Electrical equipment state monitoring system and method based on Internet of Things

Abstract

The invention discloses an electrical equipment state monitoring system and method based on the Internet of Things, and belongs to the field of the Internet of Things, and the method comprises the following steps: installing an intelligent sensor node with a built-in edge calculation and self-diagnosis function, and carrying out the self-adaptive adjustment of frequency collection data according to the working state and historical data of electrical equipment; an edge server is added between the gateway device and the cloud server, and hierarchical data analysis is carried out; training the historical data of the electrical equipment by using a machine learning algorithm, constructing an equipment health state model, and performing fault prediction and health assessment; an early warning condition value is set according to the prediction result, and the intelligent alarm system carries out fault processing according to the early warning condition value. The maintenance cost can be reduced, and the efficiency and accuracy of maintenance work can be improved.

Description

Technical Field

[0001] The present invention relates to the field of Internet of Things, and more specifically, to an electrical equipment status monitoring system and method based on the Internet of Things. Background Art

[0002] Internet technology is the foundation for the development of the Internet of Things. The core and foundation of the Internet of Things is still the Internet. It is an extended and expanded network based on the Internet. The Internet of Things integrates with the Internet through various wired and wireless networks to transmit the information of objects in real time and accurately. It is a ubiquitous network built on the Internet. In the past two decades, the rapid development of the Internet has brought extraordinary opportunities to the world. Looking at the history of the development of the Internet, it can be seen that the development of the Internet has the characteristics of industrialization of operations, commercialization of applications, globalization of interconnection, broadband interconnection, multi-business integrated platform and intelligence. The development and application of the Internet of Things will far exceed the Internet and will further change people's production, work, life and learning methods.

[0003] The power industry is also ushering in a new technological change with the development of the Internet of Things. Electrical cabinets are generally composed of electrical equipment such as electromagnetic relays, fuses, AC contactors, residual current protective devices, capacitors and meters. These devices have strict requirements on the environmental conditions of the working environment, especially in corrosive environments. Electrical equipment is affected by corrosion factors, resulting in reduced operating stability and reliability. The failure to replace accessories or new equipment in time often increases the risk of failure in electrical cabinets.

[0004] In the IoT system, data is at risk of being intercepted, tampered with, and forged during transmission from sensors to gateways, edge servers, and then to cloud servers. Sensor data and historical data stored in the cloud are easily targeted by cyber attacks, and there is a risk of data leakage and loss. Sensor data may contain noise and errors, affecting the accuracy and reliability of the monitoring system. Sensor failures, network delays, and transmission errors may cause data loss and incompleteness, affecting the real-time and accuracy of the system. There are incompatibilities in protocols and interfaces between electrical equipment and sensors from different manufacturers and models, which increases the difficulty of system integration. Summary of the invention

[0005] The purpose of the present invention is to overcome the deficiencies of the prior art and provide an electrical equipment status monitoring system and method based on the Internet of Things.

[0006] The object of the present invention is achieved through the following solutions:

[0007] An embodiment of the present invention provides an electrical equipment status monitoring method based on the Internet of Things, comprising the steps of: installing intelligent sensor nodes with built-in edge computing and self-diagnosis functions, and adaptively adjusting the frequency of data collection according to the working status and historical data of the electrical equipment; adding an edge server between a gateway device and a cloud server to perform hierarchical data analysis; using a machine learning algorithm to train the historical data of the electrical equipment, constructing an equipment health status model, and performing fault prediction and health assessment; setting a warning condition value according to the prediction result, and the intelligent alarm system performing fault processing according to the warning condition value.

[0008] Furthermore, the frequency of data collection is adaptively adjusted according to the working status and historical data of the electrical equipment, including: classifying according to equipment operating parameters to determine the collection priority of each parameter; using historical data to establish a baseline model to define characteristic thresholds of the equipment under four working conditions: normal, warning, abnormal and fault; comparing the real-time collected data with the baseline model to identify the current working status of the equipment, and using a machine learning algorithm or a rule engine to perform real-time analysis on the collected data to identify status changes.

[0009] Furthermore, the characteristic thresholds of the equipment under four working conditions, namely normal, warning, abnormal and fault, are defined, including: when the equipment is in normal operating state, the collection frequency is set to low-frequency mode, and in low-frequency mode, the collection interval is set to every 10 minutes; when the equipment parameters are close to the warning threshold, the system automatically increases the collection frequency to capture potential abnormalities. In the warning state, the collection interval is shortened to every 1 minute; when the equipment parameters exceed the warning threshold but do not reach the fault threshold, the system further increases the collection frequency. In the abnormal state, the collection interval is every 10 seconds; when the equipment parameters exceed the fault threshold, the system immediately enters the high-frequency collection mode to ensure that detailed data of the fault is captured. In the fault state, the collection interval is every second, and the alarm mechanism is triggered at the same time.

[0010] Furthermore, an edge server is added between the gateway device and the cloud server to perform hierarchical data analysis, including: the edge server receives sensor data from multiple gateway devices through a local area network or a dedicated communication link, cleans and preprocesses the received data, removes noise and fills in missing values, and performs preliminary formatting processing, and filters out redundant or irrelevant data according to preset rules, retaining only key data.

[0011] Furthermore, the machine learning algorithm is used to train the historical data of the electrical equipment, build an equipment health status model, and perform fault prediction and health assessment, including: selecting characteristic parameters related to the health status of the equipment, extracting time series features, and standardizing the data to improve the model training effect; building a random forest model, dividing the data into a training set, a validation set, and a test set to ensure the generalization ability of the model; using the training set to train the model, adjusting the model parameters so that it can effectively capture the health status characteristics of the equipment, and evaluating the stability and generalization ability of the model through cross-validation to prevent overfitting; using a sliding window method to predict time series data, and evaluating the health status of the equipment in real time. According to historical data and expert experience, thresholds for health status and fault status are set, and when the predicted value exceeds the threshold, a fault warning is issued; and according to the health status score output by the prediction model, the health index trend analysis of the equipment is calculated.

[0012] Further, the construction of the random forest model includes: performing statistical analysis on sensor data under normal operating conditions, calculating the mean μ and the standard deviation σ, setting the threshold range of the health state to [μ-kσ, μ+kσ], where k is a safety factor, and the threshold range of the health state = [μ-kσ, μ+kσ], using the trained random forest model to predict the real-time sensor data to obtain a health state score or a predicted value ypred, calculating the health index of the device according to the output of the model, and setting the value range of the health index to 0 to 100,

[0013]

[0014] When the health index or predicted value ypred of the device exceeds the health status threshold range and approaches or exceeds the fault status threshold, a fault warning is issued.

[0015] Furthermore, the warning condition value is set according to the prediction result, and the intelligent alarm system performs fault processing according to the warning condition value, including: if the vibration sensor data range is [3,7], the equipment is operating within the healthy state range; if the vibration sensor data warning range is (7,8], the equipment state is close to abnormal and a fault will occur, and a warning is required and an inspection is arranged; if the vibration sensor data fault range is >8, the equipment state is abnormal and a fault will occur, and immediate measures need to be taken.

[0016] In the second aspect, an embodiment of the present invention provides an electrical equipment status monitoring system based on the Internet of Things, including: a data acquisition module, which is used to install intelligent sensor nodes with built-in edge computing and self-diagnosis functions, and adaptively adjust the frequency of data collection according to the working status and historical data of the electrical equipment; a data analysis module, which is used to add an edge server between the gateway device and the cloud server to perform hierarchical data analysis; a model building module, which is used to use a machine learning algorithm to train the historical data of the electrical equipment, build an equipment health status model, and perform fault prediction and health assessment; an early warning processing module, which is used to set early warning condition values ​​according to the prediction results, and the intelligent alarm system performs fault processing according to the early warning condition values.

[0017] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the processor executes the computer program, any step of the above-mentioned electrical equipment status monitoring method based on the Internet of Things is implemented.

[0018] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the above-mentioned electrical equipment status monitoring method based on the Internet of Things is implemented.

[0019] The beneficial effects of the present invention include:

[0020] The present invention uses machine learning algorithms to analyze and model historical data, predict equipment failure trends in advance, set warning thresholds, and issue alarms. Take preventive measures before a failure occurs to reduce equipment failure rates and extend equipment life. Improve equipment operational reliability and reduce unexpected downtime and failure events through real-time monitoring and fault warnings. Ensure the continuity and stability of the production process, and improve production efficiency and product quality. Based on the data from equipment status monitoring, adopt a predictive maintenance (PdM) strategy to optimize maintenance plans and resource allocation. Reduce unnecessary scheduled maintenance, reduce maintenance costs, and improve the efficiency and accuracy of maintenance work. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] 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 labor.

[0022] Figure 1 A flowchart of an electrical equipment status monitoring method based on the Internet of Things provided by an embodiment of the present invention;

[0023] Figure 2 A schematic diagram of the structure of an electrical equipment status monitoring system based on the Internet of Things provided by an embodiment of the present invention;

[0024] Figure 3 An internal structural diagram of a computer device of an electrical equipment status monitoring system and method based on the Internet of Things provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0025] All features disclosed in all embodiments in this specification, or steps in all methods or processes implicitly disclosed, except for mutually exclusive features and / or steps, can be combined and / or expanded or replaced in any manner.

[0026] like Figure 1 to Figure 3 As shown in FIG. 1 , an embodiment of the present invention is provided, which provides an electrical equipment status monitoring method based on the Internet of Things, including:

[0027] S1: Install smart sensor nodes with built-in edge computing and self-diagnosis functions, and adaptively adjust the frequency of data collection according to the working status and historical data of electrical equipment.

[0028] Preferably, the equipment is classified according to its operating parameters to determine the collection priority of each parameter; historical data is used to establish a baseline model to define characteristic thresholds of the equipment in four working states: normal, warning, abnormal, and faulty; real-time collected data is compared with the baseline model to identify the current working state of the equipment, and machine learning algorithms or rule engines are used to analyze the collected data in real time to identify state changes.

[0029] Furthermore, when the equipment is in normal operation, the collection frequency is set to low-frequency mode. In low-frequency mode, the collection interval is set to every 10 minutes; when the equipment parameters are close to the warning threshold, the system automatically increases the collection frequency to capture potential abnormalities. In the warning state, the collection interval is shortened to every 1 minute; when the equipment parameters exceed the warning threshold but do not reach the fault threshold, the system further increases the collection frequency. In the abnormal state, the collection interval is every 10 seconds; when the equipment parameters exceed the fault threshold, the system immediately enters the high-frequency collection mode to ensure that detailed data of the fault is captured. In the fault state, the collection interval is every second, and the alarm mechanism is triggered at the same time.

[0030] S2: Add an edge server between the gateway device and the cloud server to perform hierarchical data analysis.

[0031] Preferably, the edge server receives sensor data from multiple gateway devices through a local area network or a dedicated communication link, cleans and preprocesses the received data, removes noise and completes missing values, and performs preliminary formatting, filtering out redundant or irrelevant data according to preset rules, and retaining only key data.

[0032] Furthermore, the edge server is connected to multiple gateway devices via a local area network or a dedicated communication link.

[0033] The gateway device collects sensor data from electrical equipment, including voltage, current, temperature, vibration and other parameters, and sends the data to the edge server.

[0034] Furthermore, the temperature sensor generates an abnormally high value due to external interference at a certain moment. Through median filtering, the high value is filtered out and replaced by the middle value of the adjacent time point.

[0035] Furthermore, if a sensor has no data recorded for a period of time during the collection process, the data of the missing time period can be estimated using the data points before and after through linear interpolation.

[0036] Furthermore, all sensor data are converted to JSON format, the timestamp is unified to UTC time, and temperature data in different units (such as Celsius and Fahrenheit) are converted to the same unit. For current data with a high collection frequency, if there is little change within a certain period of time, key points are retained at a certain time interval, and data points with little change are filtered out.

[0037] S3: Use machine learning algorithms to train historical data of electrical equipment, build equipment health status models, and perform fault prediction and health assessment.

[0038] Preferably, characteristic parameters related to the health status of the equipment are selected, time series features are extracted, and the data is standardized to improve the model training effect; a random forest model is constructed, and the data is divided into a training set, a validation set, and a test set to ensure the generalization ability of the model; the model is trained using the training set, and the model parameters are adjusted so that it can effectively capture the health status characteristics of the equipment. The stability and generalization ability of the model are evaluated through cross-validation to prevent overfitting; the time series data is predicted using a sliding window method, and the health status of the equipment is evaluated in real time. According to historical data and expert experience, thresholds for health status and fault status are set, and when the predicted value exceeds the threshold, a fault warning is issued; according to the health status score output by the prediction model, the health index trend analysis of the equipment is calculated.

[0039] Preferably, the sensor data under normal operation is statistically analyzed to calculate the mean μ and standard deviation σ, and the threshold range of the health state is set to [μ-kσ,μ+kσ], where k is the safety factor,

[0040] Health status threshold range = [μ-kσ, μ+kσ]

[0041] Use the trained random forest model to predict the real-time sensor data and obtain the health status score or predicted value ypred. According to the output of the model, calculate the health index of the device and set the value range of the health index to 0 to 100.

[0042]

[0043] When the health index or predicted value ypred of the device exceeds the health status threshold range and approaches or exceeds the fault status threshold, a fault warning is issued.

[0044] Furthermore, sensor data of normal operation over a period of time (such as one month) is collected.

[0045] Calculate the mean μ and standard deviation σ of the sensor data.

[0046] Assume that the data collected by the temperature sensor in one month is as follows (unit: °C):

[0047] [22.1,22.5,22.3,22.7,22.2,22.6,22.4,22.5,22.3,22.6,22.4,22.5,22.3,22.5,22.4][22.1,22.5,22.3,22.7,22.2,22.6,22.4,22.5,22.3,22.6,22.4,22.5,22.3,22.5,22.4][22.1,22.5,22.3,22.7,22.2,22.6,22.4,22.5,22.3,22.6,22.4,22.5,22.3,22.5,22.4]

[0048] Calculation results: mean μ = 22.4, standard deviation σ = 0.15, assuming k = 2; health status threshold range [μ-2σ, μ+2σ] = [22.4-2×0.15, 22.4+2×0.15] = [22.1, 22.7]. Use the trained random forest model to predict the real-time sensor data to obtain the health status score or predicted value. Collect real-time sensor data and input it into the trained random forest model. The model outputs the health status score or predicted value.

[0049] Assume that the real-time collected temperature data is 22.8℃, and the health status score ypred=75 predicted by the random forest model. Calculate the health index of the device based on the output of the model, and set the value range of the health index to 0 to 100. The device health index HI can directly take the predicted value, that is, HI=75.

[0050] If the health index is lower than the threshold of 50, an early warning signal is issued. The health status of the equipment is evaluated based on the health index HI, and corresponding maintenance decisions are made. The real-time temperature data of 22.8℃ exceeds the health status threshold range [22.1, 22.7], indicating that the equipment is abnormal; the health index HI = 75, although it exceeds the health status threshold range, the score is still at a high level, indicating that the equipment is still within an acceptable range, but needs to be monitored and inspected.

[0051] S4: Set the warning condition value according to the prediction result, and the intelligent alarm system performs fault processing according to the warning condition value.

[0052] Preferably, if the vibration sensor data range is [3,7], the equipment is operating within the healthy state range; if the vibration sensor data warning range is (7,8], the equipment state is close to abnormal and a fault will occur, and an early warning is required and an inspection is arranged; if the vibration sensor data fault range is >8, the equipment state is abnormal and a fault will occur, and immediate measures must be taken.

[0053] According to one aspect of an embodiment of the present invention, a computer program product or a computer program is provided, the computer program product or the computer program includes a computer instruction, and the computer instruction is stored in a computer-readable storage medium. A processor of a computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the method provided in the above various optional implementations.

[0054] As another aspect, an embodiment of the present invention further provides a computer-readable medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements the method described in the above embodiment.

[0055] In addition to the above examples, those skilled in the art may obtain other embodiments based on the above disclosure or by using the knowledge or technology in the relevant field to make changes. The features of each embodiment may be interchangeable or replaced. The changes and modifications made by those skilled in the art do not depart from the spirit and scope of the present invention and should be within the scope of protection of the claims attached to the present invention.

Claims

1. A method for monitoring the state of electrical equipment based on the Internet of Things, characterized in that: The following steps are involved: Install smart sensor nodes with built-in edge computing and self-diagnosis functions to adaptively adjust the frequency of data collection based on the working status and historical data of electrical equipment; Add edge servers between gateway devices and cloud servers to perform hierarchical data analysis; Use machine learning algorithms to train historical data of electrical equipment, build equipment health status models, and perform fault prediction and health assessment; The warning condition value is set according to the prediction result, and the intelligent alarm system performs fault processing according to the warning condition value.

2. The electrical equipment status monitoring method based on the Internet of Things according to claim 1 is characterized in that: The method of adaptively adjusting the frequency of data collection based on the working status and historical data of the electrical equipment includes: classifying the equipment according to its operating parameters to determine the collection priority of each parameter; using historical data to establish a baseline model to define characteristic thresholds of the equipment under four working conditions: normal, warning, abnormal, and fault; comparing the real-time collected data with the baseline model to identify the current working status of the equipment, and using a machine learning algorithm or a rule engine to perform real-time analysis on the collected data to identify status changes.

3. The electrical equipment status monitoring method based on the Internet of Things according to claim 2 is characterized in that: The characteristic thresholds of the device under four working states of normal, warning, abnormal and fault are defined, including: when the device is in normal operation, the collection frequency is set to low-frequency mode, and in the low-frequency mode, the collection interval is set to every 10 minutes; when the device parameters are close to the warning threshold, the system automatically increases the collection frequency to capture potential abnormalities, and in the warning state, the collection interval is shortened to every 1 minute; when the device parameters exceed the warning threshold but do not reach the fault threshold, the system further increases the collection frequency, and in the abnormal state, the collection interval is every 10 seconds; when the device parameters exceed the fault threshold, the system immediately enters the high-frequency collection mode to ensure that the detailed data of the fault is captured. In the fault state, the collection interval is every second, and the alarm mechanism is triggered at the same time.

4. The electrical equipment status monitoring method based on the Internet of Things according to claim 1 is characterized in that: The edge server is added between the gateway device and the cloud server to perform hierarchical data analysis, including: The edge server receives sensor data from multiple gateway devices through a local area network or a dedicated communication link, cleans and preprocesses the received data, removes noise, fills in missing values, and performs preliminary formatting. According to preset rules, it filters out redundant or irrelevant data and retains only key data.

5. The electrical equipment status monitoring method based on the Internet of Things according to claim 1 is characterized in that: The machine learning algorithm is used to train the historical data of electrical equipment, build an equipment health status model, and perform fault prediction and health assessment, including: Select characteristic parameters related to the health status of the equipment, extract time series features, standardize the data, and improve the model training effect; build a random forest model, divide the data into training set, validation set and test set to ensure the generalization ability of the model; use the training set to train the model, adjust the model parameters so that it can effectively capture the health status characteristics of the equipment, and evaluate the stability and generalization ability of the model through cross-validation to prevent overfitting; use the sliding window method to predict time series data, evaluate the health status of the equipment in real time, set the thresholds of health status and fault status based on historical data and expert experience, and issue a fault warning when the predicted value exceeds the threshold; calculate the health index trend analysis of the equipment based on the health status score output by the prediction model.

6. The method for monitoring the state of electrical equipment based on the Internet of Things according to claim 1, characterized in that: The random forest model is constructed, comprising: Perform statistical analysis on sensor data under normal operating conditions, calculate the mean μ and standard deviation σ, and set the threshold range of the health state to [μ-kσ, μ+kσ], where k is the safety factor, and the threshold range of the health state = [μ-kσ, μ+kσ]; Use the trained random forest model to predict the real-time sensor data and obtain the health status score or predicted value ypred. According to the output of the model, calculate the health index of the device and set the value range of the health index to 0 to 100. When the health index or predicted value ypred of the device exceeds the health status threshold range and approaches or exceeds the fault status threshold, a fault warning is issued.

7. The electrical equipment status monitoring method based on the Internet of Things according to claim 1 is characterized in that: The step of setting the warning condition value according to the prediction result, and the intelligent alarm system performing fault processing according to the warning condition value, includes: If the vibration sensor data range is [3,7], the equipment is operating within the healthy state range; if the vibration sensor data warning range is (7,8], the equipment state is close to abnormality and a fault will occur, and a warning is required and an inspection is arranged; if the vibration sensor data fault range is >8, the equipment state is abnormal and a fault will occur, and immediate measures must be taken.

8. An electrical equipment status monitoring system based on the Internet of Things, characterized in that: include: Data acquisition module, which is used to install intelligent sensor nodes with built-in edge computing and self-diagnosis functions, and adaptively adjust the frequency of data acquisition according to the working status and historical data of electrical equipment; Data analysis module, used to add edge servers between gateway devices and cloud servers to perform hierarchical data analysis; Model building module, which is used to use machine learning algorithms to train historical data of electrical equipment, build equipment health status models, and perform fault prediction and health assessment; The early warning processing module is used to set the early warning condition value according to the prediction result, and the intelligent alarm system performs fault processing according to the early warning condition value.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the electrical equipment status monitoring method based on the Internet of Things according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the electrical equipment status monitoring method based on the Internet of Things according to any one of claims 1 to 7 are implemented.

Images