Causal discovery-oriented time series data fault root cause diagnosis method and related device

Abstract

The invention discloses a time sequence data fault root cause diagnosis method and a related device for causal discovery, and relates to the technical field of fault identification, and the method comprises the steps: obtaining to-be-processed satellite telemetering time sequence data; calculating a Hotelling T2 statistic of each data sample by using a principal component analysis algorithm, and determining a fault sample according to a comparison result of the Hotelling T2 statistic and a Hotelling T2 statistic control limit; for each fault sample, screening out a fault candidate variable corresponding to the fault sample by using an LASSO-based fault isolation method; applying a channel attention sparse causal network model to the fault candidate variables of each fault sample to obtain a causal matrix between the fault candidate variables; and determining a causal topology among the fault candidate variables according to the causal matrix so as to realize identification of a fault propagation path and positioning of a fault root cause. The method can effectively identify the propagation path of the associated fault and accurately position the root cause.

Description

Technical Field

[0001] The present application relates to the technical field of fault identification, and in particular to a time series data fault root cause diagnosis method and related device for causal discovery. Background Art

[0002] During the operation of satellites, material flows, information flows, and energy flows interact with each other. Once a system unit fails, due to the characteristics of fault propagation, it is easy to cause abnormalities or failures in other system units. Therefore, root cause diagnosis (RCD) has become an important task. RCD methods are mainly divided into methods based on traditional causal inference algorithms and methods based on deep learning. Traditional causal inference algorithms face great challenges in dealing with highly complex and large-scale systems. With the in-depth integration of industrialization and informatization, multiple sensors are installed on various components of on-orbit satellites through distributed measurement systems to collect and store operating data as telemetry data. The operating status of the satellite system can be understood through the analysis of telemetry data. In recent years, the development of deep learning technology has made the application of deep learning-based methods in the field of RCD more and more popular. By performing RCD on faulty satellite systems, the propagation path of associated faults can be effectively identified and the root cause can be accurately located, which helps to reduce the risk of fault propagation.

[0003] RCD mainly includes two steps: fault isolation and fault root cause location. Common fault isolation methods include contribution graph method and reconstruction contribution graph method. When a fault is detected, the related variables causing the fault are determined through fault isolation method. Contribution graph method is the most commonly used method. It calculates the contribution value of each sample variable to the monitoring statistic. The one with a large contribution value is considered to be more likely to cause the fault. However, when using the contribution graph method for fault location, due to the influence of the fault variable, the contribution value of the non-fault variable will increase, which is easy to cause residual contamination and misdiagnosis. Although the reconstruction contribution graph method has improved the traditional contribution graph method to address this problem, the commonly used diagnostic indicators (T 2 The failure problem and residual contamination problem of the fault-related variables (and SPE statistics) cannot be perfectly solved. In addition, the contribution graph method can locate the variables related to the fault, but it cannot identify the root variable of the fault and identify the propagation path. This is because the contribution graph method always considers the correlation between the fault variables, but does not consider the causality between them.

[0004] At present, the commonly used causal analysis methods mainly include transfer entropy (TE), convergent cross mapping (CCM) and Granger causality (GC). Among them, the TE method has a large amount of calculation and cannot distinguish between direct and indirect causal relationships. CCM is only applicable to weak to medium coupling systems. For strong coupling systems, it will face the problem of generalized synchronization. Although the GC method can be used to analyze the causal relationship between variable data, there are still some problems that have not been well solved: (1) The selection of variables involved in the GC method analysis depends entirely on the subjective experience of engineers, and the uncertainty of the number of selections will lead to uncertainty in the results; (2) The traditional GC method cannot analyze nonlinear time series data and is prone to false redundant causal relationships; (3) The traditional GC method is designed to infer the causal relationship between two variables and is not suitable for satellite systems with multiple variables.

[0005] Therefore, there is an urgent need for a technology that can effectively identify the propagation path of associated faults and accurately locate the root cause. Summary of the invention

[0006] The purpose of this application is to provide a time series data fault root cause diagnosis method and related devices for causal discovery, which can effectively identify the propagation path of associated faults and accurately locate the root cause.

[0007] To achieve the above objectives, this application provides the following solutions:

[0008] In a first aspect, the present application provides a time series data fault root cause diagnosis method for causal discovery, comprising:

[0009] Obtaining satellite telemetry time series data to be processed; in the satellite telemetry time series data to be processed, satellite telemetry data at one time point is a data sample; and a data sample includes a plurality of satellite telemetry variables;

[0010] The principal component analysis algorithm is used to calculate the Hotelling T of each data sample. 2 statistic, and according to the Hotelling T 2 Statistic and Hotelling's T 2 The comparison results of statistical control limits determine the data samples that are regarded as fault samples;

[0011] For each fault sample, a fault isolation method based on LASSO is used to screen out the fault candidate variables corresponding to the fault sample; the fault candidate variables are some variables among the several satellite telemetry variables;

[0012] Applying a channel attention sparse causal network model to the fault candidate variables of each of the fault samples to obtain a causal matrix between the fault candidate variables;

[0013] The causal topology between the fault candidate variables is determined according to the causal matrix to identify the fault propagation path and locate the root cause of the fault.

[0014] In a second aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned method for diagnosing the root cause of a time series data fault oriented to causal discovery.

[0015] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for diagnosing the root cause of a time series data fault oriented to causal discovery.

[0016] In a fourth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above-mentioned time series data fault root cause diagnosis method oriented to causal discovery.

[0017] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0018] The present application provides a time series data fault root cause diagnosis method and related device for causal discovery, obtaining satellite telemetry time series data to be processed; satellite telemetry data at a time point is a data sample; the Hotelling T of each data sample is calculated using the principal component analysis algorithm 2 statistic, and according to the Hotelling T 2 Statistic and Hotelling's T 2 The comparison result of the statistical control limit determines the data sample regarded as the fault sample; for each fault sample, the fault candidate variable corresponding to the fault sample is screened out using the fault isolation method based on LASSO; the causal matrix between the fault candidate variables is obtained by applying the channel attention sparse causal network model to the fault candidate variables of each fault sample; the causal topology between the fault candidate variables is determined according to the causal matrix to realize the identification of the fault propagation path and the location of the root cause of the fault. In the present invention, the fault isolation method based on LASSO is used to screen the fault candidate variables, which solves the diagnostic indicator failure and residual pollution problems of the traditional contribution graph, and determines the candidate set of fault variables. At the same time, in view of the defect that the traditional Granger causal analysis method cannot analyze the multivariate causal relationship and nonlinear causal relationship, a causal relationship analysis model of the channel attention sparse causal network is proposed. The proposed channel attention sparse causal network model can be used to analyze the nonlinear causal relationship between the fault variables, determine the fault propagation causal topology, and locate the root cause of the fault. Therefore, the method of the present invention can not only accurately perform fault detection, but also accurately locate the root cause of the fault. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 This is an application environment diagram of a time series data fault root cause diagnosis method for causal discovery in one embodiment of the present application;

[0021] Figure 2 A flowchart of a method for diagnosing root causes of time series data faults for causal discovery provided in one embodiment of the present application;

[0022] Figure 3 A schematic diagram of the principle of a method for diagnosing root causes of time series data faults for causal discovery provided by an embodiment of the present application;

[0023] Figure 4 A schematic diagram of the structure of a channel attention sparse causal network model provided in one embodiment of the present application;

[0024] Figure 5 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0026] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0027] The time series data fault root cause diagnosis method for causal discovery provided by the embodiment of the present application can be applied to Figure 1 In the application environment shown in FIG. 1 , the terminal communicates with the server through the network. The data storage system can store the data that the server needs to process. The data storage system can be set up separately, integrated on the server, or placed on the cloud or other servers. The terminal can send the satellite telemetry time series data to be processed to the server. After the server receives the satellite telemetry time series data to be processed, the server obtains the Hotelling T of each data sample calculated using the principal component analysis algorithm.2 statistic and Hotelling T 2 The comparison result of the statistical control limit determines the fault sample; for each fault sample, the fault candidate variable corresponding to the fault sample is screened out using the fault isolation method based on LASSO; the channel attention sparse causal network model is applied to the fault candidate variables of each fault sample to obtain the causal matrix between the fault candidate variables; the causal topology between the fault candidate variables is determined according to the causal matrix to realize the identification of the fault propagation path and the location of the root cause of the fault. The server can feed back the obtained causal topology between the fault candidate variables to the terminal. In addition, in some embodiments, the time series data fault root cause diagnosis method for causal discovery can also be implemented by the server or the terminal alone, such as the terminal can directly perform fault detection and root cause location processing on the satellite telemetry time series data to be processed, or the server can obtain the satellite telemetry time series data to be processed from the data storage system and perform fault detection and root cause location processing.

[0028] The terminals may be, but are not limited to, various desktop computers, laptops, smart phones, tablet computers, IoT devices and portable wearable devices. IoT devices may be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, etc. Portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. The server may be implemented as an independent server or a server cluster consisting of multiple servers, or a cloud server.

[0029] In an exemplary embodiment, Figure 2 and Figure 3 As shown, a time series data fault root cause diagnosis method for causal discovery is provided. The method is executed by a computer device, and can be executed by a computer device such as a terminal or a server alone, or by a terminal and a server together. In the embodiment of the present application, the method is applied to Figure 1 The server in the example is used to illustrate, including the following steps 101 to 105. Among them:

[0030] Step 101, obtaining satellite telemetry time series data to be processed.

[0031] The telemetry data to be processed is stored in the form of a table with M rows and N columns; the first row and first column of the table are spaces; the telemetry data to be processed includes N-1 telemetry variables, and are stored in the first row of the table except spaces; the telemetry data to be processed includes M-1 time points, and are stored in the first column of the table except spaces; the telemetry data to be processed includes (M-1)*(N-1) data points, which are stored in the table except the first column and the first row; wherein the first row of the mark column of the table is used to store the marked telemetry variables, and the other rows of the mark column are used to store the data points collected at different time points of the marked telemetry variables; the mark column is any other column of the table except the first column, and the marked telemetry variable is any telemetry variable. Therefore, in the satellite telemetry time series data to be processed, the satellite telemetry data at a time point is a data sample; a data sample includes several satellite telemetry variables.

[0032] Step 102, using the principal component analysis algorithm to calculate the Hotelling T of each data sample 2 statistic, and according to the Hotelling T 2 Statistic and Hotelling's T 2 The comparison results of the statistical control limits determine the data samples that are considered to be faulty samples.

[0033] Step 103: for each fault sample, a fault isolation method based on LASSO is used to screen out the fault candidate variables corresponding to the fault sample; the fault candidate variables are some variables among the several satellite telemetry variables.

[0034] Step 104: Apply a channel attention sparse causal network model to the fault candidate variables of each of the fault samples to obtain a causal matrix between the fault candidate variables.

[0035] Step 105 , determining the causal topology between the fault candidate variables according to the causal matrix to identify the fault propagation path and locate the root cause of the fault.

[0036] After obtaining the causal matrix between all fault variables in the fault candidate variable set, the causal topological structure between the fault variables can be determined. Based on the causal topological structure, the fault propagation path can be generated. Variables without in-degree nodes are regarded as the root dependent variables causing the fault.

[0037] Implement the above steps 101 to 105. First, the principal component analysis method is used to detect the fault. If the fault is detected, the subsequent root cause location step of the fault is continued. Secondly, the present invention proposes a fault isolation method based on LASSO to determine the appropriate number of fault-related variables to participate in the subsequent causal relationship analysis. Finally, a causal relationship analysis method based on deep learning called channel attention sparse causal network is proposed to locate the root cause of the fault. The present invention combines causal relationship analysis with fault root cause location, which solves the problem that most methods in the current fault identification field only focus on fault detection. The present invention introduces a fault isolation method based on LASSO, overcomes the diagnostic indicator failure and residual pollution problems of the traditional contribution graph, introduces a channel attention sparse causal network model, and overcomes the shortcomings of the traditional Granger causal analysis method that cannot analyze the causal relationship between multiple variables and nonlinear causal relationship. Ground engineers can locate the root cause of the fault of the on-orbit satellite in time through the method proposed by the present invention, which can effectively avoid the risk of fault propagation. In addition, the method of the present invention is a method that combines GC and neural network, which is completely data-driven and does not require expert knowledge in related fields. In addition, the present invention detects faults in industrial equipment through telemetry time series data from on-orbit satellites, identifies subsequent propagation paths of the detected faults, and locates the root causes of the faults.

[0038] In another exemplary embodiment of the present application, for the fault detection in step 102, when the telemetry data to be processed is analyzed using the principal component analysis algorithm, it is necessary to first use the principal component analysis method to train the telemetry data of the satellite during normal operation, and the standard T obtained by training 2 Control limits (i.e. Hotelling T 2 Statistical control limit); the standard T 2 The control limits are used to determine the T of the telemetry data to be processed. 2 Therefore, in step 102, the Hotelling T of each data sample is calculated using the principal component analysis algorithm. 2 statistic, and according to the Hotelling T 2 Statistic and Hotelling's T 2 The comparison results of the statistical control limits determine the data samples that are considered as fault samples, including:

[0039] (a1) Obtain telemetry data on the normal operation of the satellite.

[0040] (a2) The principal component analysis method is used to train the telemetry data of the normal operation of the satellite to obtain the Hotelling T 2 Statistical control limits.

[0041] (a3) Calculate the Hotelling T of each data sample using the principal component analysis algorithm2 Statistics.

[0042] (a4) Determine the Hotelling T 2 Is the statistic greater than the Hotelling T 2 Statistical control limits.

[0043] (a5) If yes, the corresponding data sample is regarded as a fault sample. If no, it is a normal sample and no subsequent fault root cause location is required.

[0044] In another exemplary embodiment of the present application, for step 103, after a fault is detected, a fault candidate variable set is screened according to a fault isolation method based on LASSO; specifically, when a traditional contribution graph is used for fault location, T is often used. 2 and SPE as indicators, but in some special cases, these two indicators will fail. Special cases are:

[0045] (1) When the number of principal components in the PCA model of the original data is 1, based on T 2 The RCP method cannot effectively locate faults;

[0046] (2) When the number of principal components is one less than the number of variables, the SPE-based RCP method will also fail to diagnose.

[0047] Therefore, the present invention transforms the fault reconstruction problem into a constrained combination optimization problem of the same form as LASSO. Based on solving the optimization problem, the variables that contribute most to the fault reconstruction are calculated. Therefore, step 103, for each fault sample, the fault isolation method based on LASSO is used to screen out the fault candidate variables corresponding to the fault sample, specifically including:

[0048] (b1) For each fault sample, a constrained optimization model for fault sample reconstruction is constructed.

[0049] Reconstruct the faulty sample to eliminate the impact of the fault on the faulty sample, thereby restoring the faulty sample to a normal sample. Assume that there is a faulty sample x, and the fault reconstruction is expressed by equation (1). * represents the normal sample after reconstruction; Ξ is a vector representing the fault direction, whose elements are 0 or 1, used to represent the variable affected by the detected fault; e represents the magnitude of the fault, that is, the intensity of the fault's impact on the variable; ° represents the Hadamard product. After reconstruction, x * should be close to a normal sample, which means that for x * Calculate T 2 The statistic will be below the control limit.

[0050]

[0051] Since faults in satellite telemetry processes are sparse and usually only affect some variables, the number of non-zero elements in the fault vector f must be limited, which can be expressed as a constrained optimization problem as shown in equation (2). ||·||1 denotes the L1 norm, and c is the maximum number of faulty variables specified in advance.

[0052]

[0053] (b2) Converting the constrained optimization model for reconstructing the fault samples into a constrained combination optimization model with the same form as LASSO.

[0054]

[0055] Where P is the loading matrix, which refers to the loading matrix in the PCA algorithm, A is a diagonal matrix consisting of the first k eigenvalues, the eigenvalue is the eigenvalue obtained by the PCA algorithm, and L is A -1 The lower triangular matrix obtained after Cholesky decomposition, λ is the weight. That is, the constrained optimization model for solving (2) is transformed into y = (PL) T x, Z = (PL) T , the LASSO problem with β = f, that is, the constrained combination optimization model with the same form as LASSO:

[0056] (b3) Solve the constrained combination optimization model with the same form as LASSO to obtain the fault candidate variables corresponding to each fault sample.

[0057] The above fault reconstruction problem is transformed into a constrained combinatorial optimization problem with the same form as LASSO. The fault vector f is obtained by solving the optimization problem, in which the non-zero elements are the screened fault variables.

[0058] The solution of LASSO is a sparse variable set, which is a subset of all variables. The variables in this sparse variable set are considered to be the variables that contribute most to the fault, and the fault candidate variable set is determined based on this principle.

[0059] In another exemplary embodiment of the present application, corresponding to step 104, based on the proposed channel attention sparse causal network model and the above-mentioned fault candidate variable set, a causal matrix between the fault candidate variables is obtained. First, based on the proposed channel attention sparse causal network model, the prediction of the satellite telemetry variable (prediction of the fault candidate variable) is obtained, as shown in formula (3). Where X represents the fault candidate variable in the satellite telemetry data, CASCN (·) represents the channel sparse causal network model, Represents the prediction of the fault candidate variable.

[0060]

[0061] Secondly, by applying sparse penalty to the weights of the channel sparse causal network model, sparse weights are obtained. This weight is the causal matrix obtained based on the model. The non-zero elements of the causal matrix indicate that there is a causal relationship between the corresponding fault variables. The sparse weight is obtained by solving the optimization problem of formula (4), which is the loss calculation of the model. Where W represents the model weight; λ GL Represents,LASSO sparse penalty coefficient; λ Ridge represents the ridge regression coefficient, which is used to prevent the model from overfitting; ||·||2 represents the L2 norm.

[0062]

[0063] Therefore, step 104, applying the channel attention sparse causal network model to the fault candidate variables of each fault sample to obtain a causal matrix between the fault candidate variables, specifically includes:

[0064] (c1) Taking the fault candidate variables of the first preset time period as the model input, the channel attention sparse causal network model is used to output the fault candidate variable prediction data of the second preset time period; the time point of the second preset time period is greater than the time point of the first preset time period.

[0065] (c2) Calculating the loss of the model based on the fault candidate variables of the first preset time period, the fault candidate variable prediction data of the second preset time period and the relevant regularization applied to the model weight value, that is, formula (4).

[0066] (c3) When determining the model with the minimum loss, the model weight value is the causal matrix between the fault candidate variables.

[0067] In another exemplary embodiment of the present application, Figure 4 As shown, the channel attention sparse causal network model includes multiple channel attention sparse causal networks; the input of each channel attention sparse causal network is the fault candidate variable of each fault sample, and each channel attention sparse causal network outputs the prediction data of a fault candidate variable; the number of channel attention sparse causal networks is equal to the number of fault candidate variables.

[0068] Each of the channel-attention sparse causal networks includes an attention module and a multi-channel time-domain convolution module connected in sequence.

[0069] The attention module includes a sequentially connected time series data input layer, an expansion dimension and reshaping layer, an input embedding layer and a Transformer encoder layer; the input of the time series data input layer is the fault candidate variable of each of the fault samples.

[0070] The multi-channel time domain convolution module includes a linear head layer, a reshaping layer, a multi-channel temporal convolution network layer and a univariate prediction output layer connected in sequence; the input of the linear head layer is the output of the Transformer encoder layer; the output of the univariate prediction output layer is the prediction data of a fault candidate variable.

[0071] Among them, the dimension expansion and reshaping layers are the precursor operations of the input embedding layer; the input embedding layer is an essential operation before the Transformer performs self-attention, including position embedding and token embedding.

[0072] The linear head layer is usually composed of linear layers, which are mainly used to align the output dimensions in the present invention; the reshape layer is used to reshape the output to meet the input requirements of the multi-channel convolutional network layer; the temporal convolutional network is a popular network, and the multi-channel temporal convolutional network is its extension; the univariate prediction output layer is a linear layer in the present invention, and its purpose is to process the multi-dimensional output of the multi-channel temporal convolutional network into a single-dimensional output.

[0073] The present invention is a method based on deep learning, which proposes a time series data fault root cause diagnosis method for causal discovery. Different from the traditional fault diagnosis method, the proposed method focuses more on identifying the propagation path after the fault occurs and the root cause of the fault. First, in view of the failure problem of diagnostic indicators in the fault location process, a fault isolation method based on LASSO is studied to avoid the problem of diagnostic failure. The method is combined with PCA to realize the detection of faults and the screening of fault variable candidate sets, which is used to select a suitable number of variables to participate in the construction of subsequent fault propagation causal topology. Finally, a causal analysis method of channel-attention sparse causal network is proposed to establish a fault propagation causal topology and locate the root cause of the fault. Satellite ground management personnel can use this method to locate the root cause of the fault of the on-orbit satellite in time, so that relevant measures can be taken quickly to repair it, suppress the risk of fault propagation, and ensure the smooth execution of space missions.

[0074] The present application also provides an application scenario, which applies the above-mentioned time series data fault root cause diagnosis method for causal discovery. Specifically: The time series data fault root cause diagnosis method for causal discovery provided in this embodiment can be applied in a satellite system fault diagnosis scenario, which scenario includes a data acquisition link, a data processing link and a data result display link. The data acquisition link is used to collect satellite telemetry time series data to be processed; the data processing link is used to calculate the Hotelling T of each data sample using a principal component analysis algorithm. 2Statistics, and based on this, determine the fault sample. For each fault sample, use the LASSO-based fault isolation method to screen out the fault candidate variables corresponding to the fault sample; apply the channel attention sparse causal network model to the fault candidate variables of each fault sample to obtain the causal matrix between the fault candidate variables, and then determine the causal topology between the fault candidate variables to achieve the identification of the fault propagation path and the location of the root cause of the fault. The data display link is used to display the results of the fault propagation path and the root cause of the fault. The time series data fault root cause diagnosis method for causal discovery provided in this embodiment belongs to the data processing link.

[0075] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. 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, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store fault root cause diagnosis data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a time series data fault root cause diagnosis method for causal discovery is implemented.

[0076] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0077] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0078] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0079] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0080] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and 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 embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile 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), magnetic 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 may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0081] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0082] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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, they should be considered to be within the scope of this specification.

[0083] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A time series data fault root cause diagnosis method for causal discovery, characterized in that: The time series data fault root cause diagnosis method for causal discovery includes: Obtaining satellite telemetry time series data to be processed; in the satellite telemetry time series data to be processed, satellite telemetry data at one time point is a data sample; and a data sample includes a plurality of satellite telemetry variables; The principal component analysis algorithm is used to calculate the Hotelling T of each data sample. 2 statistic, and according to the Hotelling T 2 Statistic and Hotelling's T 2 The comparison results of statistical control limits determine the data samples that are regarded as fault samples; For each fault sample, a fault isolation method based on LASSO is used to screen out the fault candidate variables corresponding to the fault sample; the fault candidate variables are some variables among the several satellite telemetry variables; Applying a channel attention sparse causal network model to the fault candidate variables of each of the fault samples to obtain a causal matrix between the fault candidate variables; The causal topology between the fault candidate variables is determined according to the causal matrix to identify the fault propagation path and locate the root cause of the fault.

2. The time series data fault root cause diagnosis method for causal discovery according to claim 1 is characterized in that: The principal component analysis algorithm is used to calculate the Hotelling T of each data sample. 2 statistic, and according to the Hotelling T 2 Statistic and Hotelling T 2 The comparison results of the statistical control limits determine the data samples that are considered as fault samples, including: Obtaining telemetry data on normal satellite operation; The principal component analysis method is used to train the telemetry data of the normal operation of the satellite, and the Hotelling T 2 Statistical control limits; The principal component analysis algorithm is used to calculate the Hotelling T of each data sample. 2 Statistics; Judging the Hotelling T 2 Is the statistic greater than the Hotelling T 2 Statistical control limits; If yes, the corresponding data sample is regarded as a fault sample.

3. The time series data fault root cause diagnosis method for causal discovery according to claim 1, characterized in that: For each fault sample, the fault isolation method based on LASSO is used to screen out the fault candidate variables corresponding to the fault sample, including: For each fault sample, a constrained optimization model for fault sample reconstruction is constructed; Converting the constrained optimization model reconstructed by the fault sample into a constrained combination optimization model with the same form as LASSO; Solve the constrained combination optimization model with the same form as LASSO to obtain the fault candidate variables corresponding to each fault sample.

4. The time series data fault root cause diagnosis method for causal discovery according to claim 1, characterized in that: The constrained combination optimization model with the same form as LASSO is: where y = (PL) T x, Z = (PL) T , β = f; Where P is the loading matrix; A is a diagonal matrix consisting of the first k eigenvalues, and L is A -1 The lower triangular matrix obtained after Cholesky decomposition; λ is the weight; x is the fault sample; f is the fault vector.

5. The time series data fault root cause diagnosis method for causal discovery according to claim 1, characterized in that: A channel attention sparse causal network model is applied to the fault candidate variables of each fault sample to obtain a causal matrix between the fault candidate variables, specifically including: Taking the fault candidate variables of the first preset time period as the model input, the channel attention sparse causal network model is used to output the fault candidate variable prediction data of the second preset time period; the time point of the second preset time period is greater than the time point of the first preset time period; Calculating the loss of the model based on the fault candidate variables of the first preset time period, the fault candidate variable prediction data of the second preset time period, and the relevant regularization applied to the model weight value; The model weight value when determining the model with the minimum loss is the causal matrix between the fault candidate variables.

6. The time series data fault root cause diagnosis method for causal discovery according to claim 1, characterized in that: The calculation expression of the model weight value when the model loss is the smallest is: in, represents the model weight; λ GL represents the LASSO sparse penalty coefficient; λ Ridge represents the ridge regression coefficient; X represents the fault candidate variable of the first preset time period; represents the fault candidate variable prediction data of the second preset time period; ||·||2 represents the L2 paradigm.

7. The time series data fault root cause diagnosis method for causal discovery according to claim 1, characterized in that: The channel attention sparse causal network model includes multiple channel attention sparse causal networks; the input of each channel attention sparse causal network is the fault candidate variable of each fault sample, and each channel attention sparse causal network outputs prediction data of a fault candidate variable; Each of the channel attention sparse causal networks comprises an attention module and a multi-channel time-domain convolution module connected in sequence; The attention module comprises a sequentially connected time series data input layer, an expansion dimension and reshaping layer, an input embedding layer and a Transformer encoder layer; the input of the time series data input layer is a fault candidate variable of each of the fault samples; The multi-channel time domain convolution module includes a linear head layer, a reshaping layer, a multi-channel temporal convolution network layer and a univariate prediction output layer connected in sequence; the input of the linear head layer is the output of the Transformer encoder layer; the output of the univariate prediction output layer is the prediction data of a fault candidate variable.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the time series data fault root cause diagnosis method for causal discovery according to any one of claims 1 to 7.

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 method for diagnosing the root cause of a time series data fault oriented to causal discovery according to any one of claims 1 to 7 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for diagnosing the root cause of a time series data fault oriented to causal discovery according to any one of claims 1 to 7 is implemented.

Images