CN117408574A - Chip production monitoring management method, device, equipment and readable storage medium - Google Patents
Chip production monitoring management method, device, equipment and readable storage medium Download PDFInfo
- Publication number
- CN117408574A CN117408574A CN202311705619.8A CN202311705619A CN117408574A CN 117408574 A CN117408574 A CN 117408574A CN 202311705619 A CN202311705619 A CN 202311705619A CN 117408574 A CN117408574 A CN 117408574A
- Authority
- CN
- China
- Prior art keywords
- data information
- chip
- data
- model
- random forest
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 67
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 66
- 238000007726 management method Methods 0.000 title claims abstract description 56
- 238000003860 storage Methods 0.000 title claims abstract description 14
- 238000004364 calculation method Methods 0.000 claims abstract description 48
- 238000007637 random forest analysis Methods 0.000 claims abstract description 45
- 238000012216 screening Methods 0.000 claims abstract description 42
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 27
- 238000012549 training Methods 0.000 claims abstract description 18
- 230000009467 reduction Effects 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims abstract description 11
- 238000000034 method Methods 0.000 claims description 80
- 239000011159 matrix material Substances 0.000 claims description 49
- 238000003066 decision tree Methods 0.000 claims description 40
- 230000008569 process Effects 0.000 claims description 22
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 claims description 10
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 5
- 238000011156 evaluation Methods 0.000 claims description 5
- 230000000737 periodic effect Effects 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 abstract description 6
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000004891 communication Methods 0.000 description 8
- 235000012431 wafers Nutrition 0.000 description 5
- 230000005236 sound signal Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000001914 filtration Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 230000003211 malignant effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000002360 preparation method Methods 0.000 description 2
- 238000003908 quality control method Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- KLDZYURQCUYZBL-UHFFFAOYSA-N 2-[3-[(2-hydroxyphenyl)methylideneamino]propyliminomethyl]phenol Chemical compound OC1=CC=CC=C1C=NCCCN=CC1=CC=CC=C1O KLDZYURQCUYZBL-UHFFFAOYSA-N 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000739 chaotic effect Effects 0.000 description 1
- 201000001098 delayed sleep phase syndrome Diseases 0.000 description 1
- 208000033921 delayed sleep phase type circadian rhythm sleep disease Diseases 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- ZXQYGBMAQZUVMI-GCMPRSNUSA-N gamma-cyhalothrin Chemical compound CC1(C)[C@@H](\C=C(/Cl)C(F)(F)F)[C@H]1C(=O)O[C@H](C#N)C1=CC=CC(OC=2C=CC=CC=2)=C1 ZXQYGBMAQZUVMI-GCMPRSNUSA-N 0.000 description 1
- 238000000227 grinding Methods 0.000 description 1
- 239000007943 implant Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 238000012946 outsourcing Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- 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
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/088—Non-supervised learning, e.g. competitive learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
Abstract
The invention provides a chip production monitoring management method, a device, equipment and a readable storage medium, which relate to the technical field of chip monitoring management and comprise the steps of obtaining the category corresponding to a chip under each production node and screening out optimal data; training and learning the optimal data by adopting a random forest algorithm to obtain a random forest model, and performing dimension reduction calculation to obtain readable data; setting an initial threshold value, dynamically tracking the temperature characteristics of the chip by using the constructed prediction model, and calculating a residual autocorrelation coefficient between a prediction result and a preset value to obtain a difference result; and carrying out real-time monitoring and management on the chip production process, judging whether the chip is in a continuous safe state, and if so, continuing to adaptively update the model and the threshold value. The chip management method has the beneficial effects that the chip is subjected to standardized processing, so that the management difficulty of the chip is reduced; the operation condition of the equipment is directly determined by monitoring the chip data, and the visual monitoring degree is improved.
Description
Technical Field
The present invention relates to the field of chip monitoring management technologies, and in particular, to a method, an apparatus, a device, and a readable storage medium for chip production monitoring management.
Background
Integrated circuits, also known as microcircuits, microchips, wafers, or chips, are one way of miniaturizing circuits in electronics and are often manufactured on semiconductor wafer surfaces, and the preparation of chips is generally divided into the following steps: 1. a wafer preparation stage, in which the wafer is subjected to back grinding and thinning treatment; 2. wafer dicing and scribing; 3. assembly language, i.e. chip connection wire bonding, etc.
However, because these devices are compact and complex in structure, it is not possible to add redundant monitoring devices to monitor the processing in real time, and it is difficult to monitor the processing of the devices for producing chips. Furthermore, as the production process of chips increasingly depends on outsourcing, the security of the chips faces a great hidden trouble. In the production process of the chip, an attacker may implant malicious circuits, and the malicious circuits may bring risks of chip function change, denial of service, information leakage and the like. Thus, monitoring and managing chip production also becomes indispensible.
Disclosure of Invention
The invention aims to provide a chip production monitoring management method, device and equipment and a readable storage medium, so as to solve the problems. In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
in a first aspect, the present application provides a method for monitoring and managing chip production, including:
acquiring the category corresponding to the chip under each production node, respectively acquiring the data information in each chip based on the category, preprocessing the data information, and screening out the optimal data;
training and learning the optimal data by adopting a random forest algorithm to obtain a random forest model, and performing dimension reduction calculation on the preprocessed data information by utilizing the random forest model to obtain readable data;
constructing a prediction model according to the readable data, and setting an initial threshold value, wherein the method comprises model initialization and coefficient matrix solving;
carrying out dynamic tracking on the temperature characteristics of the chip by using the constructed prediction model to obtain a prediction result; carrying out residual autocorrelation coefficient calculation on the prediction result and a preset value to obtain a difference result;
based on the difference result, carrying out real-time monitoring and management on the chip production process, judging whether the chip is in a continuous safe state, and if so, continuing to adaptively update the model and the threshold value; if not, early warning is carried out.
Preferably, the training and learning are performed on the optimal data by using a random forest algorithm to obtain a random forest model, and the dimension reduction calculation is performed on the preprocessed data information by using the random forest model to obtain readable data, which includes:
setting decision trees with different attributes in the data information in a random mode, constructing decision tree branches according to the decision tree with each attribute, wherein the decision tree branches comprise set root nodes and branch nodes, and distributing multiple branches into one tree, wherein the tree branches comprise a first sample, a second sample and a third sample, and the decision tree output calculation formula is as follows:
MAPE=
wherein MAPE is a decision tree output algorithm model,for the actual data information the data quantity is periodic, +.>Outputting evaluation values for the data information decision tree, wherein n is the set different life after the input decision tree modelThe number of life cycle prediction points;
respectively extracting different data branches of the first sample, the second sample and the third sample, screening out different types of data information, and obtaining first screening data information, second screening data information and third screening data information;
constructing a sample learning library according to the first screening data information, the second screening data information and the third screening data information, inputting the sample learning library into a random forest algorithm for training and learning, and further obtaining a random forest model;
and calculating the data information processed by the random forest model by using an unsupervised learning algorithm to obtain readable data.
Preferably, the constructed prediction model is utilized to dynamically track the temperature characteristics of the chip, so as to obtain a prediction result; residual autocorrelation coefficient calculation is carried out on the prediction result and a preset value to obtain a difference result, wherein the method comprises the following steps:
the temperature characteristic updated value at the moment K-1 is obtained to predict the temperature characteristic updated value at the moment K, and a first predicted value is obtained, wherein the calculation formula is as follows:
in the method, in the process of the invention,and->The temperature characteristic matrix after correction and update at the moment k-1 and the temperature characteristic estimation matrix at the moment k-1 and the moment k are respectively +.>The initial matrix is +.>A and B are coefficient matrixes;
correcting and updating the first predicted value by using the actually measured temperature characteristic matrix to obtain a second predicted value;
and solving a deviation between the second predicted value and the actual running state, namely a residual matrix, wherein a calculation formula is as follows:
in the method, in the process of the invention,for residual matrix +.>The temperature characteristic matrix after actual noise reduction at the moment k;
comparing the deviation with a preset threshold value, and if the difference value is smaller than the preset value, preliminarily confirming safety; if the difference is larger than the preset value, the prediction is continued.
Preferably, the data information includes any one or any combination of status data information, temperature data information, power data information, usage data information, and process data information of the chip.
In a second aspect, the present application further provides a chip production monitoring management device, including an acquisition module, a first calculation module, a construction module, a second calculation module and a judgment module, where:
the acquisition module is used for: the method comprises the steps of acquiring category types corresponding to chips under each production node, respectively acquiring data information in each chip based on the category types, preprocessing the data information, and screening out optimal data;
a first calculation module: the method comprises the steps of training and learning the optimal data by adopting a random forest algorithm to obtain a random forest model, and performing dimension reduction calculation on preprocessed data information by utilizing the random forest model to obtain readable data;
the construction module comprises: the method comprises the steps of constructing a prediction model according to readable data, and setting an initial threshold value, wherein the method comprises model initialization and coefficient matrix solving;
a second calculation module: the method comprises the steps of carrying out dynamic tracking on temperature characteristics of a chip by using the constructed prediction model to obtain a prediction result; carrying out residual autocorrelation coefficient calculation on the prediction result and a preset value to obtain a difference result;
and a judging module: the method is used for carrying out real-time monitoring and management on the chip production process based on the difference result, judging whether the chip is in a continuous safe state, and if so, continuing to adaptively update the model and the threshold value; if not, early warning is carried out.
In a third aspect, the present application further provides a chip production monitoring management apparatus, including:
a memory for storing a computer program;
and the processor is used for realizing the steps of the chip production monitoring and management method when executing the computer program.
In a fourth aspect, the present application further provides a readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the above-described chip-based production monitoring management method.
The beneficial effects of the invention are as follows:
the invention realizes the standardized processing of the chip and reduces the management difficulty of the chip; the operation condition of the equipment is directly determined by monitoring the chip data, and the visual monitoring degree is improved.
The invention improves the visual monitoring degree of the chip by improving the random forest model, improves the visual recognition rate and has a certain technical contribution in the aspect of chip production monitoring management; the provided non-supervision learning method can utilize a random forest model to process data information, reduce the number of random variables, obtain a group of processes of irrelevant main variables, extract and synthesize effective information and reject useless information.
The invention carries out noise reduction processing on the acquired data, and the noise reduction processing is carried out on the data in order to better carry out chip safety monitoring and reduce the false alarm rate because the noise affects the situation that burrs are generated and the occurrence of burrs can bring obvious false alarm conditions to the subsequent chip monitoring management.
The invention predicts and monitors in real time through the Kalman filtering algorithm, thereby protecting and monitoring the chip more accurately and providing good guarantee for the management of the chip.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the embodiments of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a chip production monitoring and managing method according to an embodiment of the invention;
fig. 2 is a schematic structural diagram of a chip production monitoring and managing device according to an embodiment of the present invention.
In the figure: 701. an acquisition module; 702. a first computing module; 7021. a construction unit; 7022. an extraction unit; 7023. a training unit; 7024. a calculation unit; 703. constructing a module; 704. a second computing module; 7041. an acquisition unit; 7042. an updating unit; 7043. a solving unit; 7044. a comparison unit; 705. a judging module; 800. chip production monitoring and management equipment; 801. a processor; 802. a memory; 803. a multimedia component; 804. an I/O interface; 805. a communication component.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only to distinguish the description, and are not to be construed as indicating or implying relative importance.
Example 1:
the embodiment provides a chip production monitoring and management method.
Referring to fig. 1, the method is shown to include steps S100, S200, S300, and S400 and S500.
S100, acquiring the category class corresponding to the chip under each production node, respectively acquiring the data information in each chip based on the category class, preprocessing the data information, and screening out the optimal data.
It will be appreciated that in this step, the chips under the production node will comprise a plurality of nodes, each of which may be provided with a plurality of chips. The types of the chips can be various, and the information corresponding to each type of chip is different.
The production characteristics of the different steps are different, and the influence factors of the device characteristics are also different.
And S200, training and learning the optimal data by adopting a random forest algorithm to obtain a random forest model, and performing dimension reduction calculation on the preprocessed data information by utilizing the random forest model to obtain readable data.
It will be appreciated that the present step S200 includes steps S201, S202, S203 and S204, wherein:
s201, setting decision trees with different attributes in the data information in a random mode, constructing decision tree branches according to the decision tree with each attribute, wherein the decision tree branches comprise set root nodes and branch nodes, and distributing various branches into one tree, wherein the tree branches comprise a first sample, a second sample and a third sample, and the decision tree output calculation formula is as follows:
MAPE=
wherein MAPE is a decision tree output algorithm model,for the actual data information the data quantity is periodic, +.>Outputting evaluation values for the data information decision tree, wherein n is the number of the predicted points of different life cycles after the decision tree model is input;
s202, respectively extracting different data branches of the first sample, the second sample and the third sample, screening out different types of data information, and obtaining first screening data information, second screening data information and third screening data information;
s203, a sample learning library is constructed according to the first screening data information, the second screening data information and the third screening data information, and the sample learning library is input into a random forest algorithm for training and learning, so that a random forest model is obtained;
when the random forest algorithm model is constructed, root nodes and branch nodes are required to be set, various data information is distributed in a tree, the same leaf node is selected according to different data attributes, when the same parameter data information is output, the same leaf node is set to be 1, when the different parameter data information is output, the parameter is set to be 0, and then a sample database is constructed.
S204, calculating the data information processed by the random forest model by using an unsupervised learning algorithm to obtain readable data.
The visual monitoring degree of the data information is improved through the random forest model, so that the data transmission is short, the working efficiency is high, the visual recognition capability is strong, and certain technical contribution is made to the monitoring mode.
S300, constructing a prediction model according to the readable data, and setting an initial threshold, wherein the method comprises model initialization and coefficient matrix solving.
It will be appreciated that in this step, the read data may be taken [ ]) As a frequency matrix, the prediction model used in the invention is a Kalman filtering algorithm, and an error covariance matrix, a process noise covariance and a measurement noise covariance parameter are respectively set. The state transition matrix and the control matrix are solved to obtain each parameter value. Then, the chip temperature starts to be measured and the temperature characteristics are predicted in real time. The temperature is measured to characterize whether a malicious circuit exists in the chip, and the safety state of the chip can be analyzed by measuring the temperature information of the chip in the operation stage, so that whether the chip is activated by other malignant Trojan in the production management process is judged.
It should be noted that, the establishment of the prediction model is: setting a univariate chaotic event sequence asThe time interval is h, phase space reconstruction is carried out, and delay time and embedding dimension are calculated; and inputting the preprocessed readable data into a maximum Lyapunov exponent prediction model for prediction to obtain a final prediction model.
S400, dynamically tracking the temperature characteristics of the chip by using the constructed prediction model to obtain a prediction result; and carrying out residual autocorrelation coefficient calculation on the prediction result and a preset value to obtain a difference result.
The temperature characteristic data is acquired from the chip. And acquiring a temperature characteristic matrix, and processing the temperature characteristic matrix by using a threshold noise reduction method and an average filtering method.
It is understood that S401, S402, S403, and S404 are included in the present step S400, in which:
s401, acquiring a temperature characteristic updated value at the moment K-1, and predicting the temperature characteristic updated value at the moment K to obtain a first predicted value, wherein the calculation formula is as follows:
in the method, in the process of the invention,and->The temperature characteristic matrix after correction and update at the moment k-1 and the temperature characteristic estimation matrix at the moment k-1 and the moment k are respectively +.>The initial matrix is +.>A and B are coefficient matrixes;
s402, correcting and updating the first predicted value by using an actually measured temperature characteristic matrix to obtain a second predicted value;
s403, solving a deviation between the second predicted value and the actual running state, namely a residual matrix, wherein a calculation formula is as follows:
in the method, in the process of the invention,for residual matrix +.>The temperature characteristic matrix after actual noise reduction at the moment k;
s404, comparing the deviation with a preset threshold value, and if the difference value is smaller than the preset value, primarily confirming safety; if the difference is larger than the preset value, the prediction is continued.
After the first predicted value is obtained, the error covariance is predicted, a covariance predicted result is obtained, and the Kalman gain coefficient is updated, wherein the Kalman gain coefficient plays a decisive role, influences the accuracy of the model, and enables the updated value to be more accurate.
It should be noted that, the temperature characteristics can be tracked and predicted through the steps to obtain the difference, if the difference is small, the state is inconsistent with the preset state, and the state is inconsistent, which means that the prediction model is not matched with the actual running state, and this is that the prediction model is compared with the preset value for the first time, and is established in a full-safety state, and if the prediction model is accurate, the prediction model indicates that the chip may have a BUG, so that the model is not matched, and if the malignant circuit is activated, the state is inconsistent, and a great difference exists.
S500, carrying out real-time monitoring and management on the chip production process based on the difference result, judging whether the chip is in a continuous safe state, and if so, continuing to adaptively update the model and the threshold value; if not, early warning is carried out.
It can be understood that in this step, the difference result is judged for the second time, that is, compared with the threshold safety judgment result, to judge whether the chip is in a continuous safe state, if so, the residual autocorrelation is solved, and the result is smaller; if not, early warning and alarming are carried out.
It should be noted that the invention can also use a multi-element quality control chart for monitoring and management, the control chart has the function of identifying the process quality fluctuation generated by random factors or system factors, if unsafe, after the quality process is abnormal, the average value of the sample quantity collected by the system can be changed, and when the system is in a controlled state, the larger the value of the average running chain length is, the better the performance of the control chart is, and the more stable the performance of the control chart is; if the control chart is in a runaway state, the smaller the value of the average running chain length is, the better the value is, and the control chart can find the process deviation faster, so that measures are taken, and the control quality cost is reduced. And monitoring the production quality of the chip by establishing a combined decision model, wherein the combined decision model comprises a multi-element quality control diagram and a maintenance combined model.
The data information includes any one or any combination of state data information, temperature data information, power data information, usage rate data information, and process data information of the chip.
The above index types are merely examples, and in practical applications, the index types corresponding to different chip types are different. The index corresponding to each type of chip may be one or more, and is not limited herein.
Example 2:
the embodiment provides a chip production monitoring and management device, which comprises an acquisition module 701, a first calculation module 702, a construction module 703, a second calculation module 704 and a judgment module 705, wherein:
the acquisition module 701: the method comprises the steps of acquiring category types corresponding to chips under each production node, respectively acquiring data information in each chip based on the category types, preprocessing the data information, and screening out optimal data;
the first calculation module 702: the method comprises the steps of training and learning the optimal data by adopting a random forest algorithm to obtain a random forest model, and performing dimension reduction calculation on preprocessed data information by utilizing the random forest model to obtain readable data;
the construction module 703: the method comprises the steps of constructing a prediction model according to readable data, and setting an initial threshold value, wherein the method comprises model initialization and coefficient matrix solving;
the second calculation module 704: the method comprises the steps of carrying out dynamic tracking on temperature characteristics of a chip by using the constructed prediction model to obtain a prediction result; carrying out residual autocorrelation coefficient calculation on the prediction result and a preset value to obtain a difference result;
the judging module 705: the method is used for carrying out real-time monitoring and management on the chip production process based on the difference result, judging whether the chip is in a continuous safe state, and if so, continuing to adaptively update the model and the threshold value; if not, early warning is carried out.
Specifically, the first computing module 702 includes a constructing unit 7021, an extracting unit 7022, a training unit 7023, and a computing unit 7024, wherein:
building unit 7021: the method is used for setting decision trees with different attributes in the data information in a random mode, constructing decision tree branches according to the decision tree with each attribute, wherein the decision tree branches comprise a set root node and a branch node, and distributing various branches into one tree, wherein the tree branches comprise a first sample, a second sample and a third sample, and the decision tree outputs a calculation formula as follows:
MAPE=
wherein MAPE is a decision tree output algorithm model,for the actual data information the data quantity is periodic, +.>Outputting evaluation values for the data information decision tree, wherein n is the number of the predicted points of different life cycles after the decision tree model is input;
extraction unit 7022: the data processing device is used for respectively extracting different data branches of the first sample, the second sample and the third sample, screening out different types of data information and obtaining first screening data information, second screening data information and third screening data information;
training unit 7023: the method comprises the steps of constructing a sample learning library according to first screening data information, second screening data information and third screening data information, inputting the sample learning library into a random forest algorithm for training and learning, and further obtaining a random forest model;
calculation unit 7024: the method is used for calculating the data information processed by the random forest model by using an unsupervised learning algorithm to obtain readable data.
Specifically, the second calculating module 704 includes an acquiring unit 7041, an updating unit 7042, a solving unit 7043, and a comparing unit 7044, where:
acquisition unit 7041: the method is used for obtaining the temperature characteristic updated value at the moment K-1 to predict the temperature characteristic updated value at the moment K to obtain a first predicted value, and the calculation formula is as follows:
in the method, in the process of the invention,and->The temperature characteristic matrix after correction and update at the moment k-1 and the temperature characteristic estimation matrix at the moment k-1 and the moment k are respectively +.>The initial matrix is +.>A and B are coefficient matrixes;
update unit 7042: the temperature characteristic matrix is used for correcting and updating the first predicted value by utilizing the actually measured temperature characteristic matrix to obtain a second predicted value;
solving section 7043: the calculation formula is as follows, and the calculation formula is used for solving the deviation between the second predicted value and the actual running state, namely a residual matrix:
in the method, in the process of the invention,for residual matrix +.>The temperature characteristic matrix after actual noise reduction at the moment k;
comparison unit 7044: the method comprises the steps of comparing deviation with a preset threshold value, and if the difference value is smaller than the preset value, primarily confirming safety; if the difference is larger than the preset value, the prediction is continued.
Specifically, the data information in the acquisition module includes any one or any combination of state data information, temperature data information, power data information, usage rate data information, and process data information of the chip.
It should be noted that, regarding the apparatus in the above embodiments, the specific manner in which the respective modules perform the operations has been described in detail in the embodiments regarding the method, and will not be described in detail herein.
Example 3:
corresponding to the above method embodiment, a chip production monitoring and management apparatus is also provided in this embodiment, and a chip production monitoring and management apparatus described below and a chip production monitoring and management method described above may be referred to correspondingly with each other.
Fig. 2 is a block diagram of a chip production monitoring management device 800, shown in accordance with an exemplary embodiment. As shown in fig. 2, the chip production monitoring management apparatus 800 includes: a processor 801 and a memory 802. The chip production monitoring management apparatus 800 further includes one or more of a multimedia component 803, an i/O interface 804, and a communication component 805.
The processor 801 is configured to control the overall operation of the chip production monitoring and management apparatus 800 to perform all or part of the steps in the chip production monitoring and management method described above. The memory 802 is used to store various types of data to support the operation of the chip production monitoring management device 800, which may include, for example, instructions for any application or method operating on the chip production monitoring management device 800, as well as application related data such as contact data, messages, pictures, audio, video, and the like. The Memory 802 may be implemented by any type or combination of volatile or non-volatile Memory devices, such as static random access Memory (Static Random Access Memory, SRAM for short), electrically erasable programmable Read-Only Memory (ElectricallyErasable Programmable Read-Only Memory, EEPROM for short), erasable programmable Read-Only Memory (Erasable Programmable Read-Only Memory, EPROM for short), programmable Read-Only Memory (Programmable Read-Only Memory, PROM for short), read-Only Memory (ROM for short), magnetic Memory, flash Memory, magnetic disk, or optical disk. The multimedia component 803 may include a screen and an audio component. Wherein the screen may be, for example, a touch screen, the audio component being for outputting and/or inputting audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in the memory 802 or transmitted through the communication component 805. The audio assembly further comprises at least one speaker for outputting audio signals. The I/O interface 804 provides an interface between the processor 801 and other interface modules, which may be a keyboard, mouse, or buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the chip production monitoring management device 800 and other devices. Wireless communication, such as Wi-Fi, bluetooth, near field communication (Near FieldCommunication, NFC for short), 2G, 3G, or 4G, or a combination of one or more thereof, the respective communication component 805 may thus comprise: wi-Fi module, bluetooth module or NFC module.
In an exemplary embodiment, the chip production monitoring management apparatus 800 may be implemented by one or more application specific integrated circuits (Application Specific Integrated Circuit, abbreviated as ASIC), digital signal processors (DigitalSignal Processor, abbreviated as DSP), digital signal processing apparatus (Digital Signal Processing Device, abbreviated as DSPD), programmable logic devices (ProgrammableLogic Device, abbreviated as PLD), field programmable gate arrays (Field Programmable Gate Array, abbreviated as FPGA), controllers, microcontrollers, microprocessors, or other electronic components for performing the chip production monitoring management methods described above.
In another exemplary embodiment, a computer readable storage medium is also provided that includes program instructions that, when executed by a processor, implement the steps of the chip production monitoring management method described above. For example, the computer readable storage medium may be the memory 802 including the program instructions described above, which are executable by the processor 801 of the chip production monitoring management apparatus 800 to perform the chip production monitoring management method described above.
Example 4:
corresponding to the above method embodiment, a readable storage medium is also provided in this embodiment, and a readable storage medium described below and a chip production monitoring management method described above may be referred to correspondingly.
The readable storage medium stores a computer program which, when executed by a processor, implements the steps of the chip production monitoring management method of the above method embodiment.
The readable storage medium may be a usb disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RandomAccess Memory, RAM), a magnetic disk, or an optical disk, and the like.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (10)
1. The chip production monitoring and managing method is characterized by comprising the following steps:
acquiring the category corresponding to the chip under each production node, respectively acquiring the data information in each chip based on the category, preprocessing the data information, and screening out the optimal data;
training and learning the optimal data by adopting a random forest algorithm to obtain a random forest model, and performing dimension reduction calculation on the preprocessed data information by utilizing the random forest model to obtain readable data;
constructing a prediction model according to the readable data, and setting an initial threshold value, wherein the method comprises model initialization and coefficient matrix solving;
carrying out dynamic tracking on the temperature characteristics of the chip by using the constructed prediction model to obtain a prediction result; carrying out residual autocorrelation coefficient calculation on the prediction result and a preset value to obtain a difference result;
based on the difference result, carrying out real-time monitoring and management on the chip production process, judging whether the chip is in a continuous safe state, and if so, continuing to adaptively update the model and the threshold value; if not, early warning is carried out.
2. The method for monitoring and managing chip production according to claim 1, wherein the training and learning the optimal data by using a random forest algorithm to obtain a random forest model, and performing dimension reduction calculation on the preprocessed data information by using the random forest model to obtain readable data, and the method comprises the steps of:
setting decision trees with different attributes in the data information in a random mode, constructing decision tree branches according to the decision tree with each attribute, wherein the decision tree branches comprise set root nodes and branch nodes, and distributing multiple branches into one tree, wherein the tree branches comprise a first sample, a second sample and a third sample, and the decision tree output calculation formula is as follows:
MAPE=
wherein MAPE is a decision tree output algorithm model,for the actual data information the data quantity is periodic, +.>Outputting evaluation values for the data information decision tree, wherein n is the number of the predicted points of different life cycles after the decision tree model is input;
respectively extracting different data branches of the first sample, the second sample and the third sample, screening out different types of data information, and obtaining first screening data information, second screening data information and third screening data information;
constructing a sample learning library according to the first screening data information, the second screening data information and the third screening data information, inputting the sample learning library into a random forest algorithm for training and learning, and further obtaining a random forest model;
and calculating the data information processed by the random forest model by using an unsupervised learning algorithm to obtain readable data.
3. The chip production monitoring and management method according to claim 1, wherein the constructed prediction model is utilized to dynamically track the temperature characteristics of the chip to obtain a prediction result; residual autocorrelation coefficient calculation is carried out on the prediction result and a preset value to obtain a difference result, wherein the method comprises the following steps:
the temperature characteristic updated value at the moment K-1 is obtained to predict the temperature characteristic updated value at the moment K, and a first predicted value is obtained, wherein the calculation formula is as follows:
in the method, in the process of the invention,and->The temperature characteristic matrix after correction and update at the moment k-1 and the temperature characteristic estimation matrix at the moment k-1 and the moment k are respectively +.>The initial matrix is +.>A and B are coefficient matrixes;
correcting and updating the first predicted value by using the actually measured temperature characteristic matrix to obtain a second predicted value;
and solving a deviation between the second predicted value and the actual running state, namely a residual matrix, wherein a calculation formula is as follows:
in the method, in the process of the invention,for residual matrix +.>The temperature characteristic matrix after actual noise reduction at the moment k;
comparing the deviation with a preset threshold value, and if the difference value is smaller than the preset value, preliminarily confirming safety; if the difference is larger than the preset value, the prediction is continued.
4. The chip production monitoring and management method according to claim 1, wherein the data information includes any one or any combination of status data information, temperature data information, power data information, usage data information, and process data information of the chip.
5. A chip production monitoring and management device, characterized by comprising:
the acquisition module is used for: the method comprises the steps of acquiring category types corresponding to chips under each production node, respectively acquiring data information in each chip based on the category types, preprocessing the data information, and screening out optimal data;
a first calculation module: the method comprises the steps of training and learning the optimal data by adopting a random forest algorithm to obtain a random forest model, and performing dimension reduction calculation on preprocessed data information by utilizing the random forest model to obtain readable data;
the construction module comprises: the method comprises the steps of constructing a prediction model according to readable data, and setting an initial threshold value, wherein the method comprises model initialization and coefficient matrix solving;
a second calculation module: the method comprises the steps of carrying out dynamic tracking on temperature characteristics of a chip by using the constructed prediction model to obtain a prediction result; carrying out residual autocorrelation coefficient calculation on the prediction result and a preset value to obtain a difference result;
and a judging module: the method is used for carrying out real-time monitoring and management on the chip production process based on the difference result, judging whether the chip is in a continuous safe state, and if so, continuing to adaptively update the model and the threshold value; if not, early warning is carried out.
6. The chip production monitoring and management device according to claim 5, wherein the first computing module includes:
the construction unit: the method is used for setting decision trees with different attributes in the data information in a random mode, constructing decision tree branches according to the decision tree with each attribute, wherein the decision tree branches comprise a set root node and a branch node, and distributing various branches into one tree, wherein the tree branches comprise a first sample, a second sample and a third sample, and the decision tree outputs a calculation formula as follows:
MAPE=
wherein MAPE is a decision tree output algorithm model,for the actual data information the data quantity is periodic, +.>Outputting evaluation values for the data information decision tree, wherein n is the number of the predicted points of different life cycles after the decision tree model is input;
extraction unit: the data processing device is used for respectively extracting different data branches of the first sample, the second sample and the third sample, screening out different types of data information and obtaining first screening data information, second screening data information and third screening data information;
training unit: the method comprises the steps of constructing a sample learning library according to first screening data information, second screening data information and third screening data information, inputting the sample learning library into a random forest algorithm for training and learning, and further obtaining a random forest model;
a calculation unit: the method is used for calculating the data information processed by the random forest model by using an unsupervised learning algorithm to obtain readable data.
7. The chip production monitoring and management device according to claim 5, wherein the second calculation module includes:
an acquisition unit: the method is used for obtaining the temperature characteristic updated value at the moment K-1 to predict the temperature characteristic updated value at the moment K to obtain a first predicted value, and the calculation formula is as follows:
in the method, in the process of the invention,and->The temperature characteristic matrix after correction and update at the moment k-1 and the temperature characteristic estimation matrix at the moment k-1 and the moment k are respectively +.>The initial matrix is +.>A and B are coefficient matrixes;
an updating unit: the temperature characteristic matrix is used for correcting and updating the first predicted value by utilizing the actually measured temperature characteristic matrix to obtain a second predicted value;
and a solving unit: the calculation formula is as follows, and the calculation formula is used for solving the deviation between the second predicted value and the actual running state, namely a residual matrix:
in the method, in the process of the invention,for residual matrix +.>The temperature characteristic matrix after actual noise reduction at the moment k;
and a comparison unit: the method comprises the steps of comparing deviation with a preset threshold value, and if the difference value is smaller than the preset value, primarily confirming safety; if the difference is larger than the preset value, the prediction is continued.
8. The chip production monitoring and management apparatus according to claim 5, wherein the data information in the acquisition module includes any one or any combination of status data information, temperature data information, power data information, usage data information, and process data information of the chip.
9. A chip production monitoring and management device, characterized by comprising:
a memory for storing a computer program;
a processor for implementing the chip production monitoring management method according to any one of claims 1 to 4 when executing the computer program.
10. A readable storage medium, characterized by: the readable storage medium has stored thereon a computer program which, when executed by a processor, implements the chip production monitoring management method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311705619.8A CN117408574A (en) | 2023-12-13 | 2023-12-13 | Chip production monitoring management method, device, equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311705619.8A CN117408574A (en) | 2023-12-13 | 2023-12-13 | Chip production monitoring management method, device, equipment and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117408574A true CN117408574A (en) | 2024-01-16 |
Family
ID=89489317
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311705619.8A Pending CN117408574A (en) | 2023-12-13 | 2023-12-13 | Chip production monitoring management method, device, equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117408574A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111985571A (en) * | 2020-08-26 | 2020-11-24 | 国网湖南省电力有限公司 | Low-voltage intelligent monitoring terminal fault prediction method, device, medium and equipment based on improved random forest algorithm |
| CN113095511A (en) * | 2021-04-16 | 2021-07-09 | 广东电网有限责任公司 | Method and device for judging in-place operation of automatic master station |
| CN113433850A (en) * | 2021-06-04 | 2021-09-24 | 电子科技大学 | Method for repairing abnormal logic of FPGA (field programmable Gate array) |
| CN114328112A (en) * | 2021-12-29 | 2022-04-12 | 苏州浪潮智能科技有限公司 | Monitoring method, device, equipment and medium for heterogeneous chip |
| CN114580602A (en) * | 2022-03-18 | 2022-06-03 | 苏州浪潮智能科技有限公司 | Model training method, model training device, product life cycle prediction method, product life cycle prediction device, product life cycle prediction equipment and product life cycle prediction medium |
-
2023
- 2023-12-13 CN CN202311705619.8A patent/CN117408574A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111985571A (en) * | 2020-08-26 | 2020-11-24 | 国网湖南省电力有限公司 | Low-voltage intelligent monitoring terminal fault prediction method, device, medium and equipment based on improved random forest algorithm |
| CN113095511A (en) * | 2021-04-16 | 2021-07-09 | 广东电网有限责任公司 | Method and device for judging in-place operation of automatic master station |
| CN113433850A (en) * | 2021-06-04 | 2021-09-24 | 电子科技大学 | Method for repairing abnormal logic of FPGA (field programmable Gate array) |
| CN114328112A (en) * | 2021-12-29 | 2022-04-12 | 苏州浪潮智能科技有限公司 | Monitoring method, device, equipment and medium for heterogeneous chip |
| CN114580602A (en) * | 2022-03-18 | 2022-06-03 | 苏州浪潮智能科技有限公司 | Model training method, model training device, product life cycle prediction method, product life cycle prediction device, product life cycle prediction equipment and product life cycle prediction medium |
Non-Patent Citations (1)
| Title |
|---|
| 吴忠 等: "集成视频监控功能的智能设备可视化管理方法", 《研究与设计》, vol. 39, no. 9, 30 September 2023 (2023-09-30), pages 150 - 154 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109032829B (en) | Data anomaly detection method and device, computer equipment and storage medium | |
| CN111143102B (en) | Abnormal data detection method and device, storage medium and electronic equipment | |
| CN111475804A (en) | Alarm prediction method and system | |
| CN114386537B (en) | Lithium battery fault diagnosis method and device based on Catboost and electronic equipment | |
| CN115814686B (en) | State monitoring method and system for laser gas mixing production system | |
| CN115514619B (en) | Alarm convergence method and system | |
| CN113537642A (en) | Product quality prediction method, device, electronic equipment and storage medium | |
| CN113986693A (en) | Alarm response level determination method and device, electronic equipment and storage medium | |
| CN116823227A (en) | Intelligent equipment management system and method based on Internet of things | |
| CN116450482A (en) | User abnormality monitoring method and device, electronic equipment and storage medium | |
| CN113298638A (en) | Root cause positioning method, electronic device and storage medium | |
| CN117220920A (en) | Firewall policy management method based on artificial intelligence | |
| CN114707834A (en) | Alarm reminding method and device and storage medium | |
| CN115115190A (en) | Quality monitoring method based on working condition, related device and program medium product | |
| CN113313304A (en) | Power grid accident abnormity analysis method and system based on big data decision tree | |
| CN116401137B (en) | Core particle health state prediction method and device, electronic equipment and storage medium | |
| CN110490132B (en) | Data processing method and device | |
| CN117408574A (en) | Chip production monitoring management method, device, equipment and readable storage medium | |
| US20230156043A1 (en) | System and method of supporting decision-making for security management | |
| CN115589339A (en) | Network attack type identification method, device, equipment and storage medium | |
| WO2021178649A1 (en) | An algorithmic learning engine for dynamically generating predictive analytics from high volume, high velocity streaming data | |
| CN113452700A (en) | Method, device, equipment and storage medium for processing safety information | |
| CN112817827A (en) | Operation and maintenance method, device, server, equipment, system and medium | |
| CN115422555B (en) | Back door program detection method and device, electronic equipment and storage medium | |
| CN114936113B (en) | Task avalanche recovery method and device, electronic 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 |