Slag Bag Cruise Temperature Measurement Device and Method

Abstract

This invention belongs to the field of intelligent equipment design for smelting plants, and specifically relates to a slag ladle cruise temperature measurement device and method. The slag ladle cruise temperature measurement device includes: a freely movable carrier; a control cabinet, mounted on the movable carrier, including a shell and a control module inside the shell; a temperature measurement component, including a rotating gimbal, a pitch support, a dual-spectrum infrared temperature measurement unit, wherein the pitch support and the rotating gimbal are freely hinged, the dual-spectrum infrared temperature measurement unit is mounted on the pitch support, and the dual-spectrum infrared temperature measurement unit can control the rotation angle and / or pitch angle of the dual-spectrum infrared temperature measurement unit; and a vision component, including multiple visible light cameras and at least one lidar. Compared to fixed-point measurement, this method is lower in cost and more adaptable, requiring no modification to the slow cooling field, no need to build fixed supports or install fixed probes, saving on-site modification and fixed equipment investment costs.

Description

Technical Field

[0001] This invention belongs to the field of intelligent equipment design for smelting plants, and in particular relates to a slag ladle cruise temperature measurement device and method. Background Technology

[0002] In the copper smelting process, slag needs to be transported to a slow cooling area via slag bags for heat preservation and slow cooling treatment. The target cooling time for the slag in the target slow cooling area is approximately 72 hours, specifically including 24 hours of natural cooling and 48 hours of water cooling. The temperature change of the slag inside the slag bag directly affects the efficiency of subsequent mineral processing and the safety of production operations. If the slow cooling time is insufficient or the slag temperature distribution is uneven, it can easily lead to safety accidents such as slag bag "red envelopes" and bag explosions, seriously threatening the safety of on-site personnel and equipment. Taking the target slow cooling area that the applicant needs to address as an example, its total area is approximately 80,000 square meters, and it stores 380-400 slag bags simultaneously on a daily basis. It adopts a mixed single-row and double-row arrangement, specifically 10 rows with 34 slag bags per row, and a separate row of 19 slag bags is set up on the north side of the site. The slag bags have uniform specifications, with a height of 2.910 meters and a length and width of 4 meters. The operating environment of the slow cooling field is extremely complex, with a number of unfavorable conditions: Firstly, the highest surface temperature in summer can reach 60℃, and the temperature near the slag bag is significantly higher than that of the surface; there is a lot of dust on site, which affects the accuracy of observation and detection; water accumulation is likely to occur during the rainy season; and there is also a lot of steam and mist in winter, which obstructs the view of temperature measurement; at the same time, the slag bag is transported by a tanker truck, and there are moving obstacles on the work site, which further increases the difficulty of temperature measurement of the slag bag.

[0003] Current technology relies entirely on on-site workers carrying portable temperature measuring devices to inspect each slag bale. Due to the large number of slag bales and their dense stacking, this method results in extremely high labor intensity and low work efficiency. Furthermore, workers operating near the high-temperature slag bales face safety hazards such as burns and bale bursts. This method also cannot achieve continuous temperature measurement around the clock, making it difficult to monitor slag bale temperature changes in real time. Current intelligent temperature measurement methods require gantry frames to be installed above or to the side of the slag bales to house infrared temperature measuring devices. However, the construction of these frames is not only costly but also interferes with the original deployment, preventing the proper functioning of ladle transfer devices and some transport vehicles. This means they are not suitable for scenarios involving flexible slag bale transfer and dynamic arrangement, resulting in poor versatility. Moreover, interference from on-site spray cooling water and steam mist easily leads to the measurement of interfering temperatures, making it impossible to accurately obtain the actual slag temperature, resulting in poor temperature measurement accuracy. Summary of the Invention

[0004] To address the issues of low efficiency and high safety risks associated with manual temperature measurement in copper smelting slow cooling areas, the poor versatility of existing gantry-type temperature measurement devices, susceptibility to interference from water mist and dust leading to inaccurate measurements, and inability to adapt to the dynamic arrangement and safety monitoring requirements of slag bags, this application proposes a slag bag cruise temperature measurement device and method.

[0005] In a first aspect of this application, a slag bag cruise temperature measurement device is provided. The device includes: a freely movable carrier; a control cabinet mounted on the carrier, comprising a housing and a control module inside the housing; a temperature measurement component including a rotating gimbal, a pitch support, a dual-spectrum infrared temperature measurement unit, wherein the pitch support and the rotating gimbal are freely hinged, the dual-spectrum infrared temperature measurement unit is mounted on the pitch support, and the dual-spectrum infrared temperature measurement unit can control its rotation angle and / or pitch angle; and a vision component including multiple visible light cameras and at least one lidar; wherein the control module is configured to: [The text abruptly ends here, so the translation stops as well.] During operation, the real-time point cloud image of the environment collected by the LiDAR is preprocessed. Threshold judgment and region growing algorithm are used to coarsely extract the spray feedback features to form a preliminary interference region point cloud cluster. The features of the visible light region image and the real-time point cloud image of the environment are matched to obtain the interference region parameters, including the interference coordinate range and interference intensity. The dual-spectrum infrared temperature measurement unit is adjusted according to the interference coordinate range to achieve the optimal temperature measurement angle. And / or, the infrared image of the dual-spectrum infrared temperature measurement unit is subjected to interference filtering within the interference coordinate range. Based on the interference intensity, the preset filtering algorithm is adaptively called and the filtering parameters of the corresponding filtering algorithm are dynamically adjusted.

[0006] In a further embodiment of this application, the real-time point cloud image of the environment acquired by the receiving lidar is preprocessed, and a threshold judgment and region growing algorithm are used to coarsely extract the spray feedback features to form a preliminary interference region point cloud cluster. This includes: segmenting the target detection region based on a pass-through filter; calculating the mean and standard deviation of the distances to n neighboring points around each point cloud, removing anomalies whose mean distance deviation exceeds a preset standard deviation, and retaining valid point cloud data; traversing all valid point clouds, selecting point clouds whose electrical cloud parameters meet a preset threshold range, and marking them as candidate interference point clouds; using any point in the candidate interference point cloud as a seed point, traversing its neighboring points to perform region growing, continuing to grow until there are no neighboring points that meet the conditions, thus forming a preliminary interference region point cloud cluster.

[0007] In a further embodiment of this application, feature matching of a visible light region image and a real-time point cloud image of the environment is performed to obtain interference region parameters. This includes: performing grayscale and Gaussian blur processing on the visible light image; using the Otsu binarization algorithm to automatically determine the grayscale threshold and segment the image into interference regions and non-interference regions; and fusing and verifying the interference region point cloud clusters with the interference regions segmented from the visible light image to obtain interference region parameters.

[0008] In a further embodiment of this application, the dual-spectrum infrared temperature measurement unit is adjusted according to the interference coordinate range to achieve optimal temperature measurement angle adaptation. This includes: real-time acquisition of the current position coordinates, the current coordinates of the dual-spectrum infrared temperature measurement unit, and obtaining the real-time coordinates of the target measurement point on the target slag bag based on the digital twin model combined with the movement trajectory of the moving carrier; synchronously establishing a virtual connection between the current coordinates of the simulated temperature measurement unit and the target measurement point on the target slag bag, with the virtual connection synchronously compensating for the positional offset of the moving carrier; determining whether the virtual connection passes through the interference area through spatial calculation; if the virtual connection passes through the interference area, adjusting the rotation angle and pitch angle based on the position control of the interference area until the virtual connection avoids the interference area.

[0009] In a further embodiment of this application, the infrared image of the dual-spectrum infrared thermometer is subjected to interference filtering within the interference coordinate range. Based on the interference intensity, a preset filtering algorithm is adaptively invoked, and the filtering parameters of the corresponding filtering algorithm are dynamically adjusted. This includes: if the interference intensity is within a preset first range, an adaptive moving average filtering algorithm is invoked, with the moving window size set to n1 sampling points, the filtering coefficient α=q1, and the interference threshold set to ±w1℃; if the interference intensity is within a preset second range, a Kalman filtering algorithm and a moving average fusion filtering algorithm are invoked, with the moving window size... The sampling points are adjusted to n², the filter coefficient α = q², and the noise, observation noise, and interference intensity are proportional during the Kalman filtering process. The interference threshold is adjusted to ±w²℃. If the interference intensity is within the preset third range, the wavelet threshold filtering algorithm and the Kalman filtering fusion algorithm are called. The sliding window size is set to n³ sampling points, the filter coefficient α = q³, the wavelet basis is selected as db4, and the wavelet threshold is set to 1 / X of the current interference amplitude. The noise, observation noise, and interference intensity are proportional during the Kalman filtering process. Among these, n1 < n2 < n3, q1 > q2 > q3, and w1 > w2.

[0010] In a further embodiment of this application, the control module is further configured to: after interference filtering, compare the effective temperature signal with the historical temperature data of the slag bag and the on-site interference intensity data to calculate the filtering error; if the filtering error is ≤ T1℃, maintain the current filtering parameters; if the filtering error is > T1℃, automatically fine-tune the filtering coefficient and window size based on a preset ratio until the error meets the standard; the parameters after each adjustment will be stored in the corresponding interference filtering strategy algorithm for continuous iteration to optimize the parameters.

[0011] In a further embodiment of this application, the slag bag cruise temperature measurement device also includes a path supplement light and a collection supplement light. The path supplement light is set on the front and rear sides of the moving carrier, with the field of view of the path supplement light facing the moving path of the moving carrier, and the field of view of the collection supplement light facing the slag bag side. The control module is also configured to trigger the path supplement light and the collection supplement light to start supplementing light when the brightness of the image in the visible light area is lower than a preset brightness value.

[0012] The slag bag cruise temperature measurement device also includes millimeter-wave radar. The control module is also configured to: pre-build an environmental digital twin model before the mobile carrier moves, combine the historical measured temperature and cooling cycle curve of each slag bag to calculate the dynamic path; control the mobile carrier to move according to the dynamic path, and at the same time receive the feedback signal from the millimeter-wave radar. If the feedback signal meets the preset avoidance conditions, calculate the avoidance path and control the mobile carrier to avoid the obstacle and then return to the dynamic path.

[0013] The target measurement points are: starting from the bottom of the contour curve of each target slag bag, select the first point, the second point, and the third point at equal intervals.

[0014] A method for temperature measurement during slag bag cruise includes: Before the mobile carrier moves, a digital twin model of the environment is pre-constructed, and a dynamic path is calculated by combining the historical measured temperature and cooling cycle curve of each slag bag; the mobile carrier is controlled to move according to the dynamic path, and during the journey, real-time environmental point cloud images collected by lidar are received and preprocessed, and threshold judgment and region growing algorithms are used to coarsely extract spray feedback features to form a preliminary interference region point cloud cluster; the visible light region image and the real-time environmental point cloud image are matched to obtain interference region parameters, including interference coordinate range and interference intensity; the dual-spectrum infrared temperature measurement unit is adjusted according to the interference coordinate range to achieve optimal temperature measurement angle adaptation, and / or, the infrared image of the dual-spectrum infrared temperature measurement unit is subjected to interference removal filtering within the interference coordinate range, and based on the interference intensity, a preset filtering algorithm is adaptively called, and the filtering parameters of the corresponding filtering algorithm are dynamically adjusted; the measured temperature obtained by the dual-spectrum infrared temperature measurement unit is received, and the measured temperature is compared with the calculated safe temperature corresponding to the cooling cycle curve of the corresponding slag bag; if the measured temperature exceeds the safe temperature, an alarm is triggered.

[0015] Finally, this application also provides a computer-readable storage medium storing one or more programs, which can be executed by one or more processors to implement the detection method described above.

[0016] Beneficial effects:

[0017] The cruise-type slag bag cruise temperature measurement device used in this application is lower in cost and more adaptable than fixed-point measurement. It does not require modification of the slow cooling field, construction of fixed supports, or installation of fixed probes, saving the investment costs of on-site modification and fixed equipment. Relying on a mobile carrier, it can move freely and adapt to complex working conditions such as double-row placement of slag bags in the slow cooling field, transportation channels, high temperature water accumulation, and dust. It does not require adjustment of fixed measurement points according to the position of the slag bags, significantly improving adaptability.

[0018] Compared to the fixed-rule-based interference removal algorithms in traditional multi-fusion sensing, this solution combines the characteristics of interference from the slow-cooling field spray steam, adaptively calls the preset filtering algorithm based on the interference intensity, and dynamically adjusts the corresponding filtering parameters. It can specifically filter the temperature measurement interference caused by spray steam of different concentrations, effectively avoid the influence of interference data, and ensure the accuracy of the slag bag temperature measurement data.

[0019] Based on filtering and interference removal, the rotation and pitch angles of the dual-spectrum infrared temperature measurement unit are adjusted by the drive unit according to the extracted interference coordinate range. This achieves optimal temperature measurement angle adaptation from a physical perspective, ensuring that the temperature measurement line of sight completely avoids the interference area of ​​spray steam and focuses on the effective measurement point of the slag bag.

[0020] Reduce manual intervention and improve temperature measurement efficiency and safety: No manual on-site operation is required for temperature measurement, nor is it necessary to manually adjust the measurement angle or switch the filter mode. The device can autonomously complete interference identification, angle adjustment, filtering and temperature measurement, which reduces the intensity of manual labor and avoids the safety hazards caused by people approaching the high-temperature slag bag and spray area. At the same time, it can achieve periodic full-coverage temperature measurement and improve inspection and temperature measurement efficiency.

[0021] Combining the uncertainties of the slow cooling field spray steam interference (location and concentration changes), the device adopts a dual interference removal mechanism of "adaptive filtering and physical angle adjustment" to adapt to different interference scenarios. At the same time, combined with digital twin model and dynamic path planning, it can be flexibly adjusted according to the slag bag cooling cycle and on-site working conditions to ensure long-term stable operation of the device under complex working conditions and achieve 24-hour uninterrupted accurate temperature measurement.

[0022] Other features and advantages of the embodiments of the present invention will be described in the following detailed description section. Attached Figure Description

[0023] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of the working environment of the copper smelting slow cooling field provided in an embodiment of the present invention;

[0025] Figure 2 This is a schematic diagram of the slag bag cruise temperature measurement device provided in an embodiment of the present invention;

[0026] Figure 3 This is an exploded schematic diagram of the slag bag cruise temperature measuring device provided in an embodiment of the present invention;

[0027] Figure 4 This is a partial structural diagram of the slag bag cruise temperature measurement device provided in an embodiment of the present invention;

[0028] Figure 5 This is a schematic diagram of a portion of the slag bag cruise temperature measurement device provided in an embodiment of the present invention from another perspective.

[0029] Figure 6 This is a partial structural diagram of the slag bag cruise temperature measurement device provided in an embodiment of the present invention;

[0030] Figure 7 Demonstrates the online testing of slag bag temperature data using the slag bag cruise temperature measurement system;

[0031] Figure 8 Demonstrates online image data from the slag bag cruise temperature measurement device;

[0032] Figure 9 A schematic diagram demonstrating the path of the slag bag cruise temperature measurement device;

[0033] Figure 10 A front-view illustration of the slag bag cruise temperature measurement device.

[0034] Figure 11 A real-world photograph taken from the side view of the slag bag cruise temperature measurement device in cruise mode.

[0035] Figure 12 A real-life demonstration of the slag bag cruise temperature measurement device performing automatic charging.

[0036] Figure Labels

[0037] 100. Slag bag cruise temperature measurement device;

[0038] 10. Mobile carrier; 11. Tracked chassis; 12. Drive motor; 13. Shock absorption components; 14. Charging interface;

[0039] 20. Control cabinet; 21. Housing; 22. Control module; 23. Fan module; 24. Partition plate;

[0040] 221. Industrial control host; 222. Image module; 223. Electrical control auxiliary module; 224. Edge computing module; 225. Step-down module;

[0041] 30. Temperature measurement component; 31. Gimbal; 32. Pitch support; 33. Drive unit; 34. Dual-spectrum infrared temperature measurement unit; 35. Positioning radar;

[0042] 40. Vision components; 41. Visible light camera; 42. LiDAR; 43. Millimeter-wave radar;

[0043] 50. Auxiliary components; 51. Path fill light; 52. Acquisition fill light. Detailed Implementation

[0044] To make the above and other features and advantages of the present invention clearer, the invention will be further described below with reference to the accompanying drawings. It should be understood that the specific embodiments given herein are for the purpose of explanation to those skilled in the art and are exemplary only, not restrictive.

[0045] Continuing from the previous approach, this application's embodiments abandon the previous method of manually or on-site setting up to install movable probes, and instead design a device similar to a cruising mobile carrier. This device is required to be able to autonomously plan its path and periodically detect slag bags, as well as to rely on the current specific environment in terms of algorithms.

[0046] It should be noted that the on-site environment, such as Figure 1 The demonstration shows slag ladles in a slow cooling zone for copper smelting, arranged in double rows on both sides. The steam mist from the spraying system typically diffuses over the upper part of the slag ladles, but may also appear in other locations due to wind interference or natural descent. The double rows of slag ladles are arranged in a line, with ample space between adjacent rows to facilitate movement of the transport vehicles carrying the slag ladles.

[0047] Based on the above, as shown in the corresponding appendix... Figures 2 to 6 ,as well as Figures 10 to 12 As shown in the figure, the slag bag cruise temperature measurement device 100 includes a mobile carrier 10, a control cabinet 20, a temperature measurement component 30, a vision component 40, and an auxiliary component 50. The installation position, connection relationship, and function of each component are as follows:

[0048] The mobile carrier 10 serves as the mobile foundation for the entire device, enabling it to move freely on the ground of the slow cooling field. It is suitable for complex working conditions such as high summer temperatures (surface temperature up to 60°C), rainy season water accumulation (depth up to 1m), and high dust levels in the slow cooling field. A tracked mobile carrier is preferred, and the specific configuration is as follows:

[0049] The mobile carrier 10 includes a tracked chassis 11, a drive motor 12, a shock-absorbing component 13, and a charging interface 14. The tracked chassis 11 is made of high-strength wear-resistant rubber with anti-slip textures on the surface, which can effectively avoid slippage caused by water accumulation and dust, and has better stability. The drive motor 12 is an explosion-proof and high-temperature resistant motor with a rated power of 5.5kW, which can drive the mobile carrier 10 to cruise at a constant speed of 0.5m / s, while supporting stepless speed adjustment from 0.1 to 0.8m / s, which can be flexibly adjusted according to the on-site working conditions (such as areas with a high density of tank trucks). The charging interface 14 can be adapted to the charging device in the slow cooling field charging room to realize automatic charging and ensure that the device cruises and measures temperature 24 hours a day. The side of the mobile carrier 10 facing away from the charging interface 14 in the figure is the electrical interface, so that the mobile carrier 10 and the control cabinet 20 are physically connected and electrically connected at the same time. The shock-absorbing component 13 is set between the tracked chassis 11 and the device mounting platform to buffer the vibration caused by the uneven ground of the slow cooling field and ensure the working stability of each component.

[0050] The control cabinet 20 is fixedly mounted on the installation platform of the mobile carrier 10, located at the center of gravity of the mobile carrier 10. This prevents the center of gravity of the device from shifting during movement and ensures that the mobile carrier 10 will not tip over when moving on uneven ground or waterlogged sections in the slow cooling area, thus guaranteeing the overall stability of the device during movement. The specific structure is as follows:

[0051] The control cabinet 20 includes a housing 21, a control module 22 disposed inside the housing 21, and a fan module 23 disposed on the lower side of the housing 21. The control module 22 includes an industrial control host 221, an image module 222, an electrical control auxiliary module 223, an edge computing module 224, and a step-down module 225. The electrical control auxiliary module 223 includes a relay, a serial port server, a switch, and a transformer. The relay can realize precise control of circuit on / off to ensure the power supply safety of each component. The serial port server realizes the conversion between serial port signals and network signals to ensure the stable transmission of control signals and acquired data. The switch realizes data interaction between modules. The transformer provides adaptation voltage for each module to avoid damage to the module due to voltage instability. The step-down module can convert the external input voltage into the working voltage required by each module of the control module 22 to ensure the stable operation of the module.

[0052] The housing 21 is provided with at least one partition plate 24, which divides the interior of the housing 21 into at least a first cavity and a second cavity. The first cavity is closer to the mobile carrier 10 than the second cavity, that is, the side closer to the fan module 23. The sides of the first cavity and the second cavity are provided with ventilation grilles corresponding to those on the housing 21. The industrial control host 221, the image module 222, and the electrical control auxiliary module 223 are located in the first cavity, while the edge computing module 224 and the step-down module 225 are located in the second cavity, so as to achieve thermal isolation of each module and form a reasonable heat dissipation channel adapted to the current working conditions.

[0053] Understandably, the fan module 23 is designed to draw air upwards from the bottom, and the heat generated by most of the modules in the first cavity is directed to the ventilation grilles on both sides. Ventilation holes are set in specific areas on the partition plate 24, and these ventilation holes avoid the installation position of the temperature measuring component 30, so that the heat will not interfere with the temperature measuring component 30. In addition, because the industrial control host, image module, etc. generate a lot of heat when they are working, thermal isolation is achieved with the edge computing module and step-down module to avoid mutual heat interference that could cause the edge computing module 224 to overheat and interfere with the measurement of the temperature measuring component 30. It can also effectively extend the service life of each module, while ensuring the overall stability and reliability of the control module 22. The fan module 23 is used to dissipate heat inside the housing 21, adapting to the high temperature conditions of the slow cooling field. It can promptly dissipate the heat generated by the operation of each module inside the housing 21, preventing the control module 22 from being damaged by overheating. It ensures that the control module 22 can still operate stably in the high temperature environment of the slow cooling field in summer, where the surface temperature can reach up to 60°C, and ensures that the device can continuously monitor and measure temperature 24 hours a day.

[0054] Furthermore, the temperature measurement component 30 includes a rotating gimbal 31, a pitch support 32, a drive unit 33, a dual-spectrum infrared temperature measurement unit 34, and a positioning radar 35; the pitch support 32 and the rotating gimbal 31 are freely hinged, allowing for flexible adjustment of the pitch angle; the dual-spectrum infrared temperature measurement unit 34 is mounted on the pitch support 32 and moves synchronously with the pitch support 32; the drive unit 33 is connected to the rotating gimbal 31 and the pitch support 32 respectively, and can control the rotation angle and / or pitch angle of the dual-spectrum infrared temperature measurement unit 34 to meet the temperature measurement requirements of different directions and different heights; the positioning radar 35 is used to locate the current coordinate position.

[0055] The driving method of the drive unit 33 can be at least one of motor, hydraulic, or magnetic. In a preferred embodiment, the drive unit 33 can adopt a combination of magnetic force and articulated motor drive. The pitch angle of the pitch support 32 can be adjusted by magnetic force. A strong magnetic drive structure is selected, and the pitch support connection part is made of magnetic material. This can achieve stepless fine adjustment of the pitch angle, which can meet the precise alignment requirements of measuring points at different heights of the slag bag. At the same time, the magnetic drive has good self-locking properties. After adjustment, the pitch angle can be stably locked, which can effectively counteract the interference of the dual-spectrum infrared temperature measuring unit's own gravity and the vibration during the robot's movement, and avoid temperature measurement deviation caused by angle offset. The rotation angle of the rotating gimbal 31 is adjusted by the articulated motor. The two work together to accurately control the rotation angle and / or pitch angle of the dual-spectrum infrared temperature measuring unit 34.

[0056] The vision component 40 is fixedly mounted at the front end of the installation platform of the mobile carrier 10, located on one side of the temperature measurement component 30, and is used to collect environmental data of the slow cooling field, images of the slag bag, and data on spray interference. The specific structure is as follows:

[0057] The vision component 40 includes a visible light camera 41, at least one lidar 42, and a millimeter-wave radar 43. The visible light camera 41 is used to acquire visible light images of the site and capture the visual features of the slag bag outline and the spray area. The lidar 42 is used to acquire real-time point cloud images of the environment and capture the feedback features of the spray particles. The millimeter-wave radar 43 is used to assist in detecting moving obstacles (such as tank trucks) on site and ensure the safety of the device's movement. The three components work together to provide data support for the extraction of interference areas and path adjustment.

[0058] The auxiliary component 50 is fixedly mounted on the mobile carrier 10 to assist the device in achieving nighttime patrol and accurate data collection. The specific structure is as follows: The auxiliary component 50 includes a path supplement light 51 and a data collection supplement light 52. The path supplement light 51 is located on the front and rear sides of the mobile carrier 10 to illuminate the movement path of the mobile carrier 10 and ensure safe movement at night. The field of view of the data collection supplement light 52 is directed towards the movement path of the mobile carrier 10 to illuminate the slag bag and surrounding area, improve the data collection accuracy of the visible light camera 41 and the lidar 42, and adapt to working conditions at night or in low light.

[0059] The control module 22 is configured as follows:

[0060] Step S10: Before the mobile carrier moves, a digital twin model of the environment is pre-built, and the dynamic path is calculated by combining the historical measured temperature and cooling cycle curve of each slag bag.

[0061] Specifically, the vision component 40 collects overall environmental data of the slow cooling field in advance (including the position, spacing, and width of the transport channel of the double-row slag bags), constructs a digital twin model consistent with the physical slow cooling field, and imports historical temperature measurement data and 72-hour cooling cycle curves of each slag bag. The industrial control host 221, combined with the edge computing module 224, calculates the dynamic inspection path with "optimal efficiency and minimal interference" to ensure that the device can periodically cover all slag bags and complete the temperature measurement operation.

[0062] Step S20: Control the mobile carrier to move according to the dynamic path. During the journey, receive the real-time point cloud image of the environment collected by the lidar for preprocessing. Use threshold judgment and region growing algorithm to coarsely extract the spray feedback features to form a preliminary interference area point cloud cluster.

[0063] Specifically, the mobile carrier 10 cruises along a dynamic path at a constant speed of 0.5 m / s, the lidar 42 collects environmental point cloud data in real time, the image module 222 performs noise reduction preprocessing on the original point cloud data to remove invalid point clouds, and then uses a threshold judgment algorithm to select point clouds that meet the spray feedback characteristics. Then, the adjacent spray feature point clouds are merged by the region growing algorithm to form a preliminary interference region point cloud cluster, thus completing the coarse extraction of the spray region.

[0064] Step S30: Match the features of the visible light region image and the real-time point cloud image of the environment to obtain the interference region parameters, which include the interference coordinate range and interference intensity.

[0065] Specifically, the image module 222 performs pixel-level feature matching between the visible light image and the preprocessed point cloud image, aligns their spatial coordinates, removes false interference areas where the point cloud and image do not match, retains overlapping spray areas, and finally outputs a precise range of interference coordinates (i.e., the spatial coordinates of the spray area). Simultaneously, it calculates the interference intensity based on the point cloud density, providing data support for subsequent angle adjustment and filtering processing.

[0066] Step S40: Adjust the dual-spectrum infrared temperature measurement unit according to the interference coordinate range to achieve the optimal temperature measurement angle, and / or, perform interference removal filtering on the infrared image of the dual-spectrum infrared temperature measurement unit within the interference coordinate range, and adaptively call the preset filtering algorithm based on the interference intensity, and dynamically adjust the filtering parameters of the corresponding filtering algorithm.

[0067] Specifically, the industrial control host 221 controls the drive unit 33 to adjust the angles of the rotating gimbal 31 and the pitch support 32 according to the range of interference coordinates, so that the temperature measurement line of the dual-spectrum infrared temperature measurement unit 34 avoids the interference area, focuses on the effective measurement point of the slag bag, achieves optimal temperature measurement angle adaptation, adaptively calls the interference removal filtering algorithm, and dynamically adjusts the filtering parameters to filter the temperature measurement interference caused by the spray steam, ensuring accurate temperature measurement data.

[0068] Step S50: Receive the measured temperature obtained by the dual-spectrum infrared temperature measurement unit, compare the measured temperature with the safe temperature calculated by the cooling cycle curve of the corresponding slag bag, and trigger an alarm if the measured temperature exceeds the safe temperature.

[0069] The dual-spectrum infrared temperature measurement unit 34 transmits the collected slag bag temperature data to the industrial control host 221. The industrial control host 221 retrieves the safe temperature threshold corresponding to the cooling cycle curve of the slag bag and performs a real-time comparison. If the measured temperature of several bags exceeds the safe temperature threshold, the control module 22 immediately triggers an alarm and records the abnormal data and the coordinates of the abnormal slag bags, so that the staff can handle the situation in a timely manner.

[0070] In summary, the slag bag cruise temperature measurement device 100 of this embodiment abandons the traditional method of manual temperature measurement and on-site setup of a support frame to install a movable probe. Instead, it adopts a structural design similar to a cruise mobile carrier, and combined with optimized control algorithms, it has the following advantages:

[0071] 1. Compared with fixed-point measurement, this device does not require modification of the slow cooling field, construction of fixed supports, or installation of fixed probes, saving the investment costs of on-site modification and fixed equipment; relying on a mobile carrier, it can move freely and adapt to complex working conditions such as double-row placement of slag bags in the slow cooling field, transportation channels, high temperature water accumulation, and dust, without the need to adjust the fixed measurement points according to the position of the slag bags, significantly improving adaptability.

[0072] 2. High accuracy of interference filtering, adapted to the requirements of this working condition: Compared with the interference filtering algorithm based on fixed rules in traditional multi-fusion sensing, this solution combines the characteristics of interference from the slow cooling field spray steam, adaptively calls the preset filtering algorithm based on the interference intensity, and dynamically adjusts the corresponding filtering parameters. It can specifically filter the temperature measurement interference caused by spray steam of different concentrations, effectively avoid the influence of interference data, and ensure the accuracy of slag bag temperature measurement data.

[0073] 3. Physical angle adaptation, double avoidance of spray steam interference: Based on the filtering and interference removal, according to the extracted interference coordinate range, the rotation angle and pitch angle of the dual-spectrum infrared temperature measurement unit are adjusted by the drive unit to achieve the optimal temperature measurement angle adaptation from a physical level, so that the temperature measurement line of sight completely avoids the spray steam interference area and focuses on the effective measurement point of the slag bag, thus doubly ensuring that the temperature measurement data is not interfered with.

[0074] 4. Reduce manual intervention and improve temperature measurement efficiency and safety: No manual on-site temperature measurement is required, nor is manual adjustment of measurement angle or switching of filter mode. The device can autonomously complete interference identification, angle adjustment, filtering and temperature measurement, which reduces the intensity of manual labor and avoids the safety hazards caused by manual approach to high-temperature slag bags and spray areas. At the same time, it can achieve periodic full-coverage temperature measurement and improve inspection and temperature measurement efficiency.

[0075] 5. High adaptability and stability, ensuring long-term reliable operation: Combining the uncertainty of the spray steam interference in the slow cooling field (location and concentration changes), it adapts to different interference scenarios through a dual interference removal mechanism of "adaptive filtering + physical angle adjustment"; at the same time, combined with digital twin model and dynamic path planning, it can be flexibly adjusted according to the slag bag cooling cycle and on-site working conditions; the layout of each component is reasonable, the control cabinet 20 achieves thermal isolation through the partition plate 24, the fan module 23 ensures heat dissipation, the mobile carrier 10 adopts a tracked structure to adapt to complex ground environment, and the auxiliary component 50 improves the ability to work at night. The overall stability is strong, and it can achieve 24-hour uninterrupted accurate temperature measurement, adapting to the long-term operation requirements of the slow cooling field.

[0076] The following provides feasible implementation methods for specific embodiments:

[0077] In a further embodiment of this application, the real-time point cloud image of the environment acquired by the receiving lidar is preprocessed, and a threshold judgment and region growing algorithm are used to coarsely extract the spray feedback features to form a preliminary point cloud cluster of the interference region, including:

[0078] S11. Segment the target detection region based on pass-through filtering;

[0079] S12. Calculate the mean and standard deviation of the distances to n neighboring points around each point cloud, remove outliers whose mean distance deviation exceeds the preset standard deviation, and retain the valid point cloud data.

[0080] S13. Traverse all valid point clouds, filter point clouds whose electric cloud parameters meet the preset threshold range, and mark them as candidate interference point clouds.

[0081] S14. Using any point in the candidate interference point cloud as the seed point, traverse its neighboring points to perform region growth, and continue growing until there are no neighboring points that meet the conditions, forming a preliminary interference region point cloud cluster.

[0082] It is understandable that, based on the laser pulse emitted by the lidar into the detection area (using a 905nm wavelength, suitable for slow-cooling dust and fog environments), the spray particles (cooling water mist with a diameter of 10-100μm) will strongly scatter and reflect the laser, forming a differentiated feedback characteristic with no obvious scattering with the air and mirror reflection on the slag bag surface, resulting in stable signal intensity. Based on the applicant's experiments, its core characteristics include: Scattered signal intensity: The laser scattering signal intensity in the spray area (5000-8000AU) is significantly higher than that in the air (<1000AU) and lower than that on the slag bag surface (>10000AU); In terms of reflection time difference, the spray particles are discretely distributed, and the laser reflection time difference fluctuates greatly (±50ns), while the reflection time difference between the air and the slag bag surface is stable (fluctuation ≤10ns); In terms of point cloud density: The laser point cloud density in the spray area (150-200 points / cm²) is much higher than that in the air (<50 points / cm²), and the point cloud is discretely distributed (without continuous contours). Through on-site calibration, threshold ranges for three types of features are preset as the basis for subsequent spray (interference area) identification.

[0083] The raw point cloud data acquired by the lidar is subjected to noise reduction processing. The embodiment of this application adopts a combination algorithm of direct-pass filtering and statistical filtering: the direct-pass filtering sets the Z-axis (height) threshold according to the height of the slag bag in the slow cooling field (2.910m) and the height of the spray area (1-3m), filtering out invalid point clouds on the ground (Z<1m) and at high altitude (Z>3m), and focusing on the target detection area; the statistical filtering calculates the mean and standard deviation of the distance to 50 neighboring points around each point cloud, removes outliers with a distance mean deviation exceeding 2 times the standard deviation (such as isolated point clouds reflected by dust and debris), retains the effective point cloud data, and improves the accuracy of interference area identification;

[0084] Traverse all valid point clouds and select those that meet the preset threshold range for "scattered signal intensity, reflection time difference, and point cloud density" as candidate interference point clouds. For region growing, select any point in the candidate interference point cloud as the seed point and traverse its neighboring points (distance < 5cm). If the neighboring points also meet the threshold conditions, they are included in the same region. Continue growing until there are no neighboring points that meet the conditions, forming a preliminary interference region point cloud cluster, and completing the coarse extraction.

[0085] In the above scheme, features of the visible light region image and the real-time point cloud image of the environment are matched to obtain the parameters of the interference region, including:

[0086] S21. Perform grayscale conversion and Gaussian blur processing on the visible light image;

[0087] S22. Using the Otsu binarization algorithm, the grayscale threshold is automatically determined, and the image is segmented into interference regions and non-interference regions.

[0088] S23. The interference region point cloud clusters and the interference regions segmented from the visible light image are fused and verified to obtain the interference region parameters.

[0089] The system utilizes a visible light camera to simultaneously acquire images of the detection area, and performs pixel-level registration with the LiDAR point cloud. This is achieved through the SIFT feature matching algorithm, which enables spatial alignment between the point cloud and the image.

[0090] First, the visible light image is converted to grayscale and Gaussian blurred to suppress dust and light interference and enhance the interference area. Specifically, the interference area is white / grayish white to contrast with the dark surface of the slag bag and the air.

[0091] The Otsu binarization algorithm is used to automatically determine the grayscale threshold and segment the image into spray areas (foreground) and non-spray areas (background, slag packs, ground, etc.). The spray point cloud clusters coarsely extracted by the lidar are fused and verified with the spray areas segmented by the visible light image. Areas where the point cloud and the image do not overlap (such as isolated point clouds and false spray areas in the image) are removed, and the areas that overlap are retained. Finally, the accurate coordinate range of the spray area is obtained, and the parameters of the interference area are extracted.

[0092] Furthermore, the dual-spectrum infrared thermometry unit is adjusted according to the interference coordinate range to achieve optimal temperature measurement angle adaptation, including:

[0093] S31. Real-time acquisition of current position coordinates, current coordinates of dual-spectrum infrared temperature measurement unit, and real-time coordinates of target measurement points on the target slag bag based on digital twin model combined with the movement trajectory of the mobile carrier;

[0094] S32. Synchronize the virtual connection between the current coordinates of the simulated temperature measurement unit and the target measurement point on the current target slag bag, and the virtual connection synchronously compensates for the positional offset of the moving carrier.

[0095] S33. Determine whether the virtual connection passes through the interference area through spatial calculation;

[0096] S34. If the virtual connection passes through the interference area, adjust the rotation angle and pitch angle based on the position of the interference area until the virtual connection avoids the interference area.

[0097] Specifically, the temperature measurement carrier for the double-row slag bag moves along the outer side of the double rows, and the dual-panel independently controls the dual-spectrum infrared temperature measurement component. At the same time, the visible light camera and lidar work together. The extraction of the spray area is optimized by combining the moving speed of the carrier. Based on the spray feedback characteristics of the lidar, the spray area is accurately extracted using a standardized algorithm. The specific process and algorithm are as follows:

[0098] Based on the robot's movement speed, the LiDAR and visible light camera are activated synchronously, and the two are synchronized via timestamps (synchronization error ≤10ms) to ensure the spatial and temporal consistency of the point cloud and the image, avoiding misalignment of the point cloud and image caused by robot movement.

[0099] Combining the robot's uniform movement state, the system acquires the robot's current position coordinates based on the positioning radar (or, alternatively, through quadruple fusion positioning technology, updating the coordinates every 100ms to ensure accurate coordinates during movement), the current installation coordinates of the dual-spectrum infrared temperature measurement unit (bound to the robot's position coordinates and updated synchronously), and simultaneously obtains the real-time coordinates of the target measurement points (upper, middle, and lower points) on the target slag bag. Based on the digital twin model, the system predicts the measurement point positions based on the robot's movement trajectory, locking the target measurement points 0.2s in advance to avoid measurement point positioning delays caused by movement. Based on the above real-time coordinates, the interference area is simulated in the digital twin model, and the virtual connection between the temperature measurement unit and the target measurement points on the target slag bag in space is simulated and calculated.

[0100] By combining the robot's movement speed, the processing efficiency of the spatial calculation algorithm is optimized, completing a spatial judgment every 100ms, synchronized with the robot coordinate update frequency and the spray area update frequency (10Hz). Through spatial coordinate calculation, it is determined whether the simulated virtual connection passes through the interference area. During the calculation process, the positional offset caused by the robot's movement is compensated synchronously (based on a speed of 0.5m / s, the robot's movement distance within 100ms is predicted, and the starting coordinates of the virtual connection are corrected to ensure accurate judgment results and avoid misjudgments caused by movement).

[0101] If spatial calculations determine that the virtual connection passes through the interference area, the dynamic adjustment mechanism of the temperature measuring unit's rotating gimbal is immediately activated. This involves optimizing the gimbal's response speed based on the moving vehicle's speed and precisely adjusting the temperature measuring unit's rotation angle (horizontal) and pitch angle (vertical) based on the coordinate range of the interference area. During the adjustment process, the robot's movement trajectory is simultaneously considered to predict its position within 100ms, allowing for pre-adjustment of the angles. This ensures that the adjusted virtual connection not only avoids the current interference area but also adapts to the robot's subsequent movements, preventing the robot from re-entering the interference area. This process continues until spatial calculations confirm that the virtual connection completely avoids the interference area and accurately focuses on the target measurement point, achieving optimal temperature measurement angle adaptation.

[0102] In the above scheme, the infrared image of the dual-spectrum infrared thermometer is subjected to interference filtering within the interference coordinate range. Based on the interference intensity, a preset filtering algorithm is adaptively invoked, and the filtering parameters of the corresponding filtering algorithm are dynamically adjusted, including:

[0103] S31. If the interference intensity is within the preset first range, call the adaptive moving average filtering algorithm, set the sliding window size to n1 sampling points, the filtering coefficient α=q1, and the interference threshold to ±w1℃.

[0104] S32. If the interference intensity is within the preset second range, call the Kalman filter algorithm and the moving average fusion filter algorithm. Adjust the sliding window size to n2 sampling points, and the filter coefficient α=q2. The noise, observation noise and interference intensity are proportional in the Kalman filter process. Adjust the interference threshold to ±w2℃.

[0105] S33. If the interference intensity is within the preset third range, call the wavelet threshold filtering algorithm and the Kalman filter fusion algorithm. Set the sliding window size to n3 sampling points, the filter coefficient α=q3, the wavelet basis to db4, and the wavelet threshold to 1 / X of the current interference amplitude. The noise, observation noise and interference intensity are proportional during the Kalman filtering process.

[0106] Where n1 < n2 < n3, q1 > q2 > q3, w1 > w2.

[0107] In a specific scheme, typical interference signals and effective temperature signals (slag bag surface temperature signals) of the slow cooling field are first collected by a dual-spectrum infrared temperature measurement component, a visible light camera, and a lidar in a coordinated manner. The core features of the two types of signals are then extracted and thresholds are calibrated.

[0108] Effective temperature signal: stable frequency (1-5Hz), small signal amplitude fluctuation;

[0109] Dust interference signal: frequency disorder (5-20Hz), large amplitude fluctuation (±5-10℃), signal exhibits irregular spikes;

[0110] Steam mist interference signal: low frequency (0.5-2Hz), amplitude fluctuates slowly (±3-8℃), and the signal has obvious trailing phenomenon.

[0111] Meanwhile, a correlation model between "dust concentration and interference signal intensity" and "vapor mist concentration and interference signal amplitude" is established as the basis for subsequent adaptive adjustment of parameters (e.g., for every 10 mg / m³ increase in dust concentration, the interference signal intensity increases by 15%).

[0112] The algorithm receives auxiliary data in real time from lidar (detecting dust and mist particle density) and visible light cameras (detecting field of view clarity), and combines it with the raw temperature signal collected by the infrared thermometer to identify the current interference type (dust, steam, or mist) and interference intensity through a "feature matching algorithm".

[0113] Calculate the frequency and amplitude fluctuation values ​​of the original temperature signal and compare them with the interference characteristic threshold calibrated in the preprocessing stage to preliminarily determine whether interference exists and the type of interference.

[0114] Combining the dust particle density (unit: mg / m³) and vapor cloud density collected by lidar, the interference intensity is quantified (divided into 3 levels: weak interference, medium interference, and strong interference). For example: dust concentration < 50 mg / m³ and vapor cloud density < 80 points / cm² is considered weak interference; dust concentration 50-100 mg / m³ and vapor cloud density 80-150 points / cm² is considered medium interference; and values ​​exceeding these levels indicate strong interference.

[0115] Based on the identified interference type and intensity, the algorithm automatically invokes a preset parameter adjustment strategy to dynamically optimize the core filtering parameters (sliding window size, filter coefficients, threshold range). The specific adjustment rules are as follows:

[0116] 1) Weak interference (quantized data interference intensity in the first range): Adaptive moving average filtering is used, with the sliding window size set to 5-8 sampling points (sampling frequency synchronized with temperature measurement frequency, 0.8s / time), filtering coefficient α=0.7 (the closer α is to 1, the more original signal is retained), and the interference threshold is set to ±3℃ to filter out a small amount of glitches and slight fluctuation interference, and retain the effective temperature signal to the maximum extent.

[0117] 2) Medium interference (quantized data interference intensity in the second range): Switch to "Kalman filter + moving average fusion filter", adjust the sliding window size to 8-12 sampling points, filter coefficient α=0.5, Kalman filter process noise Q=0.05, observation noise R=0.1 (dynamically adjusted according to the interference intensity; the stronger the interference, the larger the R value), and tighten the interference threshold to ±2℃, focusing on filtering medium-intensity fluctuation interference, while taking into account signal smoothness and accuracy.

[0118] 3) Strong interference (quantized data interference intensity in the third range): The "wavelet threshold filtering + Kalman filtering fusion algorithm" is adopted, the sliding window size is set to 12-15 sampling points, the filtering coefficient α=0.3, the wavelet basis is selected as db4 (adapted to the characteristics of slow cooling field interference signal), the wavelet threshold is set to 1 / 2 of the current interference amplitude, and the Kalman filter Q=0.08, R=0.15 to further suppress the amplitude fluctuation of strong interference signal and eliminate obvious false temperature data (such as instantaneous high temperature / low temperature jump caused by steam obstruction).

[0119] Interference data filtering and signal correction: Based on the adjusted filtering parameters, the original temperature signal is processed as follows: Moving average filtering: A weighted average is taken for the sampling points within the window to reduce random spike interference; Kalman filtering: The temperature signal is predicted through the state equation, and the predicted value is corrected by combining it with the observation equation to reduce slow fluctuation interference (such as temperature drift caused by steam and mist); Wavelet threshold filtering: The signal is decomposed into wavelets to set high-frequency interference components (dust spikes) to zero, and a smooth and effective temperature signal is reconstructed. Simultaneously, the algorithm automatically removes abnormal data exceeding the interference threshold. If three consecutive sampling points are abnormal data, a secondary verification by the lidar and visible light camera is triggered to confirm whether it is caused by strong interference, thus avoiding the accidental deletion of effective data.

[0120] Furthermore, in this application, the control module 22 is also configured to:

[0121] S34. After interference filtering, the effective temperature signal is compared with the historical temperature data of the slag bag and the on-site interference intensity data to calculate the filtering error.

[0122] S35. If the filtering error is ≤ T1℃, maintain the current filtering parameters. If the filtering error is > T1℃, automatically fine-tune the filtering coefficient and window size based on the preset ratio until the error meets the standard.

[0123] S36. The adjusted parameters will be stored in the corresponding interference filtering strategy algorithm for continuous iteration to optimize the parameters.

[0124] After filtering, the algorithm compares the effective temperature signal with historical temperature data of the slag bag and on-site interference intensity data, and calculates the filtering error (the deviation between the filtered signal and the actual temperature): ① If the filtering error is ≤2℃, maintain the current filtering parameters; ② If the filtering error is >2℃, automatically fine-tune the filtering coefficient and window size (adjustment range not exceeding 10%) until the error meets the standard; ③ The parameters after each adjustment will be stored in the algorithm database, and the parameter strategy will be continuously iterated and optimized in combination with changes in on-site interference conditions to improve the algorithm's adaptability to different concentrations of dust and steam mist, ensuring long-term temperature measurement accuracy.

[0125] Furthermore, control module 22 is also configured to:

[0126] S101. When the brightness of the image in the visible light area is lower than the preset brightness value, the path fill light and the acquisition fill light are triggered to start fill light.

[0127] In this embodiment of the application, the control module 22 is further configured to:

[0128] Step S101: Before the mobile carrier moves, a digital twin model of the environment is pre-built, and the dynamic path is calculated by combining the historical measured temperature and cooling cycle curve of each slag bag.

[0129] Step S102: Control the mobile carrier to move according to the dynamic path, and at the same time receive the feedback signal from the millimeter-wave radar. If the feedback signal meets the preset avoidance conditions, calculate the avoidance path and control the mobile carrier to avoid the obstacle and then return to the dynamic path.

[0130] Before the mobile carrier (inspection robot) moves, the core task of the control module is to pre-build an environmental digital twin model. Based on this model, and combined with the historical measured temperature data and complete cooling cycle curves of each slag bag, it calculates an initial dynamic path adapted to the on-site working conditions (such as...). Figure 9 (As demonstrated) The core design logic of this initial path is to plan the initial inspection trajectory for the mobile carrier. The core idea is to accurately identify target slag bags that may face problems in advance. Specifically, by comparing and analyzing the historical temperature measurement data and cooling cycle curves of each slag bag, high-risk target slag bags with abnormal cooling rates, uneven temperature distribution, or a history of "red packets" or bursting hazards are selected. These target slag bags are used as the core locations, and stop points and slow inspection points are reasonably selected. For non-target slag bags (slag bags with no obvious abnormal risks and whose cooling process meets the standards), a rapid sweep mode is adopted, without setting stop points or slow points, and only basic temperature measurement is completed. Through the differentiated design of "precise monitoring of target slag bags + rapid sweeping of non-target slag bags", a scientific and reasonable initial dynamic path is constructed, which not only ensures the accuracy of target slag bag monitoring, but also improves the overall inspection efficiency.

[0131] The robot (mobile platform) starts according to the dynamic path preset by the control module. The positioning and navigation module (including millimeter-wave radar) collects position data in real time and performs positioning calibration (accuracy ±5cm) through a four-fold fusion technology of "LiDAR + millimeter-wave radar + UWB positioning + visual calibration". With the trajectory memory function enabled, after the millimeter-wave radar detects an obstacle and completes the avoidance, it can quickly return to the optimal dynamic path to ensure that the inspection trajectory does not deviate, thus addressing the pain point of moving obstacles on site.

[0132] The "target measurement points" defined in this application are: points 1, 2, and 3 selected at equal intervals from top to bottom of the contour curve of each target slag bag (e.g., ...). Figure 8 (As demonstrated).

[0133] By combining historical cooling curves of slag bags, ambient temperature, and spray conditions, the temperature, measurement time, extreme values, and average values ​​at three points (top, middle, and bottom) of each slag bag are marked. Simultaneously, the temperature change trend over the next 12 hours is predicted, potential anomalies are identified, and proactive warnings are issued at the visualization front end (e.g., ...). Figure 7 (As demonstrated), to prevent safety accidents in advance; bind temperature data with slag bag material, number of uses, cooling cycle, spraying duration and robot number to establish a "full life cycle temperature file" for slag bags, support cooling effect traceability and analysis, and provide a basis for cooling process optimization.

[0134] Meanwhile, after completing the full-area inspection, the robot automatically returns to the charging room to recharge; the backend evaluates the next inspection focus based on the inspection data and adjusts the route and duration accordingly; when the battery level drops below 20%, the robot automatically pauses inspection and charges at the nearest charging station, resuming operations after charging is complete (e.g., ...). Figure 12 (As demonstrated).

[0135] Based on some modified embodiments of this application, it can also be linked with the original slag bag system. Based on the temperature data and cooling trend prediction of the slag bag throughout its entire life cycle, the system automatically optimizes the operating time and spray intensity of the spray system to adapt to a 72-hour cooling cycle. According to the cooling status of the slag bag, the system recommends the turnover sequence to improve the utilization rate of the slag bag, reduce production costs, and ensure uniform cooling while avoiding safety hazards.

[0136] Furthermore, those skilled in the art should understand that if all or part of the sub-modules involved in the products provided in the embodiments of the present invention are combined or replaced by means of fusion, simple changes, mutual transformation, etc., such as moving the position of each component; or setting the product they constitute as a whole; or having a detachable design; as long as the combined components can form a device / apparatus / system with a specific function, using such a device / apparatus / system to replace the corresponding components of the present invention also falls within the protection scope of the present invention.

[0137] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0138] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A ladle cruise temperature measuring device, characterized by, include: A mobile carrier capable of free movement; A control cabinet, mounted on the mobile carrier, includes a housing and a control module inside the housing; The temperature measurement component includes a rotating gimbal, a pitch support, a drive unit, and a dual-spectrum infrared temperature measurement unit. The pitch support and the rotating gimbal are freely hinged together. The dual-spectrum infrared temperature measurement unit is mounted on the pitch support. The drive unit can control the rotation angle and / or pitch angle of the dual-spectrum infrared temperature measurement unit. The vision component includes multiple visible light cameras and at least one lidar; The control module is configured as follows: During the driving process, the real-time point cloud image of the environment collected by the LiDAR is preprocessed, and the spray feedback features are coarsely extracted using threshold judgment and region growing algorithm to form a preliminary point cloud cluster of interference area. The interference region parameters are obtained by matching the features of the visible light region image and the real-time point cloud image of the environment. The interference region parameters include the interference coordinate range and the interference intensity. The dual-spectral infrared temperature measurement unit is adjusted according to the interference coordinate range to achieve optimal temperature measurement angle adaptation, and / or, the infrared image of the dual-spectral infrared temperature measurement unit is subjected to interference removal filtering within the interference coordinate range, and a preset filtering algorithm is adaptively called based on the interference intensity, and the filtering parameters corresponding to the filtering algorithm are dynamically adjusted.

2. The ladle cruise temperature measuring device according to claim 1, characterized by The preprocessing of the real-time environmental point cloud image acquired by the receiving lidar involves using threshold judgment and region growing algorithms to coarsely extract spray feedback features and form preliminary interference region point cloud clusters, including: Target detection region segmentation based on pass-through filtering; Calculate the mean and standard deviation of the distances to n neighboring points around each point cloud, remove outliers whose mean distance deviation exceeds the preset standard deviation, and retain the valid point cloud data; Traverse all valid point clouds, filter point clouds whose electric cloud parameters meet the preset threshold range, and mark them as candidate interference point clouds; Using any point in the candidate interference point cloud as a seed point, traverse its neighboring points to perform region growth, continuing until there are no neighboring points that meet the conditions, thus forming a preliminary interference region point cloud cluster.

3. The ladle cruise temperature measuring device according to claim 2, characterized in that, The step of matching features between a visible light region image and a real-time point cloud image of the environment to obtain parameters of the interference region includes: Perform grayscale conversion and Gaussian blurring on visible light images; The Otsu binarization algorithm is used to automatically determine the grayscale threshold and segment the image into interference regions and non-interference regions. The interference region point cloud clusters and the interference regions segmented from the visible light image are fused and verified to obtain the interference region parameters.

4. The slag bag cruise temperature measuring device according to claim 3, characterized in that, The step of adjusting the dual-spectrum infrared temperature measurement unit according to the interference coordinate range to achieve optimal temperature measurement angle adaptation includes: The current position coordinates and the current coordinates of the dual-spectrum infrared temperature measurement unit are collected in real time. Based on the digital twin model and the movement trajectory of the mobile carrier, the real-time coordinates of the target measuring point on the target slag bag are obtained. A virtual connection is established between the current coordinates of the simulated temperature measurement unit and the target measurement point on the current target slag bag, and the virtual connection synchronously compensates for the positional offset of the moving carrier. Spatial calculations are used to determine whether the virtual connection passes through the interference area; If the virtual connection passes through the interference area, the rotation and pitch angles are adjusted based on the position of the interference area until the virtual connection avoids the interference area.

5. The slag bag cruise temperature measuring device according to claim 3, characterized in that, The infrared image of the dual-spectrum infrared thermometer is subjected to interference-removing filtering within the interference coordinate range. Based on the interference intensity, a preset filtering algorithm is adaptively invoked, and the filtering parameters corresponding to the filtering algorithm are dynamically adjusted, including: If the interference intensity is within a preset first range, the adaptive moving average filtering algorithm is invoked, with the sliding window size set to n1 sampling points, the filtering coefficient α=q1, and the interference threshold set to ±w1℃. If the interference intensity is within a preset second range, the Kalman filter algorithm and the moving average fusion filter algorithm are invoked, the sliding window size is adjusted to n2 sampling points, the filter coefficient α=q2, the noise in the Kalman filter process, the observation noise and the interference intensity are proportional, and the interference threshold is adjusted to ±w2℃. If the interference intensity is within a preset third range, the wavelet threshold filtering algorithm and the Kalman filter fusion algorithm are invoked. The sliding window size is set to n3 sampling points, the filter coefficient α=q3, the wavelet basis is selected as db4, and the wavelet threshold is set to 1 / X of the current interference amplitude. The noise, observation noise and interference intensity in the Kalman filtering process are proportional. Where n1 < n2 < n3, q1 > q2 > q3, w1 > w2.

6. The slag bag cruise temperature measuring device according to claim 5, characterized in that, The control module is also configured to: After the interference filtering process, the effective temperature signal is compared with the historical temperature data of the slag bag and the on-site interference intensity data to calculate the filtering error. If the filtering error is ≤ T1℃, maintain the current filtering parameters; if the filtering error is > T1℃, automatically fine-tune the filtering coefficient and window size based on the preset ratio until the error meets the standard. Each adjusted parameter will be stored in the corresponding interference filtering strategy algorithm for continuous iteration to optimize the parameters.

7. The slag bag cruise temperature measuring device according to any one of claims 1 to 6, characterized in that, The slag bag cruise temperature measurement device also includes a path supplement light and a collection supplement light. The path supplement light is set on the front and rear sides of the moving carrier, and the field of view of the path supplement light is facing the moving path of the moving carrier. The field of view of the collection supplement light is facing the slag bag side. The control module is also configured to: When the brightness of the image in the visible light area is lower than a preset brightness value, the path fill light and the acquisition fill light are triggered to start supplementary lighting.

8. The slag bag cruise temperature measuring device according to any one of claims 1 to 6, characterized in that, The slag bag cruise temperature measurement device also includes a millimeter-wave radar, and the control module is further configured to: Before the mobile carrier moves, an environmental digital twin model is pre-built, and the dynamic path is calculated by combining the historical measured temperature and cooling cycle curve of each slag bag; The mobile carrier is controlled to move according to a dynamic path, while receiving feedback signals from the millimeter-wave radar. If the feedback signals meet preset avoidance conditions, an avoidance path is calculated and the mobile carrier is controlled to avoid the obstacle and then return to the dynamic path.

9. The slag bag cruise temperature measuring device according to any one of claims 4 to 6, characterized in that, The target measuring points are: starting from the bottom of the contour curve of each target slag bag, the first point, the second point, and the third point are selected at equal intervals from top to top.

10. A method for slag bag cruise temperature measurement, characterized in that, include: Before the mobile carrier moves, an environmental digital twin model is pre-built, and the dynamic path is calculated by combining the historical measured temperature and cooling cycle curve of each slag bag; The mobile carrier is controlled to move according to a dynamic path. During the journey, it receives real-time environmental point cloud images collected by the lidar for preprocessing. Threshold judgment and region growing algorithms are used to coarsely extract the spray feedback features to form a preliminary interference region point cloud cluster. The interference region parameters are obtained by matching the features of the visible light region image and the real-time point cloud image of the environment. The interference region parameters include the interference coordinate range and the interference intensity. The dual-spectrum infrared temperature measurement unit is adjusted according to the interference coordinate range to achieve optimal temperature measurement angle adaptation, and / or, the infrared image of the dual-spectrum infrared temperature measurement unit is subjected to interference removal filtering within the interference coordinate range, and based on the interference intensity, a preset filtering algorithm is adaptively called, and the filtering parameters corresponding to the filtering algorithm are dynamically adjusted. The system receives the measured temperature from the dual-spectrum infrared temperature measurement unit, compares the measured temperature with the safe temperature calculated from the cooling cycle curve of the corresponding slag bag, and triggers an alarm if the measured temperature exceeds the safe temperature.

Images