Fault diagnosis method and device, equipment, storage medium and product

Abstract

The invention discloses a fault diagnosis method and device, equipment, a storage medium and a product, and the method comprises the steps: extracting target data corresponding to a to-be-diagnosed variable for the to-be-diagnosed variable; inputting target data into the trained target quantile regression model to obtain a quantile prediction value output by the model; and comparing the quantile prediction value with the actual observation value of the to-be-diagnosed variable, and determining a fault diagnosis result of the to-be-diagnosed variable.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fault diagnosis, and in particular to a fault diagnosis method, device, equipment, storage medium and product. BACKGROUND

[0002] With the continuous accumulation of operation data in industrial systems, how to accurately identify the degradation trend of key variables under noise interference has become a key problem in intelligent operation and maintenance.

[0003] Traditional methods have many limitations in trend identification. First, they often rely on static modeling results for state judgment, lacking the ability to continuously track the evolution process of time series. Second, such methods usually construct models for a single trend pattern, making it difficult to accommodate diverse trend patterns such as slow degradation, periodic fluctuation or nonlinear growth, resulting in insufficient trend expression ability. Third, in terms of model response mechanism, traditional methods do not fully consider the driving effect of recent data on trend changes, resulting in weak perception of sudden turning points or local deviations, affecting the timely identification and early warning effect of abnormal trends. SUMMARY

[0004] The present application provides a fault diagnosis method, device, equipment, storage medium and product to improve the effect of fault diagnosis.

[0005] According to one aspect of the present application, a fault diagnosis method is provided, comprising:

[0006] For the variable to be diagnosed, the target data corresponding to the variable to be diagnosed is extracted;

[0007] The target data is input into a trained target quantile regression model to obtain a quantile prediction value output by the model;

[0008] The quantile prediction value is compared with the actual observation value of the variable to be diagnosed to determine the fault diagnosis result of the variable to be diagnosed.

[0009] Further, the target data includes real-time observation data of related variables of the variable to be diagnosed.

[0010] Further, the training method of the target quantile regression model comprises:

[0011] Extracting historical operation data in a healthy state as training data; wherein the historical operation data includes historical data of the variable to be diagnosed and historical data of related variables of the variable to be diagnosed;

[0012] According to a set quantile point, a quantile regression model for the variable to be diagnosed is established;

[0013] inputting the training data into the quantile regression model, determining a loss function value according to a model output result and adjusting a model parameter until the loss function value meets a set requirement.

[0014] Further, the historical running data in a healthy state is extracted as the training data, including:

[0015] The healthy state data in the historical running data is acquired.

[0016] The healthy state data is verified according to a set sliding window and a set index, and if the verification is passed, the healthy state data is processed.

[0017] The healthy state data after the data processing is taken as the training data.

[0018] Further, the quantile prediction value is compared with an actual observation value of the to-be-diagnosed variable to determine a fault diagnosis result of the to-be-diagnosed variable, including:

[0019] A normal range corresponding to the quantile prediction value is determined.

[0020] If the actual observation value of the to-be-diagnosed variable belongs to the normal range, it is determined that the fault diagnosis result is that the actual observation value of the to-be-diagnosed variable is normal, otherwise, it is determined that the fault diagnosis result is that the actual observation value of the to-be-diagnosed variable is an abnormal data point.

[0021] Further, after it is determined that the fault diagnosis result is that the actual observation value of the to-be-diagnosed variable is an abnormal data point, the method further includes:

[0022] The number of the abnormal data points in a set monitoring time length is counted.

[0023] The counting result is analyzed in combination with a confidence test algorithm.

[0024] According to another aspect of the present application, a fault diagnosis device is provided, including:

[0025] A target data extraction module is configured to extract target data corresponding to a to-be-diagnosed variable for the to-be-diagnosed variable.

[0026] A quantile prediction value output module is configured to input the target data into a trained target quantile regression model to obtain a quantile prediction value output by the model.

[0027] A fault diagnosis result determination module is configured to compare the quantile prediction value with an actual observation value of the to-be-diagnosed variable to determine a fault diagnosis result of the to-be-diagnosed variable.

[0028] Optionally, the target data comprises real-time observation data of a related variable of the to-be-diagnosed variable.

[0029] Optionally, the device further comprises a model training module configured to:

[0030] extract historical operation data in a healthy state as training data, wherein the historical operation data comprises historical data of the to-be-diagnosed variable and historical data of a related variable of the to-be-diagnosed variable;

[0031] establish a quantile regression model for the to-be-diagnosed variable according to a set quantile point;

[0032] input the training data into the quantile regression model, determine a loss function value according to a model output result, and adjust model parameters until the loss function value meets a set requirement.

[0033] Optionally, the model training module is further configured to:

[0034] obtain healthy state data in the historical operation data;

[0035] verify the healthy state data according to a set sliding window and a set index, and perform data processing on the healthy state data if the verification is passed;

[0036] use the healthy state data after data processing as the training data.

[0037] Optionally, the fault diagnosis result determination module is further configured to:

[0038] determine a normal range corresponding to the quantile prediction value;

[0039] if the actual observation value of the to-be-diagnosed variable belongs to the normal range, determine that the fault diagnosis result is that the actual observation value of the to-be-diagnosed variable is normal, and otherwise, determine that the fault diagnosis result is that the actual observation value of the to-be-diagnosed variable is an abnormal data point.

[0040] Optionally, the device is further configured to:

[0041] count a number of the abnormal data points within a set monitoring time length;

[0042] analyze the statistical result in combination with a confidence test algorithm.

[0043] According to another aspect of the present application, an electronic device is provided, which comprises:

[0044] at least one processor; and

[0045] a memory in communication connection with the at least one processor; wherein

[0046] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to perform the fault diagnosis method according to any one of the embodiments of the present application.

[0047] According to another aspect of the present application, there is provided a computer readable storage medium storing computer instructions for enabling a processor to perform the fault diagnosis method according to any one of the embodiments of the present application when executed by the processor.

[0048] According to another aspect of the present application, there is provided a computer program product comprising computer program / instructions for enabling a processor to perform the steps of the fault diagnosis method according to any one of the embodiments of the present application when executed by the processor.

[0049] The fault diagnosis method disclosed by the present application firstly acquires an image sequence, firstly extracts target data corresponding to a to-be-diagnosed variable for the to-be-diagnosed variable; then inputs the target data into a trained target quantile regression model to obtain quantile prediction values output by the model; and finally compares the quantile prediction values with actual observation values of the to-be-diagnosed variable to determine a fault diagnosis result of the to-be-diagnosed variable. The fault diagnosis method disclosed by the present application can improve modeling accuracy and early warning capability of the to-be-diagnosed parameter by using a quantile regression algorithm for data analysis, thereby providing clear decision basis for operation and maintenance personnel, and optimizing maintenance plans and resource scheduling arrangements.

[0050] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0052] Figure 1 is a flowchart of a fault diagnosis method according to an embodiment of the present application;

[0053] Figure 2 is a structural schematic diagram of a fault diagnosis device according to an embodiment of the present application;

[0054] Figure 3Fig. 1 is a structural schematic diagram of an electronic device for implementing the fault diagnosis method of Embodiment Three of the present application. DETAILED DESCRIPTION

[0055] In order to make the personnel in the technical field better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all the other embodiments obtained by the personnel in the field without creative labor should belong to the protection scope of the present application.

[0056] It should be noted that the terms "first", "second" and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to the process, method, product or device.

[0057] Embodiment One

[0058] Figure 1 Fig. 1 is a flowchart of a fault diagnosis method provided for Embodiment One of the present application. The present embodiment can be applied to the case of diagnosing parameters in a production process. The method can be executed by a fault diagnosis device, which can be realized in the form of hardware and / or software, and can be configured in an electronic device. As shown in Fig. 1, the method comprises the following steps. Figure 1

[0059] S110, for a to-be-diagnosed variable, extracting target data corresponding to the to-be-diagnosed variable.

[0060] In the present embodiment, the to-be-diagnosed variable is a key variable that currently needs to be diagnosed.

[0061] Preferably, the target data includes real-time observation data of a related variable of the to-be-diagnosed variable.

[0062] In the present embodiment, for the fault diagnosis of the to-be-diagnosed variable, real-time observation data of a related variable corresponding to the to-be-diagnosed variable can be extracted as target data after the to-be-diagnosed variable is determined.

[0063] ​S120, input the target data into the trained target quantile regression model to obtain a quantile prediction value output by the model.

[0064] The quantile regression model is a non-parametric method of regression analysis, which reveals more comprehensive distribution characteristics between variables by estimating the conditional relationship between dependent variables at different quantile points (such as median, quartile) and independent variables.

[0065] In this embodiment, applying the quantile regression model to fault diagnosis can capture the characteristics of how the variables to be diagnosed (such as equipment performance indicators) change with related variables (such as sensor readings, working condition parameters) at different quantile points (especially the tail), thereby identifying abnormal behavior beyond the normal fluctuation range.

[0066] Specifically, the expression of the quantile regression model is:

[0067]

[0068] where τ is a set quantile point, X is an input feature vector, p is the number of features, B j (X) is a basis function with X as the independent variable, β j is a regression coefficient.

[0069] After inputting the target data into the trained target quantile regression model, the model can output quantile prediction values according to the target data, i.e., the trained quantile regression model can predict the values of the variables to be diagnosed at different quantile points, and these prediction values collectively constitute a dynamic normal value range interval that depends on the working condition (X).

[0070] Optionally, the training method of the target quantile regression model can be: extracting historical running data in a healthy state as training data; wherein the historical running data includes historical data of the variables to be diagnosed and historical data of related variables of the variables to be diagnosed; establishing a quantile regression model for the variables to be diagnosed according to a set quantile point; inputting the training data into the quantile regression model, determining a loss function value according to the model output result and adjusting the model parameters until the loss function value meets the set requirement.

[0071] Specifically, the historical running data in a healthy state is historical running data extracted for the variables to be diagnosed and their related variables under healthy working conditions (i.e., without faults). Collecting historical running data in a healthy state as training data, and then one or more quantile regression models can be trained for the variables to be diagnosed by selecting different quantile points. During the model training process, the loss function value corresponding to each model output result can be calculated according to the loss function of the model, and the model parameters can be adjusted to make the loss function value meet the set requirement (e.g., less than a set threshold), so that the training can be stopped.

[0072] Preferably, the loss function can be:

[0073]

[0074] where Y t and Y ref are the model prediction value and the actual value respectively, and σ is the standard deviation.

[0075] Further, the method of extracting historical running data in a healthy state as training data can be: obtaining healthy state data in the historical running data; verifying the healthy state data according to a set sliding window and a set index, and if the verification is passed, performing data processing on the healthy state data; and taking the healthy state data after the data processing as the training data.

[0076] Specifically, after obtaining the historical running data, the healthy state data therein can be preprocessed, and the healthy state data after the processing can be taken as the training data. In the processing process, in order to ensure the stability of the reference sample, the sliding window statistical characteristics can be used to perform secondary verification on the healthy reference data. That is, for each batch of candidate healthy state data, the mean, variance, skewness, kurtosis and the like in the set sliding window are defined as the verification vector, and a constraint interval is set for screening, so as to ensure that the volatility of the data is within the expected range. After the verification is passed, the data can also be processed, such as smoothing processing, for enhancing the anti-interference ability of the trend recognition algorithm to short-time disturbance, transient fluctuation and local anomaly. Let the variable to be diagnosed be y t , and the corresponding d related variables be The smoothing formula is:

[0077]

[0078] wherein, is the result after smoothing; y t-i represents the observation value of the i-th time step forward from the current time t; v i is a smoothing weight coefficient, satisfying v i >0, and ∑v i =1, which can be set to be linearly decreasing or exponentially decaying according to the requirement; and n is the size of the sliding window. Similarly:

[0079]

[0080] wherein, is the smoothing processing result of the k-th related variable, and k=1,2,…d.

[0081] Furthermore, when the historical data of the variable to be diagnosed shows significant fluctuations, the processed data can be preferentially used for modeling, and a weighted residual term can be embedded in the loss function to enhance the stability of trend extraction in quantile modeling under severe disturbances.

[0082]

[0083] in, Q is the sequence of variables to be diagnosed after data processing. y ((τ|X t )) represents the model output corresponding to the observed value of the relevant variable at time t, which is the predicted τth quantile.

[0084] S130. Compare the predicted values ​​of the digits with the actual observed values ​​of the variable to be diagnosed to determine the fault diagnosis result of the variable to be diagnosed.

[0085] In this embodiment, after obtaining the quantile prediction values ​​output by the model, an adaptive threshold range can be constructed by combining the fluctuation characteristics of the data itself. Based on the actual observed values ​​of the variable to be diagnosed, the fault diagnosis result of the variable to be diagnosed can be determined.

[0086] Optionally, the method for determining the fault diagnosis result of the variable to be diagnosed by comparing the predicted quantile values ​​with the actual observed values ​​of the variable to be diagnosed can be: determining the normal range corresponding to the predicted quantile values; if the actual observed values ​​of the variable to be diagnosed are within the normal range, then the fault diagnosis result is determined to be that the actual observed values ​​of the variable to be diagnosed are normal; otherwise, the fault diagnosis result is determined to be that the actual observed values ​​of the variable to be diagnosed are abnormal data points.

[0087] Specifically, for the same variable to be diagnosed, setting different quantiles can cause the model to output different quantile prediction values. For example, for two quantiles τ... low and τ high The predicted quantile values ​​obtained are Q low and Q high Then the predicted normal range can be [Q]. low Q high If the actual observed value of the variable to be diagnosed is Y actual If Y actual low Or Y actual Q high This indicates that under the current operating conditions, the actual value of the variable to be diagnosed has deviated significantly from its historical normal behavior pattern, which may indicate a fault or abnormal state.

[0088] ​Further, after determining the target data corresponding to the to-be-diagnosed variable as an abnormal data point, the following can also be performed: counting the number of abnormal data points within a set monitoring time length; and analyzing the counting result in combination with a confidence test algorithm.

[0089] Specifically, a statistical determination logic based on binomial distribution confidence test can be used. Assuming that the total number of data points of a batch is m, and q points are abnormal data points, the determination condition can be expressed as:

[0090]

[0091] wherein l is an abnormal reference probability set by a model; z α / 2 is a critical value under a standard normal distribution (set according to a confidence level); and ml is a theoretically expected number of abnormal points.

[0092] If the determination condition is met, it is determined that the data trend of the to-be-diagnosed variable is abnormal.

[0093] The fault diagnosis method disclosed by the present application first acquires an image sequence, first extracts target data corresponding to a to-be-diagnosed variable for the to-be-diagnosed variable, then inputs the target data into a trained target quantile regression model to obtain a quantile prediction value output by the model, and finally compares the quantile prediction value with an actual observation value of the to-be-diagnosed variable to determine a fault diagnosis result of the to-be-diagnosed variable. The fault diagnosis method disclosed by the present application can flexibly depict the asymmetric trend change of a key parameter under different working conditions by using a quantile regression algorithm for data analysis, enhances the recognition ability of the model to a nonlinear degradation mode, can improve the modeling accuracy and early warning capability of the to-be-diagnosed parameter, and thus provides clear decision basis for an operation and maintenance personnel, optimizes a maintenance plan and resource scheduling arrangement. In addition, a dynamic determination threshold strategy based on confidence test can reduce the false alarm and missed alarm risks.

[0094] Embodiment Two

[0095] Figure 2 A structural schematic diagram of a fault diagnosis device provided by Embodiment Two of the present application is shown in FIG. 3, which includes a target data extraction module 310, a quantile prediction value output module 320, and a fault diagnosis result determination module 330. Figure 2

[0096] The target data extraction module 310 is configured to extract target data corresponding to a to-be-diagnosed variable for the to-be-diagnosed variable.

[0097] The quantile prediction value output module 320 is configured to input the target data into a trained target quantile regression model to obtain a quantile prediction value output by the model.

[0098] ​The fault diagnosis result determination module 330 is configured to compare the quantile prediction value with the actual observation value of the to-be-diagnosed variable, and determine the fault diagnosis result of the to-be-diagnosed variable.

[0099] Optionally, the target data comprises real-time observation data of a related variable of the to-be-diagnosed variable.

[0100] Optionally, the device further comprises a model training module configured to:

[0101] extract historical operation data in a healthy state as training data, wherein the historical operation data comprises historical data of the to-be-diagnosed variable and historical data of a related variable of the to-be-diagnosed variable; establish a quantile regression model for the to-be-diagnosed variable according to a set quantile point; input the training data into the quantile regression model, determine a loss function value according to a model output result, and adjust model parameters until the loss function value meets a set requirement.

[0102] Optionally, the model training module is further configured to:

[0103] acquire healthy state data in the historical operation data; verify the healthy state data according to a set sliding window and a set index, and perform data processing on the healthy state data if the verification is passed; and use the healthy state data after the data processing as the training data.

[0104] Optionally, the fault diagnosis result determination module 330 is further configured to:

[0105] determine a normal range corresponding to the quantile prediction value; if the actual observation value of the to-be-diagnosed variable belongs to the normal range, determine that the fault diagnosis result is that the actual observation value of the to-be-diagnosed variable is normal; otherwise, determine that the fault diagnosis result is that the actual observation value of the to-be-diagnosed variable is an abnormal data point.

[0106] Optionally, the device is further configured to:

[0107] count the number of abnormal data points in a set monitoring duration; and analyze the statistical result in combination with a confidence test algorithm.

[0108] The fault diagnosis device provided in the embodiments of the present application can execute the fault diagnosis method provided in any of the embodiments of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0109] Embodiment three

[0110] Figure 3A structural diagram of an electronic device 10 that can be used to implement embodiments of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smartphones, wearable devices (e.g., headsets, glasses, watches, etc.), and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present application described and / or claimed in this document.

[0111] As shown in Figure 3 The electronic device 10 includes at least one processor 11, and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., communicatively connected to the at least one processor 11, where the memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer programs stored in the read-only memory (ROM) 12 or loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the RAM 13. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0112] Various components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc., an output unit 17, such as various types of displays, speakers, etc., a storage unit 18, such as a magnetic disk, an optical disk, etc., and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0113] The processor 11 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs various methods and processes described above, such as the fault diagnosis method.

[0114] In some embodiments, the fault diagnosis method can be implemented as a computer program tangibly embodied in a computer readable storage medium, e.g., storage unit 18. In some embodiments, parts or all of the computer program can be loaded and / or installed onto electronic device 10 via, e.g., ROM 12 and / or communication unit 19. When the computer program is loaded onto RAM 13 and executed by processor 11 as described above, one or more of the steps of the fault diagnosis described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the fault diagnosis method by other means, e.g., with the aid of firmware.

[0115] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a programmable logic device (PLD), a computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0116] Computer programs used to implement the methods of the present application can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the computer program, when executed by the processor of the machine, implements the functions / acts specified in the flowcharts and / or block diagrams. The computer program can be executed entirely on a machine, partially on a machine, partially on a machine as a stand-alone software package, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0117] In the context of the present application, a computer-readable storage medium can be a tangible medium that can contain or store computer programs for use by or in connection with an instruction execution system, apparatus, or device. Computer-readable storage media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of a machine-readable storage medium will include one or more lines of a program of instructions in a transitory signal, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0118] To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0119] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0120] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system, to solve the defects of large management difficulty and weak business scalability in traditional physical host and VPS service.

Claims

1. A fault diagnosis method, characterized in that, include: For the variable to be diagnosed, extract the target data corresponding to the variable to be diagnosed; The target data is input into the trained target quantile regression model to obtain the quantile prediction value output by the model; By comparing the predicted quantile values ​​with the actual observed values ​​of the variable to be diagnosed, the fault diagnosis result of the variable to be diagnosed is determined.

2. The method according to claim 1, characterized in that, The target data includes: real-time observation data of the relevant variables of the variable to be diagnosed.

3. The method according to claim 1, characterized in that, The training method for the target quantile regression model includes: Historical operational data under healthy conditions is extracted as training data; wherein, the historical operational data includes historical data of the variable to be diagnosed and historical data of related variables of the variable to be diagnosed; Based on the set quantiles, a quantile regression model is established for the variable to be diagnosed; The training data is input into the quantile regression model, and the loss function value is determined based on the model output. The model parameters are then adjusted until the loss function value meets the set requirements.

4. The method according to claim 3, characterized in that, Historical operational data under healthy conditions is extracted as training data, including: Obtain health status data from the historical operational data; The health status data is verified according to the set sliding window and set indicators. If the verification is successful, the health status data is processed. The processed health status data is used as the training data.

5. The method according to claim 1, characterized in that, By comparing the predicted quantile values ​​with the actual observed values ​​of the variable to be diagnosed, the fault diagnosis result of the variable to be diagnosed is determined, including: Determine the normal range corresponding to the predicted quantile values; If the actual observed value of the variable to be diagnosed falls within the normal range, the fault diagnosis result is determined to be that the actual observed value of the variable to be diagnosed is normal; otherwise, the fault diagnosis result is determined to be that the actual observed value of the variable to be diagnosed is an abnormal data point.

6. The method according to claim 5, characterized in that, After determining that the fault diagnosis result is that the actual observed value of the variable to be diagnosed is an outlier data point, the method further includes: Count the number of abnormal data points within the set monitoring period; The statistical results were analyzed using a confidence test algorithm.

7. A fault diagnosis device, characterized in that, include: The target data extraction module is used to extract target data corresponding to the variable to be diagnosed. The quantile prediction output module is used to input the target data into the trained target quantile regression model to obtain the quantile prediction value output by the model. The fault diagnosis result determination module is used to compare the predicted quantile values ​​with the actual observed values ​​of the variable to be diagnosed, and determine the fault diagnosis result of the variable to be diagnosed.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the fault diagnosis method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the fault diagnosis method according to any one of claims 1-6.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the fault diagnosis method as described in any one of claims 1-6.

Images