CN113822379B - Process process anomaly analysis method and device, electronic equipment and storage medium - Google Patents
Process process anomaly analysis method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN113822379B CN113822379B CN202111381935.5A CN202111381935A CN113822379B CN 113822379 B CN113822379 B CN 113822379B CN 202111381935 A CN202111381935 A CN 202111381935A CN 113822379 B CN113822379 B CN 113822379B
- Authority
- CN
- China
- Prior art keywords
- training
- groups
- sets
- characteristic
- procedure
- 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
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
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- General Factory Administration (AREA)
Abstract
The invention relates to the technical field of production monitoring, and discloses a method and a device for analyzing process abnormity, electronic equipment and a storage medium, wherein the method comprises the following steps: receiving a process procedure training set and a label set, performing characteristic classification on the process procedure training set to obtain a plurality of groups of procedure characteristic training sets, constructing a plurality of corresponding isolated tree models according to the plurality of groups of procedure characteristic training sets, sequentially inputting each group of procedure characteristic training sets to the corresponding isolated tree models to perform training, selecting the isolated tree models with training results meeting preset requirements by using the label set to obtain one or more groups of quality tree models, and inputting a process procedure data set to be judged as abnormal to the quality tree models to perform abnormal analysis to obtain abnormal analysis results. The invention also provides a process anomaly analysis device, electronic equipment and a storage medium. The invention can solve the technical problem of poor detection accuracy caused by insufficient utilization of industrial process data in the prior art.
Description
Technical Field
The invention relates to the technical field of production monitoring, in particular to a method and a device for analyzing process abnormity, electronic equipment and a storage medium.
Background
Along with the development of science and technology, the intelligent degree of each industry is higher and higher, manpower and material resources can be effectively saved, particularly, in the process of manufacturing the process, the manufacturing fineness of each process product in each process link is judged through industrial process data, and the yield of the process product can be effectively predicted.
In the prior art, the detection is mainly realized through a traditional machine learning algorithm such as a support vector machine, and although the yield detection of the technical product can be realized, the detection accuracy is poor due to insufficient utilization of industrial process data by the traditional machine learning algorithm.
Disclosure of Invention
The invention provides a method and a device for analyzing process anomaly and a computer readable storage medium, and mainly aims to solve the technical problem of poor detection accuracy caused by insufficient utilization of industrial process data in the prior art.
In order to achieve the above object, the present invention provides a method for analyzing process anomaly, comprising:
receiving a process procedure training set and a label set, and performing characteristic classification on the process procedure training set to obtain a plurality of groups of procedure characteristic training sets;
constructing a plurality of groups of corresponding isolated tree models according to the plurality of groups of process characteristic training sets;
sequentially inputting each group of process characteristic training sets to the corresponding isolated tree models to perform training, and selecting the isolated tree models with training results meeting preset requirements by using the label sets to obtain one or more groups of high-quality tree models;
receiving a process procedure data set to be judged as abnormal, inputting the process procedure data set to be judged as abnormal into the high-quality tree model to perform abnormal analysis, and obtaining a process procedure data ranking;
and extracting the process data with the process data rank higher than a preset threshold value to obtain an abnormal analysis result.
Optionally, the constructing a plurality of corresponding sets of isolated tree models according to the plurality of sets of process characteristic training sets includes:
constructing a target function of the isolated tree model according to the type of the label set;
optimizing the objective function by using a Taylor function to obtain an optimized function;
constructing an information entropy calculation method corresponding to the multiple groups of process characteristic training sets;
and integrating the optimization function and the information entropy calculation method into a plurality of groups of pre-constructed CART trees to obtain a plurality of groups of isolated tree models.
Optionally, the performing feature classification on the process recipe training set to obtain a plurality of sets of recipe feature training sets includes:
extracting a process name from the process training set to obtain a process name set;
according to different process products, performing product classification on the process name set to obtain a product-process name set;
and dividing the process procedure training set according to the corresponding relation between the product-procedure name set and the process procedure training set to obtain a plurality of groups of procedure characteristic training sets.
Optionally, the optimizing the objective function by using a taylor function to obtain an optimization function includes:
receiving a Taylor expansion series input by a user, and solving partial derivative functions of the target functions with the same number as the Taylor expansion series;
and substituting the same number of partial derivative functions into the Taylor function to obtain the optimization function.
Optionally, the sequentially inputting each group of the process characteristic training sets to the corresponding isolated tree models to perform training, and selecting the isolated tree models with training results meeting preset requirements by using the label set to obtain one or more groups of high-quality tree models, including:
calculating the information entropy value of the process characteristic training set by using the information entropy calculation method;
performing path division on the process characteristic training set in the CART tree according to the information entropy to obtain path length;
calculating the evaluation value of the path length according to a pre-constructed evaluation function;
calculating a call value of the evaluation value and the label set;
and starting the optimization function to optimize the CART tree by combining the call value to obtain one or more groups of high-quality tree models.
Optionally, the starting the optimization function to optimize the CART tree in combination with the call value to obtain one or more sets of the quality tree models includes:
judging the size relationship between the call value and a preset call threshold value;
if the call value is larger than or equal to the call threshold value, starting the optimization function to optimize the internal parameters of the CART tree;
and obtaining one or more groups of high-quality tree models until the recall value is smaller than the recall threshold value.
Optionally, the calculating the information entropy value of the process characteristic training set by using the information entropy calculation method includes:
extracting all process characteristics included in the process characteristic training set, and combining all the process characteristics to obtain a process characteristic arrangement set;
calculating the information entropy value of each group of arranged process characteristics in the process characteristic arrangement set by using the information entropy calculation method to obtain an information entropy value set corresponding to the process characteristic arrangement set;
and extracting the information entropy value with the minimum value from the information entropy value set to obtain the information entropy value of the process characteristic training set.
In order to solve the above problems, the present invention further provides an apparatus for analyzing process anomaly, the apparatus comprising:
the characteristic classification module is used for receiving a process procedure training set and a label set, and performing characteristic classification on the process procedure training set to obtain a plurality of groups of procedure characteristic training sets;
the model construction module is used for constructing a plurality of groups of corresponding isolated tree models according to the plurality of groups of process characteristic training sets;
the model training module is used for sequentially inputting each group of process characteristic training sets to the corresponding isolated tree models to perform training, and selecting the isolated tree models with the training results meeting the preset requirements by using the label sets to obtain one or more groups of high-quality tree models;
the anomaly analysis module is used for receiving a process procedure data set to be judged, inputting the process procedure data set to be judged to be abnormal into the high-quality tree model to execute anomaly analysis, and obtaining a process procedure data ranking;
and the abnormal data extraction module is used for extracting the process procedure data with the process procedure data ranking higher than the preset threshold value to obtain an abnormal analysis result.
In order to solve the above problem, the present invention also provides an electronic device, including:
a memory storing at least one instruction; and
and the processor executes the instructions stored in the memory to realize the process anomaly analysis method.
In order to solve the above problem, the present invention further provides a computer-readable storage medium, in which at least one instruction is stored, and the at least one instruction is executed by a processor in an electronic device to implement the method for analyzing process recipe exception described above.
Compared with the background art: the invention discloses a method for detecting the defect of poor detection accuracy by using a traditional machine learning algorithm, which comprises the steps of receiving a process procedure training set and a label set, classifying the process procedure training set by executing characteristics to obtain a plurality of groups of procedure characteristic training sets in order to improve the subsequent full utilization of the process procedure data, establishing a plurality of groups of corresponding isolated tree models according to the plurality of groups of procedure characteristic training sets on the basis of the purpose of fully utilizing each group of procedure characteristic training sets and avoiding the defects of the traditional machine learning, sequentially inputting each group of procedure characteristic training sets to the corresponding isolated tree models to execute training, and selecting the isolated tree models with the training results meeting the preset requirements by using the label set, one or more sets of quality tree models are obtained. And finally, receiving a process procedure data set to be judged as abnormal, inputting the process procedure data set to be judged as abnormal into the high-quality tree model to perform abnormal analysis to obtain a process procedure data ranking, and extracting the process procedure data with the process procedure data ranking higher than a preset threshold value to obtain an abnormal analysis result. Therefore, the method, the device, the electronic equipment and the computer readable storage medium for analyzing the process procedure abnormity can solve the problem of consuming excessive resources such as manpower and material resources when special equipment is managed.
Drawings
Fig. 1 is a schematic flow chart illustrating a method for analyzing process anomalies according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart showing a detailed implementation of one of the steps in FIG. 1;
FIG. 3 is a schematic flow chart showing another step of FIG. 1;
FIG. 4 is a functional block diagram of an apparatus for analyzing process anomalies according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device for implementing the method for analyzing process anomaly according to an embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The embodiment of the application provides a method for analyzing process anomaly. The execution subject of the process anomaly analysis method includes, but is not limited to, at least one of electronic devices such as a server and a terminal, which can be configured to execute the method provided by the embodiments of the present application. In other words, the method for analyzing the process recipe abnormality may be implemented by software or hardware installed in the terminal equipment or the server equipment, and the software may be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like.
Fig. 1 is a schematic flow chart of a method for analyzing process anomaly according to an embodiment of the present invention. In this embodiment, the method for analyzing process anomalies includes:
and S1, receiving the process procedure training set and the label set, and performing characteristic classification on the process procedure training set to obtain a plurality of groups of procedure characteristic training sets.
It should be explained that, in the panel manufacturing process, the process time is one of the key factors affecting the yield of the panel, wherein the process time mainly includes process time (product production time), Qtime (time interval between two production processes of a product), DelayTime (idle time of equipment), and the like, and each product has a different process name. Therefore, in the embodiment of the present invention, the process training set includes a process name and a process time, for example, the process name is a product a-process time and a corresponding process time, the product B-process name Qtime and a corresponding time, and the product C-process name DelayTime and a corresponding time.
Further, the tag sets and the process procedure training sets have a corresponding relationship, each tag set corresponds to the normal process time and the abnormal process time of each process, and for example, in the process procedure training sets with the process procedure name process time and the corresponding process time, 1000 sets are provided, wherein the process time of the 50 th set of products X is 32 minutes, but the process time of the 50 th set of products X is recorded as the abnormal process time of 32 minutes in the tag sets.
In detail, the performing feature classification on the process recipe training set to obtain a plurality of sets of recipe feature training sets includes:
extracting a process name from the process training set to obtain a process name set;
according to different process products, performing product classification on the process name set to obtain a product-process name set;
and dividing the process procedure training set according to the corresponding relation between the product-procedure name set and the process procedure training set to obtain a plurality of groups of procedure characteristic training sets.
Illustratively, the process recipe training set has 10000 groups, which include product a, product B, product C, and product X, where product a includes 3 process recipes, product B includes 2 process recipes, product C includes 4 process recipes, and product X includes 6 process recipes. Therefore, 2000 sets of process feature training sets of the product A-ProcessTime, the product A-Qtime and the product A-DelayTime, 1000 sets of process feature training sets of the product B-ProcessTime and the product B-Qtime and the like are obtained according to the corresponding relation between the product and the process name, and the like.
And S2, constructing a plurality of corresponding isolated tree models according to the plurality of groups of process characteristic training sets.
In detail, referring to fig. 2, the constructing a plurality of sets of corresponding isolated tree models according to the plurality of sets of process characteristic training sets includes:
s21, constructing a target function of the isolated tree model according to the type of the label set;
s22, optimizing the objective function by using a Taylor function to obtain an optimization function;
s23, constructing information entropy calculation methods corresponding to the multiple groups of process characteristic training sets;
s24, integrating the optimization function and the information entropy calculation method into a plurality of groups of pre-constructed CART trees to obtain a plurality of groups of isolated tree models.
Illustratively, if the tag set is normal process time and abnormal process time, the tag set is classified into two categories, and an objective function of the two categories is correspondingly constructed, and further, the objective function is as follows:
wherein Tree represents the CART Tree,
representing the objective function
The target value of (a) is determined,
representing the number of CART trees into which the objective function is integrated,
representing the process feature training set
Training data.
Further, the optimizing the objective function by using the taylor function to obtain an optimization function includes:
receiving a Taylor expansion series input by a user, and solving partial derivative functions of the target functions with the same number as the Taylor expansion series;
and substituting the same number of partial derivative functions into the Taylor function to obtain the optimization function.
Illustratively, if the taylor expansion series input by the user is 2, a 2 nd order partial derivative function for solving the objective function is expressed and is substituted into the taylor function, so as to obtain the following optimization function:
wherein the content of the first and second substances,
training data representing the training set of process features,
an increment representing the training data is represented by a delta,
as the objective function
In the short-hand form of (1),
a first order partial derivative function representing the objective function,
a second order partial derivative function representing the objective function.
Further, the information entropy is a measure index of data disorder or uncertainty, and in the embodiment of the present invention, a calculation method of the information entropy of the CART tree is used, which is not described herein again.
It should be understood that the optimization function and the information entropy calculation method are integrated into a plurality of sets of pre-constructed CART trees to obtain a plurality of sets of the isolated tree models, wherein each set of the isolated tree models is still a CART tree model in nature, internal parameters of the CART tree need to be trained through a process characteristic training set, and the information entropy calculation method is used for guiding the optimization function to optimize the internal parameters of the CART tree, so that a model training process is completed.
And S3, sequentially inputting each group of process characteristic training sets to the corresponding isolated tree models to perform training, and selecting the isolated tree models with the training results meeting the preset requirements by using the label sets to obtain one or more groups of high-quality tree models.
In detail, referring to fig. 3, the sequentially inputting each set of the process characteristic training sets to the corresponding isolated tree models to perform training, and selecting the isolated tree models with training results meeting preset requirements by using the label set to obtain one or more sets of high-quality tree models includes:
s31, calculating the information entropy value of the process characteristic training set by using the information entropy calculation method;
s32, performing path division on the process characteristic training set in the CART tree according to the information entropy to obtain path length;
s33, calculating the evaluation value of the path length according to a pre-constructed evaluation function;
s34, calculating the call values of the evaluation value and the label set;
and S35, starting the optimization function to optimize the CART tree by combining the call value to obtain one or more groups of high-quality tree models.
In detail, the calculating the information entropy value of the process characteristic training set by using the information entropy calculation method includes:
extracting all process characteristics included in the process characteristic training set, and combining all the process characteristics to obtain a process characteristic arrangement set;
calculating the information entropy value of each group of arranged process characteristics in the process characteristic arrangement set by using the information entropy calculation method to obtain an information entropy value set corresponding to the process characteristic arrangement set;
and extracting the information entropy value with the minimum value from the information entropy value set to obtain the information entropy value of the process characteristic training set.
It should be explained that, in the embodiment of the present invention, the information entropy of each set of process characteristic training set can be calculated by using the information entropy calculation method of the CART tree, for example, the process characteristic training set of the product a is 2000 sets of the product a-process time, the product a-Qtime, and the product a-DelayTime, if the information entropy calculation is performed according to the process time, the Qtime, and the DelayTime, the information entropy of the process characteristic training set of the product a is 0.8, and if the information entropy calculation is performed according to the Qtime, the process time, and the DelayTime, the information entropy of the process characteristic training set of the product a is finally determined to be 0.2.
Further, in the embodiments of the present invention, the path division is performed by using huffman coding to obtain the path length, wherein the huffman coding is a disclosed technical implementation means and is not described herein again.
In addition, the invention calculates the evaluation value of the path length by using a square error function or MSE calculation method, and calculates the square of the difference value between the evaluation value and the label set to obtain the call value.
In detail, the step of starting the optimization function to optimize the CART tree in combination with the call value to obtain one or more sets of the quality tree models includes:
judging the size relationship between the call value and a preset call threshold value;
if the call value is larger than or equal to the call threshold value, starting the optimization function to optimize the internal parameters of the CART tree;
and obtaining one or more groups of high-quality tree models until the recall value is smaller than the recall threshold value.
S4, receiving a process procedure data set to be judged as abnormal, inputting the process procedure data set to be judged as abnormal into the high-quality tree model to execute abnormal analysis, and obtaining a process procedure data ranking.
For example, if a process recipe data set corresponding to the product Y is received, where the process recipe data set includes first executing the process time, then executing the Qtime, and finally executing the process recipe of the DelayTime, and also includes directly executing the process recipe of the DelayTime after executing the process time, the process recipe arrangement of each group is input to the trained good quality tree model to obtain the information entropy of the process recipe arrangement of each group, and the ranking is executed on the process recipe arrangement according to the size of the information entropy to obtain the process recipe data ranking.
It should be noted that, in the embodiment of the present invention, the received process recipe data set with the abnormality to be determined is ranked according to the order of the entropy values of the information from large to small, so that the data set with the top rank has the highest probability of abnormality.
And S5, extracting the process data with the process data rank higher than a preset threshold value to obtain an abnormal analysis result.
For example, if the process sequence corresponding to the top 3 ranked process sequences are extracted, the data sets of the process sequence corresponding to the top 3 process sequences are summarized to obtain abnormal process sequence data.
Compared with the background art: the invention discloses a method for detecting the defect of poor detection accuracy by using a traditional machine learning algorithm, which comprises the steps of receiving a process procedure training set and a label set, classifying the process procedure training set by executing characteristics to obtain a plurality of groups of procedure characteristic training sets in order to improve the subsequent full utilization of the process procedure data, establishing a plurality of groups of corresponding isolated tree models according to the plurality of groups of procedure characteristic training sets on the basis of the purpose of fully utilizing each group of procedure characteristic training sets and avoiding the defects of the traditional machine learning, sequentially inputting each group of procedure characteristic training sets to the corresponding isolated tree models to execute training, and selecting the isolated tree models with the training results meeting the preset requirements by using the label set, one or more sets of quality tree models are obtained. And finally, receiving a process procedure data set to be judged as abnormal, inputting the process procedure data set to be judged as abnormal into the high-quality tree model to perform abnormal analysis to obtain a process procedure data ranking, and extracting the process procedure data with the process procedure data ranking higher than a preset threshold value to obtain an abnormal analysis result. Therefore, the method, the device, the electronic equipment and the computer readable storage medium for analyzing the process procedure abnormity can solve the problem of consuming excessive resources such as manpower and material resources when special equipment is managed.
Fig. 4 is a functional block diagram of an apparatus for analyzing process anomaly according to an embodiment of the present invention.
The process anomaly analysis device 100 of the present invention can be installed in an electronic device. According to the implemented functions, the apparatus 100 for analyzing process recipe abnormality may include a feature classification module 101, a model construction module 102, a model training module 103, an abnormality analysis module 104, and an abnormality data extraction module 105. The module of the present invention, which may also be referred to as a unit, refers to a series of computer program segments that can be executed by a processor of an electronic device and that can perform a fixed function, and that are stored in a memory of the electronic device.
The characteristic classification module 101 is configured to receive a process procedure training set and a tag set, perform characteristic classification on the process procedure training set, and obtain a plurality of groups of procedure characteristic training sets;
the model building module 102 is configured to build a plurality of corresponding isolated tree models according to the plurality of sets of process characteristic training sets;
the model training module 103 is configured to sequentially input each set of the process characteristic training sets to the corresponding isolated tree model to perform training, and select an isolated tree model with a training result meeting a preset requirement by using the label set to obtain one or more sets of high-quality tree models;
the anomaly analysis module 104 is configured to receive a process recipe data set to be determined as an anomaly, input the process recipe data set to be determined as an anomaly into the quality tree model, and perform anomaly analysis to obtain a process recipe data rank;
the abnormal data extraction module 105 is configured to extract the process data with the process data rank higher than a preset threshold to obtain an abnormal analysis result.
In detail, in the embodiment of the present invention, when the modules in the process anomaly analysis apparatus 100 are used, the same technical means as the process anomaly analysis method described in fig. 1 above are adopted, and the same technical effects can be produced, which is not described herein again.
Fig. 5 is a schematic structural diagram of an electronic device for implementing a method for analyzing process anomalies according to an embodiment of the present invention.
The electronic device 1 may include a processor 10, a memory 11 and a bus, and may further include a computer program, such as a process recipe abnormality analysis program 12, stored in the memory 11 and executable on the processor 10.
The memory 11 includes at least one type of readable storage medium, which includes flash memory, removable hard disk, multimedia card, card-type memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device 1, such as a removable hard disk of the electronic device 1. The memory 11 may also be an external storage device of the electronic device 1 in other embodiments, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the electronic device 1. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device 1. The memory 11 may be used to store not only the application software installed in the electronic device 1 and various data, such as the code of the abnormal process procedure analysis program 12, but also temporarily store data that has been output or will be output.
The processor 10 may be composed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or may be composed of a plurality of integrated circuits packaged with the same or different functions, including one or more Central Processing Units (CPUs), microprocessors, digital Processing chips, graphics processors, and combinations of various control chips. The processor 10 is a Control Unit (Control Unit) of the electronic device, connects various components of the electronic device by using various interfaces and lines, and executes various functions and processes data of the electronic device 1 by running or executing programs or modules (e.g., process recipe abnormality analysis programs, etc.) stored in the memory 11 and calling data stored in the memory 11.
The bus may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is arranged to enable connection communication between the memory 11 and at least one processor 10 or the like.
Fig. 5 only shows an electronic device with components, and it will be understood by a person skilled in the art that the structure shown in fig. 5 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than shown, or a combination of certain components, or a different arrangement of components.
For example, although not shown, the electronic device 1 may further include a power supply (such as a battery) for supplying power to each component, and preferably, the power supply may be logically connected to the at least one processor 10 through a power management device, so as to implement functions of charge management, discharge management, power consumption management, and the like through the power management device. The power supply may also include any component of one or more dc or ac power sources, recharging devices, power failure detection circuitry, power converters or inverters, power status indicators, and the like. The electronic device 1 may further include various sensors, a bluetooth module, a Wi-Fi module, and the like, which are not described herein again.
Further, the electronic device 1 may further include a network interface, and optionally, the network interface may include a wired interface and/or a wireless interface (such as a WI-FI interface, a bluetooth interface, etc.), which are generally used for establishing a communication connection between the electronic device 1 and other electronic devices.
Optionally, the electronic device 1 may further comprise a user interface, which may be a Display (Display), an input unit (such as a Keyboard), and optionally a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, or the like. The display, which may also be referred to as a display screen or display unit, is suitable for displaying information processed in the electronic device 1 and for displaying a visualized user interface, among other things.
It is to be understood that the described embodiments are for purposes of illustration only and that the scope of the appended claims is not limited to such structures.
The process recipe abnormality analysis program 12 stored in the memory 11 of the electronic device 1 is a combination of a plurality of instructions, and when running in the processor 10, can realize:
receiving a process procedure training set and a label set, and performing characteristic classification on the process procedure training set to obtain a plurality of groups of procedure characteristic training sets;
constructing a plurality of groups of corresponding isolated tree models according to the plurality of groups of process characteristic training sets;
sequentially inputting each group of process characteristic training sets to the corresponding isolated tree models to perform training, and selecting the isolated tree models with training results meeting preset requirements by using the label sets to obtain one or more groups of high-quality tree models;
receiving a process procedure data set to be judged as abnormal, inputting the process procedure data set to be judged as abnormal into the high-quality tree model to perform abnormal analysis, and obtaining a process procedure data ranking;
and extracting the process data with the process data rank higher than a preset threshold value to obtain an abnormal analysis result.
Specifically, the specific implementation method of the processor 10 for the instruction may refer to the description of the relevant steps in the embodiments corresponding to fig. 1 to fig. 5, which is not repeated herein.
Further, the integrated modules/units of the electronic device 1, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. The computer readable storage medium may be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying said computer program code, recording medium, U-disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM).
The present invention also provides a computer-readable storage medium, storing a computer program which, when executed by a processor of an electronic device, may implement:
receiving a process procedure training set and a label set, and performing characteristic classification on the process procedure training set to obtain a plurality of groups of procedure characteristic training sets;
constructing a plurality of groups of corresponding isolated tree models according to the plurality of groups of process characteristic training sets;
sequentially inputting each group of process characteristic training sets to the corresponding isolated tree models to perform training, and selecting the isolated tree models with training results meeting preset requirements by using the label sets to obtain one or more groups of high-quality tree models;
receiving a process procedure data set to be judged as abnormal, inputting the process procedure data set to be judged as abnormal into the high-quality tree model to perform abnormal analysis, and obtaining a process procedure data ranking;
and extracting the process data with the process data rank higher than a preset threshold value to obtain an abnormal analysis result.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method can be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is only one logical functional division, and other divisions may be realized in practice.
The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
In addition, functional modules in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional module.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof.
The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
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.
Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the system claims may also be implemented by one unit or means in software or hardware. The terms second, etc. are used to denote names, but not any particular order.
Finally, it should be noted that the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.
Claims (9)
1. A method for analyzing process anomaly, the method comprising:
receiving a process procedure training set and a label set, and performing characteristic classification on the process procedure training set to obtain a plurality of groups of procedure characteristic training sets;
constructing a target function of the isolated tree model according to the type of the label set;
optimizing the objective function by using a Taylor function to obtain an optimized function;
constructing an information entropy calculation method corresponding to the multiple groups of process characteristic training sets;
integrating the optimization function and the information entropy calculation method into a plurality of groups of pre-constructed CART trees to obtain a plurality of groups of isolated tree models;
sequentially inputting each group of process characteristic training sets to the corresponding isolated tree models to perform training, and selecting the isolated tree models with training results meeting preset requirements by using the label sets to obtain one or more groups of high-quality tree models;
receiving a process procedure data set to be judged as abnormal, inputting the process procedure data set to be judged as abnormal into the high-quality tree model to perform abnormal analysis, and obtaining a process procedure data ranking;
and extracting the process data with the process data rank higher than a preset threshold value to obtain an abnormal analysis result.
2. The method of claim 1, wherein the performing feature classification on the process recipe training set to obtain a plurality of sets of recipe feature training sets comprises:
extracting a process name from the process training set to obtain a process name set;
according to different process products, performing product classification on the process name set to obtain a product-process name set;
and dividing the process procedure training set according to the corresponding relation between the product-procedure name set and the process procedure training set to obtain a plurality of groups of procedure characteristic training sets.
3. The method of claim 1, wherein the optimizing the objective function using the taylor function to obtain an optimization function comprises:
receiving a Taylor expansion series input by a user, and solving partial derivative functions of the target functions with the same number as the Taylor expansion series;
and substituting the same number of partial derivative functions into the Taylor function to obtain the optimization function.
4. The method of claim 1, wherein the step of sequentially inputting each set of the process characteristic training sets to the corresponding isolated tree models to perform training and selecting the isolated tree models with training results meeting preset requirements by using the label set to obtain one or more sets of quality tree models comprises:
calculating the information entropy value of the process characteristic training set by using the information entropy calculation method;
performing path division on the process characteristic training set in the CART tree according to the information entropy to obtain path length;
calculating the evaluation value of the path length according to a pre-constructed evaluation function;
calculating a call value of the evaluation value and the label set;
and starting the optimization function to optimize the CART tree by combining the call value to obtain one or more groups of high-quality tree models.
5. The method of claim 4, wherein the initiating the optimization function in conjunction with the call value to optimize the CART tree to obtain one or more sets of the quality tree models comprises:
judging the size relationship between the call value and a preset call threshold value;
if the call value is larger than or equal to the call threshold value, starting the optimization function to optimize the internal parameters of the CART tree;
and obtaining one or more groups of high-quality tree models until the recall value is smaller than the recall threshold value.
6. The method for analyzing process recipe exceptions according to claim 4, wherein the calculating the entropy of information of the training set of recipe features using the entropy of information calculation method comprises:
extracting all process characteristics included in the process characteristic training set, and combining all the process characteristics to obtain a process characteristic arrangement set;
calculating the information entropy value of each group of arranged process characteristics in the process characteristic arrangement set by using the information entropy calculation method to obtain an information entropy value set corresponding to the process characteristic arrangement set;
and extracting the information entropy value with the minimum value from the information entropy value set to obtain the information entropy value of the process characteristic training set.
7. An apparatus for analyzing process anomaly, the apparatus comprising:
the characteristic classification module is used for receiving a process procedure training set and a label set, and performing characteristic classification on the process procedure training set to obtain a plurality of groups of procedure characteristic training sets;
the model building module is used for building a target function of the isolated tree model according to the type of the tag set;
optimizing the objective function by using a Taylor function to obtain an optimized function;
constructing an information entropy calculation method corresponding to the multiple groups of process characteristic training sets;
integrating the optimization function and the information entropy calculation method into a plurality of groups of pre-constructed CART trees to obtain a plurality of groups of isolated tree models;
the model training module is used for sequentially inputting each group of process characteristic training sets to the corresponding isolated tree models to perform training, and selecting the isolated tree models with the training results meeting the preset requirements by using the label sets to obtain one or more groups of high-quality tree models;
the anomaly analysis module is used for receiving a process procedure data set to be judged, inputting the process procedure data set to be judged to be abnormal into the high-quality tree model to execute anomaly analysis, and obtaining a process procedure data ranking;
and the abnormal data extraction module is used for extracting the process procedure data with the process procedure data ranking higher than the preset threshold value to obtain an abnormal analysis result.
8. An electronic device, characterized in that the electronic device comprises:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.
9. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the method of analyzing process recipe exceptions according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111381935.5A CN113822379B (en) | 2021-11-22 | 2021-11-22 | Process process anomaly analysis method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111381935.5A CN113822379B (en) | 2021-11-22 | 2021-11-22 | Process process anomaly analysis method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113822379A CN113822379A (en) | 2021-12-21 |
| CN113822379B true CN113822379B (en) | 2022-02-22 |
Family
ID=78917944
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111381935.5A Active CN113822379B (en) | 2021-11-22 | 2021-11-22 | Process process anomaly analysis method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113822379B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114331761B (en) * | 2022-03-15 | 2022-07-08 | 浙江万胜智能科技股份有限公司 | Equipment parameter analysis and adjustment method and system for special transformer acquisition terminal |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109214612A (en) * | 2018-11-20 | 2019-01-15 | 广东机场白云信息科技有限公司 | One kind being based on XGBOOST airport traffic spatial and temporal distributions prediction technique |
| KR20190030898A (en) * | 2017-09-15 | 2019-03-25 | 충남대학교산학협력단 | Apparatus for anomaly detection of internet of things device based on data distribution service using machine learning |
| CN111178396A (en) * | 2019-12-12 | 2020-05-19 | 国网北京市电力公司 | Method and device for identifying abnormal electricity consumption user |
| CN111338897A (en) * | 2020-02-24 | 2020-06-26 | 京东数字科技控股有限公司 | Identification method of abnormal node in application host, monitoring equipment and electronic equipment |
| CN111475804A (en) * | 2020-03-05 | 2020-07-31 | 浙江省北大信息技术高等研究院 | Alarm prediction method and system |
| CN111598438A (en) * | 2020-05-14 | 2020-08-28 | 哈尔滨工业大学（威海） | Civil aviation engine gas circuit abnormity detection method based on segmented fitting analysis and evaluation |
| CN111669368A (en) * | 2020-05-07 | 2020-09-15 | 宜通世纪科技股份有限公司 | End-to-end network sensing abnormity detection and analysis method, system, device and medium |
| CN111931834A (en) * | 2020-07-31 | 2020-11-13 | 广东工业大学 | Method, equipment and storage medium for detecting abnormal flow data in aluminum profile extrusion process based on isolated forest algorithm |
| TWI738016B (en) * | 2019-06-26 | 2021-09-01 | 元翎精密工業股份有限公司 | Intelligent learning mechanism and adaptive regulation system for liquid gas filling |
| CN113454630A (en) * | 2019-02-07 | 2021-09-28 | 维莎公司 | Method and system for detecting manufacturing process violations in the manufacture of three-dimensional parts |
| EP3896543A1 (en) * | 2020-04-15 | 2021-10-20 | Siemens Aktiengesellschaft | Device for evaluating a classification made for a measured data point |
| CN113537642A (en) * | 2021-08-20 | 2021-10-22 | 日月光半导体制造股份有限公司 | Product quality prediction method, device, electronic equipment and storage medium |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190087737A1 (en) * | 2017-09-18 | 2019-03-21 | Callidus Software, Inc. | Anomaly detection and automated analysis in systems based on fully masked weighted directed |
| CN108777873B (en) * | 2018-06-04 | 2021-03-02 | 江南大学 | Wireless sensor network abnormal data detection method based on weighted mixed isolated forest |
| US11972334B2 (en) * | 2019-08-13 | 2024-04-30 | Sony Corporation | Method and apparatus for generating a combined isolation forest model for detecting anomalies in data |
| CN111833172A (en) * | 2020-05-25 | 2020-10-27 | 百维金科(上海)信息科技有限公司 | Consumption credit fraud detection method and system based on isolated forest |
| CN113205125A (en) * | 2021-04-27 | 2021-08-03 | 河海大学 | XGboost-based extra-high voltage converter valve operation state evaluation method |
| CN113112188B (en) * | 2021-05-14 | 2022-05-17 | 北京邮电大学 | Power dispatching monitoring data anomaly detection method based on pre-screening dynamic integration |
| CN113420073A (en) * | 2021-08-23 | 2021-09-21 | 平安科技（深圳）有限公司 | Abnormal sample detection method based on improved isolated forest and related equipment |
-
2021
- 2021-11-22 CN CN202111381935.5A patent/CN113822379B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20190030898A (en) * | 2017-09-15 | 2019-03-25 | 충남대학교산학협력단 | Apparatus for anomaly detection of internet of things device based on data distribution service using machine learning |
| CN109214612A (en) * | 2018-11-20 | 2019-01-15 | 广东机场白云信息科技有限公司 | One kind being based on XGBOOST airport traffic spatial and temporal distributions prediction technique |
| CN113454630A (en) * | 2019-02-07 | 2021-09-28 | 维莎公司 | Method and system for detecting manufacturing process violations in the manufacture of three-dimensional parts |
| TWI738016B (en) * | 2019-06-26 | 2021-09-01 | 元翎精密工業股份有限公司 | Intelligent learning mechanism and adaptive regulation system for liquid gas filling |
| CN111178396A (en) * | 2019-12-12 | 2020-05-19 | 国网北京市电力公司 | Method and device for identifying abnormal electricity consumption user |
| CN111338897A (en) * | 2020-02-24 | 2020-06-26 | 京东数字科技控股有限公司 | Identification method of abnormal node in application host, monitoring equipment and electronic equipment |
| CN111475804A (en) * | 2020-03-05 | 2020-07-31 | 浙江省北大信息技术高等研究院 | Alarm prediction method and system |
| EP3896543A1 (en) * | 2020-04-15 | 2021-10-20 | Siemens Aktiengesellschaft | Device for evaluating a classification made for a measured data point |
| CN111669368A (en) * | 2020-05-07 | 2020-09-15 | 宜通世纪科技股份有限公司 | End-to-end network sensing abnormity detection and analysis method, system, device and medium |
| CN111598438A (en) * | 2020-05-14 | 2020-08-28 | 哈尔滨工业大学（威海） | Civil aviation engine gas circuit abnormity detection method based on segmented fitting analysis and evaluation |
| CN111931834A (en) * | 2020-07-31 | 2020-11-13 | 广东工业大学 | Method, equipment and storage medium for detecting abnormal flow data in aluminum profile extrusion process based on isolated forest algorithm |
| CN113537642A (en) * | 2021-08-20 | 2021-10-22 | 日月光半导体制造股份有限公司 | Product quality prediction method, device, electronic equipment and storage medium |
Non-Patent Citations (4)
| Title |
|---|
| Fault detection by an ensemble framework of Extreme Gradient Boosting（XGBoost） in the operation of offshore wind turbines;Pavlos Trizoglou 等;《Renewable Energy》;20210723;第945-962页 * |
| Flotation Fault Diagnosis Method Using Statistical Approaches;Jin Zhang 等;《2019 Seventh International Conference on Advanced Cloud and Big Data》;20191128;第266-271页 * |
| 基于多次抽样和维度熵的异常点检测算法研究;罗斌;《中国优秀博硕士学位论文全文数据库（硕士）信息科技辑》;20200115;第I138-906页 * |
| 基于孤立森林采样策略的企业异常用水模式检测;林青轩 等;《复杂系统与复杂性科学》;20200930;第17卷(第3期);第47-51页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113822379A (en) | 2021-12-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112035453A (en) | GBDT high-order feature combination-based recommendation method and device and storage medium | |
| CN113592019B (en) | Fault detection method, device, equipment and medium based on multi-model fusion | |
| CN113780691A (en) | Data testing method and device, electronic equipment and storage medium | |
| CN112541745A (en) | User behavior data analysis method and device, electronic equipment and readable storage medium | |
| CN111694844B (en) | Enterprise operation data analysis method and device based on configuration algorithm and electronic equipment | |
| CN111652280A (en) | Behavior-based target object data analysis method and device and storage medium | |
| CN111652279A (en) | Behavior evaluation method and device based on time sequence data and readable storage medium | |
| CN112579621A (en) | Data display method and device, electronic equipment and computer storage medium | |
| CN113327136A (en) | Attribution analysis method and device, electronic equipment and storage medium | |
| CN114491047A (en) | Multi-label text classification method and device, electronic equipment and storage medium | |
| CN112463530A (en) | Anomaly detection method and device for micro-service system, electronic equipment and storage medium | |
| CN112528013A (en) | Text abstract extraction method and device, electronic equipment and storage medium | |
| CN111339072B (en) | User behavior-based change value analysis method and device, electronic equipment and medium | |
| CN113504935A (en) | Software development quality evaluation method and device, electronic equipment and readable storage medium | |
| CN113822379B (en) | Process process anomaly analysis method and device, electronic equipment and storage medium | |
| CN116993532A (en) | Method and system for improving preparation efficiency of battery parts | |
| CN113657385B (en) | Data detection method and device of electronic metering device and electronic equipment | |
| CN112486957B (en) | Database migration detection method, device, equipment and storage medium | |
| CN111460293B (en) | Information pushing method and device and computer readable storage medium | |
| CN112101481A (en) | Method, device and equipment for screening influence factors of target object and storage medium | |
| CN112783989A (en) | Data processing method and device based on block chain | |
| CN112346838A (en) | Intelligent scheduling method and device based on Internet of things | |
| CN112541688A (en) | Service data checking method and device, electronic equipment and computer storage medium | |
| CN111652282A (en) | Big data based user preference analysis method and device and electronic equipment | |
| CN115796398A (en) | Intelligent demand analysis method, system, equipment and medium based on electric power materials |
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 |