Electrician instrument precision monitoring method
Through deep learning models, the temperature difference captured by infrared thermal imager and dynamically adjust the color scale range is solved, and the temperature monitoring problem is inaccurate due to the fixed color scale range of infrared thermal imager, achieving more accurate fault detection and equipment status monitoring .
Patent Information
- Application Number
- CN202411200350.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-29
- Publication Date
- 2024-11-29
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
When existing infrared thermal cameras map electrical signals to color or grayscale, the fixed color scale range may be too wide or too narrow, resulting in inaccurate temperature monitoring accuracy and inaccurate temperature changes or misjudgment in the equipment in time. Abnormal temperatures lead to equipment failure delays or unnecessary maintenance.
Through deep learning models, the temperature difference on the surface of objects is analyzed in real time, the changes in high span, normal and low spans are divided, and the color scale range is dynamically adjusted to ensure accurate monitoring of temperature changes and avoid important temperature fluctuations being masked or misjudged.
It improves the accuracy and reliability of fault detection, ensures that the equipment operates within the safe temperature range, avoids equipment failures and misjudgment, and significantly improves the accuracy of temperature monitoring.
Smart Images

Figure CN119043499A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of electrician instrument precision monitoring, and in particular to an electrician instrument precision monitoring method. Background Art
[0002] Electrician instrument accuracy monitoring refers to the regular or irregular accuracy testing of various instruments and equipment used by electricians (such as voltmeters, ammeters, insulation resistance meters, infrared thermal imagers, etc.) to ensure the accuracy and reliability of their measurement results. This work usually includes calibration, testing and adjustment of instruments to ensure that they comply with relevant standards and technical specifications to avoid power system failures or safety hazards caused by measurement errors.
[0003] As one of the electrical instruments, the main function of the infrared thermal imager is to detect the temperature distribution of electrical equipment through infrared imaging technology. It can obtain thermal images of the equipment surface without contact, helping electricians quickly find possible overheating problems in electrical equipment, such as poor contact, overload, insulation degradation, etc. By timely detecting and analyzing these thermal anomalies, electrical failures and fire risks can be prevented, and the safety and reliability of the power system can be improved.
[0004] The working principle of infrared thermal imagers is based on thermal radiation and infrared imaging technology. Any object, as long as its temperature is higher than absolute zero (-273.15°C), will radiate energy in the form of electromagnetic waves, including infrared rays that are invisible to the human eye. Infrared thermal imagers capture these infrared radiations through built-in detectors and convert them into electrical signals. The following is its detailed working principle: Infrared radiation capture: The core component of the infrared thermal imager is the infrared detector, which can capture the infrared radiation emitted by the surface of an object. Objects of different temperatures radiate different infrared energies. The higher the temperature, the greater the radiated energy.
[0005] Signal processing: The detector converts the captured infrared radiation energy into an electrical signal. Due to the temperature difference at different locations on the surface of the object, the generated electrical signal will also be different.
[0006] Image generation: After signal processing, these electrical signals are mapped into pixels of different colors or grayscales, and finally a two-dimensional image representing the temperature distribution of the object is generated. This image is usually called a "thermal image" or "thermal map", in which different colors or grayscales represent different temperature ranges.
[0007] By monitoring the thermal images generated by infrared thermal imagers, the temperature distribution information of electrical equipment can be obtained in real time. Infrared thermal imagers can capture the temperature of different parts of the equipment surface and convert this temperature information into a visual thermal image. Areas of different colors or grayscales represent different temperature ranges, so that electricians can intuitively observe the temperature distribution of the equipment and quickly identify possible overheating areas or temperature anomalies, so as to carry out timely maintenance and treatment to avoid potential equipment failures and safety issues.
[0008] The prior art has the following deficiencies: When mapping electrical signals to colors or grayscales, infrared thermal imagers in the prior art usually set the color scale range in a fixed manner. Due to differences in temperature changes of different electrical equipment, the fixed color scale range may cause the color scale range to be set too wide or too narrow. Setting the color scale range too wide or too narrow will lead to inaccurate temperature monitoring accuracy and cause serious consequences. The following is a specific explanation: A wide color scale range means a large temperature variation span. When the color scale range is too wide, subtle temperature changes cannot be clearly presented in the image, temperature differences may be masked, and some important areas of the equipment with small temperature differences (such as local overheating or uneven cooling) may not be detected. Because subtle temperature changes are not obvious, early equipment failures (such as poor contact or insulation damage) may not be identified in time, delaying maintenance and repair time, leading to more serious failures; Too narrow a color scale range means a smaller temperature range is covered. When the color scale range is too narrow, small temperature differences in the thermal image may be over-amplified and displayed as larger color or grayscale differences. Users may misjudge normal temperature fluctuations as abnormal temperatures, resulting in unnecessary maintenance or intervention. Extremely high or low temperature areas may be compressed to the limit of the color scale and displayed as a single color, resulting in the loss of accurate temperature monitoring of these areas. This may lead to the inability to accurately judge the extreme values of temperature, thereby ignoring potential serious problems.
[0009] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention
[0010] The purpose of the present invention is to provide an electrical instrument accuracy monitoring method, which analyzes the temperature difference on the surface of an object in real time, uses a deep learning model to effectively divide the temperature changes of high span, normal span and low span, and timely identifies the temperature anomaly of the equipment to improve the accuracy and reliability of fault detection. In addition, the system dynamically adjusts the color code range according to the temperature difference, flexibly widens or narrows the range, accurately monitors the temperature change, ensures that the high temperature information is not compressed, and the subtle temperature change is not ignored, so as to more accurately reflect the operating status of the equipment and avoid the important temperature fluctuations from being covered up, so as to solve the problems in the above-mentioned background technology.
[0011] In order to achieve the above object, the present invention provides the following technical solution: a method for monitoring the accuracy of an electrical instrument, comprising the following steps: The infrared radiation emitted by the surface of the object is captured by an infrared detector, the captured radiation is converted into an electrical signal, and the electrical signal is converted into digital data representing the surface temperature of the object; Process the real-time temperature information obtained from the surface of the object, extract the key features reflecting the temperature difference, input the extracted key features into the pre-trained deep learning model for predictive analysis, and evaluate the temperature difference on the surface of the object; Based on the prediction results of the deep learning model, the temperature difference on the surface of the object is analyzed in real time, and the temperature difference on the surface of the object is divided into high span change, normal span change and low span change; Set the initial color scale range to color map the temperature difference of normal span change; In view of the temperature difference between high span change and low span change, the actual color scale range under high span change and low span change is widened and narrowed through the initial color scale range, so as to accurately monitor the different temperature differences of electrical equipment.
[0012] Preferably, the captured radiation is converted into an electrical signal, and the electrical signal is converted into digital data representing the surface temperature of the object through a built-in processing system. The specific steps are as follows: The infrared detector of the infrared camera captures the infrared radiation emitted by the surface of the object; The received infrared radiation energy is converted into electrical signals through the photosensitive material inside the detector; The electrical signal is amplified by the amplifier to improve the availability of the signal. At the same time, during the amplification process, the signal is filtered to remove noise and other interference; The amplified and filtered analog electrical signal is converted into a digital signal through an analog-to-digital converter. Next, the digital signal is further processed by the processing system to ultimately generate digital data that can accurately reflect the surface temperature of the object.
[0013] Preferably, the extracted key features reflecting the temperature difference include heat flux and temperature gradient on the surface of the object, where the heat flux represents the heat flow rate through a unit area, and the temperature gradient represents the temperature change rate over a unit distance.
[0014] Preferably, the temperature gradient and heat flux of the object surface are extracted from the real-time temperature information obtained from the object surface, and the specific steps are as follows: Obtain real-time temperature distribution data on the surface of an object from an infrared thermal imager; Based on the acquired temperature distribution data, the temperature gradient on the surface of the object is calculated. The temperature gradient refers to the rate of change of temperature with distance, which is obtained by numerically differentiating the temperature data. After obtaining the temperature gradient, the heat flux on the surface is calculated by combining it with the thermal conductivity of the object material. Heat flux refers to the amount of heat flowing through a unit area per unit time. It is expressed by Fourier's law as heat flux density equals thermal conductivity multiplied by temperature gradient. The calculated heat flux data is visualized to generate a heat flux distribution map, which can show the flow of heat on the surface of the object.
[0015] Preferably, after performing an abnormal analysis on the heat flux and temperature gradient of the extracted object surface, a heat flux difference index and a temperature gradient difference index are generated respectively, and the heat flux difference index and the temperature gradient difference index obtained after the analysis are input into a pre-trained deep learning model for predictive analysis to generate a temperature difference coefficient, and the temperature difference on the object surface is evaluated by the temperature difference coefficient.
[0016] Preferably, the local heat flux change rate is calculated from the heat flux data extracted from the surface of the object. The local heat flux change rate is obtained by analyzing the heat flux change relationship between each pixel and its adjacent pixels. The calculation expression is: , In the formula, represents the rate of change of local heat flux, Indicates location The heat flux value, Indicates location The heat flux value, Indicates location The surrounding neighborhood, Indicates location and the location in the neighborhood The Euclidean distance between , , is the adjustment parameter for adjusting the sensitivity of heat flux difference; The temperature gradient data on the surface of the object is processed to calculate the local temperature gradient change rate. The calculation expression is: , represents the rate of change of local temperature gradient, Indicates location The temperature gradient value, Indicates location The temperature gradient value, , , Represents the tuning parameter that controls the response to temperature gradient differences.
[0017] Preferably, when obtaining the local heat flux change rate Afterwards, the heat flux difference index is calculated by the local heat flux change rate, and the calculation expression is: , In the formula, Indicates location The Laplace operator of heat flux represents the second-order derivative of heat flux, reflecting the curvature change of heat flux. represents the heat flux difference index, is a parameter to adjust the sensitivity of the heat flux difference index; The local temperature gradient change rate , calculate the temperature gradient difference index, the calculation expression is: , In the formula, Indicates location The temperature gradient Laplace operator represents the second-order derivative of the temperature gradient and reflects the curvature of the temperature change. represents the temperature gradient difference index, Represents the parameter that adjusts the exponential sensitivity of the temperature gradient difference.
[0018] Preferably, the temperature difference coefficient generated after analyzing the real-time temperature information obtained from the surface of the object is compared and analyzed with a preset first temperature difference coefficient reference threshold and a second temperature difference coefficient reference threshold, wherein the first temperature difference coefficient reference threshold is less than the second temperature difference coefficient reference threshold, and then the temperature difference on the surface of the object is divided, and the specific steps of division are as follows: If the temperature difference coefficient is less than the first temperature difference coefficient reference threshold, classifying the temperature difference on the surface of the object as a low span change; If the temperature difference coefficient is greater than or equal to the first temperature difference coefficient reference threshold and less than the second temperature difference coefficient reference threshold, the temperature difference on the surface of the object is classified as a normal span change; If the temperature difference coefficient is greater than or equal to the second temperature difference coefficient reference threshold, the temperature difference on the surface of the object is classified as a high span change.
[0019] Preferably, for the temperature difference between the high span change and the low span change, the actual color scale range under the high span change and the low span change is widened and narrowed by the initial color scale range, and the specific steps are as follows: According to the temperature difference coefficient extracted from the object surface, the control factor is calculated to determine the adjustment direction and amplitude of the actual color scale range. The calculation expression is: , represents the actual measured temperature difference coefficient, represents the first temperature difference coefficient reference threshold, represents the second temperature difference coefficient reference threshold, represents regulatory factors; The calculated regulatory factors , adjust the actual color scale range, the specific expression of adjustment is: , , where Indicates the initial color scale range, Indicates the lower limit of the initial color scale range, Indicates the upper limit of the initial color scale range, Indicates the lower limit of the adjusted color scale, reflecting the actual lowest color value after widening or narrowing. Indicates the upper limit of the adjusted color scale, reflecting the actual highest color value after widening or narrowing. and Respectively represent the control coefficients for narrowing and widening the initial color scale range under low span change and high span change, and adjust the width of the actual color scale range; The adjusted color scale range and Applied to the color mapping process of thermal images to generate more accurate temperature distribution maps, where is the actual color scale range under low span change, is the actual color scale range under high span change.
[0020] In the above technical solution, the technical effects and advantages provided by the present invention are: The present invention analyzes the temperature difference of the surface of the object in real time. The system can quickly identify and evaluate the temperature changes in different areas. Based on the prediction results of the deep learning model, it can effectively divide the temperature changes of high span, normal span and low span, and detect the temperature anomaly of the equipment in real time. This real-time analysis capability enables the system to identify possible fault areas at an early stage, such as local overheating or uneven cooling, so that maintenance measures can be taken in time to avoid potential equipment failures, greatly improving the accuracy and reliability of equipment monitoring and ensuring that the equipment can operate within a safe temperature range.
[0021] The present invention uses the initial color scale range to dynamically adjust the actual color scale range according to the temperature difference between high span changes and low span changes. The system can flexibly widen or narrow the color scale range to accurately monitor different temperature differences. When the temperature difference is large, the system can prevent the temperature information in the high temperature area from being compressed by widening the color scale range, thereby accurately reflecting the extreme temperature conditions; when the temperature difference is small, the system can amplify subtle temperature changes by narrowing the color scale range to ensure that key temperature changes are not ignored. This dynamic adjustment mechanism significantly improves the accuracy of temperature monitoring, can better reflect the actual operating status of the equipment, and avoids important temperature fluctuations from being concealed. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.
[0023] Figure 1 The present invention is a method flow chart of an electrical instrument accuracy monitoring method. DETAILED DESCRIPTION
[0024] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.
[0025] The present invention provides Figure 1 The method for monitoring the accuracy of an electrical instrument shown comprises the following steps: The infrared radiation emitted by the surface of the object is captured by an infrared detector, the captured radiation is converted into an electrical signal, and the electrical signal is converted into digital data representing the surface temperature of the object; The captured radiation is converted into an electrical signal, which is then converted into digital data representing the surface temperature of the object through a built-in processing system. The specific steps are as follows: The infrared detector of the infrared camera captures the infrared radiation emitted by the surface of the object; These infrared radiations are electromagnetic waves emitted by objects when their temperature is above absolute zero. Infrared detectors usually use technologies such as focal plane arrays (FPA) or microbolometers to receive these radiant energies. The core part of the detector is a sensitive array of photosensitive elements that can sense and receive infrared radiation of different wavelengths separately. The key to this process lies in the sensitivity of the detector, which determines whether the infrared thermal imager can accurately capture tiny temperature changes.
[0026] The received infrared radiation energy is converted into electrical signals through the photosensitive material inside the detector; Specifically, when infrared radiation hits the photosensitive elements on the detector surface, it causes a change in the charge distribution or resistance in these elements, generating an electrical signal proportional to the radiation intensity. For example, when a microwave bolometer is exposed to infrared radiation, the temperature change of the material causes its resistance to change, and this resistance change is converted into a corresponding voltage or current signal. The key to this step is to accurately convert the energy of infrared radiation into an electrical signal that can be processed.
[0027] The electrical signal is amplified by an amplifier (since the electrical signal obtained by the initial conversion is usually very weak) to improve the availability of the signal. At the same time, during the amplification process, the signal is filtered to remove noise and other interference; Filters are usually used to eliminate high-frequency or low-frequency noise introduced by the detector or the external environment to ensure the purity of the signal. After amplification and filtering, the electrical signal becomes clearer and more stable, and can accurately reflect the infrared radiation intensity of the object surface. The accuracy of amplification and filtering is crucial because it directly affects the accuracy of the final temperature data.
[0028] The amplified and filtered analog electrical signal is converted into a digital signal through an analog-to-digital converter (ADC). The analog-to-digital converter converts a continuous voltage signal into discrete digital values, which correspond to different temperatures on the surface of the object. Next, the digital signal is further processed by the processing system (such as calibration, temperature compensation and linearization) to finally generate digital data that can accurately reflect the surface temperature of the object. This digital data can be used to generate thermal images or perform temperature analysis, helping users to intuitively observe and analyze the temperature distribution of objects. The accuracy of analog-to-digital conversion and digital signal processing directly determines the measurement accuracy and image quality of the infrared thermal imager.
[0029] Process the real-time temperature information obtained from the surface of the object, extract the key features reflecting the temperature difference, input the extracted key features into the pre-trained deep learning model for predictive analysis, and evaluate the temperature difference on the surface of the object; The key features extracted to reflect temperature differences include heat flux and temperature gradient on the surface of the object. Heat flux represents the rate of heat flow through a unit area, and temperature gradient represents the rate of temperature change per unit distance.
[0030] The temperature gradient and heat flux of the object surface are extracted from the real-time temperature information obtained from the object surface. The specific steps are as follows: Obtain real-time temperature distribution data on the surface of an object from an infrared thermal imager; Infrared detectors convert infrared radiation from different locations on the surface of an object into electrical signals and generate temperature data through processing. These data represent the temperature distribution of the surface in the form of a two-dimensional matrix. The core of this step is to obtain high-resolution temperature data to ensure that subtle temperature differences can be captured. The measurement accuracy needs to be guaranteed during the acquisition process, taking into account factors such as the material properties of the object surface, the ambient temperature, and the measurement angle.
[0031] Based on the acquired temperature distribution data, the temperature gradient on the surface of the object is calculated. The temperature gradient refers to the rate of change of temperature with distance, which is obtained by numerically differentiating the temperature data. In practice, the temperature gradient can be calculated by calculating the temperature difference of adjacent points around each pixel. The direction and magnitude of the temperature gradient can be represented in the form of a vector field, indicating the direction and intensity of heat flow. The key to this step is to accurately calculate the temperature gradient to ensure that significant temperature change areas on the surface can be identified.
[0032] After obtaining the temperature gradient, the heat flux on the surface is calculated by combining it with the thermal conductivity of the object material. Heat flux refers to the amount of heat flowing through a unit area per unit time. It is expressed by Fourier's law as heat flux density equals thermal conductivity multiplied by temperature gradient. Thermal conductivity is an inherent property of a material, indicating the thermal conductivity of the material. Therefore, when calculating heat flux, the corresponding thermal conductivity must be selected according to the different material properties of the surface of the object. The accuracy of this step depends on the accurate understanding of the material properties and the correct calculation of the temperature gradient.
[0033] Visualize the calculated heat flux data to generate a heat flux distribution map, which can show the flow of heat on the surface of the object; The heat flux distribution map can be used to visually identify areas where heat flux is concentrated, heat flow paths, and possible heat accumulation points. The purpose of this step is to help users better understand the thermal behavior of the surface of an object through image presentation and identify possible heat dissipation problems or abnormal hot spots. This step can also be combined with other temperature-related data for further analysis to make more comprehensive thermal management decisions.
[0034] Under the detection window, after abnormal analysis of the heat flux and temperature gradient on the extracted object surface, the heat flux difference index and temperature gradient difference index are generated respectively. The heat flux difference index and temperature gradient difference index obtained after analysis are input into the pre-trained deep learning model for predictive analysis to generate the temperature difference coefficient, and the temperature difference on the object surface is evaluated by the temperature difference coefficient.
[0035] From the heat flux data extracted from the surface of the object, the local heat flux change rate is calculated. The local heat flux change rate is obtained by analyzing the heat flux change relationship between each pixel and its adjacent pixels. The calculation expression is: , In the formula, represents the rate of change of local heat flux, Indicates location The heat flux value, Indicates location The heat flux value, Indicates location The surrounding neighborhood (usually a 3x3 or 5x5 neighborhood window), Indicates location and the location in the neighborhood The Euclidean distance between , , are the adjustment parameters for adjusting the sensitivity of heat flux differences. By adjusting these parameters, the response to different heat flux changes can be enhanced or weakened. , , The specific size is not limited here and can be set according to actual conditions and needs.
[0036] The local heat flux rate of change formula calculates the difference in heat flux between each pixel and its neighborhood, and takes into account the distance factor, so that distant points have less impact on the rate of change. The purpose of this step is to generate a distribution map of the local heat flux rate of change to provide basic data for subsequent difference index calculations.
[0037] The temperature gradient data on the surface of the object is processed to calculate the local temperature gradient change rate. The calculation expression is: , represents the rate of change of local temperature gradient, Indicates location The temperature gradient value, Indicates location The temperature gradient value, , , Represents the adjustment parameters that control the temperature gradient difference response. By adjusting these parameters, the response to different temperature gradient changes can be enhanced or weakened. Similarly, , , The specific size is not limited here and can be set according to actual conditions and needs; The local temperature gradient change formula is applied to the temperature gradient, taking into account the local change and spatial distribution characteristics of the temperature gradient, so that subtle temperature change trends can be detected. The purpose of this step is to obtain a distribution map of the temperature gradient change on the surface of the object.
[0038] In obtaining the local heat flux change rate Afterwards, the heat flux difference index is calculated by the local heat flux change rate, and the calculation expression is: , In the formula, Indicates location The Laplace operator of heat flux represents the second-order derivative of heat flux, reflecting the curvature change of heat flux. The heat flux difference index is generated by combining the local heat flux change rate with the Laplace operator of the heat flux. The Laplace operator is used to capture the mutation point of the heat flux. The influence of the gently changing area is suppressed, making the heat flux difference index more sensitive to local thermal anomalies. The parameters for adjusting the sensitivity of the heat flux difference index are not specifically limited here.
[0039] The local temperature gradient change rate , calculate the temperature gradient difference index, the calculation expression is: , In the formula, Indicates location The temperature gradient Laplace operator represents the second-order derivative of the temperature gradient and reflects the curvature of the temperature change. represents the temperature gradient difference index, The parameter for adjusting the sensitivity of the temperature gradient difference index is not specifically limited here. The calculation formula of the temperature gradient difference index combines the temperature gradient change rate and the temperature gradient Laplace operator to generate the temperature gradient difference index, which can keenly capture local anomalies of temperature changes, especially sensitive to temperature mutation points and areas with uneven temperature distribution.
[0040] Within the detection window, the larger the performance value of the heat flux difference index and the larger the performance value of the temperature gradient difference index generated after the abnormal analysis of the heat flux and temperature gradient of the extracted object surface, the greater the temperature difference on the object surface. Specifically, a larger heat flux difference index means that the heat flow in certain areas of the object surface is significantly different, and there may be local overheating or uneven cooling. A larger temperature gradient difference index indicates that the speed and amplitude of temperature change are significantly different in different areas, which is also usually related to significant temperature non-uniformity or temperature mutation. Conversely, if the performance values of the heat flux difference index and the temperature gradient difference index are smaller, it indicates that the temperature distribution on the object surface is more uniform, the temperature difference is smaller, and the changes in heat flow and temperature gradient are relatively gentle.
[0041] The machine learning model is not specifically limited here, and can achieve the heat flux difference index and temperature gradient difference index Perform comprehensive analysis to generate temperature difference coefficient In order to realize the technical solution of the present invention, the present invention provides a specific implementation method; Temperature coefficient of variation The resulting calculation formula is: , where , Heat flux difference index and temperature gradient difference index The preset scaling factor of , Both are greater than 0.
[0042] It can be seen from the temperature difference coefficient that within the detection window, the performance value of the heat flux difference index generated after the abnormal analysis of the heat flux and temperature gradient on the extracted object surface is larger, and the performance value of the generated temperature gradient difference index is larger. That is, the larger the performance value of the generated temperature difference coefficient is, the greater the temperature difference on the object surface is, and vice versa.
[0043] Based on the prediction results of the deep learning model, the temperature difference on the surface of the object is analyzed in real time, and the temperature difference on the surface of the object is divided into high span change, normal span change and low span change; The temperature difference coefficient generated after analyzing the real-time temperature information obtained from the surface of the object is compared and analyzed with the preset first temperature difference coefficient reference threshold and the second temperature difference coefficient reference threshold, wherein the first temperature difference coefficient reference threshold is less than the second temperature difference coefficient reference threshold, and then the temperature difference of the surface of the object is divided, and the specific division steps are as follows: If the temperature difference coefficient is less than the first temperature difference coefficient reference threshold, classifying the temperature difference on the surface of the object as a low span change; If the temperature difference coefficient is greater than or equal to the first temperature difference coefficient reference threshold and less than the second temperature difference coefficient reference threshold, the temperature difference on the surface of the object is classified as a normal span change; If the temperature difference coefficient is greater than or equal to the second temperature difference coefficient reference threshold, the temperature difference on the surface of the object is classified as a high span change.
[0044] Set the initial color scale range to map the temperature difference of normal span change. For the temperature difference of high span change and low span change, widen and narrow the actual color scale range under high span change and low span change through the initial color scale range, and accurately monitor the different temperature differences of electrical equipment. In order to more accurately reflect the temperature distribution, we first need to set an initial color scale range for the "normal span change" area. This means that in the thermal image, this initial color scale range is used to map the temperature differences within the normal range, so that these temperature changes can be accurately presented in the image in the form of color or grayscale, thereby helping users to intuitively identify and analyze the temperature distribution.
[0045] The initial color scale range is a set of color or grayscale value intervals that are pre-set for a specific temperature range when generating a thermal image. This range determines the color changes corresponding to different temperatures in the image. For example, a temperature range may fade from blue (indicating low temperature) to red (indicating high temperature). The initial color scale range is set to ensure that under normal working conditions, temperature changes can be clearly reflected in the thermal image, allowing users to accurately identify and analyze subtle differences in temperature distribution.
[0046] In view of the temperature difference between high span change and low span change, the actual color scale range under high span change and low span change is widened and narrowed by the initial color scale range. The specific steps are as follows: According to the temperature difference coefficient extracted from the object surface, the control factor is calculated to determine the adjustment direction and amplitude of the actual color scale range. The calculation expression is: , represents the actual measured temperature difference coefficient, represents the first temperature difference coefficient reference threshold, represents the second temperature difference coefficient reference threshold, represents regulatory factors; The calculated regulatory factors , adjust the actual color scale range, the specific expression of adjustment is: , , where Indicates the initial color scale range, Indicates the lower limit of the initial color scale range, corresponding to the color value of the lowest temperature. Indicates the upper limit of the initial color scale range, corresponding to the color value of the highest temperature. Indicates the lower limit of the adjusted color scale, reflecting the actual lowest color value after widening or narrowing. Indicates the upper limit of the adjusted color scale, reflecting the actual highest color value after widening or narrowing. and Respectively represent the control coefficients for narrowing and widening the initial color scale range under low span change and high span change, and adjust the width of the actual color scale range; For low span changes (i.e. ), color scale lower limit shrinks to enhance sensitivity to small temperature differences; whereas for large span changes (i.e. ), upper limit of color scale Will expand to accommodate larger temperature differences.
[0047] The adjusted color scale range and Applied to the color mapping process of thermal images to generate more accurate temperature distribution maps, where is the actual color scale range under low span change, is the actual color scale range under high span change; High span change: Color scale upper limit is widened to show more high temperature areas and prevent the details of temperature differences from being compressed.
[0048] Low span change: color scale lower limit is narrowed to better show subtle temperature differences and prevent important temperature fluctuations from being overlooked.
[0049] The present invention analyzes the temperature difference on the surface of an object in real time, and the system can quickly identify and evaluate the temperature changes in different areas. Based on the prediction results of the deep learning model, the temperature changes of high span, normal span and low span can be effectively divided, and the temperature anomalies of the equipment can be detected in real time. This real-time analysis capability enables the system to identify possible fault areas at an early stage, such as local overheating or uneven cooling, so that maintenance measures can be taken in time to avoid potential equipment failures, greatly improving the accuracy and reliability of equipment monitoring and ensuring that the equipment can operate within a safe temperature range.
[0050] The present invention uses the initial color scale range to dynamically adjust the actual color scale range according to the temperature difference between high span changes and low span changes. The system can flexibly widen or narrow the color scale range to accurately monitor different temperature differences. When the temperature difference is large, the system can prevent the temperature information in the high temperature area from being compressed by widening the color scale range, thereby accurately reflecting the extreme temperature conditions; when the temperature difference is small, the system can amplify subtle temperature changes by narrowing the color scale range to ensure that key temperature changes are not ignored. This dynamic adjustment mechanism significantly improves the accuracy of temperature monitoring, can better reflect the actual operating status of the equipment, and avoids important temperature fluctuations from being concealed.
[0051] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.
[0052] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
[0053] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.
Claims
1. A method for monitoring the accuracy of an electrical instrument, characterized in that: The following steps are involved: The infrared radiation emitted by the surface of the object is captured by an infrared detector, the captured radiation is converted into an electrical signal, and the electrical signal is converted into digital data representing the surface temperature of the object; Process the real-time temperature information obtained from the surface of the object, extract the key features reflecting the temperature difference, input the extracted key features into the pre-trained deep learning model for predictive analysis, and evaluate the temperature difference on the surface of the object; Based on the prediction results of the deep learning model, the temperature difference on the surface of the object is analyzed in real time, and the temperature difference on the surface of the object is divided into high span change, normal span change and low span change; Set the initial color scale range to color map the temperature difference of normal span change; In view of the temperature difference between high span change and low span change, the actual color scale range under high span change and low span change is widened and narrowed through the initial color scale range, so as to accurately monitor the different temperature differences of electrical equipment.
2. The method for monitoring the accuracy of an electrician instrument according to claim 1, characterized in that: The captured radiation is converted into an electrical signal, which is then converted into digital data representing the surface temperature of the object through a built-in processing system. The specific steps are as follows: The infrared detector of the infrared camera captures the infrared radiation emitted by the surface of the object; The received infrared radiation energy is converted into electrical signals through the photosensitive material inside the detector; The electrical signal is amplified by an amplifier, and at the same time, the signal is filtered during the amplification process; The amplified and filtered analog electrical signal is converted into a digital signal through an analog-to-digital converter. Next, the digital signal is further processed by the processing system to ultimately generate digital data that can accurately reflect the surface temperature of the object.
3. The method for monitoring the accuracy of an electrician instrument according to claim 1, characterized in that: The key features extracted to reflect temperature differences include heat flux and temperature gradient on the surface of the object. Heat flux represents the rate of heat flow through a unit area, and temperature gradient represents the rate of temperature change per unit distance.
4. The method for monitoring the accuracy of an electrician instrument according to claim 3, characterized in that: The temperature gradient and heat flux on the surface of the object are extracted from the real-time temperature information obtained from the surface of the object. The specific steps are as follows: Obtain real-time temperature distribution data on the surface of an object from an infrared thermal imager; Based on the acquired temperature distribution data, the temperature gradient on the surface of the object is calculated. The temperature gradient refers to the rate of change of temperature with distance, which is obtained by numerically differentiating the temperature data. After obtaining the temperature gradient, the heat flux on the surface is calculated by combining it with the thermal conductivity of the object material. Heat flux refers to the amount of heat flowing through a unit area per unit time. It is expressed by Fourier's law as heat flux density equals thermal conductivity multiplied by temperature gradient. The calculated heat flux data is visualized to generate a heat flux distribution map, which can show the flow of heat on the surface of the object.
5. The method for monitoring the accuracy of an electrician instrument according to claim 4, characterized in that: Under the detection window, after abnormal analysis of the heat flux and temperature gradient on the extracted object surface, the heat flux difference index and temperature gradient difference index are generated respectively. The heat flux difference index and temperature gradient difference index obtained after analysis are input into the pre-trained deep learning model for predictive analysis to generate the temperature difference coefficient, and the temperature difference on the object surface is evaluated by the temperature difference coefficient.
6. The method for monitoring the accuracy of an electrician instrument according to claim 5, characterized in that: From the heat flux data extracted from the surface of the object, the local heat flux change rate is calculated. The local heat flux change rate is obtained by analyzing the heat flux change relationship between each pixel and its adjacent pixels. The calculation expression is: , In the formula, represents the rate of change of local heat flux, Indicates location The heat flux value, Indicates location The heat flux value, Indicates location The surrounding neighborhood, Indicates location and the location in the neighborhood The Euclidean distance between , , is the adjustment parameter for adjusting the sensitivity of heat flux difference; The temperature gradient data on the surface of the object is processed to calculate the local temperature gradient change rate. The calculation expression is: , represents the rate of change of local temperature gradient, Indicates location The temperature gradient value, Indicates location The temperature gradient value, , , Represents the tuning parameter that controls the response to temperature gradient differences.
7. The method for monitoring the accuracy of an electrician instrument according to claim 6, characterized in that: In obtaining the local heat flux change rate Afterwards, the heat flux difference index is calculated by the local heat flux change rate, and the calculation expression is: , In the formula, Indicates location The Laplace operator of heat flux represents the second-order derivative of heat flux, reflecting the curvature change of heat flux. represents the heat flux difference index, is a parameter to adjust the sensitivity of the heat flux difference index; The local temperature gradient change rate , calculate the temperature gradient difference index, the calculation expression is: , In the formula, Indicates location The temperature gradient Laplace operator represents the second-order derivative of the temperature gradient and reflects the curvature of the temperature change. represents the temperature gradient difference index, Represents the parameter that adjusts the exponential sensitivity of the temperature gradient difference.
8. The method for monitoring the accuracy of an electrician instrument according to claim 5, characterized in that: The temperature difference coefficient generated after analyzing the real-time temperature information obtained from the surface of the object is compared and analyzed with the preset first temperature difference coefficient reference threshold and the second temperature difference coefficient reference threshold, wherein the first temperature difference coefficient reference threshold is less than the second temperature difference coefficient reference threshold, and then the temperature difference of the surface of the object is divided, and the specific division steps are as follows: If the temperature difference coefficient is less than the first temperature difference coefficient reference threshold, the temperature difference on the surface of the object is classified as a low span change; If the temperature difference coefficient is greater than or equal to the first temperature difference coefficient reference threshold and less than the second temperature difference coefficient reference threshold, the temperature difference on the surface of the object is classified as a normal span change; If the temperature difference coefficient is greater than or equal to the second temperature difference coefficient reference threshold, the temperature difference on the surface of the object is classified as a high span change.
9. The method for monitoring the accuracy of an electrician instrument according to claim 5, characterized in that: In view of the temperature difference between high span change and low span change, the actual color scale range under high span change and low span change is widened and narrowed by the initial color scale range. The specific steps are as follows: According to the temperature difference coefficient extracted from the object surface, the control factor is calculated to determine the adjustment direction and amplitude of the actual color scale range. The calculation expression is: , represents the actual measured temperature difference coefficient, represents the first temperature difference coefficient reference threshold, represents the second temperature difference coefficient reference threshold, represents regulatory factors; The calculated regulatory factors , adjust the actual color scale range, the specific expression of adjustment is: , , where Indicates the initial color scale range, Indicates the lower limit of the initial color scale range, Indicates the upper limit of the initial color scale range, Indicates the lower limit of the adjusted color scale, reflecting the actual lowest color value after widening or narrowing. Indicates the upper limit of the adjusted color scale, reflecting the actual highest color value after widening or narrowing. and Respectively represent the control coefficients for narrowing and widening the initial color scale range under low span change and high span change, and adjust the width of the actual color scale range; The adjusted color scale range and Applied to the color mapping process of thermal images to generate more accurate temperature distribution maps, where is the actual color scale range under low span change, is the actual color scale range under high span change.
