Fault detection method and device, computer device, readable storage medium and program product

Abstract

The application relates to a fault detection method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring sample monitoring data collected in a preset historical period for an entity device; acquiring first sample monitoring data and second sample monitoring data; determining a covariance matrix according to the first sample monitoring data and the second sample monitoring data, so as to further obtain a conversion matrix; acquiring third sample monitoring data and fourth sample monitoring data collected at a current moment; calculating a standard variable matrix according to the conversion matrix, the third sample monitoring data and the fourth sample monitoring data, and calculating an information increment matrix and a first average value of element values in the information increment matrix; calculating a dynamic threshold value according to the first average value and a preset threshold coefficient, and judging whether the entity device is faulty according to the relationship between the dynamic threshold value and the first average value. The method can improve the accuracy of fault detection.

Description

TECHNICAL FIELD

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

[0002] With the development of sensor technology and computer technology, it is possible to collect and store a large amount of data in civil industrial processes such as chemical production and military equipment such as missiles and weapons. In order to ensure their safe and stable operation, experts and scholars have proposed multivariate statistical process monitoring (MSPM) technology. MSPM technology mainly includes principal component analysis, partial least squares, independent component analysis (ICA) and canonical variate analysis (CVA). Among them, the basic independent component analysis, principal component analysis and partial least squares algorithm are all based on the assumption that the process data has no dynamic characteristics for fault detection, while CVA is born for considering the dynamic characteristics of process data based on the optimal statistical inference.

[0003] Although the control limit obtained by the prior art can distinguish between normal and abnormal states, it cannot reflect the dynamic changes of process data in real time, resulting in inaccurate fault judgment results. SUMMARY

[0004] Therefore, it is necessary to provide a fault detection method, device, computer device, computer readable storage medium and computer program product capable of improving the accuracy of fault detection results in view of the above technical problems.

[0005] In a first aspect, the present application provides a fault detection method, comprising:

[0006] For the entity device, sample monitoring data collected in a preset historical period is obtained;

[0007] For a historical time in the preset historical period, sample monitoring data collected before the historical time and in the preset historical period is taken as first sample monitoring data, and sample monitoring data collected after the historical time and in the preset historical period is taken as second sample monitoring data;

[0008] Based on the first sample monitoring data, a first Hankel matrix is determined, and based on the second sample monitoring data, a second Hankel matrix is determined;

[0009] Based on the first Hankel matrix and the second Hankel matrix, a covariance matrix is calculated, and singular value decomposition is performed on the covariance matrix to obtain a conversion matrix;

[0010] Third sample monitoring data collected at the current moment and fourth sample monitoring data collected within a preset time length corresponding to a period before the current moment are obtained.

[0011] According to the conversion matrix, the third sample monitoring data and the fourth sample monitoring data, a canonical variable matrix is calculated.

[0012] Based on the canonical variable matrix and the third sample monitoring data, an information increment matrix and a first average value of element values in the information increment matrix are calculated.

[0013] According to the relationship between the first average value and a preset threshold coefficient, a dynamic threshold value is calculated, and according to the relationship between the dynamic threshold value and the first average value, it is judged whether the entity device is faulty.

[0014] In one of the embodiments, based on the first sample monitoring data, a first Hankel matrix is determined, and based on the second sample monitoring data, a second Hankel matrix is determined, which includes:

[0015] The first sample monitoring data and the second sample monitoring data are respectively standardized.

[0016] According to the standardized first sample monitoring data, a plurality of first vectors are constructed, and according to the standardized second sample monitoring data, a plurality of second vectors are constructed.

[0017] A first average vector of the plurality of first vectors and a second average vector of the plurality of second vectors are calculated, a first Hankel matrix is determined according to the plurality of first vectors and the first average vector, and a second Hankel matrix is determined according to the plurality of second vectors and the second average vector.

[0018] In one of the embodiments, the canonical variable matrix is calculated according to the conversion matrix, the third sample monitoring data and the fourth sample monitoring data, which includes:

[0019] The third sample monitoring data and the fourth sample monitoring data are respectively standardized.

[0020] According to the standardized third sample monitoring data and the standardized fourth sample monitoring data, a plurality of third vectors are constructed, and a fourth vector is constructed according to the plurality of third vectors, and a third average vector of the plurality of third vectors is calculated.

[0021] The fourth vector is subtracted from the third average vector, multiplied by the conversion matrix, and the canonical variable matrix is obtained.

[0022] In one embodiment, the calculating the information increment matrix based on the standardized variable matrix and the third sample monitoring data comprises:

[0023] The pre-processed standardized variable matrix is obtained based on a difference between each element value in the standardized variable matrix and an average value of element values in a row to which the element value belongs.

[0024] The first local matrix at the current time and the second local matrix at the next time of the current time are determined based on the third sample monitoring data and the pre-processed standardized variable matrix.

[0025] The information increment matrix is calculated based on the first local matrix and the second local matrix.

[0026] In one embodiment, the calculating the information increment matrix based on the first local matrix and the second local matrix comprises:

[0027] The third local matrix is constructed based on a corresponding vector at an overlapping time between the first local matrix and the second local matrix, and a third average vector corresponding to the vectors in the third local matrix is calculated.

[0028] The first covariance matrix at the current time and the second covariance matrix at the next time of the current time are calculated based on the first local matrix, the second local matrix and the third average vector.

[0029] The information increment matrix is obtained by subtracting the first covariance matrix from the second covariance matrix.

[0030] In one embodiment, the determining whether the entity device is faulty based on the relationship between the dynamic threshold and the first average value comprises:

[0031] When the dynamic threshold is less than the first average value, the entity device is in a faulty state.

[0032] When the dynamic threshold is greater than the first average value, the entity device is in a normal state.

[0033] In a second aspect, the application further provides a fault detection device, comprising:

[0034] An obtaining module is configured to obtain sample monitoring data collected in a preset historical period for an entity device.

[0035] The acquisition module is further configured to, for a historical moment within the preset historical period, take sample monitoring data collected before the historical moment and within the preset historical period as first sample monitoring data, and take sample monitoring data collected after the historical moment and within the preset historical period as second sample monitoring data.

[0036] The determination module is configured to determine a first Hankel matrix based on the first sample monitoring data, and determine a second Hankel matrix based on the second sample monitoring data.

[0037] The determination module is further configured to calculate a covariance matrix based on the first Hankel matrix and the second Hankel matrix, and perform singular value decomposition on the covariance matrix to obtain a conversion matrix.

[0038] The acquisition module is further configured to acquire third sample monitoring data collected at a current moment and fourth sample monitoring data collected within a preset time length corresponding to a period before the current moment.

[0039] The calculation module is configured to calculate a canonical variable matrix based on the conversion matrix, the third sample monitoring data and the fourth sample monitoring data.

[0040] The calculation module is further configured to calculate an information increment matrix and a first average value of element values in the information increment matrix based on the canonical variable matrix and the third sample monitoring data.

[0041] The judgment module is configured to calculate a dynamic threshold value based on the first average value and a preset threshold coefficient, and determine whether the entity device is faulty based on a relationship between the dynamic threshold value and the first average value.

[0042] In a third aspect, the present application further provides a computer device, comprising a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0043] For an entity device, acquiring sample monitoring data collected within a preset historical period;

[0044] For a historical moment within the preset historical period, taking sample monitoring data collected before the historical moment and within the preset historical period as first sample monitoring data, and taking sample monitoring data collected after the historical moment and within the preset historical period as second sample monitoring data;

[0045] Determining a first Hankel matrix based on the first sample monitoring data, and determining a second Hankel matrix based on the second sample monitoring data.

[0046] based on the first Hankel matrix and the second Hankel matrix, a covariance matrix is calculated, singular value decomposition is performed on the covariance matrix, and a conversion matrix is obtained;

[0047] third sample monitoring data collected at the current moment and fourth sample monitoring data collected within a preset time period before the current moment are obtained;

[0048] based on the conversion matrix, the third sample monitoring data and the fourth sample monitoring data, a canonical variable matrix is calculated;

[0049] based on the canonical variable matrix and the third sample monitoring data, an information increment matrix and a first mean value of element values in the information increment matrix are calculated;

[0050] a dynamic threshold is calculated according to the first mean value and a preset threshold coefficient, and it is determined whether the entity device is faulty according to the relationship between the dynamic threshold and the first mean value.

[0051] In a fourth aspect, the present application further provides a computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the following steps:

[0052] For an entity device, sample monitoring data collected within a preset historical period is obtained;

[0053] For a historical moment within the preset historical period, sample monitoring data collected before the historical moment and within the preset historical period is taken as first sample monitoring data, and sample monitoring data collected after the historical moment and within the preset historical period is taken as second sample monitoring data;

[0054] based on the first sample monitoring data, a first Hankel matrix is determined, and based on the second sample monitoring data, a second Hankel matrix is determined;

[0055] based on the first Hankel matrix and the second Hankel matrix, a covariance matrix is calculated, singular value decomposition is performed on the covariance matrix, and a conversion matrix is obtained;

[0056] third sample monitoring data collected at the current moment and fourth sample monitoring data collected within a preset time period before the current moment are obtained;

[0057] based on the conversion matrix, the third sample monitoring data and the fourth sample monitoring data, a canonical variable matrix is calculated;

[0058] calculating an information increment matrix and a first mean value of element values in the information increment matrix based on the canonical variable matrix and the third sample monitoring data;

[0059] calculating a dynamic threshold value according to the first mean value and a preset threshold coefficient, and judging whether the physical device is faulty according to a relationship between the dynamic threshold value and the first mean value.

[0060] In a fifth aspect, the present application further provides a computer program product comprising a computer program which, when executed by a processor, implements the following steps:

[0061] For the physical device, sample monitoring data collected in a preset historical period is acquired;

[0062] For a historical time in the preset historical period, sample monitoring data collected before the historical time and in the preset historical period is taken as first sample monitoring data, and sample monitoring data collected after the historical time and in the preset historical period is taken as second sample monitoring data;

[0063] A first Hankel matrix is determined based on the first sample monitoring data, and a second Hankel matrix is determined based on the second sample monitoring data;

[0064] A covariance matrix is calculated based on the first Hankel matrix and the second Hankel matrix, and a transformation matrix is obtained by singular value decomposition of the covariance matrix;

[0065] Third sample monitoring data collected at a current time and fourth sample monitoring data collected in a preset time length corresponding to the current time are acquired;

[0066] A canonical variable matrix is calculated according to the transformation matrix, the third sample monitoring data and the fourth sample monitoring data;

[0067] An information increment matrix and a first mean value of element values in the information increment matrix are calculated based on the canonical variable matrix and the third sample monitoring data;

[0068] A dynamic threshold value is calculated according to the first mean value and a preset threshold coefficient, and whether the physical device is faulty is judged according to a relationship between the dynamic threshold value and the first mean value.

[0069] The fault detection method, device, computer device, computer readable storage medium and computer program product, first, the normative variable matrix of the third monitoring data and the fourth monitoring data of the real-time sample is calculated, then the normative variable matrix is taken as the input, and the information increment matrix of the improved covariance matrix is introduced. Then the mean of the matrix is calculated, and the dynamic threshold is set according to the state (normal or fault) of the previous sample. Finally, the current working state is judged by comparing the size of the mean and the dynamic threshold, which combines the dynamic change of the process data, so that the fault detection result is more accurate. BRIEF DESCRIPTION OF DRAWINGS

[0070] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed to be used in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0071] Figure 1 The application environment diagram of the fault detection method in an embodiment;

[0072] Figure 2 The flowchart of the fault detection method in an embodiment;

[0073] Figure 3 The flowchart of the fault detection method in another embodiment;

[0074] Figure 4 The structural block diagram of the fault detection device in an embodiment;

[0075] Figure 5 The internal structure diagram of the computer device in an embodiment. DETAILED DESCRIPTION

[0076] In order to make the purpose, technical scheme and advantages of the present application more clear, the present application will be further described in detail below in combination 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.

[0077] The fault detection method provided by the embodiments of the present application can be applied to, for example Figure 1The application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the sample monitoring data required by the server 104 to process. The data storage system can be integrated on the server 104, or placed on the cloud or other network servers. Among them, the terminal 102 can be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, Internet of Things devices and portable wearable devices, Internet of Things devices can be smart speakers, smart televisions, smart air conditioners, smart car devices, projection devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The head-mounted device can be a virtual reality (VR) device, an augmented reality (AR) device, smart glasses, etc. The server 104 can be a standalone physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0078] In an exemplary embodiment, as shown in Figure 2 , a fault detection method is provided, which is applied to Figure 1 the terminal 102 in the example for illustration, including the following steps 202 to step 216. Among them:

[0079] Step 202, for the entity device, the sample monitoring data collected in the preset historical period is obtained.

[0080] Among them, the entity device can be an entity device used in civil industrial processes such as chemical production.

[0081] In some embodiments, the sample monitoring data of the entity device is collected in the preset historical period, and the multivariate state space model of the entity device is as shown in formula (1) and formula (2).

[0082]

[0083]

[0084] Among them, the data matrix X , A is the state matrix, C is the output matrix, x(k) is the state of the entity device at time k, y(k) is the output of the entity device at time k, w(k) and e(k) are modeling errors obeying non-Gaussian distribution.

[0085] Step 204, for the historical time in the preset historical period, the sample monitoring data collected before the historical time and in the preset historical period is taken as the first sample monitoring data; the sample monitoring data collected after the historical time and in the preset historical period is taken as the second sample monitoring data.

[0086] In some embodiments, a historical time is set in a preset historical period, q sample monitoring data y(k) collected before the historical time and within the preset historical period are taken as first sample monitoring data, and q sample monitoring data y(k) collected after the historical time and within the preset historical period are taken as second sample monitoring data.

[0087] In step 206, a first Hankel matrix is determined based on the first sample monitoring data, and a second Hankel matrix is determined based on the second sample monitoring data.

[0088] The Hankel matrix refers to a matrix in which elements on each sub-diagonal line are equal.

[0089] In some embodiments, the first sample monitoring data and the second sample monitoring data are standardized respectively, a plurality of first vectors are constructed according to the standardized first sample monitoring data, a plurality of second vectors are constructed according to the standardized second sample monitoring data, a first average vector of the plurality of first vectors and a second average vector of the plurality of second vectors are calculated, the first Hankel matrix is determined according to the plurality of first vectors and the first average vector, and the second Hankel matrix is determined according to the plurality of second vectors and the second average vector. .

[0090] In step 208, a covariance matrix is calculated based on the first Hankel matrix and the second Hankel matrix, and a singular value decomposition is performed on the covariance matrix to obtain a conversion matrix.

[0091] The covariance matrix is a matrix describing the mutual relationship between characteristics of multi-dimensional data, and the singular value decomposition is to decompose a matrix into a product of an orthogonal matrix, a diagonal matrix and an orthogonal matrix.

[0092] In some embodiments, the covariance matrix and the cross-covariance matrix are calculated according to the first Hankel matrix and the second Hankel matrix , and the specific calculation formulas are shown in formulas (3), (4) and (5).

[0093]

[0094]

[0095]

[0096] N = n - 2q + 1, n is the number of samples, is the transpose matrix of , and ​For the transpose matrix of V.

[0097] The standardized variable matrix of the historical data is obtained by decomposing the covariance matrix through singular values, and the specific calculation is shown in formula (6), and the conversion matrix is further obtained, and the specific calculation is shown in formula (8).

[0098]

[0099]

[0100]

[0101] wherein, , , the transpose matrix of V.

[0102] In step 210, the third sample monitoring data collected at the current time and the fourth sample monitoring data collected in the corresponding period of the preset time length before the current time are obtained.

[0103] In some embodiments, the third sample monitoring data at the current time and the fourth sample monitoring data in the corresponding period of the preset time length before the current time are collected.

[0104] In step 212, the standardized variable matrix is calculated according to the conversion matrix, the third sample monitoring data and the fourth sample monitoring data.

[0105] In some embodiments, the third sample monitoring data and the fourth sample monitoring data are standardized respectively; a plurality of third vectors are constructed according to the standardized third sample monitoring data and the standardized fourth sample monitoring data, and a fourth vector is constructed according to the plurality of third vectors, and a third average vector of the plurality of third vectors is calculated; the fourth vector is subtracted from the third average vector, multiplied by the conversion matrix, and the standardized variable matrix is obtained.

[0106] In step 214, the information increment matrix and the first average value of the element values in the information increment matrix are calculated based on the standardized variable matrix and the third sample monitoring data.

[0107] In some embodiments, the preprocessed standardized variable matrix is obtained based on the difference between each element value in the standardized variable matrix and the average value of the element values in the row to which each element value belongs; and the first local matrix at the current time and the second local matrix at the next time of the current time are determined according to the third sample monitoring data and the preprocessed standardized variable matrix.

[0108] According to the vectors corresponding to the overlapping time between the first local matrix and the second local matrix, a third local matrix is constructed, and a third average vector corresponding to the vectors in the third local matrix is calculated; according to the first local matrix, the second local matrix and the third average vector, a first covariance matrix at the current time and a second covariance matrix at the next time of the current time are calculated; the second covariance matrix is subtracted by the first covariance matrix to obtain an information increment matrix, and a first mean value of the elements in the information increment matrix is calculated, and the specific calculation formula of the first mean value is shown as (9).

[0109]

[0110] wherein, is the information increment matrix.

[0111] In step 216, a dynamic threshold is calculated according to the first mean value and a preset threshold coefficient, and it is judged whether the entity device is faulty according to the relationship between the dynamic threshold and the first mean value.

[0112] In some embodiments, the dynamic threshold is calculated according to the first mean value and a preset threshold coefficient, and the specific calculation formula is shown as (10) and (11). When there is a dynamic threshold smaller than the first mean value, the entity device is in a faulty state; when all the dynamic thresholds are greater than the first mean value, the entity device is in a normal state.

[0113]

[0114]

[0115] wherein, is a different threshold coefficient, and generally ; represents the first mean value of the elements in the information increment matrix at s time.

[0116] In the above fault detection method, the standardized variable matrix of the third monitoring data and the fourth monitoring data of the real-time sample is calculated, and then the standardized variable matrix is taken as the input to introduce the information increment matrix of the improved covariance matrix. Then the mean value of the matrix is calculated, and the dynamic threshold is set according to the state (normal or fault) of the previous sample. Finally, the current working state is judged by comparing the mean value and the dynamic threshold, which combines the dynamic change of the process data, so that the detection result of the fault is more accurate.

[0117] In an exemplary embodiment, based on the first sample monitoring data, the first Hankel matrix is determined, and based on the second sample monitoring data, the second Hankel matrix is determined, including: respectively normalizing the first sample monitoring data and the second sample monitoring data; constructing a plurality of first vectors according to the normalized first sample monitoring data, and constructing a plurality of second vectors according to the normalized second sample monitoring data; calculating a first average vector of the plurality of first vectors and a second average vector of the plurality of second vectors, and determining the first Hankel matrix according to the plurality of first vectors and the first average vector; and determining the second Hankel matrix according to the plurality of second vectors and the second average vector.

[0118] In some embodiments, the first sample monitoring data and the second sample monitoring data are respectively normalized, a plurality of first vectors are constructed according to the normalized first sample monitoring data, and a plurality of second vectors are constructed according to the normalized second sample monitoring data, the plurality of first vectors constitute The plurality of second vectors constitute The first average vector of the plurality of first vectors is calculated and the second average vector of the plurality of second vectors is calculated Each first vector is subtracted by the first average vector to constitute an element in the first Hankel matrix, and the specific calculation formula is shown as formula (14); each second vector is subtracted by the second average vector to constitute an element in the second Hankel matrix, and the specific calculation formula is shown as formula (15).

[0119]

[0120]

[0121]

[0122]

[0123] The first Hankel matrix and the second Hankel matrix are obtained by synthesizing each element in the first Hankel matrix and each element in the second Hankel matrix , and the specific calculation formulas are shown as formula (16) and formula (17).

[0124]

[0125]

[0126] In the above embodiment, by standardizing the sample monitoring data, the influence of data of different units is avoided, and the dimensional effect of multiple variables is eliminated. The conversion matrix is obtained by calculating the Hankel matrix of the past sample monitoring data, so that the result of using the conversion matrix to judge is more accurate.

[0127] In some example embodiments, the standardized third sample monitoring data and the standardized fourth sample monitoring data are used to construct a plurality of third vectors, and a fourth vector is constructed according to the plurality of third vectors, and a third average vector of the plurality of third vectors is calculated.

[0128] In some embodiments, the standardized third sample monitoring data and the standardized fourth sample monitoring data are used to construct a plurality of third vectors, and a fourth vector is constructed according to the plurality of third vectors, and a third average vector of the plurality of third vectors is calculated. The fourth vector is subtracted from the third average vector, and multiplied by the conversion matrix to obtain the standardized variable matrix. The specific calculation formula is shown in equation (18).

[0129]

[0130] wherein, is the fourth vector representing a total of q third sample monitoring data and fourth sample monitoring data collected in a corresponding period of a preset time length from the current time, represents the standardized variable matrix of the current time data.

[0131] In the above embodiment, the standardized variable matrix is obtained by combining the conversion matrix obtained from the historical data with the monitoring data of the current time, so that the standardized variable matrix is more in line with the actual situation, and the final result is more accurate.

[0132] In some example embodiments, based on the standardized variable matrix and the third sample monitoring data, an information increment matrix is calculated, including: based on the difference between each element value in the standardized variable matrix and the average value of the element values in the row to which each element value belongs, a preprocessed standardized variable matrix is obtained; according to the third sample monitoring data and the preprocessed standardized variable matrix, a first local matrix of the current time and a second local matrix of the next time of the current time are determined; and according to the first local matrix and the second local matrix, the information increment matrix is calculated.

[0133] ​In practice, Given an m*n matrix, obtain mean vector The specific calculation formula is shown in (19). The mean vector is calculated based on the average value of each element in the normalized variable matrix and the average value of the elements in the row to which each element belongs. To determine the degree of difference between them, the normalized variable matrix is ​​preprocessed, and the specific preprocessing formula is shown in (20).

[0134]

[0135]

[0136] in, , for transpose, This is the preprocessed normalized variable matrix.

[0137] In the third sample monitoring data, select monitoring data of L normal samples with a duration of L, and obtain the first local matrix at the current time and the second local matrix at the next time. The specific expression formulas are shown in (21) and (22).

[0138]

[0139]

[0140] in, This represents the first local matrix at the current moment. represent Monitoring data collected in real time This represents the second local matrix of the next time step from the current time step.

[0141] Furthermore, the information increment matrix is ​​calculated based on the first local matrix and the second local matrix.

[0142] In the above embodiments, data from multiple samples and multiple time points are fully considered when calculating the local matrix, making the final result more accurate.

[0143] In some exemplary embodiments, calculating the information increment matrix based on the first local matrix and the second local matrix includes: constructing a third local matrix based on the vectors corresponding to the overlapping time between the first local matrix and the second local matrix, and calculating the third average vector corresponding to the vectors in the third local matrix; calculating the first covariance matrix at the current time and the second covariance matrix at the next time based on the first local matrix, the second local matrix and the third average vector; and subtracting the first covariance matrix from the second covariance matrix to obtain the information increment matrix.

[0144] In practice, the first and second local matrices are compared, and a third local matrix is ​​constructed based on the vectors corresponding to the overlapping moments between them. Calculate the third average vector corresponding to the vector in the third local matrix. The specific calculation formula is shown in (23).

[0145]

[0146] in, .

[0147] Based on the first local matrix, the second local matrix, and the third average vector, calculate the first covariance matrix at the current time and the second covariance matrix at the next time. The specific calculation formulas are shown in (24) and (25).

[0148]

[0149]

[0150] in, Let be the first covariance matrix at the current time. Let be the second covariance matrix of the next time step from the current time step. The mean vector at the current moment is specifically expressed as: Similarly, the mean vector of the next time step can be obtained from the current time step. For ease of calculation, , .

[0151] Subtracting the first covariance matrix from the second covariance matrix yields the information increment matrix. The specific calculation formula is shown in (26).

[0152]

[0153] In the above embodiments, by calculating the information increment matrix of the monitoring data, a large amount of information in the monitoring data is presented in matrix form, making subsequent processing simpler and more efficient.

[0154] In some exemplary embodiments, determining whether a physical device is faulty based on the relationship between a dynamic threshold and a first mean includes: when a dynamic threshold is less than the first mean, the physical device is in a faulty state; when all dynamic thresholds are greater than the first mean, the physical device is in a normal state.

[0155] In practice, when , When a dynamic threshold is less than the first mean, the physical device is in a fault state. , All the dynamic values are greater than the first average value, the entity device is in a normal state.

[0156] In the above embodiment, the state of the entity device is determined by multiple dynamic thresholds, so that the final determination result is more accurate, and the occurrence of false determination is reduced.

[0157] To illustrate the fault detection method in the present application, an embodiment is described below, and the specific flowchart is shown in FIG. 3. The present application is described by way of example with a specific fault detection method in a specific scenario.

[0158] First, sample monitoring data of the entity device is collected within a preset historical period, and a multivariate state space model of the entity device is shown in formulas (1) and (2). A historical time is set in the preset historical period, q sample monitoring data y(k) collected before the historical time and within the preset historical period are taken as first sample monitoring data, and q sample monitoring data y(k) collected after the historical time and within the preset historical period are taken as second sample monitoring data.

[0159] The first sample monitoring data and the second sample monitoring data are standardized respectively, a plurality of first vectors are constructed according to the standardized first sample monitoring data, and a plurality of second vectors are constructed according to the standardized second sample monitoring data, the plurality of first vectors constitute a first Hankel matrix, and the plurality of second vectors constitute a second Hankel matrix. The specific representation is shown in formula (12), the plurality of second vectors constitute The specific representation is shown in formula (13), a first average vector of the plurality of first vectors is calculated and a second average vector of the plurality of second vectors is calculated Each first vector is subtracted from the first average vector to constitute an element in the first Hankel matrix, and the specific calculation formula is shown in formula (14), and each second vector is subtracted from the second average vector to constitute an element in the second Hankel matrix, and the specific calculation formula is shown in formula (15).

[0160] The elements in the first Hankel matrix and the elements in the second Hankel matrix are integrated to obtain a first Hankel matrix and a second Hankel matrix , and the specific calculation formulas are shown in formulas (16) and (17).

[0161] The first Hankel matrix and the second Hankel matrix , the covariance matrix and the cross-covariance matrix are calculated respectively, and the specific calculation formulas are shown in formulas (3), (4) and (5). The covariance matrix is decomposed by singular value, and the specific calculation is shown in formula (6), to obtain the standardized variable matrix of the historical data, and the specific calculation is shown in formula (7), so as to further obtain the conversion matrix, and the specific calculation is shown in formula (8).

[0162] The third sample monitoring data and the fourth sample monitoring data are standardized, a plurality of third vectors are constructed according to the standardized third sample monitoring data and the standardized fourth sample monitoring data, and a fourth vector is constructed according to the plurality of third vectors , a third average vector of the plurality of third vectors is calculated , the fourth vector is subtracted from the third average vector, and multiplied by the conversion matrix to obtain the standardized variable matrix , and the specific calculation formula is shown in formula (18).

[0163] Let be an m*n matrix, and the mean vector of is obtained , and the specific calculation formula is shown in formula (19). According to the difference between each element value in the standardized variable matrix and the average value of the element values in the row to which each element value belongs, that is, the mean vector , the standardized variable matrix is preprocessed, and the specific preprocessing formula is shown in formula (20). In the third sample monitoring data, the monitoring data of L normal samples is selected, and the first local matrix at the current time and the second local matrix at the next time of the current time are obtained, and the specific expression formulas are shown in formulas (21) and (22). The first local matrix and the second local matrix are compared, and a third local matrix is constructed according to the corresponding vectors of the overlapping time between the two, and the third average vector of the vectors in the third local matrix is calculated, and the specific calculation formula is shown in formula (23).

[0164] According to the first local matrix, the second local matrix and the third average vector, the first covariance matrix at the current time and the second covariance matrix at the next time of the current time are calculated, and the specific calculation formulas are shown in formulas (24) and (25). The second covariance matrix is subtracted from the first covariance matrix to obtain the information increment matrix , and the specific calculation formula is shown in formula (26), and the first mean of the elements in the information increment matrix is calculated, and the specific calculation formula of the first mean is shown in formula (9).

[0165] According to the first mean and the preset threshold coefficient, the dynamic threshold is calculated, and the specific calculation formulas are shown in formulas (10) and (11). When , there is a dynamic threshold less than the first mean, the entity device is in a fault state, and when 、 All the dynamics in the middle are greater than the first average, and the entity device is in a normal state.

[0166] The application fully considers the shortcomings of obtaining a control limit of a statistic through distribution on the basis of an existing fault detection method, and starts from the perspective of reflecting dynamic changes of process data in real time to achieve a fault detection method that is more accurate and can reflect real-time changes of data, thereby improving the accuracy of fault detection.

[0167] It should be understood that although each step in the flowchart involved in each embodiment as described above is displayed in sequence according to the direction of the arrow, these steps are not necessarily executed in sequence according to the direction of the arrow. Unless explicitly stated herein, there is no strict order limitation for the execution of these steps, and these steps can be executed in other orders. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or stages, which are not necessarily executed at the same time but can be executed at different times, and the execution order of these steps or stages is not necessarily sequential but can be executed in rotation or alternation with at least part of other steps or steps or stages in other steps.

[0168] Based on the same inventive concept, the embodiments of the application also provide a fault detection device for implementing the above-mentioned fault detection method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, and therefore the specific limitations in one or more fault detection device embodiments provided below can refer to the limitations of the fault detection method in the foregoing, which will not be described here again.

[0169] In one exemplary embodiment, as shown in Figure 4 a fault detection device is provided, comprising: an acquisition module 401, a determination module 402, a calculation module 403, and a judgment module 404, wherein:

[0170] The acquisition module is configured to acquire, for an entity device, sample monitoring data collected within a preset historical period.

[0171] The acquisition module is further configured to, for a historical time within the preset historical period, collect sample monitoring data before the historical time and within the preset historical period as first sample monitoring data, and collect sample monitoring data after the historical time and within the preset historical period as second sample monitoring data.

[0172] The determination module is configured to determine a first Hankel matrix based on the first sample monitoring data, and determine a second Hankel matrix based on the second sample monitoring data.

[0173] The determination module is further configured to calculate a covariance matrix based on the first Hankel matrix and the second Hankel matrix, and perform singular value decomposition on the covariance matrix to obtain a conversion matrix.

[0174] The acquisition module is further configured to acquire third sample monitoring data collected at a current time and fourth sample monitoring data collected within a preset time period before the current time.

[0175] The calculation module is configured to calculate a standardized variable matrix based on the conversion matrix, the third sample monitoring data, and the fourth sample monitoring data.

[0176] The calculation module is further configured to calculate an information increment matrix and a first average value of element values in the information increment matrix based on the standardized variable matrix and the third sample monitoring data.

[0177] The determination module is further configured to calculate a dynamic threshold value based on the first average value and a preset threshold coefficient, and determine whether the entity device is faulty based on a relationship between the dynamic threshold value and the first average value.

[0178] In some embodiments, the determination module is further configured to perform standardization processing on the first sample monitoring data and the second sample monitoring data respectively, construct a plurality of first vectors based on the standardized first sample monitoring data, and construct a plurality of second vectors based on the standardized second sample monitoring data, calculate a first average vector of the plurality of first vectors and a second average vector of the plurality of second vectors, determine the first Hankel matrix based on the plurality of first vectors and the first average vector, and determine the second Hankel matrix based on the plurality of second vectors and the second average vector.

[0179] In some embodiments, the calculation module is further configured to perform standardization processing on the third sample monitoring data and the fourth sample monitoring data respectively, construct a plurality of third vectors based on the standardized third sample monitoring data and the standardized fourth sample monitoring data, construct a fourth vector based on the plurality of third vectors, and calculate a third average vector of the plurality of third vectors, and obtain the standardized variable matrix by subtracting the third average vector from the fourth vector and multiplying the result by the conversion matrix.

[0180] In some embodiments, the calculation module is further configured to obtain a preprocessed standardized variable matrix based on a difference between each element value in the standardized variable matrix and an average value of element values in a row to which the element value belongs, determine a first local matrix at the current time and a second local matrix at a next time of the current time based on the third sample monitoring data and the preprocessed standardized variable matrix, and calculate the information increment matrix based on the first local matrix and the second local matrix.

[0181] In some embodiments, the calculation module is further configured to construct a third local matrix according to vectors corresponding to overlapping time points between the first local matrix and the second local matrix, and calculate a third average vector corresponding to the vectors in the third local matrix; calculate a first covariance matrix at the current time point and a second covariance matrix at the next time point of the current time point according to the first local matrix, the second local matrix and the third average vector; and obtain an information increment matrix by subtracting the first covariance matrix from the second covariance matrix.

[0182] In some embodiments, the judgment module is further configured to determine that the entity device is in a fault state when the dynamic threshold values are all greater than the first average value.

[0183] The modules in the fault detection apparatus can be implemented in whole or in part by software, hardware and combinations thereof. The modules can be embedded in or independent of a processor in a computer device in hardware form, or stored in a memory in the computer device in software form, so as to be called and executed by a processor to perform operations corresponding to the modules.

[0184] In an exemplary embodiment, a computer device is provided, which can be a server, and an internal structure diagram of the computer device can be as shown in Figure 5 The computer device includes a processor, a memory, an input / output interface (I / O), a communication interface, a display unit and an input device. 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. The processor of the computer device is configured to provide calculation 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 running the operating system and the computer program in the non-volatile storage medium. The database of the computer device is configured to store sample monitoring data. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to communicate with external terminals through network connection. The computer program is executed by the processor to implement a fault detection method.

[0185] The display unit of the computer device is configured 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 overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0186] Those skilled in the art can understand that, Figure 5 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0187] In one exemplary embodiment, a computer device is provided, comprising a memory and a processor, the memory storing a computer program, and the processor implementing the following steps when executing the computer program:

[0188] For the entity device, sample monitoring data collected in a preset historical period is acquired;

[0189] For a historical time in the preset historical period, sample monitoring data collected before the historical time and in the preset historical period is taken as first sample monitoring data, and sample monitoring data collected after the historical time and in the preset historical period is taken as second sample monitoring data;

[0190] Based on the first sample monitoring data, a first Hankel matrix is determined, and based on the second sample monitoring data, a second Hankel matrix is determined;

[0191] Based on the first Hankel matrix and the second Hankel matrix, a covariance matrix is calculated, and singular value decomposition is performed on the covariance matrix to obtain a conversion matrix;

[0192] Third sample monitoring data collected at a current time and fourth sample monitoring data collected in a preset time length corresponding to the period before the current time are acquired;

[0193] According to the conversion matrix, the third sample monitoring data and the fourth sample monitoring data, a canonical variable matrix is calculated;

[0194] Based on the canonical variable matrix and the third sample monitoring data, an information increment matrix and a first mean value of element values in the information increment matrix are calculated;

[0195] According to the first mean value and a preset threshold coefficient, a dynamic threshold is calculated, and according to the relationship between the dynamic threshold and the first mean value, it is judged whether the entity device is faulty.

[0196] In one embodiment, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the following steps:

[0197] For the entity device, sample monitoring data collected in a preset historical period is acquired;

[0198] For a historical time within the preset historical period, sample monitoring data collected before the historical time and within the preset historical period is taken as first sample monitoring data, and sample monitoring data collected after the historical time and within the preset historical period is taken as second sample monitoring data;

[0199] Based on the first sample monitoring data, a first Hankel matrix is determined, and based on the second sample monitoring data, a second Hankel matrix is determined;

[0200] Based on the first Hankel matrix and the second Hankel matrix, a covariance matrix is calculated, and singular value decomposition is performed on the covariance matrix to obtain a conversion matrix;

[0201] Third sample monitoring data collected at a current time and fourth sample monitoring data collected in a preset time length corresponding period before the current time are acquired;

[0202] According to the conversion matrix, the third sample monitoring data and the fourth sample monitoring data, a canonical variable matrix is calculated;

[0203] Based on the canonical variable matrix and the third sample monitoring data, an information increment matrix and a first mean value of element values in the information increment matrix are calculated;

[0204] According to the first mean value and a preset threshold coefficient, a dynamic threshold is calculated, and according to the relationship between the dynamic threshold and the first mean value, it is judged whether the entity device is faulty.

[0205] In one embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the following steps:

[0206] For the entity device, sample monitoring data collected in a preset historical period is acquired;

[0207] For a historical time within the preset historical period, sample monitoring data collected before the historical time and within the preset historical period is taken as first sample monitoring data, and sample monitoring data collected after the historical time and within the preset historical period is taken as second sample monitoring data;

[0208] Based on the first sample monitoring data, a first Hankel matrix is determined, and based on the second sample monitoring data, a second Hankel matrix is determined;

[0209] Based on the first Hankel matrix and the second Hankel matrix, a covariance matrix is calculated, singular value decomposition is performed on the covariance matrix, and a conversion matrix is obtained;

[0210] The third sample monitoring data collected at the current time and the fourth sample monitoring data collected in a preset time period before the current time are obtained.

[0211] According to the conversion matrix, the third sample monitoring data and the fourth sample monitoring data, a standard variable matrix is calculated.

[0212] Based on the standard variable matrix and the third sample monitoring data, an information increment matrix and a first mean value of element values in the information increment matrix are calculated.

[0213] According to the relationship between the first mean value and a preset threshold coefficient, a dynamic threshold is calculated, and according to the relationship between the dynamic threshold and the first mean value, it is judged whether the entity device is faulty.

[0214] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0215] Those skilled 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. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to 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. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, 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. The volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0216] The technical features of the above embodiments can be combined in any manner. To make the description concise, not all possible combinations of the technical features in the above embodiments are described, but as long as the combinations of the technical features do not exist contradictions, they should be considered as the scope of the present application.

[0217] The above-described embodiments are merely illustrative of several embodiments of the present application, and the description is relatively specific and detailed, but should not be understood as a limitation on the scope of the patent. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of protection of the present application should be subject to the appended claims.

Claims

1. A fault detection method characterized by, The method comprises: For an entity device, sample monitoring data collected in a preset historical period is acquired; For a historical time in the preset historical period, sample monitoring data collected before the historical time and in the preset historical period is taken as first sample monitoring data, and sample monitoring data collected after the historical time and in the preset historical period is taken as second sample monitoring data; Based on the first sample monitoring data, a first Hankel matrix is determined, and based on the second sample monitoring data, a second Hankel matrix is determined; Based on the first Hankel matrix and the second Hankel matrix, a covariance matrix is calculated, singular value decomposition is performed on the covariance matrix, and a conversion matrix is obtained; Third sample monitoring data collected at a current time and fourth sample monitoring data collected in a preset time length corresponding period before the current time are acquired; According to the conversion matrix, the third sample monitoring data and the fourth sample monitoring data, a canonical variable matrix is calculated; Based on the canonical variable matrix and the third sample monitoring data, an information increment matrix and a first average value of element values in the information increment matrix are calculated; According to the first average value and a preset threshold coefficient, a dynamic threshold is calculated, and according to the relationship between the dynamic threshold and the first average value, whether the entity device is faulty is determined.

2. The method of claim 1, wherein, The method comprises: The first sample monitoring data and the second sample monitoring data are standardized respectively; According to the standardized first sample monitoring data, a plurality of first vectors are constructed, and according to the standardized second sample monitoring data, a plurality of second vectors are constructed; A first average vector of the plurality of first vectors and a second average vector of the plurality of second vectors are calculated, the first Hankel matrix is determined according to the plurality of first vectors and the first average vector, and the second Hankel matrix is determined according to the plurality of second vectors and the second average vector.

3. The method of claim 1, wherein, The method comprises: The third sample monitoring data and the fourth sample monitoring data are standardized respectively; According to the standardized third sample monitoring data and the standardized fourth sample monitoring data, a plurality of third vectors are constructed, a fourth vector is constructed according to the plurality of third vectors, and a third average vector of the plurality of third vectors is calculated; The fourth vector is subtracted from the third average vector, multiplied by the conversion matrix, and the canonical variable matrix is obtained.

4. The method of claim 1, wherein, The method comprises: Based on the difference between each element value in the canonical variable matrix and the average value of element values in the row to which each element value belongs, a preprocessed canonical variable matrix is acquired; According to the third sample monitoring data and the preprocessed canonical variable matrix, a first local matrix at the current time and a second local matrix at the next time of the current time are determined; According to the first local matrix and the second local matrix, an information increment matrix is calculated.

5. The method of claim 4, wherein, The calculating of the information increment matrix according to the first local matrix and the second local matrix comprises: According to vectors corresponding to overlapping time between the first local matrix and the second local matrix, a third local matrix is constructed, and a third average vector corresponding to vectors in the third local matrix is calculated; According to the first local matrix, the second local matrix and the third average vector, a first covariance matrix at a current time and a second covariance matrix at a next time of the current time are calculated; The second covariance matrix is subtracted by the first covariance matrix to obtain the information increment matrix.

6. The method of claim 1, wherein, The judging of whether the entity device is faulty according to the relationship between the dynamic threshold and the first average value comprises: When the dynamic threshold is less than the first average value, the entity device is in a faulty state; When the dynamic threshold is greater than the first average value, the entity device is in a normal state.

7. A fault detection apparatus characterized by comprising: The apparatus comprises: An acquisition module is configured to acquire sample monitoring data collected in a preset historical period for an entity device; The acquisition module is further configured to, for a historical time in the preset historical period, acquire sample monitoring data collected before the historical time and in the preset historical period as first sample monitoring data, and acquire sample monitoring data collected after the historical time and in the preset historical period as second sample monitoring data; A determination module is configured to determine a first Hankel matrix based on the first sample monitoring data, and determine a second Hankel matrix based on the second sample monitoring data; The determination module is further configured to calculate a covariance matrix based on the first Hankel matrix and the second Hankel matrix, and perform singular value decomposition on the covariance matrix to obtain a conversion matrix; The acquisition module is further configured to acquire third sample monitoring data collected at a current time, and fourth sample monitoring data collected in a preset time period before the current time; A calculation module is configured to calculate a canonical variable matrix according to the conversion matrix, the third sample monitoring data and the fourth sample monitoring data; The calculation module is further configured to calculate an information increment matrix and a first average value of element values in the information increment matrix based on the canonical variable matrix and the third sample monitoring data; A judgment module is configured to calculate a dynamic threshold according to the first average value and a preset threshold coefficient, and judge whether the entity device is faulty according to the relationship between the dynamic threshold and the first average value.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor executes the computer program to implement the steps of the method in any one of claims 1 to 6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 6.

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

Images