Method and apparatus for detecting a leak in a data center cold source station

Abstract

The application discloses a kind of data center cold source station leak detection method and device, it is related to pipeline leak detection technical field, the method comprises: the combined photo of the pipeline equipment in cold source station is collected;The combined photo includes visible light photo and infrared thermal imaging photo;Determine whether the combined photo appears abnormal point;For the position of abnormal point, repeat the combined photo collection, and analyze and determine whether the area of abnormal point changes;If change, stop the step of repeated collection of combined photo, and issue the alarm information that the abnormal point is pipeline equipment leak point;Until determining the area of abnormal point changes, or the collection interval of repeated collection of combined photo is greater than preset value;And when the time interval of repeated collection of combined photo is greater than preset value, issue the notification information that the abnormal point is pipeline equipment suspected leak point.The application is used to improve the detection efficiency, detection accuracy and detection timeliness of data center cold source station leak.

Description

Technical Field

[0001] This invention relates to the field of pipeline leak detection technology, and in particular to a method and apparatus for detecting leaks in data center cold source stations. Background Technology

[0002] This section is intended to provide background or context for the embodiments of the invention set forth in the claims. The description herein is not an admission that it is prior art simply because it is included in this section.

[0003] With the rapid development of the internet and information technology, and the proposal of new infrastructure, the construction of various data centers has accelerated, and they have become an indispensable and important infrastructure for social and economic development. However, the physical infrastructure of data centers is the foundation upon which information technology and application services depend. Among these, the data center cooling station is the main source of cooling for the data center; therefore, the operation and maintenance management of the data center cooling station is crucial.

[0004] Traditional data center chiller control systems use a single controller to collect data from various sensors installed on the chiller's piping to monitor and control related equipment within the chiller. However, traditional chiller monitoring has drawbacks:

[0005] First, traditional cold source stations collect and monitor information through sensors. However, sensor values ​​cannot be verified or compared, resulting in low accuracy. For example, when a cold source station leaks liquid occasionally due to equipment wear or aging, it is mainly detected by liquid detectors, which suffers from poor timeliness and large errors. Maintenance personnel mainly go to the cold source station during routine inspections to check whether there are any abnormalities in the equipment and pipelines.

[0006] Secondly, traditional cold source station operation and maintenance mainly relies on manual inspection, depending on the experience of the staff, and using sight and hearing to determine faults such as leaks in the cold source station equipment. However, data center cold source stations are large and the environment is more complex, and manual inspection has blind spots and cannot detect faults. Summary of the Invention

[0007] This invention provides a method for detecting leaks in data center cooling stations, which improves the detection efficiency, accuracy, and timeliness of data center cooling station leak detection, ensures the normal operation of data center cooling facilities, and reduces data center operating costs. The method includes:

[0008] Collect combined photographs of the pipeline equipment in the cold source station; the combined photographs include visible light photographs and infrared thermal imaging photographs.

[0009] Based on machine vision analysis and deep learning technologies, it is determined whether there are any abnormal points in the combined photos and the area of ​​any abnormal points; the abnormal points include: hot and cold spots in infrared thermal imaging photos and chromatic aberration points in visible light photos.

[0010] If an anomaly is found, the combined images are repeatedly captured at different acquisition intervals for the location of the anomaly, and the area of ​​the anomaly in the repeatedly captured combined images is determined; the analysis is performed to determine whether the area of ​​the anomaly has changed; the acquisition interval increases with the number of acquisitions.

[0011] If the situation changes, the process of repeatedly collecting and combining photos will stop, and an alarm message will be issued indicating that the abnormal point is a leak in the pipeline equipment.

[0012] If no change is observed, the process of repeatedly collecting combined photos and analyzing to determine whether the area of ​​the anomaly point has changed continues until it is determined that the area of ​​the anomaly point has changed, or the collection interval of the combined photos is greater than a preset value. When the time interval of the combined photos is greater than the preset value, a notification message is issued that the anomaly point is a suspected leak point in the pipeline equipment.

[0013] This invention also provides a data center cooling station leakage detection device to improve the detection efficiency, accuracy, and timeliness of data center cooling station leaks, ensuring the normal operation of data center cooling facilities and reducing data center operating costs. The device includes:

[0014] A combined photo acquisition module is used to acquire combined photos of pipeline equipment in a cold source station; the combined photos include visible light photos and infrared thermal imaging photos.

[0015] The anomaly analysis module is used to determine whether there are anomalies in the combined photos and the area of ​​the anomalies based on machine vision analysis technology and deep learning technology; the anomalies include: hot and cold spots in infrared thermal imaging photos and chromatic aberration points in visible light photos.

[0016] The repeated acquisition module is used to repeatedly acquire combined photos at different acquisition intervals for the location of the abnormal point if an abnormal point is found, and to determine the area of ​​the abnormal point in the repeatedly acquired combined photos; analyze and determine whether the area of ​​the abnormal point has changed; the acquisition interval increases with the increase of the number of acquisitions.

[0017] The alarm module is used to stop the repeated acquisition of combined photos if the situation changes, and to issue an alarm message that the abnormal point is a leak in the pipeline equipment.

[0018] The notification module is used to continuously execute the steps of repeatedly collecting combined photos and analyzing and determining whether the area of ​​the anomaly point has changed if no change is made, until it is determined that the area of ​​the anomaly point has changed, or the collection interval of repeatedly collecting combined photos is greater than a preset value; and when the time interval of repeatedly collecting combined photos is greater than the preset value, a notification message is issued that the anomaly point is a suspected leak point of the pipeline equipment.

[0019] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described data center cold source station leakage detection method.

[0020] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described data center cold source station leakage detection method.

[0021] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described data center cold source station leakage detection method.

[0022] In this embodiment of the invention, combined photographs of the piping equipment in the cold source station are acquired. These combined photographs include visible light photographs and infrared thermal imaging photographs. Based on machine vision analysis and deep learning technologies, it is determined whether the combined photographs contain anomalies and the area of ​​any anomalies. The anomalies include hot and cold spots in the infrared thermal imaging photographs and color difference spots in the visible light photographs. If anomalies are found, combined photographs are repeatedly acquired at different acquisition intervals for the location of the anomalies, and the area of ​​the anomalies in the repeatedly acquired combined photographs is determined. Analysis is performed to determine if the area of ​​the anomalies has changed. The acquisition interval increases with the number of acquisitions. If the interval changes, the step of repeatedly acquiring combined photographs is stopped, and an alarm message indicating a leak in the piping equipment is issued. If the interval does not change, the steps of repeatedly acquiring combined photographs and analyzing whether the area of ​​the anomalies has changed are continued until the area of ​​the anomalies is determined to have changed, or the acquisition interval of repeatedly acquiring combined photographs is greater than a preset value. During the repeated acquisition of combined photographs... When the interval exceeds a preset value, a notification is issued indicating that the abnormal point is a suspected leak point in the pipeline equipment. Compared with existing technologies that require sensors for leak detection and manual judgment of whether a leak has occurred, this method can collect visible light and infrared thermal images of the pipeline equipment in the cold source station. By using machine vision analysis and deep learning technologies, the collected images are analyzed and judged for anomalies, which can promptly and accurately determine whether a leak has occurred in the cold source station. This solves the problem of low accuracy caused by the inability to verify and compare sensor values ​​in existing technologies, and also solves the problem of misjudgment that is unavoidable due to manual judgment of pipeline equipment leaks in existing technologies. It improves the detection efficiency, accuracy, and timeliness of leaks in data center cold source stations, solves the problem that liquid detectors or manual inspections cannot detect leaks in cold source station pipeline equipment, improves the timeliness of fault detection in data center cold source stations, ensures the normal operation of data center cooling facilities, and reduces data center operating costs. Attached Figure Description

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

[0024] Figure 1 This is a flowchart illustrating a data center cold source station leakage detection method according to an embodiment of the present invention;

[0025] Figure 2 This is a specific example diagram of a cold source station inspection robot in an embodiment of the present invention;

[0026] Figure 3 This is a specific example diagram of a data center cold source station leakage detection method according to an embodiment of the present invention;

[0027] Figure 4 This is a specific example diagram of a data center cold source station leakage detection method according to an embodiment of the present invention;

[0028] Figure 5 This is a specific example diagram of a data center cold source station leakage detection method according to an embodiment of the present invention;

[0029] Figure 6 This is a structural example diagram of a data center cold source station leakage detection device according to an embodiment of the present invention;

[0030] Figure 7 This is a specific example diagram of a data center cold source station leakage detection device according to an embodiment of the present invention;

[0031] Figure 8 This is a specific example diagram of a data center cold source station leakage detection device according to an embodiment of the present invention;

[0032] Figure 9 This is a schematic diagram of a computer device used for leak detection in a data center cold source station according to an embodiment of the present invention. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0034] In this document, the term "and / or" merely describes a relationship, indicating that three relationships can exist. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Furthermore, the term "at least one" in this document means any combination of at least two of any one or more elements. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C.

[0035] In the description of this specification, the terms "comprising," "including," "having," and "containing" are open-ended terms, meaning that they include but are not limited to. The terms "an embodiment," "a specific embodiment," "some embodiments," and "for example," etc., refer to specific features, structures, or characteristics described in connection with that embodiment or example that are included in at least one embodiment or example of this application. 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, or characteristics described can be combined in any suitable manner in one or more embodiments or examples. The order of steps involved in the various embodiments is used to illustrate the implementation of this application, and the order of steps is not limited and can be adjusted appropriately as needed.

[0036] The acquisition, storage, use, and processing of data in this application all comply with the relevant provisions of national laws and regulations.

[0037] With the rapid development of the internet and information technology, and the proposal of new infrastructure, the construction of various data centers has accelerated, and they have become an indispensable and important infrastructure for social and economic development. However, the physical infrastructure of data centers is the foundation upon which information technology and application services depend. Among these, the data center cooling station is the main source of cooling for the data center; therefore, the operation and maintenance management of the data center cooling station is crucial.

[0038] Traditional data center chiller control systems use a single controller to collect data from various sensors installed on the chiller's piping to monitor and control related equipment within the chiller. However, traditional chiller monitoring has drawbacks:

[0039] First, traditional cold source stations collect and monitor information through sensors, but sensor values ​​cannot be verified or compared, which reduces the accuracy of the values.

[0040] Secondly, traditional cold source station operation and maintenance mainly relies on manual inspection, depending on the experience of the staff, and using sight and hearing to determine faults such as leaks in the cold source station equipment. However, data center cold source stations are large and the environment is more complex, and manual inspection has blind spots and cannot detect faults.

[0041] Specifically, when occasional leaks occur at the cold source station due to equipment wear or aging, they are mainly detected by liquid detectors, which have problems such as poor timeliness and large errors. Maintenance personnel mainly go to the cold source station during routine inspections to check whether there are any abnormalities in the equipment and pipelines.

[0042] "Leaks" refer to the phenomena of gas leaks, water spills, drips, and liquids occurring during the storage and transmission of liquids and gases due to aging equipment, poor management, and improper operation. This not only wastes resources but also leads to equipment corrosion, environmental pollution, and potential safety hazards. Data center cold source stations have numerous pipelines and two water systems—chilled water and cooling water—making inspections complex and requiring high operational stability.

[0043] To address the aforementioned issues, this invention provides a method for detecting leaks in data center cooling systems. This method improves the detection efficiency, accuracy, and timeliness of data center cooling system leak detection, ensuring the normal operation of data center cooling facilities and reducing data center operating costs. (See also...) Figure 1 The method may include:

[0044] Step 101: Collect combined photographs of the piping and equipment in the cold source station; the combined photographs include visible light photographs and infrared thermal imaging photographs;

[0045] Step 102: Based on machine vision analysis technology and deep learning technology, determine whether there are any abnormal points in the combined photos and the area of ​​any abnormal points; the above-mentioned abnormal points include: hot and cold spots in infrared thermal imaging photos and color difference points in visible light photos.

[0046] Step 103: If an anomaly is found, take a series of photos at different acquisition intervals at the location of the anomaly, and determine the area of ​​the anomaly in the repeated acquisition photos; analyze to determine whether the area of ​​the anomaly has changed; the acquisition interval increases with the number of acquisitions.

[0047] Step 104: If the situation changes, stop repeatedly collecting and combining photos, and issue an alarm message that the above-mentioned abnormal point is a leak in the pipeline equipment.

[0048] Step 105: If no change is observed, continue to perform the steps of repeatedly collecting combined photos and analyzing to determine whether the area of ​​the anomaly point has changed, until it is determined that the area of ​​the anomaly point has changed, or the collection interval of repeatedly collecting combined photos is greater than a preset value; and when the time interval of repeatedly collecting combined photos is greater than a preset value, issue a notification message that the above-mentioned anomaly point is a suspected leak point of the pipeline equipment.

[0049] In this embodiment of the invention, combined photographs of the piping equipment in the cold source station are acquired. These combined photographs include visible light photographs and infrared thermal imaging photographs. Based on machine vision analysis and deep learning technologies, it is determined whether the combined photographs contain anomalies and the area of ​​any anomalies. The anomalies include hot and cold spots in the infrared thermal imaging photographs and color difference spots in the visible light photographs. If anomalies are found, combined photographs are repeatedly acquired at different acquisition intervals for the location of the anomalies, and the area of ​​the anomalies in the repeatedly acquired combined photographs is determined. Analysis is performed to determine if the area of ​​the anomalies has changed. The acquisition interval increases with the number of acquisitions. If the interval changes, the step of repeatedly acquiring combined photographs is stopped, and an alarm message indicating a leak in the piping equipment is issued. If the interval does not change, the steps of repeatedly acquiring combined photographs and analyzing whether the area of ​​the anomalies has changed are continued until the area of ​​the anomalies is determined to have changed, or the acquisition interval of repeatedly acquiring combined photographs is greater than a preset value. During the repeated acquisition of combined photographs... When the interval exceeds a preset value, a notification is issued indicating that the above-mentioned abnormal point is a suspected leak point in the pipeline equipment. Compared with the existing technology that requires the installation of sensors for leak detection and manual judgment of whether a leak has occurred, this method can collect visible light photos and infrared thermal images of the pipeline equipment in the cold source station. By using machine vision analysis technology and deep learning technology, the collected photos are analyzed and judged for anomalies, which can promptly and accurately determine whether a leak has occurred in the cold source station. This solves the problem of low accuracy caused by the inability to verify and compare sensor values ​​in the existing technology. It also solves the problem of misjudgment that is inevitable due to manual judgment of pipeline equipment leaks in the existing technology. This improves the detection efficiency, accuracy and timeliness of leaks in data center cold source stations. It solves the problem that liquid detectors or manual inspections cannot detect leaks in the pipeline equipment of cold source stations, improves the timeliness of fault detection in data center cold source stations, ensures the normal operation of data center cooling facilities, and reduces data center operating costs.

[0050] In practice, the first step is to collect combined photographs of the pipeline equipment in the cold source station; the combined photographs include visible light photographs and infrared thermal imaging photographs.

[0051] In this embodiment, combined photographs of the piping and equipment in the cold source station are collected, including:

[0052] During the inspection of the machine room by the inspection robot at the cold source station, the robot simultaneously captured visible light photos and infrared thermal images of the pipeline equipment.

[0053] In this embodiment, a visible light camera can capture images of water droplets or water saplings present in the pipeline equipment when it leaks, while an infrared thermal imaging camera can capture images of the temperature difference between the leaking substance and the pipeline equipment at the leak point.

[0054] In the above embodiments, the inspection robot can be equipped with visible light cameras and infrared thermal imaging cameras. During continuous inspection of the cold source station, both the visible light cameras and infrared thermal imaging cameras can capture video and take pictures. Through video and picture means, and with the help of deep learning image recognition algorithms, liquid leakage scenarios within the monitoring range can be identified and analyzed in a timely manner. At the same time, the infrared thermal imaging camera can perform thermal imaging on the surface of pipeline equipment. In scenarios such as leaks, liquids or gases will generate temperature differences around the leak point. By analyzing and comparing the temperature changes on the pipeline surface, local hot and cold spots can be determined. Therefore, it has a high recognition rate for leak scenarios such as leaks and drips, and a low false alarm rate.

[0055] In one embodiment, it also includes:

[0056] During the inspection of the machine room by the inspection robot of the cold source station, the gas detection results of the pipeline equipment in the cold source station are collected by the gas sensor carried by the robot.

[0057] Collect photos of the combined piping and equipment in the cooling station, including:

[0058] When gas detection results indicate a leak, take a combined photograph of the piping and equipment in the cold source station.

[0059] In this embodiment, the inspection robot can also be equipped with gas sensors and temperature and humidity sensors to directly detect abnormal gas leaks in the pipeline during continuous inspection of the cold source station. By combining the gas sensors integrated on the robot, the location of the leak can be determined by the abnormal temperature of the leak point in infrared thermal imaging.

[0060] In one embodiment, it also includes:

[0061] During the inspection of the machine room by the inspection robot of the cold source station, the temperature and humidity detection results of the pipeline equipment in the cold source station are collected by the temperature and humidity sensor carried by the robot.

[0062] Collect photos of the combined piping and equipment in the cooling station, including:

[0063] When the temperature and humidity detection results do not meet the preset threshold, take a combined photo of the pipeline equipment in the cold source station.

[0064] In this embodiment, when the inspection robot is inspecting the machine room, it simultaneously takes visible light photos and infrared thermal images of the pipeline equipment, and stores video recordings. The gas sensor can detect the ambient gas concentration information of the cold source station, and the temperature and humidity sensor can detect the ambient temperature and humidity.

[0065] In the above embodiments, by using robots to collect data, the robot inspection system can analyze and process the monitoring screen in real time 24 / 7, accurately identify pipeline leaks, avoid production accidents, reduce the intensity and difficulty of manual inspection, ensure personnel safety and reduce operation and maintenance costs.

[0066] In practice, after collecting combined photos of the pipeline equipment in the cold source station, machine vision analysis technology and deep learning technology are used to determine whether there are any abnormal points in the combined photos and the area of ​​any abnormal points. The abnormal points include hot and cold spots in infrared thermal imaging photos and color difference points in visible light photos.

[0067] In this embodiment, the monitoring of leaks in the cold source station can utilize machine vision and deep learning technologies to monitor and photograph the pipeline equipment area of ​​the cold source station in real time. When a liquid leak is detected, the machine vision technology is used to process the pipeline images based on the visible light camera photos and infrared thermal imaging camera photos, extract image sample points for comparison, and use deep learning technology to perform algorithmic analysis on the processed photos to determine whether a leak has occurred.

[0068] In one embodiment, determining whether anomalies exist in the combined photos based on machine vision analysis and deep learning techniques includes:

[0069] Based on machine vision analysis technology, it is determined whether there are hot or cold spots in the infrared thermal imaging photos in the composite photos, and whether there are color differences in the visible light photos in the composite photos.

[0070] Based on deep learning technology, the above-mentioned combined photos are sent to the anomaly detection deep learning model, and the feedback results of the anomaly detection deep learning model are received. The anomaly detection deep learning model is trained using historical combined photo anomaly analysis data as training data. The historical combined photo anomaly analysis data includes: historical data of different combined photos, and the results of whether anomalies appear in the historical data of different combined photos.

[0071] When it is determined that there are hot or cold spots in the infrared thermal image of the combined image, color difference points in the visible light image of the combined image, or when the feedback results indicate that there are abnormal points in the combined image, the abnormal points in the combined image are identified.

[0072] In this embodiment, an anomaly detection deep learning model can be used to identify whether there are hot or cold spots in the infrared thermal imaging photos in the composite photos, and whether there are color difference points in the visible light photos in the composite photos. This helps to improve the detection efficiency, accuracy and timeliness of data center cold source station leaks.

[0073] In one embodiment, determining whether anomalies exist in the combined photos based on machine vision analysis and deep learning techniques includes:

[0074] If the infrared thermal image in the combined image does not show any hot or cold spots, the visible light image in the combined image does not show any color difference points, or the feedback results indicate that there are no abnormal points in the combined image, then the combined image is confirmed to be free of abnormal points, and a notification message indicating that there is no leakage in the pipeline equipment is issued.

[0075] For example, when taking visible light and infrared photos for the first time, machine vision is used to analyze the infrared photo for localized cold or hot spots, and the visible light photo for color differences. Simultaneously, deep learning technology is used to analyze anomalies in both photos. If no anomalies are found simultaneously, the pipeline equipment location inspection result is considered normal and leak-free.

[0076] In practice, based on machine vision analysis and deep learning technologies, after determining whether there are any abnormal points in the combined photos and the area of ​​any abnormal points, if abnormal points are found, the combined photos are repeatedly collected at different collection intervals for the location of the abnormal points, and the area of ​​the abnormal points in the repeatedly collected combined photos is determined; the analysis is then conducted to determine whether the area of ​​the abnormal points has changed; the above collection interval increases with the number of collections.

[0077] In the embodiment, the above-mentioned acquisition interval increases exponentially with the number of acquisitions.

[0078] For example, if we take the interval t1 between the first two shots, we can calculate the third shot period t2 = t1 × 2. After the interval t2, the third shot is taken. If a fourth shot is needed, the shooting period is extended. The method for calculating the extended shooting period is as follows:

[0079] Taking the interval t2 between the first two shots, calculate the fourth shot period t3 = t2 × 2. After the interval t3, perform the fourth shot. If repeated data acquisition is still required, extend the shooting period t4 = t3 × 2 again, and so on. The acquisition interval for this shot is t. n =t n-1 ×2.

[0080] In one embodiment, at different acquisition intervals, combined images are repeatedly acquired at the locations where anomalies occur, and the area of ​​the anomalies in the repeatedly acquired combined images is determined, such as... Figure 4 As shown, it includes:

[0081] Step 401: At different acquisition intervals, repeatedly acquire combined photos at the locations where anomalies occur;

[0082] Step 402: Perform image processing on the repeatedly acquired composite photos, and analyze the processed composite photos based on machine vision analysis technology and deep learning technology to determine whether there are any abnormal points and the area of ​​any abnormal points in the repeatedly acquired composite photos.

[0083] For example, if a second set of visible light and infrared photos is taken simultaneously, the pipeline images are processed, image sample points are extracted for comparison, and deep learning technology is used to perform algorithmic analysis on the processed photos, while also comparing them with the first set of photos. If the analysis shows that the area of ​​an anomaly point has increased or decreased, it is determined that a leak has occurred at this inspection point.

[0084] In one embodiment, analysis determines whether the area of ​​the outlier has changed, such as... Figure 5 As shown, it includes:

[0085] Step 501: Compare and analyze the area of ​​abnormal points in the current combined photos obtained by repeated acquisition with the area of ​​abnormal points in the previously acquired combined photos;

[0086] Step 502: If the comparative analysis determines that the area of ​​the outlier has not changed, then the analysis confirms that the area of ​​the outlier has not changed.

[0087] If the area of ​​the abnormal point is determined by comparative analysis, then the area of ​​the abnormal point is determined by analysis.

[0088] In practice, at different acquisition intervals, composite images are repeatedly acquired at the locations where anomalies appear, and the area of ​​the anomalies in the repeatedly acquired composite images is determined. Analysis is then performed to determine whether the area of ​​the anomalies has changed.

[0089] If the situation changes, the step of repeatedly collecting combined photos will be stopped, and an alarm message will be issued indicating that the above-mentioned abnormal point is a leak point in the pipeline equipment. If the situation does not change, the steps of repeatedly collecting combined photos and analyzing whether the area of ​​the abnormal point has changed will continue until it is determined that the area of ​​the abnormal point has changed, or the collection interval of repeatedly collecting combined photos is greater than a preset value. When the time interval of repeatedly collecting combined photos is greater than the preset value, a notification message will be issued indicating that the above-mentioned abnormal point is a suspected leak point in the pipeline equipment.

[0090] In this embodiment, when the time interval between repeatedly collecting combined photos is greater than a preset value, that is, when there are always abnormal points and multiple comparisons are consistent, a notification message is issued that the above-mentioned abnormal points are suspected leak points of pipeline equipment. For example, if a robot monitoring system is used, a leak abnormality reminder is issued to remind the operation and maintenance personnel to conduct on-site verification.

[0091] In one embodiment, it also includes:

[0092] The process of repeatedly collecting and combining photos and analyzing them to determine whether the area of ​​the anomaly point has changed continues until it is determined that the area of ​​the anomaly point has changed. At this point, the process of repeatedly collecting and combining photos is stopped, and an alarm message is issued indicating that the anomaly point is a leak in the pipeline equipment.

[0093] The following is a specific embodiment to illustrate the application of the method of the present invention. This embodiment combines a cold source station inspection robot to implement the above method.

[0094] In this specific embodiment, the method described above is implemented using a cold source station inspection robot, which also solves the following technical problems: Currently, most cold source stations rely mainly on manual inspections to detect leaks. On the one hand, inspectors must make full use of their eyes to carefully check every potential hazard. During inspections, they need to pay attention to whether there are any leaks in the instruments and valves, and deal with them in a timely manner to prevent problems before they occur and ensure safe production. On the other hand, inspectors also need to promptly detect changes in the odor of the equipment, whether there are any abnormal sounds, to determine the location of the fault or leak; and listen for any surge in the valves, which is also an important factor in determining whether the valves are in a normal state. They can also use their noses to determine whether there are any leaks during normal production by observing the physical properties of the materials, ensuring personal safety and the safety of the equipment.

[0095] However, manual inspection is inefficient, inaccurate, and untimely. Over-reliance on manual labor can easily lead to under-measurement or missed measurements. Furthermore, in the event of a dangerous gas or liquid leak, it can threaten the personal safety of the inspection personnel and the safety of the equipment.

[0096] This specific embodiment overcomes the shortcomings of the prior art and provides a robotic leak detection method for data center cold source stations. It solves the problem that liquid detectors or manual inspections could not detect leaks in cold source station pipeline equipment, improves the timeliness of fault detection in data center cold source stations, and ensures the normal operation of data center cooling facilities.

[0097] In this specific embodiment, the cold source station inspection robot may be simply referred to as a robot, such as... Figure 2As shown, by installing visible light cameras, infrared thermal imaging cameras, gas sensors, and temperature and humidity sensors, both visible light cameras and infrared thermal imaging cameras can capture video and images during continuous inspections of the cold source station. Through video and image capture, and using deep learning image recognition algorithms, liquid leakage scenarios within the monitored area can be quickly identified and analyzed. Simultaneously, the infrared thermal imaging camera can perform thermal imaging on the surface of pipeline equipment. In scenarios such as leaks, liquids or gases will generate temperature differences around the leak point. By analyzing and comparing the temperature changes on the pipeline surface, local hot and cold spots can be identified, resulting in a high recognition rate and low false alarm rate for leak scenarios such as leaks and drips. Furthermore, the gas sensors integrated on the robot can also directly monitor abnormal gas leaks in the pipeline, and by combining this with the abnormal temperature at the leak point from the infrared thermal imaging, the location of the leak can be determined.

[0098] This incorporates a specific implementation of a cold source station inspection robot. When implementing the above method, as follows: Figure 3 As shown, the steps may include the following:

[0099] (1) When the inspection robot is inspecting the machine room, it simultaneously takes visible light photos and infrared thermal images of the pipeline equipment, and stores video recordings. The gas sensor can detect the ambient gas concentration information of the cold source station, and the temperature and humidity sensor can detect the ambient temperature and humidity.

[0100] (2) First, visible light and infrared photos are taken. The infrared photos are analyzed by machine vision to check for local cold spots or hot spots, and the visible light photos are analyzed to check for color differences. At the same time, deep learning technology is used to analyze the abnormal points in both photos. If no abnormalities are found in both photos, the inspection result of the pipeline equipment location is normal and there is no leakage. If the algorithm analysis of one photo shows an abnormality, a second photo is taken for the local abnormal point.

[0101] (3) A second simultaneous visible light and infrared photograph is taken. Image processing is performed on the pipeline images, image sample points are extracted for comparison, and deep learning technology is used to perform algorithmic analysis on the processed photographs. Simultaneously, the images are compared with the first photograph. When the analysis shows that the area of ​​an anomaly point has increased or decreased, it is determined that a leak has occurred at this inspection point. When the comparison result shows that the area of ​​the anomaly point is consistent between the two photographs, the shooting cycle is extended, and a third photograph is taken.

[0102] (4) Calculate the extended shooting period: Take the interval t1 between the first two shooting sessions, and calculate the third shooting period t2 = t1 × 2. After the interval t2, take the third shot and judge the comparison results of the two images. If they are still consistent, extend the shooting period and take the fourth shot.

[0103] (5) Calculate the extended shooting cycle: Take the interval t2 between the first two shooting sessions, and calculate the fourth shooting cycle t3 = t2 × 2. After the interval t3, perform the fourth shooting and compare the two images with the previous shooting results. If the area changes, it is determined that the inspection result of this pipeline equipment location is abnormal and a leak has occurred, and a leak alarm is issued; if they are still consistent, the shooting cycle is extended again t4 = t3 × 2...t n =t n-1 ×2, take photos and compare the results.

[0104] (6) Set the inspection anomaly alert time period T, t n When the time interval is ≥T, if an anomaly persists and matches after multiple comparisons, the robot monitoring system will issue a leakage anomaly alert, reminding maintenance personnel to conduct on-site verification. Here, T can be 24 hours.

[0105] In this specific embodiment, a visible light camera can capture images of water droplets or condensation in the pipeline equipment when a leak occurs, while an infrared thermal imaging camera can capture the temperature difference between the leaking substance and the pipeline equipment at the leak point. The cold source station leak monitoring system utilizes machine vision and deep learning technologies to monitor and capture images of the pipeline equipment area in real time. When a liquid leak is detected, the system uses machine vision technology to process the pipeline images based on the visible light and infrared thermal imaging camera photos, extracting image sample points for comparison. Deep learning technology is then used to perform algorithmic analysis on the processed images to determine if a leak has occurred. When a leak is detected, the robot monitoring system can promptly issue an alarm and send the alarm information to the backend monitoring system to remind relevant personnel to handle the situation promptly, effectively improving the efficiency of leak inspection and enhancing the ability to manage safety risks in various areas of the data center cold source station. The robot inspection system can continuously analyze and process monitoring images in real time 24 / 7, accurately identifying pipeline leaks, preventing production accidents, reducing the intensity and difficulty of manual inspections, ensuring personnel safety, and reducing operation and maintenance costs.

[0106] For the challenging issue of leaks, the robot utilizes its onboard visible light and infrared cameras and algorithmic perception to precisely locate leak points, capture images, and send them to the robot's backend monitoring platform. The platform promptly displays the coordinates of the leak location and issues an alarm, enabling comprehensive detection and immediate handling of all leaks. Simultaneously, the robot's real-time operational data can be transmitted to the data center monitoring platform. On-duty personnel can monitor the operational status through the data window, directly control and operate the robot to analyze leak locations from multiple angles, thus achieving early detection and diagnosis of faults. In addition to leak detection, the robot can also perform unmanned autonomous inspections, meter readings, equipment infrared temperature measurement, intelligent obstacle avoidance, temperature and humidity detection, and environmental gas detection. This enables automatic fault diagnosis of data center cooling stations, intelligent operation and maintenance of environmental facilities, production safety protection, and hazard identification, facilitating unmanned inspections of data centers. Even with personnel shortages amidst the ongoing global pandemic, the robot can successfully complete the infrastructure operation and maintenance of data center cooling stations.

[0107] This invention introduces inspection robots, artificial intelligence, machine vision, and deep learning technologies into the fault monitoring and maintenance of data center cooling stations. It enables 24 / 7 monitoring of cooling station pipeline equipment, intelligent analysis of photo and video information, improved image data processing capabilities, and timely detection of pipeline equipment leaks. This significantly reduces the labor intensity of manual inspections, ensures personnel safety, and reduces data center operating costs.

[0108] Of course, it is understood that there may be other variations of the above detailed process, and all such variations should fall within the protection scope of this invention.

[0109] In this embodiment of the invention, combined photographs of the piping equipment in the cold source station are acquired. These combined photographs include visible light photographs and infrared thermal imaging photographs. Based on machine vision analysis and deep learning technologies, it is determined whether the combined photographs contain anomalies and the area of ​​any anomalies. The anomalies include hot and cold spots in the infrared thermal imaging photographs and color difference spots in the visible light photographs. If anomalies are found, combined photographs are repeatedly acquired at different acquisition intervals for the location of the anomalies, and the area of ​​the anomalies in the repeatedly acquired combined photographs is determined. Analysis is performed to determine if the area of ​​the anomalies has changed. The acquisition interval increases with the number of acquisitions. If the interval changes, the step of repeatedly acquiring combined photographs is stopped, and an alarm message indicating a leak in the piping equipment is issued. If the interval does not change, the steps of repeatedly acquiring combined photographs and analyzing whether the area of ​​the anomalies has changed are continued until the area of ​​the anomalies is determined to have changed, or the acquisition interval of repeatedly acquiring combined photographs is greater than a preset value. During the repeated acquisition of combined photographs... When the interval exceeds a preset value, a notification is issued indicating that the above-mentioned abnormal point is a suspected leak point in the pipeline equipment. Compared with the existing technology that requires the installation of sensors for leak detection and manual judgment of whether a leak has occurred, this method can collect visible light photos and infrared thermal images of the pipeline equipment in the cold source station. By using machine vision analysis technology and deep learning technology, the collected photos are analyzed and judged for anomalies, which can promptly and accurately determine whether a leak has occurred in the cold source station. This solves the problem of low accuracy caused by the inability to verify and compare sensor values ​​in the existing technology. It also solves the problem of misjudgment that is inevitable due to manual judgment of pipeline equipment leaks in the existing technology. This improves the detection efficiency, accuracy and timeliness of leaks in data center cold source stations. It solves the problem that liquid detectors or manual inspections cannot detect leaks in the pipeline equipment of cold source stations, improves the timeliness of fault detection in data center cold source stations, ensures the normal operation of data center cooling facilities, and reduces data center operating costs.

[0110] This invention also provides a data center cold source station leakage detection device, as described in the following embodiments. Since the principle by which this device solves the problem is similar to the data center cold source station leakage detection method, the implementation of this device can refer to the implementation of the data center cold source station leakage detection method; repeated details will not be elaborated further.

[0111] This invention also provides a data center cooling station leakage detection device to improve the detection efficiency, accuracy, and timeliness of data center cooling station leaks, ensuring the normal operation of data center cooling facilities and reducing data center operating costs. Figure 6 As shown, the device includes:

[0112] The combined photo acquisition module 601 is used to acquire combined photos of pipeline equipment in the cold source station; the combined photos include visible light photos and infrared thermal imaging photos.

[0113] The anomaly analysis module 602 is used to determine whether there are anomalies in the combined photos and the area of ​​the anomalies based on machine vision analysis technology and deep learning technology; the anomalies include hot and cold spots in infrared thermal imaging photos and chromatic aberrations in visible light photos.

[0114] The repeated acquisition module 603 is used to repeatedly acquire combined photos at different acquisition intervals for the location of the abnormal point if an abnormal point is found, and to determine the area of ​​the abnormal point in the repeatedly acquired combined photos; analyze and determine whether the area of ​​the abnormal point has changed; the above acquisition interval increases with the increase of the number of acquisitions.

[0115] The alarm module 604 is used to stop the repeated acquisition of combined photos if the situation changes, and to issue an alarm message that the above-mentioned abnormal point is a leak point in the pipeline equipment.

[0116] The notification module 605 is used to continuously execute the steps of repeatedly collecting combined photos and analyzing and determining whether the area of ​​the abnormal point has changed if there is no change, until it is determined that the area of ​​the abnormal point has changed, or the collection interval of repeatedly collecting combined photos is greater than a preset value; and when the time interval of repeatedly collecting combined photos is greater than the preset value, it issues a notification message that the above-mentioned abnormal point is a suspected leak point of the pipeline equipment.

[0117] In one embodiment, the combined photo acquisition module is specifically used for:

[0118] During the inspection of the machine room by the inspection robot at the cold source station, the robot simultaneously captured visible light photos and infrared thermal images of the pipeline equipment.

[0119] In one embodiment, the anomaly analysis module is specifically used for:

[0120] Based on machine vision analysis technology, it is determined whether there are hot or cold spots in the infrared thermal imaging photos in the composite photos, and whether there are color differences in the visible light photos in the composite photos.

[0121] Based on deep learning technology, the above-mentioned combined photos are sent to the anomaly detection deep learning model, and the feedback results of the anomaly detection deep learning model are received. The anomaly detection deep learning model is trained using historical combined photo anomaly analysis data as training data. The historical combined photo anomaly analysis data includes: historical data of different combined photos, and the results of whether anomalies appear in the historical data of different combined photos.

[0122] When it is determined that there are hot or cold spots in the infrared thermal image of the combined image, color difference points in the visible light image of the combined image, or when the feedback results indicate that there are abnormal points in the combined image, the abnormal points in the combined image are identified.

[0123] In one embodiment, the anomaly analysis module is specifically used for:

[0124] If the infrared thermal image in the combined image does not show any hot or cold spots, the visible light image in the combined image does not show any color difference points, or the feedback results indicate that there are no abnormal points in the combined image, then the combined image is confirmed to be free of abnormal points, and a notification message indicating that there is no leakage in the pipeline equipment is issued.

[0125] In one embodiment, the above-mentioned acquisition interval doubles as the number of acquisitions increases.

[0126] In one embodiment, the repetitive data acquisition module is specifically used for:

[0127] At different acquisition intervals, composite photos were repeatedly acquired at the locations where anomalies occurred;

[0128] Image processing is performed on repeatedly acquired composite photos. Based on machine vision analysis and deep learning techniques, the processed composite photos are analyzed to determine whether there are any outliers and the area of ​​any outliers.

[0129] In one embodiment, the repetitive data acquisition module is specifically used for:

[0130] The area of ​​abnormal points in the current combined photos obtained by repeated collection is compared and analyzed with the area of ​​abnormal points in the previously collected combined photos;

[0131] If the comparative analysis determines that the area of ​​the outlier has not changed, then the analysis confirms that the area of ​​the outlier has not changed.

[0132] If the area of ​​the abnormal point is determined by comparative analysis, then the area of ​​the abnormal point is determined by analysis.

[0133] In one embodiment, it also includes:

[0134] The second alarm module is used for:

[0135] The process of repeatedly collecting and combining photos and analyzing them to determine whether the area of ​​the anomaly point has changed continues until it is determined that the area of ​​the anomaly point has changed. At this point, the process of repeatedly collecting and combining photos is stopped, and an alarm message is issued indicating that the anomaly point is a leak in the pipeline equipment.

[0136] In one embodiment, such as Figure 7 As shown, it also includes:

[0137] Gas detection result acquisition module 701 is used for:

[0138] During the inspection of the machine room by the inspection robot of the cold source station, the gas detection results of the pipeline equipment in the cold source station are collected by the gas sensor carried by the robot.

[0139] The combined photo acquisition module is specifically used for:

[0140] When gas detection results indicate a leak, take a combined photograph of the piping and equipment in the cold source station.

[0141] In one embodiment, such as Figure 8 As shown, it also includes:

[0142] Temperature and humidity detection result acquisition module 801 is used for:

[0143] During the inspection of the machine room by the inspection robot of the cold source station, the temperature and humidity detection results of the pipeline equipment in the cold source station are collected by the temperature and humidity sensor carried by the robot.

[0144] The combined photo acquisition module is specifically used for:

[0145] When the temperature and humidity detection results do not meet the preset threshold, take a combined photo of the pipeline equipment in the cold source station.

[0146] This invention provides an embodiment of a computer device for implementing all or part of the above-described data center cold source station leakage detection method. The computer device specifically includes the following components:

[0147] The computer device comprises a processor, memory, a communications interface, and a bus; wherein the processor, memory, and communications interface communicate with each other via the bus; the communications interface is used to realize information transmission between related devices; the computer device can be a desktop computer, tablet computer, or mobile terminal, etc., and this embodiment is not limited to these. In this embodiment, the computer device can be implemented with reference to the embodiments for implementing a data center cold source station leakage detection method and the embodiments for implementing a data center cold source station leakage detection device, the contents of which are incorporated herein by reference, and repeated details will not be described again.

[0148] Figure 9 This is a schematic block diagram illustrating the system configuration of the computer device 1000 according to an embodiment of this application. Figure 9 As shown, the computer device 1000 may include a central processing unit 1001 and a memory 1002; the memory 1002 is coupled to the central processing unit 1001. It is worth noting that... Figure 9This is an example; other types of structures can also be used to supplement or replace this structure to achieve telecommunications functions or other functions.

[0149] In one embodiment, the data center cold source station leakage detection function can be integrated into the central processing unit 1001.

[0150] The central processing unit 1001 can be configured to perform the following control:

[0151] Collect combined photographs of the pipeline equipment in the cold source station; the combined photographs include visible light photographs and infrared thermal imaging photographs.

[0152] Based on machine vision analysis and deep learning technologies, it is determined whether there are any abnormal points in the combined photos and the area of ​​any abnormal points; the abnormal points include: hot and cold spots in infrared thermal imaging photos and chromatic aberration points in visible light photos.

[0153] If an anomaly is found, the combined images are repeatedly captured at different acquisition intervals for the location of the anomaly, and the area of ​​the anomaly in the repeatedly captured combined images is determined; the analysis is performed to determine whether the area of ​​the anomaly has changed; the acquisition interval increases with the number of acquisitions.

[0154] If the situation changes, the process of repeatedly collecting and combining photos will stop, and an alarm message will be issued indicating that the abnormal point is a leak in the pipeline equipment.

[0155] If no change is observed, the process of repeatedly collecting combined photos and analyzing to determine whether the area of ​​the anomaly point has changed continues until it is determined that the area of ​​the anomaly point has changed, or the collection interval of the combined photos is greater than a preset value. When the time interval of the combined photos is greater than the preset value, a notification message is issued that the anomaly point is a suspected leak point in the pipeline equipment.

[0156] In another embodiment, the data center cold source station leakage detection device can be configured separately from the central processing unit 1001. For example, the data center cold source station leakage detection device can be configured as a chip connected to the central processing unit 1001, and the data center cold source station leakage detection function can be realized through the control of the central processing unit.

[0157] like Figure 9 As shown, the computer device 1000 may further include: a communication module 1003, an input unit 1004, an audio processor 1005, a display 1006, and a power supply 1007. It is worth noting that the computer device 1000 does not necessarily need to include... Figure 9 All components shown; in addition, the computer device 1000 may also include Figure 9 For components not shown, please refer to existing technologies.

[0158] like Figure 9 As shown, the central processing unit 1001, sometimes also referred to as a controller or operation control, may include a microprocessor or other processor device and / or logic device. The central processing unit 1001 receives input and controls the operation of various components of the computer device 1000.

[0159] The memory 1002 may be, for example, one or more of a cache, flash memory, hard drive, removable medium, volatile memory, non-volatile memory, or other suitable device. It may store the aforementioned failure-related information, and also store a program for executing that information. The central processing unit 1001 may execute the program stored in the memory 1002 to perform information storage or processing, etc.

[0160] Input unit 1004 provides input to central processing unit 1001. This input unit 1004 may be, for example, a keypad or touch input device. Power supply 1007 provides power to computer device 1000. Display 1006 displays images, text, and other display objects. This display may be, for example, an LCD display, but is not limited to this.

[0161] The memory 1002 can be a solid-state memory, such as a read-only memory (ROM), random access memory (RAM), a SIM card, etc. It can also be a memory that retains information even when power is off, can be selectively erased, and contains more data; examples of this type of memory are sometimes referred to as EPROMs, etc. The memory 1002 can also be some other type of device. The memory 1002 includes a buffer memory 1021 (sometimes referred to as a buffer). The memory 1002 may include an application / function storage unit 1022 for storing application programs and function programs or processes for executing operations of the computer device 1000 via the central processing unit 1001.

[0162] The memory 1002 may also include a data storage unit 1023 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the computer device. The driver storage unit 1024 of the memory 1002 may include various drivers for the computer device for communication functions and / or for performing other functions of the computer device (such as messaging applications, address book applications, etc.).

[0163] The communication module 1003 is a transmitter / receiver 1003 that transmits and receives signals via the antenna 1008. The communication module (transmitter / receiver) 1003 is coupled to the central processing unit 1001 to provide input signals and receive output signals, which can be the same as in a conventional mobile communication terminal.

[0164] Based on different communication technologies, multiple communication modules 1003 can be configured in the same computer device, such as cellular network modules, Bluetooth modules, and / or wireless LAN modules. The communication module (transmitter / receiver) 1003 is also coupled to a speaker 1009 and a microphone 1010 via an audio processor 1005 to provide audio output via the speaker 1009 and receive audio input from the microphone 1010, thereby realizing typical telecommunications functions. The audio processor 1005 may include any suitable buffer, decoder, amplifier, etc. Additionally, the audio processor 1005 is also coupled to a central processing unit 1001, enabling on-device recording via the microphone 1010 and on-device playback of stored sound via the speaker 1009.

[0165] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described data center cold source station leakage detection method.

[0166] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described data center cold source station leakage detection method.

[0167] In this embodiment of the invention, combined photographs of the piping equipment in the cold source station are acquired. These combined photographs include visible light photographs and infrared thermal imaging photographs. Based on machine vision analysis and deep learning technologies, it is determined whether the combined photographs contain anomalies and the area of ​​any anomalies. The anomalies include hot and cold spots in the infrared thermal imaging photographs and color difference spots in the visible light photographs. If anomalies are found, combined photographs are repeatedly acquired at different acquisition intervals for the location of the anomalies, and the area of ​​the anomalies in the repeatedly acquired combined photographs is determined. Analysis is performed to determine if the area of ​​the anomalies has changed. The acquisition interval increases with the number of acquisitions. If the interval changes, the step of repeatedly acquiring combined photographs is stopped, and an alarm message indicating a leak in the piping equipment is issued. If the interval does not change, the steps of repeatedly acquiring combined photographs and analyzing whether the area of ​​the anomalies has changed are continued until the area of ​​the anomalies is determined to have changed, or the acquisition interval of repeatedly acquiring combined photographs is greater than a preset value. During the repeated acquisition of combined photographs... When the interval exceeds a preset value, a notification is issued indicating that the abnormal point is a suspected leak point in the pipeline equipment. Compared with existing technologies that require sensors for leak detection and manual judgment of whether a leak has occurred, this method can collect visible light and infrared thermal images of the pipeline equipment in the cold source station. By using machine vision analysis and deep learning technologies, the collected images are analyzed and judged for anomalies, which can promptly and accurately determine whether a leak has occurred in the cold source station. This solves the problem of low accuracy caused by the inability to verify and compare sensor values ​​in existing technologies, and also solves the problem of misjudgment that is unavoidable due to manual judgment of pipeline equipment leaks in existing technologies. It improves the detection efficiency, accuracy, and timeliness of leaks in data center cold source stations, solves the problem that liquid detectors or manual inspections cannot detect leaks in cold source station pipeline equipment, improves the timeliness of fault detection in data center cold source stations, ensures the normal operation of data center cooling facilities, and reduces data center operating costs.

[0168] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0169] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0170] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0171] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0172] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for detecting leaks in a data center cooling station, characterized in that, include: When the inspection robot inspects the computer room, it collects combined photos of the pipeline equipment in the cold source station; the combined photos include visible light photos and infrared thermal imaging photos. Based on machine vision analysis and deep learning technologies, it is determined whether there are any abnormal points in the combined photos and the area of ​​any abnormal points; the abnormal points include: hot and cold spots in infrared thermal imaging photos and chromatic aberration points in visible light photos. If an anomaly is found, the combined images are repeatedly captured at different acquisition intervals for the location of the anomaly, and the area of ​​the anomaly in the repeatedly captured combined images is determined; the analysis is performed to determine whether the area of ​​the anomaly has changed; the acquisition interval increases with the number of acquisitions. The process of repeatedly capturing combined images at different acquisition intervals for locations where anomalies occur, and determining the area of ​​the anomalies in the repeatedly captured combined images, includes: At different acquisition intervals, composite photos were repeatedly acquired at the locations where anomalies occurred; Image processing is performed on repeatedly acquired composite photos, and the processed composite photos are analyzed based on machine vision analysis technology and deep learning technology to determine whether there are any outliers and the area of ​​any outliers in the composite photos. If the situation changes, the process of repeatedly collecting and combining photos will stop, and an alarm message will be issued indicating that the abnormal point is a leak in the pipeline equipment. If no change is observed, the shooting cycle is extended to continuously perform the steps of repeatedly collecting combined photos and analyzing to determine whether the area of ​​the abnormal point has changed, until it is determined that the area of ​​the abnormal point has changed, or the collection interval of repeatedly collecting combined photos is greater than a preset value; and when the time interval of repeatedly collecting combined photos is greater than a preset value, a notification message is issued that the abnormal point is a suspected leak point of the pipeline equipment. During the inspection of the machine room by the inspection robot of the cold source station, the gas detection results of the pipeline equipment in the cold source station are collected by the gas sensor carried by the robot. Collect photos of the combined piping and equipment in the cooling station, including: When gas detection results indicate a leak, take a combined photograph of the piping and equipment in the cold source station; or During the inspection of the machine room by the inspection robot of the cold source station, the temperature and humidity detection results of the pipeline equipment in the cold source station are collected by the temperature and humidity sensor carried by the robot. Collect photos of the combined piping and equipment in the cooling station, including: When the temperature and humidity detection results do not meet the preset threshold, take a combined photo of the pipeline equipment in the cold source station.

2. The method as described in claim 1, characterized in that, Collect photos of the combined piping and equipment in the cooling station, including: During the inspection of the machine room by the cold source station inspection robot, the robot simultaneously captures visible light photos and infrared thermal images of the pipeline equipment.

3. The method as described in claim 1, characterized in that, Based on machine vision analysis and deep learning technologies, determine whether there are outliers in the composite photos, including: Based on machine vision analysis technology, it is determined whether there are hot or cold spots in the infrared thermal imaging photos in the composite photos, and whether there are color differences in the visible light photos in the composite photos. Based on deep learning technology, the combined photos are sent to an anomaly detection deep learning model, and the feedback results of the anomaly detection deep learning model are received; the anomaly detection deep learning model is trained using historical combined photo anomaly analysis data as training data; the historical combined photo anomaly analysis data includes: historical data of different combined photos, and the results of whether anomalies appear in the historical data of different combined photos. When it is determined that there are hot or cold spots in the infrared thermal imaging image of the combined image, color difference points in the visible light image of the combined image, or when the feedback result indicates that there are abnormal points in the combined image, the abnormal points in the combined image are determined.

4. The method as described in claim 3, characterized in that, Based on machine vision analysis and deep learning technologies, determine whether there are outliers in the composite photos, including: When it is determined that there are no hot or cold spots in the infrared thermal image of the combined image, no color difference in the visible light image of the combined image, or the feedback result indicates that there are no abnormal points in the combined image, it is determined that there are no abnormal points in the combined image, and a notification message indicating that there is no leakage in the pipeline equipment is issued.

5. The method as described in claim 1, characterized in that, The acquisition interval increases exponentially with the number of acquisitions.

6. The method as described in claim 1, characterized in that, The analysis determines whether the area of ​​outliers has changed, including: The area of ​​abnormal points in the current combined photos obtained by repeated collection is compared and analyzed with the area of ​​abnormal points in the previously collected combined photos; If the comparative analysis determines that the area of ​​the outlier has not changed, then the analysis confirms that the area of ​​the outlier has not changed. If the area of ​​the abnormal point is determined by comparative analysis, then the area of ​​the abnormal point is determined by analysis.

7. The method as described in claim 1, characterized in that, Also includes: The process of repeatedly collecting and combining photos and analyzing them to determine whether the area of ​​the anomaly point has changed continues until it is determined that the area of ​​the anomaly point has changed. At this point, the process of repeatedly collecting and combining photos stops, and an alarm message is issued indicating that the anomaly point is a leak in the pipeline equipment.

8. A data center cold source station leakage detection device, characterized in that, include: A combined photo acquisition module is used by the inspection robot to acquire combined photos of pipeline equipment in the cold source station during inspection of the computer room; the combined photos include visible light photos and infrared thermal imaging photos. The anomaly analysis module is used to determine whether there are anomalies in the combined photos and the area of ​​the anomalies based on machine vision analysis technology and deep learning technology; the anomalies include: hot and cold spots in infrared thermal imaging photos and chromatic aberration points in visible light photos. The repeated acquisition module is used to repeatedly acquire combined photos at different acquisition intervals for the location of the abnormal point if an abnormal point is found, and to determine the area of ​​the abnormal point in the repeatedly acquired combined photos; analyze and determine whether the area of ​​the abnormal point has changed; the acquisition interval increases with the increase of the number of acquisitions. The repeated acquisition module is specifically used for: At different acquisition intervals, composite photos were repeatedly acquired at the locations where anomalies occurred; Image processing is performed on repeatedly acquired composite photos, and the processed composite photos are analyzed based on machine vision analysis technology and deep learning technology to determine whether there are any outliers and the area of ​​any outliers in the composite photos. The alarm module is used to stop the repeated acquisition of combined photos if the situation changes, and to issue an alarm message that the abnormal point is a leak in the pipeline equipment. The notification module is used to extend the shooting cycle and continue to perform the steps of repeatedly collecting combined photos and analyzing and determining whether the area of ​​the abnormal point has changed if no change is found, until it is determined that the area of ​​the abnormal point has changed, or the collection interval of repeatedly collecting combined photos is greater than a preset value; and when the time interval of repeatedly collecting combined photos is greater than the preset value, a notification message is issued that the abnormal point is a suspected leak point of the pipeline equipment. The gas detection result acquisition module is used for: During the inspection of the machine room by the inspection robot of the cold source station, the gas detection results of the pipeline equipment in the cold source station are collected by the gas sensor carried by the robot. The combined photo acquisition module is specifically used for: When gas detection results indicate a leak, take a combined photograph of the piping and equipment in the cold source station; or The temperature and humidity detection result acquisition module is used for: During the inspection of the machine room by the inspection robot of the cold source station, the temperature and humidity detection results of the pipeline equipment in the cold source station are collected by the temperature and humidity sensor carried by the robot. The combined photo acquisition module is specifically used for: When the temperature and humidity detection results do not meet the preset threshold, take a combined photo of the pipeline equipment in the cold source station.

9. The apparatus as claimed in claim 8, characterized in that, The combined photo acquisition module is specifically used for: During the inspection of the machine room by the cold source station inspection robot, the robot simultaneously captures visible light photos and infrared thermal images of the pipeline equipment.

10. The apparatus as claimed in claim 8, characterized in that, The anomaly analysis module is specifically used for: Based on machine vision analysis technology, it is determined whether there are hot or cold spots in the infrared thermal imaging photos in the composite photos, and whether there are color differences in the visible light photos in the composite photos. Based on deep learning technology, the combined photos are sent to an anomaly detection deep learning model, and the feedback results of the anomaly detection deep learning model are received. The anomaly identification deep learning model is trained using historical composite photo anomaly analysis data as training data. The historical composite photo outlier analysis data includes: historical data of different composite photos, and the results of whether outliers appear in the historical data of different composite photos; When it is determined that there are hot or cold spots in the infrared thermal imaging image of the combined image, color difference points in the visible light image of the combined image, or when the feedback result indicates that there are abnormal points in the combined image, the abnormal points in the combined image are determined.

11. The apparatus as claimed in claim 10, characterized in that, The anomaly analysis module is specifically used for: When it is determined that there are no hot or cold spots in the infrared thermal image of the combined image, no color difference in the visible light image of the combined image, or the feedback result indicates that there are no abnormal points in the combined image, it is determined that there are no abnormal points in the combined image, and a notification message indicating that there is no leakage in the pipeline equipment is issued.

12. The apparatus as claimed in claim 8, characterized in that, The acquisition interval increases exponentially with the number of acquisitions.

13. The apparatus as claimed in claim 8, characterized in that, The repeat acquisition module is specifically used for: The area of ​​abnormal points in the current combined photos obtained by repeated collection is compared and analyzed with the area of ​​abnormal points in the previously collected combined photos; If the comparative analysis determines that the area of ​​the outlier has not changed, then the analysis confirms that the area of ​​the outlier has not changed. If the area of ​​the abnormal point is determined by comparative analysis, then the area of ​​the abnormal point is determined by analysis.

14. The apparatus as claimed in claim 8, characterized in that, Also includes: The second alarm module is used for: The process of repeatedly collecting and combining photos and analyzing them to determine whether the area of ​​the anomaly point has changed continues until it is determined that the area of ​​the anomaly point has changed. At this point, the process of repeatedly collecting and combining photos stops, and an alarm message is issued indicating that the anomaly point is a leak in the pipeline equipment.

15. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 7.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 7.

17. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method of any one of claims 1 to 7.

Images