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

Abstract

The invention relates to a fault diagnosis method and device, equipment, a storage medium and a program product. The method comprises the steps that historical process data of a target system is collected, a fault detection model and a fault subspace library are constructed based on the historical process data, and the fault subspace library comprises at least one candidate fault type; calculating a historical error statistical value of the target system based on the fault detection model, and calculating a fault threshold based on the historical error statistical value; obtaining a to-be-tested sample, and calculating a current error statistical value of the to-be-tested sample; and under the condition that the current error statistical value is not smaller than the fault threshold value, sequentially reconstructing the to-be-detected sample based on the candidate fault types, and under the condition that the reconstruction error statistical value of the reconstructed sample is smaller than the fault threshold value, determining the candidate fault type currently used for reconstruction as the fault type of the to-be-detected sample. The method can be applied to a target system with a dynamic characteristic and a nonlinear characteristic, and the application range is expanded.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and particularly to a fault diagnosis method, apparatus, device, storage medium, and program product. Background Art

[0002] With the development and progress of control theory, computer technology, etc., the civil and military industrial systems have been continuously expanding, the system complexity has been increasing, and the development of sensor technology and computer technology has also made it possible to collect and store a large amount of process data. Based on this, data-driven fault diagnosis methods have been widely applied in civil and military industrial systems.

[0003] However, some civil and military industrial systems are affected by time operation, and the process data collected therefrom has characteristics such as dynamic characteristics and non-linearity, while traditional data-driven methods are difficult to be applied to the fault diagnosis of civil and military industrial systems with the above characteristics. Summary of the Invention

[0004] Based on this, it is necessary to provide a fault diagnosis method, apparatus, device, storage medium, and program product that can be applied to dynamic data and non-linear systems for the above technical problems.

[0005] In a first aspect, the present application provides a fault diagnosis method, including:

[0006] Collect historical process data of a target system, and construct a fault detection model and a fault subspace library based on the historical process data, where the historical process data includes normal operation data and fault data, and the fault subspace library includes at least one candidate fault type;

[0007] Calculate a historical error statistic value of the target system based on the fault detection model, and calculate a fault threshold based on the historical error statistic value;

[0008] Obtain a sample to be measured, and calculate a current error statistic value of the sample to be measured;

[0009] In the case where the current error statistic value is not less than the fault threshold, reconstruct the sample to be measured in sequence based on the candidate fault types, and in the case where the reconstruction error statistic value of the reconstructed sample is less than the fault threshold, determine the candidate fault type currently used for reconstruction as the fault type of the sample to be measured.

[0010] In one embodiment, the constructing a fault subspace library based on the historical process data includes:

[0011] Classify the fault data in the historical process data to obtain multiple groups of different types of fault data, and determine the classified fault types as candidate fault types;

[0012] Based on the fault subspace extraction method, the fault features of each type of fault data are extracted, and the fault features are associated with the candidate fault types to obtain a corresponding fault subspace library.

[0013] In one embodiment, the building of a fault detection model based on the historical process data includes:

[0014] The fault detection model is constructed based on a canonical variable analysis method, wherein the input data of the fault detection model is the normal operation data, and the output of the fault detection model is a first canonical variable corresponding to the normal operation data.

[0015] In one embodiment, the calculating the historical error statistics of the target system based on the fault detection model, and the calculating the fault threshold based on the historical error statistics, includes:

[0016] Calculating a squared prediction error statistic of the historical process data based on a residual between the historical process data and the first standard variable, and determining the squared prediction error statistic as a historical error statistic;

[0017] A control limit of the historical error statistic is calculated based on a sampling distribution, and the control limit is determined as the fault threshold.

[0018] In one embodiment, the obtaining of the sample to be tested and calculating the current error statistic of the sample to be tested includes:

[0019] Preprocessing the sample to be tested, and calculating a second normalized variable of the preprocessed sample to be tested;

[0020] Based on the residual between the sample to be tested and the second standard variable, a square prediction error statistic of the sample to be tested is calculated, and the square prediction error statistic is determined as the current error statistic.

[0021] In one embodiment, the method further comprises:

[0022] When the current error statistic is less than the fault threshold, determining the state of the sample to be tested as no fault;

[0023] After reconstructing the sample to be tested based on all candidate fault types in the fault subspace library, the fault manifestation of the sample to be tested is recorded when the statistical values ​​of reconstruction errors of the reconstructed samples are not less than the fault threshold.

[0024] In a second aspect, the present application also provides a fault diagnosis device, comprising:

[0025] A building module for collecting historical process data of a target system, constructing a fault detection model and a fault subspace library based on the historical process data, wherein the historical process data includes normal operation data and fault data, and the fault subspace library includes at least one candidate fault type;

[0026] A first calculation module for calculating a historical error statistic value of the target system based on the fault detection model and calculating a fault threshold based on the historical error statistic value;

[0027] A second calculation module for obtaining a sample to be tested and calculating a current error statistic value of the sample to be tested;

[0028] A reconstruction module for, when the current error statistic value is not less than the fault threshold, reconstructing the sample to be tested in sequence based on the candidate fault types, and when the reconstruction error statistic value of the reconstructed sample is less than the fault threshold, determining the candidate fault type currently used for reconstruction as the fault type of the sample to be tested.

[0029] In a third aspect, the present application further provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0030] Collect historical process data of a target system, construct a fault detection model and a fault subspace library based on the historical process data, wherein the historical process data includes normal operation data and fault data, and the fault subspace library includes at least one candidate fault type;

[0031] Calculate a historical error statistic value of the target system based on the fault detection model and calculate a fault threshold based on the historical error statistic value;

[0032] Obtain a sample to be tested and calculate a current error statistic value of the sample to be tested;

[0033] When the current error statistic value is not less than the fault threshold, reconstruct the sample to be tested in sequence based on the candidate fault types, and when the reconstruction error statistic value of the reconstructed sample is less than the fault threshold, determine the candidate fault type currently used for reconstruction as the fault type of the sample to be tested.

[0034] In a fourth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0035] Collect historical process data of the target system, and construct a fault detection model and a fault subspace library based on the historical process data, where the historical process data includes normal operation data and fault data, and the fault subspace library includes at least one candidate fault type;

[0036] Calculate the historical error statistical value of the target system based on the fault detection model, and calculate the fault threshold based on the historical error statistical value;

[0037] Obtain a sample to be measured, and calculate the current error statistical value of the sample to be measured;

[0038] When the current error statistical value is not less than the fault threshold, reconstruct the sample to be measured in sequence based on the candidate fault types, and when the reconstruction error statistical value of the reconstructed sample is less than the fault threshold, determine the candidate fault type currently used for reconstruction as the fault type of the sample to be measured.

[0039] In a fifth aspect, the present application further provides a computer program product, including a computer program, which when executed by a processor implements the following steps:

[0040] Collect historical process data of the target system, and construct a fault detection model and a fault subspace library based on the historical process data, where the historical process data includes normal operation data and fault data, and the fault subspace library includes at least one candidate fault type;

[0041] Calculate the historical error statistical value of the target system based on the fault detection model, and calculate the fault threshold based on the historical error statistical value;

[0042] Obtain a sample to be measured, and calculate the current error statistical value of the sample to be measured;

[0043] When the current error statistical value is not less than the fault threshold, reconstruct the sample to be measured in sequence based on the candidate fault types, and when the reconstruction error statistical value of the reconstructed sample is less than the fault threshold, determine the candidate fault type currently used for reconstruction as the fault type of the sample to be measured.

[0044] The above-mentioned fault diagnosis method, device, equipment, storage medium and program product first obtain relevant process data, and then introduce the fault reconstruction technology into the canonical variable analysis method to extract the fault direction of the process data, thereby eliminating abnormal states and restoring normal states, providing a complete canonical variable analysis fault diagnosis solution based on fault reconstruction, and providing support for the fault diagnosis of target systems with dynamic characteristics. It can be applied to various types of big data systems with dynamic and non-linear characteristics, expanding the application scope. Description of the Drawings

[0045] To more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments of the present application or related technologies. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0046] Figure 1 It is a schematic flowchart of a fault diagnosis method in one embodiment;

[0047] Figure 2 It is a schematic flowchart of a fault diagnosis method in another embodiment;

[0048] Figure 3 It is a schematic flowchart of a fault diagnosis method in yet another embodiment;

[0049] Figure 4 It is a schematic diagram of the experimental results of a fault diagnosis method in one embodiment;

[0050] Figure 5 It is a schematic diagram of the experimental results of a fault diagnosis method in another embodiment;

[0051] Figure 6 It is a schematic diagram of the experimental results of a fault diagnosis method in yet another embodiment;

[0052] Figure 7 It is a structural block diagram of a fault diagnosis device in one embodiment;

[0053] Figure 8 It is an internal structure diagram of a computer device in one embodiment. Detailed implementation manners

[0054] In order to make the purpose, technical solutions and advantages of the present application clearer, the following further details the present application in conjunction with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0055] With the development and progress of control theory, computer technology, etc., the civil and military industrial systems are constantly growing, and the system complexity is constantly increasing. Once a fault occurs and cannot be detected and diagnosed in time, it will cause unimaginable harm and losses. Since 1967, after half a century of development of fault diagnosis technology, there have been many methods and classifications. The common fault diagnosis classifications are mainly qualitative and quantitative.

[0056] Among them, the qualitative methods mainly include: (1) graph theory method; (2) expert system; (3) qualitative simulation. The quantitative methods mainly include: (1) method based on analytical model; (2) data-driven method. In industrial systems such as military and civilian industries, the system complexity is increasing proportionally with the increase of the system. At the same time, the development of sensor technology and computer technology also makes it possible to collect and store a large amount of process data.

[0057] In this background environment, it has become increasingly difficult to diagnose faults due to system modeling problems for the method based on analytical model. The data-driven methods mainly include: 1) machine learning methods such as neural networks and support vector machines; 2) multivariate statistical methods such as principal component analysis, canonical variate analysis (CVA), and independent component analysis; 3) signal processing methods such as wavelet transform; 4) information fusion, etc. Multivariate statistical methods such as CVA are one of the most commonly used methods for fault diagnosis in data-driven. Algorithms such as principal component analysis and independent component analysis are often used for fault diagnosis on the premise that the process data is in a steady state without disturbance, while actual military and civilian industrial systems are often affected by time operation, and the process data collected from them has dynamic characteristics. Therefore, the CVA fault diagnosis method specifically designed to solve the dynamic characteristics of process data has emerged.

[0058] The most commonly used data-driven fault diagnosis methods are contribution graph and fault reconstruction. Although the contribution graph has the advantages of simple method and no need for prior knowledge, it also has three fatal disadvantages: (1) Nonlinearity is a common characteristic of military and civilian industrial systems, but it cannot be applied to systems with nonlinear characteristics; (2) The fault diagnosis method based on the contribution graph is carried out under the assumption that the adopted statistic has a good fault detection effect; (3) The fuzzy effect between process variables may lead to chaotic fault diagnosis results.

[0059] Based on this, the present application provides a fault diagnosis method based on fault reconstruction, as Figure 1 shown. In this embodiment, taking the application of this method to the terminal as an example, it can be understood that this method can also be applied to the server, and can also be applied to a system including the terminal and the server, and is realized through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0060] Step 101, collect the historical process data of the target system, and construct a fault detection model and a fault subspace library based on the historical process data, where the historical process data includes normal operation data and fault data, and the fault subspace library includes at least one candidate fault type;

[0061] Step 102, calculate the historical error statistic value of the target system based on the fault detection model, and calculate the fault threshold based on the historical error statistic value;

[0062] Step 103: Obtain a sample to be measured and calculate the current error statistic value of the sample to be measured.

[0063] Step 104: When the current error statistic value is not less than the fault threshold, reconstruct the sample to be measured based on the candidate fault types in sequence. When the reconstructed error statistic value of the reconstructed sample is less than the fault threshold, determine the candidate fault type currently used for reconstruction as the fault type of the sample to be measured.

[0064] Among them, the target system refers to various types of systems with a large amount of process data and dynamic data characteristics, including civil and military industrial systems, etc., and is not specifically limited here. Please refer to Figure 2 , The historical process data refers to a large amount of process data collected from the dynamic multi-variable target system, including the normal operation data of the system during normal operation and the fault data under different fault modes. After collecting the process data, it is necessary to perform data integration to integrate the fault data under different fault modes into a fault data set. After completing the above work, data preprocessing operations such as removing singular samples are performed.

[0065] The historical error statistic value refers to the error statistic value obtained based on the historical process data and the fault detection model. The sample to be measured refers to the sample collected at the current moment during online fault diagnosis. The current error statistic value refers to the error statistic value corresponding to the sample at the current moment. The reconstruction process is used to determine the fault type when the sample to be measured has a fault.

[0066] Exemplarily, refer to Figure 2 , This embodiment is divided into an offline process and an online process. Among them, the processes of establishing a fault detection model, a fault subspace library, and calculating the fault threshold can be implemented offline; when applied to the fault diagnosis of a sample to be measured, it can be performed online. When performing fault diagnosis, the previously offline obtained fault detection model, fault threshold, and fault subspace library can be directly called.

[0067] For the above-mentioned fault diagnosis method, relevant process data is first obtained, and then the fault reconstruction technology is introduced into the canonical variable analysis method to extract the fault direction of the process data, thereby eliminating abnormal states and restoring the normal state. This embodiment provides a complete set of canonical variable analysis fault diagnosis solutions based on fault reconstruction, providing support for the fault diagnosis of target systems with dynamic characteristics. It can be applied to civil and military industrial systems with dynamic and non-linear characteristics, expanding the application scope.

[0068] In an exemplary embodiment, constructing the fault detection model based on the historical process data includes:

[0069] The fault detection model is constructed based on Canonical variate analysis (CVA). Among them, the input data of the fault detection model is the normal operation data, and the output of the fault detection model is the first canonical variate corresponding to the normal operation data.

[0070] Canonical variate analysis is a linear dimensionality reduction technique. By maximizing the correlation between two variable sets, it realizes the dimensionality reduction of high-dimensional data and obtains a set of canonical variates that can maximize the explanation of the information in the variable set. Using canonical variate analysis, the best prediction of future output can be achieved based on the past and current states of the system, so as to achieve the purpose of process identification. Among them, the modeling process of canonical variate analysis can be referred to:

[0071] Preprocess the historical process data to obtain the data matrix X. The variable state space model identified by CVA is:

[0072] Formula 1:

[0073] ;

[0074] In the formula, , A, and C are the measurement variable matrix, state matrix, and output matrix respectively. and respectively represent the system state and output at time in the historical process data. and are the modeling errors that usually follow non-Gaussian distributions.

[0075] To avoid the influence of values with different units, it is necessary to eliminate the dimensional effects of multivariate variables. Standardize the preprocessed data matrix X:

[0076] Formula 2:

[0077] ;

[0078] Among them, E(X) represents the expectation of X.

[0079] If the output vector is composed of variables, expand the output vector using past and future measurement values, and the past and future output vectors and can be obtained:

[0080] Formula 3:

[0081] ;

[0082] Obtain the Hankel matrices of past and future outputs and where the Hankel Matrix is a matrix in which the elements on each secondary diagonal are equal.

[0083] Formula 4:

[0084] ;

[0085] Formula 5:

[0086] ;

[0087] Formula 6:

[0088] ;

[0089] Formula 7:

[0090] ;

[0091] where and are the average values of and respectively, and and are the estimated values of and respectively. , is the number of samples.

[0092] and The covariance and cross-covariance matrices of

[0093] Formula 8:

[0094] ;

[0095] The objective of CVA can be obtained through the Hankel matrix scaled by singular value decomposition:

[0096] Formula 9:

[0097] ;

[0098] where , and .

[0099] where U is an orthogonal matrix of , called the left singular vector; V is an orthogonal matrix of The orthogonal matrix is called the right singular vector. Σ is a diagonal matrix with non - negative real numbers as diagonal elements, called singular values, usually arranged in descending order. The Hankel matrix can help identify the dynamic characteristics of the system. Through singular value decomposition, the Hankel matrix can perform data compression and dimensionality reduction.

[0100] So far, the first canonical variable can be obtained:

[0101] Equation 10:

[0102] ;

[0103] where represents the transformation matrix and , which transforms the past observation samples in the historical process data into the canonical variable space.

[0104] In one embodiment, calculating the historical error statistic value of the target system based on the fault detection model and calculating the fault threshold based on the historical error statistic value includes: calculating the squared prediction error statistic value of the historical process data based on the residual between the historical process data and the first canonical variable, and determining the squared prediction error statistic value as the historical error statistic value; calculating the control limit of the historical error statistic value based on the sampling distribution and determining the control limit as the fault threshold.

[0105] Specifically, calculating the SPE statistic value and its control limit, where the SPE statistic value is used to measure the difference between the observed value and the value predicted by the CVA model.

[0106] The SPE statistic value is calculated as follows:

[0107] Equation 11:

[0108] ;

[0109] where represents the residual (the difference between the actual observed value and the estimated value, that is, the residual of the first canonical variable). The distribution is used to calculate the control limit of the statistic. It should be noted that the SPE statistic value here is the SPE statistic value of each data in the historical process data. Therefore, the data volume of the SPE statistic value is equal to the data volume of the historical process data.

[0110] In an exemplary embodiment, constructing the fault subspace library based on the historical process data includes: classifying the fault data in the historical process data to obtain multiple groups of fault data of different types, and determining the classified fault types as candidate fault types; extracting the fault features of each type of fault data based on the fault subspace extraction method, and associating the fault features with the candidate fault types to obtain the corresponding fault subspace library.

[0111] Based on the fault subspace extraction method:

[0112] If represents the th fault sample in the historical process data, then refer to Equation 12:

[0113] ;

[0114] where represents the correction value of the canonical variable space, which satisfies the zero-mean condition and is therefore usually ignored. represents the estimated fault amplitude in the canonical variable space. Equation 12 is rewritten as Equation 13:

[0115] ;

[0116] In the formula, represents the sample obtained after being processed by Equation 12, represents the fault amplitude obtained after being processed by Equation 12. The average fault data matrix is calculated as follows:

[0117] Equation 14:

[0118] ;

[0119] Singular value decomposition :

[0120] Equation 15:

[0121] ;

[0122] where the diagonal matrix has non-zero singular values in descending order, and select . The dimension of the fault subspace is the minimum dimension that makes the reconstructed statistic within the control limits.

[0123] Suppose a system has 10 types of faults. Then, use the above fault subspace extraction method to extract the fault features of these 10 types of faults respectively and store them for use when diagnosing faults online.

[0124] In an exemplary embodiment, obtaining the sample to be measured and calculating the current error statistic value of the sample to be measured includes: preprocessing the sample to be measured and calculating the second canonical variable of the preprocessed sample to be measured; calculating the squared prediction error statistic value of the sample to be measured based on the residual between the sample to be measured and the second canonical variable, and determining the squared prediction error statistic value as the current error statistic value.

[0125] Please refer to Figure 3 , in an exemplary embodiment, the method further includes:

[0126] Step 301, when the current error statistic value is less than the fault threshold, determining the state of the sample to be measured as fault-free;

[0127] Step 302, after reconstructing the sample to be measured based on all candidate fault types in the fault subspace library, when the reconstruction error statistic values of the reconstructed samples are all not less than the fault threshold, recording the fault manifestation of the sample to be measured, and creating a new fault and fault characteristics of the corresponding type of the fault manifestation in the fault subspace library.

[0128] Collect the sample at the current moment , first perform standardization preprocessing:

[0129] Formula 16:

[0130] ;

[0131] Then calculate the second canonical variable of the sample at the current moment :

[0132] Formula 17:

[0133] ;

[0134] Wherein, represents the output vector extended by q past measurement values of the current sample, represents the average value of.

[0135] Formula 18:

[0136] ;

[0137] Wherein, represents the canonical variable of the current sample.

[0138] The statistic of the online sample is calculated as follows:

[0139] Formula 19:

[0140] ;

[0141] Among them, represents the residual of the sample to be measured and the second canonical variable.

[0142] If the statistic of the current sample is lower than the control limit, no failure has occurred at the current moment, and continue to monitor whether a failure occurs at the next moment. Otherwise, a failure has occurred at the current moment, and it is necessary to diagnose what kind of failure has occurred.

[0143] Suppose there are 10 failure modes. If the currently collected sample is detected as a failure, the failure data of the 10 failure modes are used to reconstruct the current failure sample in turn. If the monitoring statistic reconstructed by the second failure mode is lower than the control limit, it is considered that the second mode of failure has occurred. If the monitoring statistics reconstructed by the failure data of the 10 failure modes are not lower than the control limit after reconstructing the current failure sample, it is considered that an unknown failure has occurred, and the process engineer needs to extract the failure direction and improve the relevant failure subspace library. If the current failure is an unknown failure, create a new failure of this type in the failure subspace library for subsequent failure detection.

[0144] To better explain the technology of this embodiment and clearly and intuitively display the fault diagnosis effect of canonical variable analysis based on fault reconstruction, the TEP experimental results shown in Figure 4 , Figure 5 , Figure 6 are given.

[0145] In Figure 4 , Figure 5 , Figure 6 , the solid line represents the statistic, the dashed line represents the control limit, and the point set on the abscissa represents the reconstructed statistic. When the statistic is higher than the control limit, it is considered that a failure has occurred at the current moment, and it is necessary to reconstruct and diagnose the failure. Figures 4 to 6 The failures in

[0146] This embodiment has the following beneficial effects:

[0147] A set of fault diagnosis techniques based on canonical variable analysis of fault reconstruction is proposed for the fault diagnosis problem of process monitoring of dynamic multi-variable target systems.

[0148] Different from the fault diagnosis method of canonical variable analysis based on contribution graph, it can be applied to target systems with dynamic and nonlinear characteristics, effectively reducing the usage assumptions of CVA fault diagnosis technology and expanding the application scope.

[0149] The evaluation criteria are more perfect and the process is more systematic. The fault diagnosis technology of this application uses the process data of TEP as the test input, and mainly based on eliminating the assumptions that the adopted statistics have good fault detection effects and the fuzzy effects between process variables. It verifies its effectiveness and accuracy according to the digital results of TEP case studies. Compared with the fault diagnosis method of canonical variable analysis based on contribution graph, the evaluation criteria are more perfect and the process is more systematic.

[0150] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are sequentially shown according to the indications of the arrows, these steps do not necessarily need to be executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages do not necessarily need to be executed at the same moment, but can be executed at different moments. The execution order of these steps or stages does not necessarily need to be sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0151] Based on the same inventive concept, the embodiments of this application also provide a fault diagnosis device for implementing the above-mentioned fault diagnosis method. The implementation solutions provided by this device to solve problems are similar to the implementation solutions described in the above method. Therefore, the specific limitations in one or more embodiments of the following fault diagnosis devices can refer to the limitations on the fault diagnosis method in the above text, and will not be repeated here.

[0152] In an exemplary embodiment, as Figure 7 shown, a fault diagnosis device 600 is provided, including: a construction module 601, a first calculation module 602, a second calculation module 603, and a reconstruction module 604, where:

[0153] The construction module 601 is configured to collect historical process data of the target system, and construct a fault detection model and a fault subspace library based on the historical process data. Among them, the historical process data includes normal operation data and fault data, and the fault subspace library includes at least one candidate fault type;

[0154] The first calculation module 602 is configured to calculate the historical error statistical value of the target system based on the fault detection model, and calculate a fault threshold based on the historical error statistical value;

[0155] A second calculation module 603, configured to obtain a sample to be measured and calculate a current error statistic value of the sample to be measured;

[0156] A reconstruction module 604, configured to, when the current error statistic value is not less than the fault threshold, reconstruct the sample to be measured in sequence based on the candidate fault types, and when the reconstruction error statistic value of the reconstructed sample is less than the fault threshold, determine the candidate fault type currently used for reconstruction as the fault type of the sample to be measured.

[0157] The construction module 601 is further configured to classify the fault data in the historical process data to obtain multiple groups of fault data of different types, and determine the classified fault types as candidate fault types;

[0158] Extract fault features of each type of fault data based on the fault subspace extraction method, and associate the fault features with the candidate fault types to obtain a corresponding fault subspace library.

[0159] The construction module 601 is further configured to construct the fault detection model based on the canonical variate analysis method, where the input data of the fault detection model is the normal operation data, and the output of the fault detection model is the first canonical variate corresponding to the normal operation data.

[0160] The construction module 601 is further configured to calculate a squared prediction error statistic value of the historical process data based on the residual between the historical process data and the first canonical variate, and determine the squared prediction error statistic value as the historical error statistic value;

[0161] Calculate a control limit of the historical error statistic value based on the sampling distribution, and determine the control limit as the fault threshold.

[0162] The second calculation module 603 is further configured to preprocess the sample to be measured and calculate a second canonical variate of the preprocessed sample to be measured;

[0163] Calculate a squared prediction error statistic value of the sample to be measured based on the residual between the sample to be measured and the second canonical variate, and determine the squared prediction error statistic value as the current error statistic value.

[0164] The reconstruction module 604 is further configured to:

[0165] When the current error statistic value is less than the fault threshold, determine the state of the sample to be measured as fault-free;

[0166] After reconstructing the to-be-tested sample based on all candidate fault types in the fault subspace library, when the reconstruction error statistical values of the reconstructed sample are not less than the fault threshold, record the fault manifestation of the to-be-tested sample.

[0167] Each module in the above fault diagnosis device can be implemented in whole or in part by software, hardware, and their combination. Each of the above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0168] In an exemplary embodiment, a computer device is provided. The computer device can be a terminal, and its internal structure diagram can be as Figure 8 shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface, the display unit, and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communicating with external terminals in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC), or other technologies. The computer program, when executed by the processor, implements a fault diagnosis method. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0169] In an exemplary embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0170] Collect historical process data of the target system, and construct a fault detection model and a fault subspace library based on the historical process data, where the historical process data includes normal operation data and fault data, and the fault subspace library includes at least one candidate fault type;

[0171] Calculate the historical error statistical value of the target system based on the fault detection model, and calculate the fault threshold based on the historical error statistical value;

[0172] Obtain a sample to be tested, and calculate the current error statistical value of the sample to be tested;

[0173] When the current error statistical value is not less than the fault threshold, reconstruct the sample to be tested in sequence based on the candidate fault types. When the reconstructed error statistical value of the reconstructed sample is less than the fault threshold, determine the candidate fault type currently used for reconstruction as the fault type of the sample to be tested.

[0174] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0175] Collect the historical process data of the target system, and construct a fault detection model and a fault subspace library based on the historical process data, where the historical process data includes normal operation data and fault data, and the fault subspace library includes at least one candidate fault type;

[0176] Calculate the historical error statistical value of the target system based on the fault detection model, and calculate the fault threshold based on the historical error statistical value;

[0177] Obtain a sample to be tested, and calculate the current error statistical value of the sample to be tested;

[0178] When the current error statistical value is not less than the fault threshold, reconstruct the sample to be tested in sequence based on the candidate fault types. When the reconstructed error statistical value of the reconstructed sample is less than the fault threshold, determine the candidate fault type currently used for reconstruction as the fault type of the sample to be tested.

[0179] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the following steps are implemented:

[0180] Collect the historical process data of the target system, and construct a fault detection model and a fault subspace library based on the historical process data, where the historical process data includes normal operation data and fault data, and the fault subspace library includes at least one candidate fault type;

[0181] Calculate the historical error statistical value of the target system based on the fault detection model, and calculate the fault threshold based on the historical error statistical value;

[0182] Obtain a sample to be tested, and calculate the current error statistical value of the sample to be tested;

[0183] When the current error statistical value is not less than the fault threshold, the candidate fault types are used to reconstruct the sample to be tested in sequence. When the reconstruction error statistical value of the reconstructed sample is less than the fault threshold, the candidate fault type currently used for reconstruction is determined as the fault type of the sample to be tested.

[0184] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above method embodiments. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., and are not limited thereto. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, artificial intelligence (AI) processors, etc., and are not limited thereto.

[0185] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this application.

[0186] The above-described embodiments merely represent several implementation manners of this application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of this application. It should be noted that for those of ordinary skill in the art, without departing from the concept of this application, several modifications and improvements can still be made, and these all belong to the protection scope of this application. Therefore, the protection scope of this application shall be subject to the appended claims.

Claims

1. A fault diagnosis method, characterized in that: The method comprises: Collecting historical process data of the target system, and building a fault detection model and a fault subspace library based on the historical process data, wherein the historical process data includes normal operation data and fault data, and the fault subspace library includes at least one candidate fault type; Calculate historical error statistics of the target system based on the fault detection model, and calculate a fault threshold based on the historical error statistics; Obtaining a sample to be tested, and calculating a current error statistic value of the sample to be tested; When the current error statistic is not less than the fault threshold, the samples to be tested are reconstructed in sequence based on the candidate fault types. When the reconstruction error statistic of the reconstructed samples is less than the fault threshold, the candidate fault type currently used for reconstruction is determined as the fault type of the sample to be tested.

2. The method according to claim 1, characterized in that The constructing of a fault subspace library based on the historical process data comprises: Classifying the fault data in the historical process data to obtain multiple groups of fault data of different types, and determining the fault types obtained by classification as candidate fault types; Based on the fault subspace extraction method, the fault features of each type of fault data are extracted, and the fault features are associated with the candidate fault types to obtain a corresponding fault subspace library.

3. The method according to claim 1, characterized in that The constructing of a fault detection model based on the historical process data comprises: The fault detection model is constructed based on a canonical variable analysis method, wherein the input data of the fault detection model is the normal operation data, and the output of the fault detection model is a first canonical variable corresponding to the normal operation data.

4. The method according to claim 3, characterized in that The calculating the historical error statistics of the target system based on the fault detection model, and calculating the fault threshold based on the historical error statistics, includes: Calculating a squared prediction error statistic of the historical process data based on a residual between the historical process data and the first standard variable, and determining the squared prediction error statistic as a historical error statistic; A control limit of the historical error statistic is calculated based on a sampling distribution, and the control limit is determined as the fault threshold.

5. The method according to claim 1, characterized in that The obtaining of the sample to be tested and calculating the current error statistic of the sample to be tested includes: Preprocessing the sample to be tested, and calculating a second normalized variable of the preprocessed sample to be tested; Based on the residual between the sample to be tested and the second standard variable, a square prediction error statistic of the sample to be tested is calculated, and the square prediction error statistic is determined as the current error statistic.

6. The method according to claim 1, characterized in that The method further comprises: When the current error statistic is less than the fault threshold, determining the state of the sample to be tested as no fault; After reconstructing the sample to be tested based on all candidate fault types in the fault subspace library, the fault manifestation of the sample to be tested is recorded when the statistical values ​​of reconstruction errors of the reconstructed samples are not less than the fault threshold.

7. A fault diagnosis device, characterized in that: The device comprises: A construction module, used for collecting historical process data of the target system, and constructing a fault detection model and a fault subspace library based on the historical process data, wherein the historical process data includes normal operation data and fault data, and the fault subspace library includes at least one candidate fault type; A first calculation module, configured to calculate historical error statistics of the target system based on the fault detection model, and calculate a fault threshold based on the historical error statistics; A second calculation module is used to obtain a sample to be tested and calculate a current error statistic value of the sample to be tested; A reconstruction module is used to reconstruct the samples to be tested in sequence based on the candidate fault types when the current error statistic is not less than the fault threshold, and to determine the candidate fault type currently used for reconstruction as the fault type of the sample to be tested when the reconstruction error statistic of the reconstructed sample is less than the fault threshold.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images