Server fan fault diagnosis method and diagnosis equipment

Abstract

The invention discloses a server fan fault diagnosis method and diagnosis equipment, and relates to the technical field of data center server operation and maintenance. According to the method, on the basis of a server fan fault diagnosis method based on a power spectral density map, temperature characteristics of a server fan infrared thermal imaging map are combined; the method solves the problem that in practical application of a method based on a power spectral density map, different types of server fan faults may show similar power density curve characteristics in a low-frequency component region of the power spectral density map, so that the fault types are difficult to distinguish accurately, and improves the accuracy and reliability of server fan fault diagnosis. And a solid foundation is provided for the maintenance of the data center server.

Description

Technical Field

[0001] This application relates to the field of data center server operation and maintenance technology, and in particular to a diagnostic method and diagnostic device for server fan failure. Background Technology

[0002] With the continuous expansion of data center scale and the increasing density of servers, the reliability of server cooling systems is crucial for ensuring the stable operation of data centers. As a core component of the server cooling system, the health of server fans directly affects the normal operation and lifespan of the server. In practical applications, server fans may experience various types of failures, such as bearing wear, blade imbalance, motor failure, and shaft misalignment. If these failures are not detected and addressed in a timely manner, they may lead to server overheating, performance degradation, or even system crashes, resulting in data loss and business interruption. Therefore, accurately diagnosing server fan failures is a vital step in ensuring the stable operation of data centers.

[0003] Traditional server fan fault detection methods primarily rely on monitoring basic parameters such as fan current, status, and speed to determine if the fan is functioning correctly. However, this parameter-based approach can only quickly identify obvious faults, such as a complete fan stoppage, abnormally low speed, or abnormally high or low current, and cannot detect early-stage faults or non-electrical problems. To improve the accuracy of server fan fault diagnosis, a power spectral density (PSD)-based diagnostic method has been developed. This method collects the sound generated by the server fan's rotation and converts the sound signal into a PSD. By analyzing different frequency ranges of the PSD (including high-frequency components, low-frequency components, and the normal operating range), the health status of the server fan can be determined.

[0004] Although power spectral density map-based server fan fault diagnosis methods represent an improvement over traditional methods, in practical applications, different types of server fan faults may exhibit similar power density curve characteristics in the low-frequency component region of the power spectral density map, making it difficult to accurately distinguish the fault type. For example, bearing wear and blade imbalance exhibit similar power density curve characteristics in the low-frequency component region of the power spectral density map, but the actual causes and solutions are quite different. Summary of the Invention

[0005] This application provides a diagnostic method and device for server fan failures, which solves the problem that in existing server fan failure detection methods based on power spectral density maps, different types of server fan failures exhibit similar power density characteristics in the low-frequency component region of the power spectral density map, making it impossible to accurately distinguish the failure type.

[0006] In a first aspect, this application provides a method for diagnosing server fan failures, comprising: a diagnostic device dividing an infrared thermal image of a server fan into a core area and a peripheral area, wherein the core area is the region where the motor and bearing are located, and the peripheral area is the region where the blades and air duct are located; the diagnostic device extracting a core temperature difference index of the core area and a circumferential uniformity index of the peripheral area, wherein the core temperature difference index is the difference between the highest temperature point in the core area and the average temperature at the edge of the core area, and the circumferential uniformity index is the degree of temperature non-uniformity in the circumferential direction of the blade area; based on the power density curve characteristics of the low-frequency component region in the current power spectral density map corresponding to the sound signal generated by the rotation of the server fan, the diagnostic device determining that the server fan is a first type of failure, wherein the first type of failure includes at least blade imbalance failure and bearing wear failure; if the core temperature difference index is less than a first preset threshold and the circumferential uniformity index is greater than or equal to a second preset threshold, the diagnostic device diagnoses the current failure of the server fan as a blade imbalance failure; if the core temperature difference index is greater than or equal to the first preset threshold and the circumferential uniformity index is less than the second preset threshold, the diagnostic device diagnoses the current failure of the server fan as a bearing wear failure.

[0007] By employing the aforementioned technical solution, the acoustic and temperature characteristics of server fans are integrated. By dividing the infrared thermal image of the server fan into a core area and a peripheral area, clear spatial localization is provided for feature extraction of different fault types. The core temperature difference index extracted from the core area directly reflects the degree of thermal anomaly in the bearing and motor areas, while the circumferential uniformity index of the peripheral area accurately quantifies the non-uniformity of temperature distribution in the blade area. Based on the power density curve characteristics of the low-frequency component region of the current power spectral density map corresponding to the sound signal generated by the server fan rotation, the first fault type is first determined, and then the fault subtype is diagnosed through temperature characteristics, establishing a multi-dimensional fault diagnosis framework. This effectively solves the problem that different types of server fan faults exhibit similar power density curve characteristics in the low-frequency component region of the power spectral density map in practical applications, making it difficult to accurately distinguish specific fault types. It significantly improves the accuracy and reliability of fault diagnosis, providing a solid foundation for data center maintenance.

[0008] In conjunction with some embodiments of the first aspect, in some embodiments, the diagnostic device divides the infrared thermal image of the server fan into a core area and a peripheral area. Specifically, the diagnostic device divides the infrared thermal image of the server fan into a core area and a peripheral area based on the fan information built into the server. When a hot spot is detected in an area within the core area that has no obvious hot spot but contains a hot spot within 15% of the server fan radius, the diagnostic device expands the boundary of the core area to include the hot spot. When a temperature gradient is detected at the inner diameter boundary and the outer diameter boundary of the peripheral area that is greater than or equal to a preset threshold, the diagnostic device adjusts the inner diameter and outer diameter boundaries of the peripheral area accordingly.

[0009] By adopting the above technical solution, the diagnostic equipment first initializes the division of the core and peripheral areas of the infrared thermal image based on the fan information built into the server, ensuring the initial accuracy of the division. Furthermore, a dynamic adjustment mechanism is introduced: when a hot spot is detected within a 15% fan radius area but no obvious hot spot is detected in the core area, the core area boundary is automatically expanded to include the hot spot, ensuring that critical heat source information is not missed due to installation deviations or slight hot spot shifts in the early stages of a fault. Simultaneously, when a large temperature gradient appears at the inner and outer diameter boundaries of the peripheral area, the system adjusts the boundary range of the peripheral area accordingly. These dynamic adjustment strategies make the definition of the core and peripheral areas more precise and adaptive, ensuring that the extracted core temperature difference index and circumferential uniformity index more accurately reflect the actual thermal condition of the corresponding fan components, providing a high-quality data foundation, thereby improving the accuracy and stability of fault diagnosis.

[0010] In conjunction with some embodiments of the first aspect, in some embodiments, the diagnostic device extracts the core temperature difference index of the core area and the circumferential uniformity index of the surrounding area, specifically including: in the infrared thermal image of the core area, the diagnostic device extracts the highest temperature value among all pixels and the average temperature value among the edge pixels; the diagnostic device subtracts the average temperature value from the highest temperature value to obtain the core temperature difference index; the diagnostic device draws multiple rays from the center outward at equal angular intervals in the surrounding area; the diagnostic device extracts a set of sample points representing the temperature at different positions around the circumference from the rays or near the rays; the diagnostic device calculates the standard deviation of the sample points to obtain the circumferential uniformity index.

[0011] By employing the above technical solutions, the degree of localized heat concentration in the motor or bearing area caused by faults (such as increased friction due to bearing wear) can be intuitively and sensitively captured. The circumferential uniformity index, obtained through standard deviation, effectively quantifies the dispersion or non-uniformity of temperature distribution on the blade's plane of rotation, thus reflecting conditions such as blade imbalance, damage, or abnormal airflow. These specific quantitative calculation steps provide key indicators for distinguishing fault types, enhancing the feasibility and accuracy of the diagnostic method.

[0012] In conjunction with some embodiments of the first aspect, in some embodiments, after the diagnostic device calculates the standard deviation of the sample point to obtain the circumferential uniformity index, the method further includes: the diagnostic device using a low-pass filter to filter out high-frequency fluctuations in the core temperature difference index and the circumferential uniformity index acquired in consecutive multiple frames.

[0013] By adopting the above technical solution, the stability and reliability of temperature characteristic indicators are significantly improved. Applying low-pass filtering to both the core temperature difference index and the circumferential uniformity index collected from multiple consecutive frames improves the signal-to-noise ratio and temporal consistency of these two key indicators, making fault diagnosis results more stable and reliable, reducing the risk of false alarms and false negatives, and providing a more stable basis for accurately distinguishing server fan fault types. Especially in server operating environments with temperature fluctuations or airflow disturbances, this filtering process can eliminate short-term interference from environmental factors. By performing time-series processing on multiple consecutive frames of data, a smooth indicator change curve is formed, avoiding potential misjudgments from single-frame data.

[0014] In conjunction with some embodiments of the first aspect, in some embodiments, the method is characterized in that, after the step of the diagnostic device using a low-pass filter to filter out high-frequency fluctuations in the core temperature difference index and the circumferential uniformity index collected in multiple consecutive frames, the method further includes: based on the core temperature difference index and the circumferential uniformity index when the server fan is in a fault-free operating state, the diagnostic device establishes a dynamic health status database; the diagnostic device determines a first preset threshold and a second preset threshold based on the dynamic health status database, the first preset threshold being the arithmetic mean of the core temperature difference index plus three times the standard deviation, and the second preset threshold being the arithmetic mean of the circumferential uniformity index plus three times the standard deviation.

[0015] By adopting the above technical solution, and dynamically adjusting the first and second preset thresholds through the constructed dynamic health status database, the judgment criteria can be dynamically adjusted based on the actual health operation data of the fan, rather than relying on fixed empirical values. This better adapts to individual differences in fans of different models, batches, and even aging degrees, significantly improving the rationality and accuracy of fault detection thresholds, thereby reducing the risk of missed or false alarms due to improper threshold settings. Secondly, determining the preset thresholds based on statistical principles ensures that, under the assumption of normal distribution, approximately 99.7% of normal status data fall within the thresholds, providing a reliable statistical basis for fault judgment.

[0016] In conjunction with some embodiments of the first aspect, in some embodiments, the method further includes, after the diagnostic device determines the first preset threshold and the second preset threshold based on the dynamic health status database, the method further includes: when the current power spectral density map corresponding to the sound signal generated by the server fan indicates normal operation and the core temperature difference index and the circumferential uniformity index are abnormal, the diagnostic device determines the server fan to be a second fault type, wherein the abnormality refers to the core temperature difference index being lower than the first preset threshold in multiple consecutive frames and the difference from the first preset threshold gradually increasing to 0, or the circumferential uniformity index being lower than the second preset threshold in multiple consecutive frames and the difference from the second preset threshold gradually increasing to 0, and the second fault type includes at least early blade imbalance fault and early bearing wear fault; when the core temperature difference index is determined to be normal and the circumferential uniformity index is abnormal, the diagnostic device diagnoses the current fault of the server fan as an early blade imbalance fault; when the core temperature difference index is determined to be abnormal and the circumferential uniformity index is normal, the diagnostic device diagnoses the current fault of the server fan as an early bearing wear fault.

[0017] By employing the above technical solution, in certain situations, such as the early stages of a fault, the power spectral density map corresponding to the sound signal generated by the fan rotation may still show a normal operating state, which is difficult to detect with existing technologies. However, at this time, the core temperature difference index and the circumferential uniformity index may have begun to show persistent, slight abnormal deviations. For example, they may be below a preset threshold for multiple consecutive frames, and the difference from the preset threshold may gradually increase to 0. By monitoring this subtle but trending change in thermal characteristics, it can be determined that the fan has entered the second type of fault, namely the early fault stage. Furthermore, by determining whether the abnormality is in the core temperature difference index or the circumferential uniformity index, a preliminary distinction between early fault types is achieved, allowing maintenance personnel to take intervention measures before the fault worsens.

[0018] In conjunction with some embodiments of the first aspect, in some embodiments, the method further includes: when the core temperature difference index is greater than or equal to a first preset threshold and the circumferential uniformity index is greater than or equal to a second preset threshold, the diagnostic device diagnoses the current server fan's fault as a composite fault, the composite fault being at least simultaneously a bearing wear fault and a blade imbalance fault; when the absolute value of the difference between the core temperature difference index and the first preset threshold is greater than the absolute value of the difference between the circumferential uniformity index and the second preset threshold, the diagnostic device diagnoses the current server fan's primary fault as a bearing wear fault and the secondary fault as a blade imbalance fault; when the absolute value of the difference between the core temperature difference index and the first preset threshold is less than the absolute value of the difference between the circumferential uniformity index and the second preset threshold, the diagnostic device diagnoses the current server fan's primary fault as a blade imbalance fault and the secondary fault as a bearing wear fault; when the absolute value of the difference between the core temperature difference index and the first preset threshold is equal to the absolute value of the difference between the circumferential uniformity index and the second preset threshold, the diagnostic device diagnoses the current server fan as simultaneously having a blade imbalance fault and a bearing wear fault, and the fault degrees are the same.

[0019] By adopting the above technical solution, the absolute values ​​of the core temperature difference index and the first preset threshold difference, as well as the absolute values ​​of the circumferential uniformity index and the second preset threshold difference, are compared to accurately distinguish the relative severity of each component fault in a compound fault and determine the primary and secondary relationships of each fault. This provides maintenance personnel with more refined and guiding diagnostic conclusions, enabling them to formulate better maintenance priorities and resource allocation strategies, and significantly improving the efficiency and effectiveness of fault handling.

[0020] In a second aspect, embodiments of this application provide a diagnostic device, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, the computer program code including computer instructions, and the one or more processors call the computer instructions to cause the diagnostic device to perform the method described in the first aspect and any possible implementation thereof.

[0021] Thirdly, embodiments of this application provide a computer program product containing instructions that, when the computer program product is run on a diagnostic device, cause the diagnostic device to perform the method described in the first aspect and any possible implementation thereof.

[0022] Fourthly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a diagnostic device, cause the diagnostic device to perform the method described in the first aspect and any possible implementation thereof.

[0023] Understandably, the diagnostic device provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the methods provided in the embodiments of this application. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. By integrating the acoustic and temperature characteristics of server fans, and dividing the infrared thermal image of the server fan into a core area and a peripheral area, a clear spatial localization is provided for feature extraction of different fault types. The core temperature difference index extracted from the core area directly reflects the degree of thermal anomaly in the bearing and motor areas, while the circumferential uniformity index of the peripheral area accurately quantifies the non-uniformity of temperature distribution in the blade area. Based on the power spectral density map, the first fault type is determined first, and then the fault subtype is diagnosed through temperature characteristics, establishing a multi-dimensional fault diagnosis method. This effectively solves the problem that in practical applications, different types of server fan faults based on power spectral density maps exhibit similar power density curve characteristics in the low-frequency component region of the power spectral density map, making it difficult to accurately distinguish specific fault types. This significantly improves the accuracy and reliability of fault diagnosis, providing a solid foundation for data center maintenance.

[0025] 2. By initializing the division of the core and peripheral areas of the infrared thermal imaging image based on the server's built-in fan information, the initial accuracy of the division is ensured. Furthermore, a dynamic adjustment mechanism is introduced. When a hotspot is detected within a 15% fan radius area but no obvious hotspot is detected in the core area, the core area boundary is automatically expanded to include the hotspot. This ensures that critical heat source information is not missed due to installation deviations or slight hotspot shifts in the early stages of a fault. Simultaneously, when a large temperature gradient appears at the inner and outer diameter boundaries of the peripheral area, the system adjusts the boundary range accordingly. These dynamic adjustment strategies make the definition of the core and peripheral areas more precise and adaptive, ensuring that the extracted core temperature difference index and circumferential uniformity index more accurately reflect the actual thermal condition of the corresponding fan components. This provides a high-quality data foundation, thereby improving the accuracy and stability of fault diagnosis.

[0026] 3. By monitoring subtle changes in the core temperature difference index and circumferential uniformity index when the power spectral density map corresponding to the sound signal generated by the fan is displayed as normal operation, the diagnostic equipment can determine that the fan has entered the second fault type, i.e., the early fault stage, when both the core temperature difference index and the circumferential uniformity index are below the preset threshold for multiple consecutive frames and the difference from the preset threshold gradually increases to 0. Furthermore, by determining whether the abnormality is in the core temperature difference index or the circumferential uniformity index, a preliminary distinction between early fault types is achieved, allowing maintenance personnel to take intervention measures before the fault worsens. Attached Figure Description

[0027] Figure 1 This is a topology diagram of existing methods for diagnosing server fan failures.

[0028] Figure 2 This is a flowchart illustrating a method for diagnosing server fan failure in an embodiment of this application.

[0029] Figure 3 This is another flowchart illustrating a method for diagnosing server fan failure in an embodiment of this application; Figure 4 This is a schematic diagram of the physical device structure of a diagnostic device in the embodiments of this application. Detailed Implementation

[0030] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to and includes any or all possible combinations of one or more of the listed items.

[0031] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0032] Since the embodiments of this application involve the application of data center server operation and maintenance technology, for ease of understanding, the relevant terms and concepts involved in the embodiments of this application will be introduced below.

[0033] 1. MEMS microphone: MEMS microphones, short for Micro-Electro-Mechanical Systems microphones, are also often referred to as microphone chips or silicon microphones. They are miniature acoustic sensors that convert acoustic signals into electrical signals.

[0034] In this embodiment, a MEMS microphone is used to convert acoustic signals into electrical signals. Compared to traditional microphones, it offers a wider acquisition bandwidth, greater onboard stability, improved noise cancellation performance, and better RF (Radio Frequency) and EMI (Electromagnetic Interference) suppression capabilities. Compared to vibration sensors, it is better able to detect the overall fan noise level. MEMS (Micro-Electro-Mechanical Systems) microphones are manufactured using MEMS technology; simply put, it's a microphone with a capacitor integrated onto a micro-silicon chip. It can be manufactured using surface-mount technology, can withstand high reflow soldering temperatures, and is easily integrated with CMOS (Complementary Metal Oxide Semiconductor) processes and other audio circuits.

[0035] 2. ADC (Analog-to-Digital Converter): An ADC (Analog-to-Digital Converter) is an electronic device or integrated circuit whose core function is to convert continuously changing analog signals into discrete digital signals.

[0036] 3. BMC (Baseboard Management Controller): BMC, short for Baseboard Management Controller, is a dedicated microcontroller or small computer embedded on the motherboard of a server, workstation, or other complex computing device.

[0037] In this embodiment, the BMC ensures the system is in normal operation by monitoring the system's power supply, temperature, etc. It mainly plays a management and maintenance role in the server system. It is an independent system that does not depend on other hardware on the system (such as the central processing unit, memory, etc.) or the BIOS (basic input / output system) or OS (operating system). However, the BMC can interact with the BIOS and OS, which can play a better platform management role. Since the BMC itself is a small system with an external processor, it can also read the voltage timing signal acquired by the ADC independently.

[0038] This application provides a method and device for diagnosing server fan failures, which addresses the problem that in practical applications, different types of server fan failures may exhibit similar power density curve characteristics in the low-frequency component region of the power spectral density diagram, making it difficult to accurately distinguish the failure type.

[0039] Please see Figure 1 , Figure 1 This is a topology diagram of a current method for diagnosing server fan failures.

[0040] like Figure 1 As shown, existing technology uses a MEMS microphone to collect the sound signal generated by the server fan's rotation. This signal is then amplified and filtered through a signal conditioning link to become a voltage timing signal that an ADC can acquire. The BMC (Browser Controlled Component) then reads the ADC data. A cloud server receives this voltage timing signal from the BMC and uses a feature extraction module to extract sound characteristics, generating a power spectral density (PSD) map of the voltage timing signal. This PSD map is then divided into frequency ranges, and the PSD range corresponding to different health states is analyzed to divide the PSD map into multiple regions, including a high-frequency component region, a low-frequency component region, and a normal operating region. The cloud server's prediction module compares the current PSD of the server fan with each of these multiple regions to determine the current health state of the server fan.

[0041] Below Figure 1 Based on the existing server fan failure diagnosis methods shown, this application describes a server fan failure diagnosis method in its embodiments: Please see Figure 2 This is a flowchart illustrating a method for diagnosing server fan failure in an embodiment of this application.

[0042] S201. The diagnostic equipment divides the infrared thermal image of the server fan into a core area and a peripheral area.

[0043] The core area is where the motor and bearings are located, mainly including the fan's motor and bearing components. These components will produce obvious thermal characteristics when specific faults occur (such as bearing wear or motor overheating). The peripheral area is where the blades and air ducts are located, mainly including the fan blades and the surrounding air ducts. Abnormal temperature distribution in these areas is usually related to blade imbalance, foreign object interference, or airflow problems.

[0044] Specifically, since front and rear ventilation is a standard design feature of modern server racks, the diagnostic equipment first receives infrared thermal images of the server fans from an infrared thermal imager installed in the center of the rear door of the rack (covering the most server units). To distinguish between the core and peripheral areas, the diagnostic equipment utilizes the server's built-in fan information, typically stored in the server's Baseboard Management Controller (BMC) or system firmware. Based on parameters such as the fan diameter, bearing diameter, motor housing diameter, distance from the bearing to the blade tip, and blade length, the diagnostic equipment proportionally maps the physical dimensions to pixel dimensions, thereby dividing the server fan's infrared thermal image into core and peripheral areas.

[0045] Understandably, there are many ways to divide the infrared thermal image of a server fan into a core area and a peripheral area: In some embodiments, a method for dynamically adjusting the core area and the surrounding area can be adopted: Specifically, one method for dynamically adjusting the core area is as follows: During the initial wear of the bearing, hot spots may deviate from the core area. Without adaptive adjustment, these critical hot spots might be incorrectly classified as peripheral features, leading to a misjudgment of blade imbalance. Therefore, the diagnostic equipment continuously monitors the core area. When it detects no obvious hot spots within the core area but hot spots exist around it, it expands the core area boundary to include the hot spot. In this method, to ensure that the core area is not excessively expanded to include the blade region, a ring-shaped area bordering the core area, equal to 15% of the server fan radius, is selected.

[0046] One method for dynamically adjusting the peripheral zone is as follows: The diagnostic equipment first calculates the radial temperature gradient of the heat map, which is the rate of temperature change along the direction extending outward from the fan center. Then, the gradient peak near the core area is determined as the inner diameter of the peripheral zone, and the gradient peak near the edge of the server fan is determined as the outer diameter. A peripheral ring-shaped region is then defined based on these inner and outer diameters. For example, blade imbalance faults can cause uneven temperature gradient distribution at the junction of the blade and the air duct; while air duct blockage can alter the overall distribution of the peripheral temperature gradient. By dynamically adjusting the peripheral zone boundary, the system can more accurately capture fault characteristics.

[0047] It is understandable that other methods can be used to adjust the core area and the surrounding area to collect more accurate temperature characteristics, such as adjusting the core area and the surrounding area according to the specific position of the server fan with a special structure. This is not limited here.

[0048] S202, the diagnostic equipment extracts the core temperature difference index of the core area and the circumferential uniformity index of the surrounding area.

[0049] The core temperature difference index (CTDI) is an indicator used to quantify the intensity of hot spots in the core area, reflecting whether there is abnormal heating in the fan bearings or motor components. This index is calculated by the difference between the highest temperature point within the core area and the average temperature at the core edge. A higher CTDI usually indicates overheating of core components due to bearing wear or motor failure.

[0050] The circumferential uniformity index is used to quantify whether the temperature distribution in a blade area is uniform along the circumference. When a blade is unbalanced, damaged, or has foreign objects attached, it will cause uneven airflow and vibration during rotation, resulting in irregular fluctuations in the temperature distribution in the blade area along the circumference. The higher the circumferential uniformity index, the more uneven the temperature distribution.

[0051] Specifically, the process by which the diagnostic equipment extracts these two core temperature difference indices and the circumferential uniformity index is as follows: (1) Extraction of the core temperature difference index: 1. Identify the highest temperature point in the core area. Specifically, within the defined core area, the diagnostic equipment scans the temperature values ​​of all pixels. To ensure accurate identification, filtering is first performed to reduce noise interference, and then the highest temperature point and its temperature value are precisely located by finding the maximum value. 2. Calculate the average temperature at the edge of the core area. Specifically, the diagnostic device defines an annular band with a width of 10%-15% of the core area radius inside the core area boundary and identifies all pixels falling within this annular band. This annular band represents the transition area between the core area and the surrounding area. To ensure the representativeness of the calculation, the diagnostic device selects a point every 15° along the annular band, for a total of 24 sampling points, and calculates the average temperature value of the sampling points. 3. Calculate the core temperature difference index. Specifically, subtract the average temperature value at the edge of the core from the temperature value at the highest temperature point in the core area. The difference is the core temperature difference index.

[0052] (2) Extraction of the circumferential uniformity index: 1. The diagnostic equipment defines a circular sampling path within the peripheral area based on the midpoint between the inner and outer diameters of the peripheral area; 2. The diagnostic equipment samples uniformly along a circular sampling path at 1° intervals, acquiring 360 temperature sampling points; 3. The diagnostic equipment calculates the standard deviation of the temperature values ​​at this set of sample points. This standard deviation directly reflects the dispersion or non-uniformity of the temperature distribution in the circumferential direction of the blade area. The larger the standard deviation, the more non-uniform the temperature distribution, and the higher the circumferential uniformity index.

[0053] It is understandable that other methods can be used to extract the core temperature difference index of the core area and the circumferential uniformity index of the surrounding area, such as the ray-based circumferential uniformity index extraction method, which is not limited here.

[0054] S203. Based on the power density curve characteristics of the low-frequency component region in the current power spectral density map corresponding to the sound signal generated by the rotation of the server fan, the diagnostic device determines that the server fan is a first type of fault.

[0055] The sound signal is generated by the mechanical vibration and airflow disturbance during fan operation. Fans in different health states and with different fault types will produce sounds with different characteristics. The power spectral density plot is a method of converting the time-domain sound signal to the frequency domain for analysis; it shows the distribution of signal power at different frequencies. The power density curve characteristics in the low-frequency component region are usually related to macroscopic mechanical faults such as overall fan imbalance and bearing problems. The first fault type refers to a group of highly probable fault modes identified after preliminary analysis of the low-frequency characteristics of the power spectral density plot.

[0056] Based on the power density curve characteristics of the low-frequency component region in the current power spectral density map corresponding to the sound signal generated by the server fan rotation, the step of the diagnostic device determining that the server fan has a first fault type is based on, for example: Figure 1 The specific implementation method of the aforementioned existing technology is as follows: 1. The sound signal generated by the server fan rotation is collected by the MEMS microphone and amplified and filtered by the signal conditioning link into a voltage timing signal that can be collected by the ADC. Then the BMC reads the data of the ADC and receives the voltage timing signal sent by the BMC through the cloud server. 2. The cloud server extracts sound characteristics through the feature extraction module, generates the power spectral density map of the voltage time sequence signal, divides the power spectral density map into frequency intervals, and analyzes the power spectral density range corresponding to different health states, so as to divide the power spectral density map into multiple regions, including high frequency component region, low frequency component region and normal operation region. 3. When similar power density curve characteristics are exhibited in the low-frequency component region of the power spectral density map, the cloud server has difficulty accurately distinguishing the fault type. This situation is classified as "Type 1 Fault", indicating that the fan has one or more faults (such as bearing wear faults and blade imbalance faults), and further differentiation is required by combining thermal characteristics.

[0057] S204. The diagnostic device determines whether the server fan is in a situation where the core temperature difference index is less than the first preset threshold and the circumferential uniformity index is greater than or equal to the second preset threshold.

[0058] The first preset threshold is a judgment threshold set for the core temperature difference index, used to distinguish whether the temperature in the core area is significantly abnormal. The second preset threshold is a judgment threshold set for the circumferential uniformity index, used to distinguish whether the temperature distribution in the surrounding area is significantly uneven.

[0059] In some embodiments, a method for determining a first preset threshold and a second preset threshold based on a dynamic health status database can be adopted: Specifically, based on the core temperature difference index and circumferential uniformity index of the server fan in a fault-free operating state, the diagnostic equipment establishes a dynamic health status database. This database includes fan identification information, time series data, historical data of the core temperature difference index, historical data of the circumferential uniformity index, and the arithmetic mean, standard deviation, maximum value, and minimum value of the core temperature difference index, as well as the arithmetic mean, standard deviation, maximum value, and minimum value of the circumferential uniformity index. This data is stored in a structured format using a time-series database structure, supporting efficient time-series query and statistical analysis. It facilitates the calculation of dynamic thresholds and automatically updates threshold parameters as new data accumulates, achieving adaptive optimization of diagnostic criteria.

[0060] The diagnostic equipment determines preset thresholds based on statistical principles. The first preset threshold is the arithmetic mean of the core temperature difference index stored in the dynamic health status database plus three times the standard deviation. The second preset threshold is the arithmetic mean of the circumferential uniformity index stored in the dynamic health status database plus three times the standard deviation. This method ensures that, under the assumption of normal distribution, approximately 99.7% of the normal state data are within the thresholds, providing a reliable statistical basis for fault diagnosis.

[0061] It is understandable that other methods for more accurately determining the first and second preset thresholds can also be used, and no limitation is made here.

[0062] After determining that the current server fan is the first type of fault, the diagnostic device determines whether the server fan is in a situation where the core temperature difference index is less than the first preset threshold and the circumferential uniformity index is greater than or equal to the second preset threshold. If so, step S205 is executed; otherwise, step S206 is executed.

[0063] S205. The diagnostic equipment diagnoses the current server fan's fault as a blade imbalance fault.

[0064] Specifically, if the server fan is found to have a core temperature difference index less than a first preset threshold and a circumferential uniformity index greater than or equal to a second preset threshold, the diagnostic device diagnoses the current server fan's fault as a blade imbalance fault.

[0065] Specifically, when the core temperature difference index is less than the first preset threshold, it indicates that there is no significant concentrated overheating in the core area of ​​the fan (motor and bearings), ruling out abnormalities in the low-frequency component region of the power spectral density diagram caused by severe bearing wear or motor failure. Simultaneously, the circumferential uniformity index is greater than or equal to the second preset threshold, indicating that the temperature distribution around the fan blades / duct exhibits significant non-uniformity in the circumferential direction. This non-uniformity is a typical thermal characteristic of blade imbalance, as unbalanced blades lead to uneven airflow disturbance and vibration, which is reflected in the temperature distribution. Thus, after the diagnostic equipment determines the current server fan to be of the first fault type, it further combines the thermal characteristics to confirm that the current server fan fault is a blade imbalance fault.

[0066] S206. When it is determined that the server fan is in a state where the core temperature difference index is greater than or equal to the first preset threshold and the circumferential uniformity index is less than the second preset threshold, the diagnostic device diagnoses the current fault of the server fan as a bearing wear fault.

[0067] When the core temperature difference index is greater than or equal to the first preset threshold and the circumferential uniformity index is less than the second preset threshold, it indicates that a significant concentrated overheating phenomenon has occurred in the core area of ​​the fan. This is a typical thermal characteristic of bearing wear, because worn bearings generate abnormally high temperatures due to increased friction, leading to a significant increase in the core area temperature. Motor faults will exhibit harmonics related to the power supply frequency or rotational speed on the power spectral density graph. The main characteristic of bearing wear faults on the power spectral density graph is the appearance of specific fault characteristic frequencies and their harmonics related to the bearing geometry and rotational speed. When the diagnostic equipment determines that the current server fan is the first type of fault, it has ruled out the possibility of motor faults by analyzing the power spectral density graph and the current signal monitored by the BMC.

[0068] The above embodiments solve the problem Figure 1 The server fan fault diagnosis method based on power spectral density maps shown herein faces challenges in practical applications. Different types of server fan faults may exhibit similar power density curve characteristics in the low-frequency component region of the power spectral density map, making it difficult to accurately distinguish the fault type. However... Figure 1 The server fan fault diagnosis method based on power spectral density maps shown herein may exhibit similar power density curve characteristics in the low-frequency component region of the power spectral density map in practical applications, leading to difficulties in accurately identifying primary and secondary faults. Furthermore... Figure 1The server fan fault diagnosis method based on power spectral density maps shown above faces the problem that the acoustic features of early faults are too weak to be identified when detecting early faults.

[0069] Please refer to the following: Figure 3 This is another flowchart illustrating a method for diagnosing server fan failure in an embodiment of this application. S301, the diagnostic equipment divides the infrared thermal image of the server fan into a core area and a peripheral area.

[0070] S302, the diagnostic equipment extracts the core temperature difference index of the core area and the circumferential uniformity index of the surrounding area.

[0071] Step S301 is similar to step S201, and step S302 is similar to step S202, so they will not be described again here.

[0072] S303. The diagnostic device determines whether the current power spectral density map corresponding to the sound signal generated by the server fan rotation indicates a fault in the low-frequency component region.

[0073] The step S303, which involves the diagnostic device determining whether the current power spectral density map corresponding to the sound signal generated by the server fan indicates a fault in the low-frequency component region, has already been described in step S203 and will not be repeated here.

[0074] Specifically, after determining in step S303 that the current power spectral density map corresponding to the sound signal generated by the server fan indicates a fault in the low-frequency component region, step S307 is executed; after determining in step S303 that the current power spectral density map corresponding to the sound signal generated by the server fan does not indicate a fault in the low-frequency component region, step S304 is executed.

[0075] S304. If the power spectral density graph of the server fan indicates normal operation, the diagnostic device determines whether the core temperature difference index and circumferential uniformity index are abnormal.

[0076] Anomalies are defined as follows: the core temperature difference index is lower than the first preset threshold in multiple consecutive frames and the difference between the core temperature difference index and the first preset threshold gradually increases to 0; or the circumferential uniformity index is lower than the second preset threshold in multiple consecutive frames and the difference between the core temperature difference index and the second preset threshold gradually increases to 0.

[0077] Specifically, the first preset threshold is a judgment threshold set for the core temperature difference index, used to distinguish whether the core area temperature is significantly abnormal. The second preset threshold is a judgment threshold set for the circumferential uniformity index, used to distinguish whether the temperature distribution in the surrounding area is significantly uneven. Therefore, if the core temperature difference index of the server fan is lower than the first preset threshold for multiple consecutive frames and the difference between the core and the first preset threshold gradually increases to 0, or if the circumferential uniformity index is lower than the second preset threshold for multiple consecutive frames and the difference between the core and the second preset threshold gradually increases to 0, it indicates that the server fan is in the early stage of failure development when it is not performing high-power tasks.

[0078] After the diagnostic device determines in step S304 that the core temperature difference index and the circumferential uniformity index are abnormal, step S305 is executed. After the diagnostic device determines in step S304 that the core temperature difference index and the circumferential uniformity index are not abnormal, step S306 is executed.

[0079] S305. The diagnostic equipment diagnoses the current server fan failure as an early-stage failure.

[0080] Specifically, when the diagnostic equipment determines that the core temperature difference index is normal and the circumferential uniformity index is abnormal, since the circumferential uniformity index is used to quantify whether the temperature distribution in the surrounding area is uniform in the circumferential direction, when the blades are unbalanced, damaged, or have foreign objects attached, it will cause uneven airflow and vibration during the rotation of the blades, which will cause irregular fluctuations in the temperature distribution in the blade area in the circumferential direction. The higher the circumferential uniformity index, the more uneven the temperature distribution. Therefore, the diagnostic equipment diagnoses the current server fan's fault as an early blade imbalance fault based on the abnormal circumferential uniformity index of the current server fan.

[0081] The core temperature difference index (CTDI) is used to quantify the intensity of hot spots in the core area, reflecting whether there is abnormal heating in the fan bearings or motor. This index is calculated by the difference between the highest temperature point within the core area and the average temperature at the edge of the core area. A higher CTDI usually indicates overheating of core components due to bearing wear or motor failure. Therefore, based on the abnormal CTDI of the current server fan, the diagnostic equipment diagnoses the current server fan's fault as an early bearing wear failure.

[0082] In this step, the diagnostic equipment enables an initial differentiation of early fault types, allowing maintenance personnel to take intervention measures before the fault worsens.

[0083] S306. The diagnostic equipment indicates that the current server fan is operating normally.

[0084] After steps S303 and S304 determine that the acoustic and temperature characteristics of the current server fan are in normal condition, the diagnostic equipment diagnoses that the current server fan is operating normally.

[0085] S307. Based on the power density curve characteristics of the low-frequency component region in the current power spectral density map corresponding to the sound signal generated by the rotation of the server fan, the diagnostic device determines that the server fan is a first type of fault.

[0086] Step S307 is similar to step S203, and will not be described again here.

[0087] S308. When it is determined that the server fan is in a state where the core temperature difference index is greater than or equal to the first preset threshold and the circumferential uniformity index is greater than or equal to the second preset threshold, the diagnostic device diagnoses the current server fan fault as a composite fault.

[0088] Since the core temperature difference index is used to quantify the intensity of hot spots in the core area, and the circumferential uniformity index is used to quantify whether the temperature distribution in the surrounding area is uniform in the circumferential direction, if the diagnostic device determines that the server fan is in a situation where the core temperature difference index is greater than or equal to the first preset threshold and the circumferential uniformity index is greater than or equal to the second preset threshold, the diagnostic device diagnoses that there is a fault in both the core area and the surrounding area of ​​the current server fan, that is, the server fan has a combined fault.

[0089] S309. The diagnostic device determines whether the server fan is in a situation where the absolute value of the difference between the core temperature difference index and the first preset threshold is greater than the absolute value of the difference between the circumferential uniformity index and the second preset threshold.

[0090] After the diagnostic device determines in step S309 that the absolute value of the difference between the core temperature difference index and the first preset threshold is greater than the absolute value of the difference between the circumferential uniformity index and the second preset threshold, step S310 is executed. After the diagnostic device determines in step S309 that the server fan is not in a situation where the absolute value of the difference between the core temperature difference index and the first preset threshold is greater than the absolute value of the difference between the circumferential uniformity index and the second preset threshold, step S311 is executed.

[0091] S310. The diagnostic equipment diagnoses the primary fault of the server fan as bearing wear and the secondary fault as blade imbalance.

[0092] Since the diagnostic equipment determined that the absolute value of the difference between the core temperature difference index and the first preset threshold of the server fan was greater than the absolute value of the difference between the circumferential uniformity index and its threshold, it indicated that the anomaly in the core area was more severe. Therefore, the diagnostic equipment determined that the primary fault was bearing wear (corresponding to the core area anomaly), and the secondary fault was blade imbalance (corresponding to the peripheral area anomaly). In this case, the bearing wear was more severe than the blade imbalance, and the bearing problem should be addressed first during maintenance.

[0093] S311. The diagnostic device determines whether the server fan is in a situation where the absolute value of the difference between the core temperature difference index and the first preset threshold is less than the absolute value of the difference between the circumferential uniformity index and the second preset threshold.

[0094] After the diagnostic device determines in step S311 that the absolute value of the difference between the core temperature difference index and the first preset threshold is less than the absolute value of the difference between the circumferential uniformity index and the second preset threshold, step S312 is executed. After the diagnostic device determines in step S311 that the server fan is not in the situation where the absolute value of the difference between the core temperature difference index and the first preset threshold is less than the absolute value of the difference between the circumferential uniformity index and the second preset threshold, step S313 is executed.

[0095] S312. The diagnostic equipment diagnoses the primary fault of the server fan as blade imbalance and the secondary fault as bearing wear.

[0096] Similar to the diagnostic logic in step S310, when the circumferential uniformity index deviates more significantly from the second preset threshold, it indicates a more severe problem with the blade or surrounding area. Therefore, the diagnostic equipment determines the primary fault to be blade imbalance and the secondary fault to be bearing wear. In this case, the blade problem should be addressed first during maintenance.

[0097] S313. When the diagnostic device determines that the absolute value of the difference between the core temperature difference index and the first preset threshold is equal to the absolute value of the difference between the circumferential uniformity index and the second preset threshold, the diagnostic device diagnoses and determines that the current server fan has both blade imbalance fault and bearing wear fault, and the degree of fault is the same.

[0098] Similar to the diagnostic logic in steps S310 and S312, since the core temperature difference index and circumferential uniformity index deviate from the preset thresholds to the same extent, this indicates that the severity of the two faults, bearing wear and blade imbalance, is comparable. The diagnostic equipment determines that the server fan simultaneously suffers from blade imbalance and bearing wear faults, and that the two faults are of equal severity. In this case, maintenance requires attention to both bearing and blade issues.

[0099] This application addresses the problem that, in certain situations, such as the early stages of a fault, the power spectral density map corresponding to the sound signal generated by the fan rotation may still show a normal operating state, making it difficult for existing technologies to detect. It enables preliminary differentiation of early fault types, allowing maintenance personnel to take intervention measures before the fault worsens. Furthermore, by comparing the absolute values ​​of the differences between the core temperature difference index and the first preset threshold, and the absolute values ​​of the differences between the circumferential uniformity index and the second preset threshold, when the diagnostic equipment determines whether the server fan is at a state where the core temperature difference index is greater than or equal to a first preset threshold and the circumferential uniformity index is greater than or equal to a second preset threshold, the relative severity of each component fault in a complex fault is accurately distinguished, and the primary and secondary relationships of each fault are determined. This provides maintenance personnel with more refined and guiding diagnostic conclusions, enabling the formulation of better maintenance priorities and resource allocation strategies, significantly improving the efficiency and effectiveness of fault handling.

[0100] The above describes a method for diagnosing server fan failure in an embodiment of this application. The following describes an exemplary diagnostic device 100 provided in an embodiment of this application.

[0101] Figure 4 This is an exemplary hardware structure diagram of the diagnostic device 100 provided in an embodiment of this application. In some embodiments, the diagnostic device 100 is a computer device. The computer device includes a processor, a memory, and a network interface connected via a system bus. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with other external terminals or servers via a network connection. In some embodiments, the network interface can be a wired network interface; in some embodiments, the network interface can also be a wireless network interface. When the computer program is executed by the processor, it implements a method for diagnosing server fan failure according to an embodiment of this application.

[0102] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0103] In some embodiments of this application, a computer-readable storage medium is also provided, including instructions that, when executed on the diagnostic device 100, cause the diagnostic device 100 to perform a server fan failure diagnostic method according to an embodiment of this application.

[0104] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0105] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as meaning "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as meaning "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0106] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0107] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for diagnosing server fan failure, applied to diagnostic equipment, characterized in that, The method includes: The diagnostic equipment divides the infrared thermal image of the server fan into a core area and a peripheral area. The core area is where the motor and bearing are located, and the peripheral area is where the blades and air duct are located. The diagnostic device extracts the core temperature difference index of the core area and the circumferential uniformity index of the surrounding area. The core temperature difference index is the difference between the highest temperature point in the core area and the average temperature at the edge of the core area, and the circumferential uniformity index is the degree of temperature non-uniformity in the circumferential direction of the blade area. Based on the power density curve characteristics of the low-frequency component region in the current power spectral density map corresponding to the sound signal generated by the server fan rotation, the diagnostic device determines that the server fan is a first type of fault, which includes at least blade imbalance fault and bearing wear fault. If the core temperature difference index is less than a first preset threshold and the circumferential uniformity index is greater than or equal to a second preset threshold, the diagnostic device diagnoses the current server fan's fault as a blade imbalance fault. If the core temperature difference index is greater than or equal to the first preset threshold and the circumferential uniformity index is less than the second preset threshold, the diagnostic device diagnoses the current server fan fault as a bearing wear fault.

2. The method according to claim 1, characterized in that, The diagnostic device divides the infrared thermal image of the server fan into a core area and a peripheral area, specifically including: The diagnostic equipment divides the infrared thermal image of the server fan into a core area and a peripheral area based on the fan information built into the server. When a hotspot is detected in an area within the core region that has no obvious hotspots but contains an area of ​​15% of the server fan radius, the diagnostic device expands the core region boundary to include the hotspot. When the temperature gradient detected at the inner and outer diameter boundaries of the peripheral area is greater than or equal to a preset threshold, the diagnostic device adjusts the inner and outer diameter boundaries of the peripheral area accordingly.

3. The method according to claim 1, characterized in that, The diagnostic equipment extracts the core temperature difference index of the core area and the circumferential uniformity index of the surrounding area, specifically including: In the infrared thermal image of the core area, the diagnostic device extracts the highest temperature value among all pixels and the average temperature value among the edge pixels. The diagnostic device subtracts the average temperature value from the highest temperature value to obtain the core temperature difference index. The diagnostic device draws multiple rays from the center outwards at equal angular intervals in the peripheral area; The diagnostic device extracts a set of sample points representing the temperature at different locations around the circumference from the ray or near the ray. The diagnostic device calculates the standard deviation of the sample points to obtain the circumferential uniformity index.

4. The method according to claim 3, characterized in that, After the diagnostic device calculates the standard deviation of the sample points to obtain the circumferential uniformity index, the method further includes: The diagnostic device uses a low-pass filter to filter out high-frequency fluctuations in the core temperature difference index and the circumferential uniformity index acquired in multiple consecutive frames.

5. The method according to claim 4, characterized in that, After the diagnostic device uses a low-pass filter to filter out high-frequency fluctuations in the core temperature difference index and the circumferential uniformity index acquired in multiple consecutive frames, the method further includes: Based on the core temperature difference index and the circumferential uniformity index indicating that the server fan is operating without faults, the diagnostic device establishes a dynamic health status database. The diagnostic device determines a first preset threshold and a second preset threshold based on the dynamic health status database. The first preset threshold is the arithmetic mean of the core temperature difference index plus three times the standard deviation, and the second preset threshold is the arithmetic mean of the circumferential uniformity index plus three times the standard deviation.

6. The method according to claim 5, characterized in that, After the diagnostic device determines the first preset threshold and the second preset threshold based on the dynamic health status database, the method further includes: If the current power spectral density map corresponding to the sound signal generated by the server fan indicates normal operation and the core temperature difference index and the circumferential uniformity index are abnormal, the diagnostic device determines that the server fan is a second type of fault. The abnormality refers to the core temperature difference index being lower than the first preset threshold in multiple consecutive frames and the difference between the core temperature difference index and the first preset threshold gradually increasing to 0, or the circumferential uniformity index being lower than the second preset threshold in multiple consecutive frames and the difference between the core temperature difference index and the second preset threshold gradually increasing to 0. The second type of fault includes at least early blade imbalance fault and early bearing wear fault. When the core temperature difference index is determined to be normal and the circumferential uniformity index is abnormal, the diagnostic device diagnoses the current server fan fault as an early blade imbalance fault. When the core temperature difference index is determined to be abnormal and the circumferential uniformity index is normal, the diagnostic equipment diagnoses the current server fan failure as an early bearing wear failure.

7. The method according to any one of claims 1-6, characterized in that, The method further includes: When the core temperature difference index is greater than or equal to a first preset threshold and the circumferential uniformity index is greater than or equal to a second preset threshold, the diagnostic device diagnoses the current server fan's fault as a composite fault, which is the simultaneous presence of bearing wear fault and blade imbalance fault. When the absolute value of the difference between the core temperature difference index and the first preset threshold is greater than the absolute value of the difference between the circumferential uniformity index and the second preset threshold, the diagnostic device diagnoses the primary fault of the current server fan as bearing wear fault and the secondary fault as blade imbalance fault. When the absolute value of the difference between the core temperature difference index and the first preset threshold is less than the absolute value of the difference between the circumferential uniformity index and the second preset threshold, the diagnostic device diagnoses the primary fault of the current server fan as blade imbalance fault and the secondary fault as bearing wear fault. When the absolute value of the difference between the core temperature difference index and the first preset threshold is equal to the absolute value of the difference between the circumferential uniformity index and the second preset threshold, the diagnostic device diagnoses that the current server fan has both blade imbalance fault and bearing wear fault, and the faults are of the same degree.

8. A diagnostic device, characterized in that, The diagnostic device includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code including computer instructions, and the one or more processors call the computer instructions to cause the diagnostic device to perform the method as described in any one of claims 1-7.

9. A computer program product containing instructions, characterized in that, When the computer program product is run on a diagnostic device, the diagnostic device performs the method as described in any one of claims 1-7.

10. A computer-readable storage medium comprising instructions, characterized in that, When the instructions are executed on a diagnostic device, the diagnostic device performs the method as described in any one of claims 1-7.

Images