Power transmission and distribution line iron accessory temperature abnormity control method and system

Abstract

The invention discloses a power transmission and distribution line iron accessory temperature abnormity control method and system, and relates to the technical field of power component control, and the method comprises the steps: obtaining an infrared thermal image of a target fitting; performing hot spot separation on the infrared thermal image according to a preset separation temperature threshold to obtain a hot spot region; performing boundary sharpness evaluation on the hot spot region to obtain a boundary sharpness evaluation result; performing shape regularity analysis on the hot spot area to obtain a shape regularity analysis result; and determining a heat source of the hot spot area according to the boundary sharpness evaluation result and the shape regularity analysis result. According to the method, the boundary sharpness and the shape regularity can be combined to analyze a heat source so as to realize temperature anomaly control of the hardware fitting, the temperature anomaly detection accuracy is improved, and the iron accessory hardware fitting is effectively controlled to maintain a normal state.

Description

Technical Field

[0001] The present invention relates to the technical field of power component control, and in particular to a method and system for controlling abnormal temperature of iron accessory hardware of a power transmission and distribution line. Background Art

[0002] Transmission and distribution lines are crucial components of the power system, and their safe and stable operation is directly linked to the normal operation of society. Iron fittings, as key components in these lines, connect conductors, secure insulators, and carry current. Their operational condition is crucial. Internal connection problems in these fittings, such as loose bolts or deteriorated contact surfaces, can increase contact resistance, leading to abnormally high temperatures. This is often an early sign of a major line failure. To promptly detect and address these potential hazards, technicians typically use drones equipped with infrared thermal imaging devices for inspections, measuring the surface temperature of the fittings to assess their health. However, in outdoor environments, the measured temperature of the fittings reflects not only internal electrical heating but also external environmental factors, such as solar radiation. A single temperature reading is a mixed signal, incorporating factors such as internal electrical heating and solar radiation. Existing control methods struggle to effectively decompose this mixed signal, separating fault signals from complex environmental noise. Temperature anomaly detection suffers from low accuracy, making it difficult to maintain the normal operation of the iron fittings.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0004] The main purpose of the embodiments of the present invention is to propose a method and system for controlling temperature anomalies of iron accessories of power transmission and distribution lines, which can analyze the heat source by combining edge sharpness and shape regularity to achieve temperature anomaly control of the hardware, improve the accuracy of temperature anomaly detection, and effectively control the iron accessories to maintain a normal state.

[0005] In one aspect, an embodiment of the present invention provides a method for controlling abnormal temperature of iron accessories of a power transmission and distribution line, comprising the following steps: Acquire infrared thermal images of target hardware; performing hot spot separation on the infrared thermal image according to a preset separation temperature threshold to obtain a hot spot area; Performing a boundary sharpness evaluation on the hot spot area to obtain a boundary sharpness evaluation result; Performing shape regularity analysis on the hot spot area to obtain a shape regularity analysis result; The heat source of the hot spot area is determined according to the boundary sharpness evaluation result and the shape regularity analysis result.

[0006] In some embodiments, performing boundary sharpness evaluation on the hot spot area to obtain a boundary sharpness evaluation result includes: Select a pixel point from the boundary of the hot spot area as a target pixel point; Constructing a temperature difference analysis window, wherein the temperature difference analysis window is centered on the target pixel point; Calculate the temperature difference between each pixel in the temperature difference analysis window and the target pixel; averaging the absolute values ​​of the plurality of temperature differences to obtain a local temperature change rate; If the local temperature change rate is greater than a preset temperature gradient threshold, determining that the boundary sharpness evaluation result is high sharpness; If the local temperature change rate is less than a preset temperature gradient threshold, the boundary sharpness evaluation result is determined to be low sharpness.

[0007] In some embodiments, performing shape regularity analysis on the hot spot area to obtain a shape regularity analysis result includes: Extracting morphological features of the hot spot area and calculating the area and perimeter of the hot spot area; Calculating circularity based on the area and the perimeter; Calculating the aspect ratio of the minimum circumscribed rectangle of the hot spot area; If the roundness is greater than a preset roundness threshold and the aspect ratio is greater than a preset aspect ratio threshold, the shape regularity analysis result is determined to be high regularity; otherwise, the shape regularity analysis result is determined to be low regularity.

[0008] In some embodiments, determining the heat source of the hot spot area according to the boundary sharpness evaluation result and the shape regularity analysis result includes: Collecting a visible light image of the target hardware, where the timestamp of the visible light image is the same as the timestamp of the infrared thermal image; Performing contour extraction on the infrared thermal image to obtain a hot spot boundary contour; Performing contour extraction on the visible light image to obtain a shadow boundary contour; Calculating a degree of boundary overlap based on the hot spot boundary outline and the shadow boundary outline; The heat source of the hot spot area is determined according to the boundary sharpness evaluation result, the shape regularity analysis result and the boundary overlap.

[0009] In some embodiments, performing contour extraction on the infrared thermal image to obtain a hot spot boundary contour includes: Performing Gaussian smoothing on the infrared thermal image to obtain a Gaussian smoothed image; Calculating the gradient strength and direction of the Gaussian smoothed image; According to the gradient strength and direction, the boundary of the Gaussian smoothed image is refined by a double threshold suppression method to obtain the hot spot boundary contour.

[0010] In some embodiments, performing contour extraction on the visible light image to obtain a shadow boundary contour includes: Performing color space conversion on the visible light image so as to convert the visible light image from an RGB color space to an HSV color space, wherein parameters of the HSV color space include hue, saturation, and brightness; Binarizing the brightness component of the visible light image after color space conversion by a Gaussian adaptive threshold method to obtain a binary image, so as to separate the shadow area from the non-shadow area; Boundaries of the binary image are extracted using an edge detection algorithm to obtain the shadow boundary contour.

[0011] In some embodiments, binarizing the brightness component of the visible light image after color space conversion by using a Gaussian adaptive threshold method to obtain a binarized image includes: Acquire a brightness change region of the brightness component from the visible light image after color space conversion; Extracting the brightness value range of the brightness change area; Calculating the brightness distribution of the brightness value range; calculating the skewness and kurtosis of the brightness distribution; Determining an upper limit of the brightness value according to the skewness and kurtosis; updating the brightness value range according to the brightness value upper limit; Determining a shadow boundary brightness threshold according to the updated brightness value range; The brightness component is binarized according to the shadow boundary brightness threshold to obtain a binarized image.

[0012] In some embodiments, determining the heat source of the hot spot area according to the boundary sharpness evaluation result, the shape regularity analysis result, and the boundary overlap includes: If the boundary sharpness evaluation result is low sharpness and the shape regularity analysis result is low regularity, determining that the heat source is solar radiation; If the boundary sharpness evaluation result is high sharpness and the shape regularity analysis result is high regularity, determining whether the boundary overlap is greater than a preset overlap threshold; If the boundary overlap is greater than a preset overlap threshold, determining that the heat source is an environmental thermal artifact; If the boundary overlap is less than a preset overlap threshold, it is determined that the heat source is internal electrical heating.

[0013] In some embodiments, the method further comprises: Acquire infrared thermal image sequences; Selecting an image from the infrared thermal image sequence as an image to be processed; Extracting a temperature peak from the image to be processed as a hot spot center point; Constructing a hot spot core area according to a preset pixel side length and the center point of the hot spot; Select a pixel point from the hot spot core area as a pixel point to be processed; Combining the temperature values ​​corresponding to each image of the pixel to be processed in the infrared thermal image sequence according to the coordinate information of the pixel to be processed to obtain a temperature value sequence; Calculating an average temperature based on the temperature value sequence; Calculating a standard deviation as a temperature fluctuation amplitude based on the temperature value sequence and the average temperature; Performing statistical analysis on the temperature fluctuation amplitude corresponding to each pixel point in the hot spot core area to obtain a temperature fluctuation rate; If the temperature fluctuation rate is greater than a preset fluctuation threshold and the heat source is internal electrical heating, the heat source is updated to solar radiation.

[0014] On the other hand, an embodiment of the present invention provides a temperature abnormality control system for iron accessories of power transmission and distribution lines, comprising: An image acquisition module is used to acquire infrared thermal images of target hardware; a hot spot identification module, configured to separate hot spots from the infrared thermal image according to a preset separation temperature threshold to obtain a hot spot area; A boundary sharpness evaluation module is used to evaluate the boundary sharpness of the hot spot area and obtain a boundary sharpness evaluation result; A shape regularity analysis module is used to perform shape regularity analysis on the hot spot area to obtain a shape regularity analysis result; The hot spot source judgment module is used to determine the heat source of the hot spot area based on the boundary sharpness evaluation result and the shape regularity analysis result.

[0015] The embodiments of the present application include at least the following beneficial effects: the embodiments of the present application first obtain an infrared thermal image of the target hardware, and then perform hot spot separation on the infrared thermal image according to a preset separation temperature threshold to obtain a hot spot area, and then perform boundary sharpness evaluation on the hot spot area to obtain a boundary sharpness evaluation result, and perform shape regularity analysis on the hot spot area to obtain a shape regularity analysis result, and finally determine the heat source of the hot spot area based on the boundary sharpness evaluation result and the shape regularity analysis result, so that the heat source can be analyzed in combination with the boundary sharpness and shape regularity to achieve hardware temperature anomaly control, thereby improving the accuracy of temperature anomaly detection and effectively controlling the iron accessories hardware to maintain a normal state.

[0016] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained through the structures particularly pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0018] Figure 1 This is a flow chart of a method for controlling abnormal temperature of iron accessories of power transmission and distribution lines according to an embodiment of the present invention; Figure 2 The present invention is a schematic structural diagram of a temperature anomaly control system for iron accessories of power transmission and distribution lines according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the objectives, technical solutions, and advantages of this application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only intended to illustrate this application and are not intended to limit this application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements.

[0020] Before explaining the embodiments of the present application in detail, some of the nouns and terms involved in the embodiments of the present application are first explained. The nouns and terms involved in the embodiments of the present application are subject to the following explanations.

[0021] Hardware: Iron or aluminum metal accessories used to connect, secure, and protect overhead transmission lines and substation equipment in power systems, collectively referred to as power hardware. Their core functions are to transmit mechanical loads, ensure stable electrical contact, and handle tasks such as conductor suspension, splicing, and spacing control, directly impacting the safe operation of power lines. Based on their intended use, they can be categorized as line hardware, substation hardware, and busbar hardware, encompassing dozens of types, including suspension clamps, tension clamps, anti-vibration hammers, and spacer bars.

[0022] Transmission and distribution lines are crucial components of the power system, and their safe and stable operation is directly related to the normal operation of society. Iron fittings, as key components in these lines, connect conductors, secure insulators, and carry current, making their operational condition crucial. Problems with the internal connections of these fittings, such as loose bolts or deteriorated contact surfaces, can increase contact resistance, leading to abnormally high local temperatures. This is often an early sign of a major line failure. To promptly detect and address these potential hazards, power companies typically use drones equipped with infrared thermal imaging devices for inspections, measuring the surface temperature of the fittings to assess their health. However, in actual outdoor environments, the measured temperature of the fittings not only reflects internal electrical heating but is also significantly affected by external environmental factors, particularly solar radiation. Further complicating matters, the condition of the fittings' surface, including their condition, cleanliness, and degree of corrosion, can change dynamically over time. These changes directly affect the fittings' ability to absorb solar radiation, making existing methods for detecting temperature anomalies challenging and prone to false or missed detections.

[0023] Specifically, power transmission and distribution networks are critical infrastructure for the normal operation of society, and their stability and reliability are paramount. Within these lines, conductors carrying enormous currents are secured to insulator strings and suspended from transmission towers via various metal fittings, such as tension clamps and suspension clamps. These fittings serve as critical nodes in the current path, and their operational status directly impacts the safety of the entire line. Due to the inherent resistance of the fittings, as well as the contact resistance between the conductor and the clamps, high currents generate heat due to the Joule effect. Under normal circumstances, this heat generation is stable and controllable. However, if the fastening bolts of the fittings loosen due to long-term exposure to wind, sunlight, and vibration, or if the contact surfaces degrade due to environmental corrosion or electrochemical reactions, the contact resistance can increase dramatically. This can significantly increase the heat output at that point, leading to localized high temperatures. This abnormal temperature rise is a typical precursor to a major line failure. Sustained overheating can reduce the mechanical strength of metal materials, causing annealing. In severe cases, it can even cause the fittings to melt, the conductors to fall off, and cause widespread power outages.

[0024] To prevent such accidents, technicians typically monitor the temperatures of these critical nodes through regular inspections. Traditional manual inspections using infrared thermometers are inefficient and dangerous, so drones equipped with infrared thermal imaging devices have become the mainstream. Drones fly along a pre-set route, capturing thermal images of each hardware fixture along the route. The captured thermal images are transmitted to a backend analysis system in real time or afterward. After receiving the thermal images, the backend system extracts the highest temperature point for each hardware fixture. To determine whether this temperature is abnormal, the system typically uses a comparative analysis method: the temperature of the target hardware fixture is compared with other hardware fixtures on the same phase or tower, or with hardware in the same location on different phases. Under ideal operating conditions and environments, these temperatures should be similar. If the temperature of a hardware fixture is significantly higher than that of other reference hardware, the system identifies it as an abnormally hot spot and generates an alarm, notifying operations and maintenance personnel to conduct an inspection.

[0025] However, in real-world outdoor environments, this seemingly reliable temperature difference comparison method faces significant challenges. A key interfering factor is direct sunlight. During a drone inspection on a clear day, sunlight directly strikes the tower and transmission lines. Metal fittings absorb solar radiation, causing them to heat up. This solar-induced warming is not uniform. For example, due to the sun's varying position in the sky, fittings facing different directions receive varying degrees of sunlight at varying intensities and angles. An east-facing fitting will be warmer than a west-facing one in the morning, while the opposite is true in the afternoon. Furthermore, the tower's structure can obstruct the fittings, creating complex shadows. This can result in fittings on the same tower being exposed to sunlight while others are in the shadows, resulting in distinct solar warming effects.

[0026] The root of the problem lies in the fact that a single temperature reading is a composite of multiple factors, including electrical heating, ambient heat exchange, and solar warming. These factors are in turn modulated by a surface parameter that varies over time and is difficult to quantify. Existing control methods are unable to effectively decompose this complex information and accurately separate the true fault signal from the complex environmental noise.

[0027] During the temperature monitoring and fault diagnosis process of iron fittings of power transmission and distribution lines, the measured temperature of the fittings is a complex situation in which internal electrical heating and external heat absorption from solar radiation are superimposed, which increases the difficulty of controlling the fittings. It is necessary to distinguish and separate the "internal electrical heating component" and "solar radiation component" in the fitting temperature, so as to accurately identify the real abnormal temperature rise caused by the increase in contact resistance, and avoid misjudging old fittings with intact electrical connections as faults.

[0028] In view of this, the embodiment of the present application performs boundary sharpness evaluation and shape regularity analysis on the hot spot area in the infrared thermal image, and comprehensively determines the heat source of the hot spot, thereby effectively distinguishing internal electrical heating from external environmental influences, improving the accuracy of fault diagnosis, and effectively controlling iron accessories to maintain normal status.

[0029] The following is a detailed explanation of the embodiments of the present application with reference to the accompanying drawings: Figure 1 This is an optional flow chart of a method for controlling abnormal temperature of iron accessories of power transmission and distribution lines provided in an embodiment of the present application. Figure 1 The method may include but is not limited to steps S101 to S105.

[0030] Step S101: Acquire an infrared thermal image of a target hardware; Step S102: performing hot spot separation on the infrared thermal image according to a preset separation temperature threshold to obtain a hot spot area; Step S103: performing boundary sharpness evaluation on the hot spot area to obtain a boundary sharpness evaluation result; Step S104: performing shape regularity analysis on the hot spot area to obtain a shape regularity analysis result; Step S105: Determine the heat source of the hot spot area based on the boundary sharpness evaluation result and the shape regularity analysis result.

[0031] Steps S101 to S105 shown in the embodiment of the present application can combine boundary sharpness and shape regularity to analyze the heat source to achieve abnormal temperature control of hardware, thereby improving the accuracy of temperature anomaly detection and effectively controlling the iron accessory hardware to maintain a normal state.

[0032] In some embodiments, steps S101-S105 may first acquire an infrared thermal image of the target hardware. For example, this can be done by manually scanning and photographing the hardware with an infrared thermal imager, or by periodically capturing images using an infrared thermal imaging device mounted on a transmission and distribution line tower. Alternatively, drones equipped with infrared thermal imaging equipment can be used for patrol inspections to acquire infrared thermal image data over a wide area. This image data can then be transmitted to a processing unit for subsequent analysis.

[0033] Then, based on a preset separation temperature threshold, the infrared thermal image is subjected to hot spot separation to obtain hot spot regions. For example, a set of pixels in the image with temperatures above the preset separation temperature threshold can be identified as potential hot spot regions. Alternatively, hot spot regions can be directly designated on the infrared thermal image through manual selection. This allows for the preliminary identification of regions with abnormal temperatures. The preset separation temperature threshold can be set based on expert experience or the median temperature in the infrared thermal image.

[0034] Then, the hot spot area is evaluated for boundary sharpness to obtain a boundary sharpness evaluation result. For example, a technician can judge the edge clarity of the hot spot area by visually inspecting it, or by calculating the temperature difference between the edge pixels of the hot spot area and the pixels of its adjacent non-hot spot area. If the temperature difference is large, it is considered that the boundary sharpness is high; otherwise, it is low. A shape regularity analysis is performed on the hot spot area to obtain a shape regularity analysis result. For example, a technician can judge the geometric shape of the hot spot area based on experience, such as whether it is close to a circle or a rectangle. Alternatively, an image processing method can be used, such as calculating the minimum circumscribed rectangle of the hot spot area and judging the regularity of the shape based on its aspect ratio.

[0035] Finally, based on the edge sharpness assessment and shape regularity analysis results, the heat source of the hot spot area is determined. For example, this can be determined using pre-set judgment rules or by consulting a pre-set judgment table. For example, low edge sharpness and low shape regularity can be determined as solar radiation, while high edge sharpness and high shape regularity can be determined as internal electrical heating. These judgment rules can be set based on historical data and expert experience.

[0036] Through the above technical solution, this embodiment introduces the boundary sharpness evaluation and shape regularity analysis of the hot spot area in the infrared thermal image, which can more deeply explore the intrinsic physical characteristics of the hot spot and effectively distinguish the heat caused by internal electrical faults in the hardware from the temperature increase caused by external environmental factors such as solar radiation. This embodiment can more accurately identify the true source of heat by comprehensively analyzing the geometric characteristics of the hot spot, such as the clarity of its edges and the regularity of its overall shape. In this way, the problems of false alarms and missed alarms can be effectively avoided, and the accuracy and reliability of the temperature anomaly diagnosis of iron accessories hardware of the transmission and distribution lines can be significantly improved, thereby ensuring the safe and stable operation of the power system.

[0037] In some embodiments, in step S103, performing a boundary sharpness evaluation on the hot spot area to obtain a boundary sharpness evaluation result may include but is not limited to the following steps: Select a pixel from the boundary of the hot spot area as the target pixel; Construct a temperature difference analysis window with the target pixel as the center; Calculate the temperature difference between each pixel in the temperature difference analysis window and the target pixel; The absolute values ​​of multiple temperature differences are averaged to obtain the local temperature change rate; If the local temperature change rate is greater than a preset temperature gradient threshold, the boundary sharpness evaluation result is determined to be high sharpness; If the local temperature change rate is less than the preset temperature gradient threshold, the boundary sharpness evaluation result is determined to be low sharpness.

[0038] In some embodiments, a pixel can be selected from the boundary of the hot spot area as a target pixel. This target pixel serves as the reference point for subsequent temperature gradient analysis. A temperature difference analysis window is then constructed with the target pixel as the center. This temperature difference analysis window can be a rectangular or circular area of ​​a preset size, sufficient to cover the target pixel and the pixels within a certain range around it, allowing for local temperature change analysis. For example, the temperature difference analysis window can be set to a 3x3 or 5x5 pixel matrix.

[0039] The temperature difference between each pixel in the temperature difference analysis window and the target pixel is then calculated. These temperature differences reflect the local temperature variation at the hot spot boundary. To quantify this variation, the absolute values ​​of these temperature differences can be averaged to obtain the local temperature change rate. Using the absolute value eliminates the influence of positive and negative temperature differences, focusing only on the magnitude of the temperature change. A higher local temperature change rate indicates a greater temperature gradient and a sharper boundary in that area.

[0040] Finally, the calculated local temperature change rate is compared with a preset temperature gradient threshold. This preset temperature gradient threshold is determined based on experience or experimental data and is used to distinguish between high-sharp and low-sharp boundaries. If the local temperature change rate is greater than the preset temperature gradient threshold, the boundary sharpness assessment result of the hot spot area is determined to be high sharpness, which generally means that heat is concentrated and the boundary is clear. Conversely, if the local temperature change rate is less than the preset temperature gradient threshold, the boundary sharpness assessment result is determined to be low sharpness, which may indicate heat diffusion or blurred boundaries.

[0041] This embodiment can effectively evaluate the sharpness of the hot spot boundary by quantitatively analyzing the local temperature change rate of the hot spot area boundary. When internal electrical heating occurs in the iron accessories of the power transmission and distribution line, the heat is usually concentrated at the fault point, resulting in a significant temperature gradient between the hot spot area and the surrounding environment, thereby forming a high-sharpness boundary. On the contrary, when the hardware is heated up by external environmental factors such as sunlight, the heat distribution is usually more uniform, and the temperature gradient of the hot spot boundary is relatively small, showing as a low-sharpness boundary. By calculating the local temperature change rate and comparing it with the preset threshold, the physical characteristics of the hot spot boundary can be objectively reflected, providing a key basis for the subsequent judgment of the heat source.

[0042] Through the above technical solution, this embodiment precisely calculates the local temperature change rate at the hot spot boundary and classifies it based on a preset threshold, avoiding the subjectivity and inaccuracy of traditional manual visual judgment. This makes the identification of hot spot boundary characteristics more accurate and automated. This lays a solid foundation for the subsequent precise determination of heat sources and improves the accuracy and reliability of temperature anomaly control for iron accessories and hardware on power transmission and distribution lines.

[0043] In some embodiments, in step S104, performing shape regularity analysis on the hot spot area to obtain a shape regularity analysis result may include but is not limited to the following steps: Extract morphological features of the hot spot area and calculate the area and perimeter of the hot spot area; Calculate circularity based on area and perimeter; Calculate the aspect ratio of the minimum bounding rectangle of the hot spot area; If the roundness is greater than the preset roundness threshold and the aspect ratio is greater than the preset aspect ratio threshold, the shape regularity analysis result is determined to be high regularity; otherwise, the shape regularity analysis result is determined to be low regularity.

[0044] In some embodiments, morphological features can be extracted from the hot spot region to calculate its area and perimeter. Morphological feature extraction involves using image processing techniques to quantitatively describe the hot spot region's geometric shape. Specifically, methods such as connected domain analysis and contour tracing can be used to accurately capture the hot spot region's pixel set, and then calculate the number of pixels it occupies (i.e., area) and the pixel length of its boundary (i.e., perimeter), providing basic data for subsequent shape regularity quantification.

[0045] Then, the circularity is calculated based on the area and perimeter. The circularity is an indicator to measure the degree to which the shape of the hot spot area is close to a circle. The calculation method is: , where is the roundness, is the area, The closer the roundness value is to 1, the closer the shape of the hot spot area is to a circle. Its purpose is to characterize the compactness of the hot spot through a dimensionless parameter.

[0046] Next, calculate the aspect ratio of the minimum bounding rectangle of the hot spot area. It's understood that the aspect ratio of the minimum bounding rectangle is the ratio of the long side to the short side of the smallest rectangle that can completely enclose the hot spot area. This parameter is used to assess the ductility or flatness of the hot spot area. For example, an elongated hot spot area will have a larger aspect ratio, while an area that is approximately square or circular will have an aspect ratio close to 1. Its purpose is to reflect the geometric characteristics of the hot spot from another dimension.

[0047] Finally, thresholds can be used to distinguish hot spot areas with different shape characteristics. If the roundness is greater than a preset roundness threshold and the aspect ratio is greater than a preset aspect ratio threshold, it indicates that the shape of the hot spot area is relatively regular and may be related to a specific heat source. The shape regularity analysis result can be determined as high regularity. Otherwise, it indicates that the shape is irregular, such as caused by environmental factors, and the shape regularity analysis result can be determined as low regularity. The preset roundness threshold and the preset aspect ratio threshold are critical values ​​pre-set based on experience or a large amount of experimental data.

[0048] This embodiment can quantitatively evaluate the degree of shape regularity of the hot spot area by comprehensively analyzing the morphological features of the hot spot area, such as the area, perimeter, roundness, and aspect ratio of the minimum circumscribed rectangle. Specifically, the area and perimeter provide basic size information of the hot spot area; the roundness reflects the compactness of the hot spot from the perspective of circular similarity; and the aspect ratio of the minimum circumscribed rectangle reveals the ductility of the hot spot. The combined use of these parameters enables the shape characteristics of the hot spot to be captured comprehensively and accurately. By comparing these quantitative features with the preset threshold, the hot spot area can be objectively divided into high regularity or low regularity, thereby providing key shape feature basis for subsequent heat source judgment.

[0049] Through the above technical solution, this embodiment can perform a refined shape regularity analysis on the hot spot area in the infrared thermal image of the iron accessories of the power transmission and distribution line. This analysis method not only takes into account the overall size of the hot spot, but also quantifies its geometric features in more depth, such as compactness and ductility. As a result, hot spots with specific shape characteristics can be more accurately identified, such as hot spots that are usually more regular due to internal electrical heating, or hot spots with irregular shapes caused by external environmental factors (such as sunlight). This refined shape analysis capability significantly improves the accuracy and reliability of the hot spot source judgment, helps to avoid misjudgment, and thus improves the level of intelligence in the abnormal temperature control of the iron accessories of the power transmission and distribution line.

[0050] In some embodiments, in step S105, determining the heat source of the hot spot area based on the boundary sharpness evaluation result and the shape regularity analysis result may include but is not limited to the following steps: Step S201: Acquire a visible light image of the target hardware, where the timestamp of the visible light image is the same as the timestamp of the infrared thermal image; Step S202: performing contour extraction on the infrared thermal image to obtain the hot spot boundary contour; Step S203: performing contour extraction on the visible light image to obtain a shadow boundary contour; Step S204: Calculate the boundary overlap based on the hot spot boundary outline and the shadow boundary outline; Step S205: Determine the heat source of the hot spot area based on the boundary sharpness evaluation result, the shape regularity analysis result and the boundary overlap.

[0051] In some embodiments, relying solely on edge sharpness and shape regularity can make it difficult to accurately distinguish between surface temperature increases caused by external environmental factors (such as sunlight and environmental heating artifacts) and actual heating caused by internal electrical faults in hardware. For example, strong sunlight can create high-temperature areas on the surface of a hardware fixture. These areas, as seen on infrared thermal images, may resemble hot spots caused by internal electrical heating in terms of edge sharpness and shape regularity. This can lead to misjudgments, unnecessary on-site inspections, increased maintenance costs, and even delayed detection and resolution of actual electrical faults.

[0052] To this end, a visible light image of the target hardware can be acquired first. For example, a visible light image of the same target hardware can be acquired simultaneously with an infrared thermal image. The visible light image provides complementary visual information to the infrared thermal image, such as the object's shape, color, texture, and the presence of shadows. The visible light image's timestamp is identical to the infrared thermal image's, ensuring that the two images capture the scene at the same moment. This ensures consistency and correspondence in image content, providing an accurate foundation for subsequent image fusion and analysis.

[0053] Contour extraction is then performed on the infrared thermal image to obtain the hot spot boundary outline. Image processing algorithms can be used to identify and delineate the precise boundaries of the hot spot area in the infrared thermal image. This hot spot boundary outline reflects the actual range and shape of the heat distribution. Contour extraction is also performed on the visible light image to obtain the shadow boundary outline. Shadow boundaries, created by lighting conditions, can be identified and extracted in the visible light image. This shadow boundary outline indicates the degree of obstruction of an object under visible light.

[0054] Based on the hot spot boundary outlines and the shadow boundary outlines, the boundary overlap is calculated to quantify the degree of match between the hot spot boundary in the infrared thermal image and the shadow boundary in the visible light image. Boundary overlap can be calculated using various image matching or similarity measurement algorithms, such as by calculating the ratio of the intersection to the union of two contours or by calculating the average distance between contour points. A high degree of overlap generally indicates a high degree of spatial correspondence between the hot spot area and the shadow area.

[0055] Finally, the heat source of the hot spot area is determined based on the results of boundary sharpness assessment, shape regularity analysis, and boundary overlap. This involves combining features extracted from infrared thermal images (boundary sharpness and shape regularity) with features extracted from visible light images (boundary overlap). This fusion of multimodal information makes the determination of the heat source more comprehensive and accurate.

[0056] This embodiment effectively addresses the limitations of relying solely on infrared thermal image features to determine heat sources by introducing visible light images and calculating the overlap between the hot spot boundary outline and the shadow boundary outline. Traditional methods can be ambiguous when distinguishing between solar radiation, environmental thermal artifacts, and internal electrical heating. For example, solar radiation can cause the surface temperature of hardware to rise, forming a hot spot on an infrared thermal image. The sharpness and regularity of its boundary may be similar to those of a hot spot caused by internal electrical heating. However, solar radiation typically produces a distinct shadow in a visible light image, and the boundaries of the hot spot area often correspond to those of the shadow. By acquiring a visible light image with the same timestamp as the infrared thermal image and extracting the hot spot boundary outline and the shadow boundary outline separately, this embodiment can calculate the overlap between the two boundaries. When the heat source of the hot spot area is solar radiation, the hot spot boundary outline and the shadow boundary outline in the visible light image typically show a high degree of overlap because there is a clear temperature gradient and boundary between the area blocked from sunlight (the shadow) and the unblocked area (heated by direct sunlight). In contrast, for internal electrical heating, the heat comes from the current effect inside the hardware, which usually does not produce a corresponding shadow boundary in the visible light image. Therefore, the overlap between the hot spot boundary outline and the shadow boundary outline will be low. For environmental thermal artifacts, such as hot spots caused by reflections from nearby high-temperature objects, they may show specific boundaries and shape characteristics on infrared thermal images, but they usually do not have corresponding shadows in visible light images, or the overlap between their boundaries and shadow boundaries is low. By comprehensively considering the boundary sharpness evaluation results, shape regularity analysis results, and the newly added boundary overlap, a more sophisticated judgment logic can be constructed to more accurately distinguish different types of heat sources.

[0057] In order to more clearly illustrate the technical solution, a specific example is used for explanation below. Suppose that a hot spot area is found during infrared thermal imaging detection of a certain power transmission and distribution line iron accessory hardware. First, an infrared thermal image of the hardware is obtained, and the hot spot is separated to obtain the hot spot area. Then, the boundary sharpness of the hot spot area is evaluated to obtain a boundary sharpness evaluation result; and the shape regularity of the hot spot area is analyzed to obtain a shape regularity analysis result. On this basis, in order to more accurately determine the source of heat, a visible light image of the hardware with the same timestamp as the infrared thermal image is further collected. Subsequently, the infrared thermal image is contour extracted to obtain the hot spot boundary contour; at the same time, the visible light image is contour extracted to obtain the shadow boundary contour.

[0058] For example, if the hot spot region's boundary sharpness assessment results in low sharpness and the shape regularity analysis results in low regularity, this initially suggests possible solar radiation. By calculating the degree of overlap between the hot spot boundary outline and the shadow boundary outline, if the overlap is high (for example, greater than a preset overlap threshold), it can be further confirmed that the hot spot is caused by solar radiation. This indicates that the formation of the hot spot corresponds spatially to the shadow region in the visible light image, consistent with the characteristics of solar radiation. If the hot spot region's boundary sharpness assessment results in high sharpness and the shape regularity analysis results in high regularity, this initially suggests possible internal electrical heating or environmental heating artifacts. If the calculated boundary overlap is low (for example, less than a preset overlap threshold), it can be clearly determined that the hot spot is caused by internal electrical heating. This is because internal electrical heating generally does not produce a corresponding shadow in the visible light image, and the overlap between the hot spot boundary and the shadow boundary is naturally low. Conversely, if the boundary overlap is high, it may be an environmental thermal artifact. For example, a hot spot caused by reflection from a nearby high-temperature object may appear sharp and regular in the infrared image, but there may be some overlap with the shadow boundary in the visible light image (for example, the reflection source itself casts a shadow in visible light). In this way, combining the inherent characteristics of infrared thermal images with the environmental information provided by visible light images can effectively avoid misjudgments and improve the accuracy of heat source determination, thereby providing a more reliable basis for fault diagnosis and maintenance of power transmission and distribution lines.

[0059] Through the above technical solution, this embodiment significantly improves the accuracy and reliability of the heat source judgment in the hot spot area of ​​the iron accessories of the power transmission and distribution lines. By introducing visible light images and calculating the overlap between the hot spot boundary contour and the shadow boundary contour, it is possible to effectively distinguish between the surface temperature increase caused by external environmental factors (such as solar radiation, environmental thermal artifacts) and the real heating caused by internal electrical faults. This avoids the misjudgment that may occur in traditional methods, such as misjudging hot spots caused by solar radiation as internal electrical heating, thereby reducing unnecessary on-site inspections and maintenance costs. At the same time, it also reduces the risk of missing real internal electrical faults, ensuring the safe and stable operation of the power transmission and distribution lines. This fusion analysis of multimodal image information makes the judgment of the heat source more comprehensive and objective, and provides more solid technical support for the intelligent operation and maintenance of the power transmission and distribution lines.

[0060] In some embodiments, in step S202, contour extraction is performed on the infrared thermal image to obtain the hot spot boundary contour, which may include but is not limited to the following steps: Perform Gaussian smoothing on the infrared thermal image to obtain a Gaussian smoothed image; Compute the gradient strength and direction of a Gaussian-smoothed image; According to the gradient intensity and direction, the boundary of the Gaussian smoothed image is refined by the double threshold suppression method to obtain the hot spot boundary contour.

[0061] In some embodiments, Gaussian smoothing can be performed on the infrared thermal image to produce a Gaussian smoothed image. This denoises the image and reduces random noise, thereby providing a clearer image foundation for subsequent edge detection. Gaussian smoothing is a linear smoothing filter whose weight coefficients follow a Gaussian distribution. It effectively blurs the image, reducing image detail while preserving the overall image structure.

[0062] The gradient strength and direction of the Gaussian-smoothed image are then calculated. For example, each pixel in the Gaussian-smoothed image can be processed using a specific operator (such as the Sobel, Prewitt, or Canny operator) to obtain the horizontal and vertical grayscale change rate for that pixel. The gradient strength indicates the severity of the grayscale change, while the gradient direction indicates the direction of the grayscale change. This information is key to identifying image edges.

[0063] Then, based on the gradient strength and direction, the double threshold suppression method is used to refine the boundaries of the Gaussian smoothed image to obtain the hot spot boundary contour in order to accurately locate the edges in the image. The double threshold suppression method usually includes two thresholds: a high threshold and a low threshold. First, all pixels with gradient strength higher than the high threshold are determined to be strong edge pixels. Then, for pixels with gradient strength between the high threshold and the low threshold, if they are connected to strong edge pixels, they are also considered to be edge pixels. At the same time, pixels below the low threshold are suppressed. This method helps to eliminate false edges and connect broken edges to obtain a continuous and refined hot spot boundary contour.

[0064] This embodiment aims to optimize the contour extraction process of infrared thermal images by introducing Gaussian smoothing, gradient calculation and double threshold suppression methods. Specifically, Gaussian smoothing can effectively filter out noise in the image, avoid interference of noise on subsequent edge detection, and ensure that the extracted edge information is purer. Subsequently, by calculating the gradient intensity and direction of the Gaussian smoothed image, the areas with drastic temperature changes in the image can be accurately captured. These areas usually correspond to the boundaries of hot spots. Finally, the double threshold suppression method is used to process these gradient information, which can accurately screen out the real hot spot boundaries, suppress pseudo edges, and connect possible broken edges, thereby forming a complete and detailed hot spot boundary contour. This step-by-step processing method ensures that the hot spot boundary contour extracted from the original infrared thermal image has higher accuracy and robustness, and provides a reliable image basis for subsequent heat source judgment.

[0065] Through the above technical solution, this embodiment can achieve accurate extraction of the hot spot boundary contour in the infrared thermal image. The use of Gaussian smoothing can effectively reduce the interference of image noise on edge detection, making the extracted boundary clearer. The calculation of gradient strength and direction ensures sensitive capture of temperature changes at the edge of the hot spot. Furthermore, the application of the dual threshold suppression method effectively avoids the misjudgment of weak edges and the omission of strong edges, thereby obtaining a continuous and refined hot spot boundary contour. As a result, the obtained hot spot boundary contour has higher accuracy and reliability, significantly improving the accuracy of subsequent heat source judgment, and helping to more accurately identify the cause of temperature anomalies in iron accessories of power transmission and distribution lines.

[0066] In some embodiments, in step S203, contour extraction is performed on the visible light image to obtain the shadow boundary contour, which may include but is not limited to the following steps: Step S301: performing color space conversion on the visible light image, so as to convert the visible light image from the RGB color space to the HSV color space, where parameters of the HSV color space include hue, saturation, and brightness; Step S302: binarizing the brightness component of the visible light image after the color space conversion using a Gaussian adaptive threshold method to obtain a binarized image so as to separate the shadow area from the non-shadow area; Step S303: extract the boundary of the binary image using an edge detection algorithm to obtain the shadow boundary contour.

[0067] In some embodiments, the visible light image can be first converted into a color space so that the visible light image is converted from the RGB color space to the HSV color space, wherein the parameters of the HSV color space include hue, saturation, and brightness. The RGB color space is an additive color model that represents colors through different combinations of the three primary colors of red, green, and blue. The HSV color space (Hue, Saturation, Value / Brightness) is a color model that is more in line with human visual perception, wherein hue represents the color type, saturation represents the purity of the color, and brightness represents the lightness or darkness of the color. The purpose of converting the visible light image from the RGB color space to the HSV color space is to separate the brightness information of the image from the color information, because the recognition of shadows mainly depends on changes in brightness, and the brightness component is more sensitive to changes in illumination, which helps in the subsequent recognition of shadow areas.

[0068] The luminance component of the color-space-converted visible light image is then binarized using the Gaussian adaptive thresholding method to produce a binary image that separates shadow and non-shadow areas. The Gaussian adaptive thresholding method dynamically determines the threshold based on the pixel value distribution of a local region of the image, rather than using a single global threshold. Specifically, for each pixel in the image, the method considers the pixel values ​​in its surrounding neighborhood and calculates a local threshold based on the statistical properties of these pixel values ​​(e.g., mean or weighted mean). This method effectively separates shadow and non-shadow areas, generating a binary image, even in the presence of uneven illumination or varying shadow depths. After binarization, shadow areas in the image are labeled with one value (e.g., black), and non-shadow areas are labeled with another value (e.g., white), achieving a clear separation between shadow and non-shadow areas.

[0069] The binary image is then subjected to edge detection algorithms to extract boundaries and obtain the shadow boundary outline. Edge detection algorithms are used to identify areas in an image where significant changes in brightness or color occur. These areas typically correspond to the boundaries or outlines of objects. After obtaining the binary image, there is a significant brightness difference between shadow and non-shadow areas, so edge detection algorithms can be used to accurately extract the shadow boundary outline. In practical applications, a variety of mature edge detection algorithms can be used, such as the Canny algorithm, Sobel operator, Prewitt operator, Roberts operator, or Laplacian operator. These algorithms identify edges by calculating the gradient information of the image.

[0070] This embodiment can effectively decouple the brightness information of the image from the color information by converting the visible light image from the RGB color space to the HSV color space, so that the subsequent shadow recognition process can focus more on the core feature of brightness change. The use of the Gaussian adaptive threshold method to binarize the brightness component can overcome the limitations of the traditional global threshold method in uneven lighting scenes, ensuring that the shadow area can be accurately separated from the non-shadow area to generate a clear binary image. On this basis, by applying the edge detection algorithm, the boundary between the shadow area and the non-shadow area in the binary image can be accurately identified and extracted, thereby obtaining an accurate shadow boundary contour. This series of processing steps works together to ensure that shadow information can be reliably obtained even under complex lighting conditions, providing key auxiliary data for subsequent heat source judgment.

[0071] Through the above technical solution, this embodiment can overcome the challenge of traditional methods that are difficult to accurately identify shadow areas under complex lighting conditions. By combining color space conversion and Gaussian adaptive threshold method, the identification of shadow areas is no longer limited to a single global brightness threshold, thereby improving the accuracy and robustness of the separation of shadow areas from non-shadow areas. As a result, the extracted shadow boundary contour is more accurate, providing a more reliable basis for the subsequent judgment of the heat source in the hot spot area, and effectively improving the accuracy of temperature anomaly control of iron accessories of power transmission and distribution lines.

[0072] In some embodiments, in step S302, binarizing the brightness component of the visible light image after color space conversion by using a Gaussian adaptive threshold method to obtain a binarized image may include but is not limited to the following steps: Obtaining a brightness change area of ​​a brightness component from a visible light image after color space conversion; Extract the brightness value range of the brightness change area; Calculate the brightness distribution of a range of brightness values; Calculate the skewness and kurtosis of the brightness distribution; Determine the upper limit of brightness value based on skewness and kurtosis; Update the brightness value range according to the upper limit of the brightness value; Determine the shadow boundary brightness threshold according to the updated brightness value range; According to the brightness threshold of the shadow boundary, the brightness component is binarized to obtain a binary image.

[0073] In some embodiments, the brightness variation regions of the luminance component can be first obtained from the visible light image after color space conversion. For example, by analyzing the brightness value differences of pixels in the image, regions with significant brightness changes can be identified. These regions typically correspond to edges or light-dark junctions in the image, such as shadow boundaries. In practical applications, this can be achieved by calculating local brightness gradients or using specific filters.

[0074] The brightness range of the brightness variation region is then extracted. The minimum and maximum brightness values ​​of all pixels within the identified brightness variation region can be determined, thereby defining a brightness interval encompassing both potential shadow and non-shadow areas as the brightness range. Simultaneously, the brightness distribution of the brightness range is calculated, allowing statistical analysis of the pixel brightness values ​​within the range, such as generating a brightness histogram. This histogram visually displays the frequency of occurrence of different brightness values ​​as a brightness distribution. Statistical analysis of the brightness distribution can also be performed to calculate the skewness and kurtosis of the brightness distribution. Skewness measures the symmetry of the brightness distribution, indicating whether the tail of the distribution skews toward high or low brightness. Kurtosis measures the sharpness or flatness of the brightness distribution, reflecting the degree of concentration or dispersion of the brightness values. These statistics provide a more detailed characterization of the brightness distribution and help distinguish brightness variations caused by shadows from those caused by other factors.

[0075] Then, based on the skewness and kurtosis, an upper limit for the brightness value is determined. For example, when the brightness distribution exhibits a significant low-brightness skew (i.e., a negative skewness) and a high kurtosis, this may indicate the presence of a concentrated shadow region. In this case, an appropriate upper brightness limit can be adaptively set based on these statistical characteristics to better define the brightness range of the shadow region. At the same time, based on the upper brightness limit, the brightness range is updated. The original brightness range can be adjusted based on the newly determined upper brightness limit, allowing subsequent analysis to focus more closely on the brightness range associated with shadows.

[0076] Finally, based on the updated brightness range, the shadow boundary brightness threshold is determined. The brightness component is binarized based on this threshold to produce a binary image. The shadow boundary brightness threshold is a key parameter for binarizing the brightness component in an image, dividing pixels into shadow and non-shadow areas. This threshold is adaptive and can be adjusted based on the actual brightness distribution characteristics of the image, thereby improving the accuracy of shadow segmentation.

[0077] This embodiment performs in-depth statistical analysis of the brightness components of visible light images, specifically using skewness and kurtosis to characterize the brightness distribution, thereby adaptively determining a precise shadow boundary brightness threshold. This approach overcomes the problem of inaccurate shadow region segmentation under complex lighting conditions using traditional fixed threshold or simple adaptive threshold methods. This is particularly true in visible light images of iron fittings of power transmission and distribution lines, where factors such as ambient lighting and structural complexity can lead to diverse shadow morphology and brightness variations. By analyzing the skewness and kurtosis of the brightness distribution, the brightness characteristics of shadow regions can be more accurately captured, resulting in a more robust separation of shadow and non-shadow regions.

[0078] Through the above-mentioned technical solution, this embodiment can significantly improve the accuracy and robustness of binarization processing of shadow areas in visible light images. Because the threshold is adaptively determined based on the statistical characteristics of the brightness distribution (skewness and kurtosis), this method can better adapt to shadow changes under different lighting conditions and scenarios, avoiding the over-segmentation or under-segmentation problems that may occur with fixed thresholds or simple adaptive threshold methods. This makes the subsequent extraction of shadow boundary contours more accurate, providing a reliable image processing foundation for accurately determining the heat source in hot spot areas, effectively improving the overall performance of the temperature anomaly control method for iron accessories of power transmission and distribution lines.

[0079] In some embodiments, in step S205, determining the heat source of the hot spot area based on the boundary sharpness evaluation result, the shape regularity analysis result, and the boundary overlap may include but is not limited to the following steps: If the boundary sharpness evaluation result is low sharpness and the shape regularity analysis result is low regularity, it is determined that the heat source is solar radiation; If the boundary sharpness evaluation result is high sharpness and the shape regularity analysis result is high regularity, then determine whether the boundary coincidence is greater than a preset coincidence threshold; If the boundary overlap is greater than a preset overlap threshold, the heat source is determined to be an environmental thermal artifact; If the boundary overlap is less than a preset overlap threshold, it is determined that the heat source is internal electrical heating.

[0080] In some embodiments, the boundary sharpness assessment results, shape regularity analysis results, and boundary overlap can be used to determine the heat source in the hot spot area. If the boundary sharpness assessment result is low sharpness and the shape regularity analysis result is low regularity, it indicates that the heat diffusion range is wide, the boundaries are blurred, and the shape is irregular, which is consistent with the phenomenon of surface temperature increase caused by solar radiation. The heat source can be determined to be solar radiation. Solar radiation generally causes the entire surface of an object to be heated, with relatively uniform heat distribution and unclear boundaries. The shape of the hot spot is often irregular, affected by the surface geometry and the angle of illumination.

[0081] If the boundary sharpness assessment results are high and the shape regularity analysis results are high, this indicates that the hot spot area has concentrated heat, clear boundaries, and a relatively regular shape. This characteristic is often associated with a localized temperature increase caused by internal heat sources (such as electrical heating) or specific environmental factors (such as reflections and shadows). Further determination is needed to determine whether the boundary overlap exceeds the preset overlap threshold to determine the source of the heat.

[0082] If the boundary overlap exceeds a preset threshold, meaning the hot spot boundary outline closely matches the shadow boundary outline in the visible light image, the hot spot is likely caused by shadows or reflections in the environment, rather than internal heating of the hardware itself. This indicates that the heat source is an environmental thermal artifact. For example, when sunlight is blocked by nearby objects, creating a shadow, or when light reflects off a smooth surface, it can create a hot spot-like area in the infrared image.

[0083] If the boundary overlap is less than the preset overlap threshold, that is, when the overlap between the hot spot boundary outline and the shadow boundary outline is low, the possibility of environmental thermal artifacts is ruled out. At this time, the hot spot is more likely to be caused by an electrical fault or abnormal heating inside the hardware, and it can be determined that the source of heat is internal electrical heating.

[0084] This embodiment achieves precise identification of heat sources by comprehensively analyzing the hot spot region's boundary sharpness assessment results, shape regularity analysis results, and overlap with shadow boundaries in visible light images. Specifically, when a hot spot exhibits low sharpness and regularity, its heat distribution characteristics are highly consistent with surface heating caused by solar radiation, and therefore it is determined to be solar radiation. However, when a hot spot exhibits high sharpness and regularity, it indicates the presence of a localized, concentrated heat source, requiring further analysis using visible light information. By calculating the overlap between the hot spot boundary outline and the shadow boundary outline, it is possible to effectively distinguish between pseudo-hot spots caused by environmental factors (such as reflections and shadows) and actual electrical heating within the hardware. High overlap indicates environmental heating artifacts, as environmental factors often leave traces in both visible and infrared images; low overlap indicates internal electrical heating, as internal heating is generally not highly correlated with external shadows or reflections. This multi-dimensional, cross-validated judgment mechanism significantly improves the accuracy and reliability of heat source identification.

[0085] Through the above technical solution, this embodiment can effectively distinguish between various heat sources that can cause abnormal temperatures in iron fittings on power transmission and distribution lines, including solar radiation, environmental thermal artifacts, and internal electrical heating. This prevents misidentification of non-fault-related heat increases as internal electrical faults, thereby reducing unnecessary on-site inspections and maintenance costs. Furthermore, genuine internal electrical heating can be identified promptly and accurately, providing a reliable basis for subsequent fault diagnosis and resolution, ensuring the safe and stable operation of power transmission and distribution lines.

[0086] In some embodiments, the method further comprises: Acquire infrared thermal image sequences; Select an image from the infrared thermal image sequence as an image to be processed; Extract the temperature peak from the image to be processed as the center point of the hot spot; Construct the hot spot core area according to the preset pixel side length and hot spot center point; Select a pixel from the hot spot core area as the pixel to be processed; According to the coordinate information of the pixel to be processed, the temperature value corresponding to each image of the pixel to be processed in the infrared thermal image sequence is combined to obtain a temperature value sequence; Calculate the average temperature based on the temperature value sequence; According to the temperature value series and the average temperature, the standard deviation is calculated as the temperature fluctuation amplitude; Statistical analysis is performed on the temperature fluctuation amplitude corresponding to each pixel point in the core area of ​​the hot spot to obtain the temperature fluctuation rate; If the temperature fluctuation rate is greater than the preset fluctuation threshold and the heat source is internal electrical heating, the heat source is updated to solar radiation.

[0087] In some embodiments, there may be limitations in determining the source of heat due to relying solely on the static features of infrared thermal images. For example, the infrared characteristics of hot spots caused by certain environmental factors (such as solar radiation) may be similar to those of hot spots caused by internal electrical heating under certain conditions, leading to misjudgment and misidentification of heat caused by solar radiation as internal electrical heating, which may in turn cause unnecessary maintenance or waste of resources. To this end, the source of heat can be further determined by temperature fluctuation information within a certain period of time. A sequence of infrared thermal images can be first obtained. For example, infrared thermal images of target hardware can be continuously or periodically collected over a period of time to form an image set containing time dimension information. The acquisition of this sequence is intended to capture the temperature dynamics of the hot spot area over time.

[0088] Then, an image is selected from the infrared thermal image sequence as the image to be processed, and the temperature peak is extracted from the image to be processed as the center point of the hot spot. The pixel with the highest temperature can be located in the identified hot spot area, and its coordinates can be used as the central reference point of the hot spot. At the same time, based on the preset pixel side length and the hot spot center point, a hot spot core area is constructed. A sub-area of ​​a preset size (for example, a 5×5 pixel rectangular area or a circular area with a preset radius) can be delineated with the hot spot center point as the center. This area is considered to be the core part of the hot spot where the temperature change is most significant. A pixel point is selected from the hot spot core area as the pixel point to be processed. In the subsequent statistical analysis, statistical analysis can be performed on each pixel point in the hot spot core area.

[0089] Then, based on the coordinate information of the pixel to be processed, the temperature values ​​corresponding to each image in the infrared thermal image sequence of the pixel to be processed are combined to obtain a temperature value sequence. For example, by tracking the temperature values ​​of the pixel at the same spatial position at different time points, a sequence reflecting the change of the temperature of the pixel over time can be formed. For example, if the core area of ​​the hot spot contains multiple pixels, a corresponding temperature value sequence is generated for each pixel. For each pixel, the average temperature within the observation time period can be calculated based on the temperature value sequence, and the standard deviation can be calculated as the temperature fluctuation amplitude based on the temperature value sequence and the average temperature. The standard deviation in statistics can be used to quantify the degree of discreteness of the temperature value sequence of each pixel. The larger the standard deviation, the more severe the temperature fluctuation of the pixel.

[0090] Finally, the temperature fluctuation amplitude corresponding to each pixel in the core area of ​​the hot spot is statistically analyzed to obtain the temperature fluctuation rate. For example, the temperature fluctuation amplitudes of all pixels in the core area of ​​the hot spot can be summarized and analyzed. For example, the average, maximum or certain percentile of these fluctuation amplitudes can be calculated to obtain a comprehensive indicator representing the degree of temperature fluctuation in the entire core area of ​​the hot spot, namely the temperature fluctuation rate. The temperature fluctuation rate reflects the overall temperature stability of the hot spot. If the temperature fluctuation rate is greater than the preset fluctuation threshold and the heat source is internal electrical heating, the heat source is updated to solar radiation. This means that when the static image features preliminarily determine that the heat source is internal electrical heating, if further dynamic temperature fluctuation analysis shows that the hot spot has significant temperature fluctuations (that is, the temperature fluctuation rate exceeds the preset threshold), the judgment of the heat source will be revised to solar radiation. The preset fluctuation threshold can be set based on actual experience or experimental data to distinguish between temperature fluctuations caused by environmental factors (such as solar radiation, cloud cover, wind changes) and relatively stable heating caused by internal electrical faults.

[0091] This embodiment effectively overcomes the shortcomings of relying solely on static image features to determine the source of heat by analyzing the temperature fluctuation characteristics of hot spot areas. Specifically, the temperature of hot spots caused by solar radiation is often affected by factors such as sunlight intensity, cloud cover, and ambient wind speed, exhibiting significant temperature fluctuations. In contrast, the temperature of heat caused by internal electrical faults is generally relatively stable or exhibits a continuous upward trend during the duration of the fault, with less fluctuation. Therefore, by acquiring a sequence of infrared thermal images and quantitatively analyzing the temperature fluctuation amplitude in the core area of ​​the hot spot, important temporal dimension information can be provided for determining the source of heat. When a hot spot is detected to have a high temperature fluctuation rate, even if its static characteristics (such as edge sharpness, shape regularity, and edge overlap) are similar to internal electrical heating, it can be more accurately attributed to solar radiation based on its dynamic characteristics, thus avoiding misjudgment.

[0092] To more clearly illustrate this technical solution, a specific example is provided below. Suppose, during an inspection, the system analyzes infrared thermal images of a target hardware fixture and, based on the sharpness of the hot spot's boundaries, its regular shape, and the degree of boundary overlap with the shadow in the visible light image, preliminarily determines that the heat source of a particular hot spot is "internal electrical heating." However, to further verify the accuracy of this determination, the system activates a temperature fluctuation analysis module. This module continuously acquires a sequence of infrared thermal images of the hardware fixture, for example, one image every five minutes for 30 minutes. During this period, the system tracks the temperature changes of the pixels within the core area of ​​the hot spot. If the temperature fluctuation rate of the hot spot's core area (for example, by calculating the average standard deviation of the temperature of the pixels within the core area) is found to be greater than a preset fluctuation threshold (for example, 2°C), the system will update the heat source determination to "sunlight," even if the previous determination was "internal electrical heating." This correction mechanism can effectively prevent temporary high temperatures caused by sunlight from being misjudged as electrical faults, thereby avoiding unnecessary on-site inspections and power outages, saving manpower and material costs, and improving the reliability of transmission and distribution line operations.

[0093] Through the above technical solution, this embodiment can significantly improve the accuracy of heat source identification, especially in distinguishing between solar radiation and internal electrical heating. This helps reduce unnecessary maintenance and resource waste caused by misjudgment, improves the accuracy of fault diagnosis, and thus optimizes the operation and maintenance efficiency of power transmission and distribution lines, ensuring the safe and stable operation of the power system.

[0094] The beneficial effects of implementing the embodiments of the present invention include: the embodiments of the present application first obtain an infrared thermal image of the target hardware, and then perform hot spot separation on the infrared thermal image according to a preset separation temperature threshold to obtain a hot spot area, and then perform boundary sharpness evaluation on the hot spot area to obtain a boundary sharpness evaluation result, and perform shape regularity analysis on the hot spot area to obtain a shape regularity analysis result, and finally determine the heat source of the hot spot area based on the boundary sharpness evaluation result and the shape regularity analysis result, so that the heat source can be analyzed in combination with the boundary sharpness and shape regularity to achieve hardware temperature anomaly control, thereby improving the accuracy of temperature anomaly detection and effectively controlling the iron accessories hardware to maintain a normal state.

[0095] like Figure 2 As shown, an embodiment of the present invention further provides a temperature abnormality control system for iron accessories of power transmission and distribution lines, comprising: Image acquisition module 401, used to acquire infrared thermal images of target hardware; The hot spot identification module 402 is used to separate hot spots from the infrared thermal image according to a preset separation temperature threshold to obtain hot spot areas; A boundary sharpness evaluation module 403 is used to evaluate the boundary sharpness of the hot spot area and obtain a boundary sharpness evaluation result; A shape regularity analysis module 404 is used to perform shape regularity analysis on the hot spot area to obtain a shape regularity analysis result; The hot spot source determination module 405 is used to determine the heat source of the hot spot area based on the boundary sharpness evaluation result and the shape regularity analysis result.

[0096] The contents of the above method embodiments are all applicable to the present system embodiments. The functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0097] The embodiments described in the embodiments of this application are intended to more clearly illustrate the technical solutions of the embodiments of this application and do not constitute a limitation on the technical solutions provided by the embodiments of this application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

Claims

1. A method for controlling abnormal temperature of iron accessories of power transmission and distribution lines, characterized in that: The following steps are involved: Acquire infrared thermal images of target hardware; performing hot spot separation on the infrared thermal image according to a preset separation temperature threshold to obtain a hot spot area; Performing a boundary sharpness evaluation on the hot spot area to obtain a boundary sharpness evaluation result; Performing shape regularity analysis on the hot spot area to obtain a shape regularity analysis result; The heat source of the hot spot area is determined according to the boundary sharpness evaluation result and the shape regularity analysis result.

2. The method according to claim 1, characterized in that The performing boundary sharpness evaluation on the hot spot area to obtain a boundary sharpness evaluation result includes: Select a pixel point from the boundary of the hot spot area as a target pixel point; Constructing a temperature difference analysis window, wherein the temperature difference analysis window is centered on the target pixel point; Calculate the temperature difference between each pixel in the temperature difference analysis window and the target pixel; averaging the absolute values ​​of the plurality of temperature differences to obtain a local temperature change rate; If the local temperature change rate is greater than a preset temperature gradient threshold, determining that the boundary sharpness evaluation result is high sharpness; If the local temperature change rate is less than a preset temperature gradient threshold, the boundary sharpness evaluation result is determined to be low sharpness.

3. The method according to claim 1, characterized in that The performing of shape regularity analysis on the hot spot area to obtain a shape regularity analysis result includes: Extracting morphological features of the hot spot area and calculating the area and perimeter of the hot spot area; Calculating circularity based on the area and the perimeter; Calculating the aspect ratio of the minimum circumscribed rectangle of the hot spot area; If the roundness is greater than a preset roundness threshold and the aspect ratio is greater than a preset aspect ratio threshold, the shape regularity analysis result is determined to be high regularity; otherwise, the shape regularity analysis result is determined to be low regularity.

4. The method according to claim 1, wherein The determining the heat source of the hot spot area according to the boundary sharpness evaluation result and the shape regularity analysis result includes: Collecting a visible light image of the target hardware, where the timestamp of the visible light image is the same as the timestamp of the infrared thermal image; Performing contour extraction on the infrared thermal image to obtain a hot spot boundary contour; Performing contour extraction on the visible light image to obtain a shadow boundary contour; Calculating a degree of boundary overlap based on the hot spot boundary outline and the shadow boundary outline; The heat source of the hot spot area is determined according to the boundary sharpness evaluation result, the shape regularity analysis result and the boundary overlap.

5. The method according to claim 4, characterized in that The step of extracting the contour of the infrared thermal image to obtain the hot spot boundary contour includes: Performing Gaussian smoothing on the infrared thermal image to obtain a Gaussian smoothed image; Calculating the gradient strength and direction of the Gaussian smoothed image; According to the gradient strength and direction, the boundary of the Gaussian smoothed image is refined by a double threshold suppression method to obtain the hot spot boundary contour.

6. The method according to claim 4, characterized in that The step of extracting the contour of the visible light image to obtain the shadow boundary contour includes: Performing color space conversion on the visible light image so as to convert the visible light image from an RGB color space to an HSV color space, wherein parameters of the HSV color space include hue, saturation, and brightness; Binarizing the brightness component of the visible light image after color space conversion by a Gaussian adaptive threshold method to obtain a binary image, so as to separate the shadow area from the non-shadow area; Boundaries of the binary image are extracted using an edge detection algorithm to obtain the shadow boundary contour.

7. The method according to claim 6, characterized in that The binarization process of the brightness component of the visible light image after the color space conversion by the Gaussian adaptive threshold method to obtain a binarized image includes: Acquire a brightness change region of the brightness component from the visible light image after color space conversion; Extracting the brightness value range of the brightness change area; Calculating the brightness distribution of the brightness value range; calculating the skewness and kurtosis of the brightness distribution; Determining an upper limit of the brightness value according to the skewness and kurtosis; updating the brightness value range according to the brightness value upper limit; Determining a shadow boundary brightness threshold according to the updated brightness value range; The brightness component is binarized according to the shadow boundary brightness threshold to obtain a binarized image.

8. The method according to claim 4, characterized in that The determining the heat source of the hot spot area according to the boundary sharpness evaluation result, the shape regularity analysis result and the boundary overlap includes: If the boundary sharpness evaluation result is low sharpness and the shape regularity analysis result is low regularity, determining that the heat source is solar radiation; If the boundary sharpness evaluation result is high sharpness and the shape regularity analysis result is high regularity, determining whether the boundary overlap is greater than a preset overlap threshold; If the boundary overlap is greater than a preset overlap threshold, determining that the heat source is an environmental thermal artifact; If the boundary overlap is less than a preset overlap threshold, it is determined that the heat source is internal electrical heating.

9. The method according to claim 1, characterized in that The method further comprises: Acquire infrared thermal image sequences; Selecting an image from the infrared thermal image sequence as an image to be processed; Extracting a temperature peak from the image to be processed as a hot spot center point; Constructing a hot spot core area according to a preset pixel side length and the center point of the hot spot; Select a pixel point from the hot spot core area as a pixel point to be processed; Combining the temperature values ​​corresponding to each image of the pixel to be processed in the infrared thermal image sequence according to the coordinate information of the pixel to be processed to obtain a temperature value sequence; Calculating an average temperature based on the temperature value sequence; Calculating a standard deviation as a temperature fluctuation amplitude based on the temperature value sequence and the average temperature; Performing statistical analysis on the temperature fluctuation amplitude corresponding to each pixel point in the hot spot core area to obtain a temperature fluctuation rate; If the temperature fluctuation rate is greater than a preset fluctuation threshold and the heat source is internal electrical heating, the heat source is updated to solar radiation.

10. A temperature abnormality control system for iron accessories of power transmission and distribution lines, characterized in that: include: An image acquisition module is used to acquire infrared thermal images of target hardware; a hot spot identification module, configured to separate hot spots from the infrared thermal image according to a preset separation temperature threshold to obtain a hot spot area; A boundary sharpness evaluation module is used to evaluate the boundary sharpness of the hot spot area and obtain a boundary sharpness evaluation result; A shape regularity analysis module is used to perform shape regularity analysis on the hot spot area to obtain a shape regularity analysis result; The hot spot source judgment module is used to determine the heat source of the hot spot area based on the boundary sharpness evaluation result and the shape regularity analysis result.

Images