CN113283503A - Method, device, equipment and medium for detecting equipment state based on feature similarity - Google Patents
Method, device, equipment and medium for detecting equipment state based on feature similarity Download PDFInfo
- Publication number
- CN113283503A CN113283503A CN202110567243.3A CN202110567243A CN113283503A CN 113283503 A CN113283503 A CN 113283503A CN 202110567243 A CN202110567243 A CN 202110567243A CN 113283503 A CN113283503 A CN 113283503A
- Authority
- CN
- China
- Prior art keywords
- current
- standard
- historical data
- processed
- similarity
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000012545 processing Methods 0.000 claims abstract description 19
- 230000008859 change Effects 0.000 claims description 32
- 238000004590 computer program Methods 0.000 claims description 26
- 238000003066 decision tree Methods 0.000 claims description 25
- 238000001514 detection method Methods 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 16
- 238000003860 storage Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 abstract description 3
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000003754 machining Methods 0.000 description 13
- 238000004519 manufacturing process Methods 0.000 description 13
- 238000010606 normalization Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- FFBHFFJDDLITSX-UHFFFAOYSA-N benzyl N-[2-hydroxy-4-(3-oxomorpholin-4-yl)phenyl]carbamate Chemical compound OC1=C(NC(=O)OCC2=CC=CC=C2)C=CC(=C1)N1CCOCC1=O FFBHFFJDDLITSX-UHFFFAOYSA-N 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Abstract
The present application relates to the field of artificial intelligence, and in particular, to a method, an apparatus, a device, and a medium for detecting a device status based on feature similarity. The method comprises the following steps: acquiring a current to be processed of a device to be judged, and acquiring a historical current corresponding to the current to be processed to obtain a current sequence to be processed; acquiring a standard feature generated in advance according to small sample historical data, and reading a standard parameter corresponding to the standard feature; processing the current sequence to be processed based on the standardized parameters to obtain current characteristics; and determining the state of the equipment to be judged according to the similarity and an equipment state threshold value. The method can improve the accuracy of judging the state of the equipment, and in addition, the invention also relates to a block chain technology, and the standard characteristics can be stored in the block chain nodes.
Description
Technical Field
The present application relates to the field of artificial intelligence technologies, and in particular, to a method, an apparatus, a device, and a medium for detecting a device status based on feature similarity.
Background
With the development of the requirements of modern industry, the processing requirements on manufacturing equipment are higher and higher. Whether the operation of the manufacturing equipment normally directly influences the processing efficiency and the processing speed of the manufacturing equipment, the abnormal condition of the manufacturing equipment in the working state is accurately detected, and the equipment abnormality can be processed, so that the purpose of improving the processing efficiency of the manufacturing equipment is achieved.
In a conventional manner, the current of the manufacturing equipment during operation is usually collected at a low frequency by a current collecting function carried by the manufacturing equipment, and the equipment state is directly determined based on if based on the collected current.
However, for semi-automatic equipment, i.e. equipment requiring human intervention during operation, sometimes it is in a standby state, waiting for human intervention to perform subsequent operations, and from a business perspective, this waiting process should be calculated within working hours. For the low-power working state, the distribution interval is smaller, and the low-power working state is easily judged to be other working states by mistake due to the error of the acquisition equipment. The above-mentioned determination directly by a current leads to inaccurate determination.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a method, an apparatus, a device, and a medium for detecting a device state based on feature similarity, which can improve accuracy of determining the device state.
A device state detection method based on feature similarity comprises the following steps:
acquiring a current to be processed of a device to be judged, and acquiring a historical current corresponding to the current to be processed to obtain a current sequence to be processed;
acquiring a standard feature generated in advance according to small sample historical data, and reading a standard parameter corresponding to the standard feature;
processing the current sequence to be processed based on the standardized parameters to obtain current characteristics;
and determining the state of the equipment to be judged according to the similarity and an equipment state threshold value.
In one embodiment, before the obtaining of the standard features generated in advance according to the small sample historical data, the method further includes:
reading a preset amount of historical data from a database, and classifying the historical data according to the equipment state;
extracting a standardized parameter from the historical data in each classification, and standardizing the read historical data based on the standardized parameter;
and calculating to obtain standard characteristics according to the standardized historical data.
In one embodiment, after calculating the standard feature according to the normalized historical data, the method further includes:
determining the amount of historical data for which state classification has been completed;
and determining the updating frequency of the standard features according to the quantity of the historical data of which the state classification is completed, wherein the updating frequency of the standard features is slower when the quantity of the historical data of which the state classification is completed is larger.
In one embodiment, after calculating the standard feature according to the normalized historical data, the method further includes:
acquiring standard characteristics of historical data classified according to the completed states;
comparing the standard features with historically generated standard features to obtain the change rate of the standard features;
and adjusting the updating frequency of the standard features according to the change rate of the standard features, wherein the updating frequency of the standard features is faster when the change rate of the standard features is larger.
In one embodiment, the determining the state of the device to be determined according to the similarity and the device state threshold includes:
and correspondingly subtracting the current characteristic from each characteristic in the standard characteristic, and calculating the sum of the subtracted difference values as the similarity.
In one embodiment, the subtracting the current feature from each of the standard features, and calculating the sum of the subtracted differences as the similarity includes:
acquiring the weight corresponding to each feature;
subtracting the current characteristic from each characteristic in the standard characteristic correspondingly to obtain a difference value;
and calculating the sum of the subtracted difference values according to the weight corresponding to each feature to serve as the similarity.
In one embodiment, before the obtaining the weight corresponding to each feature, the method further includes:
obtaining a decision tree, each branch of the decision tree being one of the features;
and training the decision tree according to the historical data to obtain the weight corresponding to the characteristic of each fork.
An apparatus for detecting device status based on feature similarity, the apparatus comprising:
the device comprises a to-be-processed current acquisition module, a to-be-processed current acquisition module and a to-be-processed current sequence acquisition module, wherein the to-be-processed current acquisition module is used for acquiring a to-be-processed current of a to-be-judged device and acquiring a historical current corresponding to the to-be-processed current to obtain the to-be-processed current sequence;
the standardized parameter acquisition module is used for acquiring a standard feature generated in advance according to the historical data of the small sample and reading a standardized parameter corresponding to the standard feature;
the current characteristic calculation module is used for processing the current sequence to be processed based on the standardized parameters to obtain current characteristics;
and the state detection module is used for determining the state of the equipment to be judged according to the similarity and the equipment state threshold value.
A computer device comprising a memory storing a computer program and a processor implementing the steps of the above method when executing the computer program.
A computer storage medium on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method.
According to the method, the device, the equipment and the medium for detecting the equipment state based on the characteristic similarity, the judgment based on the historical standard characteristic does not need a data label, and the standard characteristic is generated in advance based on the small sample historical data, namely, the priori knowledge is introduced to obtain the standardized parameter, so that when the current sequence to be processed is processed, the standardized parameter is adopted to enable the model to be compatible with the state of unbalanced data, namely, a certain state is not acquired, the current sequence to be processed can be standardized to obtain the current characteristic, the similarity calculation with the standard characteristic is carried out to judge the equipment state, and the accuracy of the equipment state judgment is improved.
Drawings
FIG. 1 is a diagram illustrating an exemplary application of a method for detecting device states based on feature similarity;
FIG. 2 is a schematic flow chart illustrating a method for device status detection based on feature similarity according to an embodiment;
FIG. 3 is a block diagram of an embodiment of an apparatus for detecting device status based on feature similarity;
FIG. 4 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The device state detection method based on the feature similarity can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the database 104 via a network. The terminal 102 may obtain the current to be processed of the device to be determined stored in the database 104, and obtain the historical current corresponding to the current to be processed to obtain a current sequence to be processed, where the current to be processed of the device to be determined obtained by the terminal 102 may also be directly read from the device to be determined, and the terminal 102 obtains a standard feature pre-generated according to the historical data of the small sample, and reads a standardized parameter corresponding to the standard feature; processing the current sequence to be processed based on the standardized parameters to obtain current characteristics; and finally, determining the state of the equipment to be judged according to the similarity and the equipment state threshold, so that the judgment based on the historical standard characteristic does not need a data label, and because the standard characteristic is generated in advance based on small sample historical data, namely, priori knowledge is introduced to obtain a standardized parameter, when the current sequence to be processed is processed, the standardized parameter is adopted to enable the model to be compatible with the state of unbalanced data, namely, a certain state is not acquired, the current sequence to be processed can be standardized to obtain the current characteristic, so that the similarity calculation is carried out on the current characteristic and the standard characteristic to judge the state of the equipment, and the accuracy of the equipment state judgment is improved. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, and the database 104 may be implemented by an independent database or a database cluster composed of a plurality of databases.
In an embodiment, as shown in fig. 2, a method for detecting a device status based on feature similarity is provided, which is described by taking the method as an example applied to the terminal in fig. 1, and includes the following steps:
s202: obtaining the current to be processed of the equipment to be judged, and obtaining the historical current corresponding to the current to be processed to obtain the current sequence to be processed.
Specifically, the device to be determined may be a manufacturing device or the like, including a semi-automatic device, a low-power device, and the like. The current to be processed refers to data obtained by collecting the operating current of the manufacturing equipment through an external collection device, and may specifically be data collected at a relatively low frequency, for example, data generated by collecting the operating current once in 1 minute. The current may particularly refer to current signal data.
In this embodiment, the historical current refers to a plurality of current data before the current to be processed, for example, the first N-1(N > -3) historical current values that are adjacent to each other are expected, so that the corresponding current sequence to be processed can be obtained by arranging the current data according to the chronological order, and the current sequence to be processed may be data of the manufacturing equipment in a preset detection interval, for example, half-day current data or one-day current data, and the like, which is not limited in this application.
In this embodiment, the manufacturing apparatus may include a plurality of apparatus states, for example, a machining state, a non-machining state, a standby state, or a rough machining state, a finish machining state, a machining process 1 state, a machining process 2 state, or the like, for a semi-automatic machining apparatus, which is not limited in this application.
In this embodiment, the current sequence to be processed of the device to be determined is data of the manufacturing device in the continuous operation state, that is, the current to be processed may include current data of the manufacturing device in the machining state, may also include current data in the non-machining state, may include current data in the rough machining state, may also include current data in the fine machining state, or may also include current data in the machining process 1 state, and may also include current data in the machining process 2 state. That is, the current to be processed is data in a preset time interval, and is data of the whole.
It should be noted that the current to be processed acquired by the terminal does not carry the device status, and the device status needs to be determined according to the characteristics of the current to be processed.
S204: and acquiring standard features generated in advance according to the small sample historical data, and reading the standardized parameters corresponding to the standard features.
Specifically, the standard features are obtained by processing a part of data manually marked in advance, for example, extracting a certain amount of historical data, which can ensure that the amount of the extracted historical data of each equipment state is balanced for convenience, so as to ensure the accuracy of the standard features. The historical data for each device state is then processed to obtain the corresponding standard features. And preferably, the terminal first performs a normalization process on the historical data before obtaining the standard feature from the historical data, wherein the normalization parameter may be a parameter that is not changed in the normalization process. For example, if the processing is performed by the minmax normalization, the normalization parameter may be a maximum value and a minimum value of the extracted history data, and each history data may be normalized by the maximum value and the minimum value, that is, the history data after the normalization is the history data before the normalization/(maximum value-minimum value). And then, calculating according to the standardized historical data to obtain the standard characteristics.
S206: and processing the current sequence to be processed based on the standardized parameters to obtain current characteristics.
Specifically, the method for standardizing the current sequence to be processed is the same as the method for standardizing the historical data, and the same standardization parameters are adopted, so that the obtained current characteristics and the standardization characteristics are comparable.
S208: and obtaining the similarity corresponding to the current sequence to be processed according to the standard characteristic and the current characteristic.
Specifically, the standard features include, but are not limited to, at least one of a median, a variance, a 75% quantile, and a coefficient of variation, and when the similarity is calculated, the similarities of the median, the variance, the 75% quantile, and the coefficient of variation are calculated respectively, and then the final similarity is obtained by weighting.
S210: and determining the state of the equipment to be judged according to the similarity and the equipment state threshold value.
In the above embodiment, the judgment of the historical standard features does not need data labels, and the standard features are generated in advance based on the historical data of the small sample, that is, the priori knowledge is introduced to obtain the standardized parameters, so that when the current sequence to be processed is processed, the standardized parameters are adopted to enable the model to be compatible with the state of data imbalance, that is, a certain state is not acquired, the current sequence to be processed can still be standardized to obtain the current features, and therefore, the similarity calculation is performed on the current features and the standard features to judge the state of the equipment, and the accuracy of the equipment state judgment is improved.
In one embodiment, before the obtaining of the standard features generated in advance according to the small sample historical data, the method further includes: reading a preset amount of historical data from a database, and classifying the historical data according to the equipment state; extracting standardized parameters from the historical data in each classification, and standardizing the read historical data based on the standardized parameters; and calculating to obtain standard characteristics according to the standardized historical data.
It should be noted that, before the standard feature generated in advance according to the small sample historical data is acquired, the added steps are mainly the standard feature generation manner, and the steps of the standard feature generation manner are not required to be executed before the standard feature is acquired every time, but are periodically or according to needs to generate the standard feature.
Specifically, in this embodiment, the historical data of a large data size is processed by the historical data of a small sample, so that the sample data size is increased, and the accuracy of the standardized parameters and the standard features is improved.
Specifically, as for the device status, as described above, the terminal marks the obtained small sample data to obtain a corresponding label, and then processes each class separately, for example, parallel processing may be performed, where before parallel processing, extraction of a normalization parameter, that is, processing of a current maximum value and a current minimum value, may be performed first, and then the historical data in each class is normalized according to the obtained normalization parameter, that is, the normalized historical data is the historical data before normalization/(maximum-minimum value).
The standard features include, but are not limited to, at least one of a median, a variance, a 75% quantile, and a coefficient of variation, so that the terminal extracts the standard features for each class to obtain the standard features corresponding to each class, that is, the standard features corresponding to each device state.
In the above embodiment, a certain amount of historical data is labeled in a classified manner, then other large amounts of historical data are processed, a large sample is labeled with a small sample, and after the processing is completed, the labeled data is added to the small sample, so that the sample amount is gradually increased, and further, the standardized parameters and the standard features are more accurate.
In one embodiment, after calculating the standard features according to the normalized historical data, the method further includes: determining the amount of historical data for which state classification has been completed; and determining the updating frequency of the standard features according to the quantity of the historical data of which the state classification is completed, wherein the more the quantity of the historical data of which the state classification is completed, the slower the updating frequency of the standard features is.
Specifically, the data amount refers to the number of historical data used for generating the standard feature corresponding to each device state, wherein the number of the standard features corresponding to each device state may be counted respectively in this embodiment. The update frequency refers to the update frequency of the standard feature and the normalized parameter, which is related to the amount of the historical data of the completed state classification, and the greater the amount of the historical data of the completed state classification is, the more stable the standard feature is, and the slower the update frequency thereof is.
Therefore, the terminal can acquire the number of the historical data of the standard feature corresponding to each equipment state, then judge the updating frequency corresponding to the number, acquire the last updating time and the current time, or acquire the number of the historical data of the standard feature corresponding to each equipment state and the number of the historical data of the standard feature corresponding to each equipment state at the last updating time, determine whether the updating condition is met according to the data, and update the standardized parameters and the standard features according to the generation method of the standardized parameters and the standard features if the updating condition is met.
If the update frequency is time, calculating a time interval between the last update time and the current time, and judging whether the time interval meets the time corresponding to the update frequency. If the update frequency is the data volume, calculating the difference between the number of the historical data of the standard feature corresponding to each equipment state at the last update and the number of the historical data of the standard feature corresponding to each current equipment state, and judging whether the difference meets the data volume corresponding to the update frequency to judge whether the update condition is met.
In the embodiment, the standard features and the standardized parameters are updated according to the sample data size, so that the stability of the standard features and the standardized parameters is improved, and the accuracy is also ensured.
In one embodiment, after calculating the standard features according to the normalized historical data, the method further includes: acquiring standard characteristics of historical data classified according to the completed states; comparing the standard features with the historically generated standard features to obtain the change rate of the standard features; and adjusting the updating frequency of the standard features according to the change rate of the standard features, wherein the larger the change rate of the standard features is, the faster the updating frequency of the standard features is.
Specifically, in this embodiment, the standard features and the normalized parameters are updated in association with the change rates thereof, that is, the change rates of the currently updated standard features and the historically generated standard features can be determined, that is, the maximum value of the change rates of the above features, that is, the median, the variance, the 75% quantile and the coefficient of variation, is used as the target change rate, and then the level of the change rate is determined, so as to obtain the relationship between the preset change rate level and the update frequency, and further determine the change time of the next standard feature, so that when the next time arrives, the standard features and the normalized parameters are updated, where it is to be noted that the greater the change rate of the standard features is, the faster the update frequency of the standard features is.
In practical application, after the standard feature and the standardized parameter are updated, the change rate of the standard feature can be calculated, and then the level of the change rate is determined, so that the relation between the preset change rate level and the update frequency is obtained, the change time of the next standard feature is further determined, the current time is obtained in real time, and if the current time reaches the change time, the standard feature and the standardized parameter are updated.
In the above embodiment, the standard features and the standardized parameters are updated by combining the change rate of the standard features, so that the stability of the standard features and the standardized parameters is improved, and the accuracy is also ensured.
In one embodiment, the standard features include at least one of median, variance, 75% quantile, coefficient of variation; determining the state of the device to be judged according to the similarity and the device state threshold, comprising: and correspondingly subtracting the current characteristic from each characteristic in the standard characteristic, and calculating the sum of the subtracted differences as the similarity.
Specifically, the standard features in this embodiment include at least one of median, variance, 75% quantile, coefficient of variation, and in other embodiments the normalization parameters may also include other statistics of the historical data.
In order to ensure that if can be directly used during threshold judgment, no additional judgment condition is needed, and implementation is easy, when the similarity of the standard characteristic and the current characteristic is calculated, subtraction is directly used instead of using the mean square sum or the absolute value sum, so that the sign bit can be reserved.
In one embodiment, to ensure the simplification of the subsequent calculation, the difference corresponding to each feature is summed to obtain the final similarity.
In the above embodiment, the subtraction is directly adopted, so that the sign bit can be retained, and if can be directly used when the threshold is determined, and no additional determination condition is required, and the implementation is easier.
In one embodiment, the subtracting the current feature from each of the standard features, and calculating the sum of the subtracted differences as the similarity includes: acquiring the weight corresponding to each feature; subtracting each characteristic in the current characteristic and the standard characteristic correspondingly to obtain a difference value; and calculating the sum of the subtracted difference values according to the weight corresponding to each feature to serve as the similarity.
Specifically, in the embodiment, when the similarity is calculated, a corresponding weight is assigned to each feature, and the weight may be manually set according to a calculation scenario or calculated according to historical data, so that the weight of the most critical feature in the scenario is increased, and the similarity is more accurate.
Optionally, before the obtaining the weight corresponding to each feature, the method further includes: obtaining a decision tree, wherein each branch of the decision tree is a feature; and training the decision tree according to historical data to obtain the weight corresponding to the characteristic of each fork.
Before the weight corresponding to each feature is obtained, the added step is a weight generation mode, and the step of the weight generation mode is not required to be obtained before the weight is obtained every time, but is generated periodically or according to the requirement.
Specifically, for convenience, in the present embodiment, a structure of the decision tree is first constructed according to each feature, wherein a plurality of decision trees may be constructed in an arrangement manner with each feature, and finally, a user selects one of the decision trees. And in order to obtain the weight corresponding to each feature in the decision tree, the terminal inputs the historical data into the decision tree to obtain the weight corresponding to each branch of the decision tree, so as to obtain the weight corresponding to each feature.
In this embodiment, in order to ensure the accuracy of the similarity, a weight is assigned to each feature, so that the calculation of the similarity is more representative.
It is emphasized that, in order to further ensure the privacy and safety of the standard features, the currents to be processed and the historical currents, the standard features, the currents to be processed and the historical currents can also be stored in the nodes of a block chain.
It should be understood that, although the steps in the flowchart of fig. 2 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a portion of the steps in fig. 2 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
In one embodiment, as shown in fig. 3, there is provided a device status detection apparatus based on feature similarity, including: a current to be processed acquisition module 100, a standardized parameter acquisition module 200, a current feature calculation module 300, a similarity calculation module 400, and a state detection module 500, wherein:
the current to be processed acquiring module 100 is configured to acquire a current to be processed of a device to be determined, and acquire a historical current corresponding to the current to be processed to obtain a current sequence to be processed;
a standardized parameter obtaining module 200, configured to obtain a standard feature pre-generated according to the small sample historical data, and read a standardized parameter corresponding to the standard feature;
the current feature calculation module 300 is configured to process the current sequence to be processed based on the standardized parameter to obtain a current feature;
a similarity calculation module 400, configured to obtain a similarity corresponding to the current sequence to be processed according to the standard feature and the current feature;
and the state detection module 500 is configured to determine the state of the device to be determined according to the similarity and the device state threshold.
In one embodiment, the device state detection apparatus based on feature similarity further includes:
the classification module is used for reading a preset amount of historical data from the database and classifying the historical data according to the equipment state;
the standardization module is used for extracting standardization parameters from the historical data in each classification and standardizing the read historical data based on the standardization parameters;
and the calculation module is used for calculating to obtain the standard characteristics according to the standardized historical data.
In one embodiment, the device state detection apparatus based on feature similarity further includes:
the quantity determining module is used for determining the quantity of the historical data which is subjected to state classification;
and the first updating frequency adjusting module is used for determining the updating frequency of the standard features according to the quantity of the historical data which have already finished the state classification, wherein the more the quantity of the historical data which have already finished the state classification is, the slower the updating frequency of the standard features is.
In one embodiment, the device state detection apparatus based on feature similarity further includes:
the characteristic acquisition module is used for acquiring standard characteristics of the historical data classified according to the finished states;
the comparison module is used for comparing the standard features with the historical standard features to obtain the change rate of the standard features;
and the second updating frequency adjusting module is used for adjusting the updating frequency of the standard feature according to the change rate of the standard feature, wherein the larger the change rate of the standard feature is, the faster the updating frequency of the standard feature is.
In one embodiment, the standard features include at least one of median, variance, 75% quantile, coefficient of variation; the state detection module 500 is configured to subtract each of the current features from each of the standard features, and calculate a sum of the subtracted differences as a similarity.
In one embodiment, the status detection module 500 includes:
the weight obtaining unit is used for obtaining the weight corresponding to each feature;
the difference value calculating unit is used for correspondingly subtracting each characteristic in the current characteristic and the standard characteristic to obtain a difference value;
and the similarity calculation unit is used for calculating the sum of the subtracted difference values according to the weight corresponding to each feature as the similarity.
In one embodiment, the device state detection apparatus based on feature similarity further includes:
the decision tree acquisition module is used for acquiring a decision tree, and each branch of the decision tree is a feature;
and the weight calculation module is used for training the decision tree according to the historical data to obtain the weight corresponding to the characteristic of each fork.
For the specific definition of the device state detection apparatus based on the feature similarity, refer to the above definition of the device state detection method based on the feature similarity, and are not described herein again. All or part of each module in the device state detection device based on the feature similarity can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 4. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of device state detection based on feature similarity. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
Those skilled in the art will appreciate that the architecture shown in fig. 4 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, there is provided a computer device comprising a memory storing a computer program and a processor implementing the following steps when the processor executes the computer program: obtaining a current to be processed of a device to be judged, and obtaining a historical current corresponding to the current to be processed to obtain a current sequence to be processed; acquiring a standard feature generated in advance according to the historical data of the small sample, and reading a standard parameter corresponding to the standard feature; processing the current sequence to be processed based on the standardized parameters to obtain current characteristics; obtaining the similarity corresponding to the current sequence to be processed according to the standard characteristic and the current characteristic; and determining the state of the equipment to be judged according to the similarity and the equipment state threshold value.
In one embodiment, the processor, involved in executing the computer program, further performs the following steps prior to said obtaining the standard features pre-generated from the small sample history data: reading a preset amount of historical data from a database, and classifying the historical data according to the equipment state; extracting standardized parameters from the historical data in each classification, and standardizing the read historical data based on the standardized parameters; and calculating to obtain standard characteristics according to the standardized historical data.
In one embodiment, after the processor, involved in executing the computer program, calculates the standard features according to the normalized historical data, the method further comprises: determining the amount of historical data for which state classification has been completed; and determining the updating frequency of the standard features according to the quantity of the historical data of which the state classification is completed, wherein the more the quantity of the historical data of which the state classification is completed, the slower the updating frequency of the standard features is.
In one embodiment, after the processor, involved in executing the computer program, calculates the standard features according to the normalized historical data, the method further comprises: acquiring standard characteristics of historical data classified according to the completed states; comparing the standard features with the historically generated standard features to obtain the change rate of the standard features; and adjusting the updating frequency of the standard features according to the change rate of the standard features, wherein the larger the change rate of the standard features is, the faster the updating frequency of the standard features is.
In one embodiment, the determining the state of the device to be determined according to the similarity and the device state threshold involved in the execution of the computer program by the processor includes: and correspondingly subtracting the current characteristic from each characteristic in the standard characteristic, and calculating the sum of the subtracted differences as the similarity.
In one embodiment, the subtracting the current signature from each of the standard signatures and calculating the sum of the subtracted differences as the similarity involved in the execution of the computer program by the processor comprises: acquiring the weight corresponding to each feature; subtracting each characteristic in the current characteristic and the standard characteristic correspondingly to obtain a difference value; and calculating the sum of the subtracted difference values according to the weight corresponding to each feature to serve as the similarity.
In one embodiment, the processor, when executing the computer program, further performs the following steps before the obtaining the weight corresponding to each feature: obtaining a decision tree, wherein each branch of the decision tree is a feature; and training the decision tree according to historical data to obtain the weight corresponding to the characteristic of each fork.
In one embodiment, a computer storage medium is provided, having a computer program stored thereon, the computer program, when executed by a processor, implementing the steps of: obtaining a current to be processed of a device to be judged, and obtaining a historical current corresponding to the current to be processed to obtain a current sequence to be processed; acquiring a standard feature generated in advance according to the historical data of the small sample, and reading a standard parameter corresponding to the standard feature; processing the current sequence to be processed based on the standardized parameters to obtain current characteristics; obtaining the similarity corresponding to the current sequence to be processed according to the standard characteristic and the current characteristic; and determining the state of the equipment to be judged according to the similarity and the equipment state threshold value.
In one embodiment, the computer program, when executed by the processor, further performs the following steps prior to said obtaining the standard features pre-generated from the small sample history data: reading a preset amount of historical data from a database, and classifying the historical data according to the equipment state; extracting standardized parameters from the historical data in each classification, and standardizing the read historical data based on the standardized parameters; and calculating to obtain standard characteristics according to the standardized historical data.
In one embodiment, the computer program, after being executed by the processor, further comprises, after calculating the standard feature based on the normalized historical data: determining the amount of historical data for which state classification has been completed; and determining the updating frequency of the standard features according to the quantity of the historical data of which the state classification is completed, wherein the more the quantity of the historical data of which the state classification is completed, the slower the updating frequency of the standard features is.
In one embodiment, the computer program, after being executed by the processor, further comprises, after calculating the standard feature based on the normalized historical data: acquiring standard characteristics of historical data classified according to the completed states; comparing the standard features with the historically generated standard features to obtain the change rate of the standard features; and adjusting the updating frequency of the standard features according to the change rate of the standard features, wherein the larger the change rate of the standard features is, the faster the updating frequency of the standard features is.
In one embodiment, the determining the state of the device to be determined according to the similarity and the device state threshold, when executed by the processor, includes: and correspondingly subtracting the current characteristic from each characteristic in the standard characteristic, and calculating the sum of the subtracted differences as the similarity.
In one embodiment, the computer program, when executed by the processor, involves subtracting each of the current signature and the standard signature respectively, and calculating a sum of the subtracted differences as the similarity, including: acquiring the weight corresponding to each feature; subtracting each characteristic in the current characteristic and the standard characteristic correspondingly to obtain a difference value; and calculating the sum of the subtracted difference values according to the weight corresponding to each feature to serve as the similarity.
In one embodiment, the computer program, when executed by the processor, further performs the following steps prior to said obtaining the weight corresponding to each feature: obtaining a decision tree, wherein each branch of the decision tree is a feature; and training the decision tree according to historical data to obtain the weight corresponding to the characteristic of each fork.
The block chain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism, an encryption algorithm and the like. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile 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), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A device state detection method based on feature similarity comprises the following steps:
acquiring a current to be processed of a device to be judged, and acquiring a historical current corresponding to the current to be processed to obtain a current sequence to be processed;
acquiring a standard feature generated in advance according to small sample historical data, and reading a standard parameter corresponding to the standard feature;
processing the current sequence to be processed based on the standardized parameters to obtain current characteristics;
obtaining the similarity corresponding to the current sequence to be processed according to the standard characteristic and the current characteristic;
and determining the state of the equipment to be judged according to the similarity and an equipment state threshold value.
2. The method of claim 1, wherein prior to said obtaining standard features pre-generated from small sample historical data, the method further comprises:
reading a preset amount of historical data from a database, and classifying the historical data according to the equipment state;
extracting a standardized parameter from the historical data in each classification, and standardizing the read historical data based on the standardized parameter;
and calculating to obtain standard characteristics according to the standardized historical data.
3. The method of claim 2, wherein after calculating the standard features from the normalized historical data, further comprising:
determining the amount of historical data for which state classification has been completed;
and determining the updating frequency of the standard features according to the quantity of the historical data of which the state classification is completed, wherein the updating frequency of the standard features is slower when the quantity of the historical data of which the state classification is completed is larger.
4. The method of claim 2, wherein after calculating the standard features from the normalized historical data, further comprising:
acquiring standard characteristics of historical data classified according to the completed states;
comparing the standard features with historically generated standard features to obtain the change rate of the standard features;
and adjusting the updating frequency of the standard features according to the change rate of the standard features, wherein the updating frequency of the standard features is faster when the change rate of the standard features is larger.
5. The method according to any one of claims 1 to 4, wherein the determining the state of the device to be determined according to the similarity and a device state threshold value comprises:
and correspondingly subtracting the current characteristic from each characteristic in the standard characteristic, and calculating the sum of the subtracted difference values as the similarity.
6. The method of claim 5, wherein the subtracting the current signature from each of the standard signatures and calculating a sum of the subtracted differences as a similarity comprises:
acquiring the weight corresponding to each feature;
subtracting the current characteristic from each characteristic in the standard characteristic correspondingly to obtain a difference value;
and calculating the sum of the subtracted difference values according to the weight corresponding to each feature to serve as the similarity.
7. The method of claim 6, wherein prior to said obtaining the weight corresponding to each feature, the method further comprises:
obtaining a decision tree, each branch of the decision tree being one of the features;
and training the decision tree according to the historical data to obtain the weight corresponding to the characteristic of each fork.
8. An apparatus for detecting device status based on feature similarity, the apparatus comprising:
the device comprises a to-be-processed current acquisition module, a to-be-processed current acquisition module and a to-be-processed current sequence acquisition module, wherein the to-be-processed current acquisition module is used for acquiring a to-be-processed current of a to-be-judged device and acquiring a historical current corresponding to the to-be-processed current to obtain the to-be-processed current sequence;
the standardized parameter acquisition module is used for acquiring a standard feature generated in advance according to the historical data of the small sample and reading a standardized parameter corresponding to the standard feature;
the current characteristic calculation module is used for processing the current sequence to be processed based on the standardized parameters to obtain current characteristics;
the similarity calculation module is used for obtaining the similarity corresponding to the current sequence to be processed according to the standard characteristic and the current characteristic;
and the state detection module is used for determining the state of the equipment to be judged according to the similarity and the equipment state threshold value.
9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 7 when executing the computer program.
10. A computer storage medium on which a computer program is stored, characterized in that the computer program, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110567243.3A CN113283503A (en) | 2021-05-24 | 2021-05-24 | Method, device, equipment and medium for detecting equipment state based on feature similarity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110567243.3A CN113283503A (en) | 2021-05-24 | 2021-05-24 | Method, device, equipment and medium for detecting equipment state based on feature similarity |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113283503A true CN113283503A (en) | 2021-08-20 |
Family
ID=77281145
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110567243.3A Pending CN113283503A (en) | 2021-05-24 | 2021-05-24 | Method, device, equipment and medium for detecting equipment state based on feature similarity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113283503A (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105548814A (en) * | 2016-01-13 | 2016-05-04 | 国网浙江省电力公司 | Method and system for determining ground fault of distribution network line |
| CN106124885A (en) * | 2016-06-13 | 2016-11-16 | 四川网达科技有限公司 | Switch breakdown detection apparatus and method |
| CN108052628A (en) * | 2017-12-19 | 2018-05-18 | 河北省科学院应用数学研究所 | Switch starting electric current detecting method, system and terminal device |
| CN108416362A (en) * | 2018-01-29 | 2018-08-17 | 同济大学 | A kind of track switch abnormity early warning and method for diagnosing faults |
| CN108693469A (en) * | 2018-06-12 | 2018-10-23 | 广东电网有限责任公司 | The method for diagnosing faults and device of GIS device |
| CN109613432A (en) * | 2019-01-08 | 2019-04-12 | 广州小鹏汽车科技有限公司 | Estimate method, equipment and the computer readable storage medium of battery charge state |
| CN110163094A (en) * | 2019-04-15 | 2019-08-23 | 深圳壹账通智能科技有限公司 | Biopsy method, device, equipment and storage medium based on gesture motion |
| CN110674149A (en) * | 2019-09-12 | 2020-01-10 | 金蝶软件（中国）有限公司 | Service data processing method and device, computer equipment and storage medium |
| CN110766269A (en) * | 2019-09-02 | 2020-02-07 | 平安科技（深圳）有限公司 | Task allocation method and device, readable storage medium and terminal equipment |
| CN110780150A (en) * | 2019-10-21 | 2020-02-11 | 江苏轶一电力科技有限公司 | Transmission line fault positioning device and method based on transmission tower leakage current |
| CN111882395A (en) * | 2020-07-29 | 2020-11-03 | 平安国际融资租赁有限公司 | Monitoring method and system for rental equipment, computer equipment and storage medium |
| CN112184090A (en) * | 2020-11-30 | 2021-01-05 | 广东浩迪创新科技有限公司 | Standard electricity utilization feature library establishing method, environment-friendly monitoring method, system and monitor |
-
2021
- 2021-05-24 CN CN202110567243.3A patent/CN113283503A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105548814A (en) * | 2016-01-13 | 2016-05-04 | 国网浙江省电力公司 | Method and system for determining ground fault of distribution network line |
| CN106124885A (en) * | 2016-06-13 | 2016-11-16 | 四川网达科技有限公司 | Switch breakdown detection apparatus and method |
| CN108052628A (en) * | 2017-12-19 | 2018-05-18 | 河北省科学院应用数学研究所 | Switch starting electric current detecting method, system and terminal device |
| CN108416362A (en) * | 2018-01-29 | 2018-08-17 | 同济大学 | A kind of track switch abnormity early warning and method for diagnosing faults |
| CN108693469A (en) * | 2018-06-12 | 2018-10-23 | 广东电网有限责任公司 | The method for diagnosing faults and device of GIS device |
| CN109613432A (en) * | 2019-01-08 | 2019-04-12 | 广州小鹏汽车科技有限公司 | Estimate method, equipment and the computer readable storage medium of battery charge state |
| CN110163094A (en) * | 2019-04-15 | 2019-08-23 | 深圳壹账通智能科技有限公司 | Biopsy method, device, equipment and storage medium based on gesture motion |
| CN110766269A (en) * | 2019-09-02 | 2020-02-07 | 平安科技（深圳）有限公司 | Task allocation method and device, readable storage medium and terminal equipment |
| CN110674149A (en) * | 2019-09-12 | 2020-01-10 | 金蝶软件（中国）有限公司 | Service data processing method and device, computer equipment and storage medium |
| CN110780150A (en) * | 2019-10-21 | 2020-02-11 | 江苏轶一电力科技有限公司 | Transmission line fault positioning device and method based on transmission tower leakage current |
| CN111882395A (en) * | 2020-07-29 | 2020-11-03 | 平安国际融资租赁有限公司 | Monitoring method and system for rental equipment, computer equipment and storage medium |
| CN112184090A (en) * | 2020-11-30 | 2021-01-05 | 广东浩迪创新科技有限公司 | Standard electricity utilization feature library establishing method, environment-friendly monitoring method, system and monitor |
Non-Patent Citations (1)
| Title |
|---|
| M.NARASIMHAMURTY 等: "《模式识别 算法及实现方法》", 31 December 2017 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108460346B (en) | Fingerprint identification method and device | |
| CN111314173A (en) | Monitoring information abnormity positioning method and device, computer equipment and storage medium | |
| CN112016318A (en) | Triage information recommendation method, device, equipment and medium based on interpretation model | |
| CN111767192B (en) | Business data detection method, device, equipment and medium based on artificial intelligence | |
| CN111400126A (en) | Network service abnormal data detection method, device, equipment and medium | |
| CN112365070A (en) | Power load prediction method, device, equipment and readable storage medium | |
| CN110888625B (en) | Method for controlling code quality based on demand change and project risk | |
| CN111178537A (en) | Feature extraction model training method and device | |
| CN113110961B (en) | Equipment abnormality detection method and device, computer equipment and readable storage medium | |
| CN107077617B (en) | Fingerprint extraction method and device | |
| CN113283501A (en) | Deep learning-based equipment state detection method, device, equipment and medium | |
| CN111738356A (en) | Object feature generation method, device, equipment and storage medium for specific data | |
| CN113283503A (en) | Method, device, equipment and medium for detecting equipment state based on feature similarity | |
| CN115660817A (en) | Classification model training method and device | |
| CN114860608A (en) | Scene construction based system automation testing method, device, equipment and medium | |
| CN107463564A (en) | The characteristic analysis method and device of data in server | |
| CN112966965A (en) | Import and export big data analysis and decision method, device, equipment and storage medium | |
| CN113283504B (en) | Device state detection method and device based on clustering, computer device and medium | |
| CN113095963A (en) | Real estate cost data processing method, real estate cost data processing device, computer equipment and storage medium | |
| CN113689020A (en) | Service information prediction method, device, computer equipment and storage medium | |
| CN113269246B (en) | Method, device, equipment and medium for detecting equipment state based on transaction logic | |
| CN111382925A (en) | Production actual performance data analysis device | |
| CN111340349A (en) | Product reliability evaluation method and device, computer equipment and storage medium | |
| CN111984636B (en) | Data modeling method, device, equipment and storage medium | |
| CN114356405B (en) | Matching method and device of open source component function, computer equipment and storage medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210820 |