Fault detection method, system and device for whole server cabinet and electronic equipment

Abstract

The invention provides a fault detection method, system and device for a whole server cabinet and electronic equipment. The method comprises the following steps: acquiring a noise signal acquired by pickup equipment of a whole server cabinet; performing frequency division processing on the noise signal to obtain high-frequency information and low-frequency information; extracting high-frequency features of the high-frequency information and low-frequency features of the low-frequency information; and fault detection is carried out based on the high-frequency characteristics and the low-frequency characteristics to obtain a fault detection result of the whole server cabinet, and fault components in the whole server cabinet can be further determined. The method is used for achieving the effect of improving the timeliness of fault detection.

Description

Technical Field

[0001] This application relates to the field of fault detection and the field of artificial intelligence, and particularly relates to a method, system, device and electronic device for fault detection of a server rack. Background Art

[0002] Servers are important devices in data centers and network environments. For ensuring the stability of data centers and network environments, the normal operation of components in a server rack is crucial.

[0003] In the related art, the method of manually inspecting regularly is used to determine whether components in a server rack are abnormal. Such a method is difficult to detect faults in a server rack in a timely manner and lacks timeliness. Summary of the Invention

[0004] Embodiments of this application provide a method, system, device and electronic device for fault detection of a server rack, so as to achieve the effect of improving the timeliness of fault detection.

[0005] In a first aspect, an embodiment of this application provides a method for fault detection of a server rack, including:

[0006] Obtaining a noise signal collected by a sound pickup device of the server rack; performing frequency division processing on the noise signal to obtain high-frequency information and low-frequency information; extracting high-frequency features of the high-frequency information and low-frequency features of the low-frequency information; performing fault detection based on the high-frequency features and low-frequency features to obtain a fault detection result of the server rack.

[0007] In a second aspect, an embodiment of this application provides a server fault detection system, including: a server rack and a fault detection device; the server rack includes a sound pickup device for collecting a noise signal; the fault detection device is configured to receive the noise signal and perform fault detection based on the noise signal according to the method in the first aspect.

[0008] In a third aspect, an embodiment of this application provides a device for fault detection of a server rack, including:

[0009] A noise signal acquisition module for obtaining a noise signal collected by a sound pickup device of the server rack;

[0010] An audio separation module for performing frequency division processing on the noise signal to obtain high-frequency information and low-frequency information;

[0011] A feature extraction module for extracting high-frequency features of the high-frequency information and low-frequency features of the low-frequency information;

[0012] A detection module for performing fault detection based on the high-frequency features and low-frequency features to obtain a fault detection result of the server rack.

[0013] Fourthly, an embodiment of the present application provides an electronic device, including: a memory and a processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementation manners of the first aspect.

[0014] Fifthly, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, they are used to implement the above first aspect and / or various possible implementation manners of the first aspect.

[0015] Sixthly, an embodiment of the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the above first aspect and / or various possible implementation manners of the first aspect.

[0016] For the fault detection method, system, device and electronic device of the server rack provided by the embodiments of the present application, a noise signal collected by a sound pickup device of the server rack is obtained, the noise signal is frequency-divided to obtain high-frequency information and low-frequency information; high-frequency features of the high-frequency information and low-frequency features of the low-frequency information are extracted; fault detection is performed based on the high-frequency features and low-frequency features to obtain a fault detection result of the server rack; since the noise signal generated when a component in the server rack fails is different from the noise information generated during normal operation, fault detection based on the noise signal can detect abnormal situations in the first time when a fault occurs, improving the timeliness of fault detection; and since there may be some components in the server rack, the noise signal generated when a fault occurs has a prominent high-frequency part, and the noise signal generated when some other components fail has a prominent low-frequency part, the high-frequency information and low-frequency information extracted from the noise signal, and further the high-frequency features and low-frequency features are extracted. When the server rack has a fault, the high-frequency features and low-frequency features include unique fault features. Fault detection based on the high-frequency features and low-frequency features can obtain an accurate fault detection result, improving the comprehensiveness and accuracy of fault detection for the server rack. Description of the Drawings

[0017] The drawings here are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0018] Figure 1 It is a schematic diagram of the application scenario of the fault detection method for the server rack provided by the present application;

[0019] Figure 2 It is a flowchart of the fault detection method for the server rack provided by the present application Figure 1 ;

[0020] Figure 3 Interaction schematic diagram between the wide-spectrum microphone provided by this application and the electronic device;

[0021] Figure 4 Flow schematic diagram for extracting high-frequency features and low-frequency features provided by this application;

[0022] Figure 5 Flow schematic diagram for giving an early warning when the fault detection result provided by this application is abnormal;

[0023] Figure 6 Flow schematic of the fault detection method for the entire server cabinet provided by this application Figure 2 ;

[0024] Figure 7 Flow schematic of the fault detection method for the entire server cabinet provided by this application Figure 3 ;

[0025] Figure 8 Structural schematic diagram of the fault detection device for the entire server cabinet provided by this application;

[0026] Figure 9 Structural schematic diagram of the electronic device provided by this application.

[0027] Through the above-mentioned drawings, specific embodiments of this application have been shown, and there will be more detailed descriptions hereinafter. These drawings and text descriptions are not intended to limit the scope of the concept of this application in any way, but to illustrate the concept of this application to those skilled in the art by referring to specific embodiments. Specific Embodiments

[0028] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. On the contrary, they are merely examples of devices and methods consistent with some aspects of this application as detailed in the appended claims.

[0029] The fault detection method for the entire server cabinet provided by the embodiments of this application can be applied to such as Figure 1In the application environment shown. Among them, the fault detection system of the server rack includes a server rack 101 and a fault detection device 102. The server rack 101 communicates with the fault detection device 102 through a network. Among them, the fault detection device 102 can be integrated in the server rack, or the fault detection device 102 can be a separately implemented device. For example, it can be implemented by an independent server or a server cluster composed of multiple servers; the separately implemented fault detection device 102 can be deployed inside the server rack 101 or outside the server rack 101.

[0030] The following uses specific embodiments to describe in detail the technical solution of the present application and how the technical solution of the present application solves the above technical problems. The following several specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0031] Figure 2 Schematic flow of the fault detection method for the server rack provided by the present application Figure 1 , the fault detection method of the server rack can be applied to an electronic device. Taking the fault detection device in the electronic device as an example for description; as Figure 1 shown, the fault detection method of the server rack includes: Figure 2

[0032] S201. Obtain the noise signal collected by the sound pickup device of the server rack.

[0033] Among them, the sound pickup device can be a wide-spectrum microphone. The wide-spectrum microphone has an ultra-wide frequency response range. Through the wide-spectrum microphone, sound signals from the low-frequency band to the high-frequency band can be collected. Exemplarily, the frequency response range of the wide-spectrum microphone can be 0.1 Hz to 140 kHz.

[0034] Optionally, the sound pickup device is set inside the server rack to collect the noise signal; the collected noise signal includes the noise signal generated by the operation of the server rack and the noise signal of the external environment.

[0035] In practical applications, the sound pickup device is set inside the server rack, so that the sound pickup device is in the sound field generated by the operation of the components inside the server rack, and the cabinet of the server rack has a certain sound insulation effect on the internal environment and the external environment inside the server rack. Therefore, the collected noise signal mainly includes the noise information generated by the operation of the server rack, and the noise information generated by the external environment is less.

[0036] Optionally, the sound pickup device is set outside the server rack, close to the server rack.

[0037] The server rack integrates servers and related components, and is a highly integrated, easy-to-deploy and manage whole. When the server rack is running, the internal servers and multiple components will generate noise signals. There are differences in the noise signals generated when the server rack is running normally and abnormally. Therefore, the noise signals can be used as a basis for detecting whether the server rack is abnormal.

[0038] There are differences in the noise signals generated when the server rack is running normally and abnormally, including but not limited to the following situations:

[0039] 1) Fan. When the server rack is running abnormally (such as overheating), the fan will increase its speed, resulting in different noise signals generated by the fan compared to when it is running normally. When there is an abnormality in the fan itself, the generated noise signals are also different from those generated during normal operation;

[0040] 2) Liquid cooling components. The liquid flow and pipeline friction will generate noise signals. In case of abnormal situations such as liquid leakage, the change in the flow rate of the liquid will cause different noise signals to be generated;

[0041] 3) Capacitor. When the capacitor operates in an abnormal frequency band, it will generate a slight whistling sound or buzzing sound, thus resulting in different noise signals generated by it;

[0042] 4) Coils and inductors. When high-frequency current passes through coils and inductors, it will cause abnormal vibrations, thus resulting in different noise signals generated by them;

[0043] 5) Power supply. When the transformer or inductor of the power supply operates abnormally, the buzzing sound generated by it is different from that during normal operation, resulting in different noise signals;

[0044] 6) Server rack cabinet. When the server rack operates abnormally, it may cause the server rack cabinet to vibrate, resulting in different noise signals.

[0045] When different components in the above-mentioned server rack are running, some may generate high-frequency noise signals while others may generate low-frequency noise signals. The noise signals in each frequency band can be collected through a wide-spectrum microphone, achieving a comprehensive collection of the noise signals.

[0046] Specifically, such as Figure 3As shown in the figure, a sound pickup device (such as a wide-spectrum microphone) is set inside the server cabinet. The sound pickup device collects noise signals in real time. The collected noise signals may include the noise signals generated when the server cabinet operates normally, or may include the noise signals generated when the server cabinet operates abnormally. The sound pickup device sends the noise signals to the data forwarding module, and the data forwarding module sends the noise signals to the fault detection device. Then, the fault detection device obtains the noise signals collected by the sound pickup device.

[0047] Among them, the data forwarding module, as the data transmission bridge between the sound pickup device and the fault detection device, can convert the noise signals into data packets suitable for network transmission, and then send the data packets to the fault detection device. The fault detection device restores the obtained data packets to obtain the noise signals. The data forwarding module can be implemented by a microcontroller.

[0048] S202. Perform frequency division processing on the noise signals to obtain high-frequency information and low-frequency information.

[0049] Among them, the high-frequency information can represent the high-frequency part in the noise signals, and the low-frequency information can represent the low-frequency part in the noise signals. The high-frequency information and the low-frequency information respectively correspond to different frequency ranges. The frequency range corresponding to the high-frequency information is greater than the frequency range corresponding to the low-frequency information. The frequency ranges respectively corresponding to the high-frequency information and the low-frequency information can be set according to actual needs, and the embodiments of the present application do not limit this.

[0050] It should be noted that when some components in the server cabinet fail, the generated noise signals may have prominent high-frequency parts, and when some other components fail, the generated noise signals may have prominent low-frequency parts. Therefore, the high-frequency information and the low-frequency information determined according to the noise signals may include the fault sound characteristics unique to different components.

[0051] Exemplarily, the high-frequency parts of the noise signals generated by the following components are prominent:

[0052] 1) Fan. When its blades rotate, it will generate relatively high-frequency noise, especially when rotating at high speed, the high-frequency noise is obvious.

[0053] 2) Capacitor. It will generate a whistling noise when working at high frequencies.

[0054] 3) Coil and capacitor. They will vibrate when high-frequency current passes through, so the high-frequency part is prominent.

[0055] The low-frequency parts of the noise signals generated by the following components are prominent:

[0056] 1) Power supply. The power supply transformer and inductor may emit low-frequency humming sounds when working.

[0057] 2) Fan. The overall noise of the fan may contain high-frequency components, but there are also some low-frequency components;

[0058] 3) Chassis resonance. When the chassis is vibrated, it may generate low-frequency resonance sounds.

[0059] Optionally, frequency division can be implemented in software; the fault detection device includes a digital audio processing unit, which can be used to process frequency division; the digital audio processing unit performs frequency division processing on the noise signal according to the preset frequency division parameters to obtain high-frequency information and low-frequency information; performing frequency division processing in software can improve the efficiency of determining high-frequency information and low-frequency information.

[0060] Optionally, frequency division can be implemented in hardware; the fault detection device includes a frequency divider. The fault detection device inputs the noise signal into the frequency divider, and the frequency divider can use components such as coils and capacitors to divide the frequency to obtain high-frequency information and low-frequency information; performing frequency division processing in hardware requires less computing resources, reducing the computing resource pressure on the fault detection device.

[0061] S203. Extract the high-frequency features of the high-frequency information and the low-frequency features of the low-frequency information.

[0062] Among them, the high-frequency features can reflect the sharpness, clarity, harmonic richness, and transient characteristics of the high-frequency information; the low-frequency features can reflect the depth, duration, amplitude change, and resonance characteristics of the low-frequency information.

[0063] Both the high-frequency features and the low-frequency features can be represented by feature vectors.

[0064] Specifically, the fault detection device can extract the high-frequency features of the high-frequency information through a high-frequency information processing module (HFM, High Frequency Module). For example, the high-frequency information is input into the HFM module, and the high-frequency features are obtained through the HFM module. In practical applications, the HFM module can be implemented by a convolutional neural network.

[0065] The fault detection device can extract the low-frequency features of the low-frequency information through a low-frequency information processing module (LFM, Low Frequency Module). For example, the low-frequency information is input into the LFM module, and the low-frequency features are obtained through the LFM module. In practical applications, the LFM can be implemented by the encoder of the Transformer, where the Transformer consists of an encoder and a decoder, and both the encoder and the decoder are connected through a number of self-attention layers.

[0066] S204. Perform fault detection based on the high-frequency features and the low-frequency features to obtain the fault detection result of the entire server cabinet.

[0067] Among them, the fault detection result is normal or abnormal. If the fault detection result is normal, it means that no component in the server's entire cabinet has failed. If the fault detection result is abnormal, it means that there is a component in the server's entire cabinet that has failed.

[0068] Optionally, the fault detection device fuses the high-frequency feature and the low-frequency feature to obtain a comprehensive audio feature, and performs fault detection based on the comprehensive audio feature to obtain the abnormal detection result of the server's entire cabinet.

[0069] Specifically, a detection module is used to perform fault detection on the comprehensive audio feature to obtain an abnormal probability. When the abnormal probability is greater than a preset probability threshold, it is determined that the fault detection result is abnormal. When the abnormal probability is not greater than the preset probability threshold, it is determined that the fault detection result is normal; among them, the detection module can be a classifier that performs a binary classification task.

[0070] Optionally, the fault detection device performs fault detection based on the high-frequency feature to obtain a high-frequency fault detection result, and performs fault detection based on the low-frequency feature to obtain a low-frequency fault detection result. When both the high-frequency fault detection result and the low-frequency fault detection result are normal, it is determined that the fault detection result of the server's entire cabinet is normal. When the high-frequency fault detection result and / or the low-frequency fault detection result is abnormal, it is determined that the fault detection result of the server's entire cabinet is abnormal.

[0071] Specifically, a detection module is used to perform fault detection on the high-frequency feature to obtain a high-frequency abnormal probability. When the high-frequency abnormal probability is greater than a preset probability threshold, it is determined that the high-frequency fault detection result is abnormal. When the high-frequency abnormal probability is not greater than the preset probability threshold, it is determined that the high-frequency fault detection result is normal; a detection module is used to perform fault detection on the low-frequency feature to obtain a low-frequency abnormal probability. When the low-frequency abnormal probability is greater than a preset probability threshold, it is determined that the low-frequency fault detection result is abnormal. When the low-frequency abnormal probability is not greater than the preset probability threshold, it is determined that the low-frequency fault detection result is normal.

[0072] The fault detection method for the server rack provided by the embodiment of the present application obtains the noise signal collected by the sound pickup device of the server rack, performs frequency division processing on the noise signal to obtain high-frequency information and low-frequency information; extracts the high-frequency characteristics of the high-frequency information and the low-frequency characteristics of the low-frequency information; performs fault detection based on the high-frequency characteristics and the low-frequency characteristics to obtain the fault detection result of the server rack; since the noise signal generated when the components in the server rack fail is different from the noise information generated during normal operation, fault detection based on the noise signal can detect abnormal situations in the first time when the fault occurs, improving the timeliness of fault detection; and because there may be some components in the server rack, the noise signal generated when they fail has a prominent high-frequency part, and the noise signal generated when some other components fail has a prominent low-frequency part. The high-frequency information and the low-frequency information extracted from the noise signal, and further the high-frequency characteristics and the low-frequency characteristics are extracted. When the server rack has a fault, the high-frequency characteristics and the low-frequency characteristics include unique fault characteristics. Fault detection based on the high-frequency characteristics and the low-frequency characteristics can obtain accurate fault detection results, improving the comprehensiveness and accuracy of fault detection for the server rack.

[0073] In a possible implementation manner, extracting the high-frequency characteristics of the high-frequency information and the low-frequency characteristics of the low-frequency information includes: performing shallow feature extraction on the high-frequency information and the low-frequency information to obtain shallow high-frequency characteristics and shallow low-frequency characteristics, and fusing the shallow high-frequency characteristics and the shallow low-frequency characteristics to obtain shallow fusion characteristics; performing deep feature extraction on the shallow fusion characteristics and the shallow low-frequency characteristics to obtain deep high-frequency characteristics and low-frequency characteristics, and fusing the deep high-frequency characteristics and the low-frequency characteristics to obtain high-frequency characteristics.

[0074] Optionally, the fault detection device inputs the high-frequency information into the shallow high-frequency processing module to obtain shallow high-frequency characteristics; inputs the low-frequency information into the shallow low-frequency processing module to obtain shallow low-frequency characteristics. The shallow high-frequency processing module can be implemented by a convolutional neural network; the shallow low-frequency processing module can be implemented by the encoder of a Transformer.

[0075] Fusing the shallow high-frequency characteristics and the shallow low-frequency characteristics to obtain shallow fusion characteristics includes: inputting the shallow low-frequency characteristics into the calibration module (Calibration Module, CM) to obtain the calibrated shallow low-frequency characteristics, and splicing the calibrated shallow low-frequency characteristics and the shallow high-frequency characteristics to obtain shallow fusion characteristics; wherein, the calibration module can be a neural network module with a self-attention mechanism.

[0076] Input the shallow low-frequency features into the deep low-frequency processing module to obtain low-frequency features. Input the low-frequency features into the correction module to obtain corrected low-frequency features. Fuse the deep high-frequency features and the corrected low-frequency features to obtain deep fusion features. Input the deep fusion features into the deep high-frequency processing module to obtain high-frequency features.

[0077] Specifically, as Figure 4 shown, the shallow high-frequency processing module includes: a first high-frequency processing unit, a second high-frequency processing unit, and a third high-frequency processing unit. The deep high-frequency processing module includes: a fourth high-frequency processing unit, a fifth high-frequency processing unit, and a sixth high-frequency processing unit; each high-frequency processing unit includes two cascaded convolutional layers; the shallow low-frequency processing unit includes: a first low-frequency processing unit, a second low-frequency processing unit, and a third low-frequency processing unit; the deep low-frequency processing module includes: a fourth low-frequency processing unit, a fifth low-frequency processing unit, and a sixth low-frequency processing unit; each low-frequency processing unit includes an encoder of a Transformer.

[0078] Input the high-frequency information into the first high-frequency processing unit to obtain the first intermediate high-frequency features; input the low-frequency information into the first low-frequency processing unit to obtain the first intermediate low-frequency features; correct the first intermediate low-frequency features through the correction module to obtain the first corrected low-frequency features, splice the first corrected low-frequency features and the first intermediate high-frequency features to obtain the first fusion features, input the first fusion features into the second high-frequency processing unit to obtain the second intermediate high-frequency features, and input the first intermediate low-frequency features into the second low-frequency processing unit to obtain the second intermediate low-frequency features.

[0079] Process the second intermediate low-frequency features and the second intermediate high-frequency features according to the above process of processing the first intermediate low-frequency features and the first intermediate high-frequency features to obtain the third intermediate high-frequency features and the third intermediate low-frequency features, and then continue to process the third intermediate low-frequency features and the third intermediate high-frequency features until the sixth intermediate high-frequency features and the sixth intermediate low-frequency features are obtained; among them, the third intermediate high-frequency features are shallow high-frequency features, the third intermediate low-frequency features are shallow low-frequency features, the sixth intermediate high-frequency features are high-frequency features, and the sixth intermediate low-frequency features are low-frequency features.

[0080] In the above embodiment, the shallow high-frequency features and the shallow low-frequency features are extracted, the shallow high-frequency features and the shallow low-frequency features are fused to obtain shallow fusion features, and then the low-frequency features and the high-frequency features are determined based on the shallow fusion features and the shallow low-frequency features. Through multi-level feature extraction and fusion processing, the potential features in the noise signal are fully excavated, and the quality of the low-frequency features and the high-frequency features is improved.

[0081] In some embodiments, fault detection is performed based on high-frequency features and low-frequency features to obtain the fault detection result of the entire server cabinet, including: performing feature fusion processing on the high-frequency features and low-frequency features to obtain comprehensive audio features; performing fault detection on the comprehensive audio features through a first classifier to obtain the fault detection result of the entire server cabinet; the fault detection result is abnormal or normal.

[0082] Specifically, the fault detection device can splice the high-frequency features and low-frequency features to obtain comprehensive audio features; or can correct the low-frequency features through a correction module, and splice the corrected low-frequency features and high-frequency features to obtain comprehensive audio features.

[0083] Input the comprehensive audio features into the first classifier, obtain the abnormal probability through the first classifier. In the case where the abnormal probability is greater than the preset probability threshold, determine that the fault detection result of the entire server cabinet is abnormal; in the case where the abnormal probability is not greater than the preset probability threshold, determine that the fault detection result of the entire server cabinet is normal. The preset probability threshold can be set according to actual needs, and the specific value of the preset probability threshold in the embodiments of the present application is not limited.

[0084] Among them, the first classifier can be implemented by a Multilayer Perceptron (MLP).

[0085] In the above embodiments, feature fusion processing is performed on the high-frequency features and low-frequency features, and then fault detection is performed on the comprehensive audio features through the first classifier, which improves the accuracy of fault detection.

[0086] In a possible implementation manner, fault detection is performed based on high-frequency features and low-frequency features to obtain the fault detection result of the entire server cabinet, including: performing fault detection on the high-frequency features through a second classifier to obtain a first detection result; performing fault detection on the low-frequency features through the second classifier to obtain a second detection result; determining the fault detection result of the entire server cabinet based on the first detection result and the second detection result.

[0087] Among them, the second classifier is used to perform a binary classification task for fault detection of high-frequency features, and the second classifier can be implemented by a multilayer perceptron; the third classifier is used to perform a binary classification task for fault detection of low-frequency features, and the third classifier can also be implemented by a multilayer perceptron.

[0088] The second classifier and the third classifier are trained through different training samples. For example, the second classifier is trained through high-frequency feature samples and corresponding fault labels, and the third classifier is trained through low-frequency feature samples and corresponding fault labels.

[0089] Specifically, the high-frequency features are input into the second classifier to obtain the high-frequency anomaly probability. When the high-frequency anomaly probability is greater than the first preset threshold, it is determined that the first detection result is abnormal; when the high-frequency anomaly probability is not greater than the first preset threshold, it is determined that the first detection result is normal. The low-frequency features are input into the third classifier to obtain the low-frequency anomaly probability. When the low-frequency anomaly probability is greater than the second preset threshold, it is determined that the second detection result is abnormal; when the low-frequency anomaly probability is not greater than the second preset threshold, it is determined that the second detection result is normal.

[0090] It should be noted that the first preset threshold and the second preset threshold can be set according to actual needs. The first preset threshold and the second preset threshold can be equal or unequal; the first preset threshold and the second preset threshold can also be equal to the preset probability threshold.

[0091] When both the first detection result and the second detection result are normal, it is determined that the fault detection result of the server whole cabinet is normal; when the first detection result and / or the second detection result is abnormal, it is determined that the fault detection result of the server whole cabinet is abnormal.

[0092] In the above embodiments, the fault detection is respectively performed according to the high-frequency features and the low-frequency features, which can improve the accuracy of the determined first detection result and the second detection result. Then, the fault detection result of the server whole cabinet is determined by comprehensively considering the first detection result and the second detection result, thereby improving the accuracy of the fault detection.

[0093] In a possible implementation manner, the number of the sound pickup devices is multiple, and the multiple sound pickup devices are respectively arranged in multiple detection areas inside the server whole cabinet. The fault detection method of the server whole cabinet further includes: when the fault detection result is abnormal, determining the target detection area to which the sound pickup device that collects the noise signal belongs among the multiple detection areas; based on the high-frequency features and the low-frequency features, performing fault detection on the components included in the target detection area to obtain the faulty components.

[0094] Among them, the multiple sound pickup devices are respectively arranged in multiple different detection areas, and each detection area inside the server whole cabinet includes at least one component.

[0095] Specifically, when the fault detection result obtained by performing fault detection on the collected noise information is abnormal, determining the sound pickup device used to collect the noise information, and obtaining the target detection area to which the sound pickup device belongs among the multiple detection areas.

[0096] When the target detection area contains at least two components, the fault detection device performs fusion processing on high-frequency features and low-frequency features to obtain comprehensive audio features, obtains the fourth classifier corresponding to the detection area, inputs the comprehensive audio features into the fourth classifier to obtain the fault probabilities of the components within the target detection area, and takes the component with the highest fault probability as the faulty component of the entire server cabinet.

[0097] Among them, the fourth classifier is used to perform a multi-classification task. The fourth classifier can be implemented by a multi-layer perceptron, and the fourth classifiers corresponding to different detection areas may be different; for example, if there are 3 components in the detection area where the sound pickup device belongs, the fourth classifier corresponding to this detection area can be used to detect the fault probabilities of the 3 components.

[0098] In a possible implementation manner, when the target detection area contains one component, the component contained in the target detection area is taken as the faulty component.

[0099] In the above embodiment, when the fault detection result of the entire server cabinet is abnormal, further fault detection is performed on the components within the target detection area, and the specific faulty component can be located, realizing fault detection at the component level and improving the fault detection effect.

[0100] In a possible implementation manner, when the fault detection result of the entire server cabinet is abnormal, warning is given through sound and light and / or sending a warning message to the terminal.

[0101] Among them, the sound and light warning includes playing a warning audio and lighting a warning lamp.

[0102] Specifically, when the fault detection result is abnormal, the fault can be prompted by the warning lamp and playing the warning audio. The light color of the warning lamp can be used to distinguish different fault detection results. The warning audio can be a specific prompt music or a voice indicating the existence of a fault.

[0103] The fault detection device can also report the warning information to the terminal in real time. The terminal can display the warning information in the corresponding application program. For example, the warning information is displayed through a small program; the user can send a warning cancellation instruction to the fault detection device through the application program. For example, when the warning information is displayed, the entire server cabinet is controlled to shut down through the displayed warning information to send a warning cancellation instruction to the fault detection device, or, when it is determined that the fault has been resolved, an operation for the resolved fault is performed in the application program to send a warning cancellation instruction to the fault detection device.

[0104] For example, Figure 5As shown, read the fault detection results of the entire server cabinet. When the fault detection result is abnormal, control the warning light to emit red light. When the fault detection result is normal, control the warning light to emit green light. The fault detection device can also send a warning instruction to the terminal, causing the terminal to vibrate or ring according to the warning instruction, and determine whether a warning cancellation instruction is received. If a warning cancellation instruction is received, end the current process. If not, continue to determine whether the preset duration has elapsed since the first warning instruction was sent. The preset duration can be set according to requirements. For example, the preset duration can be 3 minutes. If the preset duration has not elapsed, continue to send the warning instruction. If the preset duration has elapsed, a call can be made to the terminal, and a preset warning voice can be played after the call is connected. Determine whether a warning cancellation instruction is received. If the call is not connected, determine whether a warning cancellation instruction is received. If a warning cancellation instruction is received, end the current process.

[0105] In a possible implementation, when the fault detection result of the entire server cabinet is normal, end the current detection process. When the fault detection result of the entire server cabinet is abnormal, record relevant information, such as obtaining and saving the noise signal of this fault detection, so as to optimize the shallow high-frequency processing module, deep high-frequency processing module, shallow low-frequency processing module, deep low-frequency processing module, first classifier, second classifier, third classifier, and fourth classifier through the noise signal later.

[0106] In a possible implementation, after each fault detection is completed, the used noise signal can be converted into audio data in a specific format, and the noise signal and audio data are uploaded to the cloud platform and stored in a specified location on the cloud platform, so as to facilitate subsequent review of the noise signal and audio data, or use the noise signal and audio data as samples for deep learning.

[0107] In a possible implementation, after each fault detection is completed, log information can be generated according to the process of this fault detection, and the log information is saved to the log folder for subsequent analysis and processing.

[0108] In a possible implementation, after each fault detection is completed, the noise signal, fault detection result, and faulty component used in this fault detection can be written into a file package, and the file package is uploaded to the specified location corresponding to the fault detection result to achieve the upload of classification results for subsequent processing.

[0109] Exemplarily, Figure 6 is the flow diagram of the fault detection method for the entire server cabinet provided by this application Figure 2 as Figure 6As shown in the figure, the fault detection method for the entire server cabinet includes: collecting the noise signal through the sound pickup device set in the entire server cabinet, uploading the noise signal to the fault detection device, and the fault detection device performs frequency division processing on the noise signal to obtain high-frequency information and low-frequency information; using the high-frequency feature extraction module (HFM) to extract features from the high-frequency information to obtain high-frequency features, and using the low-frequency feature extraction module (LFM) to extract features from the low-frequency information to obtain low-frequency features; fusing the low-frequency features and high-frequency features to obtain comprehensive audio features, and processing the comprehensive audio features through the first classifier to obtain the fault detection result. If the fault detection result is abnormal, a warning prompt is given and the next fault detection is performed. If the fault detection result is normal, the next fault detection is performed. In addition, the noise signal is subjected to audio conversion to obtain audio data in a specific format, and the noise signal, audio data, and fault detection result are written into the log, and the log is added to the log folder. The data in the log folder can be used for subsequent training of the high-frequency feature extraction module and the low-frequency feature extraction module.

[0110] Figure 7 is the process schematic of the fault detection method for the entire server cabinet provided by this application Figure 3 , as Figure 7 shown, this method includes:

[0111] S701. Obtain the noise signal collected by the sound pickup device of the entire server cabinet;

[0112] S702. Perform frequency division processing on the noise signal to obtain high-frequency information and low-frequency information;

[0113] S703. Perform shallow feature extraction on the high-frequency information and low-frequency information to obtain shallow high-frequency features and shallow low-frequency features, and fuse the shallow high-frequency features and shallow low-frequency features to obtain shallow fusion features; perform deep feature extraction on the shallow fusion features and shallow low-frequency features to obtain deep high-frequency features and low-frequency features, and fuse the deep high-frequency features and low-frequency features to obtain high-frequency features;

[0114] S704A. Perform feature fusion processing on the high-frequency features and low-frequency features to obtain comprehensive audio features; perform fault detection on the comprehensive audio features through the first classifier to obtain the fault detection result of the entire server cabinet; the fault detection result is abnormal or normal;

[0115] S704B. Perform fault detection on the high-frequency features through the second classifier to obtain the first detection result; perform fault detection on the low-frequency features through the third classifier to obtain the second detection result; based on the first detection result and the second detection result, determine the fault detection result of the entire server cabinet;

[0116] S705. The number of sound pickup devices is multiple, and the multiple sound pickup devices are respectively arranged in multiple detection areas within the server cabinet. In the case where the fault detection result is abnormal, determine the target detection area to which the sound pickup device that collects the noise signal belongs among the multiple detection areas. Based on the high-frequency feature and the low-frequency feature, perform fault detection on the components included in the target detection area to obtain the faulty component.

[0117] S706. In the case where the fault detection result is abnormal, give an early warning through acoustic and optical prompts and / or send a warning message to the terminal.

[0118] The fault detection method for the server cabinet provided by the embodiment of the present application collects the noise signal through a sound pickup device (such as a wide-spectrum microphone), realizing the comprehensive collection of the noise signals generated when each component fails. Since multiple wide-spectrum microphones are deployed inside the server cabinet, in the case where it is recognized that there is a fault in the server cabinet, it is possible to determine which detection area's component has failed according to the detection areas where the multiple wide-spectrum microphones are located inside the server cabinet, and then further detect the faulty component, effectively reducing the noise interference between different machines and improving the accuracy and stability of fault location.

[0119] The embodiment of the present application also provides a full process from noise signal collection, frequency division, feature extraction to fault prediction and early warning, enabling the fault detection method for the server cabinet to be applied to a complex noise environment such as a server cabinet. Among them, a neural network is used to realize feature extraction and fault detection, which can process noise information in different frequency bands and improve the processing efficiency.

[0120] The embodiment of the present application adopts feature extraction and fusion, fully excavating the potential features in the noise signal, significantly improving the generalization ability of the model and the accuracy of fault detection. The noise signal and the fault detection result are used to establish a data set covering various faults and working scenarios, and the data set is used for targeted model training, so that the fault detection method for the server cabinet can be applied to the fault detection in more complex environments, improving the practicality of fault detection.

[0121] The various early warning prompt methods provided by the embodiment of the present application can ensure timely feedback of the fault situation, enabling the fault to be solved in time and reducing the risks brought by the fault.

[0122] It should be understood that although the steps in the flowcharts involved in the above embodiments are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0123] This application also provides a fault detection system for a server rack, including: a server rack and a fault detection device;

[0124] The server rack includes a sound pickup device for collecting noise signals;

[0125] The fault detection device is used to receive the noise signal, perform frequency division processing on the noise signal to obtain high-frequency information and low-frequency information; extract the high-frequency characteristics of the high-frequency information and the low-frequency characteristics of the low-frequency information; and perform fault detection based on the high-frequency characteristics and the low-frequency characteristics to obtain the fault detection result of the server rack.

[0126] Among them, the fault detection device can be integrated in the server rack, or it can be a separately implemented device. The separately implemented fault detection device can be deployed inside the server rack or outside the server rack.

[0127] The sound pickup device can be a wide-spectrum microphone. The sound pickup device can be set inside the server rack to collect noise signals; or it can be set outside the server rack and close to the cabinet of the server rack to collect noise signals and send the collected noise signals to the fault detection device.

[0128] After the fault detection device obtains the noise signal collected by the sound pickup device, the specific process of performing frequency division processing on the noise signal to obtain high-frequency information and low-frequency information, extracting the high-frequency characteristics of the high-frequency information and the low-frequency characteristics of the low-frequency information, and performing fault detection based on the high-frequency characteristics and the low-frequency characteristics to obtain the fault detection result of the server rack can refer to the description of the fault detection method for the server rack in the above embodiments.

[0129] The server fault detection system provided by this application includes: a server rack and a fault detection device. The server rack includes a sound pickup device for collecting noise signals. The fault detection device performs fault detection based on the noise signals. Since the noise signals generated when components in the server rack fail are different from the noise information generated during normal operation, performing fault detection based on the noise signals can detect abnormal situations at the first moment of a fault occurrence, improving the timeliness of fault detection. Also, because there may be some components in the server rack where the noise signals generated during a fault have prominent high-frequency parts, and for other components, the noise signals generated during a fault have prominent low-frequency parts, the high-frequency information and low-frequency information are extracted from the noise signals, and further high-frequency features and low-frequency features are extracted. When the server rack has a fault, the high-frequency features and low-frequency features include unique fault features. Performing fault detection based on the high-frequency features and low-frequency features can obtain accurate fault detection results, improving the comprehensiveness and accuracy of fault detection for the server rack.

[0130] This application also provides a server rack, which includes a sound pickup device and a server. The sound pickup device is used to collect noise signals; the server is used to receive the noise signals, perform frequency division processing on the noise signals to obtain high-frequency information and low-frequency information; extract the high-frequency features of the high-frequency information and the low-frequency features of the low-frequency information; and perform fault detection based on the high-frequency features and low-frequency features to obtain the fault detection result of the server rack.

[0131] In practical applications, the server rack further includes: a cabinet, a network communication device, a storage device, a power supply device, an auxiliary device, and an accessory device arranged in the cabinet; the network communication device includes, but is not limited to, a network switch and a router; the power supply device includes an uninterruptible power supply and a power distribution unit; the auxiliary device includes, but is not limited to, a cooling component and a detection sensor; and the accessory device includes, but is not limited to, a patch panel and an alarm unit.

[0132] After the server obtains the noise signals collected by the sound pickup device, it performs frequency division processing on the noise signals to obtain high-frequency information and low-frequency information, extracts the high-frequency features of the high-frequency information and the low-frequency features of the low-frequency information, and performs fault detection based on the high-frequency features and low-frequency features to obtain the fault detection result of the server rack. The specific process can refer to the description of the fault detection method for the server rack in the above embodiments.

[0133] Figure 8 It is a schematic structural diagram of the fault detection device for the server rack provided by this application, as Figure 8 shown, the fault detection device 80 for the server rack provided in this embodiment includes:

[0134] A noise signal acquisition module 801, which is used to acquire the noise signals collected by the sound pickup device of the server rack;

[0135] An audio separation module 802, configured to perform frequency division processing on a noise signal to obtain high-frequency information and low-frequency information;

[0136] A feature extraction module 803, configured to extract high-frequency features of the high-frequency information and low-frequency features of the low-frequency information;

[0137] A detection module 804, configured to perform fault detection based on the high-frequency features and the low-frequency features to obtain a fault detection result of the entire server cabinet.

[0138] In a possible implementation manner, the feature extraction module 803 is further configured to perform shallow feature extraction on the high-frequency information and the low-frequency information to obtain shallow high-frequency features and shallow low-frequency features, and fuse the shallow high-frequency features and the shallow low-frequency features to obtain shallow fusion features; perform deep feature extraction on the shallow fusion features and the shallow low-frequency features to obtain deep high-frequency features and low-frequency features, and fuse the deep high-frequency features and the low-frequency features to obtain high-frequency features.

[0139] In a possible implementation manner, the detection module 804 is further configured to perform feature fusion processing on the high-frequency features and the low-frequency features to obtain comprehensive audio features; perform fault detection on the comprehensive audio features through a first classifier to obtain a fault detection result of the entire server cabinet; the fault detection result is abnormal or normal.

[0140] In a possible implementation manner, the detection module 804 is further configured to perform fault detection on the high-frequency features through a second classifier to obtain a first detection result; the first detection result; perform fault detection on the low-frequency features through a third classifier to obtain a second detection result; the second detection result; determine the fault detection result of the entire server cabinet based on the first detection result and the second detection result.

[0141] In a possible implementation manner, the number of sound pickup devices is multiple, and the multiple sound pickup devices are respectively arranged in multiple detection areas inside the entire server cabinet;

[0142] The detection module 804 is further configured to, when the fault detection result is abnormal, determine a target detection area to which the sound pickup device that collects the noise signal belongs among the multiple detection areas; perform fault detection on the components included in the target detection area based on the high-frequency features and the low-frequency features to obtain faulty components.

[0143] In a possible implementation manner, the fault detection device of the entire server cabinet further includes: a warning module, configured to, when the fault detection result is abnormal, perform warning prompts through sound and light and / or send a warning message to a terminal for warning.

[0144] The fault detection device for the entire server cabinet provided in this embodiment can execute the fault detection method for the entire server cabinet provided in the above method embodiment. The implementation principle and technical effect are similar, and will not be elaborated here in this embodiment.

[0145] The fault detection device for the entire server cabinet provided in the embodiment of the present application acquires the noise signal collected by the sound pickup device of the entire server cabinet, performs frequency division processing on the noise signal to obtain high-frequency information and low-frequency information; extracts the high-frequency characteristics of the high-frequency information and the low-frequency characteristics of the low-frequency information; performs fault detection based on the high-frequency characteristics and the low-frequency characteristics to obtain the fault detection result of the entire server cabinet; since the noise signal generated when a component in the entire server cabinet fails is different from the noise information generated during normal operation, fault detection based on the noise signal can detect abnormal conditions at the first time when the fault occurs, improving the timeliness of fault detection; and because there may be some components in the entire server cabinet, the noise signal generated when a fault occurs has a prominent high-frequency part, and the noise signal generated when some other components fail has a prominent low-frequency part. The high-frequency information and low-frequency information extracted from the noise signal, and further the high-frequency characteristics and low-frequency characteristics are extracted. When there is a fault in the entire server cabinet, the high-frequency characteristics and low-frequency characteristics include unique fault characteristics. Fault detection based on the high-frequency characteristics and the low-frequency characteristics can obtain an accurate fault detection result, improving the comprehensiveness and accuracy of fault detection for the entire server cabinet.

[0146] Figure 9 It is a schematic structural diagram of the electronic device provided in the present application. As Figure 9 shown, the electronic device 90 provided in this embodiment includes: at least one processor 901 and a memory 902. Optionally, the device 90 further includes a communication component 903. Among them, the processor 901, the memory 902, and the communication component 903 are connected through a bus.

[0147] In the specific implementation process, at least one processor 901 executes the computer execution instructions stored in the memory 902, so that at least one processor 901 executes the above method.

[0148] The specific implementation process of the processor 901 can refer to the above method embodiment. The implementation principle and technical effect are similar, and will not be elaborated here in this embodiment.

[0149] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU for short), or may also be other general-purpose processors, digital signal processors (DSP for short), application specific integrated circuits (ASIC for short), etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the method disclosed in combination with the invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor.

[0150] The memory may include a random access memory (RAM), and may also include non-volatile memory (NVM), such as at least one disk memory.

[0151] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of representation, the bus in the drawings of this application is not limited to only one bus or one type of bus.

[0152] This application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0153] This application also provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the processor executes the computer-executable instructions, the above method is implemented.

[0154] The above-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, a disk or an optical disc. The readable storage medium can be any available medium accessible by a general-purpose or special-purpose computer.

[0155] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an Application Specific Integrated Circuits (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in a device.

[0156] The division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Additionally, the couplings or direct couplings or communication connections shown or discussed among each other can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0157] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0158] Furthermore, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0159] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in each embodiment of the present invention. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks or optical discs and other various media that can store program codes.

[0160] Those of ordinary skill in the art will understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments; and the foregoing storage medium includes: various media such as ROM, RAM, magnetic disks, or optical discs that can store program codes.

[0161] Finally, it should be noted that those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses, or adaptations of the present invention, which follow the general principles of the present invention and include known common knowledge or conventional technical means in the technical field not disclosed in the present invention. It is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. A method for detecting a fault in a server cabinet, characterized in that: include: Obtain the noise signal collected by the sound pickup device of the entire server cabinet; Performing frequency division processing on the noise signal to obtain high-frequency information and low-frequency information; Extracting high-frequency features of the high-frequency information and low-frequency features of the low-frequency information; Fault detection is performed based on the high-frequency characteristics and the low-frequency characteristics to obtain a fault detection result of the entire server cabinet.

2. The method according to claim 1, characterized in that The extracting the high-frequency features of the high-frequency information and the low-frequency features of the low-frequency information includes: Performing shallow feature extraction on the high-frequency information and the low-frequency information to obtain shallow high-frequency features and shallow low-frequency features, and fusing the shallow high-frequency features and the shallow low-frequency features to obtain shallow fusion features; Deep feature extraction is performed on the shallow fusion features and the shallow low-frequency features to obtain deep high-frequency features and low-frequency features, and the deep high-frequency features and the low-frequency features are fused to obtain high-frequency features.

3. The method according to claim 1 or 2, characterized in that: The performing fault detection based on the high-frequency feature and the low-frequency feature to obtain the fault detection result of the server cabinet includes: Performing feature fusion processing on the high-frequency feature and the low-frequency feature to obtain a comprehensive audio feature; The comprehensive audio feature is subjected to fault detection by a first classifier to obtain a fault detection result of the entire server cabinet; the fault detection result is abnormal or normal.

4. The method according to claim 1 or 2, characterized in that: The performing fault detection based on the high-frequency feature and the low-frequency feature to obtain the fault detection result of the server cabinet includes: Performing fault detection on the high-frequency feature by a second classifier to obtain a first detection result; Performing fault detection on the low-frequency feature by a third classifier to obtain a second detection result; Based on the first detection result and the second detection result, a fault detection result of the server cabinet is determined.

5. The method according to claim 1 or 2, characterized in that: There are multiple sound pickup devices, and the multiple sound pickup devices are respectively arranged in multiple detection areas in the server cabinet; the method also includes: In the case where the fault detection result is abnormal, determining, among the multiple detection areas, a target detection area to which the sound pickup device that collects the noise signal belongs; Based on the high-frequency features and the low-frequency features, fault detection is performed on the components included in the target detection area to obtain faulty components.

6. The method according to claim 1 or 2, characterized in that: The method further comprises: When the fault detection result is abnormal, an early warning is given by sound and light prompts and / or sending an early warning message to the terminal.

7. A server fault detection system, characterized in that: include: Server cabinets and fault detection devices; The server cabinet includes a sound pickup device, and the sound pickup device is used to collect noise signals; The fault detection device is used to receive the noise signal and perform fault detection according to the method according to any one of claims 1 to 6 based on the noise signal.

8. A fault detection device for a server cabinet, characterized in that: include: A noise signal acquisition module is used to acquire the noise signal collected by the sound pickup device of the server cabinet; An audio separation module, used to perform frequency division processing on the noise signal to obtain high-frequency information and low-frequency information; A feature extraction module, used to extract high-frequency features of the high-frequency information and low-frequency features of the low-frequency information; The detection module is used to perform fault detection based on the high-frequency characteristics and the low-frequency characteristics to obtain a fault detection result of the server cabinet.

9. A computer device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 6 when executed by a processor.

11. A computer program product, characterized in that The method comprises computer-executable instructions, which implement the method according to any one of claims 1 to 6 when executed by a processor.

Images