Electrical equipment thermal fault monitoring and early warning system fused with infrared image processing

Abstract

The invention relates to the technical field of power failure prediction, and discloses a power equipment thermal failure monitoring and early warning system fused with infrared image processing. According to the invention, the infrared image acquisition module acquires and standardizes equipment temperature field data, and the electrical parameter synchronous acquisition module acquires load current and environment temperature in real time; the hot spot dynamic analysis module identifies and tracks an overheat area based on a dynamic threshold value, extracts spatio-temporal evolution characteristics of the overheat area, and comprehensively calculates a fault hot spot intensity factor capable of representing an early fault; the electrical fault diagnosis module starts deep diagnosis when the factor is abnormal, and accurately estimates the contact resistance value of the key part through a nonlinear recursive algorithm; and the comprehensive early warning module performs fusion evaluation on the multi-source information, and generates multi-level early warning information based on a layering strategy. According to the method, post-event alarm is converted into beforehand intervention, the sudden failure risk of the equipment is remarkably reduced, and accurate and reliable decision support is provided for predictive maintenance.

Description

Technical Field

[0001] This invention relates to the field of power fault prediction technology, and in particular to a power equipment thermal fault monitoring and early warning system that integrates infrared image processing. Background Technology

[0002] Power equipment is a core component of the power grid, and its operating status is directly related to the safety and stability of the entire power system. Statistics show that most electrical equipment faults undergo a process of increased internal or contact resistance leading to heating and abnormal temperature rise before eventually evolving into accidents. Therefore, real-time and accurate monitoring of the temperature of power equipment is a key technical means to prevent major faults and achieve condition-based maintenance.

[0003] In the field of thermal fault monitoring of power equipment, current technologies mainly rely on setting absolute temperature thresholds for alarms. This method is essentially a reactive alarm and cannot effectively address potential and developing faults such as poor contact and insulation degradation. These faults initially manifest as abnormal heating trends or minor localized temperature rises, with their absolute temperature values ​​often remaining within safe thresholds. By the time the equipment temperature eventually exceeds the fixed threshold and triggers an alarm, the fault has usually developed into a serious defect, missing the optimal early intervention opportunity and resulting in passive and costly maintenance. Therefore, achieving accurate detection of faults in their nascent stage is a core technical problem that urgently needs to be solved in this field. Summary of the Invention

[0004] The technical problem to be solved by this invention is that the existing technology relies on setting an absolute temperature threshold for alarm, which cannot provide early warning and leads to equipment damage. To address this, we propose a power equipment thermal fault monitoring and early warning system that integrates infrared image processing.

[0005] To achieve the above objectives, this application adopts the following technical solution: a power equipment thermal fault monitoring and early warning system integrating infrared image processing, comprising:

[0006] The infrared image acquisition module is used to acquire infrared image sequences of power equipment through an infrared thermal imager, perform non-uniformity correction, noise filtering and temperature field calibration on the original infrared images, generate standardized temperature field matrix data, and store the processed time-series temperature field data in the image cache library.

[0007] The electrical parameter synchronous acquisition module is used to collect the operating electrical parameters of power equipment in real time through the power monitoring system. The operating electrical parameters include load current and ambient temperature. The acquired electrical data is cleaned and timestamped to ensure time synchronization with infrared image data and stored in the electrical parameter database.

[0008] The hot spot dynamic analysis module is used to call time-series temperature field data from the image cache library, identify and extract the overheated area on the surface of the power equipment through the image segmentation algorithm, track the spatiotemporal evolution characteristics of the overheated area in real time, and calculate the fault hot spot intensity factor based on the spatiotemporal evolution characteristics.

[0009] The electrical fault diagnosis module is used to call the corresponding operating electrical parameters from the electrical parameter database. When the fault hot spot intensity factor indicates an abnormality, the deep diagnosis process is initiated, and the contact resistance value of the key connection parts of the power equipment is estimated in real time through a nonlinear recursive estimation algorithm.

[0010] The integrated early warning module is used to receive the estimated values ​​of fault hot spot intensity factor and contact resistance, and to comprehensively assess the health status of power equipment based on a preset hierarchical early warning strategy, generating multi-level early warning information.

[0011] The system management module is used to monitor the operating status of various modules within the system and manage user permissions and system configuration.

[0012] Preferably, the hot spot dynamic analysis module includes:

[0013] The hot spot recognition and tracking unit is used to automatically segment the overheated area from the temperature field matrix according to the set dynamic temperature threshold and an improved region growing algorithm. It calculates the pixel area of ​​the overheated area in each frame of the image as its instantaneous area, records the highest temperature value and the average temperature value of the area, and establishes a motion trajectory database of the overheated area to track the position change and morphological evolution of the same hot spot in multiple frames of images. The dynamic temperature threshold is adjusted in real time according to the ambient temperature and load current.

[0014] The spatiotemporal feature extraction unit is used to calculate the rate of change of the area of ​​the overheated region over time, obtain the instantaneous growth rate of the hot spot through the discrete differential method, calculate the weighted average of the normal components of the temperature gradient of all pixels on the boundary line of the overheated region to characterize the sharpness of the hot spot boundary, and calculate the expected temperature rise of the power equipment under the current operating conditions in real time through the electrothermal coupling model based on the load current and ambient temperature, and establish the spatiotemporal correlation matrix of the temperature field distribution.

[0015] The fault hot spot intensity factor calculation unit is used to comprehensively calculate the fault hot spot intensity factor by using the instantaneous growth rate of the hot spot, the weighted average sharpness of the hot spot boundary, and the relative ratio of the measured temperature rise to the expected temperature rise in the overheated area through a nonlinear combination algorithm, and then normalizes the calculation results.

[0016] Preferably, the electrical fault diagnosis module includes:

[0017] The diagnostic triggering unit is used to trigger deep electrical diagnostics when the fault hot spot intensity factor continuously exceeds the set dynamic threshold. The dynamic threshold is adaptively adjusted according to the equipment type, years of operation and environmental conditions, and the historical trend of the fault hot spot intensity factor is recorded at the same time.

[0018] The contact resistance recursive estimation unit is used to calculate the apparent thermal resistance observation value through a thermoelectric conversion model based on real-time data of the highest temperature value, ambient temperature and load current. The recursive estimation algorithm with nonlinear correction is used to update the contact resistance estimate value in real time and establish a database of historical changes in contact resistance. The recursive estimation algorithm includes an adaptive gain coefficient and a nonlinear correction exponent.

[0019] The diagnostic report generation unit is used to generate a comprehensive electrical fault diagnostic report based on the contact resistance estimate and its changing trend, combined with the equipment's historical operating data. This report includes the current resistance value, the rate of change relative to the initial value, the assessment of the rate of deterioration, and the prediction of future trends.

[0020] Preferably, the integrated early warning module includes:

[0021] The multi-source information fusion unit is used to perform multi-scale weighted fusion of fault hot spot intensity factor and contact resistance estimate, establish a dynamic allocation mechanism of index weight based on entropy weight method, use fuzzy reasoning method to handle the uncertainty relationship between different parameters, and evaluate the confidence of fusion results.

[0022] The early warning level decision unit is used to determine the early warning level based on the multi-dimensional evaluation results and through the established early warning decision tree model, and to set up an automatic escalation and downgrade mechanism for the early warning level, while taking into account the importance of the equipment and the operating environment factors.

[0023] The early warning information release unit is used to generate formatted early warning information that includes equipment identification, fault type, risk level, recommended measures, and diagnostic data source according to the early warning level, and release it through multiple communication channels to ensure the timeliness and accuracy of the information.

[0024] Preferably, the hot spot recognition and tracking unit is also used for;

[0025] A database of historical movement trajectories of the overheated region was established, and the Kalman filter algorithm was used to predict the next movement trend of the hot spot.

[0026] Calculate the morphological parameters of the overheated region, including the rate of change of the region's perimeter, the rate of change of the shape factor, and the centroid offset distance;

[0027] The frequency, duration, and spatial distribution characteristics of overheated areas were statistically analyzed.

[0028] A spatiotemporal correlation model of hot spots was established to analyze the mutual influence between different hot spots.

[0029] Preferably, the contact resistance recursive estimation unit is also used for:

[0030] Establish a multi-dimensional historical database of contact resistance to record the resistance variation patterns under different operating conditions and environmental conditions;

[0031] Calculate the acceleration and jerk of the change in contact resistance, and predict the trend of resistance change over a future period using cubic spline interpolation.

[0032] Based on equipment type, years of operation, and maintenance history, an adaptive parameter adjustment mechanism is established to dynamically adjust the gain coefficient and nonlinear exponent of the recursive algorithm.

[0033] A data validity verification mechanism is set up to identify and filter out abnormal observations, ensuring the reliability of the estimation results.

[0034] Preferably, the early warning level decision unit is also used for:

[0035] Establish a differentiated early warning threshold system based on equipment importance, and set stricter early warning standards for critical equipment;

[0036] Set up an automatic escalation mechanism for the warning level, which will automatically escalate the warning level when a single parameter continues to deteriorate or multiple parameters are abnormal at the same time.

[0037] Establish an early warning cancellation mechanism so that the early warning is automatically cancelled after a set observation period once the equipment status returns to normal and remains stable.

[0038] Record the early warning decision-making process, including the contribution of each parameter and the basis for the decision, to facilitate subsequent analysis and optimization.

[0039] Preferably, the fault hotspot intensity factor is calculated using the following formula:

[0040] ;

[0041] in: The area change rate of the overheated region is calculated and provided by the hot spot identification and tracking unit. The instantaneous area of ​​the overheated region is calculated by the hot spot identification and tracking unit; The line integral value representing the normal component of the temperature gradient on the boundary line of the overheated region is provided by the spatiotemporal feature extraction unit. Represents the temperature gradient; Represents the boundary unit normal vector; This represents the measured temperature rise, calculated from the highest temperature value and the ambient temperature. The expected temperature rise is calculated by the spatiotemporal feature extraction unit.

[0042] Preferably, the contact resistance estimate is updated using the following recursive formula:

[0043] ;

[0044] in: The estimated contact resistance value at the previous moment is stored and provided by the contact resistance recursive estimation unit; The highest temperature of the overheated area at the current moment is provided by the hot spot identification and tracking unit; The ambient temperature at the current moment is provided by the electrical parameter synchronous acquisition module; The load current at the current moment is provided by the electrical parameter synchronous acquisition module; Indicates the adaptive gain coefficient; This represents a nonlinear correction exponent greater than 1.

[0045] Preferably, the hierarchical early warning strategy executed by the early warning level decision unit is as follows:

[0046] When the fault hot spot intensity factor If the threshold is exceeded for an extended period, a warning of concern will be issued.

[0047] When the fault hot spot intensity factor If the value continuously exceeds the second threshold, or if the increase in the estimated contact resistance exceeds a preset percentage, a warning level alert will be issued.

[0048] When the fault hot spot intensity factor An alarm-level warning will be issued when the third threshold is exceeded, or when the estimated contact resistance reaches the absolute danger threshold, or when its rate of increase per unit time exceeds the deterioration rate threshold.

[0049] The technical effects and advantages of this invention are as follows:

[0050] In this invention, the actual temperature extracted from infrared images is compared in real time with the expected normal temperature calculated based on load current and ambient temperature. This allows for the sensitive detection of abnormal heating caused by potential faults such as increased contact resistance, enabling early warning of progressive faults such as poor contact and insulation degradation. This transforms the operation and maintenance mode from reactive alarm to proactive intervention, significantly reducing the risk of power outages and safety accidents caused by sudden equipment failures. Furthermore, the use of a nonlinear recursive algorithm to accurately estimate the contact resistance, which represents the health status of the equipment, effectively eliminates the interference of environmental and operating condition fluctuations on the monitoring results, greatly reducing the false alarm rate and missed alarm rate of the system. This provides accurate data support for predictive maintenance and improves the level of equipment lifecycle management. Attached Figure Description

[0051] The disclosure of this invention is illustrated with reference to the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. In the drawings, the same reference numerals are used to refer to the same parts:

[0052] Figure 1 This is a schematic diagram of the module unit of the present invention. Detailed Implementation

[0053] It is readily understood that, based on the technical solution of this invention, those skilled in the art can propose various interchangeable structural methods and implementations without altering the essential spirit of the invention. Therefore, the following detailed embodiments and accompanying drawings are merely illustrative examples of the technical solution of this invention and should not be considered as the entirety of the invention or as limitations or restrictions on the technical solution of this invention.

[0054] Reference Figure 1 This invention provides a power equipment thermal fault monitoring and early warning system that integrates infrared image processing. The system mainly includes an infrared image acquisition module, an electrical parameter synchronous acquisition module, a hot spot dynamic analysis module, an electrical fault diagnosis module, a comprehensive early warning module, and a system management module.

[0055] The infrared image acquisition module is responsible for acquiring thermal image information of power equipment. This module continuously or periodically scans the power equipment using a high-precision infrared thermal imager to acquire raw infrared image sequences. To ensure the quality and consistency of the image data, the raw infrared images undergo a series of preprocessing steps, including non-uniformity correction to eliminate errors caused by inconsistent response of the thermal imager detector, noise filtering to remove random noise from the image, and temperature field calibration processing to convert the pixel values ​​of the infrared image into actual temperature values ​​and generate standardized temperature field matrix data. These processed time-series temperature field data are then stored in the image cache library for subsequent modules to access.

[0056] The electrical parameter synchronization acquisition module is responsible for acquiring the operating status parameters of the power equipment. This module interfaces with the power monitoring system to acquire the operating electrical parameters of the power equipment in real time, mainly including load current and ambient temperature. In order to ensure the correspondence between infrared image data and electrical parameter data in time, the acquired electrical data is cleaned to remove outliers or redundant data, and timestamped to ensure that each frame of infrared image can find the electrical parameters that are synchronized with it. These synchronized electrical data are stored in the electrical parameter database.

[0057] The hot spot dynamic analysis module is one of the core components of this system, used to identify, track, and quantify the risk of hot spots from infrared images. This module retrieves time-series temperature field data from the image cache library and uses advanced image segmentation algorithms, such as region growing, edge detection, or deep learning-based segmentation methods, to identify and extract overheated areas, i.e., hot spots, on the surface of power equipment. This module tracks the spatiotemporal evolution characteristics of these overheated areas in real time, including changes in their location, size, shape, and temperature distribution over time. Based on these spatiotemporal evolution characteristics, this module calculates the fault hot spot intensity factor, which comprehensively reflects the activity level and potential danger of the hot spot.

[0058] The hot spot dynamic analysis module includes a hot spot identification and tracking unit, a spatiotemporal feature extraction unit, and a fault hot spot intensity factor calculation unit.

[0059] The hotspot identification and tracking unit automatically segments overheated regions from the temperature field matrix using an improved region growing algorithm based on a set dynamic temperature threshold. The dynamic temperature threshold is adjusted in real-time according to ambient temperature and load current. For example, a functional relationship between a reference temperature and ambient temperature and load current can be established based on historical data or an empirical model. When the local temperature exceeds this reference temperature plus a safety margin, it is considered an overheated region. This unit calculates the pixel area of ​​the overheated region in each frame as its instantaneous area and records the highest and average temperature values ​​of the region. Simultaneously, this unit establishes the operational tracking mechanism for the overheated region. The hotspot tracking and identification unit utilizes a dynamic trajectory database, employing image matching or target tracking algorithms such as Kalman filtering to track the positional changes and morphological evolution of the same hotspot across multiple image frames. In a preferred embodiment, the hotspot tracking and identification unit further includes the following functions: establishing a historical motion trajectory database of the overheated region; using a Kalman filtering algorithm to predict the next movement trend of the hotspot; calculating the morphological parameters of the overheated region, including the rate of change of the region's perimeter, the rate of change of the shape factor, and the centroid offset distance; statistically analyzing the frequency of occurrence, duration, and spatial distribution characteristics of the overheated region; and establishing a spatiotemporal correlation model of the hotspots to analyze the mutual influence relationships between different hotspots. These additional functions make the tracking and analysis of hotspots more comprehensive and in-depth.

[0060] The spatiotemporal feature extraction unit is used to further quantify the dynamic characteristics of hot spots. This unit calculates the rate of change of the area of ​​the overheated region over time and obtains the instantaneous growth rate of the hot spot through the discrete differential method. At the same time, it calculates the weighted average of the normal components of the temperature gradient of all pixels on the boundary line of the overheated region to characterize the sharpness of the hot spot boundary. The higher the sharpness, the higher the heat concentration and the more severe the fault. In addition, based on the load current and ambient temperature, the expected temperature rise of the power equipment under the current operating conditions is calculated in real time through the established electrothermal coupling model, for example, based on finite element analysis or empirical formulas, and a spatiotemporal correlation matrix of the temperature field distribution is established for subsequent comparative analysis.

[0061] The fault hotspot intensity factor calculation unit is the core output of the hotspot dynamic analysis module. This unit uses the instantaneous growth rate of the hotspot, the weighted average sharpness of the hotspot boundary, and the relative ratio of the measured temperature rise to the expected temperature rise in the overheated region to comprehensively calculate the fault hotspot intensity factor through a nonlinear combination algorithm. The measured temperature rise can be obtained by subtracting the ambient temperature from the highest temperature value of the overheated region. Finally, the calculation results are normalized to facilitate comparison and threshold setting under different equipment and operating conditions. In a preferred embodiment, the fault hotspot intensity factor... Calculated using the following formula:

[0062] ;

[0063] in:

[0064] The area change rate of the overheated region is calculated and provided by the hot spot identification and tracking unit.

[0065] The instantaneous area of ​​the overheated region is calculated by the hot spot identification and tracking unit;

[0066] The line integral value representing the normal component of the temperature gradient on the boundary line of the overheated region is provided by the spatiotemporal feature extraction unit.

[0067] Represents the temperature gradient;

[0068] Represents the boundary unit normal vector;

[0069] This represents the measured temperature rise, calculated from the highest temperature value and the ambient temperature.

[0070] The expected temperature rise is calculated by the spatiotemporal feature extraction unit.

[0071] This formula comprehensively considers the rate of change of hot spot area, boundary temperature gradient, and the degree of deviation between actual temperature rise and normal temperature rise, and can comprehensively and accurately quantify the danger of hot spot.

[0072] The electrical fault diagnosis module is used to deeply diagnose the electrical nature of the fault when the hot spot intensity factor indication is abnormal. This module calls the corresponding operating electrical parameters from the electrical parameter database. When the fault hot spot intensity factor indication is abnormal, for example, when it continuously exceeds the set dynamic threshold, the module starts the deep diagnosis process and estimates the contact resistance value of the key connection parts of the power equipment in real time through a nonlinear recursive estimation algorithm. The abnormal increase in contact resistance is one of the root causes of local overheating.

[0073] Specifically, the electrical fault diagnosis module includes a diagnosis triggering unit, a contact resistance recursive estimation unit, and a diagnosis report generation unit.

[0074] The diagnostic trigger unit is used to trigger in-depth electrical diagnostics when the fault hot spot intensity factor continuously exceeds a set dynamic threshold. The dynamic threshold is adaptively adjusted according to the equipment type, years of operation, and environmental conditions to adapt to the characteristics and operating environment of different equipment. At the same time, the unit records the historical trend of the fault hot spot intensity factor to provide a reference for subsequent diagnostics.

[0075] The contact resistance recursive estimation unit, based on the highest temperature value provided by the hot spot identification and tracking unit, and real-time data of ambient temperature and load current, calculates the observed apparent thermal resistance value using an established thermoelectric conversion model, such as based on Joule's law and the thermal balance equation. Then, a recursive estimation algorithm with nonlinear correction, such as extended Kalman filtering, unscented Kalman filtering, or particle filtering, is used to update the contact resistance estimate in real time. This recursive estimation algorithm includes an adaptive gain coefficient and a nonlinear correction exponent to improve the accuracy and robustness of the estimation and adapt to the nonlinear variation characteristics of the contact resistance. Simultaneously, this unit establishes a historical database of contact resistance changes, recording its variation over time. In a preferred embodiment, the contact resistance estimate is updated using the following recursive formula:

[0076] ;

[0077] in:

[0078] The estimated contact resistance value at the previous moment is stored and provided by the contact resistance recursive estimation unit;

[0079] The highest temperature of the overheated area at the current moment is provided by the hot spot identification and tracking unit;

[0080] The ambient temperature at the current moment is provided by the electrical parameter synchronous acquisition module;

[0081] The load current at the current moment is provided by the electrical parameter synchronous acquisition module;

[0082] Indicates the adaptive gain coefficient;

[0083] This represents a nonlinear correction exponent greater than 1;

[0084] This formula compares the ratio of the observed temperature rise to the square of the current (i.e., the apparent thermal resistance) with the estimated contact resistance value from the previous moment, and introduces an adaptive gain coefficient and a nonlinear correction exponent to achieve a real-time, nonlinear recursive estimate of the contact resistance.

[0085] In another preferred embodiment, the contact resistance recursive estimation unit further includes the following functions: establishing a multi-dimensional historical database of contact resistance to record the resistance change patterns under different operating conditions and environmental conditions; calculating the acceleration and jerk of contact resistance change, and predicting the resistance change trend over a future period using cubic spline interpolation; establishing a parameter adaptive adjustment mechanism based on equipment type, operating years, and maintenance history to dynamically adjust the gain coefficient and nonlinear exponent of the recursive algorithm; and setting a data validity verification mechanism to identify and filter abnormal observations to ensure the reliability of the estimation results. These functions further enhance the accuracy and predictive ability of contact resistance estimation.

[0086] The diagnostic report generation unit is used to generate a comprehensive electrical fault diagnostic report based on the contact resistance estimate and its changing trend, combined with the equipment's historical operating data. This report includes the current resistance value, the rate of change relative to the initial value, an assessment of the rate of deterioration, and a prediction of future trends.

[0087] The integrated early warning module is used to comprehensively assess the health status of power equipment and generate early warning information. This module receives the estimated values ​​of fault hot spot intensity factor and contact resistance, and performs a comprehensive assessment of the health status of power equipment based on a preset hierarchical early warning strategy to generate multi-level early warning information.

[0088] Specifically, the integrated early warning module includes a multi-source information fusion unit, an early warning level decision-making unit, and an early warning information dissemination unit.

[0089] The multi-source information fusion unit is used to perform multi-scale weighted fusion of the fault hot spot intensity factor and the contact resistance estimate. This unit establishes a dynamic weight allocation mechanism for indicators based on the entropy weight method, and dynamically adjusts the weight of each indicator according to its information content and importance. Fuzzy reasoning is used to handle the uncertainty relationship between different parameters. For example, when the hot spot intensity factor and the contact resistance value are both at a medium level, fuzzy reasoning can determine their comprehensive risk. The unit also performs a confidence assessment on the fusion results to reflect the reliability of the assessment results.

[0090] The early warning level decision unit determines the early warning level based on multi-dimensional evaluation results using an established early warning decision tree model. This unit sets up an automatic escalation and de-escalation mechanism for the early warning level; for example, it automatically escalates the early warning level when the fault continues to worsen and automatically de-escalates it when the equipment status returns to normal. Simultaneously, this unit considers equipment importance and operating environment factors, setting stricter early warning standards for equipment of different importance or equipment operating in harsh environments. In a preferred embodiment, the early warning level decision unit also includes the following functions: establishing a differentiated early warning threshold system based on equipment importance, setting stricter early warning standards for critical equipment; setting an automatic escalation mechanism for the early warning level, automatically escalating the early warning level when a single parameter continues to worsen or multiple parameters are simultaneously abnormal; establishing an early warning cancellation mechanism, automatically canceling the early warning after a set observation period once the equipment status returns to normal and remains stable; and recording the early warning decision process, including the contribution of each parameter and the decision basis, for subsequent analysis and optimization. These functions make early warning decision-making more intelligent, flexible, and traceable.

[0091] In another preferred embodiment, the hierarchical early warning strategy executed by the early warning level decision unit is: when the fault hotspot intensity factor When the intensity factor of the fault hotspot continuously exceeds the first threshold, a warning of concern level is issued; when the intensity factor of the fault hotspot exceeds the first threshold, a warning of concern level is issued. A warning level alert is issued when the value of the fault hotspot intensity factor continuously exceeds the second threshold or the increase in the estimated contact resistance exceeds a preset percentage; An alarm-level warning is issued when the third threshold is exceeded, or the estimated contact resistance reaches the absolute danger threshold, or its rate of increase per unit time exceeds the degradation rate threshold. This tiered warning strategy can provide different levels of warning information based on the severity and development trend of the fault, making it easier for maintenance personnel to take differentiated response measures.

[0092] The early warning information release unit is used to generate formatted early warning information that includes equipment identification, fault type, risk level, recommended measures, and diagnostic data source according to the early warning level. This information is released through various communication channels, such as SMS, email, APP push, and audible and visual alarms, to ensure the timeliness and accuracy of the information so that maintenance personnel can respond quickly.

[0093] The system management module is used to monitor the operating status of each module within the system to ensure the stable operation of the entire system. This module is also responsible for managing user permissions, assigning corresponding operation and viewing permissions according to different user roles, and managing system configurations, including parameter settings, threshold adjustments, data storage strategies, etc., to adapt to different application scenarios and needs.

[0094] Through the coordinated operation of the above modules, this invention achieves comprehensive and intelligent monitoring and early warning of thermal faults in power equipment. From the refined processing of infrared images to the dynamic tracking of hot spots, from the synchronous acquisition of electrical parameters to the in-depth diagnosis of contact resistance, and then to the fusion and early warning of multi-source information, the entire system forms an intelligent monitoring and early warning system, which significantly improves the safety, reliability and maintenance efficiency of power equipment operation.

[0095] The technical scope of this invention is not limited to the content described above. Those skilled in the art can make various modifications and variations to the above embodiments without departing from the technical concept of this invention, and all such modifications and variations should fall within the protection scope of this invention.

Claims

1. A power equipment thermal fault monitoring and early warning system fusing infrared image processing, characterized in that, The application relates to a power equipment fault diagnosis system based on infrared image and electrical parameter, which comprises the following modules: An infrared image acquisition module is used for acquiring an infrared image sequence of power equipment through an infrared thermal imager, performing non-uniformity correction, noise filtering and temperature field calibration on original infrared images, generating standardized temperature field matrix data, and storing the processed time-series temperature field data into an image cache library; An electrical parameter synchronous acquisition module is used for acquiring running electrical parameters of the power equipment through a power monitoring system in real time, wherein the running electrical parameters comprise a load current and an ambient temperature, the acquired electrical data is cleaned and time-stamped, time synchronization with infrared image data is ensured, and the data is stored into an electrical parameter database; A hot spot dynamic analysis module is used for calling time-series temperature field data from the image cache library, identifying and extracting a superheating area on the surface of the power equipment through an image segmentation algorithm, tracking time-space evolution characteristics of the superheating area in real time, and calculating a fault hot spot intensity factor based on the time-space evolution characteristics; An electrical fault diagnosis module is used for calling corresponding running electrical parameters from the electrical parameter database, starting a deep diagnosis process when the fault hot spot intensity factor indicates that there is an anomaly, and estimating a contact resistance value of a key connection part of the power equipment in real time through a nonlinear recursive estimation algorithm; A comprehensive early warning module is used for receiving the fault hot spot intensity factor and the estimated value of the contact resistance, comprehensively evaluating the health state of the power equipment based on a preset hierarchical early warning strategy, and generating multi-level early warning information; A system management module is used for monitoring the running state of each module in the system, managing user permissions and system configuration.

2. The power equipment thermal fault monitoring and early warning system integrated with infrared image processing according to claim 1, characterized in that, The hot spot dynamic analysis module comprises: A hot spot identification and tracking unit is used for automatically segmenting a superheating area from a temperature field matrix according to a set dynamic temperature threshold value, calculating the pixel area of the superheating area in each frame of image as the instantaneous area, recording the maximum temperature value and the average temperature value of the area, and establishing a motion trajectory database of the superheating area to track the position change and morphological evolution of the same hot spot in multiple frames of image, wherein the dynamic temperature threshold value is adjusted in real time according to the ambient temperature and the load current; A time-space feature extraction unit is used for calculating the change rate of the area of the superheating area with time, obtaining the instantaneous growth speed of the hot spot through a discrete differential method, calculating the weighted average value of the normal component of the temperature gradient of all pixel points on the boundary line of the superheating area to represent the sharpness of the hot spot boundary, calculating the expected temperature rise of the power equipment under the current running condition through an electro-thermal coupling model based on the load current and the ambient temperature, and establishing a time-space correlation matrix of temperature field distribution; A fault hot spot intensity factor calculation unit is used for comprehensively calculating the fault hot spot intensity factor through a nonlinear combination algorithm by using the instantaneous growth speed of the hot spot, the weighted average sharpness of the hot spot boundary and the relative ratio of the measured temperature rise of the superheating area to the expected temperature rise, and performing normalization processing on the calculation result.

3. The power equipment thermal fault monitoring and early warning system integrated with infrared image processing according to claim 1, characterized in that, The electrical fault diagnosis module comprises: a diagnostic triggering unit configured to trigger a deep electrical diagnosis when the fault hot spot intensity factor continues to exceed a set dynamic threshold, the dynamic threshold being self-adaptively adjusted according to the device type, the operation age, and the environmental conditions, while recording the historical change trend of the fault hot spot intensity factor; a contact resistance recursive estimation unit configured to calculate an apparent thermal resistance observation value through a thermoelectric conversion model based on real-time data of the highest temperature value, the ambient temperature, and the load current, to update the contact resistance estimation value in real time using a recursive estimation algorithm with a non-linear correction, and to establish a historical change database of the contact resistance, wherein the recursive estimation algorithm includes an adaptive gain coefficient and a non-linear correction index; a diagnostic report generation unit configured to generate a comprehensive electrical fault diagnosis report including the current resistance value, the change rate relative to the initial value, the degradation speed evaluation, and the future trend prediction based on the contact resistance estimation value and its change trend, in combination with the device operation history data.

4. The power equipment thermal fault monitoring and early warning system integrated with infrared image processing according to claim 3, characterized in that, The comprehensive early warning module includes: a multi-source information fusion unit configured to perform multi-scale weighted fusion on the fault hot spot intensity factor and the contact resistance estimation value, to establish an index weight dynamic distribution mechanism based on an entropy weight method, to process the uncertainty relationship between different parameters using a fuzzy reasoning method, and to perform confidence evaluation on the fusion result; a warning level decision unit configured to determine the warning level based on the multi-dimensional evaluation result through an established early warning decision tree model, to set an automatic escalation and de-escalation mechanism of the warning level, and to consider the device importance and the operation environment factors; a warning information publishing unit configured to generate a formatted warning information including the device identification, the fault type, the risk level, the recommended measures, and the diagnostic data source according to the warning level, and to publish the information through multiple communication channels to ensure the timeliness and accuracy of the information.

5. The power equipment thermal fault monitoring and early warning system integrated with infrared image processing according to claim 4, characterized in that, The hot spot recognition and tracking unit is further configured to: establish a historical motion trajectory database of the overheated area, and predict the next moving trend of the hot spot using a Kalman filtering algorithm; calculate the morphological parameters of the overheated area, including the area perimeter change rate, the shape factor change rate, and the centroid offset distance; statistically analyze the appearance frequency, the duration, and the spatial distribution characteristics of the overheated area; establish a space-time correlation model of the hot spot, and analyze the mutual influence relationship between different hot spots.

6. The power equipment thermal fault monitoring and early warning system integrated with infrared image processing according to claim 5, characterized in that, The contact resistance recursive estimation unit is further configured to: establish a multi-dimensional historical database of the contact resistance, and record the resistance value change law under different operation conditions and different environmental conditions; calculate the change acceleration and the jerk of the contact resistance, and predict the resistance value change trend in a future period of time through a cubic spline interpolation method; establish a parameter self-adaptive adjustment mechanism according to the device type, the operation age, and the maintenance history, and dynamically adjust the gain coefficient and the non-linear index of the recursive algorithm; set a data validity verification mechanism to identify and filter abnormal observation values, and ensure the reliability of the estimation result.

7. The power equipment thermal fault monitoring and early warning system integrated with infrared image processing according to claim 6, characterized in that, The warning level decision unit is further configured to: establish a differentiated early warning threshold system based on the device importance, and set a more stringent early warning standard for critical devices; set an automatic escalation mechanism of the warning level, and automatically upgrade the warning level when a single parameter continues to deteriorate or multiple parameters are abnormal at the same time; An early warning cancellation mechanism is established, and the early warning is automatically cancelled after the device state is restored to normal and remains stable for a set observation period; The early warning decision process is recorded, including the contribution of various parameters and the decision basis, to facilitate subsequent analysis and optimization.

8. The power equipment thermal fault monitoring and early warning system integrated with infrared image processing according to claim 7, characterized in that, The fault hot spot intensity factor is calculated by the following formula: ； wherein: represents the area change rate of the overheating region, calculated by the hot spot identification and tracking unit; represents the instantaneous area of the overheating region, calculated by the hot spot identification and tracking unit; represents the line integral value of the normal component of the temperature gradient on the boundary line of the overheating region, calculated by the spatio-temporal feature extraction unit; represents the temperature gradient; represents the boundary unit normal vector; represents the measured temperature rise, calculated from the highest temperature value and the ambient temperature; represents the expected temperature rise, calculated by the spatio-temporal feature extraction unit.

9. The power equipment thermal fault monitoring and early warning system integrated with infrared image processing according to claim 8, characterized in that, The contact resistance estimation value is updated by the following recursive formula: ； wherein: represents the contact resistance estimation value at the previous time, stored and provided by the contact resistance recursive estimation unit; represents the highest temperature of the overheating area at the current time, provided by the hot spot identification and tracking unit; represents the ambient temperature at the current time, provided by the electrical parameter synchronous acquisition module; represents the load current at the current time, provided by the electrical parameter synchronous acquisition module; represents the adaptive gain coefficient; represents the nonlinear correction index greater than 1.

10. The power equipment thermal fault monitoring and early warning system integrated with infrared image processing according to claim 9, characterized in that, The hierarchical early warning strategy executed by the early warning level decision unit is: when the fault hot spot intensity factor releasing an attention level warning when the fault hot spot intensity factor continues to exceed the first threshold value; when the fault hot spot intensity factor when the contact resistance estimate value exceeds a second threshold value, or when the growth rate of the contact resistance estimate value exceeds a preset percentage, an early warning of a warning level is issued; when the fault hot spot intensity factor an alarm level warning is issued when the fault hot spot intensity factor exceeds a third threshold, or the contact resistance estimate reaches an absolute danger threshold, or the amount of growth per unit time exceeds a deterioration speed threshold.

Images