Cluster server intelligent fault prediction method, system, terminal and storage medium

Abstract

The present invention relates to the field of server technology, and more specifically to a method, system, terminal, and storage medium for intelligent fault prediction of cluster servers. The method comprises: acquiring real-time information about a device; analyzing device wear status data based on the real-time information; and obtaining device status data based on the wear status data; and predicting device faults based on the device status data. The present invention enables detection and assessment of the status of components in the server, enabling timely detection of faults.

Description

Technical Field

[0001] The present invention relates to the technical field of servers, and in particular to a cluster server intelligent fault prediction method, system, terminal and storage medium. Background Art

[0002] A server is a type of computer that runs faster, handles higher loads, and is more expensive than regular computers. It provides computing or application services to other clients on a network, such as PCs, smartphones, ATMs, and even large devices like train systems. Servers feature high-speed CPU computing power, long-term reliable operation, robust I / O external data throughput, and excellent scalability.

[0003] There may be some faults in the server during operation. If the server stops working due to a fault, it will cause certain losses. Therefore, it is necessary to predict the fault in advance. In order to solve this technical problem, a cluster server intelligent fault prediction method, system, terminal and storage medium are proposed. Summary of the Invention

[0004] In order to solve the technical problems existing in the above-mentioned prior art, the present invention provides a cluster server intelligent fault prediction method, system, terminal and storage medium.

[0005] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0006] In a first aspect, in one embodiment provided by the present invention, a cluster server intelligent fault prediction method is provided, the method comprising the following steps:

[0007] Get real-time information of the device;

[0008] Analyze the equipment's loss status data based on the equipment's real-time information, and obtain equipment status data based on the equipment's loss status data analysis;

[0009] Predict equipment failures based on equipment status data.

[0010] As a further solution of the present invention, the real-time information of the device includes data such as vibration conditions, temperature changes, and current stability.

[0011] As a further solution of the present invention, the device loss status data is analyzed according to the real-time information of the device, and the device status data is obtained according to the loss status data analysis of the device; including: the device loss status data is analyzed by a BP neural network prediction model to obtain the device status data.

[0012] As a further solution of the present invention, the BP neural network prediction model construction step includes the following steps:

[0013] S201, constructing a BP neural network prediction model based on vibration conditions, temperature changes, and current stability elements;

[0014] S202. Establish a sample data set based on {(vibration condition, temperature change, current stability), fault type};

[0015] S203, using a normalization formula to normalize the sample data set values ​​so that the range is between 0 and 1;

[0016] S204, inputting the sample data set into the constructed BP neural network prediction model, and outputting fault type data;

[0017] S205, judging the calculation error based on the fault type data, and adjusting the weights from the hidden layer to the output layer and from the input layer to the hidden layer of the BP neural network prediction model;

[0018] S206. Repeat the above steps S204-S205 until the error meets the set value.

[0019] As a further solution of the present invention, the calculation error is judged based on the fault type data, and the weights from the hidden layer to the output layer and from the input layer to the hidden layer of the BP neural network prediction model are adjusted, including: using the least squares method to calculate the error of the model, and updating the weights from back to front through the gradient descent method.

[0020] As a further solution of the present invention, the step S30 of predicting equipment failure based on equipment status data includes the following steps:

[0021] S301, generating a state transition probability distribution matrix for each associated component;

[0022] S302: Predict the status of associated components based on the Markov chain.

[0023] As a further solution of the present invention, the step S301, generating a state transition probability distribution matrix of each associated component, includes the following steps:

[0024] S3011. Select a historical period width T and obtain the status of the main component and the status of its associated components in each unit time period;

[0025] S3012. Set all predicted states of the main component and all states of the associated components, divide them according to the states of the main component, calculate the frequency of data state transitions of the associated components in adjacent time periods when the main component is in each state, and obtain the predicted state probability distribution of the main component and the state transition probability distribution of the associated components.

[0026] In a second aspect, in another embodiment provided by the present invention, a cluster server intelligent fault prediction system is provided, which includes: a device monitoring terminal 100, a device status analysis module 200 and a device status prediction module 300;

[0027] The device monitoring terminal 100 is used to obtain real-time information about the device, wherein the real-time information about the device includes vibration conditions, temperature changes, and current stability data;

[0028] The device status analysis module 200 is used to analyze the wear status data of the device according to the real-time information of the device, and obtain the device status data according to the wear status data of the device;

[0029] The device status prediction module 300 is used to predict device failures based on device status data.

[0030] In a third aspect, in another embodiment provided by the present invention, a terminal is provided, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the cluster server intelligent fault prediction method when loading and executing the computer program.

[0031] In a fourth aspect, in another embodiment provided by the present invention, a storage medium is provided, which stores a computer program, and when the computer program is loaded and executed by a processor, the steps of the cluster server intelligent fault prediction method are implemented.

[0032] The technical solution provided by the present invention has the following beneficial effects:

[0033] The present invention provides a cluster server intelligent fault prediction method, system, terminal and storage medium. The method obtains real-time information of the device; analyzes the loss status data of the device based on the real-time information of the device, and obtains device status data based on the loss status data of the device; and predicts device failure based on the device status data. The present invention realizes the detection and evaluation of the status of the components in the server, and can detect faults in a timely manner.

[0034] These and other aspects of the present invention will become more readily apparent in the following description of the embodiments. It should be understood that the above general description and the following detailed description are merely exemplary and explanatory and are not intended to limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can be obtained based on these drawings without paying any creative work.

[0036] Figure 1 This is a flow chart of a cluster server intelligent fault prediction method according to an embodiment of the present invention;

[0037] Figure 2 This is a flowchart of S30 in the cluster server intelligent fault prediction method according to one embodiment of the present invention;

[0038] Figure 3 This is a flowchart of S302 in the cluster server intelligent fault prediction method according to one embodiment of the present invention;

[0039] Figure 4 is a neural network model diagram;

[0040] Figure 5 This is a structural block diagram of a cluster server intelligent fault prediction system according to an embodiment of the present invention;

[0041] Figure 6 This is a structural block diagram of a terminal according to an embodiment of the present invention.

[0042] In the figure: device monitoring terminal-100, device status analysis module-200, device status prediction module-300, processor-401, communication interface-402, memory-403, communication bus-404. DETAILED DESCRIPTION

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0044] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, combined, or partially merged, so the actual execution order may vary depending on the actual situation.

[0045] It should be understood that the terms used in this specification are only for the purpose of describing particular embodiments and are not intended to limit the present invention. As used in the specification and appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0046] Specifically, the embodiments of the present invention are further described below with reference to the accompanying drawings.

[0047] See also Figure 1 , Figure 1 This is a flow chart of a cluster server intelligent fault prediction method provided by an embodiment of the present invention. Figure 1 As shown, the cluster server intelligent fault prediction method includes steps S10 to S30.

[0048] S10, obtaining real-time information of the device;

[0049] In an embodiment of the present invention, the real-time information of the device includes data such as vibration conditions, temperature changes, and current stability.

[0050] In an embodiment of the present invention, the acquiring of the real-time information of the device includes performing noise reduction processing on the acquired real-time information of the device to remove noise data.

[0051] S20, analyzing the loss status data of the device according to the real-time information of the device, and obtaining device status data according to the loss status data of the device;

[0052] The method of analyzing the loss status data of the equipment according to the real-time information of the equipment and obtaining the equipment status data according to the loss status data of the equipment includes: obtaining the equipment status data by analyzing the loss status data of the equipment through a BP neural network prediction model.

[0053] In an embodiment of the present invention, the BP neural network prediction model construction step includes the following steps:

[0054] S201, constructing a BP neural network prediction model based on vibration conditions, temperature changes, and current stability elements;

[0055] S202. Establish a sample data set based on {(vibration condition, temperature change, current stability), fault type};

[0056] S203, using a normalization formula to normalize the value so that it ranges from 0 to 1;

[0057] S204, inputting the sample data set into the constructed BP neural network prediction model, and outputting fault type data;

[0058] S205, judging the calculation error based on the fault type data, and adjusting the weights from the hidden layer to the output layer and from the input layer to the hidden layer of the BP neural network prediction model;

[0059] S206. Repeat the above steps S204-S205 until the error meets the set value.

[0060] In an embodiment of the present invention, the step S201 of constructing a BP neural network prediction model based on vibration conditions, temperature changes, and current stability elements includes:

[0061] The vibration condition, temperature change and current stability are divided into segments, and the segments constitute the variation range space, which are: V = {V1, V2, ..., V v}, T={T1,T2,...,T t}, E={E1,E2,...,E e};

[0062] Construct the early fault symptom feature space A={A1,A2,...,A m}, the component fault feature space is B={B1,B2,...,B n};

[0063] Each element Ai in the fault symptom feature space is a triplet A i ={V j ,T k ,E l}, where j∈{1,2,...,v},k∈{1,2,...,t},l∈{1,2,...,e};

[0064] Collect {A i →B j ,i∈{1,2,..m},j∈{1,2,...,n}} mapping relationship data set, and normalize the values ​​in the data set: use the normalization formula to normalize the values ​​so that their range is between 0 and 1.

[0065] In an embodiment of the present invention, the calculation error is determined based on the fault type data, and the weights from the hidden layer to the output layer and from the input layer to the hidden layer of the BP neural network prediction model are adjusted, including: calculating the error of the model using the least squares method, and updating the weights from back to front through the gradient descent method.

[0066] S30. Predicting equipment failure based on equipment status data.

[0067] In an embodiment of the present invention, the step S30 of predicting device failure based on device status data includes the following steps:

[0068] S301, generating a state transition probability distribution matrix for each associated component;

[0069] S302: Predict the status of associated components based on the Markov chain.

[0070] S301, generating a state transition probability distribution matrix of each associated component, comprises the following steps:

[0071] S3011. Select a historical period width T and obtain the status of the main component and the status of its associated components in each unit time period;

[0072] S3012. Set all predicted states of the main component and all states of the associated components, divide them according to the states of the main component, calculate the frequency of data state transitions of the associated components in adjacent time periods when the main component is in each state, and obtain the predicted state probability distribution of the main component and the state transition probability distribution of the associated components.

[0073] For example, assume that all predicted states of main component a are E={E1, E2, E3, E4}, where E1 represents a state with no fault, E2 represents a state with a slight fault, E3 represents a state with a moderate fault, and E4 represents a state with a severe fault; all states of associated component b are Q={Q1, Q2, ..., Q4}, where Q1 represents a state with no fault, Q2 represents a state with a slight fault, Q3 represents a state with a moderate fault, and Q4 represents a state with a severe fault. Divide by the state of the main component, and calculate the frequency of data state transition of associated component b in adjacent time periods when the main component is in each state. In E i In this state, the transfer distribution of the associated component b is: Then the state transition probability distribution of the associated components is:

[0074] In an embodiment of the present invention, S302, predicting the status of associated components according to a Markov chain, includes the following steps:

[0075] S3021. Obtain the predicted state probability distribution of the main component a and the current state of the associated component b.

[0076] S3022. Calculate the probability distribution of the next state of associated component b for each state of primary component a using Markov's law. Assume that all predicted states of primary component a are E = {E1, E2, E3, E4}. All states of associated component b are Q = {Q1, Q2, ..., Q4}.

[0077] For the initial prediction state of the main component a, set W0 = [w01 ,w 02 ,...,w 0i ,...,w 0m ], where w 0i Indicates that at time t=0, the predicted state of the main component is E i The probability of .

[0078]

[0079] The predicted state E of the main component a i , i∈{1,2,...,m} as an example, select any time point as the starting point, take the state at that moment as the initial state, set U0={0,...,1,...,0}, U0 represents a 1x n unit row vector, if its pth component is 1 and the other components are 0, it means that the initial state of the system is in the pth state, calculate the state probability U at the next moment i1 for:

[0080] U i1 =U0*P i =[p i (1),p i (2),...,p i (k),...,p i (n)]

[0081] S3023. Calculate the probability distribution of the next state of the associated component b based on the predicted probability distribution of the main component a:

[0082]

[0083] S3024. Summarize and derive a loss prediction of each component, including main components and associated components.

[0084] The present invention combines a neural network algorithm and a Markov chain to analyze and judge the status of equipment components. Regardless of whether component loss changes the parameters of the equipment, the system can detect and evaluate the status of the component.

[0085] It should be understood that, although the above is described in a certain order, these steps are not necessarily performed in sequence according to the above order. Unless there is clear explanation in this article, the execution of these steps does not have strict order restriction, and these steps can be performed in other orders. Moreover, a part of the steps of the present embodiment may include multiple steps or multiple stages, and these steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be performed in turn or alternately with at least a portion of the steps or stages in other steps or other steps.

[0086] In one embodiment, see Figure 3 As shown, an embodiment of the present invention further provides a cluster server intelligent fault prediction system, which includes a device monitoring terminal 100 , a device status analysis module 200 and a device status prediction module 300 .

[0087] The device monitoring terminal 100 is used to obtain real-time information about the device, wherein the real-time information about the device includes data such as vibration, temperature change, and current stability.

[0088] In an embodiment of the present invention, the acquiring of the real-time information of the device includes performing noise reduction processing on the acquired real-time information of the device to remove noise data.

[0089] The device status analysis module 200 is used to analyze the wear status data of the device according to the real-time information of the device, and obtain the device status data according to the wear status data of the device.

[0090] The equipment status analysis module 200 analyzes the equipment loss status data through a BP neural network prediction model to obtain equipment status data.

[0091] like Figure 4 As shown, the BP neural network prediction model construction step includes the following steps:

[0092] S201, constructing a BP neural network prediction model based on vibration conditions, temperature changes, and current stability elements;

[0093] S202. Establish a sample data set based on {(vibration condition, temperature change, current stability), fault type};

[0094] S203, using a normalization formula to normalize the sample data set values ​​so that the range is between 0 and 1;

[0095] S204, inputting the sample data set into the constructed BP neural network prediction model, and outputting fault type data;

[0096] S205, judging the calculation error based on the fault type data, and adjusting the weights from the hidden layer to the output layer and from the input layer to the hidden layer of the BP neural network prediction model;

[0097] S206. Repeat the above steps S204-S205 until the error meets the set value.

[0098] The step S201, constructing a BP neural network prediction model based on vibration conditions, temperature changes, and current stability elements, includes:

[0099] The vibration condition, temperature change and current stability are divided into segments, and the segments constitute the variation range space, which are: V = {V1, V2, ..., V v}, T={T1,T2,...,T t}, E={E1,E2,...,E e};

[0100] Construct the early fault symptom feature space A={A1,A2,...,A m}, the component fault feature space is B={B1,B2,...,B n};

[0101] Each element A in the fault symptom feature space i A three-element group i ={V j ,T k ,E l}, where j∈{1,2,...,v},k∈{1,2,...,t},l∈{1,2,...,e};

[0102] Collect {A i →B j ,i∈{1,2,..m},j∈{1,2,...,n}} mapping relationship data set, and normalize the values ​​in the data set: use the normalization formula to normalize the values ​​so that their range is between 0 and 1.

[0103] The method of judging and calculating the error based on the fault type data and adjusting the weights from the hidden layer to the output layer and from the input layer to the hidden layer of the BP neural network prediction model includes: calculating the error of the model using the least squares method and updating the weights from back to front using the gradient descent method.

[0104] The device status prediction module 300 is used to predict device failures based on device status data.

[0105] In an embodiment of the present invention, the step S30 of predicting device failure based on device status data includes the following steps:

[0106] S301, generating a state transition probability distribution matrix for each associated component;

[0107] S302: Predict the status of associated components based on the Markov chain.

[0108] S301, generating a state transition probability distribution matrix of each associated component, comprises the following steps:

[0109] S3011. Select a historical period width T and obtain the status of the main component and the status of its associated components in each unit time period;

[0110] S3012. Set all predicted states of the main component and all states of the associated components, divide them according to the states of the main component, calculate the frequency of data state transitions of the associated components in adjacent time periods when the main component is in each state, and obtain the predicted state probability distribution of the main component and the state transition probability distribution of the associated components.

[0111] For example, assume that all predicted states of main component a are E={E1, E2, E3, E4}, where E1 represents a state with no fault, E2 represents a state with a slight fault, E3 represents a state with a moderate fault, and E4 represents a state with a severe fault; all states of associated component b are Q={Q1, Q2, ..., Q4}, where Q1 represents a state with no fault, Q2 represents a state with a slight fault, Q3 represents a state with a moderate fault, and Q4 represents a state with a severe fault. Divide by the state of the main component, and calculate the frequency of data state transition of associated component b in adjacent time periods when the main component is in each state. In E i In this state, the transfer distribution of the associated component b is: Then the state transition probability distribution of the associated components is:

[0112] In the embodiment of the present invention, the step S302 of predicting the status of associated components according to the Markov chain includes the following steps:

[0113] S3021. Obtain the predicted state probability distribution of the main component a and the current state of the associated component b.

[0114] S3022. Calculate the probability distribution of the next state of the associated components for each state of the primary component based on Markov's law. Assume that all predicted states of primary component a are E = {E1, E2, E3, E4}. All states of associated component b are Q = {Q1, Q2, ..., Q4}.

[0115] For the initial prediction state of the main component a, set W0 = [w 01 ,w 02 ,...,w 0i ,...,w 0m ], where w 0i Indicates that at time t=0, the predicted state of the main component is E i The probability of .

[0116] The predicted state E of the main component ai , i∈{1,2,...,m} as an example, select any time point as the starting point, take the state at that moment as the initial state, set U0={0,...,1,...,0}, U0 represents a 1x n unit row vector, if its pth component is 1 and the other components are 0, it means that the initial state of the system is in the pth state, calculate the state probability U at the next moment i1 for:

[0117] U i1 =U0*P i =[p i (1),p i (2),...,p i (k),...,p i (n)]

[0118] S3023. Calculate the probability distribution of the next state of the associated components based on the predicted probability distribution of the main components.

[0119]

[0120] S3024. Summarize and derive a loss prediction of each component, such as the main component a and the associated component b.

[0121] In one embodiment, see Figure 5 As shown, an embodiment of the present invention further provides a terminal, including a processor 401, a communication interface 402, a memory 403 and a communication bus 404, wherein the processor 401, the communication interface 402, and the memory 403 communicate with each other through the communication bus 404.

[0122] Memory 403, used for storing computer programs;

[0123] The processor 401 is configured to execute the computer program stored in the memory 403 to execute the cluster server intelligent fault prediction method. The processor implements the steps in the above method embodiment when executing instructions.

[0124] The communication bus mentioned in the terminal can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is used in the figure, but this does not mean that there is only one bus or only one type of bus.

[0125] The communication interface is used for communication between the above terminal and other devices.

[0126] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage. Alternatively, the memory may be at least one storage device located away from the processor.

[0127] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0128] The terminal includes a user device and a network device. The user device includes, but is not limited to, a computer, a smartphone, a PDA, etc.; the network device includes, but is not limited to, a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of computers or network servers. Cloud computing is a type of distributed computing, a super virtual computer composed of a group of loosely coupled computers. The terminal can operate independently to implement the present invention, or it can access a network and implement the present invention through interactive operations with other terminals in the network. The network in which the terminal is located includes, but is not limited to, the Internet, a wide area network, a metropolitan area network, a local area network, a VPN network, etc.

[0129] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0130] In one embodiment of the present invention, a storage medium is further provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiment are implemented.

[0131] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes in the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein can include at least one of non-volatile and volatile memory.

[0132] It should be understood that, as used herein, the singular form "a" or "an" is intended to include the plural form as well, unless the context clearly supports an exception. It should also be understood that, as used herein, "and / or" refers to any and all possible combinations of one or more of the items listed in association. The serial numbers of the embodiments disclosed in the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0133] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the disclosure of the embodiments of the present invention (including the claims) is limited to these examples. Within the spirit of the embodiments of the present invention, the technical features of the above embodiments or different embodiments may be combined, and there are many other variations of different aspects of the above embodiments of the present invention, which are not provided in detail for the sake of simplicity. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present invention should be included in the scope of protection of the embodiments of the present invention.

Claims

1. A cluster server intelligent fault prediction method, characterized in that: The method includes: obtaining real-time information of the device, wherein the real-time information of the device includes vibration conditions, temperature changes, and current stability data; Analyzing the loss status data of the device according to the real-time information of the device, and obtaining the device status data according to the loss status data of the device, which includes: analyzing the loss status data of the device through a BP neural network prediction model to obtain the device status data; Predict equipment failures based on equipment status data, including: S301, generating a state transition probability distribution matrix for each associated component; S302, predicting the status of associated components based on the Markov chain; The BP neural network prediction model construction step includes the following steps: S201, constructing a BP neural network prediction model based on vibration conditions, temperature changes, and current stability elements; S202, establishing a sample data set based on vibration conditions, temperature changes, current stability, and fault types; S203, using a normalization formula to normalize the sample data set values ​​so that the range is between 0 and 1; S204, inputting the sample data set into the constructed BP neural network prediction model, and outputting fault type data; S205, judging the calculation error based on the fault type data, and adjusting the weights from the hidden layer to the output layer and from the input layer to the hidden layer of the BP neural network prediction model; S206. Repeat steps S204-S205 until the error meets the set value.

2. The cluster server intelligent fault prediction method according to claim 1, characterized in that: The method of judging and calculating the error based on the fault type data and adjusting the weights from the hidden layer to the output layer and from the input layer to the hidden layer of the BP neural network prediction model includes: calculating the error of the model using the least squares method and updating the weights from back to front using the gradient descent method.

3. The cluster server intelligent fault prediction method according to claim 1, wherein: The step S301, generating a state transition probability distribution matrix of each associated component, includes the following steps: S3011. Select a historical period width T and obtain the status of the main component and the status of its associated components in each unit time period; S3012. Set all predicted states of the main component and all states of the associated components, divide them according to the states of the main component, calculate the frequency of data state transitions of the associated components in adjacent time periods when the main component is in each state, and obtain the predicted state probability distribution of the main component and the state transition probability distribution of the associated components.

4. A cluster server intelligent fault prediction system, characterized in that: The system includes: an equipment monitoring terminal 100, an equipment status analysis module 200 and an equipment status prediction module 300; The device monitoring terminal 100 is used to obtain real-time information about the device, wherein the real-time information about the device includes vibration conditions, temperature changes, and current stability data; The device status analysis module 200 is used to analyze the device's wear status data based on the device's real-time information, and obtain device status data based on the wear status data of the device; The device status prediction module 300 is used to predict device failures based on device status data; The device status analysis module 200 is further used to: analyze the loss status data of the device through the BP neural network prediction model to obtain the device status data; The device state prediction module 300 is further configured to: S301, generate a state transition probability distribution matrix for each associated component; S302, perform state prediction on the associated components based on a Markov chain; The BP neural network prediction model construction step includes the following steps: S201, constructing a BP neural network prediction model based on vibration conditions, temperature changes, and current stability elements; S202, establishing a sample data set based on vibration conditions, temperature changes, current stability, and fault types; S203, using a normalization formula to normalize the sample data set values ​​so that the range is between 0 and 1; S204, inputting the sample data set into the constructed BP neural network prediction model, and outputting fault type data; S205, judging the calculation error based on the fault type data, and adjusting the weights from the hidden layer to the output layer and from the input layer to the hidden layer of the BP neural network prediction model; S206. Repeat steps S204-S205 until the error meets the set value.

5. A terminal comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the cluster server intelligent fault prediction method according to any one of claims 1 to 3 when loading and executing the computer program.

6. A storage medium storing a computer program, wherein when the computer program is loaded and executed by a processor, the computer program implements the steps of the cluster server intelligent fault prediction method according to any one of claims 1 to 3.

Images