CN111144267B - Equipment running state detection method and device, storage medium and computer equipment - Google Patents
Equipment running state detection method and device, storage medium and computer equipment Download PDFInfo
- Publication number
- CN111144267B CN111144267B CN201911336947.9A CN201911336947A CN111144267B CN 111144267 B CN111144267 B CN 111144267B CN 201911336947 A CN201911336947 A CN 201911336947A CN 111144267 B CN111144267 B CN 111144267B
- Authority
- CN
- China
- Prior art keywords
- original signal
- matrix
- singular
- signal sequence
- values
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Pure & Applied Mathematics (AREA)
- Evolutionary Biology (AREA)
- Mathematical Optimization (AREA)
- Health & Medical Sciences (AREA)
- Mathematical Analysis (AREA)
- Biomedical Technology (AREA)
- Computational Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Signal Processing (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The application relates to a method, a device, a storage medium and a computer device for detecting the running state of equipment, on one hand, the method is based on all original signal values of an original signal sequence of target equipment, then based on the corresponding relation between singular values and the running state, the corresponding running state is determined according to the singular values obtained by the signal matrix, so that more comprehensive running state information of the equipment can be obtained; on the other hand, the method can prevent the generated signal matrix from being highly correlated by adding the random number on the basis of the original signal value, thereby preventing a pathological equation set from occurring in the singular value decomposition process, ensuring that the obtained singular value does not approach 0, namely the obtained singular value has reference significance, and effectively improving the accuracy of the operation state determination result when the corresponding operation state is determined according to the obtained singular value.
Description
Technical Field
The present application relates to the field of signal processing technologies, and in particular, to a method and apparatus for detecting an operating state of a device, a storage medium, and a computer device.
Background
With the development of scientific technology, the types of equipment are more and more, and the method has important significance for detecting the running state of the equipment.
In the prior art, the detection of the state of the device is generally performed according to signals generated during the operation of the device, and a more common detection means is to determine whether the device has an operation fault by extracting time domain features (such as a mean value, a root mean square value, a peak value, a kurtosis value) of the signals, frequency domain features (such as a mean square frequency, a root mean square frequency and a frequency variance of a frequency domain), or time-frequency domain features (such as wavelet packet energy of a time-frequency domain) and the like. However, the above detection means relies only on the extracted partial features to perform the device state detection, and the accuracy of the detection result is low.
Disclosure of Invention
In view of the foregoing, it is desirable to provide an apparatus, a device, a storage medium, and a computer apparatus for detecting an operating state of an apparatus, which are advantageous for improving the accuracy of detecting the state of the apparatus.
A method for detecting an operating state of a device, comprising:
acquiring an original signal sequence generated by target equipment to be detected in the operation process, wherein the original signal sequence comprises a plurality of original signal values;
generating a signal matrix based on the original signal sequence, wherein each element in the signal matrix comprises a sum of a single original signal value and a random number in a preset range;
singular value decomposition is carried out on the signal matrix to obtain a plurality of singular values;
and carrying out state classification on the singular values based on the corresponding relation between the singular values and the running states to obtain the running state information of the target equipment.
An apparatus for detecting an operating state of a device, comprising:
the signal acquisition module is used for acquiring an original signal sequence generated in the operation process of target equipment to be detected, wherein the original signal sequence comprises a plurality of original signal values;
the matrix generation module is used for generating a signal matrix based on the original signal sequence, wherein each element in the signal matrix is the sum of a single original signal value and a random number in a preset range;
the singular value decomposition module is used for carrying out singular value decomposition on the signal matrix to obtain a plurality of singular values;
and the state classification module is used for carrying out state classification on the singular values based on the corresponding relation between the singular values and the running states to obtain the running state information of the target equipment.
A computer device comprising a memory storing a computer program and a processor implementing the steps of the method described above when the processor executes the computer program.
A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the above method.
According to the method, the device, the storage medium and the computer equipment for detecting the running state of the equipment, on one hand, the embodiment of the application is based on all original signal values of the original signal sequence of the target equipment, then the corresponding running state is determined according to the singular value obtained by the signal matrix based on the corresponding relation between the singular value and the running state, namely, the running state of the equipment is determined by taking all characteristics in the running process of the equipment into consideration, so that more comprehensive running state information of the equipment can be obtained; on the other hand, the method can prevent the generated signal matrix from being highly correlated by adding the random number on the basis of the original signal value, thereby preventing a pathological equation set from occurring in the singular value decomposition process, ensuring that the obtained singular value does not approach 0, namely the obtained singular value has reference significance, and effectively improving the accuracy of the operation state determination result when the corresponding operation state is determined according to the obtained singular value.
Drawings
FIG. 1 is a flow chart of a method for detecting an operating state of a device in one embodiment;
fig. 2 (a) is a waveform diagram of an original signal sequence of length 2000;
fig. 2 (b) is a first row data diagram of a signal matrix of 1001 x 1000 size (1001 rows, 1000 columns) generated from the original signal sequence in fig. 2 (a);
fig. 2 (c) is a data diagram of a second row in a signal matrix of 1001 x 1000 size (1001 rows, 1000 columns) generated from the original signal sequence in fig. 2 (a);
FIG. 2 (d) is a graph of singular values from a signal matrix;
FIG. 3 is a flow diagram of generating a signal matrix based on an original signal sequence in one embodiment;
FIG. 4 is a flow chart of generating a signal matrix based on an original signal sequence according to another embodiment;
FIG. 5 is a flow chart of a method for classifying states of a plurality of singular values based on a correspondence between the singular values and the operating states to obtain operating state information of a device to be detected in one embodiment;
FIG. 6 is a flow chart of a method for detecting an operating state of a device according to another embodiment;
FIG. 7 is a schematic diagram of a device operation state detecting apparatus according to an embodiment;
FIG. 8 is a schematic diagram of a device for detecting an operating state of an apparatus according to another embodiment;
fig. 9 is an internal structural diagram of a computer device in one embodiment.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
In one embodiment, as shown in fig. 1, a method for detecting an operating state of a device is provided, and the method is applied to a processor capable of detecting an operating state of a device for explanation, and mainly includes the following steps:
step S100, an original signal sequence corresponding to a target device to be detected is obtained, wherein the original signal sequence comprises a plurality of original signal values.
The original signal sequence is a signal sequence acquired by a signal acquisition device (particularly, various sensors such as a speed signal sensor and the like) when the target equipment is in an operating state, and the original signal sequence contains original signal values generated by a plurality of targets in the operating process.
When the processor detects the running state of the equipment, the processor can detect the state in real time, namely, the signal acquisition device acquires the original signal sequence of the target equipment in real time when in running and sends the original signal sequence to the processor in real time, and the processor receives the original signal sequence sent by the signal acquisition device in real time and carries out subsequent processing, so that real-time monitoring is realized. Of course, the original signal sequence may be collected in advance and stored in the memory, and when the processor needs to detect the running state of the target device, the original signal sequence corresponding to the target device is directly read from the memory. Of course, the processor may also obtain the original signal sequence from the external device. For example, the original signal sequence of the target device is stored in the cloud, and when the processor needs to detect the running state of the target device, the original signal sequence corresponding to the target device is directly obtained from the cloud. In addition, the external device may be an external storage medium, etc., and the method for acquiring the original signal sequence corresponding to the target device by the processor in this embodiment is not limited.
Step S200, generating a signal matrix based on the original signal sequence, wherein each element in the signal matrix comprises a sum of a single original signal value and a random number within a preset range.
The processor generates a signal matrix based on the original signal sequence after obtaining the original signal sequence. In the prior art, when a signal matrix is generated based on an original signal sequence, original signal values in the original signal sequence are generally directly taken as each element in the signal matrix, however, the signal matrix generated by the prior art is only different from each other by one element every two rows, so that the signal matrix is highly likely to be highly correlated, thereby causing the problem that a disease state equation set can occur when singular values are solved, the disease state equation set can cause that the obtained partial singular values are close to 0, and the partial singular values close to 0 do not belong to singular values with reference significance, thereby losing partial useful information and causing inaccurate detection results.
Specifically, as shown in fig. 2 (a) to 2 (d), a specific example of obtaining singular values by the prior art is shown. Fig. 2 (a) is a waveform diagram of an original signal sequence with a length of 2000; fig. 2 (b) and fig. 2 (c) are data diagrams of a first row and a second row in a signal matrix generated according to the original signal sequence and having a size of 1001×1000 (1001 rows and 1000 columns), respectively; fig. 2 (d) is a graph of singular values obtained from a signal matrix.
Referring to fig. 2 (b) and fig. 2 (c), it can be seen that the data of the first row and the second row are staggered by only one data point, and in fact, two adjacent rows in all 1001 rows are different by only one data point, which can cause a problem of solving a pathological equation set when the matrix of 1001 x 1000 is subjected to singular value decomposition, and as a result, a great part of singular values are very close to 0, that is, a singular value curve as shown in fig. 2 (d) appears.
The application is different from the prior art in that each element in the signal matrix generated by the application is not a single original signal value per se, but the sum of the single original signal value and random numbers in a preset range is used as the element in the signal matrix. The random number is added in the application, so as to add a tiny disturbance on the basis of the original signal value, thereby preventing the signal matrix from generating a high correlation condition, namely preventing the signal matrix from generating a problem of a disease state equation set (the disease state equation set can cause that the obtained partial singular value approaches to 0) due to the fact that only one element is different between every two rows of the signal matrix, and enabling the singular value obtained by solving to be the singular value with reference significance.
In addition, when the random number is added on the basis of the original signal value, the added random number is in a smaller preset range, so that the added random number is ensured not to influence the overall trend of the original signal value, namely the added random number cannot submerge the original signal value. For example, the random number added may be less than 1/2 of the original signal amplitude. In addition, the added random numbers are different for each original signal value (unless the generated random numbers are the same due to coincidence), thereby preventing the problem of the system of pathological equations during singular value decomposition.
And step S300, singular value decomposition is carried out on the signal matrix, and a plurality of singular values are obtained.
After the processor obtains the signal matrix, the processor can obtain a plurality of singular values corresponding to the signal matrix by performing singular value decomposition. In this step, the process of singular value decomposition by the processor may be implemented by an existing singular value decomposition method, which is not limited herein.
Step S400, based on the corresponding relation between the singular values and the running states, carrying out state classification on the singular values to obtain the running state information of the target equipment.
After obtaining the plurality of singular values, the processor can perform state classification on the plurality of singular values corresponding to the signal matrix of the target object according to the pre-established correspondence between the singular values and the operation states, so as to obtain the operation state information of the target device, wherein the operation state information can comprise normal operation, abnormal operation and the like.
In addition, in the pre-established correspondence between the singular value and the running state, the correspondence between different abnormal states and the singular value may be included, so that when the processor determines that the target device is in abnormal running, the processor may further determine what kind of abnormality the target device specifically belongs to, so that the running state detection result of the target device is more detailed and comprehensive. For example, according to the pre-established correspondence between the singular values and the running states, the features of the singular values corresponding to different anomaly types can be extracted, and the correspondence between the anomaly types and the features of the singular values can be established. After the singular value corresponding to the target equipment is obtained, extracting the characteristics of the obtained singular value, and then determining the specific anomaly type corresponding to the singular value corresponding to the target equipment according to the corresponding relation between the anomaly type and the singular value characteristics.
The embodiment provides a device running state detection method, on the one hand, the application is based on all original signal values of an original signal sequence of target device, then based on the corresponding relation between singular values and running states, the corresponding running states are determined according to the singular values obtained by the signal matrix, namely, the application considers all characteristics in the running process of the device to determine the running states of the device, so that more comprehensive device running state information can be obtained; on the other hand, the method can prevent the generated signal matrix from being highly correlated by adding the random number on the basis of the original signal value, thereby preventing a pathological equation set from occurring in the singular value decomposition process, ensuring that the obtained singular value does not approach 0, namely the obtained singular value has reference significance, and effectively improving the accuracy of the operation state determination result when the corresponding operation state is determined according to the obtained singular value.
In one embodiment, the original signal values corresponding to the elements on each of the opposite diagonals in the signal Matrix are equal, and the signal Matrix may be a Hankel Matrix (Hankel Matrix).
Specifically, if the original signal sequence x (t) is a one-dimensional time sequence with a length of N, i.e., x (t) = { x 1 ,x 2 ,…x N Then a Hankel matrix H of order p x q can be constructed as follows p*q ：
Wherein ε i,j For added random numbers, then for matrix H p*q Singular value decomposition is performed to obtain the singular values with min { p, q } greater than zero as the characteristics of the original signal sequence x (t).
Alternatively, to obtain enough features, the values of p, q can be determined by the following formula:
q＝N+1-p
for example, when the signal length N is 2000, p=1000, q=1001, so that 1000 singular values greater than 0 can be obtained as features of the original signal sequence.
In this embodiment, a Hankel matrix is constructed based on a one-dimensional original signal sequence, and then singular value decomposition is performed on the constructed matrix, and the obtained singular value is used as a feature of the original signal sequence.
In one embodiment, as shown in fig. 3, step S200 generates a signal matrix based on the original signal sequence, including steps S212 to S214.
Step S212, each original signal value in the original signal sequence is used as an initial element to generate an initial matrix;
step S214, taking the sum of each initial element and the random number as a new element, and replacing the initial element at the corresponding position in the initial matrix by using the new element to obtain the signal matrix.
Specifically, taking an example that the original signal sequence contains 2000 original signal values, a process of generating a signal matrix by the processor is explained. The signal length n=2000 of the original signal sequence acquired by the processor, i.e. comprising (x 1 ,x 2 ,…x 2000 ) A total of 2000 raw signal values. The processor then uses these 2000 raw signal values as initial elements to generate the following initial matrix:
then, each initial element x m And random number epsilon i,j Sum x of m +ε i,j As a new element, replacing the initial element of the corresponding position, and obtaining the following signal matrix:
in this embodiment, by adding a random number on the basis of the original signal value, the generated signal matrix can be prevented from being highly correlated, so that a disease state equation set is prevented from occurring in the singular value decomposition process, the obtained singular value is ensured not to approach 0, that is, the obtained singular value has a reference meaning, and the accuracy of the operation state determination result can be effectively improved when the corresponding operation state is determined according to the obtained singular value.
In one embodiment, as shown in fig. 4, step S200 generates a signal matrix based on the original signal sequence, including steps S222 to S224.
Step S222, taking the sum of each original signal value and the random number as a new signal value, and replacing the new signal value with the original signal value at the corresponding position in the original signal sequence to obtain a new signal sequence;
in step S224, a signal matrix is generated using each new signal value in the new signal sequence as an element.
Specifically, taking an example that the original signal sequence contains 2000 original signal values, a process of generating a signal matrix by the processor is explained. The signal length n=2000 of the original signal sequence acquired by the processor, i.e. comprising (x 1 ,x 2 ,…x 2000 ) A total of 2000 raw signal values. The processor then processes each initial element x m And random number epsilon i,j Sum x of m +ε i,j As new signal values and replacing the original signal values at the corresponding positions, a new signal sequence (x 1 +ε 1，1 ,x 2 +ε 1，2 ,…x 2000 +ε 1000,1001 ) The processor then uses each new signal value in the new signal sequence as an element to generate the following signal matrix:
in this embodiment, by adding a random number on the basis of the original signal value, the generated signal matrix can be prevented from being highly correlated, so that a disease state equation set is prevented from occurring in the singular value decomposition process, the obtained singular value is ensured not to approach 0, that is, the obtained singular value has a reference meaning, and the accuracy of the operation state determination result can be effectively improved when the corresponding operation state is determined according to the obtained singular value.
In one embodiment, when constructing the signal matrix based on the original signal sequence, the original signal sequence may be sampled at intervals, and then the signal matrix may be generated.
Specifically, taking the interval as an original signal value as an example for explanation, the signal length n=2000 of the original signal sequence obtained by the processor, that is, the signal length n=2000 includes (x 1 ,x 2 ,…x 2000 ) Co-production2000 raw signal values. The processor first samples the original signal sequence at intervals of 1 to obtain a signal sequence (x 1 ,x 3 ,…x 1999 ) Then based on the signal sequence (x 1 ,x 3 ,…x 1999 ) The following signal matrix is generated:
in this embodiment, after the processor obtains the original signal sequence, the processor firstly performs interval sampling to obtain new sequences with the number less than that of the original signal sequence, and then generates the signal matrix based on the new sequences, so that the data volume can be effectively reduced, and the data processing efficiency can be improved.
In one embodiment, specific examples of singular value decomposition based on a signal matrix are provided.
The signal length n=9 of the original signal sequence acquired by the processor includes (1, 2,3, … 8, 9) total 9 original signal values. Under the condition that random numbers are not added, the Hankel matrix obtained according to the original signal sequence is as follows:
after singular value decomposition is carried out on the matrix A, 5 singular values are obtained, and the matrix A is sequentially arranged from big to small: 26.9,1.86,8.37e-16,5.83e-16,4.43e-17. It can be seen that the last 3 singular values are all very close to 0, so that only the first two singular values belong to the valid singular values with reference significance.
In the application, on the basis of a matrix A, each element in the matrix A is respectively added with a random number within a preset range as noise to resist the high correlation of the matrix, and then the matrix after noise addition is subjected to singular value decomposition. For example, taking the preset range of [ -1,1] as an example, on the basis of the matrix a, adding a random number distributed between [ -1,1] to each element, a singular value decomposition result (the random number added is not shown here specifically due to the uncertainty of the random number) is obtained as follows: 27.590,1.713,1.514,0.502,0.103 (3-bit decimal place preserved). It can be seen that the singular values obtained after the addition of the random number all belong to the valid singular values having a reference meaning.
Therefore, compared with a signal matrix without random numbers, the method can extract more effective singular values with reference significance by adding the random numbers in the preset range to each element in the signal matrix on the premise of not changing the main characteristics of the original signal sequence.
In one embodiment, as shown in fig. 5, step S400 performs state classification on a plurality of singular values based on the correspondence between the singular values and the operation states, and obtaining the operation state information of the device to be detected includes steps S420 to S460.
Step S420, singular value sample data and corresponding state categories are obtained, wherein the singular value sample data comprises singular values corresponding to different state categories;
step S440, performing model training based on the singular value sample data and the corresponding state category to obtain a trained state classification model;
step S460, carrying out state classification on the singular values through a state classification model to obtain the running state information of the equipment to be detected.
Specifically, signal sequences of sufficient length in different operation states (normal state and abnormal state) can be acquired through the sensor, and for each signal sequence in the operation state, data length of at least more than 2 periods is intercepted as a sample signal and a corresponding state type label is set. Then, according to the method flow of the previous embodiment of the present application, a corresponding signal matrix is generated based on the sample signal, singular value decomposition is performed on the generated signal matrix, a singular value corresponding to the sample signal is obtained, and a state class label corresponding to the sample signal is used as a state class label of the singular value obtained according to the sample data.
After the singular value sample data and the corresponding state type label are obtained, training the initial model, namely, inputting the singular value sample data into the initial model, calculating a loss value according to a state prediction result output by the initial model and the corresponding state type label, and optimally training the initial model according to the loss value to obtain a trained state classification model. The state classification model includes a neural network, such as DNN (Deep Neural Networks, deep neural network), CNN (Convolutional Neural Networks, convolutional neural network), or RNN (Recurrent Neural Network ), etc. After the trained state classification model is obtained, the state classification of the singular values can be automatically performed through the state classification model.
It can be understood that the state classification model adopted in the embodiment may be trained in advance, and when the state classification of the singular value is required, the trained state classification model is directly called, that is, the step of model training in the application is not necessary.
In the embodiment, the singular values are subjected to state classification through the trained state classification model, so that the classification efficiency and the accuracy of classification results can be improved.
In one embodiment, as shown in fig. 6, the device operation state detection method further includes: step S500, when the running state information of the target equipment contains abnormal running state, outputting corresponding prompt information.
Specifically, when the processor detects that the operation state of the target device is abnormal, the processor may display corresponding prompt information to the user through the display interface, for example, "XX failure occurs in XX device, please note-! "etc., thereby facilitating the user to understand the relevant abnormal situation and solve the problem.
In addition, the processor can also send an alarm instruction to the alarm device, so that the alarm device sends corresponding alarm information after receiving the alarm instruction sent by the processor, and the alarm device can be particularly an audible and visual alarm device and the like, so that a user can be reminded immediately when the abnormal operation of the device is detected.
In one embodiment, the method for detecting the running state of the equipment provided by the application can be particularly applied to equipment such as CT (computed tomography) bulb tubes, bearings, gears, gear boxes, motors, engines, fans, pumps, drills, lathes and the like, so that the accuracy of the running state detection result of the equipment can be improved.
It should be understood that, under reasonable conditions, although the steps in the flowcharts referred to in the foregoing embodiments are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, and the order of execution of the sub-steps or stages is not necessarily sequential, but may be performed in rotation or alternately with at least a portion of the sub-steps or stages of other steps or other steps.
In one embodiment, as shown in fig. 7, there is provided an apparatus for detecting an operating state of a device, the apparatus mainly including the following modules:
the signal acquisition module 100 is configured to acquire an original signal sequence generated by a target device to be detected in an operation process, where the original signal sequence includes a plurality of original signal values;
a matrix generation module 200, configured to generate a signal matrix based on the original signal sequence, where each element in the signal matrix is a sum of a single original signal value and a random number within a preset range;
the singular value decomposition module 300 is configured to perform singular value decomposition on the signal matrix to obtain a plurality of singular values;
the state classification module 400 is configured to perform state classification on the plurality of singular values based on a correspondence between the singular values and the operation states, so as to obtain operation state information of the target device.
The embodiment provides a device for detecting the running state of equipment, on the one hand, the application is based on all original signal values of an original signal sequence of target equipment, then based on the corresponding relation between singular values and the running state, the corresponding running state is determined according to the singular values obtained by the signal matrix, namely, the application considers all characteristics in the running process of the equipment to determine the running state of the equipment, so that more comprehensive running state information of the equipment can be obtained; on the other hand, the method can prevent the generated signal matrix from being highly correlated by adding the random number on the basis of the original signal value, thereby preventing a pathological equation set from occurring in the singular value decomposition process, ensuring that the obtained singular value does not approach 0, namely the obtained singular value has reference significance, and effectively improving the accuracy of the operation state determination result when the corresponding operation state is determined according to the obtained singular value.
In one embodiment, the matrix generation module 200 is further configured to: taking each original signal value in the original signal sequence as an initial element to generate an initial matrix; and taking the sum of each initial element and the random number as a new element, and replacing the initial element at the corresponding position in the initial matrix by using the new element to obtain the signal matrix.
In one embodiment, the matrix generation module 200 is further configured to: taking the sum of each original signal value and the random number as a new signal value, and replacing the new signal value with the original signal value at the corresponding position in the original signal sequence to obtain a new signal sequence; and generating a signal matrix by taking each new signal value in the new signal sequence as an element.
In one embodiment, the state classification module 400 is further configured to: acquiring singular value sample data and corresponding state categories, wherein the singular value sample data comprises singular values corresponding to different state categories; model training is carried out based on the singular value sample data and the corresponding state category, and a trained state classification model is obtained; and carrying out state classification on the singular values through a state classification model to obtain the running state information of the equipment to be detected.
In one embodiment, as shown in fig. 8, the device operation state detection apparatus further includes: the abnormal prompting module 500 is configured to output corresponding prompting information when the operation state information of the target device includes an abnormal operation state.
The specific limitation of the device operation state detection apparatus may be referred to the limitation of the device operation state detection method hereinabove, and will not be described herein. The above-described respective modules in the apparatus operation state detection device may be implemented in whole or in part by software, hardware, and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of: acquiring an original signal sequence corresponding to target equipment to be detected, wherein the original signal sequence comprises a plurality of original signal values; generating a signal matrix based on the original signal sequence, wherein each element in the signal matrix comprises a sum of a single original signal value and a random number in a preset range; singular value decomposition is carried out on the signal matrix to obtain a plurality of singular values; and carrying out state classification on the plurality of singular values based on the corresponding relation between the singular values and the running states to obtain the running state information of the target equipment.
In one embodiment, the processor when executing the computer program further performs the steps of: taking each original signal value in the original signal sequence as an initial element to generate an initial matrix; and taking the sum of each initial element and the random number as a new element, and replacing the initial element at the corresponding position in the initial matrix by using the new element to obtain the signal matrix.
In one embodiment, the processor when executing the computer program further performs the steps of: taking the sum of each original signal value and the random number as a new signal value, and replacing the new signal value with the original signal value at the corresponding position in the original signal sequence to obtain a new signal sequence; and generating a signal matrix by taking each new signal value in the new signal sequence as an element.
In one embodiment, the processor when executing the computer program further performs the steps of: acquiring singular value sample data and corresponding state categories, wherein the singular value sample data comprises singular values corresponding to different state categories; model training is carried out based on the singular value sample data and the corresponding state category, and a trained state classification model is obtained; and carrying out state classification on the singular values through a state classification model to obtain the running state information of the equipment to be detected.
In one embodiment, the processor when executing the computer program further performs the steps of: when the running state information of the target equipment contains abnormal running states, outputting corresponding prompt information.
Fig. 9 is an internal structural diagram of a computer device in one embodiment. The computer device may in particular be a terminal (or a server). As shown in fig. 9, the computer device includes a processor, a memory, a network interface, an input device, and a display screen connected by a system bus. The memory includes a nonvolatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system, and may also store a computer program that, when executed by a processor, causes the processor to implement a device running state detection method. The internal memory may also store a computer program that, when executed by the processor, causes the processor to perform the device operating state detection method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by persons skilled in the art that the architecture shown in fig. 9 is merely a block diagram of some of the architecture relevant to the present inventive arrangements and is not limiting as to the computer device to which the present inventive arrangements are applicable, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of: acquiring an original signal sequence corresponding to target equipment to be detected, wherein the original signal sequence comprises a plurality of original signal values; generating a signal matrix based on the original signal sequence, wherein each element in the signal matrix comprises a sum of a single original signal value and a random number in a preset range; singular value decomposition is carried out on the signal matrix to obtain a plurality of singular values; and carrying out state classification on the plurality of singular values based on the corresponding relation between the singular values and the running states to obtain the running state information of the target equipment.
In one embodiment, the computer program when executed by the processor further performs the steps of: taking each original signal value in the original signal sequence as an initial element to generate an initial matrix; and taking the sum of each initial element and the random number as a new element, and replacing the initial element at the corresponding position in the initial matrix by using the new element to obtain the signal matrix.
In one embodiment, the computer program when executed by the processor further performs the steps of: taking the sum of each original signal value and the random number as a new signal value, and replacing the new signal value with the original signal value at the corresponding position in the original signal sequence to obtain a new signal sequence; and generating a signal matrix by taking each new signal value in the new signal sequence as an element.
In one embodiment, the computer program when executed by the processor further performs the steps of: acquiring singular value sample data and corresponding state categories, wherein the singular value sample data comprises singular values corresponding to different state categories; model training is carried out based on the singular value sample data and the corresponding state category, and a trained state classification model is obtained; and carrying out state classification on the singular values through a state classification model to obtain the running state information of the equipment to be detected.
In one embodiment, the computer program when executed by the processor further performs the steps of: when the running state information of the target equipment contains abnormal running states, outputting corresponding prompt information.
Those skilled in the art will appreciate that implementing all or part of the above-described methods may be accomplished by way of a computer program, which may be stored on a non-transitory computer readable storage medium and which, when executed, may comprise the steps of the above-described embodiments of the methods. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above-described embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above-described embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
Claims (10)
1. A method for detecting an operating state of a device, comprising:
acquiring an original signal sequence corresponding to target equipment to be detected, wherein the original signal sequence comprises a plurality of original signal values;
generating a signal matrix based on the original signal sequence, wherein each element in the signal matrix comprises a sum of a single original signal value and a random number in a preset range; the random numbers within the preset range do not submerge the original signal values;
singular value decomposition is carried out on the signal matrix to obtain a plurality of singular values;
and carrying out state classification on the singular values based on the corresponding relation between the singular values and the running states to obtain the running state information of the target equipment.
2. The method of claim 1, wherein generating a signal matrix based on the original signal sequence comprises:
generating an initial matrix by taking each original signal value in the original signal sequence as an initial element;
and taking the sum of the initial elements and the random numbers as a new element, and replacing the initial elements at the corresponding positions in the initial matrix by using the new element to obtain the signal matrix.
3. The method of claim 1, wherein generating a signal matrix based on the original signal sequence comprises:
taking the sum of each original signal value and the random number as a new signal value, and replacing the new signal value with the original signal value at the corresponding position in the original signal sequence to obtain a new signal sequence;
and taking each new signal value in the new signal sequence as an element to generate the signal matrix.
4. The method of claim 1, wherein the original signal values corresponding to the elements on each of the inverse diagonals in the signal matrix are equal.
5. The method of claim 1, wherein the classifying the plurality of singular values into the state based on the correspondence between the singular values and the operating states, the obtaining the operating state information of the target device comprises:
acquiring singular value sample data and corresponding state categories, wherein the singular value sample data comprises singular values corresponding to different state categories;
model training is carried out based on the singular value sample data and the corresponding state category, and a trained state classification model is obtained;
and carrying out state classification on the singular values through the state classification model to obtain the running state information of the target equipment.
6. The method of claim 5, wherein the state classification model comprises a neural network.
7. The method as recited in claim 1, further comprising:
and outputting corresponding prompt information when the running state information of the target equipment contains an abnormal running state.
8. An apparatus for detecting an operating state of a device, comprising:
the signal acquisition module is used for acquiring an original signal sequence generated in the operation process of target equipment to be detected, wherein the original signal sequence comprises a plurality of original signal values;
the matrix generation module is used for generating a signal matrix based on the original signal sequence, wherein each element in the signal matrix is the sum of a single original signal value and a random number in a preset range; the random numbers within the preset range do not submerge the original signal values;
the singular value decomposition module is used for carrying out singular value decomposition on the signal matrix to obtain a plurality of singular values;
and the state classification module is used for carrying out state classification on the singular values based on the corresponding relation between the singular values and the running states to obtain the running state information of the target equipment.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911336947.9A CN111144267B (en) | 2019-12-23 | 2019-12-23 | Equipment running state detection method and device, storage medium and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911336947.9A CN111144267B (en) | 2019-12-23 | 2019-12-23 | Equipment running state detection method and device, storage medium and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111144267A CN111144267A (en) | 2020-05-12 |
| CN111144267B true CN111144267B (en) | 2023-09-26 |
Family
ID=70519431
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911336947.9A Active CN111144267B (en) | 2019-12-23 | 2019-12-23 | Equipment running state detection method and device, storage medium and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111144267B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113760629A (en) * | 2020-06-15 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Method, device and equipment for monitoring running state of application system |
| CN111949938B (en) * | 2020-08-13 | 2023-06-20 | 抖音视界有限公司 | Determination method and device of transaction information, electronic equipment and computer readable medium |
| CN113111854A (en) * | 2021-04-30 | 2021-07-13 | 平安国际融资租赁有限公司 | Current signal extraction method, current signal extraction device, computer equipment and storage medium |
| CN115600131B (en) * | 2022-11-28 | 2023-03-07 | 西安弘捷电子技术有限公司 | Automatic test system based on graphical test platform |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106404386A (en) * | 2016-08-29 | 2017-02-15 | 北京工业大学 | Method for acquiring, extracting and diagnosing early stage fault characteristic signals of gear case |
| CN107192553A (en) * | 2017-06-28 | 2017-09-22 | 石家庄铁道大学 | Gear-box combined failure diagnostic method based on blind source separating |
| CN108960511A (en) * | 2018-07-05 | 2018-12-07 | 广东工业大学 | A kind of wind speed forecasting method and relevant apparatus |
| CN109058089A (en) * | 2018-06-13 | 2018-12-21 | 天津大学 | A method of the vacuum pump overload fault detection based on acoustic emission signal |
| US10299694B1 (en) * | 2018-02-05 | 2019-05-28 | King Saud University | Method of classifying raw EEG signals |
| CN110108474A (en) * | 2019-06-04 | 2019-08-09 | 山东大学 | A kind of rotating machinery operation stability on-line monitoring and appraisal procedure and system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4953239B2 (en) * | 2006-12-11 | 2012-06-13 | インターナショナル・ビジネス・マシーンズ・コーポレーション | Technology to detect anomalies in the observation target |
| WO2018025421A1 (en) * | 2016-08-05 | 2018-02-08 | 日本電気株式会社 | Object detection apparatus and object detection method |
-
2019
- 2019-12-23 CN CN201911336947.9A patent/CN111144267B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106404386A (en) * | 2016-08-29 | 2017-02-15 | 北京工业大学 | Method for acquiring, extracting and diagnosing early stage fault characteristic signals of gear case |
| CN107192553A (en) * | 2017-06-28 | 2017-09-22 | 石家庄铁道大学 | Gear-box combined failure diagnostic method based on blind source separating |
| US10299694B1 (en) * | 2018-02-05 | 2019-05-28 | King Saud University | Method of classifying raw EEG signals |
| CN109058089A (en) * | 2018-06-13 | 2018-12-21 | 天津大学 | A method of the vacuum pump overload fault detection based on acoustic emission signal |
| CN108960511A (en) * | 2018-07-05 | 2018-12-07 | 广东工业大学 | A kind of wind speed forecasting method and relevant apparatus |
| CN110108474A (en) * | 2019-06-04 | 2019-08-09 | 山东大学 | A kind of rotating machinery operation stability on-line monitoring and appraisal procedure and system |
Non-Patent Citations (2)
| Title |
|---|
| SVD analysis of GPR full-wave inversion;F. Watson et al.;《Proceedings of the 15th International Conference on Ground Penetrating Radar》;全文 * |
| 基于多分辨率技术及奇异值理论的故障检测方法;李世玲,李治,张富堂;系统工程与电子技术(04);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111144267A (en) | 2020-05-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111144267B (en) | Equipment running state detection method and device, storage medium and computer equipment | |
| Entezami et al. | Fast unsupervised learning methods for structural health monitoring with large vibration data from dense sensor networks | |
| CN107493277B (en) | Large data platform online anomaly detection method based on maximum information coefficient | |
| CN112766342A (en) | Abnormity detection method for electrical equipment | |
| CN111832019A (en) | Malicious code detection method based on generation countermeasure network | |
| CN111325159B (en) | Fault diagnosis method, device, computer equipment and storage medium | |
| CN112615888B (en) | Threat assessment method and device for network attack behavior | |
| Wang et al. | Performance degradation assessment of rolling bearing based on convolutional neural network and deep long-short term memory network | |
| CN112165484B (en) | Network encryption traffic identification method and device based on deep learning and side channel analysis | |
| CN114520736B (en) | Internet of things security detection method, device, equipment and storage medium | |
| Teodorescu | Characterization of nonlinear dynamic systems for engineering purposes–a partial review | |
| WO2019200739A1 (en) | Data fraud identification method, apparatus, computer device, and storage medium | |
| US20220367057A1 (en) | Missing medical diagnosis data imputation method and apparatus, electronic device and medium | |
| CN112668526A (en) | Bolt group loosening positioning monitoring method based on deep learning and piezoelectric active sensing | |
| Golovko et al. | Dimensionality reduction and attack recognition using neural network approaches | |
| JP2019105871A (en) | Abnormality candidate extraction program, abnormality candidate extraction method and abnormality candidate extraction apparatus | |
| US11762730B2 (en) | Selection of outlier-detection programs specific to dataset meta-features | |
| Yu et al. | MAG: A novel approach for effective anomaly detection in spacecraft telemetry data | |
| CN113110961A (en) | Equipment abnormality detection method and device, computer equipment and readable storage medium | |
| CN117454190A (en) | Log data analysis method and device | |
| CN117058451A (en) | Structural acceleration data anomaly detection method based on two-dimensional convolutional neural network | |
| CN116226770A (en) | Time sequence data anomaly detection method and device | |
| CN114492146B (en) | Bolt group loosening positioning and quantitative analysis method and system based on transfer learning | |
| Yan et al. | Mutual Information‐Assisted Wavelet Function Selection for Enhanced Rolling Bearing Fault Diagnosis | |
| US12001546B2 (en) | Systems and methods for causality-based multivariate time series anomaly detection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |