Intelligent monitoring method and system for power adapter based on internet of things

Abstract

The application relates to the technical field of power adapter monitoring, and discloses a power adapter intelligent monitoring method and system based on the Internet of Things, which comprises the following steps: collecting multi-dimensional operation data under the operation condition of a power adapter, and identifying a target monitoring area of the power adapter; adaptively performing spectral decomposition on the multi-dimensional operation data to obtain target analysis data, positioning an abnormal node in the target monitoring area to obtain an abnormal circuit area; collecting infrared thermal imaging data and electromagnetic radiation data of the power adapter, analyzing the thermal stress distribution characteristics of the power adapter, performing frequency domain normalization processing on the electromagnetic radiation data to obtain a standard frequency domain spectrum; performing three-dimensional topological analysis on the abnormal circuit area to obtain a fault form feature vector, and performing health state analysis on the power adapter to obtain a device health analysis report. The application can improve the accuracy of detecting abnormal problems of the power adapter under complex environments.

Description

TECHNICAL FIELD

[0001] The application relates to an Internet of Things-based intelligent monitoring method and system for a power adapter and belongs to the technical field of power adapter monitoring. BACKGROUND

[0002] Under the background of the vigorous development of the modern Internet of Things, the effective monitoring of the running state of the power adapter, which is a key component for stable power supply of various electronic devices, is of great significance. From ensuring the continuous and normal operation of various smart home appliances in a smart home system to ensuring stable power support for production equipment in an industrial Internet of Things environment, the reliable operation of the power adapter is the cornerstone of the stable operation of the entire Internet of Things system and is directly related to the normal order and efficiency of production and life. Timely and accurate grasp of the running state of the power adapter can detect potential faults in advance, prevent equipment downtime, and ensure uninterrupted operation of the system.

[0003] At present, the intelligent monitoring of the power adapter mostly relies on traditional basic monitoring methods. The traditional power adapter monitoring system mostly adopts a combination of threshold alarm and regular manual inspection, and determines the device state through the static threshold of basic parameters such as voltage and current. However, with the high frequency and complex topology of new power electronic devices, the device faults exhibit complex characteristics such as multi-physical field coupling (electrical-thermal-mechanical stress interaction), dynamic load mutation (millisecond-level power fluctuation), and hidden degradation accumulation (electrolytic capacitor value gradual change), which makes the static threshold determination often inaccurate (i.e., there are abnormal points in the circuit and the problem cannot be found), resulting in the inability to accurately identify the problems existing in the power adapter. SUMMARY

[0004] The application provides an Internet of Things-based intelligent monitoring method and system for a power adapter, which mainly aims to improve the accuracy of detecting abnormal problems of the power adapter in a complex environment.

[0005] To achieve the above-mentioned purpose, the Internet of Things-based intelligent monitoring method for a power adapter provided by the application comprises the following steps:

[0006] An Internet of Things device is deployed at a key node of the power adapter to collect multi-dimensional running data of the power adapter under running conditions by using the Internet of Things device, calculate the parameter deviation degree of the multi-dimensional running data, extract key parameter points of the multi-dimensional running data based on the parameter deviation degree index, and identify a target monitoring area of the power adapter by using the key parameter points.

[0007] Based on the target monitoring area, adaptive spectrum decomposition is performed on the multidimensional operating data to obtain target analysis data. The multidimensional electrical characteristics of the power adapter are identified using the target analysis data. The abnormal nodes in the target monitoring area are located using the multidimensional electrical characteristics to obtain the abnormal circuit area.

[0008] Infrared thermal imaging data and electromagnetic radiation data of the power adapter are collected. Non-uniformity correction is performed on the infrared thermal imaging data to obtain a calibrated thermal distribution map. The temperature gradient characteristics of the calibrated thermal distribution map are identified. The thermal stress distribution characteristics of the power adapter are analyzed using the temperature gradient characteristics. The electromagnetic radiation data is frequency domain normalized to obtain a standard frequency domain spectrum.

[0009] A three-dimensional topology analysis is performed on the abnormal circuit region to obtain the fault morphology feature vector. The thermal stress distribution characteristics, standard frequency domain spectrum and fault morphology feature vector are fused from multiple sources to obtain a fused diagnostic dataset. The health status of the power adapter is analyzed using the fused diagnostic dataset to obtain a device health analysis report.

[0010] Optionally, calculating the parameter deviation of the multidimensional operational data includes:

[0011] Identify the parameter measurement types of the multidimensional operational data;

[0012] Based on the parameter measurement type, the multidimensional operational data is classified to obtain categorized data;

[0013] Query the measured values ​​of the classified data and the observation times corresponding to the measured values;

[0014] Based on the measured values, the deviation of the categorical data from the observed values ​​within the observation period is calculated using the following formula:

[0015] ;

[0016] in, The value represents the deviation of the observed value, and t represents the observation time. This represents the measured value of the i-th class of data at time t. express The corresponding reference value, This represents the deviation weight coefficient of the i-th category data;

[0017] The parameter deviation of the multidimensional operational data is determined based on the deviation of the observed values.

[0018] Optionally, identifying the target monitoring area of ​​the power adapter using the key parameter points includes:

[0019] The key parameter points are mapped to a three-dimensional coordinate system to obtain a standardized coordinate dataset;

[0020] Perform time-synchronized interpolation on the standardized coordinate dataset to obtain a spatiotemporally aligned dataset;

[0021] Data clustering is performed on the spatiotemporal aligned dataset to obtain an initial monitoring cluster set;

[0022] Spatiotemporal correlation verification is performed on the initial monitoring cluster set to extract the effective dataset from the initial monitoring cluster set using the verification results of the spatiotemporal correlation verification.

[0023] Based on the effective dataset, a dynamic risk analysis is performed on the power adapter to identify the priority monitoring area of ​​the power adapter.

[0024] The priority monitoring area is processed to obtain the target monitoring area.

[0025] Optionally, the step of performing adaptive spectral decomposition on the multidimensional operational data based on the target monitoring area to obtain target analysis data includes:

[0026] Identify the regional characteristics of the target monitoring area;

[0027] Based on the aforementioned regional characteristics, the multidimensional operational data is filtered to obtain filtered data;

[0028] Calculate the signal Shannon entropy of the filtered data;

[0029] Based on the Shannon entropy of the signal, the selected data is dynamically decomposed to obtain a time-frequency matrix;

[0030] The time-frequency matrix is ​​subjected to time-frequency sub-generation optimization to obtain target analysis data.

[0031] Optionally, the step of using the multi-dimensional electrical characteristics to locate abnormal nodes in the target monitoring area to obtain the abnormal circuit area includes:

[0032] The multidimensional electrical features are filtered to obtain the filtered features;

[0033] Using the aforementioned screening features, anomaly detection is performed on the target monitoring area to obtain anomaly detection data;

[0034] The anomaly detection data is transformed into a matrix to obtain the anomaly detection matrix;

[0035] Using the anomaly detection matrix, the anomaly range of the target monitoring area is identified to obtain the preliminary identified area;

[0036] calculating an abnormal similarity of the electrical data in the preliminary identified region;

[0037] performing region segmentation on the target monitoring region based on the abnormal similarity to obtain an abnormal segmentation region;

[0038] identifying a region center point of the abnormal segmentation region;

[0039] performing abnormal positioning on the target monitoring region by using the region center point to obtain an abnormal circuit region.

[0040] Optionally, the non-uniformity correction on the infrared thermal imaging data to obtain a calibrated thermal distribution map comprises:

[0041] detecting bad points of the infrared thermal imaging data to obtain bad point coordinates;

[0042] performing matrix conversion on the bad point coordinates to obtain a bad point coordinate matrix;

[0043] performing neighborhood interpolation on the bad point coordinate matrix to obtain an interpolated image;

[0044] calculating a gain matrix of the interpolated image by using the following formula:

[0045] ;

[0046] wherein, the gain matrix, a pixel point of the interpolated image an original gray value at a high calibration temperature, a pixel point of the interpolated image an original gray value at a low calibration temperature, the high calibration temperature, the low calibration temperature;

[0047] calculating an offset matrix of the interpolated image by using the following formula based on the gain matrix:

[0048] ;

[0049] wherein, the offset matrix, a pixel point of the interpolated image an original gray value at a low calibration temperature, the low calibration temperature;

[0050] performing linear correction on the infrared thermal imaging data based on the gain matrix and the offset matrix to obtain a linear correction image;

[0051] Perform scene adaptive correction on the linear correction image to obtain a calibration thermal distribution map.

[0052] Optionally, the temperature gradient feature of the calibration thermal distribution map is identified, including:

[0053] Perform effective region mask operation on the thermal distribution map to obtain a mask image;

[0054] Perform morphological closing operation processing on the mask image to obtain an effective image

[0055] Calculate the gradient matrix of the effective image;

[0056] Perform polar coordinate conversion on the gradient matrix to obtain a gradient amplitude matrix;

[0057] Perform adaptive threshold segmentation on the gradient amplitude matrix to obtain a binary image;

[0058] The temperature gradient feature of the calibration thermal distribution map is identified using the binary image.

[0059] Optionally, the three-dimensional topological analysis of the abnormal circuit region is performed to obtain a fault morphology feature vector, including:

[0060] Perform three-dimensional finite element modeling on the abnormal circuit region to obtain a visualized circuit region;

[0061] Identify the current density distribution data, thermal stress data, and electromagnetic field intensity distribution data in the visualized circuit region;

[0062] Perform vectorization processing on the current density distribution data, the thermal stress data, and the electromagnetic field intensity distribution data to obtain a fault morphology feature vector.

[0063] Optionally, the multi-source data fusion of the thermal stress distribution characteristics, the standard frequency domain spectrum, and the fault morphology feature vector is performed to obtain a fusion diagnosis data set, including:

[0064] Perform matrix conversion on the thermal stress distribution characteristics, the standard frequency domain spectrum, and the fault morphology feature vector to obtain a first matrix, a second matrix, and a third matrix;

[0065] Perform tensor combination on the first matrix, the second matrix, and the third matrix to obtain a high-order tensor;

[0066] Perform multilinear decomposition on the high-order tensor to obtain a low-rank matrix;

[0067] Perform matrix solving on the low-rank matrix to obtain a matrix element;

[0068] Element fusion is performed on the matrix elements to obtain a fused diagnosis data set.

[0069] To solve the above problems, the application further provides an Internet of Things-based intelligent monitoring system for a power adapter, which comprises:

[0070] A target area identification module is configured to deploy an Internet of Things device at a key node of the power adapter, collect multi-dimensional operation data of the power adapter under an operation condition by using the Internet of Things device, calculate a parameter deviation degree of the multi-dimensional operation data, extract a key parameter point of the multi-dimensional operation data based on the parameter deviation degree index, and identify a target monitoring area of the power adapter by using the key parameter point.

[0071] An abnormal area identification module is configured to perform adaptive spectral decomposition on the multi-dimensional operation data based on the target monitoring area, obtain target analysis data, identify multi-dimensional electrical characteristics of the power adapter by using the target analysis data, locate an abnormal node in the target monitoring area by using the multi-dimensional electrical characteristics, and obtain an abnormal circuit area.

[0072] A multi-dimensional data processing module is configured to collect infrared thermal imaging data and electromagnetic radiation data of the power adapter, perform non-uniformity correction on the infrared thermal imaging data to obtain a calibrated thermal distribution map, identify a temperature gradient feature of the calibrated thermal distribution map, analyze thermal stress distribution characteristics of the power adapter by using the temperature gradient feature, and perform frequency domain normalization processing on the electromagnetic radiation data to obtain a standard frequency domain spectrum.

[0073] A state analysis module is configured to perform three-dimensional topological analysis on the abnormal circuit area to obtain a fault morphology feature vector, perform multi-source data fusion on the thermal stress distribution characteristics, the standard frequency domain spectrum and the fault morphology feature vector to obtain a fused diagnosis data set, analyze a health state of the power adapter by using the fused diagnosis data set, and obtain a device health analysis report.

[0074] Compared with the problems described in the background art, the embodiments of the present application deploy Internet of Things devices at the key node of the power adapter, integrate multiple sensor modules such as temperature, humidity, and vibration, and collect multi-dimensional operating data in real time under the operating condition, laying a data foundation for subsequent analysis, and calculating the parameter deviation degree of the multi-dimensional operating data to quickly identify abnormal parameters, thereby screening out points with a deviation degree higher than a preset threshold, focusing on the target monitoring area, and improving the analysis efficiency. Further, the present application identifies the multi-dimensional electrical characteristics of the target monitoring area to realize the positioning of the abnormal circuit area of the target monitoring area, facilitating accurate positioning of the abnormal circuit area and facilitating key monitoring of potential hidden dangers. Further, the present application collects infrared thermal imaging data and electromagnetic radiation data of the power adapter to understand the surface temperature and electromagnetic energy emission, and further understand the thermal mechanical properties and working stability of the power adapter. Still further, the present application performs three-dimensional topological analysis on the abnormal circuit area, generates a visual circuit area with the help of finite element modeling, and then obtains the fusion diagnosis data set of the visual circuit area, so as to evaluate the health status of the power adapter from multiple dimensions such as thermal performance, electromagnetic compatibility, and circuit failure. Therefore, the present application can improve the accuracy of detecting abnormal problems of the power adapter in a complex environment. BRIEF DESCRIPTION OF DRAWINGS

[0075] Figure 1 A flowchart of a power adapter intelligent monitoring method based on Internet of Things provided by an embodiment of the present application is shown.

[0076] Figure 2 A module diagram of a power adapter intelligent monitoring system based on Internet of Things provided by an embodiment of the present application is shown.

[0077] The object implementation, functional features, and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0078] It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0079] The embodiments of the present application provide a power adapter intelligent monitoring method based on Internet of Things. The execution subject of the power adapter intelligent monitoring method based on Internet of Things includes but is not limited to at least one of electronic devices such as a server, a terminal, etc. which can be configured to execute the method provided by the embodiments of the present application. In other words, the power adapter intelligent monitoring method based on Internet of Things can be executed by software or hardware installed in a terminal device or a server device. The server includes but is not limited to a single server, a server cluster, a cloud server, or a cloud server cluster, etc.

[0080] Embodiment 1:

[0081] Reference Figure 1As shown, it is a flowchart of the Internet of Things-based power adapter intelligent monitoring method provided by an embodiment of the application. In this embodiment, the Internet of Things-based power adapter intelligent monitoring method comprises:

[0082] S1, deploying Internet of Things devices at key nodes of the power adapter to collect multi-dimensional running data of the power adapter under running conditions by using the Internet of Things devices, calculate the parameter deviation degree of the multi-dimensional running data, extract key parameter points of the multi-dimensional running data based on the parameter deviation degree index, and identify the target monitoring area of the power adapter by using the key parameter points.

[0083] In this embodiment, the Internet of Things devices are deployed at the key nodes of the power adapter to collect multi-dimensional running data of the power adapter under running conditions by using the Internet of Things devices, which can collect various data in the running process of the power adapter in real time and accurately, thereby providing basic data support for subsequent analysis and decision-making.

[0084] The power adapter refers to an electronic device, which mainly converts alternating current provided by a power grid into direct current suitable for various electronic devices and stably supplies power to the devices. The Internet of Things device refers to a device that integrates various sensor modules for real-time sensing and collecting information related to the running of the power adapter, and has the function of transmitting the information through a network, such as temperature sensors, humidity sensors, vibration sensors, etc. The multi-dimensional running data refers to a series of data sets reflecting the running state of the power adapter from multiple dimensions, such as temperature, humidity, vibration, etc.

[0085] Optionally, the multi-dimensional running data can be obtained by deploying Internet of Things devices at the power input port, power output port and core power conversion module of the power adapter, and then collecting the data generated by the power adapter during running by using the deployed Internet of Things devices.

[0086] Further, the parameter deviation degree of the multi-dimensional running data can be calculated to quickly identify which parameters have abnormal changes, thereby helping users to quickly screen which parameters have abnormal changes.

[0087] The parameter deviation degree refers to an index for measuring the deviation degree of each parameter in the multi-dimensional running data relative to the reference value.

[0088] As an embodiment of the present application, the calculation of the parameter deviation degree of the multidimensional operation data comprises: identifying the parameter measurement type of the multidimensional operation data, classifying the multidimensional operation data according to the parameter measurement type to obtain classified data, querying the measured value of the classified data and the observation time corresponding to the measured value, and calculating the observation value deviation degree of the classified data within the observation time based on the measured value by using the following formula:

[0089] ;

[0090] wherein, represents the observation value deviation degree, t represents the observation time, represents the measured value of the i-th classified data at t, represents the corresponding reference value, represents the deviation weight coefficient of the i-th classified data;

[0091] According to the observation value deviation degree, the parameter deviation degree of the multidimensional operation data is determined.

[0092] The parameter measurement type refers to the measurement attribute category of different parameters in the multidimensional operation data under the power adapter operation condition, for example, voltage, current, temperature, and frequency, etc. The measured value refers to the actual measurement value of each parameter collected by various sensors deployed on the key nodes of the power adapter at a specific observation time. For example, at a certain time, the output voltage of the power adapter measured by the voltage sensor is 20V, and this 20V is the measured value of the voltage parameter. The observation time refers to the specific moment or time period of collecting the measured value in the multidimensional operation data.

[0093] Optionally, the parameter measurement type can analyze each parameter contained in the multidimensional operation data to determine its measurement type, such as different types of voltage, current, temperature, and humidity, etc. The classified data can be divided according to the parameter measurement type, and data of the same measurement type is classified into a category. The measured value can be obtained by extracting the specific measured value of each classified data from the classified data. The observation time can be obtained by querying the test time corresponding to the measured value using MySQL. The parameter deviation degree can be obtained by calculating the observation value deviation degree of different parameters in the multidimensional operation data according to the calculation formula of the observation value deviation degree.

[0094] It should be further explained that the observation value deviation calculation formula highlights the deviation between the two by calculating the square of the difference between the measured value and the reference value, the greater the difference, the more obvious the deviation; divided by the deviation weight coefficient, which considers the importance of different classification data on the result, can realize the deviation of the multi-dimensional running data of the power adapter and the reference state, help users quickly judge whether the operation is normal, and evaluate the stability and reliability of the power adapter running state, so as to discover potential fault risks in time.

[0095] The embodiment of the application can reduce the amount of data, focus on the most valuable information, and further improve the analysis efficiency by extracting the key parameter points of the multi-dimensional running data based on the parameter deviation index.

[0096] Among them, the key parameter point refers to a parameter point with a higher deviation degree selected according to the parameter deviation index in the multi-dimensional running data of the power adapter. These points represent the key information with large differences from the normal reference state in the running process of the power adapter, and can reflect the abnormal or running state that needs to be focused on.

[0097] Optionally, the key parameter point can extract the parameter point corresponding to the parameter deviation higher than the preset parameter deviation threshold from the multi-dimensional running data.

[0098] Further, the preset parameter deviation threshold can be set according to actual application requirements. If the data volume is large, the preset parameter deviation threshold can be set to 0.6, and if the data volume is small, the preset parameter deviation threshold can be set to 0.4. The specific setting needs to be combined with the actual application.

[0099] The embodiment of the application can improve the efficiency of fault discovery and processing and reduce maintenance cost by identifying the target monitoring area of the power adapter using the key parameter point.

[0100] Among them, the target monitoring area refers to the area selected as the focus of observation and detection, such as transformers, capacitors, etc.

[0101] As an embodiment of the present application, the target monitoring area of the power adapter is identified by using the key parameter points, which includes: performing three-dimensional coordinate system mapping on the key parameter points to obtain a standardized coordinate dataset, performing time synchronization interpolation operation on the standardized coordinate dataset to obtain a space-time alignment dataset, performing data clustering on the space-time alignment dataset to obtain an initial monitoring cluster set, performing space-time correlation verification on the initial monitoring cluster set, extracting an effective dataset in the initial monitoring cluster set by using the verification result of the space-time correlation verification, performing dynamic risk analysis on the power adapter based on the effective dataset to identify a priority monitoring area of the power adapter, and performing region boundary processing on the priority monitoring area to obtain the target monitoring area.

[0102] The standardized coordinate dataset refers to a dataset formed by mapping the key parameter points to a three-dimensional coordinate system and then normalizing the coordinate values to a uniform scale. The space-time alignment dataset refers to a dataset obtained by synchronizing and interpolating the standardized coordinate dataset according to timestamps, so that the data is uniformly corresponding in time and space dimensions. The initial monitoring cluster set refers to a collection of clusters obtained by dividing the data points in the space-time alignment dataset into different clusters. The effective dataset refers to a dataset composed of clusters that meet the rules after verifying and screening the initial monitoring cluster set according to the space-time correlation rules. The priority monitoring area refers to an area that needs to be monitored preferentially in the monitoring area of the power adapter.

[0103] Optionally, the standardized coordinate data set can be determined first to determine the three-dimensional spatial reference coordinate system of the power adapter, convert the physical position information corresponding to the key parameter point into coordinate values in the coordinate system, and then normalize the coordinate values to obtain the space-time alignment data set. The space-time alignment data set can be sorted and interpolated using the interpolation function library in Python according to the timestamp. The initial monitoring cluster set can be obtained by clustering the space-time alignment data set using DBSCAN clustering. The effective data set can be first defined to verify each cluster in the initial monitoring cluster set according to the space-time correlation judgment rule, such as the continuity in time and the proximity in space, and then the clusters meeting the space-time correlation requirement are extracted as the effective data set. The priority monitoring area can use a decision tree to build a dynamic risk analysis model, combine the numerical trend of the key parameter point, the size and density of the cluster, and other factors to evaluate the risk degree of the area corresponding to each effective data set, and sort the areas according to the risk degree to determine, such as three areas A, B and C obtained by monitoring the power adapter. The voltage in area A continuously rises greatly, the cluster is large and dense, the regression analysis shows that the voltage rising slope is 0.5, the decision tree determines that the risk is high; the current in area B fluctuates regularly, the cluster and the density are moderate, and the risk is evaluated as medium by time series and neural network; the temperature in area C normally fluctuates, the cluster is small and sparse, and the risk is low. According to the risk degree, the priority of area A is the highest, area B is monitored regularly, and area C is reduced in frequency. The target monitoring area can be obtained by smoothing the boundary of the priority monitoring area to remove noise points and irregular parts on the boundary, and making the boundary of the area more clear.

[0104] S2, based on the target monitoring area, performing adaptive frequency spectrum decomposition on the multi-dimensional running data to obtain target analysis data, identifying multi-dimensional electrical characteristics of the power adapter by using the target analysis data, and positioning an abnormal node in the target monitoring area by using the multi-dimensional electrical characteristics to obtain an abnormal circuit area.

[0105] The embodiment of the present application can highlight the information of different frequency characteristics in the data by performing adaptive frequency spectrum decomposition on the multi-dimensional running data based on the target monitoring area to obtain target analysis data, making the data characteristics more obvious, and further better capturing subtle changes and hidden information in the power adapter running data, thereby providing more valuable data basis for accurately identifying electrical characteristics.

[0106] As an embodiment of the present application, the adaptive spectral decomposition of the multi-dimensional operation data based on the target monitoring area to obtain target analysis data comprises: identifying the area characteristics of the target monitoring area, performing data screening on the multi-dimensional operation data based on the area characteristics to obtain screening data, calculating the signal Shannon entropy of the screening data, performing dynamic layer decomposition on the screening data based on the signal Shannon entropy to obtain a time-frequency matrix, and performing time-frequency sub-generation optimization on the time-frequency matrix to obtain target analysis data.

[0107] The signal Shannon entropy is a measure of uncertainty or information content in a signal, and the time-frequency matrix is a matrix representing a signal in both time and frequency dimensions.

[0108] Optionally, the area characteristics are used to conduct detailed investigation and analysis of the target monitoring area, collect the area information such as physical characteristics and electrical characteristics of the area, and then analyze the area information by using a sensor network, circuit design drawings and other tools. The screening data can be screened according to the identified area characteristics to determine the screening criteria and conditions. For example, if the area characteristics indicate that the area mainly focuses on signals in a certain frequency range, the frequency range is used as the screening condition to screen the required data from the multi-dimensional operation data. The signal Shannon entropy of the screening data can be calculated by using the Shannon algorithm. The time-frequency matrix can be obtained by wavelet decomposition of the screening data by using a wavelet analysis library to automatically adjust the decomposition layer number according to the signal Shannon entropy. The target analysis data can be evaluated and sorted by calculating the energy and correlation of each time-frequency sub-generation in the time-frequency matrix. According to the sorting result, the optimal time-frequency sub-generation is selected as the target analysis data, such as the time-frequency sub-generation with the maximum energy, the strongest correlation and other characteristics.

[0109] Further, the multi-dimensional electrical characteristics of the power adapter can be comprehensively understood by using the target analysis data to identify the multi-dimensional electrical characteristics of the power adapter, so as to facilitate the user to find potential fault hazards of the power adapter.

[0110] The multi-dimensional electrical characteristics refer to the performance of the power adapter in multiple dimensions of electrical performance, such as voltage characteristics, current characteristics and power characteristics.

[0111] Optionally, the multi-dimensional electrical characteristics can be obtained by identifying the voltage value, current value and power value of the power adapter in the target analysis data.

[0112] Further, the embodiment of the present application can accurately locate the abnormal area of the power adapter by locating the abnormal node in the target monitoring area using the multi-dimensional electrical characteristics, so as to monitor the area and find possible hidden dangers or faults in time.

[0113] The abnormal circuit area refers to an area with running abnormalities in the power adapter, such as voltage abnormalities at an interface or a power output port.

[0114] As an embodiment of the present application, the step of locating the abnormal node in the target monitoring area using the multi-dimensional electrical characteristics to obtain the abnormal circuit area comprises the following steps: performing feature screening on the multi-dimensional electrical characteristics to obtain screened features, performing abnormal detection on the target monitoring area using the screened features to obtain abnormal detection data, performing matrix conversion on the abnormal detection data to obtain an abnormal detection matrix, identifying the abnormal range of the target monitoring area using the abnormal detection matrix to obtain a preliminary identification area, calculating the abnormal similarity of electrical data in the preliminary identification area, performing area segmentation on the target monitoring area based on the abnormal similarity to obtain an abnormal segmentation area, identifying the area center point of the abnormal segmentation area, and performing abnormal positioning on the target monitoring area using the area center point to obtain the abnormal circuit area.

[0115] The abnormal detection matrix refers to a matrix used to represent the abnormality of electrical data in the target monitoring area.

[0116] Optionally, the screened features can be obtained by using a correlation analysis algorithm to calculate the correlation coefficient of each electrical feature with known abnormal conditions, and screening key features such as voltage fluctuation features and current distortion features according to the coefficient size and combining expert experience. The abnormal detection data can be obtained by monitoring and comparing the real-time electrical data of each node in the target monitoring area using the screened electrical features as indicators, judging whether the data of each node deviates from the normal range, and screening out the data deviating from the normal range. The abnormal detection matrix can be obtained by filling the abnormal detection results (such as Boolean values of whether abnormal, abnormal degree values, etc.) of each node at different times into the corresponding positions of the matrix according to the node number and time sequence. The preliminary identification area can be obtained by analyzing the abnormal detection matrix to find the range of rows (corresponding to nodes) and columns (corresponding to time) in which abnormal data appears in the matrix. The abnormal similarity of electrical data in the preliminary identification area can be calculated using the Euclidean distance. The abnormal segmentation area can be obtained by using a clustering algorithm such as K-means clustering, taking the abnormal similarity value as the clustering basis, and dividing the preliminary identification area into multiple segmentation areas. The area center point can be identified using the area center method.

[0117] S3, collect infrared thermal imaging data and electromagnetic radiation data of the power adapter, perform non-uniformity correction on the infrared thermal imaging data to obtain a calibrated thermal distribution map, identify a temperature gradient feature of the calibrated thermal distribution map, analyze thermal stress distribution characteristics of the power adapter by using the temperature gradient feature, and perform frequency domain normalization processing on the electromagnetic radiation data to obtain a standard frequency domain spectrum.

[0118] The embodiment of the present application can understand the temperature distribution of the surface of the power adapter and the emission characteristics of electromagnetic energy by collecting the infrared thermal imaging data and the electromagnetic radiation data of the power adapter.

[0119] The infrared thermal imaging data refers to data about thermal radiation information of an object obtained by infrared thermal imaging technology, and the electromagnetic radiation data refers to data about electromagnetic radiation characteristics of an electrical appliance generated during operation.

[0120] Optionally, the infrared thermal imaging data can be obtained by collecting infrared radiation of the power adapter during operation by an infrared imager, and the electromagnetic radiation data can be obtained by collecting electromagnetic energy of the power adapter during operation by an electromagnetic radiation tester.

[0121] Further, the embodiment of the present application can improve the quality and reliability of the infrared thermal imaging data by performing non-uniformity correction on the infrared thermal imaging data to obtain a calibrated thermal distribution map, so as to ensure that the obtained calibrated thermal distribution map can truly reflect the heating condition of the power adapter.

[0122] The calibrated thermal distribution map refers to an image that presents the temperature distribution of the surface of an object in an intuitive and visual manner after processing and optimization.

[0123] As an embodiment of the present application, the non-uniformity correction on the infrared thermal imaging data to obtain a calibrated thermal distribution map comprises: performing bad point detection on the infrared thermal imaging data to obtain bad point coordinates, performing matrix conversion on the bad point coordinates to obtain a bad point coordinate matrix, performing neighborhood interpolation on the bad point coordinate matrix to obtain an interpolated image, and calculating a gain matrix of the interpolated image by using the following formula:

[0124] ;

[0125] wherein, the gain matrix is represented by G, a pixel point of the interpolated image is represented by I, an original gray value at a high calibration temperature is represented by I high, a pixel point of the interpolated image is represented by I, an original gray value at a low calibration temperature is represented by I low, the high calibration temperature is represented by T high, Indicates a low calibration temperature;

[0126] Based on the gain matrix, the offset matrix of the interpolated image is calculated using the following formula:

[0127] ;

[0128] in, Represents the offset matrix. Represents the pixels of the interpolated image The original grayscale value at the low calibration temperature. Indicates a low calibration temperature;

[0129] Based on the gain matrix and the offset matrix, the infrared thermal imaging data is linearly corrected to obtain a linearly corrected image. The linearly corrected image is then subjected to scene adaptive correction to obtain a calibrated thermal distribution map.

[0130] In this context, the "dead pixel coordinates" refer to pixels in infrared thermal imaging data whose grayscale values ​​significantly deviate from the normal range due to various reasons (such as sensor malfunction, noise interference, etc.). The "interpolated image" refers to the image obtained after repairing the dead pixels in the infrared thermal imaging data. The "gain matrix" refers to the matrix obtained after linearly correcting the infrared thermal imaging data. The "offset matrix" refers to the matrix calculated based on the interpolated image and the gain matrix, used to further adjust the infrared thermal imaging data. The "high calibration temperature" refers to a relatively high temperature value set when calibrating the infrared thermal imaging device, and the "low calibration temperature" refers to a relatively low temperature value set when calibrating the infrared thermal imaging device.

[0131] Optionally, the defective pixel coordinates can be obtained using a statistical analysis-based defective pixel detection algorithm, employing an image processing library (such as OpenCV) to traverse pixels and perform grayscale value comparison calculations. The defective pixel coordinate matrix can be obtained by using matrix operation functions in a programming language (such as Python) to convert the list of defective pixel coordinates into a two-dimensional matrix, where the rows of the matrix correspond to the defective pixel index, and the columns correspond to the x and y coordinates. The neighbor interpolation of the defective pixel coordinate matrix to obtain the interpolated image can be achieved using bilinear interpolation. The linearly corrected image can be obtained by using matrix operations and image processing techniques, performing matrix multiplication and addition operations to operate the gain matrix and offset matrix with the original infrared thermal imaging data matrix to correct the grayscale value of each pixel. The calibration thermal distribution map can be obtained by using an adaptive correction algorithm based on scene analysis, such as utilizing the image's histogram statistics, edge features, etc., combined with adaptive threshold adjustment techniques to optimize the linearly corrected image pixel by pixel.

[0132] It needs to be further explained that the gain matrix calculation formula can calculate the change rate of the gray scale when the temperature changes by comparing the gray scale changes of the pixels at high and low calibration temperatures, and the change rate of each pixel is combined to obtain the gain matrix, so that the change rate of the gray scale of each pixel point can be adjusted, so that the gray scale of the pixel point at different temperatures can more accurately reflect the change of the actual temperature, thereby improving the accuracy and uniformity of the infrared thermal imaging image; the offset matrix calculation formula can determine the initial offset of the linear relationship between the pixel gray scale and the temperature in combination with the low calibration temperature and the corresponding gray scale after knowing the gray scale change rate reflected by the gain matrix, and can further adjust the gray scale of each pixel point, so that the gray scale of the pixel point at the low calibration temperature can accurately correspond to the actual low calibration temperature.

[0133] Further, the temperature gradient feature of the calibration thermal distribution map can be identified by the power adapter surface temperature variation severity and direction.

[0134] The temperature gradient feature refers to a set of characteristics for describing the temperature change of an object surface or space, which can intuitively reflect the change trend and severity of the temperature in different positions and directions.

[0135] Optionally, the temperature gradient feature can be identified by calculating the temperature difference and direction of adjacent pixel points in the calibration thermal distribution map.

[0136] As an embodiment of the present application, the temperature gradient feature of the calibration thermal distribution map includes: performing an effective area mask operation on the thermal distribution map to obtain a mask image, performing a morphological closing operation on the mask image to obtain an effective image, calculating a gradient matrix of the effective image, performing polar coordinate conversion on the gradient matrix to obtain a gradient amplitude matrix, performing adaptive threshold segmentation on the gradient amplitude matrix to obtain a binary image, and identifying the temperature gradient feature of the calibration thermal distribution map using the binary image.

[0137] The mask image refers to an image obtained by operating a specific mask on the original thermal distribution map, the effective image refers to an image obtained by performing a morphological closing operation on the mask image, the gradient matrix refers to a matrix recording the change rate of the gray scale of each pixel point in the horizontal and vertical directions of the effective image, and the gradient amplitude matrix refers to a matrix composed of the gradient amplitudes of all pixel points in the gradient matrix.

[0138] Optionally, the mask image can first determine the effective area in the thermal distribution map, for example, excluding the background area irrelevant to the analysis target. Then create a mask with the same size as the thermal distribution map, set the mask part corresponding to the effective area to a specific value (such as 1), and set the invalid area to other values (such as 0), and then perform a pixel-by-pixel multiplication operation on the mask and the thermal distribution map to obtain the effective image. The effective image can be obtained by performing an inflation operation on the mask image and then performing a corrosion operation. The gradient matrix of the effective image can be calculated using a gradient operator such as the Sobel operator. The gradient amplitude matrix can be obtained by performing polar coordinate conversion on the gradient matrix using a mathematical library (such as NumPy). The binary image can be obtained by performing adaptive threshold segmentation on the gradient amplitude matrix using an adaptive threshold segmentation function in an image processing library. The temperature gradient feature can analyze the distribution of different regions in the binary image, such as the size, shape, and position of the connected regions, etc.

[0139] The thermal stress distribution characteristics of the power adapter can be analyzed by using the temperature gradient feature, which can help understand the thermal mechanical performance of the power adapter and the stability of the power adapter during operation.

[0140] The thermal stress distribution characteristics refer to the distribution of stress in the entire adapter caused by uneven heating of internal parts, resulting in different temperature distributions during operation of the power adapter.

[0141] Optionally, the thermal stress distribution characteristics can be obtained by establishing a correlation model between the temperature gradient feature and the thermal stress, based on the temperature gradient amplitude, direction, and variation, combined with the material characteristics and structural parameters of the power adapter, to analyze the internal thermal stress size, distribution area, and variation trend.

[0142] Further, the standard frequency domain spectrum can be obtained by performing frequency domain normalization processing on the electromagnetic radiation data, which can avoid interference from abnormal electromagnetic radiation frequency components, and can more clearly understand the frequency characteristics of the electromagnetic radiation of the power adapter.

[0143] The standard frequency domain spectrum refers to the frequency component of the original electromagnetic radiation data after adjustment and standardization, and the interference of abnormal frequency components is eliminated.

[0144] Optionally, the standard frequency domain spectrum can be obtained by converting the electromagnetic radiation data from the time domain to the frequency domain, and then adjusting and scaling the amplitude of the frequency domain data based on the statistical characteristics of the data or a specific normalization algorithm, so that it has a uniform scale and range.

[0145] S4, performing three-dimensional topological analysis on the abnormal circuit area to obtain a fault morphology feature vector, performing multi-source data fusion on the thermal stress distribution characteristics, the standard frequency domain spectrum and the fault morphology feature vector to obtain a fusion diagnosis data set, and performing health state analysis on the power adapter by using the fusion diagnosis data set to obtain a device health analysis report.

[0146] The embodiment of the present application can describe the structure and morphology of the abnormal circuit area in detail from a spatial angle by performing three-dimensional topological analysis on the abnormal circuit area to obtain a fault morphology feature vector, and can provide accurate information about the fault physical morphology for fault diagnosis of the abnormal circuit area.

[0147] The fault morphology feature vector refers to a set of numerical features obtained by performing three-dimensional topological analysis on the abnormal circuit area and used for representing the fault morphology of the area.

[0148] As an embodiment of the present application, the three-dimensional topological analysis on the abnormal circuit area to obtain a fault morphology feature vector includes: performing three-dimensional finite element modeling on the abnormal circuit area to obtain a visualized circuit area, identifying current density distribution data, thermal stress data and electromagnetic field intensity distribution data in the visualized circuit area, and performing vectorization processing on the current density distribution data, the thermal stress data and the electromagnetic field intensity distribution data to obtain a fault morphology feature vector.

[0149] The visualized circuit area refers to a three-dimensional graph showing the structure, layout and spatial relationship of each part of the circuit, the current density distribution data refers to a set of data describing the current distribution in the visualized circuit area, the thermal stress data refers to data reflecting the stress distribution in the visualized circuit area due to temperature changes, and the electromagnetic field intensity distribution data refers to data representing the spatial distribution of electromagnetic field intensity in the visualized circuit area.

[0150] Optionally, the three-dimensional finite element modeling on the abnormal circuit area to obtain a visualized circuit area can be realized by using professional finite element modeling software such as ANSYS. Further, the embodiment of the present application can make full use of the advantages of various data, complement and verify each other, and improve the accuracy and reliability of fault diagnosis by performing multi-source data fusion on the thermal stress distribution characteristics, the standard frequency domain spectrum and the fault morphology feature vector to obtain a fusion diagnosis data set.

[0151] As one of the embodiments of the present application, the multi-source data fusion of the thermal stress distribution characteristics, the standard frequency domain spectrum and the fault mode feature vector to obtain the fusion diagnostic data set comprises: matrix conversion of the thermal stress distribution characteristics, the standard frequency domain spectrum and the fault mode feature vector to obtain a first matrix, a second matrix and a third matrix, tensor combination of the first matrix, the second matrix and the third matrix to obtain a high-order tensor, multi-linear decomposition of the high-order tensor to obtain a low-rank matrix, matrix solving of the low-rank matrix to obtain matrix elements, element fusion of the matrix elements to obtain the fusion diagnostic data set.

[0152] The first matrix refers to the matrix obtained by matrix conversion of the thermal stress distribution characteristics, the second matrix refers to the matrix obtained by matrix conversion of the standard frequency domain spectrum, the third matrix refers to the matrix converted from the fault mode feature vector, and the low-rank matrix refers to the matrix obtained by multi-linear decomposition of the high-order tensor.

[0153] Optionally, the matrix conversion of the thermal stress distribution characteristics, the standard frequency domain spectrum and the fault mode feature vector can be realized by using the NumPy library of Python. The high-order tensor can convert the first matrix, the second matrix and the third matrix into a dimension slice representation of a tensor, and then stack and combine the dimension slices in a specific order, such as the size of the tensor in the new dimension, to obtain. The low-rank matrix can be obtained by Tucker decomposition of the high-order tensor. The matrix solving of the low-rank matrix to obtain the matrix elements can be realized by using the NumPy library. The fusion diagnostic data set can perform weighted average processing on all elements obtained from the low-rank matrix solving, and arrange the element values after the weighted average processing in order to obtain a data set.

[0154] The embodiment of the present application can evaluate and judge the overall health status of the power adapter and provide a detailed report on the health status of the power adapter for the user by using the fusion diagnostic data set to analyze the health status of the power adapter and obtain a device health analysis report.

[0155] Optionally, the process of using the fusion diagnostic data set to analyze the health status of the power adapter and obtaining a device health analysis report comprises: training and predicting the fusion diagnostic data set by using a machine learning algorithm (such as a support vector machine, a random forest, etc.), evaluating the health status of the power adapter from multiple dimensions such as thermal performance, electromagnetic compatibility and circuit fault according to the prediction result, and constructing a health analysis report of the power adapter by using the result data of the multi-dimensional evaluation.

[0156] Embodiment 2:

[0157] As Figure 2 shown, it is the function module diagram of the power adapter intelligent monitoring system based on Internet of Things of the application.

[0158] The power adapter intelligent monitoring system based on Internet of Things 200 can be installed in an electronic device. According to the functions realized, the power adapter intelligent monitoring system based on Internet of Things can include a target area identification module 201, an abnormal area identification module 202, a multi-dimensional data processing module 203, and a state analysis module 204. The modules of the application can also be referred to as units, which refer to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0159] In the embodiments of the application, the functions of each module / unit are as follows:

[0160] The target area identification module 201 is used to deploy Internet of Things devices at key nodes of a power adapter to collect multi-dimensional running data of the power adapter under running conditions using the Internet of Things devices, calculate the parameter deviation degree of the multi-dimensional running data, extract key parameter points of the multi-dimensional running data based on the parameter deviation degree index, and identify the target monitoring area of the power adapter using the key parameter points.

[0161] The abnormal area identification module 202 is used to perform adaptive spectral decomposition on the multi-dimensional running data based on the target monitoring area to obtain target analysis data, identify multi-dimensional electrical characteristics of the power adapter using the target analysis data, locate abnormal nodes in the target monitoring area using the multi-dimensional electrical characteristics, and obtain an abnormal circuit area.

[0162] The multi-dimensional data processing module 203 is used to collect infrared thermal imaging data and electromagnetic radiation data of the power adapter, perform non-uniformity correction on the infrared thermal imaging data to obtain a calibrated thermal distribution map, identify temperature gradient characteristics of the calibrated thermal distribution map, analyze the thermal stress distribution characteristics of the power adapter using the temperature gradient characteristics, perform frequency domain normalization processing on the electromagnetic radiation data to obtain a standard frequency domain spectrum.

[0163] The state analysis module 204 is used to perform three-dimensional topological analysis on the abnormal circuit area to obtain a fault morphology feature vector, perform multi-source data fusion on the thermal stress distribution characteristics, the standard frequency domain spectrum, and the fault morphology feature vector to obtain a fusion diagnostic data set, perform health state analysis on the power adapter using the fusion diagnostic data set, and obtain a device health analysis report.

[0164] In detail, the modules in the power adapter intelligent monitoring system 200 based on the Internet of Things in the embodiments of the present application adopt the same technical means as the power adapter intelligent monitoring method based on the Internet of Things in the above-mentioned Figure 1 and can produce the same technical effects, which will not be described here.

[0165] It is obvious for those skilled in the art that the present application is not limited to the details of the above-mentioned exemplary embodiments, and the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application.

[0166] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. A power adapter intelligent monitoring method based on Internet of Things, characterized in that, The method comprises: Deploying an Internet of Things device at a key node of a power adapter to collect multi-dimensional operation data of the power adapter under operation conditions by using the Internet of Things device, calculating a parameter deviation degree of the multi-dimensional operation data, extracting a key parameter point of the multi-dimensional operation data based on the parameter deviation degree index, and identifying a target monitoring area of the power adapter by using the key parameter point, wherein the identifying the target monitoring area of the power adapter by using the key parameter point comprises: Mapping the key parameter point to a three-dimensional coordinate system to obtain a standardized coordinate data set; Performing a time synchronization interpolation operation on the standardized coordinate data set to obtain a space-time alignment data set; Performing data clustering on the space-time alignment data set to obtain an initial monitoring cluster set; Performing space-time correlation verification on the initial monitoring cluster set to extract an effective data set in the initial monitoring cluster set by using a verification result of the space-time correlation verification; Performing dynamic risk analysis on the power adapter based on the effective data set to identify a priority monitoring area of the power adapter; Performing area boundary processing on the priority monitoring area to obtain a target monitoring area; Performing self-adaptive spectral decomposition on the multi-dimensional operation data based on the target monitoring area to obtain target analysis data, identifying multi-dimensional electrical characteristics of the power adapter by using the target analysis data, positioning an abnormal node in the target monitoring area by using the multi-dimensional electrical characteristics, and obtaining an abnormal circuit area; Collecting infrared thermal imaging data and electromagnetic radiation data of the power adapter, performing non-uniformity correction on the infrared thermal imaging data to obtain a calibrated thermal distribution map, identifying temperature gradient characteristics of the calibrated thermal distribution map, analyzing thermal stress distribution characteristics of the power adapter by using the temperature gradient characteristics, performing frequency domain normalization processing on the electromagnetic radiation data to obtain a standard frequency domain spectrum; Performing three-dimensional topology analysis on the abnormal circuit area to obtain a fault morphology feature vector, performing multi-source data fusion on the thermal stress distribution characteristics, the standard frequency domain spectrum, and the fault morphology feature vector to obtain a fusion diagnosis data set, and performing health state analysis on the power adapter by using the fusion diagnosis data set to obtain a device health analysis report, wherein the performing multi-source data fusion on the thermal stress distribution characteristics, the standard frequency domain spectrum, and the fault morphology feature vector to obtain the fusion diagnosis data set comprises: Performing matrix conversion on the thermal stress distribution characteristics, the standard frequency domain spectrum, and the fault morphology feature vector to obtain a first matrix, a second matrix, and a third matrix; Performing tensor combination on the first matrix, the second matrix, and the third matrix to obtain a high-order tensor; Performing multi-linear decomposition on the high-order tensor to obtain a low-rank matrix; Solving the low-rank matrix to obtain a matrix element; Performing element fusion on the matrix element to obtain the fusion diagnosis data set.

2. The IoT-based intelligent monitoring method of a power adapter according to claim 1, wherein, The calculating the parameter deviation degree of the multi-dimensional operation data comprises: Identifying a parameter measurement type of the multi-dimensional operation data; According to the parameter measurement type, the multi-dimensional operation data is classified to obtain classified data; Inquiry of the measured value of the classified data and the observation time corresponding to the measured value; Based on the measured value, the observation value deviation degree of the classified data in the observation time is calculated by using the following formula: ； wherein, represents an observation value deviation degree, t represents an observation time, represents a measured value of the i-th classification data at the time t, represents a corresponding reference value, represents a deviation weight coefficient of the i-th classification data; According to the observation value deviation degree, the parameter deviation degree of the multi-dimensional operation data is determined.

3. The IoT-based intelligent monitoring method of a power adapter according to claim 1, wherein, The adaptive spectral decomposition of the multi-dimensional operation data is carried out based on the target monitoring area to obtain target analysis data, including: Identifying the area characteristics of the target monitoring area; Based on the area characteristics, the multi-dimensional operation data is screened to obtain screened data; The signal Shannon entropy of the screened data is calculated; Based on the signal Shannon entropy, the screened data is decomposed into a time-frequency matrix; The time-frequency matrix is subjected to time-frequency sub-generation optimization to obtain target analysis data.

4. The IoT-based intelligent monitoring method of a power adapter according to claim 1, wherein, The abnormal node in the target monitoring area is located by using the multi-dimensional electrical characteristics to obtain an abnormal circuit area, including: Feature screening is performed on the multi-dimensional electrical characteristics to obtain screened features; Abnormal detection is performed on the target monitoring area by using the screened features to obtain abnormal detection data; The abnormal detection data is subjected to matrix conversion to obtain an abnormal detection matrix; The abnormal range of the target monitoring area is identified by using the abnormal detection matrix to obtain a preliminary identified area; The abnormal similarity of the electrical data in the preliminary identified area is calculated; Based on the abnormal similarity, the target monitoring area is segmented to obtain an abnormal segmentation area; The area center point of the abnormal segmentation area is identified; The abnormal circuit area is located by using the area center point.

5. The IoT-based intelligent monitoring method of a power adapter according to claim 1, wherein, The non-uniformity correction of the infrared thermal imaging data is performed to obtain a calibrated thermal distribution map, including: Bad point detection is performed on the infrared thermal imaging data to obtain bad point coordinates; The bad point coordinates are subjected to matrix conversion to obtain a bad point coordinate matrix; Neighborhood interpolation is performed on the bad point coordinate matrix to obtain an interpolated image; The gain matrix of the interpolated image is calculated by using the following formula: ； wherein denotes a gain matrix, denotes a pixel point of the interpolated image original gray value at high calibration temperature, denotes a pixel point of the interpolated image original gray value at low calibration temperature, denotes a high calibration temperature, denotes a low calibration temperature; Based on the gain matrix, the offset matrix of the interpolated image is calculated by using the following formula: ； wherein, denotes a shift matrix, denotes a pixel point of the interpolated image original gray value at low calibration temperature, denotes a low calibration temperature; Based on the gain matrix and the offset matrix, linear correction is performed on the infrared thermal imaging data to obtain a linear correction image; Scene adaptive correction is performed on the linear correction image to obtain a calibrated thermal distribution map.

6. The IoT-based intelligent monitoring method of a power adapter according to claim 1, wherein, The temperature gradient characteristics of the calibrated thermal distribution map are identified, including: Effective area mask operation is performed on the thermal distribution map to obtain a mask image; Morphological closing operation processing is performed on the mask image to obtain an effective image; The gradient matrix of the effective image is calculated; Polar coordinate conversion is performed on the gradient matrix to obtain a gradient amplitude matrix; Adaptive threshold segmentation is performed on the gradient amplitude matrix to obtain a binary image; The temperature gradient characteristics of the calibrated thermal distribution map are identified by using the binary image.

7. The IoT-based intelligent monitoring method of a power adapter according to claim 1, wherein, The three-dimensional topological analysis of the abnormal circuit area is performed to obtain a fault morphology feature vector, including: The abnormal circuit area is modeled by three-dimensional finite elements to obtain a visualized circuit area; Current density distribution data, thermal stress data and electromagnetic field intensity distribution data in the visualized circuit area are identified; The current density distribution data, the thermal stress data and the electromagnetic field intensity distribution data are vectorized to obtain a fault mode feature vector.

8. The power adapter intelligent monitoring system based on the Internet of Things, characterized in that, The system comprises: A target area identification module is configured to deploy an Internet of Things device at a key node of a power adapter, collect multi-dimensional operation data of the power adapter under operation conditions by using the Internet of Things device, calculate a parameter deviation degree of the multi-dimensional operation data, extract a key parameter point of the multi-dimensional operation data based on the parameter deviation degree, and identify a target monitoring area of the power adapter by using the key parameter point. The identification of the target monitoring area of the power adapter by using the key parameter point comprises: mapping the key parameter point to a three-dimensional coordinate system to obtain a standardized coordinate data set; performing a time synchronization interpolation operation on the standardized coordinate data set to obtain a space-time alignment data set; performing data clustering on the space-time alignment data set to obtain an initial monitoring cluster set; performing a space-time correlation verification on the initial monitoring cluster set to extract an effective data set in the initial monitoring cluster set by using a verification result of the space-time correlation verification; performing a dynamic risk analysis on the power adapter based on the effective data set to identify a priority monitoring area of the power adapter; performing a region boundary processing on the priority monitoring area to obtain the target monitoring area; An abnormal area identification module is configured to perform an adaptive frequency spectrum decomposition on the multi-dimensional operation data based on the target monitoring area to obtain target analysis data, identify multi-dimensional electrical characteristics of the power adapter by using the target analysis data, locate an abnormal node in the target monitoring area by using the multi-dimensional electrical characteristics, and obtain an abnormal circuit area. A multi-dimensional data processing module is configured to collect infrared thermal imaging data and electromagnetic radiation data of the power adapter, perform a non-uniformity correction on the infrared thermal imaging data to obtain a calibrated thermal distribution map, identify a temperature gradient feature of the calibrated thermal distribution map, analyze thermal stress distribution characteristics of the power adapter by using the temperature gradient feature, perform a frequency domain normalization processing on the electromagnetic radiation data to obtain a standard frequency domain spectrum. A state analysis module is configured to perform a three-dimensional topology analysis on the abnormal circuit area to obtain a fault mode feature vector, perform a multi-source data fusion on the thermal stress distribution characteristics, the standard frequency domain spectrum and the fault mode feature vector to obtain a fusion diagnosis data set, and perform a health state analysis on the power adapter by using the fusion diagnosis data set to obtain a device health analysis report. The multi-source data fusion on the thermal stress distribution characteristics, the standard frequency domain spectrum and the fault mode feature vector to obtain the fusion diagnosis data set comprises: performing a matrix conversion on the thermal stress distribution characteristics, the standard frequency domain spectrum and the fault mode feature vector to obtain a first matrix, a second matrix and a third matrix. tensor combination is performed on the first matrix, the second matrix and the third matrix to obtain a high-order tensor; multilinear decomposition is performed on the high-order tensor to obtain a low-rank matrix; matrix solving is performed on the low-rank matrix to obtain a matrix element; element fusion is performed on the matrix element to obtain a fused diagnosis data set.

Images