Machine room self-adaptive inspection method, system, equipment and medium

Abstract

The invention provides a machine room self-adaptive inspection method, system and device and a medium, and the method comprises the steps: obtaining collection data, carrying out the analysis and calculation of the collection data, and obtaining the characteristics of the device, the characteristics of the device comprise the historical failure rate of the device, the importance level of the device and the stability of the device; constructing an equipment feature vector according to the equipment features, importing the equipment feature vector into a pre-established inspection probability prediction model, and outputting the probability that the equipment needs to be inspected; generating an inspection plan corresponding to the to-be-inspected equipment according to the probability, and executing the inspection plan to obtain an inspection result; and updating the inspection probability prediction model based on the inspection result to obtain an updated inspection probability prediction model, and dynamically generating a next inspection plan according to the updated training strategy model. According to the invention, the accuracy and efficiency of inspection can be improved, the availability of equipment is guaranteed, the service continuity is improved, and the optimization of inspection human resources is realized.

Description

Technical Field

[0001] The present invention relates to the field of computer room patrol management, and particularly to a computer room adaptive patrol method, system, device and medium. Background Art

[0002] The Internet Data Center (IDC) is the core infrastructure in the digital age, and its security is the basis for ensuring the information security of enterprises and users. In the data center, equipment and systems need to be regularly patrolled and maintained to ensure normal operation. Traditional patrol strategies usually conduct regular and timed patrols of all cabinets and equipment in the computer room based on a predetermined schedule, without considering factors such as the importance of equipment, the actual health status of equipment, equipment failure historical data, historical patrol data, etc. Therefore, the traditional patrol method has the following limitations:

[0003] 1. Limitations of fixed patrol frequency: The fixed patrol plan cannot be flexibly adjusted according to the actual situation of the equipment. Therefore, even if the status of some equipment is normal, it must be patrolled according to the predetermined schedule, which will result in unnecessary patrol costs.

[0004] 2. Unable to respond to equipment changes in real time: The health status and failure risk of equipment may change within a short period of time. Under the fixed plan, these changes cannot be responded to in real time, resulting in the neglect of high-risk equipment or over-patrol of low-risk equipment.

[0005] 3. The importance of equipment is not fully considered: In the traditional patrol method, all equipment is often treated equally, while in fact, some equipment may be more important for the continuous operation of the business.

[0006] 4. Historical failures and historical patrol data are not fully considered: The importance of the historical data of equipment failures lies in that they provide valuable experience and insights to help the system more accurately predict the performance and maintenance requirements of the equipment. Summary of the Invention

[0007] Embodiments of the present invention provide a computer room adaptive patrol method, system, device and medium to solve the problems existing in the related technologies. The technical solutions are as follows:

[0008] In the first aspect, embodiments of the present invention provide a computer room adaptive patrol method, including:

[0009] Obtain the collected data, analyze and calculate the collected data to obtain equipment features, where the equipment features include the historical failure rate of the equipment, the importance level of the equipment, and the stability of the equipment;

[0010] Construct an equipment feature vector according to the equipment features, import the equipment feature vector into a pre-established patrol probability prediction model, and output the probability that the equipment needs to be patrolled;

[0011] Screen the devices to be inspected whose probabilities are higher than or equal to the probability threshold, generate corresponding inspection plans according to the devices to be inspected, and execute the inspection plans to obtain inspection results;

[0012] Update the inspection probability prediction model based on the inspection results to obtain the updated inspection probability prediction model, and dynamically generate the next inspection plan according to the updated training strategy model.

[0013] In one implementation, the collected data includes real-time data and historical data, where the real-time data is obtained by real-time monitoring of sensors.

[0014] In one implementation, it further includes:

[0015] Preprocess the collected data to obtain preprocessed data; the preprocessing includes data denoising, outlier processing, data cleaning, and data format conversion processing;

[0016] Perform stationary processing on the preprocessed data to obtain the data after stationary processing; the data after stationary processing is used for analysis and calculation to obtain device characteristics.

[0017] In one implementation, the calculation method of the historical failure rate of the device is:

[0018] Screen out the historical failure data based on the collected data, count the historical failure times and the total number of inspections according to the historical failure data, and divide the historical failure times by the total number of inspections to obtain the historical failure rate of the device.

[0019] In one implementation, the calculation method of the device stability is:

[0020] Calculate the device stability according to the standard deviation formula or the coefficient of variation formula;

[0021] Among them, the standard deviation = sqrt((Σ(data point - average value)^2) / (total number of data points));

[0022] The coefficient of variation = (standard deviation / average value) * 100.

[0023] In one implementation, the calculation method of the device feature vector is:

[0024] Device feature vector = (first weight coefficient * historical failure rate of the device + second weight coefficient * device stability + third weight coefficient * device importance level).

[0025] In one implementation, the establishment method of the inspection probability prediction model is:

[0026] Taking the device feature vector as the input of the logistic regression model and the probability that the device needs to be inspected as the output of the model, training the logistic regression model to obtain an inspection probability prediction model.

[0027] In a second aspect, an embodiment of the present invention provides a computer room adaptive inspection system that executes the computer room adaptive inspection method as described above.

[0028] In a third aspect, an embodiment of the present invention provides an electronic device, which includes: a memory and a processor. Among them, the memory and the processor communicate with each other through an internal connection path. The memory is used to store instructions, and the processor is used to execute the instructions stored in the memory. And when the processor executes the instructions stored in the memory, the processor is caused to execute the method in any one of the above aspects.

[0029] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium. The computer-readable storage medium stores a computer program. When the computer program runs on a computer, the method in any one of the above aspects is executed.

[0030] The advantages or beneficial effects in the above technical solutions at least include:

[0031] The purpose of the present invention is to provide an adaptive computer room inspection method and system to make up for the deficiencies of traditional computer room inspection methods. By real-time monitoring the device status, comprehensively considering factors such as device importance, fault history data, and historical inspection data, dynamically adjusting the number of inspected devices and the inspection frequency, improving the accuracy and efficiency of inspections, ensuring the availability of devices, improving business continuity, and achieving the optimization of inspection human resources.

[0032] The above summary is only for the purpose of the specification and is not intended to be limiting in any way. In addition to the above-described illustrative aspects, embodiments, and features, further aspects, embodiments, and features of the present invention will be readily apparent by reference to the drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In the drawings, unless otherwise specified, the same reference numerals throughout the several views denote the same or similar components or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings only depict some embodiments disclosed in accordance with the present invention and should not be regarded as limiting the scope of the present invention.

[0034] Figure 1 It is a flowchart of the computer room adaptive inspection method of the present invention;

[0035] Figure 2 It is a module diagram of the computer room adaptive inspection system of the present invention;

[0036] Figure 3 The structural block diagram of an electronic device according to an embodiment of the present invention. Specific embodiments

[0037] In the following, only some exemplary embodiments are briefly described. As those skilled in the art can recognize, the described embodiments can be modified in various different ways without departing from the spirit or scope of the present invention. Therefore, the drawings and the description are considered to be exemplary in nature and not restrictive.

[0038] Embodiment 1

[0039] This embodiment provides a method for adaptive inspection of a computer room. Referring to Figure 1 as shown, it mainly includes:

[0040] Step S1: Obtain the collected data, analyze and calculate the collected data to obtain device characteristics, where the device characteristics include the historical failure rate of the device, the importance level of the device, and the stability of the device;

[0041] Step S2: Construct a device feature vector according to the device characteristics, import the device feature vector into a pre-established inspection probability prediction model, and output the probability that the device needs to be inspected;

[0042] Step S3: Screen the devices to be inspected whose probability is higher than or equal to the probability threshold, generate a corresponding inspection plan according to the devices to be inspected, and execute the inspection plan to obtain the inspection result;

[0043] Step S4: Update the inspection probability prediction model based on the inspection result to obtain an updated inspection probability prediction model, and dynamically generate the next inspection plan according to the updated training strategy model.

[0044] Among them, the collected data includes real-time data and historical data. Among them, the real-time data is collected by a variety of different sensors. The sensors can be temperature sensors, humidity sensors, vibration sensors, etc. Correspondingly, the real-time data obtained by the sensors includes types of data such as temperature data, humidity data, and vibration data.

[0045] The acquisition and detection frequency of the sensors can be set according to actual needs. The sensors regularly collect real-time data to ensure continuous detection of the device status. And the collected data in this embodiment is collected in a time series manner, that is, the collected data not only includes real-time data for regularly monitoring the real-time status of the device, but also includes historical data. The historical data includes the data collected by the sensors when the computer room fails, the data collected by the sensors when the computer room is working normally, and historical inspection data; at the same time, it is also necessary to collect the data of the importance level of the device, and store it together with all the data collected during the operation of the computer room, which together serves as an important data source for the computer room inspection evaluation in this embodiment.

[0046] All the collected data are stored in a central database. Since the current data is relatively messy, data preprocessing is required. The preprocessing mainly includes steps such as data denoising, outlier handling, data cleaning, and data format conversion.

[0047] After that, time series analysis, statistical analysis, and machine learning methods are used to analyze the preprocessed data. At the same time, the dynamic changes of device data need to be considered. Techniques such as exponential smoothing and moving average can be used to smooth the data, obtaining the smoothed data. Useful features are extracted from the smoothed data, such as the historical failure rate of the device and the importance level of the device. Among them, the importance level of the device can be pre-divided into three categories: high risk, medium risk, and low risk according to preset rules.

[0048] The historical failure rate of the device is calculated based on the historical failure data of the device. The historical failure data records the failure type, frequency, and impact. The historical failure data is crucial for evaluating the failure risk of the device. They help the system understand the past performance and problems of the device. The calculation method of the historical failure rate of the device in this embodiment is as follows:

[0049] Based on the historical collected data, the historical failure data is screened out. According to the historical failure data, the historical failure times and the total number of inspections are statistically analyzed. The historical failure rate of the device is obtained by dividing the historical failure times by the total number of inspections; its formula is the historical failure rate of the device = (the number of historical device failures / the total number of inspections).

[0050] The stability of the device can be quantified using the standard deviation or the coefficient of variation. The standard deviation represents the degree of data fluctuation, and the calculation formula is as follows:

[0051] Standard deviation = sqrt((Σ(data point - mean)²) / (total number of data points));

[0052] Coefficient of variation = (standard deviation / mean) * 100.

[0053] The construction of the device feature vector can be based on factors such as the historical failure rate of the device, the stability of the device, and the importance of the device. At the same time, the weighted sum method can be used to construct the feature vector. For example:

[0054] Device feature vector = (first weight coefficient * historical failure rate of the device + second weight coefficient * stability of the device + third weight coefficient * importance of the device); among them, the weight coefficients can be set and adjusted according to the actual situation. In this embodiment, the first weight coefficient can be 0.4, the second weight coefficient can be 0.3, and the third weight coefficient can be 0.3.

[0055] After obtaining the device feature vector in the above manner, the inspection devices can be selected and an inspection plan can be formulated through a pre-established inspection strategy model. Among them, since the inspection strategy model is mainly used to calculate the probability that a device needs to be inspected, the inspection strategy model can also be called an inspection probability prediction model.

[0056] In this embodiment, logistic regression is used as the inspection probability prediction model. The input of the logistic regression model is the device feature vector, and the output is the probability that the device needs to be inspected. The model can be trained using a large amount of historical data, and the model parameters can be estimated through maximum likelihood estimation. The following formula:

[0057] Inspection probability = 1 / (1 + e^(-z));

[0058] z = θ0 + θ1 * device feature vector 1 + θ2 * device feature vector 2 + θ3 * device feature vector 3;

[0059] Among them, θ0, θ1, θ2, and θ3 are model parameters.

[0060] For each device in the computer room, the established inspection probability prediction model can be used to predict the probability that the device needs to be inspected. The probability value output by the model ranges from 0 to 1.

[0061] Sort all devices in descending order of inspection probability to select the devices that need to be inspected. The method of selecting the devices that need to be inspected can be to preset a probability threshold, compare the probability threshold with the probability value output by the model. If the probability value output by the model is higher than or equal to the probability threshold, it means that the device needs to be inspected; if the probability value output by the model is lower than the probability threshold, it means that the device does not need to participate in the inspection. For example, if the probability threshold is 0.8, then select the devices with an inspection probability higher than or equal to 0.8 as the devices that need to participate in the inspection.

[0062] After determining the list of devices that need to be inspected, generate a corresponding inspection plan according to the devices that need to be inspected and execute the corresponding inspection tasks. After each inspection, the operation and maintenance personnel record the inspection results, including information such as the actual status of the device and whether a failure occurs. The recorded inspection results can be used for the adaptive learning of the model to further optimize the prediction accuracy of the inspection probability prediction model. Specifically:

[0063] Using the inspection results to update the parameter model of the logistic regression model. Adaptive learning can use methods such as gradient descent to update the model parameters. At the same time, re-estimate the weight parameters, and genetic algorithms or simulated annealing algorithms can be used for parameter optimization. Finally, obtain the updated inspection probability prediction model, adjust the inspection strategy according to the new data to adapt to the changes in the device status, and improve the prediction accuracy.

[0064] The inspection probability prediction model generated by adaptive learning can output more accurate probability prediction results. At this time, the next inspection plan can be dynamically generated based on the updated inspection probability prediction model. This plan will take into account the current situation of the computer room to improve the accuracy and efficiency of inspections.

[0065] The computer room adaptive inspection method in this embodiment synthesizes data analysis of four key dimensions, including equipment importance, equipment status, equipment historical failures, and equipment historical inspection data. Through this comprehensive data analysis, the system can more accurately evaluate the status and urgency of each device. Compared with traditional inspection methods, this innovation no longer requires fixed inspections of all devices regardless of their actual status, but rather screens out devices that need to be inspected first based on multi-dimensional eigenvalue calculations, thereby improving the accuracy and efficiency of computer room inspections.

[0066] A dynamic inspection plan is formulated according to the real-time changes in equipment status. This method achieves a significant advantage over traditional methods. The equipment status can be reflected by equipment stability. If the equipment status remains stable, the inspection interval can be extended, unnecessary inspection frequencies can be reduced, and maintenance costs can be lowered. However, when the equipment status shows abnormalities or the risk increases, the system will automatically conduct an early inspection to detect problems in a timely manner and take measures to repair them. This breaks the traditional fixed inspection mode, can better adapt to the actual changes in equipment status, thereby detecting equipment failures, abnormalities, or potential problems earlier, reducing the failure rate of equipment, improving the reliability of equipment, and ensuring the stable operation of computer room equipment to the greatest extent.

[0067] In this embodiment, through the adaptive learning method, this innovation can not only more accurately consider historical data and multi-dimensional analysis when formulating a plan at the initial stage of inspection, but also continuously iterate, optimize, and adjust the inspection plan according to the actual inspection results and the dynamic changes in equipment status. If the status of a certain device changes, the system will automatically adjust its eigenvalue and the next inspection time to adapt to the new situation. This self-learning process ensures the continuous improvement of the inspection strategy and makes the inspection and maintenance work of computer room equipment more intelligent and efficient. This adaptive learning method makes computer room inspections more flexible and intelligent, can be adjusted according to the actual situation, and ensures the stability and reliability of computer room equipment.

[0068] Embodiment 2

[0069] This embodiment provides a computer room adaptive inspection system that executes the computer room adaptive inspection method as in Embodiment 1. Refer to Figure 2 As shown, the system mainly includes an equipment importance evaluation module, an equipment status evaluation module, a historical failure evaluation module, a historical inspection evaluation module, an inspection decision-making module, an inspection execution module, an inspection feedback module, and an adaptive learning model.

[0070] The device importance assessment module classifies the importance of devices in the computer room. The computer room administrator assigns an importance level to each device, usually divided into three categories: high, medium, and low. This helps determine the relative importance of devices in terms of business continuity, so as to better allocate resources and decide the priority of inspections.

[0071] The device status assessment module monitors the real-time status of devices based on multiple sensors on the devices, such as parameters like temperature and noise. Real-time data is collected regularly to ensure continuous monitoring of the device status, and then the sensor data is analyzed in real time. Pre-defined thresholds and rules are used to determine whether the device is in a high-risk, medium-risk, or low-risk state.

[0072] The historical fault assessment module retrieves the historical fault data of devices, including the fault type, frequency, and impact. According to certain rules, the devices are divided into three categories: high risk, medium risk, and low risk. Historical fault data is crucial for assessing the fault risk of devices and can help the system understand the past performance and problems of the devices.

[0073] The historical inspection assessment module retrieves the historical inspection data, including the number of inspections and inspection results. By analyzing the historical inspection data, it is determined which devices usually need maintenance and which devices usually remain in a normal state.

[0074] The inspection decision-making module comprehensively considers the data in four dimensions: the importance of the device, the real-time status, historical faults, and inspection data. Based on these data, the characteristic value and inspection frequency of each device are calculated. The characteristic value is reflected in the form of a device characteristic vector. Sorted according to the inspection probability, it is used to obtain the list of devices that most need to be inspected and provides a basis for the formulation and generation of the inspection plan.

[0075] The inspection execution module is used to execute the inspection work according to the inspection plan. During the execution of the inspection work, only the high-risk devices listed in the inspection plan need to be concerned, without the need to inspect all devices in the computer room, which reduces the inspection pressure.

[0076] The inspection feedback module is used to update the status and problems of the inspected devices at any time during the inspection process using mobile terminal devices to ensure that problems are discovered and processed in a timely manner.

[0077] The adaptive learning module, based on the inspection results and the actual device status, the system conducts self-learning and adaptive optimization. The optimization process includes updating the model, adjusting the feature weights, improving the prediction algorithm, etc., to improve the accuracy of the model. Then, based on the learned knowledge, the system dynamically formulates the next inspection plan, including which devices need to be inspected, when to inspect, and adjusting the inspection frequency, etc.

[0078] Through this series of modules, the system can achieve adaptive inspection, dynamically adjust the inspection plan according to the real-time status and historical data of the equipment, ensure that the equipment with a high probability of needing inspection is covered, and at the same time reduce the inspection frequency of the equipment with a low probability. This can improve the inspection efficiency, reduce the maintenance cost, and ensure the reliability of the computer room equipment.

[0079] For the functions of the modules in the system of the embodiment of the present invention, reference can be made to the corresponding descriptions in the above method, which will not be elaborated here.

[0080] Embodiment Three

[0081] Figure 3 The structural block diagram of an electronic device according to an embodiment of the present invention is shown. As Figure 3 shown, the electronic device includes: a memory 100 and a processor 200, and a computer program that can run on the processor 200 is stored in the memory 100. When the processor 200 executes the computer program, the computer room adaptive inspection method in the above embodiment is implemented. The number of the memory 100 and the processor 200 can be one or more.

[0082] The electronic device further includes:

[0083] a communication interface 300, which is used to communicate with external devices and perform data interaction and transmission.

[0084] If the memory 100, the processor 200, and the communication interface 300 are implemented independently, the memory 100, the processor 200, and the communication interface 300 can be connected to each other through a bus and complete communication with each other. The bus can 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 convenience of representation, Figure 3 only a thick line is shown in, but it does not mean that there is only one bus or one type of bus.

[0085] Optionally, in specific implementation, if the memory 100, the processor 200, and the communication interface 300 are integrated on a chip, the memory 100, the processor 200, and the communication interface 300 can complete communication with each other through an internal interface.

[0086] The embodiment of the present invention provides a computer-readable storage medium, which stores a computer program, and when the program is executed by a processor, the method provided in the embodiment of the present invention is implemented.

[0087] An embodiment of the present invention also provides a chip, which includes a processor for calling and running instructions stored in a memory, so that a communication device installed with the chip executes the method provided by the embodiment of the present invention.

[0088] An embodiment of the present invention also provides a chip, including: an input interface, an output interface, a processor, and a memory. The input interface, the output interface, the processor, and the memory are connected through an internal connection path. The processor is configured to execute code in the memory. When the code is executed, the processor is configured to execute the method provided by the embodiment of the invention.

[0089] It should be understood that the above-mentioned processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. It is worth noting that the processor may be a processor supporting the advanced RISC machines (ARM) architecture.

[0090] Further, optionally, the above-mentioned memory may include a read-only memory and a random access memory, and may further include a non-volatile random access memory. The memory may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may include a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may include a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available. For example, static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct rambus random access memory (DR RAM).

[0091] In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it may be implemented in whole or in part 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, the processes or functions according to the present invention are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium.

[0092] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0093] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In the description of the present invention, "a plurality of" means two or more unless otherwise specifically defined.

[0094] Any process or method description represented in a flowchart or described in other ways herein can be understood as representing a module, segment, or portion of code including one or more executable instructions for implementing a specific logical function or process. And the scope of the preferred embodiments of the present invention includes additional implementations, where the functions can be executed in a substantially simultaneous manner or in a reverse order according to the involved functions, rather than in the order shown or discussed.

[0095] The logic and / or steps represented in a flowchart or described in other ways herein, for example, can be considered as a sequenced list of executable instructions for implementing a logical function, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in combination with these instruction execution systems, apparatuses, or devices.

[0096] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. All or part of the steps of the method in the above embodiments can be completed by a program instructing relevant hardware, and this program can be stored in a computer-readable storage medium. When this program is executed, it includes one or a combination of the steps of the method embodiment.

[0097] In addition, each functional unit in various embodiments of the present invention may be integrated into a processing module, may exist separately as individual physical units, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. When the above-mentioned integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium. The storage medium may be a read-only memory, a magnetic disk, an optical disc, or the like.

[0098] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of various changes or substitutions, and these should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. An adaptive inspection method for computer rooms, characterized in that, Including: Obtain the collected data, analyze and calculate the collected data to obtain device characteristics, where the device characteristics include the historical failure rate of the device, the importance level of the device, and the stability of the device; Construct a device feature vector according to the device characteristics, import the device feature vector into a pre-established inspection probability prediction model, and output the probability that the device needs to be inspected; Screen the devices to be inspected whose probability is higher than or equal to the probability threshold, generate a corresponding inspection plan according to the devices to be inspected, and execute the inspection plan to obtain an inspection result; Update the inspection probability prediction model based on the inspection result to obtain an updated inspection probability prediction model, and dynamically generate the next inspection plan according to the updated training strategy model.

2. The adaptive inspection method for computer rooms according to claim 1, characterized in that, The collected data includes real-time data and historical data, where the real-time data is obtained by real-time monitoring of sensors.

3. The adaptive inspection method for computer rooms according to claim 1, characterized in that, Also including: Preprocess the collected data to obtain preprocessed data; The preprocessing includes data denoising, outlier processing, data cleaning, and data format conversion; Perform stationary processing on the preprocessed data to obtain data after stationary processing; The data after stationary processing is used for analysis and calculation to obtain the device characteristics.

4. The adaptive inspection method for computer rooms according to claim 1, characterized in that, The calculation method of the historical failure rate of the device is: Based on the collected data, screen out historical failure data, count the number of historical failures and the total number of inspections according to the historical failure data, and divide the number of historical failures by the total number of inspections to obtain the historical failure rate of the device.

5. The adaptive inspection method for computer rooms according to claim 1, characterized in that, The calculation method of the device stability is: Calculate the device stability according to the standard deviation formula or the coefficient of variation formula; Among them, the standard deviation formula is: standard deviation = sqrt((Σ(data point - average value)^2) / (total number of data points)); The coefficient of variation formula is: coefficient of variation = (standard deviation / average value) * 100.

6. The adaptive inspection method for computer rooms according to claim 1, characterized in that, The calculation method of the device feature vector is: Device feature vector = (first weight coefficient * historical failure rate of the device + second weight coefficient * device stability + third weight coefficient * device importance level).

7. The adaptive inspection method for computer rooms according to claim 1, characterized in that, The establishment method of the inspection probability prediction model is: Use the device feature vector as the input of the logistic regression model, use the probability that the device needs to be inspected as the output of the model, and train the logistic regression model to obtain the inspection probability prediction model.

8. An adaptive inspection system for computer rooms, characterized in that, Execute the computer room adaptive inspection method according to any one of claims 1 to 7.

9. An electronic device, characterized in that, Including: A processor and a memory, where instructions are stored in the memory, and the instructions are loaded and executed by the processor to implement the computer room adaptive inspection method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, the computer room adaptive inspection method according to any one of claims 1 to 7 is implemented.

Images