in real time during their operation. The data collected by the sensors form an original data set, which is preprocessed. The preprocessing operations include data cleaning and data normalization. Data points that deviate from the mean by more than 3 times the standard deviation are marked as abnormal data points, and the abnormal data points are deleted. The high-frequency noise is removed by the sliding average filter algorithm to ensure the quality of subsequent data processing. The original data after data cleaning is normalized again, and the normalization result is mapped to the [0,1] interval, so that the data are at the same order of magnitude, which is convenient for subsequent data analysis. The convolutional neural network is used to extract features from the data and extract characteristic parameters that can reflect the operating status of the equipment.

[0026] During the S3 model construction process, an artificial intelligence monitoring model based on the attention mechanism is constructed. The monitoring model uses the operating parameters of the IoT device as input variables and the operating status of the IoT device as output variables. The attention mechanism is used to enable the monitoring model to automatically focus on key parameters that have a greater impact on the operating status of the device. The monitoring model is trained using the preprocessed original data set to optimize the monitoring model parameters. Before monitoring the operation of IoT devices, an artificial intelligence monitoring model is first constructed based on the attention mechanism. The monitoring model takes the operating parameters of IoT devices as input variables and the operating status of IoT devices as output variables. The attention mechanism is set in the monitoring model so that the model can automatically pay attention to the key parameters that have a greater impact on the operating status of the device, ensuring that the monitoring model can have greater reliability in the monitoring results of IoT devices. At the same time, a large amount of historical data is used to train the monitoring model to optimize the parameters of the monitoring model.

[0027] During the S4 monitoring and prediction process, the trained monitoring model is used to monitor and predict the IoT devices in real time. The monitoring model inputs the real-time data collected by the IoT devices and outputs the real-time operating status of the IoT devices. The monitoring model analyzes historical fault data, establishes a mapping relationship between fault modes and features, and classifies the equipment fault types and predicts the fault time. During the S5 warning and decision-making process, the operating status of the IoT devices is divided into three types according to the prediction output results of the monitoring model: the first-level status is the normal status, the second-level status is the warning status, and the third-level status is the fault status. The normal status indicates that all parameters of the IoT device are within the normal range. The warning status indicates that the parameters of the IoT device have a small range of abnormalities and there are potential hidden dangers of failure. The fault status indicates that the parameters of the IoT device have a large range of abnormalities and the IoT device has failed. In the first-level status, the device does not issue a warning message. In the second-level status, a text warning message is sent to the user mobile terminal of the manager to remind the manager to pay attention to the operating status of the device. In the third-level status, an emergency alarm message is sent to the user mobile terminal of the manager. Based on the prediction output results of the monitoring model, maintenance suggestions are provided to the manager. The maintenance suggestions include the prediction of the cause of the failure, the maintenance methods and the required parts. like Figure 1 As shown in the figure, the trained monitoring model is used to receive the real-time data collected by the IoT devices. According to the real-time output results of the monitoring model, the operating status of the IoT devices is divided into three types: the normal status is when all parameters of the IoT devices are within the normal range; the warning status is when the parameters of the IoT devices are abnormal in a small range and there is a potential fault hidden danger; and the fault status is when the parameters of the IoT devices are abnormal in a large range and the IoT devices have already failed. According to different output results, different information is sent to the user mobile terminal of the manager, so that the manager can perform different operation and maintenance operations according to the different operating status of the IoT devices. In conjunction with the prediction output results of the monitoring model, maintenance suggestions are provided to the manager. The maintenance suggestions include the prediction of the cause of the fault, the maintenance means and the required parts, thereby improving the efficiency of the manager in the operation and maintenance of the IoT devices.

[0028] Working principle: Sensors are placed on IoT devices to collect parameter data during the operation of the equipment in real time to form an original data set. By setting outlier standards, noise and abnormal data are eliminated to improve data quality. A convolutional neural network is used to extract key feature parameters that can reflect the operating status of the equipment from normalized data. An artificial intelligence monitoring model based on the attention mechanism is constructed. The equipment operating parameters are input into the artificial intelligence monitoring model, and the real-time operating status of the equipment is output to predict the type and time of equipment failure, thereby increasing the efficiency of equipment operation and maintenance by managers.

[0029] While the embodiments of the present invention have been shown and described, it will be apparent to those skilled in the art that various changes, modifications, substitutions, and alterations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. A method for monitoring the operation of IoT devices based on artificial intelligence, characterized by: The following steps are involved: S1, data collection; Collect parameter data sets during equipment operation to form original data sets; S2, data preprocessing; Processing and feature extraction of raw data sets; S3, model construction: using the processed original data set to build a monitoring model; S4, monitoring and prediction; Use monitoring models to monitor IoT device operating parameters in real time and predict device failures; S5. Early warning and decision-making: Output graded early warning information based on model prediction results to assist equipment operation and maintenance.

2. The method for monitoring the operation of an IoT device based on artificial intelligence according to claim 1, characterized in that: During the S1 data collection process, sensors are deployed on IoT devices for data collection. The sensors collect parameter data of IoT devices in real time during operation. Device identification information is added to the original parameter data collected by the sensors to form an original data set.

3. The method for monitoring the operation of an Internet of Things device based on artificial intelligence according to claim 1, characterized in that: The S2 data preprocessing process receives the collected original data set and performs preprocessing and feature extraction on the original data set. The preprocessing includes data cleaning and data normalization. The data cleaning removes noise and outliers in the data. The outliers are defined as data points that deviate from the mean by more than 3 standard deviations. The outlier data are removed. The data normalization transforms the cleaned original data set. The formula for the data normalization is as follows: in, is the original data; is the minimum value of the parameter; is the maximum value of the parameter; is the normalized data; The normalized data is subjected to feature extraction using a convolutional neural network, and feature parameters that can reflect the operating status of the device are extracted during the feature extraction process.

4. The method for monitoring the operation of IoT devices based on artificial intelligence according to claim 1, characterized in that: During the S3 model construction process, an artificial intelligence monitoring model based on the attention mechanism is constructed. The monitoring model takes the operating parameters of the IoT device as input variables and the operating status of the IoT device as output variables. The attention mechanism is used to enable the monitoring model to automatically focus on key parameters that have a greater impact on the operating status of the device. The monitoring model is trained using the preprocessed original data set to optimize the monitoring model parameters.

5. The method for monitoring the operation of IoT devices based on artificial intelligence according to claim 1, characterized in that: During the S4 monitoring and prediction process, the trained monitoring model is used to perform real-time monitoring and prediction of IoT devices. The monitoring model inputs real-time data collected by IoT devices and outputs the real-time operating status of IoT devices. The monitoring model analyzes historical fault data, establishes a mapping relationship between fault modes and features, and classifies equipment fault types and predicts fault times.

6. The method for monitoring the operation of IoT devices based on artificial intelligence according to claim 1, characterized in that: During the S5 warning and decision-making process, the operating status of the Internet of Things device is divided into three types according to the prediction output results of the monitoring model, the first-level status is the normal status, the second-level status is the warning status, and the third-level status is the fault status. The normal status indicates that all parameters of the Internet of Things device are within the normal range. The warning status indicates that there are small-scale abnormalities in the parameters of the Internet of Things device and there are potential hidden dangers of failure. The fault status indicates that there are large-scale abnormalities in the parameters of the Internet of Things device and the Internet of Things device has failed. In the first-level status, the device does not issue a warning message. In the second-level status, a text warning message is sent to the user mobile terminal of the manager to remind the manager to pay attention to the operating status of the device. In the third-level status, an emergency alarm message is sent to the user mobile terminal of the manager. According to the prediction output results of the monitoring model, maintenance suggestions are provided to the manager, and the maintenance suggestions include prediction of the cause of the failure, maintenance means and required parts.

7. An artificial intelligence-based IoT device operation monitoring method and system, characterized by: An artificial intelligence-based Internet of Things device operation monitoring method according to any one of claims 1 to 6, comprising a central processing unit and a terminal processing unit, wherein one central processing unit is connected to no less than two terminal processing units at the same time, the central processing unit comprises a central computer, the central computer has a built-in model construction module and an early warning decision module, the terminal processing unit comprises a monitoring terminal computer, the monitoring terminal computer has a built-in data acquisition module, a data preprocessing module and a real-time monitoring module, the model construction module is used for constructing a monitoring model, the early warning decision module receives real-time monitoring data of the Internet of Things device and uses the monitoring model to output the operating status of the device, the data acquisition module is connected to the sensor via a wireless signal to collect real-time operating data of the Internet of Things device, the data preprocessing module preprocesses the collected data, and the real-time monitoring module inputs the preprocessed equipment operation data into the early warning decision module after abnormal data detection to monitor and issue an early warning of the equipment operation status.

8. The artificial intelligence-based IoT device operation monitoring system according to claim 7, characterized in that: The central computer in the central processing unit is remotely connected to the user mobile terminal via a wireless signal. The central computer is communicatively connected to the monitoring terminal computer in the terminal processing unit. The monitoring terminal computer is connected to the Internet of Things device. A data anomaly detection algorithm is set between the monitoring terminal computer and the central computer. The data anomaly detection algorithm judges the real-time data input by the monitoring terminal computer. The data anomaly detection algorithm performs an anomaly judgment on the real-time data. If there is no anomaly in the real-time data, the real-time data is input into the central computer for subsequent processing; if there is an anomaly in the real-time data, the real-time data is returned to the data preprocessing module in the monitoring terminal computer for further processing.

Images