Device monitoring method, apparatus and device

Abstract

Embodiments of the present application provide a device monitoring method, device and apparatus. The method comprises: obtaining a plurality of sets of first device parameters of a device to be monitored at a plurality of times, each set of first device parameters comprising a plurality of types of device parameters; performing decomposition processing on the plurality of sets of first device parameters to obtain a plurality of first device parameter sequences, one of the first device parameter sequences comprising one type of device parameter; performing spectral residual processing on the plurality of first device parameter sequences to obtain target spectral information corresponding to the plurality of first device parameter sequences; determining abnormal data according to the target spectral information, and determining a monitoring result corresponding to the device to be monitored according to the abnormal data. The accuracy of device monitoring is improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a device monitoring method, apparatus, and device. Background Technology

[0002] During the operation of an industrial system, the equipment running in the system can be monitored to ensure the normal operation of the industrial system.

[0003] In related technologies, equipment is monitored by acquiring equipment parameters in chronological order and monitoring time points in the parameter sequence that deviate from the normal data range. The operating status of the equipment is determined based on the parameters corresponding to the abnormal time points in the parameter sequence. However, equipment parameters may be affected by external factors, making it impossible to accurately determine the abnormal time points using the above method, resulting in low accuracy in equipment monitoring. Summary of the Invention

[0004] This application provides a device, apparatus, and equipment for monitoring equipment, in order to solve the problem of low accuracy in equipment monitoring.

[0005] In a first aspect, embodiments of this application provide a device monitoring method, including:

[0006] Acquire multiple sets of first device parameters of the device under monitoring at multiple times, with each set of first device parameters including multiple types of device parameters;

[0007] The multiple sets of first equipment parameters are decomposed to obtain multiple first equipment parameter sequences, and each first equipment parameter sequence includes one type of equipment parameter;

[0008] Spectral residual processing is performed on the plurality of first device parameter sequences to obtain the target spectral information corresponding to the plurality of first device parameter sequences;

[0009] Based on the target spectrum information, abnormal data is determined, and based on the abnormal data, the monitoring result corresponding to the device to be monitored is determined.

[0010] In one possible implementation, the plurality of first device parameters are decomposed to obtain a plurality of device parameter sequences, including:

[0011] The multiple sets of first equipment parameters are normalized and whitened to obtain multiple sets of second equipment parameters.

[0012] In each group of second equipment parameters, determine the equipment parameters for each parameter type;

[0013] The plurality of device parameter sequences are determined based on the device parameters of each parameter type in each group of second device parameters.

[0014] In one possible implementation, for any set of second device parameters; determining device parameters for each parameter type within the set of second device parameters includes:

[0015] A third set of device parameters is obtained by processing the set of second device parameters and the preset basis vectors using a preset algorithm.

[0016] The set of third device parameters are projected onto the preset basis vector to obtain device parameters for each parameter type.

[0017] In one possible implementation, for any first device parameter sequence, spectral residual processing is performed on the first device parameter sequence to obtain target spectral information corresponding to the first device parameter sequence, including:

[0018] The data within a preset time range in the first device parameter sequence are determined as the second device parameter sequence;

[0019] The target spectrum information is obtained by performing wavelet transform processing, separation spectrum residual processing, and inverse wavelet transform processing on the second device parameter sequence.

[0020] In one possible implementation, the second device parameter sequence is subjected to wavelet transform processing, spectral residual separation processing, and inverse wavelet transform processing to obtain the target spectral information, including:

[0021] Determine the predicted value corresponding to the second equipment parameter sequence;

[0022] A preset number of predicted values ​​are added to the end of the second device parameter sequence to obtain a third device parameter sequence;

[0023] The wavelet transform is applied to the third device parameter sequence to obtain the frequency domain parameter sequence;

[0024] The frequency domain parameter sequence is subjected to the separated spectral residual processing to obtain the spectral residual;

[0025] The target spectral information is obtained by performing the inverse wavelet transform on the spectral residual.

[0026] In one possible implementation, the target spectral information includes multiple spectral values; determining abnormal data based on the target spectral information includes:

[0027] Obtain the average spectral value of the neighboring spectral values ​​of each spectral value in the target spectral information;

[0028] Based on the adjacent average spectral values ​​corresponding to each spectral value, it is determined whether the spectral value is abnormal data, so as to identify the abnormal data.

[0029] In one possible implementation, for any spectral value in the target spectral information; determining whether the spectral value is abnormal data based on the adjacent average spectral values ​​of the spectral value includes:

[0030] Obtain the ratio between the spectral value and the adjacent average spectral value;

[0031] If the ratio is greater than or equal to a preset threshold, the spectral value is determined to be the abnormal data;

[0032] If the ratio is less than the preset threshold, then the spectral value is determined not to be abnormal data.

[0033] Secondly, embodiments of this application provide a monitoring device, the device comprising:

[0034] The acquisition module is used to acquire multiple sets of first device parameters of the device under monitoring at multiple times, and each set of first device parameters includes multiple types of device parameters.

[0035] The first processing module decomposes the multiple sets of first device parameters to obtain multiple first device parameter sequences, wherein each first device parameter sequence includes a type of device parameter.

[0036] The second processing module performs spectral residual processing on the plurality of first device parameter sequences to obtain target spectral information corresponding to the plurality of first device parameter sequences;

[0037] The determination module determines abnormal data based on the target spectrum information, and determines the monitoring result corresponding to the device to be monitored based on the abnormal data.

[0038] In one possible implementation, the first processing module is specifically used for:

[0039] The multiple sets of first equipment parameters are normalized and whitened to obtain multiple sets of second equipment parameters.

[0040] In each group of second equipment parameters, determine the equipment parameters for each parameter type;

[0041] The plurality of device parameter sequences are determined based on the device parameters of each parameter type in each group of second device parameters.

[0042] In one possible implementation, the first processing module is specifically used for:

[0043] A third set of device parameters is obtained by processing the set of second device parameters and the preset basis vectors using a preset algorithm.

[0044] The set of third device parameters are projected onto the preset basis vector to obtain device parameters for each parameter type.

[0045] In one possible implementation, the second processing module is specifically used for:

[0046] The data within a preset time range in the first device parameter sequence are determined as the second device parameter sequence;

[0047] The target spectrum information is obtained by performing wavelet transform processing, separation spectrum residual processing, and inverse wavelet transform processing on the second device parameter sequence.

[0048] In one possible implementation, the second processing module is specifically used for:

[0049] Determine the predicted value corresponding to the second equipment parameter sequence;

[0050] A preset number of predicted values ​​are added to the end of the second device parameter sequence to obtain a third device parameter sequence;

[0051] The wavelet transform is applied to the third device parameter sequence to obtain the frequency domain parameter sequence;

[0052] The frequency domain parameter sequence is subjected to the separated spectral residual processing to obtain the spectral residual;

[0053] The target spectral information is obtained by performing the inverse wavelet transform on the spectral residual.

[0054] In one possible implementation, the determining module is specifically used for:

[0055] Obtain the average spectral value of the neighboring spectral values ​​of each spectral value in the target spectral information;

[0056] Based on the adjacent average spectral values ​​corresponding to each spectral value, it is determined whether the spectral value is abnormal data, so as to identify the abnormal data.

[0057] In one possible implementation, the determining module is specifically used for:

[0058] Obtain the ratio between the spectral value and the adjacent average spectral value;

[0059] If the ratio is greater than or equal to a preset threshold, the spectral value is determined to be the abnormal data;

[0060] If the ratio is less than the preset threshold, then the spectral value is determined not to be abnormal data.

[0061] Thirdly, embodiments of this application provide a monitoring device, including:

[0062] At least one processor; and

[0063] A memory communicatively connected to the at least one processor; wherein,

[0064] The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the method described in any of the first aspects.

[0065] Fourthly, embodiments of this application provide a non-transitory computer-readable storage medium storing computer instructions, characterized in that the computer instructions are used to cause the computer to perform the method according to any one of the first aspects.

[0066] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the method according to any one of the first aspects.

[0067] The equipment monitoring method, apparatus, and device provided in this application decompose multiple sets of equipment operating parameters input at multiple times to determine equipment parameters of the same type at multiple times, resulting in a sequence of equipment parameters of the same type. The sequence of equipment parameters is then subjected to general residual processing to obtain a target spectrum. Based on the target spectrum information, abnormal data is identified. Based on the abnormal data, the operating status of the equipment to be monitored is determined. In the above process, multiple sets of equipment parameters at multiple times are decomposed into equipment parameters of the same type. Based on the equipment parameter sequence composed of equipment parameters of the same type at multiple times, the general residual processing can accurately process the equipment parameter sequence to obtain the target spectrum. Furthermore, based on the target spectrum, the time point of the anomaly can be accurately determined, improving the accuracy of equipment monitoring. Attached Figure Description

[0068] Figure 1 A schematic diagram illustrating the application scenarios provided in the embodiments of this application;

[0069] Figure 2 A schematic flowchart illustrating the device monitoring method provided in this application embodiment;

[0070] Figure 3 This is a schematic diagram of target spectrum information provided in an embodiment of this application;

[0071] Figure 4 A schematic diagram of the process for decomposing multiple sets of first device parameters provided in the embodiments of this application;

[0072] Figure 5 This is a schematic flowchart of a method for determining target spectrum information corresponding to a first device parameter sequence, provided in an embodiment of this application.

[0073] Figure 6 A schematic diagram illustrating the process of determining the second device parameter sequence provided in this application embodiment;

[0074] Figure 7 A schematic diagram illustrating the process of determining the parameter sequence of the third device provided in this application embodiment;

[0075] Figure 8 This is a schematic diagram of the monitoring device provided in the embodiments of this application;

[0076] Figure 9 This is a schematic diagram of the monitoring device provided in an embodiment of this application. Detailed Implementation

[0077] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0078] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0079] Figure 1 This is a schematic diagram illustrating an application scenario provided in an embodiment of this application. Please refer to [link / reference]. Figure 1 The system includes a monitoring device 101 and a device to be monitored 102. The monitoring device 101 can be a computer, server, etc. The device to be monitored 102 can be a machine tool, transformer, etc. The monitoring device 101 can determine the operating status of the device to be monitored 102 based on the operating parameters of the device to be monitored 102 at multiple times.

[0080] In related technologies, operating equipment is monitored by acquiring its parameters chronologically and detecting time points in the parameter sequence that deviate from the normal data range. The operating status of the equipment is then determined based on the parameters corresponding to these abnormal time points. However, equipment parameters may be affected by external factors, making it impossible to accurately determine the abnormal time points using the above method, resulting in low accuracy in equipment monitoring.

[0081] In this embodiment, multiple sets of equipment parameters of the device under monitoring at multiple times are acquired, and each set of equipment parameters may include multiple types of equipment parameters. The multiple sets of equipment operating parameters input at multiple times are decomposed to determine the same type of equipment parameters at multiple times, resulting in a sequence of equipment parameters of the same type. This sequence of equipment parameters is then subjected to general residual processing to obtain a target spectrum. Based on the target spectrum information, abnormal data is identified. Based on the abnormal data, the operating status of the device under monitoring is determined. In the above process, multiple sets of equipment parameters at multiple times are decomposed into the same type of equipment parameters. Based on the sequence of equipment parameters composed of the same type of equipment parameters at multiple times, the general residual processing can accurately process the equipment parameter sequence to obtain the target spectrum. Furthermore, based on the target spectrum, the time point of the anomaly can be accurately determined, improving the accuracy of equipment monitoring.

[0082] The method described in this application will now be illustrated through specific embodiments. It should be noted that the following embodiments may exist independently or in combination with each other; identical or similar content will not be repeated in different embodiments.

[0083] Figure 2 This is a schematic flowchart illustrating the device monitoring method provided in an embodiment of this application. Please refer to [link / reference]. Figure 2 The method may include:

[0084] S201. Obtain multiple sets of first device parameters of the device to be monitored at multiple times.

[0085] The executing entity in this application embodiment can be a monitoring device or a monitoring apparatus installed within the monitoring device. The monitoring apparatus can be implemented through software or through a combination of software and hardware.

[0086] The equipment to be monitored can be any equipment that is currently in operation. For example, the equipment to be monitored can be a transformer, a water pump, a machine tool, etc.

[0087] Equipment parameters can be parameters related to the operation of the equipment being monitored. For example, equipment parameters can include the temperature, voltage, current, and rotational speed of the equipment being monitored.

[0088] Each group of first equipment parameters includes multiple types of equipment parameters.

[0089] For example, assuming the device to be tested is a transformer, the transformer's parameters may include temperature T, voltage V, and current A. Assuming a set of first device parameters is acquired every 1 second, then 5 sets of first device parameters can be acquired within 5 seconds. The specific details of the 5 sets of first device parameters are shown in Table 1.

[0090] Table 1

[0091]

[0092] S202. Decompose multiple sets of first equipment parameters to obtain multiple first equipment parameter sequences.

[0093] One of the first device parameter sequences includes one type of device parameter.

[0094] A sequence of device parameters includes multiple types of device parameters, and each type of device parameter can be an independent component. For example, assuming the device to be tested is a transformer, the device parameters of a transformer can include temperature T, voltage V, and current A. A first sequence of device parameters can be T1, V1, and A1, and an independent component of a sequence of device parameters can be T1.

[0095] Multiple sets of first equipment parameters can be decomposed as follows: normalize and whiten the multiple sets of first equipment parameters to obtain multiple sets of second equipment parameters; determine the equipment parameters of each parameter type in each set of second equipment parameters; and determine multiple equipment parameter sequences based on the equipment parameters of each parameter type in each set of second equipment parameters.

[0096] The first equipment parameter sequence includes parameters of the same parameter type from the equipment to be monitored, and the parameters in the first equipment parameter sequence are arranged in chronological order.

[0097] For example, assuming the device to be tested is a transformer, the transformer's parameters may include temperature T, voltage V, and current A. Then the transformer has three parameter sequences: a temperature parameter sequence, a voltage parameter sequence, and a current parameter sequence. The specific sequences of these three parameters over n time points are shown in Table 2.

[0098] Table 2

[0099] time 1 2 …… n-1 n Temperature parameter sequence <![CDATA[T1]]> <![CDATA[T2]]> …… <![CDATA[T n-1 ]]> <![CDATA[T n ]]> Voltage parameter sequence <![CDATA[V1]]> <![CDATA[V2]]> …… <![CDATA[V n-1 ]]> <![CDATA[V n ]]> Current parameter sequence <![CDATA[A1]]> <![CDATA[A2]]> …… <![CDATA[A n-1 ]]> <![CDATA[A n ]]>

[0100] Optionally, anomalous parameters can be determined based on the first sequence of device parameters. The kurtosis coefficient of each device parameter is calculated using the following formula:

[0101] k = E(y 4 )-3E(y 2 ) 2

[0102] The equipment parameter with the largest absolute value of the kurtosis coefficient among all equipment parameters is identified as the abnormal equipment parameter.

[0103] S203. Perform spectral residual processing on multiple first device parameter sequences to obtain target spectral information corresponding to multiple first device parameter sequences.

[0104] Multiple first device parameter sequences can be processed using residual processing as follows: data within a preset time range in the first device parameter sequence are determined as the second device parameter sequence; wavelet transform processing, spectral separation residual processing, and inverse wavelet transform processing are performed on the second device parameter sequence to obtain the target spectrum information.

[0105] Wavelet processing can transform the sequence of second device parameters after Fourier transform from the frequency domain to the time domain.

[0106] The general residual can be processed as follows: Smooth the second device parameter sequence after wavelet transform to determine the background of the second device parameter sequence. The difference between the second device parameter sequence and its background is determined as the general residual.

[0107] The residuals are processed by inverse wavelet transform to convert the time-domain data into frequency-domain data, thus obtaining the target spectrum.

[0108] S204. Determine abnormal data based on the target spectrum information, and determine the monitoring results corresponding to the device to be monitored based on the abnormal data.

[0109] Abnormal data can be identified as follows: obtain the average spectral value of the adjacent spectral values ​​of each spectral value in the target spectral information; determine whether the spectral value is abnormal based on the average spectral value of the adjacent spectral values ​​corresponding to each spectral value, thereby identifying abnormal data.

[0110] Below, in conjunction with Figure 3 The adjacent spectral values ​​of any spectral value in the target spectral information are described. Figure 3 This is a schematic diagram of target spectral information provided in an embodiment of this application. Point A in the target spectrum can be any point in the target spectrum. Based on the adjacent range of the time series corresponding to point A, the spectral value corresponding to each adjacent point is determined. Based on the spectral value of each adjacent point, the average spectral value of the adjacent points is determined. The preset range can be determined according to the application scenario of the device to be monitored.

[0111] For any spectral value in the target spectral information, obtain the ratio between the spectral value and the adjacent average spectral value; if the ratio is greater than or equal to a preset threshold, the spectral value is determined to be abnormal data; if the ratio is less than the preset threshold, the spectral value is determined not to be abnormal data.

[0112] Based on the ratios, determine the statistical distribution pattern of all recent ratios. Then, automatically adjust the preset threshold based on this statistical distribution pattern.

[0113] The device monitoring method provided in this application acquires multiple sets of first device parameters of the device to be monitored at multiple times, each set of first device parameters including multiple types of device parameters; decomposes the multiple sets of first device parameters to obtain multiple sequences of first device parameters; performs spectral residual processing on the multiple sequences of first device parameters to obtain target spectral information corresponding to the multiple sequences of first device parameters; determines abnormal data based on the target spectral information, and determines the monitoring result corresponding to the device to be monitored based on the abnormal data. In the above process, multiple sets of device parameters at multiple times are decomposed into device parameters of the same type. Based on the device parameter sequences composed of device parameters of the same type at multiple times, the target spectrum can be accurately processed by spectral residual processing to obtain the target spectrum, and then the time point of the anomaly can be accurately determined based on the target spectrum, thereby improving the accuracy of device monitoring.

[0114] Based on any of the above embodiments, the following, in conjunction with Figure 4 The process of decomposing multiple sets of first equipment parameters (S202) will be explained.

[0115] Figure 4 This is a schematic flowchart illustrating the process of decomposing multiple sets of first device parameters provided in the embodiments of this application. Please refer to... Figure 4 The method may include:

[0116] S401. Normalize multiple sets of first equipment parameters to obtain the first matrix.

[0117] The normalization process for multiple sets of first device parameters can be performed as follows: Based on the multiple sets of first device parameters, determine the initial matrix corresponding to each set of first device parameters. The number of rows in the matrix corresponds to the number of parameter types, and the number of columns corresponds to the number of sets of first device parameters. Calculate the average value of each row in the initial matrix. Subtract the average value of each row from each data point in the initial matrix to obtain the first difference. The new matrix formed by these first differences is the first matrix obtained after normalization.

[0118] For example, assuming the device to be tested is a transformer, the transformer's parameters may include temperature T, voltage V, and current A. If a set of first device parameters is acquired every 1 second, then 3 sets of first device parameters can be acquired within 3 seconds. The specific details of the 3 sets of first device parameters are shown in Table 3.

[0119] Table 3

[0120]

[0121] Based on Table 3, the initial matrices corresponding to multiple sets of first device parameters can be determined as follows:

[0122]

[0123] Each column in the initial matrix forms a column vector representing the first device parameters.

[0124] In the matrix corresponding to multiple sets of first device parameters, the average value of the first row of data is E1 = (T1 + V2 + V3) / 3, the average value of the second row of data is E2 = (V1 + A2 + T3) / 3, and the average value of the third row of data is E3 = (A1 + T2 + A3) / 3. Therefore, the first matrix after normalization can be:

[0125]

[0126] S402. Perform a whitening operation on the first matrix to obtain multiple sets of second device parameters.

[0127] The second equipment parameters can be obtained as follows: Perform singular value decomposition on the covariance matrix corresponding to the first matrix, and determine each element in the matrix corresponding to the second equipment parameters using the following formula;

[0128] x2=ED -1 / 2 E T x

[0129] Where E is the covariance matrix corresponding to the first matrix, D is the diagonal matrix corresponding to the covariance matrix, x is an element in the first matrix, and x2 is an element in the matrix corresponding to the second device parameters.

[0130] The matrix formed by the new elements determined after each set of singular value decomposition is used to define multiple sets of second device parameters. In the matrix formed by the new elements, each column forms a column vector, which represents a set of second device parameters.

[0131] The processing procedure for each group of second equipment parameters is the same across multiple groups. The following explanation will take the processing procedure for any one group of second equipment parameters as an example.

[0132] S403. A set of second device parameters and a set of preset basis vectors are processed by a preset algorithm to obtain a set of third device parameters.

[0133] The preset basis vector can be any unit column vector, and the number of rows in the preset basis vector is the same as the number of parameter types. For example, if a device to be monitored has two parameter types, its corresponding preset basis vector can be...

[0134] The preset algorithm can be Newton's iterative formula, specifically:

[0135] w + =E(xg(w) T x))-E(g′(w T x))w

[0136] Where w is a preset basis vector, x is a column vector composed of the second set of device parameters, and E is the averaging function. g is the function G(w T x)=-exp(-(w T x) 2 The derivative of ( / 2). Where exp is an exponential function with the natural constant e as its base.

[0137] Optionally, normalization can be performed using w / |w| during the iteration process.

[0138] According to the preset algorithm, any set of third device parameters obtained is a column vector.

[0139] S404. Project the third set of device parameters into preset basis vectors to obtain device parameters for each parameter type.

[0140] The column vector corresponding to the third device parameter can be projected onto the preset basis vector using the following formula:

[0141] f = xcosθ

[0142] Where f is the device parameter, x is the column vector corresponding to the third device parameter, and θ is the angle between the column vector corresponding to the third device parameter and the preset basis vector.

[0143] Based on the dimension of the preset basis vectors, determine the number of column vectors corresponding to any set of third device parameters that are projected onto the preset basis vectors. The number of device parameters obtained after projection has the same dimension as the preset basis vectors. For example, if the dimension of the preset basis vectors is 3, then the number of device parameters obtained after projection is 3.

[0144] Optionally, the number of projected device parameters can be adjusted as follows: The iterated basis vectors form a mixing matrix G. The number of projected device parameters is determined by the ratio of the determinant |G| to the norm ||G|| of the mixing matrix. It is then determined whether the number of projected device parameters is within a preset range. If so, the projected device parameters are defined as the device parameters for each parameter type. If not, the number of projected device parameters needs to be adjusted according to the ratio, and the iteration is repeated to determine the device parameters for each parameter type based on the adjusted number of device parameters.

[0145] Where the norm ||G|| = trace(GTG)¹ / ². `trace` is a specified function used to calculate the sum of the elements on the diagonal of the square matrix.

[0146] For example, if the preset range is 0.1 to 1, and the number of projected device parameters is 3, the determinant of the mixed matrix |G| and the norm ||G|| = trace(GTG). 1 / 2If the ratio is 0.5, then there are 3 device parameters. If the ratio of the determinant of the mixed matrix |G| to the norm ||G|| is 0.05, then the number of device parameters after projection needs to be adjusted to 2, and the iteration needs to be repeated to obtain 2 device parameters.

[0147] S405. Determine multiple device parameter sequences based on the device parameters of each parameter type.

[0148] Multiple sets of third-party device parameters are projected onto preset basis vectors to obtain device parameters for each parameter type in the multiple sets of third-party device parameters. Based on the device parameters of each parameter type corresponding to the multiple sets of third-party device parameters, they are sorted in chronological order to determine the parameter sequence of multiple devices.

[0149] For example, assuming the device to be tested is a transformer, the transformer's parameters may include temperature T, voltage V, and current A. Then, the transformer has three parameter sequences at n time points: a temperature parameter sequence, a voltage parameter sequence, and a current parameter sequence, each sequence containing n data points. These three parameter sequences are detailed in Table 4.

[0150] Table 4

[0151] Temperature parameter sequence <![CDATA[T1]]> <![CDATA[T2]]> …… <![CDATA[T n-1 ]]> <![CDATA[T n ]]> Voltage parameter sequence <![CDATA[V1]]> <![CDATA[V2]]> …… <![CDATA[V n-1 ]]> <![CDATA[V n ]]> Current parameter sequence <![CDATA[A1]]> <![CDATA[A2]]> …… <![CDATA[A n-1 ]]> <![CDATA[A n ]]>

[0152] The method for determining multiple device parameter sequences provided in this application decomposes each group of device parameters through normalization, whitening, and projection processing to determine each type of device parameter within each group. Based on each type of device parameter, a parameter sequence for multiple devices is determined. In this process, multivariate parameters can be accurately decomposed into univariate parameters.

[0153] Based on any of the above embodiments, since the process of determining the target spectrum corresponding to each first device parameter sequence is the same, for any first device parameter sequence, the following, in conjunction with... Figure 5 The detailed process of determining the target spectrum information corresponding to the first device parameter sequence is explained.

[0154] Figure 5 This is a schematic flowchart illustrating a method for determining the target spectrum information corresponding to a first device parameter sequence, provided in an embodiment of this application. Please refer to... Figure 5 The method includes:

[0155] S501. Perform a Fourier transform on the first device parameter sequence.

[0156] The first device sequence can be a temperature sequence, a voltage sequence, etc.

[0157] After performing a Fourier transform on the first device parameter sequence, the low-frequency data sequence after the transform can also be removed.

[0158] The parameter sequence for low-frequency equipment can be determined based on experience or the operating characteristics of the equipment. The low-frequency equipment parameter sequence can be a sequence of equipment parameters within a fixed period. For example, the low-frequency equipment parameter sequence can be for one day, one week, etc.

[0159] S502. The data within the preset time range in the first device parameter sequence after Fourier transform are determined as the second device parameter sequence.

[0160] The preset time range is determined based on the monitoring scenario. For example, the preset time range can be 2 minutes.

[0161] Optionally, the second device parameter sequence can be divided in real time based on the latest time monitored by the monitoring equipment.

[0162] Below, in conjunction with Figure 6 The process of determining the second device parameter sequence will be explained. Figure 6 This is a schematic diagram illustrating the process of determining the second device parameter sequence provided in an embodiment of this application. Please refer to... Figure 6 Based on the first equipment parameter sequence, a second equipment parameter sequence is determined within a preset time range.

[0163] S503. Determine the predicted value corresponding to the second equipment parameter sequence.

[0164] The predicted value corresponding to the second equipment parameter sequence can be calculated as follows: Based on the second equipment parameter sequence, determine the average slope of several points from the end of the second equipment parameter sequence; based on the average slope, determine the predicted value corresponding to the second equipment parameter sequence.

[0165] The average slope of several points from the end of the second device parameter sequence can be calculated using the following formula:

[0166]

[0167] Where x is the value of any point from the end of the second device parameter sequence, and m is the number of points from the end of the second device parameter sequence. This represents the average slope.

[0168] The number of points from the end of the second device parameter sequence is a preset value. For example, the number of points from the end of the second device parameter sequence can be 10.

[0169] The predicted value corresponding to the second equipment parameter sequence can be calculated using the following formula:

[0170]

[0171] Where x is the predicted value corresponding to the second device parameter sequence, x n-m+1For any point at the end of the second device parameter sequence, is the average slope, and m is the number of points from the end of the second device parameter sequence.

[0172] S504. Add a preset number of predicted values ​​to the end of the second equipment parameter sequence to obtain the third equipment parameter sequence.

[0173] Below, in conjunction with Figure 7 The process of determining the parameter sequence of the third device is explained. Figure 7 This is a schematic diagram illustrating the process of determining the parameter sequence of the third device provided in an embodiment of this application. Please refer to [link / reference]. Figure 7 The endpoint of the second parameter sequence is determined as point B. Based on the predicted values ​​corresponding to the second equipment parameter sequence and the preset quantity, a preset number of predicted values ​​are added to the end of the second equipment parameter sequence to obtain the third parameter sequence.

[0174] The preset quantity is the quantity preset by the monitoring equipment. The time interval corresponding to the increased predicted value is the same as the time interval in the second equipment parameter sequence.

[0175] For example, the second device parameter sequence is a sequence of device parameters within 1s to 10s, and the predicted value corresponding to the second device parameter sequence is 3, with a preset quantity of 5. Then, at the device parameter corresponding to 10s, 5 data points with a predicted value of 3 are added according to the time interval in the second device parameter sequence. The device parameter sequence formed by the second device sequence and these 5 data points is the third device parameter sequence.

[0176] S505. Perform wavelet transform on the third device parameter sequence to obtain the frequency domain parameter sequence.

[0177] Wavelet processing can transform the sequence of third-party device parameters from the frequency domain to the time domain.

[0178] S506. Perform spectral residual processing on the frequency domain parameter sequence to obtain the spectral residual.

[0179] The general residual can be processed as follows: Smooth the second device parameter sequence after wavelet transform to determine the background of the second device parameter sequence. The difference between the second device parameter sequence and its background is determined as the general residual.

[0180] Smoothing can be achieved using the following formula:

[0181] AL(k)=h q (k)*L(k)

[0182] Among them, h q (k) is the specified function, L(k) is the second device parameter sequence, and AL(k) is the background of the second device parameter sequence.

[0183] S507. Perform the wavelet inverse transform on the spectral residual to obtain the target spectral information.

[0184] Based on the generalized residual, the generalized residual is transformed from time-domain information to frequency-domain information through inverse wavelet transform. The target spectral information is then obtained from the frequency-domain information obtained after the generalized residual transformation.

[0185] The method for determining the target spectrum information corresponding to a first device parameter sequence provided in this application involves performing a Fourier transform on the first device sequence to remove low-frequency data sequences, resulting in a second device sequence. The second device sequence is then subjected to residual processing to obtain the target spectrum information. In this process, residual processing can reduce data fluctuations caused by periodic factors, thereby improving the accuracy of monitoring.

[0186] Figure 8 This is a schematic diagram of the monitoring device provided in an embodiment of this application. Please refer to... Figure 8 The monitoring device 10 may include:

[0187] The acquisition module 11 is used to acquire multiple sets of first device parameters of the device to be monitored at multiple times, and each set of first device parameters includes multiple types of device parameters;

[0188] The first processing module 12 decomposes the multiple sets of first device parameters to obtain multiple first device parameter sequences, wherein each first device parameter sequence includes a type of device parameter.

[0189] The second processing module 13 performs spectral residual processing on the plurality of first device parameter sequences to obtain target spectral information corresponding to the plurality of first device parameter sequences;

[0190] The determination module 14 determines abnormal data based on the target spectrum information, and determines the monitoring result corresponding to the device to be monitored based on the abnormal data.

[0191] The monitoring device provided in this application embodiment can execute the technical solution shown in the above method embodiment. Its implementation principle and beneficial effects are similar, and will not be described again here.

[0192] In one possible implementation, the first processing module 12 is specifically used for:

[0193] The multiple sets of first equipment parameters are normalized and whitened to obtain multiple sets of second equipment parameters.

[0194] In each group of second equipment parameters, determine the equipment parameters for each parameter type;

[0195] The plurality of device parameter sequences are determined based on the device parameters of each parameter type in each group of second device parameters.

[0196] In one possible implementation, the first processing module 12 is specifically used for:

[0197] A third set of device parameters is obtained by processing the set of second device parameters and the preset basis vectors using a preset algorithm.

[0198] The set of third device parameters are projected onto the preset basis vector to obtain device parameters for each parameter type.

[0199] In one possible implementation, the second processing module 13 is specifically used for:

[0200] The data within a preset time range in the first device parameter sequence are determined as the second device parameter sequence;

[0201] The target spectrum information is obtained by performing wavelet transform processing, separation spectrum residual processing, and inverse wavelet transform processing on the second device parameter sequence.

[0202] In one possible implementation, the second processing module 13 is specifically used for:

[0203] Determine the predicted value corresponding to the second equipment parameter sequence;

[0204] A preset number of predicted values ​​are added to the end of the second device parameter sequence to obtain a third device parameter sequence;

[0205] The wavelet transform is applied to the third device parameter sequence to obtain the frequency domain parameter sequence;

[0206] The frequency domain parameter sequence is subjected to the separated spectral residual processing to obtain the spectral residual;

[0207] The target spectral information is obtained by performing the inverse wavelet transform on the spectral residual.

[0208] In one possible implementation, the determining module 14 is specifically used for:

[0209] Obtain the average spectral value of the neighboring spectral values ​​of each spectral value in the target spectral information;

[0210] Based on the adjacent average spectral values ​​corresponding to each spectral value, it is determined whether the spectral value is abnormal data, so as to identify the abnormal data.

[0211] In one possible implementation, the determining module 14 is specifically used for:

[0212] Obtain the ratio between the spectral value and the adjacent average spectral value;

[0213] If the ratio is greater than or equal to a preset threshold, the spectral value is determined to be the abnormal data;

[0214] If the ratio is less than the preset threshold, then the spectral value is determined not to be abnormal data.

[0215] The monitoring device provided in this application embodiment can execute the technical solution shown in the above method embodiment. Its implementation principle and beneficial effects are similar, and will not be described again here.

[0216] Figure 9 This is a schematic diagram of the monitoring device provided in an embodiment of this application. Please refer to... Figure 5 The monitoring device 20 may include a memory 21 and a processor 22. Exemplarily, the memory 21 and the processor 22 are interconnected via a bus 23.

[0217] Memory 21 is used to store program instructions;

[0218] The processor 22 is used to execute the program instructions stored in the memory, so that the monitoring device 20 performs the method shown in the above method embodiment.

[0219] The monitoring device provided in this application embodiment can execute the technical solution shown in the above method embodiment. Its implementation principle and beneficial effects are similar, and will not be described again here.

[0220] This application provides a computer-readable storage medium storing computer-executable instructions, which are used to implement the above-described method when executed by a processor.

[0221] This application embodiment may also provide a computer program product, including a computer program that, when executed by a processor, can implement the above-described method.

[0222] All or part of the steps in the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a readable memory. When the program is executed, it performs the steps of the above method embodiments; and the aforementioned memory (storage medium) includes: read-only memory (ROM), RAM, flash memory, hard disk, solid-state drive, magnetic tape, floppy disk, optical disk, and any combination thereof.

[0223] This application describes embodiments with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processing unit of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processing unit of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0224] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0225] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0226] Obviously, those skilled in the art can make various modifications and variations to the embodiments of this application without departing from the spirit and scope of this application. Therefore, if these modifications and variations to the embodiments of this application fall within the scope of the claims of this application and their equivalents, this application also intends to include these modifications and variations.

[0227] In this application, the term "comprising" and its variations can refer to non-limiting inclusion; the term "or" and its variations can refer to "and / or". The terms "first", "second", etc., in this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. In this application, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

Claims

1. A method for monitoring equipment, characterized in that, include: Acquire multiple sets of first device parameters of the device under monitoring at multiple times, with each set of first device parameters including multiple types of device parameters; The multiple sets of first equipment parameters are decomposed to obtain multiple first equipment parameter sequences, and each first equipment parameter sequence includes one type of equipment parameter; Spectral residual processing is performed on the plurality of first device parameter sequences to obtain the target spectral information corresponding to the plurality of first device parameter sequences; Abnormal data is determined based on the target spectrum information, and the monitoring result corresponding to the device to be monitored is determined based on the abnormal data. The multiple sets of first equipment parameters are decomposed to obtain multiple equipment parameter sequences, including: The multiple sets of first equipment parameters are normalized and whitened to obtain multiple sets of second equipment parameters. In each group of second equipment parameters, determine the equipment parameters for each parameter type; The sequence of multiple device parameters is determined based on the device parameters of each parameter type in each group of second device parameters; For any given sequence of first device parameters; perform spectral residual processing on the first device parameter sequence to obtain the target spectral information corresponding to the first device parameter sequence, including: The data within a preset time range in the first device parameter sequence are determined as the second device parameter sequence; The target spectrum information is obtained by performing wavelet transform processing, separation spectrum residual processing, and inverse wavelet transform processing on the second device parameter sequence.

2. The method according to claim 1, characterized in that, For any set of second device parameters; determine the device parameters for each parameter type in the set of second device parameters, including: A third set of device parameters is obtained by processing the set of second device parameters and the preset basis vectors using a preset algorithm. The set of third device parameters are projected onto the preset basis vector to obtain device parameters for each parameter type.

3. The method according to claim 1, characterized in that, The target spectrum information is obtained by performing wavelet transform processing, spectral separation residual processing, and inverse wavelet transform processing on the second device parameter sequence, including: Determine the predicted value corresponding to the second equipment parameter sequence; A preset number of predicted values ​​are added to the end of the second device parameter sequence to obtain a third device parameter sequence; The wavelet transform is applied to the third device parameter sequence to obtain the frequency domain parameter sequence; The frequency domain parameter sequence is subjected to the separated spectral residual processing to obtain the spectral residual; The target spectral information is obtained by performing the inverse wavelet transform on the spectral residual.

4. The method according to any one of claims 1-3, characterized in that, The target spectral information includes multiple spectral values; determining abnormal data based on the target spectral information includes: Obtain the average spectral value of the neighboring spectral values ​​of each spectral value in the target spectral information; Based on the adjacent average spectral values ​​corresponding to each spectral value, it is determined whether the spectral value is abnormal data, so as to identify the abnormal data.

5. The method according to claim 4, characterized in that, For any spectral value in the target spectral information; Determining whether a spectral value is abnormal based on its adjacent average spectral values ​​includes: Obtain the ratio between the spectral value and the adjacent average spectral value; If the ratio is greater than or equal to a preset threshold, the spectral value is determined to be the abnormal data; If the ratio is less than the preset threshold, then the spectral value is determined not to be abnormal data.

6. A monitoring device, characterized in that, The device includes: The acquisition module is used to acquire multiple sets of first device parameters of the device under monitoring at multiple times, and each set of first device parameters includes multiple types of device parameters. The first processing module decomposes the multiple sets of first device parameters to obtain multiple first device parameter sequences, wherein each first device parameter sequence includes a type of device parameter. The second processing module performs spectral residual processing on the plurality of first device parameter sequences to obtain target spectral information corresponding to the plurality of first device parameter sequences; The determination module determines abnormal data based on the target spectrum information, and determines the monitoring result corresponding to the device to be monitored based on the abnormal data; The first processing module is specifically used to perform normalization and whitening operations on the multiple sets of first device parameters to obtain multiple sets of second device parameters; determine the device parameters of each parameter type in each set of second device parameters; and determine the multiple device parameter sequences based on the device parameters of each parameter type in each set of second device parameters. The second processing module is specifically used to determine the data within a preset time range in any first device parameter sequence as a second device parameter sequence; and to perform wavelet transform processing, separation spectrum residual processing, and inverse wavelet transform processing on the second device parameter sequence to obtain the target spectrum information.

7. A monitoring device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, in, The computer instructions are used to cause the computer to perform the method according to any one of claims 1 to 5.

Images