Electromagnetic pump internal overheating defect positioning device and method
By combining wide-angle and telephoto infrared thermal imagers with temperature measurement and image fusion technology, and integrating three-dimensional temperature field inversion, the problem of accurately locating overheating defects inside the electromagnetic pump was solved, achieving high-precision early warning and diagnosis.
Patent Information
- Application Number
- CN202511770672.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-28
- Publication Date
- 2026-03-06
AI Technical Summary
Existing methods for detecting overheating in electromagnetic pumps are difficult to achieve accurate location and early warning. In particular, due to limitations in the installation location and number of sensors, the accuracy and real-time performance of temperature detection are poor, and they are susceptible to electromagnetic interference, making it impossible to accurately locate the internal heat source.
Wide-angle and long-focus infrared thermal imagers are used for collaborative temperature measurement. Combined with image fusion and three-dimensional temperature field inversion technology, the disparity map is calculated through the SGM matching algorithm. The internal temperature rise of the electromagnetic pump is analyzed using heat transfer models and data-driven algorithms, so as to achieve accurate location of internal hot spots and defect diagnosis.
It improves the detection sensitivity and positioning accuracy of internal overheating defects in electromagnetic pumps, can penetrate complex structures, accurately capture the correlation characteristics between the outer shell and internal heat conduction, realize dynamic diagnosis, and avoid the insufficient spatial resolution and misjudgment of traditional methods.
Smart Images
Figure CN121611615A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of equipment monitoring and diagnostic technology, specifically relating to a device and method for locating internal overheating defects in an electromagnetic pump. Background Technology
[0002] Electromagnetic pumps are widely used in liquid metal cooling systems, aerospace, and nuclear engineering due to their compact structure, lack of moving mechanical parts, and high fluid transport stability. However, during prolonged operation, electromagnetic pumps are prone to internal overheating, which not only affects pump performance but can also lead to equipment failure and, in severe cases, even endanger the safety of the entire system. Internal overheating defects in electromagnetic pumps are typically caused by factors such as the electromagnetic coil, pump body material, and cooling efficiency, and are closely related to the pump's operating conditions.
[0003] Existing methods for detecting overheating in electromagnetic pumps primarily rely on the arrangement of temperature sensors and thermal imaging technology. These methods typically require the installation of multiple sensors, either externally or internally, to monitor the pump's temperature distribution. However, due to the complex internal structure of electromagnetic pumps, the installation location and number of sensors are limited, resulting in poor accuracy and real-time performance in temperature detection. Furthermore, temperature sensors are susceptible to electromagnetic interference, which can lead to measurement errors.
[0004] While infrared thermal imaging technology can provide temperature distribution maps of the pump body surface, its limited two-dimensional spatial representation capabilities result in blind spots and inaccurate hotspot localization, making it impossible to accurately locate heat sources inside the pump. This limitation means that by the time overheating defects become apparent, the equipment may already be damaged. Therefore, existing methods are insufficient for accurately locating and providing early warning of overheating defects inside electromagnetic pumps. To ensure the safe and stable operation of electromagnetic pumps, a technical solution that improves the accuracy and real-time performance of internal overheating defect detection is urgently needed. Summary of the Invention
[0005] The purpose of this invention is to provide a device and method for locating internal overheating defects in electromagnetic pumps, thereby solving the problem that existing detection methods are unable to effectively and accurately locate and provide early warning of internal overheating defects in electromagnetic pumps.
[0006] The technical solution adopted by this invention to solve its technical problem is: a method for locating overheating defects inside an electromagnetic pump, wherein the method includes the following steps:
[0007] Step 1: Large-scale thermal measurement. A wide-angle infrared thermal imager is used to perform a large-scale overall thermal measurement of the electromagnetic pump to obtain the overall thermal distribution information of the electromagnetic pump.
[0008] Step 2: Local thermal measurement. After identifying the thermal anomaly area, a long-focus infrared thermal imager is used to perform local thermal measurement on the overheated area of the electromagnetic pump to obtain detailed temperature data.
[0009] Step 3: Infrared thermal image fusion. Based on the spatial positions of the two thermal imagers, parameter calibration is performed. The image fusion module fuses the image data acquired by the two infrared thermal imagers and calibrates the parameters to correct the spatial positions, forming a unified thermal image and enhancing the display effect of the temperature field. The transformation formula from pixel coordinates to world coordinates is as follows:
[0010]
[0011] M1M2 represents the internal and external parameters of the wide-angle infrared thermal imager and the telephoto infrared thermal imager, respectively. 3×3 and T 3×1 These are the rotation and translation matrices of the camera's external parameters, respectively.
[0012] Step 4: Using the three-dimensional temperature field information obtained in the above steps, the disparity map is calculated by combining the SGM matching algorithm to realize the three-dimensional imaging and positioning of the infrared thermal zone of the electromagnetic pump shell, so as to deduce the internal temperature rise of the electromagnetic pump.
[0013] Step 5: Based on data-driven and physical constraints, derive the spatial inversion relationship between the temperature rise of the electromagnetic pump's casing and its internal temperature rise. This enables the location of hot spots and the diagnosis of thermal defects within the electromagnetic pump. Using heat transfer models and data-driven algorithms, analyze the casing temperature rise data to infer the location of hot spots and the potential degree of thermal defects within the electromagnetic pump. The physical equations are shown below:
[0014]
[0015] Where T is temperature, and φ is the heat source, i.e., the heat generated per unit volume per unit time, with units of W / m³. 3 , where k is the thermal conductivity.
[0016] The above-mentioned method for locating internal overheating defects in an electromagnetic pump involves calculating a disparity map using the SGM matching algorithm. This process consists of four steps: cost calculation, cost aggregation, disparity calculation, and disparity optimization.
[0017] In the above-mentioned method for locating internal overheating defects in an electromagnetic pump, the transformation function for cost calculation is:
[0018]
[0019] Where (x, y) is the coordinate point to be matched, and (x′, y′) are other points in the local region centered at (x, y); I(x, y) is the image gray value of the point to be matched, and I(x′, y′) is the gray value of the surrounding points;
[0020] Cost calculation using Hamming distance:
[0021] C p (x, y, d) = Ham ming(C s (x, y), C s (xd, y))
[0022] This formula calculates the Hamming distance by performing an XOR operation on the two Census transform values, Cs(x, y) and Cs(xd, y).
[0023] The above-mentioned method for locating overheating defects inside an electromagnetic pump, wherein the cost aggregation, the calculation method for any pixel p on path r, is as follows:
[0024]
[0025] Where Lr(p, d) is the matching cost of pixel p with disparity d in direction r; pr represents the previous matching pixel of pixel p in direction r; the first term on the right side of the formula is the initial matching cost of pixel p with disparity d, the second term is the minimum matching cost of the previous pixel pr in direction r, and the third term is used to prevent the matching cost from overflowing due to excessive size.
[0026] In the above-mentioned method for locating internal overheating defects in an electromagnetic pump, the parallax calculation involves selecting the minimum cost from all costs, and the parallax corresponding to this minimum cost is the optimal parallax.
[0027] In the above-mentioned method for locating internal overheating defects in an electromagnetic pump, the parallax optimization involves detecting and processing occlusion points and mismatched points in the image.
[0028] The aforementioned electromagnetic pump internal overheating defect location device includes: a binocular system with a wide-angle infrared thermal imager and a telephoto infrared thermal imager, and a terminal device connected to the binocular system. The terminal device integrates an image fusion module, which is connected to both the wide-angle and telephoto infrared thermal imagers. The image fusion module can fuse and correct the imaging data from the two thermal imagers and transmit it to a three-dimensional imaging module. This module uses the corrected data to realize three-dimensional thermal imaging of the electromagnetic pump casing and transmits it to a connected spatial inversion module. This module can analyze and deduce the temperature rise and defect location inside the electromagnetic pump.
[0029] Compared with the prior art, the beneficial effects of the present invention are as follows: The electromagnetic pump internal overheating defect location device and method provided by the present invention, on the one hand, improves the detection sensitivity and location accuracy of internal overheating defects by using wide-angle and long-focal-length infrared thermal imagers for coordinated temperature measurement, multi-source thermal imaging data fusion and three-dimensional temperature field inversion technology; on the other hand, the device can penetrate the complex electromagnetic pump structure, accurately capture the correlation characteristics of heat conduction between the outer shell and the inside, and combine physical models and data-driven algorithms to realize dynamic diagnosis of thermal defects from the surface to the inside, effectively avoiding the problems of insufficient spatial resolution and misjudgment of traditional single thermal imaging technology. Attached Figure Description
[0030] Figure 1 The diagram shown is a schematic of a module for locating overheating defects inside an electromagnetic pump according to the present invention.
[0031] The reference numerals in the attached diagram are as follows: 1. Wide-angle infrared thermal imager; 2. Telephoto infrared thermal imager; 3. Image fusion module; 4. 3D imaging module; 5. Spatial inversion module. Detailed Implementation
[0032] To address the challenge of existing detection methods failing to accurately locate and provide early warning of internal overheating defects in electromagnetic pumps, this invention provides a device and method for locating internal overheating defects in electromagnetic pumps.
[0033] The method for locating internal overheating defects in the electromagnetic pump includes the following steps:
[0034] Step 1: Large-scale thermal measurement. A large-scale overall thermal measurement of the electromagnetic pump is performed using a wide-angle infrared thermal imager 1 to obtain the overall thermal distribution information of the electromagnetic pump. During this process, the position and angle of the wide-angle infrared thermal imager 1 need to be adjusted to perform thermal imaging of the electromagnetic pump as a whole to obtain the temperature distribution information of its outer shell.
[0035] Step 2: Local thermal measurement. After identifying the thermal anomaly area, the overheated area of the electromagnetic pump is measured locally using a long-focus infrared thermal imager 2 to obtain detailed temperature data.
[0036] Step 3: Infrared thermal image fusion. Based on the spatial positions of the two thermal imagers with different angles and resolutions, parameter calibration is performed. The image fusion module 3 fuses the image data acquired by the two infrared thermal imagers and calibrates the parameters to correct the spatial position, forming a unified thermal image and enhancing the display effect of the temperature field. The transformation formula from pixel coordinates to world coordinates is as follows:
[0037]
[0038] M1M2 represents the internal and external parameters of the wide-angle infrared thermal imager 1 and the telephoto infrared thermal imager 2, respectively.3×3 and T 3×1 These are the rotation and translation matrices of the camera's external parameters, respectively.
[0039] Step 4: Using the three-dimensional temperature field information obtained in the above steps, the disparity map is calculated by combining the SGM matching algorithm to realize the three-dimensional imaging and positioning of the infrared thermal zone of the electromagnetic pump shell, so as to deduce the internal temperature rise of the electromagnetic pump and locate the thermal anomaly area of the electromagnetic pump shell.
[0040] The disparity map is calculated using the SGM matching algorithm, which consists of four steps: cost calculation, cost aggregation, disparity calculation, and disparity optimization.
[0041] The transformation function for cost calculation is:
[0042]
[0043] Where (x, y) is the coordinate point to be matched, and (x′, y′) are other points in the local region centered at (x, y); I(x, y) is the image gray value of the point to be matched, and I(x′, y′) is the gray value of the surrounding points;
[0044] Then, the cost is calculated using Hamming distance:
[0045] C p (x, y, d) = Ham ming(C s (x, y), C s (xd, y))
[0046] This formula calculates the Hamming distance by performing an XOR operation on the two Census transform values, Cs(x, y) and Cs(xd, y).
[0047] The cost aggregation, where any pixel p is calculated on path r, is as follows:
[0048]
[0049] Where Lr(p, d) is the matching cost of pixel p with disparity d in direction r; pr represents the previous matching pixel of pixel p in direction r; the first term on the right side of the formula is the initial matching cost of pixel p with disparity d, the second term is the minimum matching cost of the previous pixel pr in direction r, and the third term is used to prevent the matching cost from overflowing due to excessive size.
[0050] The disparity calculation selects the minimum cost from all costs, and the disparity corresponding to the minimum cost is the optimal disparity.
[0051] Parallax optimization involves detecting and processing occluded points and mismatched points in an image.
[0052] Step 5: Based on data-driven and physical constraints, derive the spatial inversion relationship between the temperature rise of the electromagnetic pump's casing and its internal temperature rise. This enables the location of hot spots and the diagnosis of thermal defects within the electromagnetic pump. Using heat transfer models and data-driven algorithms, analyze the casing temperature rise data to infer the location of hot spots and the potential degree of thermal defects within the electromagnetic pump. The physical equations are shown below:
[0053]
[0054] Where T is temperature, and φ is the heat source, i.e., the heat generated per unit volume per unit time, with units of W / m³. 3 , where k is the thermal conductivity.
[0055] An internal overheating defect location device for an electromagnetic pump includes: a binocular system with a wide-angle infrared thermal imager 1 and a telephoto infrared thermal imager 2, and a terminal device connected to the binocular system. The wide-angle infrared thermal imager 1 features high image resolution and is responsible for large-area thermal measurements of the electromagnetic pump; the telephoto infrared thermal imager 2 features a wide temperature measurement range and is used for localized thermal measurements of the electromagnetic pump. The terminal device integrates an image fusion module 3, which is connected to both the wide-angle infrared thermal imager 1 and the telephoto infrared thermal imager 2. The image fusion module 3 fuses and corrects the imaging data from the two thermal imagers and transmits it to a three-dimensional imaging module 4. This module uses the corrected data to achieve three-dimensional thermal imaging of the electromagnetic pump's outer casing and transmits it to a connected spatial inversion module 5. This module analyzes and derives the temperature information to obtain the internal temperature rise and defect location of the electromagnetic pump.
[0056] In the application, a high-resolution wide-angle infrared thermal imager 1 is first used to perform preliminary thermal imaging of the electromagnetic pump to identify potential overheating areas. Subsequently, a telephoto infrared thermal imager 2 is used to perform detailed measurements of these potential overheating areas to obtain more accurate temperature data. The image fusion module 3, connected to both thermal imagers, fuses the different image data to form a unified temperature field representation, enhancing the overall understanding of the electromagnetic pump's thermal state. Then, the 3D imaging module 4 performs spatial reconstruction of the fused thermal image to generate a 3D temperature distribution map of the electromagnetic pump's casing, helping to accurately locate thermal anomaly areas within the casing. Finally, the spatial inversion module 5, combining data-driven and physical constraint methods, analyzes the casing's thermal data to deduce the internal temperature rise of the electromagnetic pump, enabling the diagnosis of internal hotspot locations and thermal defects.
[0057] It should be noted that the combination of the technical features in the embodiments of the present invention is not limited to the combination methods described in the embodiments of the present invention or the combination methods described in the specific embodiments. All technical features described in the present invention can be freely combined or combined in any way, unless there is a contradiction between them.
[0058] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for locating an internal overheating defect of an electromagnetic pump, characterized in that, The method comprises the following steps: Step one: wide-range thermal measurement, the electromagnetic pump is measured as a whole by a wide-angle end infrared thermal imager (1) to obtain the thermal distribution information of the electromagnetic pump as a whole; Step two: local thermal measurement, after identifying the thermal abnormal area, the electromagnetic pump overheating area is measured locally by a long-focus end infrared thermal imager (2) to obtain detailed temperature data; Step three: infrared thermal imaging picture fusion, according to the spatial positions of the two thermal imagers, the image data obtained by the two infrared thermal imagers are fused and parameter calibrated by an image fusion module (3) to correct the spatial positions, form a unified thermal image, and enhance the display effect of the temperature field, the pixel coordinate to world coordinate transformation formula is as follows: M1M2 are internal and external parameters of the wide-angle end infrared thermal imager (1) and the long-focus end infrared thermal imager (2), R 3×3 and T 3×1 are the rotation matrix and the translation matrix of the camera external parameters, respectively; Step four: according to the three-dimensional temperature field information obtained through the above steps, the three-dimensional imaging and positioning of the electromagnetic pump shell infrared thermal area are realized by combining the SGM matching algorithm to calculate the disparity map, so as to deduce the internal temperature rise of the electromagnetic pump; Step five: according to data driving and physical constraints, the spatial inversion relationship between the electromagnetic pump shell temperature rise and internal temperature rise is deduced to realize the positioning of the electromagnetic pump internal hot spot and the thermal defect degree diagnosis, the heat transfer model and data driving algorithm are used to analyze the shell temperature rise data, and the internal hot spot position and possible thermal defect degree of the electromagnetic pump are deduced, and the physical equation is as follows: Wherein, T is temperature, φ is heat source, i.e. heat generation per unit time per unit volume, unit: W / m 3 k is thermal conductivity.
2. A method of locating internal overheating defects in an electromagnetic pump according to claim 1, characterized in that, The disparity map is calculated by the SGM matching algorithm, and the process includes four steps of cost calculation, cost aggregation, disparity calculation and disparity optimization.
3. A method of locating internal overheating defects in an electromagnetic pump according to claim 2, characterized in that, The transformation function of the cost calculation is: Where (x, y) is the coordinate point to be matched, (x', y') is the other point in the local area centered on (x, y); I(x, y) is the image gray value of the to-be-matched point, and I(x', y') is the gray value of the surrounding point; The cost is calculated by using the Hamming distance: C p (x, y, d) = Hamming(C s (x, y), C s (x-d, y)) The formula is to calculate the Hamming distance after performing XOR operation on the two Census transformation values Cs(x, y) and Cs(x-d, y).
4. A method of locating internal overheating defects in an electromagnetic pump according to claim 2, characterized in that, The cost aggregation method of any pixel p on the path r is: Where Lr(p, d) is the matching cost of pixel point p when the disparity is d in direction r; p-r represents the previous matching pixel point of pixel point p in the direction; the first term on the right side of the formula is the initial matching cost of pixel point p when the disparity is d, the second term is the minimum matching cost of the previous pixel point p-r in direction r, and the third term is used to prevent the matching cost from overflowing.
5. A method of locating internal overheating defects in an electromagnetic pump according to claim 2, characterized in that, The disparity calculation is to select the minimum cost from all costs, and the disparity corresponding to the minimum cost is the optimal disparity.
6. A method of locating internal overheating defects in an electromagnetic pump according to claim 2, characterized in that, The disparity optimization is to detect and process the occluded points and mismatched points in the image.
7. An electromagnetic pump internal overheating defect locating device, characterized by, The device comprises a binocular system with a wide-angle end infrared thermal imager (1) and a long-focus end infrared thermal imager (2) and a terminal device connected with the binocular system, an image fusion module (3) is integrated in the terminal device and connected with the wide-angle end infrared thermal imager (1) and the long-focus end infrared thermal imager (2) respectively, the image fusion module (3) can fuse and correct the imaging data from the two thermal imagers and deliver to a three-dimensional imaging module (4), the module realizes three-dimensional thermal area imaging of the electromagnetic pump shell by using the corrected data and delivers to a space inversion module (5) connected therewith, the module can analyze and deduce the temperature rise condition and defect position inside the electromagnetic pump.