CN117078105A - Production quality monitoring method and system based on artificial intelligence - Google Patents
Production quality monitoring method and system based on artificial intelligence Download PDFInfo
- Publication number
- CN117078105A CN117078105A CN202311106420.3A CN202311106420A CN117078105A CN 117078105 A CN117078105 A CN 117078105A CN 202311106420 A CN202311106420 A CN 202311106420A CN 117078105 A CN117078105 A CN 117078105A
- Authority
- CN
- China
- Prior art keywords
- production
- data
- target
- quality
- product quality
- 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
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 927
- 238000012544 monitoring process Methods 0.000 title claims abstract description 109
- 238000000034 method Methods 0.000 title claims abstract description 69
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 54
- 238000011156 evaluation Methods 0.000 claims abstract description 79
- 238000004458 analytical method Methods 0.000 claims abstract description 56
- 238000007726 management method Methods 0.000 claims abstract description 52
- 238000012216 screening Methods 0.000 claims abstract description 32
- 238000003326 Quality management system Methods 0.000 claims description 44
- 230000008859 change Effects 0.000 claims description 37
- 239000013598 vector Substances 0.000 claims description 22
- 230000006870 function Effects 0.000 claims description 16
- 238000003860 storage Methods 0.000 claims description 16
- 238000012806 monitoring device Methods 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 11
- 239000000047 product Substances 0.000 description 172
- 230000008569 process Effects 0.000 description 19
- 230000007613 environmental effect Effects 0.000 description 10
- 238000004422 calculation algorithm Methods 0.000 description 8
- 238000012545 processing Methods 0.000 description 8
- 235000015895 biscuits Nutrition 0.000 description 7
- 235000013305 food Nutrition 0.000 description 6
- 238000005457 optimization Methods 0.000 description 6
- 238000000605 extraction Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 239000000284 extract Substances 0.000 description 4
- 238000010924 continuous production Methods 0.000 description 3
- 230000002068 genetic effect Effects 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000446 fuel Substances 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000003756 stirring Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 238000001303 quality assessment method Methods 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 238000013442 quality metrics Methods 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000036962 time dependent Effects 0.000 description 1
- 238000012549 training Methods 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
- 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
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention relates to the field of artificial intelligence, and discloses a production quality monitoring method and system based on artificial intelligence, which are used for improving the accuracy of production quality monitoring so as to improve the production efficiency and quality. The method comprises the following steps: carrying out relation analysis on the production process data and the product quality data to obtain a production process influence weight, and carrying out relation analysis on the production environment data and the product quality data to obtain a production environment influence weight; combining to obtain a plurality of initial production parameter combinations; inputting a plurality of initial production parameters into a preset production quality monitoring model for production quality monitoring to obtain a target production evaluation index; screening according to the target production evaluation index to obtain a target production parameter combination, and updating the relation state among the production process data, the production environment data and the product quality data; and carrying out strategy updating based on the relation state and the target production parameter combination, and generating a target production quality management execution strategy.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to a production quality monitoring method and system based on artificial intelligence.
Background
In the field of manufacturing, production quality has always been one of the important issues of enterprise concern. Conventional production quality management methods generally rely on manual experience and rules, and are difficult to adapt to complex and varied production environments and requirements. With the rapid development of artificial intelligence technology, the production quality monitoring method based on artificial intelligence becomes an important way for improving production efficiency and optimizing product quality.
Conventional production quality monitoring methods often rely on predefined rules and thresholds, lacking the ability to adapt to complex production environments. Meanwhile, the manual intervention is high in cost and low in efficiency. Therefore, an innovative method is needed to realize intelligent production quality monitoring and improve production efficiency and quality.
Disclosure of Invention
The invention provides a production quality monitoring method and system based on artificial intelligence, which are used for improving the accuracy of production quality monitoring and further improving the production efficiency and quality.
The first aspect of the invention provides an artificial intelligence-based production quality monitoring method, which comprises the following steps:
acquiring target production associated data through a preset intelligent production quality management system, and dividing a data set of the target production associated data to obtain production process data, product quality data and production environment data;
Performing relation analysis on the production process data and the product quality data to obtain a production process influence weight, and performing relation analysis on the production environment data and the product quality data to obtain a production environment influence weight;
combining the production environment data, the production process data and the product quality data to obtain a plurality of initial production parameter combinations;
inputting the initial production parameter combinations into a preset production quality monitoring model for production quality monitoring according to the production process influence weights and the production environment influence weights, and obtaining target production evaluation indexes corresponding to each initial production parameter combination;
screening the plurality of initial production parameter combinations according to the target production evaluation indexes to obtain target production parameter combinations, and updating the relation states among the production process data, the production environment data and the product quality data according to the target production parameter combinations;
and based on the relation state and the target production parameter combination, carrying out strategy updating on an initial production quality management execution strategy of the intelligent production quality management system, and generating a target production quality management execution strategy.
With reference to the first aspect, in a first implementation manner of the first aspect of the present invention, the obtaining, by a preset intelligent production quality management system, target production associated data, and dividing a data set of the target production associated data, to obtain production process data, product quality data, and production environment data includes:
acquiring target production associated data through a preset intelligent production quality management system, and acquiring a production process label, a product quality label and a production environment label;
determining corresponding first production process data points, first product quality data points and first production environment data points according to the production process tags, the product quality tags and the production environment tags;
performing data point distance analysis on a plurality of original data points in the target production associated data based on the first production process data points to obtain a first data point distance between each original data point and the first production process data point; performing data point distance analysis on a plurality of original data points in the target production associated data based on the first product quality data points to obtain second data point distances between each original data point and the first product quality data points; based on the first production environment data point, carrying out data point distance analysis on a plurality of original data points in the target production associated data to obtain a third data point distance between each original data point and the first production environment data point;
Performing data point screening on the plurality of original data points according to the first data point distance to obtain a plurality of first target data points, and generating corresponding production process data according to the plurality of first target data points; performing data point screening on the plurality of original data points according to the second data point distance to obtain a plurality of second target data points, and generating corresponding product quality data according to the plurality of second target data points; and carrying out data point screening on the plurality of original data points according to the third data point distance to obtain a plurality of third target data points, and generating corresponding production environment data according to the plurality of third target data points.
With reference to the first aspect, in a second implementation manner of the first aspect of the present invention, the performing a relationship analysis on the production process data and the product quality data to obtain a production process influence weight, and performing a relationship analysis on the production environment data and the product quality data to obtain a production environment influence weight, where the method includes:
performing curve fitting on the production process data to obtain a plurality of production process parameter change curves, performing curve fitting on the product quality data to obtain a product quality change curve, and performing curve fitting on the production environment data to obtain a plurality of production environment parameter change curves;
Extracting features of the production process parameter change curves to obtain a plurality of production process parameter features, extracting features of the product quality change curves to obtain a plurality of product quality change features, and extracting features of the production environment parameter change curves to obtain a plurality of production environment parameter change features;
calculating first correlation coefficients between the plurality of production process parameter characteristics and the plurality of product quality variation characteristics, and generating production process influence weights according to the first correlation coefficients;
and calculating a second phase relation number between the plurality of production environment parameter change characteristics and the plurality of product quality change characteristics, and generating a production environment influence weight according to the second phase relation number.
With reference to the first aspect, in a third implementation manner of the first aspect of the present invention, the combining the production environment data, the production process data, and the product quality data to obtain a plurality of initial production parameter combinations includes:
inquiring a corresponding initial production quality management execution strategy through the intelligent production quality management system, and matching a corresponding combination evaluation function and a target combination type according to the initial production quality management execution strategy;
Acquiring a first numerical value set corresponding to the production environment data, a second numerical value set corresponding to the production process data and a third numerical value set corresponding to the product quality data;
generating a plurality of random production parameter combinations for the first value set, the second value set and the third value set according to the target combination type;
calculating a combined evaluation index of the plurality of random production parameter combinations through the combined evaluation function to obtain a combined evaluation index of each random production parameter combination;
and screening and iterating the random production parameter combinations according to the combination evaluation index of each random production parameter combination to obtain a plurality of initial production parameter combinations, wherein each initial production parameter combination comprises a plurality of production process values, a plurality of production environment values and a production quality value.
With reference to the first aspect, in a fourth implementation manner of the first aspect of the present invention, inputting the plurality of initial production parameter combinations into a preset production quality monitoring model according to the production process influence weight and the production environment influence weight to perform production quality monitoring, to obtain a target production evaluation index corresponding to each initial production parameter combination, where the method includes:
Setting weights of a plurality of production process values in each initial production parameter combination according to the production process influence weights, and simultaneously setting weights of a plurality of production environment values in each initial production parameter combination according to the production environment influence weights to obtain a plurality of weighted production parameter combinations;
respectively inputting the plurality of weighted production parameter combinations into a preset production quality monitoring model, wherein the production quality monitoring model comprises an encoding network and a decoding network;
extracting coding characteristics of each weighted production parameter combination through the coding network to obtain a coding hidden vector of each weighted production parameter combination;
and inputting the coding hidden vector of each weighted production parameter combination into the decoding network to carry out the evaluation index prediction of the production quality monitoring, so as to obtain the corresponding target production evaluation index.
With reference to the first aspect, in a fifth implementation manner of the first aspect of the present invention, the screening the plurality of initial production parameter combinations according to the target production evaluation index to obtain a target production parameter combination, and updating a relationship state among the production process data, the production environment data, and the product quality data according to the target production parameter combination includes:
Screening out the target production parameter combination with the maximum target production evaluation index from the plurality of initial production parameter combinations;
notifying the intelligent production quality management system to provide a target production quality monitoring service according to the target production parameter combination;
when the target production quality monitoring service starts, a relation state among the production process data, the production environment data and the product quality data is updated.
With reference to the first aspect, in a sixth implementation manner of the first aspect of the present invention, the performing, based on the combination of the relationship state and the target production parameter, policy update on an initial production quality management execution policy of the intelligent production quality management system, to generate a target production quality management execution policy, includes:
based on the relation state and the target production parameter combination, performing rewarding feedback calculation on an initial production quality management execution strategy of the intelligent production quality management system to obtain rewarding feedback data of the initial production quality management execution strategy;
and carrying out strategy updating on the initial production quality management execution strategy of the intelligent production quality management system based on the reward feedback data, and generating a target production quality management execution strategy.
The second aspect of the present invention provides an artificial intelligence based production quality monitoring system, comprising:
the acquisition module is used for acquiring target production associated data through a preset intelligent production quality management system, and dividing a data set of the target production associated data to obtain production process data, product quality data and production environment data;
the analysis module is used for carrying out relation analysis on the production process data and the product quality data to obtain production process influence weights, and carrying out relation analysis on the production environment data and the product quality data to obtain production environment influence weights;
the combination module is used for combining the production environment data, the production process data and the product quality data to obtain a plurality of initial production parameter combinations;
the monitoring module is used for inputting the plurality of initial production parameter combinations into a preset production quality monitoring model for production quality monitoring according to the production process influence weight and the production environment influence weight, and obtaining a target production evaluation index corresponding to each initial production parameter combination;
The screening module is used for screening the plurality of initial production parameter combinations according to the target production evaluation indexes to obtain target production parameter combinations, and updating the relation states among the production process data, the production environment data and the product quality data according to the target production parameter combinations;
and the updating module is used for carrying out strategy updating on the initial production quality management execution strategy of the intelligent production quality management system based on the relation state and the target production parameter combination, and generating a target production quality management execution strategy.
A third aspect of the present invention provides an artificial intelligence based production quality monitoring device comprising: a memory and at least one processor, the memory having instructions stored therein; the at least one processor invokes the instructions in the memory to cause the artificial intelligence based production quality monitoring device to perform the artificial intelligence based production quality monitoring method described above.
A fourth aspect of the present invention provides a computer readable storage medium having instructions stored therein which, when run on a computer, cause the computer to perform the above-described artificial intelligence based production quality monitoring method.
In the technical scheme provided by the invention, the relation analysis is carried out on the production process data and the product quality data to obtain the production process influence weight, and the relation analysis is carried out on the production environment data and the product quality data to obtain the production environment influence weight; combining to obtain a plurality of initial production parameter combinations; inputting a plurality of initial production parameters into a preset production quality monitoring model for production quality monitoring to obtain a target production evaluation index; screening according to the target production evaluation index to obtain a target production parameter combination, and updating the relation state among the production process data, the production environment data and the product quality data; the invention can analyze the production process data, the product quality data and the production environment data in real time and quickly identify potential quality problems. This enables the enterprise to respond to the problem more quickly, reducing reject rate and rejection rate. By analyzing the historical data and the real-time data, the artificial intelligence can predict potential quality problems and take measures in advance to prevent the problems from occurring. Production parameters are automatically optimized to maximize production efficiency and product quality. The production process, the product quality and the environmental data are combined for analysis, so that the influence relation among different factors can be more comprehensively known. Therefore, continuous production quality improvement is realized, and therefore, the accuracy of production quality monitoring is improved, and the production efficiency and quality are further improved.
Drawings
FIG. 1 is a schematic diagram of one embodiment of an artificial intelligence based production quality monitoring method in an embodiment of the present invention;
FIG. 2 is a flow chart of relationship analysis in an embodiment of the invention;
FIG. 3 is a flow chart of obtaining a plurality of initial combinations of production parameters according to an embodiment of the present invention;
FIG. 4 is a flow chart of production quality monitoring in an embodiment of the invention;
FIG. 5 is a schematic diagram of one embodiment of an artificial intelligence based production quality monitoring system in accordance with an embodiment of the present invention;
FIG. 6 is a schematic diagram of one embodiment of an artificial intelligence based production quality monitoring device in accordance with an embodiment of the present invention.
Detailed Description
The embodiment of the invention provides a production quality monitoring method and system based on artificial intelligence, which are used for improving the accuracy of production quality monitoring and further improving the production efficiency and quality. The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.
For ease of understanding, a specific flow of an embodiment of the present invention is described below with reference to fig. 1, where an embodiment of an artificial intelligence-based production quality monitoring method according to an embodiment of the present invention includes:
s101, acquiring target production associated data through a preset intelligent production quality management system, and dividing a data set of the target production associated data to obtain production process data, product quality data and production environment data;
it will be appreciated that the execution subject of the present invention may be an artificial intelligence based production quality monitoring system, or may be a terminal or a server, and is not limited herein. The embodiment of the invention is described by taking a server as an execution main body as an example.
Specifically, the server first obtains the target production associated data through a preset intelligent production quality management system. These data include production process data, product quality data, and production environment data. Meanwhile, a production process label, a product quality label and a production environment label are also acquired for subsequent data processing and analysis. Using the tags, corresponding first production process data points, first product quality data points, and first production environment data points are determined. These data points may be considered reference points for subsequent data point distance analysis. For the production process data, based on the first production process data point, a data point distance analysis is performed, a plurality of raw data points in the target production association data are compared with the first production process data point, and a data point distance between the raw data points is calculated. These data point distances reflect the degree of similarity or difference between the original data point and the first process data point. Similarly, for product quality data, a data point distance analysis is performed based on the first product quality data point. This will calculate for each raw data point its data point distance from the first product quality data point. For production environment data, based on the first production environment data point, data point distance analysis is performed to obtain a data point distance between each original data point and the first production environment data point. Through analysis of the data point distances, a plurality of first target data points can be screened, which are within a certain range from the first production process data point, the first product quality data point and the first production environment data point. Based on these screened data points, corresponding production process data, product quality data, and production environment data may be generated. By a similar method, a plurality of second and third target data points may be screened out, respectively, and corresponding product quality data and production environment data may then be generated from these data points. For example, suppose a certain manufacturing company produces an electronic product. The server uses this method to monitor quality during production. The server obtains production process data, product quality data and production environment data from the intelligent production quality management system, and also obtains applicable labels. For the production process data, the server selects labels such as "temperature", "humidity" and "working time". For product quality data, the server selects tags such as "power", "stability" and "efficiency". For production environment data, the server selects labels such as "factory temperature", "humidity", and "power stability". The server uses the tags to determine a first production process data point, a first product quality data point, and a first production environment data point. The server performs a data point distance analysis for each tag. For example, for a "temperature" tag, the server calculates the distance of each raw data point from the first process data point, and screens out the data point that is closer to the first process data point as the first target data point. Likewise, the server performs data point distance analysis for the product quality and the production environment label, screening out second and third target data points. The server generates updated production process data, product quality data, and production environment data based on the screened target data points. These data will be used for execution policy updates of the intelligent production quality management system to optimize the production process and improve the product quality.
S102, carrying out relation analysis on production process data and product quality data to obtain production process influence weights, and carrying out relation analysis on production environment data and product quality data to obtain production environment influence weights;
specifically, the server performs curve fitting on the production process data to obtain a plurality of production process parameter variation curves. And drawing the production process data into a curve on a time axis, and then utilizing a curve fitting technology to find a curve model which is most suitable for the data change trend. And (5) performing curve fitting on the product quality data to obtain a product quality change curve. This process is similar to the processing of production process data, but is directed to product quality indicators. And on the other hand, curve fitting is also carried out on the production environment data to obtain a plurality of production environment parameter change curves. These curves will show the change in environmental parameters over time, such as temperature, humidity, etc. And extracting the characteristics of the fitting curves to obtain important characteristics of each data. For production process data, the extracted features include mean, variance, trend, etc. For product quality data, feature extraction includes maxima, minima, volatility, etc. Also, feature extraction of production environment data may include periodicity, stability, and the like. At this stage, the result of feature extraction will be that the original curve data is represented as a feature vector, thereby reducing the data dimension. A first correlation coefficient between the production process parameter characteristic and the product quality variation characteristic is calculated. This correlation coefficient can measure the degree of linear relationship between the production process parameters and the product quality variation. If the correlation between the two is high, the correlation coefficient is close to 1, which indicates that the production process has a great influence on the product quality. By calculating the correlation coefficients, a process impact weight may be generated for each process parameter feature. Similarly, a second correlation coefficient between the production environment parameter characteristic and the product quality variation characteristic is calculated. This correlation coefficient measures the impact of the production environment on the quality variation of the product. By calculating the correlation coefficient, a production environment impact weight may be generated for each production environment parameter feature. For example, consider an automotive manufacturing plant, which the server uses to monitor the quality of automotive production. The server collects production process data such as production speed, material usage, etc., product quality data such as body stability, engine efficiency, etc., and production environment data such as plant temperature, humidity. For the production process data, the server generates a time-varying production speed curve, for the product quality data, a time-varying body stability curve, and for the production environment data, a time-varying shop temperature curve. The server extracts features from these curves, such as the mean of the production speed curve, the volatility of the body stability curve, and the periodicity of the shop temperature curve. And by calculating the correlation coefficient between the production speed characteristic and the vehicle body stability characteristic, the production process influence weight is generated, which shows that the production speed has a larger influence on the vehicle body stability. Also, the server calculates a correlation coefficient between the shop temperature characteristic and the vehicle body stability characteristic, and generates a production environment influence weight, which indicates that the shop temperature has a certain influence on the vehicle body stability.
S103, combining production environment data, production process data and product quality data to obtain a plurality of initial production parameter combinations;
it should be noted that, the intelligent production quality management system is used to query the initial production quality management execution policies and match the corresponding combination evaluation function and the target combination type according to the policies. These strategies will serve as guidelines in subsequent steps to generate initial production parameter combinations with the desired characteristics. A first set of values is obtained from the production environment data, a second set of values is obtained from the production process data, and a third set of values is obtained from the product quality data. These value sets contain data on different aspects, such as key indicators of production environment, production process and product quality. Based on the target combination type, a plurality of random production parameter combinations are generated for the sets of values. These random combinations will cover a range of parameters for optimization in subsequent evaluations. The evaluation index calculation is performed on these random production parameter combinations by combining the evaluation functions. The combined evaluation function will calculate an evaluation index for each combination based on the requirements in the initial production quality management execution strategy, in combination with the data of the production environment, the production process and the product quality. Screening and iterating according to the evaluation index of each random production parameter combination to obtain a plurality of initial production parameter combinations. These initial combinations will be used as references in subsequent quality monitoring processes to optimize the production process and improve product quality. For example, assume a food processing company uses this method to monitor the quality of biscuit production. The server firstly inquires the intelligent production quality management system to acquire an initial production quality management execution strategy, when the strategy requires that biscuits are produced, the temperature should be kept within a specific range, the humidity should be moderate, and the taste and appearance of the biscuits should meet the standard. The server acquires production environment data such as temperature, humidity and the like from a production workshop, records production process data such as production time, process steps and the like of each batch, and collects product quality data such as biscuit mouthfeel, appearance and the like. Based on the target combination type, the server generates a plurality of random combinations of production parameters. For example, random combinations include different combinations of temperature and humidity, and different arrangements of manufacturing steps. And calculating the evaluation index of each random production parameter combination by the server through the set combination evaluation function. This relates to the quality scores of mouthfeel and appearance, as well as the suitability of temperature and humidity during production. And according to the evaluation index, the server screens and iterates the random production parameter combination. For example, the server excludes combinations with lower evaluation metrics, and only retains combinations that perform well. The server obtains a plurality of initial production parameter combinations, each combination including a suitable production environment parameter, a production process parameter, and a product quality parameter. These initial combinations will be used in the quality monitoring process to optimize the biscuit production process, ensuring that the final product meets quality standards.
S104, inputting a plurality of initial production parameter combinations into a preset production quality monitoring model for production quality monitoring according to the production process influence weight and the production environment influence weight, and obtaining a target production evaluation index corresponding to each initial production parameter combination;
specifically, the weight setting is performed on the production process value and the production environment value in each initial production parameter combination according to the production process influence weight and the production environment influence weight. These weights can be derived from previous relational analysis to more accurately reflect the extent to which the production process and environment have affected the quality of the product. After the weight is set, a plurality of weighted production parameter combinations are obtained, and each combination considers weight factors in the aspects of the production process and the production environment. These weighted production parameter combinations are respectively input into a preset production quality monitoring model. This model includes an encoding network and a decoding network for processing the weighted production parameter combinations and predicting an evaluation index of the production quality. And extracting coding characteristics of each weighted production parameter combination through a coding network. The encoding network may convert the input weighted production parameter combinations into an encoded hidden vector that contains the combined significant feature information. The encoded concealment vector for each weighted production parameter combination is input into a decoding network. The decoding network will use these encoded hidden vectors to make an evaluation index prediction for production quality monitoring. The output of the decoding network will be the corresponding target production assessment index for quantifying and assessing the quality of production of the combination. Consider, for example, a pharmaceutical company. According to the previous analysis, the production process influence weight and the production environment influence weight are obtained, and the factors such as temperature, humidity and the like are considered. For each initial production parameter combination, the production process values and production environment values are weighted according to these weights. For example, if the impact weight of temperature is high, then in a particular combination, the value of temperature will be more pronounced. These weighted production parameter combinations are input into a preset production quality monitoring model, wherein the model comprises an encoding network and a decoding network. The coding network will code each weighted combination to obtain a coded hidden vector. These encoded concealment vectors will be input into a decoding network, which will predict the evaluation index of the production quality, such as product purity, efficacy, etc., from the concealment vectors. Through these indices, the pharmaceutical company can know the production quality level of each weighted production parameter combination, and further adjust the production parameters to improve the product quality.
S105, screening a plurality of initial production parameter combinations according to the target production evaluation indexes to obtain target production parameter combinations, and updating the relation states among the production process data, the production environment data and the product quality data according to the target production parameter combinations;
specifically, the server selects a target production parameter combination that maximizes a target production evaluation index from among a plurality of initial production parameter combinations. Among all combinations, those that perform best in terms of production quality are found. For example, assuming a chemical plant uses this method to monitor the product purity of a chemical reaction, the server will select a set of parameters from a variety of initial reaction conditions that will result in the highest product purity. After the target production parameter combination is selected, the intelligent production quality management system will be notified to provide the target production quality monitoring service. This triggers the system to take appropriate control and monitoring strategies to ensure that parameters and environment in the actual production process remain consistent with the target parameters to optimize product quality. When the target production quality monitoring service begins, the system will update the relationship state between the production process data, the production environment data, and the product quality data based on the target production parameter combinations. This process involves adjusting the previously established relationship models and weights to better reflect the impact of the target production parameters. For example, consider an electronics manufacturing company, which uses this method for monitoring the quality of electronic products on a production line. When the target production parameter combination is determined, the system updates the relation model and the weight between the technological parameter, the environmental factor and the electronic product performance according to the data in the actual production. If the influence of a certain parameter is found to be beyond expectation, the system adjusts the weight accordingly to reflect the actual situation more accurately. By continuously updating the relationship state, the method can help the production process adapt to changes in real time, thereby achieving higher quality standards and better production results.
S106, based on the relation state and the target production parameter combination, carrying out strategy updating on the initial production quality management execution strategy of the intelligent production quality management system, and generating a target production quality management execution strategy.
Specifically, combining the relationship state with the target production parameter involves integrating the relationship weight, the parameter impact and the target parameter obtained from the previous analysis. This can be accomplished by building a comprehensive model that can correlate production process data, environmental data, and product quality data so that the system can understand the complex relationships between them. This model may be a machine learning based model, such as a neural network or decision tree, to capture interactions between different factors. For example, consider a food processing plant where a server uses this method to monitor the quality of biscuit production. The server can correlate factors such as temperature, stirring time and the like in the production process with quality attributes such as taste, color and the like of biscuits, so that a relational model is constructed. The bonus feedback data is calculated using this integrated model. By monitoring the actual production data, the actual production parameters are compared with the target parameters, and the quality performance of each initial production strategy can be obtained. These indicators may be accuracy, consistency, or other important performance indicators of product quality. For example, at an electronic device manufacturing facility, the system may evaluate the impact of different strategies on product performance, such as battery life, response speed, etc., based on actual production data. Based on the reward feedback data, an optimization algorithm may be employed to update the initial production quality management execution policy to generate a target production quality management execution policy. The selection of the optimization algorithm depends on the actual situation, and may be a genetic algorithm, an ant colony algorithm, and the like. For example, a manufacturing facility uses genetic algorithms to find those that are excellent in terms of quality of production by randomly generating a set of policies and then evaluating those policies using rewards feedback data. The genetic algorithm then gradually optimizes these strategies by crossover, mutation, etc. to find a more optimal target quality management implementation strategy.
In the embodiment of the invention, the relation analysis is carried out on the production process data and the product quality data to obtain the production process influence weight, and the relation analysis is carried out on the production environment data and the product quality data to obtain the production environment influence weight; combining to obtain a plurality of initial production parameter combinations; inputting a plurality of initial production parameters into a preset production quality monitoring model for production quality monitoring to obtain a target production evaluation index; screening according to the target production evaluation index to obtain a target production parameter combination, and updating the relation state among the production process data, the production environment data and the product quality data; the invention can analyze the production process data, the product quality data and the production environment data in real time and quickly identify potential quality problems. This enables the enterprise to respond to the problem more quickly, reducing reject rate and rejection rate. By analyzing the historical data and the real-time data, the artificial intelligence can predict potential quality problems and take measures in advance to prevent the problems from occurring. Production parameters are automatically optimized to maximize production efficiency and product quality. The production process, the product quality and the environmental data are combined for analysis, so that the influence relation among different factors can be more comprehensively known. Therefore, continuous production quality improvement is realized, and therefore, the accuracy of production quality monitoring is improved, and the production efficiency and quality are further improved.
In a specific embodiment, the process of executing step S101 may specifically include the following steps:
(1) Acquiring target production associated data through a preset intelligent production quality management system, and acquiring a production process label, a product quality label and a production environment label;
(2) Determining corresponding first production process data points, first product quality data points and first production environment data points according to the production process tags, the product quality tags and the production environment tags;
(3) Based on the first production process data point, carrying out data point distance analysis on a plurality of original data points in the target production associated data to obtain a first data point distance between each original data point and the first production process data point; based on the first product quality data point, carrying out data point distance analysis on a plurality of original data points in the target production associated data to obtain a second data point distance between each original data point and the first product quality data point; based on the first production environment data point, carrying out data point distance analysis on a plurality of original data points in the target production associated data to obtain a third data point distance between each original data point and the first production environment data point;
(4) According to the first data point distance, carrying out data point screening on a plurality of original data points to obtain a plurality of first target data points, and generating corresponding production process data according to the plurality of first target data points; according to the second data point distance, carrying out data point screening on a plurality of original data points to obtain a plurality of second target data points, and generating corresponding product quality data according to the plurality of second target data points; and carrying out data point screening on the plurality of original data points according to the third data point distance to obtain a plurality of third target data points, and generating corresponding production environment data according to the plurality of third target data points.
Specifically, the server acquires a large amount of data from the production process through a preset intelligent production quality management system, wherein the data comprise production process parameters, product quality indexes, production environment conditions and the like. At the same time, the server attaches labels to each data point, including production process labels, product quality labels, and production environment labels, for further analysis and classification. The server selects a particular data point as the first data point based on the production process label, the product quality label, and the production environment label. For example, assume that the production process label of the server is an assembly time, the product quality label is a product durability, the production environment label is a production temperature, the server takes the assembly time as a first production process data point, the product durability as a first product quality data point, and the production temperature as a first production environment data point. The server analyzes the distance between each raw data point and the first data point. This distance may be calculated using various methods, such as euclidean distance, manhattan distance, and the like. The server obtains the distance between each original data point and the first data point by calculating the distance, and the first data point distance, the second data point distance and the third data point distance are respectively obtained. Based on the result of the data point distance analysis, the server screens the original data points and selects the data point with the smallest distance with the first data point as the first target data point. Likewise, the data point that is the smallest distance from the first data point in terms of product quality and production environment is selected as the second and third target data points, respectively, based on the second and third data point distances. These target data points will be used to generate corresponding production process data, product quality data, and production environment data. For example, consider an automobile manufacturing company, which uses this method by a server to monitor the quality of automobiles on a production line. Through the intelligent production quality management system, the server obtains data regarding production parameters, quality flags, and production environmental conditions for each car. The speed is used as a production process label, the safety is used as a product quality label, and the air temperature is used as a production environment label. And (3) selecting the vehicle with the smallest distance from the target data point as the first target data point by calculating the distance between each vehicle and the first data point (vehicle speed, safety and air temperature). Likewise, depending on the distance of the other tags, the server finds a vehicle that is similar in different respects to the target data point, thereby generating relevant production process data, product quality data, and production environment data.
In a specific embodiment, as shown in fig. 2, the process of executing step S102 may specifically include the following steps:
s201, performing curve fitting on production process data to obtain a plurality of production process parameter change curves, performing curve fitting on product quality data to obtain a product quality change curve, and performing curve fitting on production environment data to obtain a plurality of production environment parameter change curves;
s202, extracting features of a plurality of production process parameter change curves to obtain a plurality of production process parameter features, extracting features of a product quality change curve to obtain a plurality of product quality change features, and extracting features of a plurality of production environment parameter change curves to obtain a plurality of production environment parameter change features;
s203, calculating first correlation coefficients between the plurality of production process parameter characteristics and the plurality of product quality change characteristics, and generating production process influence weights according to the first correlation coefficients;
s204, calculating a second phase relation number between the plurality of production environment parameter change characteristics and the plurality of product quality change characteristics, and generating production environment influence weights according to the second phase relation number.
Specifically, the server performs curve fitting on the production process data, the product quality data and the production environment data to obtain a corresponding parameter change curve. For each data type, the server fits the raw data points to a smooth curve using a suitable curve fitting algorithm, such as polynomial fitting, exponential fitting, or neural network fitting, among others. For example, for production process data, the server fits a time-dependent variation curve of different production process parameters; for the product quality data, the server obtains a change curve of the product quality along with the production time; for production environment data, the server obtains a profile of production environment parameters over time. And on the basis of fitting the curve, the server performs feature extraction. This means that information representing its characteristics is extracted from the fitted curve. Features may include mean, variance, slope, peak, etc. of the curve. For the production process parameter variation curve, the product quality variation curve and the production environment parameter variation curve, the server respectively extracts a plurality of characteristics for subsequent analysis. The server calculates correlation coefficients between the features to determine the degree of correlation between them. For the production process parameter characteristics and the product quality variation characteristics, the server calculates a first correlation coefficient between them. This can be achieved by common correlation coefficient calculation methods, such as pearson correlation coefficients. Similarly, the server calculates a second correlation coefficient between the production environment parameter variation characteristic and the product quality variation characteristic. Based on the correlation coefficient, the server generates a production process impact weight and a production environment impact weight. The higher the correlation coefficient, the higher the degree of correlation between the two features. Thus, the server uses the correlation coefficients to quantify the importance between the production process parameter characteristics and the product quality variation characteristics, and between the production environment parameter variation characteristics and the product quality variation characteristics. For example, consider a food processing plant where a server uses this method to monitor the quality of food production. The server collects data of production process parameters (such as heating temperature, stirring time and the like), product quality indexes (such as taste, color and the like) and production environment parameters (such as humidity, air temperature and the like) of different batches of foods. Through curve fitting, the server obtains the variation curves of the parameters of the food production process, the product quality and the production environment of different batches. For each type of data, the server extracts a plurality of features from the fitted curve, such as an average of the process parameter profiles, a peak of the product quality profile, etc. The server calculates a first correlation coefficient between the production process parameter characteristic and the product quality variation characteristic, and a second correlation coefficient between the production environment parameter variation characteristic and the product quality variation characteristic. Higher correlation coefficients indicate that certain production process parameters and environmental parameters have a greater impact on product quality variation. The server generates a production process influence weight and a production environment influence weight by the correlation coefficient. These weights will be used for subsequent production quality monitoring to help the server better understand the impact of each factor in the production process, optimize the production flow, and improve the product quality.
In a specific embodiment, as shown in fig. 3, the process of executing step S103 may specifically include the following steps:
s301, inquiring a corresponding initial production quality management execution strategy through an intelligent production quality management system, and matching a corresponding combination evaluation function and a target combination type according to the initial production quality management execution strategy;
s302, acquiring a first numerical value set corresponding to production environment data, acquiring a second numerical value set corresponding to production process data and acquiring a third numerical value set corresponding to product quality data;
s303, generating a plurality of random production parameter combinations for the first numerical value set, the second numerical value set and the third numerical value set according to the target combination type;
s304, performing combined evaluation index calculation on a plurality of random production parameter combinations through a combined evaluation function to obtain combined evaluation indexes of each random production parameter combination;
s305, screening and iterating a plurality of random production parameter combinations according to the combination evaluation index of each random production parameter combination to obtain a plurality of initial production parameter combinations, wherein each initial production parameter combination comprises a plurality of production process values, a plurality of production environment values and a production quality value.
Specifically, the server intelligent production quality management system will query the initial production quality management execution policies that have been set, including the setting of production process parameters, the control of environmental conditions, and the like. The system will match the corresponding combination evaluation function and the target combination type according to the initial production quality management execution strategy. These evaluation functions relate to stability of the production process, consistency of product quality, etc., while the target combination type may be maximizing product quality, minimizing production costs, etc. The system will acquire the required production environment data, production process data and product quality data. For each data type, the system will extract a corresponding set of values from the history or sensor. Based on the target combination type, the system will generate a plurality of random production parameter combinations for these sets of values. The random parameter combinations will include a plurality of production process values, a plurality of production environment values, and a production quality value. For each random production parameter combination, the system will calculate its combined evaluation index using a pre-matched combined evaluation function. These metrics are a set of values that measure the quality and efficiency of production. The evaluation function may be calculated by means of mathematical models, statistical analysis, etc. in order to comprehensively evaluate the comprehensive effect of each parameter combination. And according to the calculated combined evaluation index, the system screens and iterates a plurality of random production parameter combinations. In the screening process, random parameter combinations with optimal evaluation indexes can be selected according to the target combination types. The iterative process includes fine-tuning of parameters, generation of new rounds of random parameters, etc., to further optimize the parameter combinations. For example, assume that an electronic product manufacturing company employs an artificial intelligence-based production quality monitoring method in order to optimize a production process of an electronic product to improve product quality. The server queries the system for initial production quality management enforcement policies, such as setting process temperatures and times for electronic components. According to these strategies, the system matches the combined evaluation functions, including product performance consistency and production cost. The server collects historical production environment data, production process data, and product quality data. From these data, the server obtains a set of values related to production process parameters, environmental conditions, and product performance, etc. Based on the target combination type, the system generates a plurality of random production parameter combinations, each comprising a plurality of production process values, a plurality of production environment values, and a production quality value. The server then calculates a combined evaluation index for each random combination of production parameters, such as product performance consistency and production cost, using the pre-matched combined evaluation function. Based on these metrics, the server screens and iterates through a number of random parameter combinations. The server obtains a plurality of initial production parameter combinations, which optimize the production process to improve the quality and production efficiency of the electronic product.
In a specific embodiment, as shown in fig. 4, the process of executing step S104 may specifically include the following steps:
s401, carrying out weight setting on a plurality of production process values in each initial production parameter combination according to the production process influence weight, and simultaneously carrying out weight setting on a plurality of production environment values in each initial production parameter combination according to the production environment influence weight to obtain a plurality of weighted production parameter combinations;
s402, respectively inputting a plurality of weighted production parameter combinations into a preset production quality monitoring model, wherein the production quality monitoring model comprises an encoding network and a decoding network;
s403, extracting coding features of each weighted production parameter combination through a coding network to obtain a coding hidden vector of each weighted production parameter combination;
s404, inputting the coding hidden vector of each weighted production parameter combination into a decoding network to carry out evaluation index prediction of production quality monitoring, and obtaining a corresponding target production evaluation index.
Specifically, the server sets the weight of the production process value and the production environment value in each initial production parameter combination according to the production process influence weight and the production environment influence weight calculated previously. This means that the system will assign different importance to each parameter based on these weights. With these weight settings, the system obtains a plurality of weighted production parameter combinations, each of which will contain a weight-adjusted production process value and a production environment value. The system will prepare a pre-set production quality monitoring model for evaluating the effect of the weighted combination of production parameters on the quality of the product. The model may consist of an encoding network responsible for converting the input weighted production parameter combinations into encoded hidden vectors and a decoding network for converting the encoded hidden vectors back into the evaluation index for production quality monitoring. Each weighted production parameter combination is input into the encoding network, which will perform feature extraction for each combination. The encoding network encodes the production process values and the production environment values in the weighted production parameter combinations to generate corresponding encoded hidden vectors. These encoded hidden vectors contain an abstract representation of the original data, capturing the relationships and features between the data. The encoded concealment vector will be passed as input into the decoding network. The decoding network decodes the coded hidden vector, thereby predicting the evaluation index of production quality monitoring, such as product performance, stability and the like. These predictors can be obtained by regression, classification, etc. methods, depending on the service requirements. For example, suppose an automotive manufacturing company employs an artificial intelligence based production quality monitoring method to improve the quality of an automotive manufacturing process. According to the historical data analysis, the server calculates the influence weights of different production processes and production environments on the automobile performance. The server sets weights for the production process values and the production environment values in each of the initial production parameter combinations based on the weights, generating a plurality of weighted production parameter combinations. The server then prepares a quality of production monitoring model that includes the encoding network and the decoding network. The server inputs each weighted production parameter combination into the coding network to obtain a corresponding coding hidden vector. These encoded hidden vectors are passed to a decoding network that predicts the production quality metrics associated with the combination, such as fuel efficiency, safety, etc. For example, consider a weighted combination of production parameters in which the production process values include vehicle assembly time and worker training levels, and the production environment values include plant temperature and humidity. These values are converted into coded hidden vectors via the coding network, and then the combined fuel efficiency of the vehicle is predicted via the decoding network. This predicted value will serve as a target production evaluation index to help the company optimize production parameters to improve vehicle performance.
In a specific embodiment, the process of executing step S105 may specifically include the following steps:
(1) Screening out the target production parameter combination with the maximum target production evaluation index from the plurality of initial production parameter combinations;
(2) Informing an intelligent production quality management system to provide a target production quality monitoring service according to the target production parameter combination;
(3) When the target production quality monitoring service starts, the relation state among the production process data, the production environment data and the product quality data is updated.
Specifically, from among a plurality of initial production parameter combinations, a combination capable of achieving the maximum target production evaluation index needs to be screened out. The system evaluates each combination and calculates a corresponding target production evaluation index value. And selecting a combination corresponding to the maximum value from the calculated values as a target production parameter combination. This step ensures that the system selects the combination of parameters that best optimize the quality of production in the current situation. When the target production parameter combination is obtained, the system will pass this information to the intelligent production quality management system. This notification triggers the system to enter a state of the target production quality monitoring service, ready for real-time monitoring and quality assessment of the new parameter combination. When the target production quality monitoring service begins, the system will begin monitoring production process data, production environment data, and product quality data. By means of real-time monitoring, the system is able to collect a large amount of actual data that will contribute to updating the state of the relationship between the previously established production process data, production environment data and product quality data. Through analysis of the new data, the system can discover new associations, patterns, and trends, thereby continuously refining the previous relationship model. For example, consider an electronics manufacturing company, where a server uses an artificial intelligence based production quality monitoring method to improve product stability. During the production process, the server adjusts a number of production parameters, such as temperature, humidity, etc., to find the optimal combination of parameters. Through calculation and analysis, the server obtains the influence weight of different parameter combinations on the stability of the product. In one production run, the system, through monitoring and evaluation, finds that a specific combination of parameters can achieve the best product stability under the current circumstances. This combination is determined as the target production parameter combination. The system will notify the intelligent production quality management system that it is required to provide the target production quality monitoring service. When the target production quality monitoring service begins, the intelligent system will continuously collect production process data, production environment data, and product quality data. These data are used to monitor and evaluate the performance of the target production parameter combinations in real time. By analyzing the data, the system can update the relationship state between the previously established production process data, production environment data, and product quality data. For example, if the system finds that the product stability is better at a particular temperature, this information will be used to update the relationship model to make more accurate predictions and optimizations in future production.
In a specific embodiment, the process of executing step S106 may specifically include the following steps:
(1) Based on the relation state and the target production parameter combination, performing rewarding feedback calculation on an initial production quality management execution strategy of the intelligent production quality management system to obtain rewarding feedback data of the initial production quality management execution strategy;
(2) And carrying out strategy updating on the initial production quality management execution strategy of the intelligent production quality management system based on the rewarding feedback data, and generating a target production quality management execution strategy.
Specifically, based on the previously established relationship state and the target production parameter combination, the system may perform a reward feedback calculation for the initial production quality management execution strategy of the intelligent production quality management system. This calculation involves combining the relationship state with the target parameter combination and analyzing its impact on the quality of production. For example, by monitoring actual production data, quality performance under different strategies, such as product performance, stability, etc., can be assessed. Based on these data, the system can calculate a reward feedback value for each initial strategy reflecting its degree of contribution to the production quality. Based on the rewarding feedback data, the system performs strategy updating on an initial production quality management execution strategy of the intelligent production quality management system to generate a target production quality management execution strategy. This process involves adjusting parameters and settings of the initial strategy based on the reward feedback data by means of machine learning, optimization algorithms, or reinforcement learning, etc., to achieve better production quality. For example, assuming an automobile manufacturing company, the server would like to optimize the efficiency and product quality of the automobile production line through an intelligent production quality management system. In the initial stage, the server establishes a set of production quality management execution strategies including production process parameter setting, quality detection standards and the like. In actual production, the system obtains production process data, product quality data and production environment data under different strategies through real-time monitoring and data collection. Based on the previously established relationship model and target production parameter combinations, the system performs a rewards feedback calculation for each initial strategy. For example, one strategy results in a faster production rate for a production line, but has some negative impact on product quality, while another strategy balances production rate and quality. And according to the calculated reward feedback data, the system starts to update the initial production quality management execution strategy. This may be achieved by adjusting production process parameters, optimizing quality control procedures, etc. The updated strategy is more in line with the actual production condition, and the production efficiency is improved without sacrificing the product quality. Through continuous iteration and optimization, the system gradually generates a target production quality management execution strategy. This strategy allows the production line to achieve better product quality while maintaining high efficiency. Through reward feedback calculation and strategy updating, the company can continuously optimize the production flow, improve the production efficiency and the product quality,
The method for monitoring production quality based on artificial intelligence in the embodiment of the present invention is described above, and the system for monitoring production quality based on artificial intelligence in the embodiment of the present invention is described below, referring to fig. 5, and one embodiment of the system for monitoring production quality based on artificial intelligence in the embodiment of the present invention includes:
the obtaining module 501 is configured to obtain target production related data through a preset intelligent production quality management system, and divide a data set of the target production related data to obtain production process data, product quality data and production environment data;
the analysis module 502 is configured to perform a relationship analysis on the production process data and the product quality data to obtain a production process influence weight, and perform a relationship analysis on the production environment data and the product quality data to obtain a production environment influence weight;
a combination module 503, configured to combine the production environment data, the production process data, and the product quality data to obtain a plurality of initial production parameter combinations;
the monitoring module 504 is configured to input the plurality of initial production parameter combinations into a preset production quality monitoring model according to the production process influence weight and the production environment influence weight to perform production quality monitoring, so as to obtain a target production evaluation index corresponding to each initial production parameter combination;
The screening module 505 is configured to screen the plurality of initial production parameter combinations according to the target production evaluation index to obtain a target production parameter combination, and update a relationship state among the production process data, the production environment data and the product quality data according to the target production parameter combination;
and an updating module 506, configured to perform policy updating on the initial production quality management execution policy of the intelligent production quality management system based on the combination of the relation state and the target production parameter, and generate a target production quality management execution policy.
Through the cooperative cooperation of the components, the relation analysis is carried out on the production process data and the product quality data to obtain the production process influence weight, and the relation analysis is carried out on the production environment data and the product quality data to obtain the production environment influence weight; combining to obtain a plurality of initial production parameter combinations; inputting a plurality of initial production parameters into a preset production quality monitoring model for production quality monitoring to obtain a target production evaluation index; screening according to the target production evaluation index to obtain a target production parameter combination, and updating the relation state among the production process data, the production environment data and the product quality data; the invention can analyze the production process data, the product quality data and the production environment data in real time and quickly identify potential quality problems. This enables the enterprise to respond to the problem more quickly, reducing reject rate and rejection rate. By analyzing the historical data and the real-time data, the artificial intelligence can predict potential quality problems and take measures in advance to prevent the problems from occurring. Production parameters are automatically optimized to maximize production efficiency and product quality. The production process, the product quality and the environmental data are combined for analysis, so that the influence relation among different factors can be more comprehensively known. Therefore, continuous production quality improvement is realized, and therefore, the accuracy of production quality monitoring is improved, and the production efficiency and quality are further improved.
The above fig. 5 describes the artificial intelligence based production quality monitoring system in the embodiment of the present invention in detail from the point of view of the modularized functional entity, and the following describes the artificial intelligence based production quality monitoring device in the embodiment of the present invention in detail from the point of view of hardware processing.
Fig. 6 is a schematic structural diagram of an artificial intelligence-based production quality monitoring device 600 according to an embodiment of the present invention, where the artificial intelligence-based production quality monitoring device 600 may have a relatively large difference due to configuration or performance, and may include one or more processors (central processing units, CPU) 610 (e.g., one or more processors) and a memory 620, and one or more storage media 630 (e.g., one or more mass storage devices) storing application programs 633 or data 632. Wherein the memory 620 and the storage medium 630 may be transitory or persistent storage. The program stored on the storage medium 630 may include one or more modules (not shown), each of which may include a series of instruction operations on the artificial intelligence based production quality monitoring device 600. Still further, the processor 610 may be configured to communicate with the storage medium 630 to execute a series of instruction operations in the storage medium 630 on the artificial intelligence based production quality monitoring device 600.
The artificial intelligence based production quality monitoring device 600 may also include one or more power supplies 640, one or more wired or wireless network interfaces 650, one or more input/output interfaces 660, and/or one or more operating systems 631, such as Windows Server, mac OS X, unix, linux, freeBSD, and the like. It will be appreciated by those skilled in the art that the artificial intelligence based production quality monitoring device structure shown in FIG. 6 is not limiting of the artificial intelligence based production quality monitoring device and may include more or fewer components than shown, or may combine certain components, or a different arrangement of components.
The invention also provides an artificial intelligence based production quality monitoring device, which comprises a memory and a processor, wherein the memory stores computer readable instructions, and the computer readable instructions, when executed by the processor, cause the processor to execute the steps of the artificial intelligence based production quality monitoring method in the above embodiments.
The present invention also provides a computer readable storage medium, which may be a non-volatile computer readable storage medium, and may also be a volatile computer readable storage medium, where instructions are stored in the computer readable storage medium, when the instructions are executed on a computer, cause the computer to perform the steps of the artificial intelligence based production quality monitoring method.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. The production quality monitoring method based on the artificial intelligence is characterized by comprising the following steps of:
acquiring target production associated data through a preset intelligent production quality management system, and dividing a data set of the target production associated data to obtain production process data, product quality data and production environment data;
performing relation analysis on the production process data and the product quality data to obtain a production process influence weight, and performing relation analysis on the production environment data and the product quality data to obtain a production environment influence weight;
Combining the production environment data, the production process data and the product quality data to obtain a plurality of initial production parameter combinations;
inputting the initial production parameter combinations into a preset production quality monitoring model for production quality monitoring according to the production process influence weights and the production environment influence weights, and obtaining target production evaluation indexes corresponding to each initial production parameter combination;
screening the plurality of initial production parameter combinations according to the target production evaluation indexes to obtain target production parameter combinations, and updating the relation states among the production process data, the production environment data and the product quality data according to the target production parameter combinations;
and based on the relation state and the target production parameter combination, carrying out strategy updating on an initial production quality management execution strategy of the intelligent production quality management system, and generating a target production quality management execution strategy.
2. The method for monitoring production quality based on artificial intelligence according to claim 1, wherein the steps of obtaining target production-related data by a preset intelligent production quality management system, and dividing a data set of the target production-related data to obtain production process data, product quality data and production environment data, comprise:
Acquiring target production associated data through a preset intelligent production quality management system, and acquiring a production process label, a product quality label and a production environment label;
determining corresponding first production process data points, first product quality data points and first production environment data points according to the production process tags, the product quality tags and the production environment tags;
performing data point distance analysis on a plurality of original data points in the target production associated data based on the first production process data points to obtain a first data point distance between each original data point and the first production process data point; performing data point distance analysis on a plurality of original data points in the target production associated data based on the first product quality data points to obtain second data point distances between each original data point and the first product quality data points; based on the first production environment data point, carrying out data point distance analysis on a plurality of original data points in the target production associated data to obtain a third data point distance between each original data point and the first production environment data point;
performing data point screening on the plurality of original data points according to the first data point distance to obtain a plurality of first target data points, and generating corresponding production process data according to the plurality of first target data points; performing data point screening on the plurality of original data points according to the second data point distance to obtain a plurality of second target data points, and generating corresponding product quality data according to the plurality of second target data points; and carrying out data point screening on the plurality of original data points according to the third data point distance to obtain a plurality of third target data points, and generating corresponding production environment data according to the plurality of third target data points.
3. The method for monitoring production quality based on artificial intelligence according to claim 1, wherein the performing a relationship analysis on the production process data and the product quality data to obtain a production process influence weight, and performing a relationship analysis on the production environment data and the product quality data to obtain a production environment influence weight, comprises:
performing curve fitting on the production process data to obtain a plurality of production process parameter change curves, performing curve fitting on the product quality data to obtain a product quality change curve, and performing curve fitting on the production environment data to obtain a plurality of production environment parameter change curves;
extracting features of the production process parameter change curves to obtain a plurality of production process parameter features, extracting features of the product quality change curves to obtain a plurality of product quality change features, and extracting features of the production environment parameter change curves to obtain a plurality of production environment parameter change features;
calculating first correlation coefficients between the plurality of production process parameter characteristics and the plurality of product quality variation characteristics, and generating production process influence weights according to the first correlation coefficients;
And calculating a second phase relation number between the plurality of production environment parameter change characteristics and the plurality of product quality change characteristics, and generating a production environment influence weight according to the second phase relation number.
4. The artificial intelligence based production quality monitoring method of claim 1, wherein the combining the production environment data, the production process data, and the product quality data to obtain a plurality of initial production parameter combinations comprises:
inquiring a corresponding initial production quality management execution strategy through the intelligent production quality management system, and matching a corresponding combination evaluation function and a target combination type according to the initial production quality management execution strategy;
acquiring a first numerical value set corresponding to the production environment data, a second numerical value set corresponding to the production process data and a third numerical value set corresponding to the product quality data;
generating a plurality of random production parameter combinations for the first value set, the second value set and the third value set according to the target combination type;
calculating a combined evaluation index of the plurality of random production parameter combinations through the combined evaluation function to obtain a combined evaluation index of each random production parameter combination;
And screening and iterating the random production parameter combinations according to the combination evaluation index of each random production parameter combination to obtain a plurality of initial production parameter combinations, wherein each initial production parameter combination comprises a plurality of production process values, a plurality of production environment values and a production quality value.
5. The method for monitoring production quality based on artificial intelligence according to claim 4, wherein inputting the plurality of initial production parameter combinations into a preset production quality monitoring model for production quality monitoring according to the production process influence weight and the production environment influence weight to obtain target production evaluation indexes corresponding to each initial production parameter combination comprises:
setting weights of a plurality of production process values in each initial production parameter combination according to the production process influence weights, and simultaneously setting weights of a plurality of production environment values in each initial production parameter combination according to the production environment influence weights to obtain a plurality of weighted production parameter combinations;
respectively inputting the plurality of weighted production parameter combinations into a preset production quality monitoring model, wherein the production quality monitoring model comprises an encoding network and a decoding network;
Extracting coding characteristics of each weighted production parameter combination through the coding network to obtain a coding hidden vector of each weighted production parameter combination;
and inputting the coding hidden vector of each weighted production parameter combination into the decoding network to carry out the evaluation index prediction of the production quality monitoring, so as to obtain the corresponding target production evaluation index.
6. The method of claim 5, wherein the screening the plurality of initial production parameter combinations according to the target production evaluation index to obtain a target production parameter combination, and updating a relationship state among the production process data, the production environment data, and the product quality data according to the target production parameter combination, comprises:
screening out the target production parameter combination with the maximum target production evaluation index from the plurality of initial production parameter combinations;
notifying the intelligent production quality management system to provide a target production quality monitoring service according to the target production parameter combination;
when the target production quality monitoring service starts, a relation state among the production process data, the production environment data and the product quality data is updated.
7. The method of claim 6, wherein the generating a target quality of production management execution policy by policy updating an initial quality of production management execution policy of the intelligent quality of production management system based on the relationship state and the target production parameter combination comprises:
based on the relation state and the target production parameter combination, performing rewarding feedback calculation on an initial production quality management execution strategy of the intelligent production quality management system to obtain rewarding feedback data of the initial production quality management execution strategy;
and carrying out strategy updating on the initial production quality management execution strategy of the intelligent production quality management system based on the reward feedback data, and generating a target production quality management execution strategy.
8. An artificial intelligence based production quality monitoring system, characterized in that the artificial intelligence based production quality monitoring system comprises:
the acquisition module is used for acquiring target production associated data through a preset intelligent production quality management system, and dividing a data set of the target production associated data to obtain production process data, product quality data and production environment data;
The analysis module is used for carrying out relation analysis on the production process data and the product quality data to obtain production process influence weights, and carrying out relation analysis on the production environment data and the product quality data to obtain production environment influence weights;
the combination module is used for combining the production environment data, the production process data and the product quality data to obtain a plurality of initial production parameter combinations;
the monitoring module is used for inputting the plurality of initial production parameter combinations into a preset production quality monitoring model for production quality monitoring according to the production process influence weight and the production environment influence weight, and obtaining a target production evaluation index corresponding to each initial production parameter combination;
the screening module is used for screening the plurality of initial production parameter combinations according to the target production evaluation indexes to obtain target production parameter combinations, and updating the relation states among the production process data, the production environment data and the product quality data according to the target production parameter combinations;
and the updating module is used for carrying out strategy updating on the initial production quality management execution strategy of the intelligent production quality management system based on the relation state and the target production parameter combination, and generating a target production quality management execution strategy.
9. An artificial intelligence based production quality monitoring device, characterized in that the artificial intelligence based production quality monitoring device comprises: a memory and at least one processor, the memory having instructions stored therein;
the at least one processor invoking the instructions in the memory to cause the artificial intelligence based production quality monitoring device to perform the artificial intelligence based production quality monitoring method of any of claims 1-7.
10. A computer readable storage medium having instructions stored thereon, which when executed by a processor, implement the artificial intelligence based production quality monitoring method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311106420.3A CN117078105A (en) | 2023-08-30 | 2023-08-30 | Production quality monitoring method and system based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311106420.3A CN117078105A (en) | 2023-08-30 | 2023-08-30 | Production quality monitoring method and system based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117078105A true CN117078105A (en) | 2023-11-17 |
Family
ID=88719362
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311106420.3A Pending CN117078105A (en) | 2023-08-30 | 2023-08-30 | Production quality monitoring method and system based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117078105A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117289668A (en) * | 2023-11-24 | 2023-12-26 | 深圳市陶氏精密技术有限公司 | Distributed speed reducer network cooperative control method, device, equipment and storage medium |
| CN117311295A (en) * | 2023-11-28 | 2023-12-29 | 深圳百通玄武技术有限公司 | Production quality improving method and system based on wireless network equipment |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109033609A (en) * | 2018-07-20 | 2018-12-18 | 中航沈飞民用飞机有限责任公司 | Aviation machine adds part towards the method for the Product Process programming simulation of intelligence manufacture |
| US20200193354A1 (en) * | 2018-12-12 | 2020-06-18 | Hitachi, Ltd. | Information processing device and production instruction support method |
| CN111553664A (en) * | 2020-05-11 | 2020-08-18 | 重庆金美通信有限责任公司 | Method for realizing intelligent management of design and production of communication equipment based on 5G technology |
| CN113506046A (en) * | 2021-09-06 | 2021-10-15 | 江苏扬子易联智能软件有限公司 | Medicine quality control analysis system based on quality management standard data tracking |
| CN114398837A (en) * | 2022-01-19 | 2022-04-26 | 广州迈科机械有限公司 | CIP cleaning method and system based on production equipment |
| CN114677025A (en) * | 2022-03-30 | 2022-06-28 | 北京数智勘实科技有限公司 | Intelligent management system and management method for catalyst operation |
| US20220396862A1 (en) * | 2019-11-12 | 2022-12-15 | Arcelormittal | Method for scheduling production on a continuous galvanizing line |
| KR20230017556A (en) * | 2021-07-28 | 2023-02-06 | 팁스밸리(주) | System and operational methods for manufacturing execution based on artificial intelligence and bigdata |
| CN115860514A (en) * | 2022-11-09 | 2023-03-28 | 国网河北省电力有限公司物资分公司 | Low-voltage power cable quality evaluation method based on production process data |
| CN116468160A (en) * | 2023-04-04 | 2023-07-21 | 大连理工大学 | Aluminum alloy die casting quality prediction method based on production big data |
-
2023
- 2023-08-30 CN CN202311106420.3A patent/CN117078105A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109033609A (en) * | 2018-07-20 | 2018-12-18 | 中航沈飞民用飞机有限责任公司 | Aviation machine adds part towards the method for the Product Process programming simulation of intelligence manufacture |
| US20200193354A1 (en) * | 2018-12-12 | 2020-06-18 | Hitachi, Ltd. | Information processing device and production instruction support method |
| US20220396862A1 (en) * | 2019-11-12 | 2022-12-15 | Arcelormittal | Method for scheduling production on a continuous galvanizing line |
| CN111553664A (en) * | 2020-05-11 | 2020-08-18 | 重庆金美通信有限责任公司 | Method for realizing intelligent management of design and production of communication equipment based on 5G technology |
| KR20230017556A (en) * | 2021-07-28 | 2023-02-06 | 팁스밸리(주) | System and operational methods for manufacturing execution based on artificial intelligence and bigdata |
| CN113506046A (en) * | 2021-09-06 | 2021-10-15 | 江苏扬子易联智能软件有限公司 | Medicine quality control analysis system based on quality management standard data tracking |
| CN114398837A (en) * | 2022-01-19 | 2022-04-26 | 广州迈科机械有限公司 | CIP cleaning method and system based on production equipment |
| CN114677025A (en) * | 2022-03-30 | 2022-06-28 | 北京数智勘实科技有限公司 | Intelligent management system and management method for catalyst operation |
| CN115860514A (en) * | 2022-11-09 | 2023-03-28 | 国网河北省电力有限公司物资分公司 | Low-voltage power cable quality evaluation method based on production process data |
| CN116468160A (en) * | 2023-04-04 | 2023-07-21 | 大连理工大学 | Aluminum alloy die casting quality prediction method based on production big data |
Non-Patent Citations (2)
| Title |
|---|
| YU, MM (YU, MIAOMIAO): "Monitoring the data quality of data streams using a two-step control scheme", 《IISE TRANSACTIONS》, vol. 51, no. 9, 2 September 2019 (2019-09-02), pages 985 - 998 * |
| 高晗: "基于模糊综合评价的混凝土生产质量评价研究", 《混凝土世界》, no. 08, 20 August 2023 (2023-08-20), pages 82 - 86 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117289668A (en) * | 2023-11-24 | 2023-12-26 | 深圳市陶氏精密技术有限公司 | Distributed speed reducer network cooperative control method, device, equipment and storage medium |
| CN117289668B (en) * | 2023-11-24 | 2024-02-02 | 深圳市陶氏精密技术有限公司 | Distributed speed reducer network cooperative control method, device, equipment and storage medium |
| CN117311295A (en) * | 2023-11-28 | 2023-12-29 | 深圳百通玄武技术有限公司 | Production quality improving method and system based on wireless network equipment |
| CN117311295B (en) * | 2023-11-28 | 2024-01-30 | 深圳百通玄武技术有限公司 | Production quality improving method and system based on wireless network equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117078105A (en) | Production quality monitoring method and system based on artificial intelligence | |
| US10921759B2 (en) | Computer system and method for monitoring key performance indicators (KPIs) online using time series pattern model | |
| CN107464023B (en) | Online learning method and equipment of life cycle prediction model | |
| US8250006B2 (en) | Inferential sensors developed using three-dimensional pareto-front genetic programming | |
| CN111222290A (en) | Large-scale equipment residual service life prediction method based on multi-parameter feature fusion | |
| CN112987666B (en) | Power plant unit operation optimization regulation and control method and system | |
| CN110757510B (en) | Method and system for predicting remaining life of robot | |
| CN116484269B (en) | Parameter processing method, device and equipment of display screen module and storage medium | |
| CN112288137A (en) | LSTM short-term load prediction method and device considering electricity price and Attention mechanism | |
| US7480535B2 (en) | Method for the computer-supported generation of prognoses for operative systems and a system for the generation of prognoses for operative systems | |
| CN114118636A (en) | Automobile spare part demand prediction system based on multi-model optimization | |
| CN115617606A (en) | Equipment monitoring method and system, electronic equipment and storage medium | |
| CN111008724A (en) | Price prediction method and device, electronic equipment and readable storage medium | |
| CN116235148A (en) | Early warning method and device for downtime of dry pump, electronic equipment, storage medium and program | |
| CN112348259A (en) | Method for predicting purchase price of power equipment | |
| CN116880867A (en) | Policy big model-based decision engine updating method and device | |
| Sagar et al. | A novel prediction algorithm for multivariate data sets | |
| CN115619539A (en) | Pre-loan risk evaluation method and device | |
| Ramadhanti et al. | Building customer churn prediction models in Indonesian telecommunication company using decision tree algorithm | |
| CN114066050A (en) | Product index prediction method, product index prediction device and readable storage medium | |
| JP2012146205A (en) | Quality prediction device, operation condition determination method, quality prediction method, computer program, and computer-readable storage medium | |
| KR102577826B1 (en) | Methods and apparatus for generating a predictive model | |
| CN116466672B (en) | Data center machine room parameter regulation and control method based on artificial intelligence and related device | |
| KR102573979B1 (en) | Predictable methods and apparatus for develop new product | |
| Yolanda et al. | Forecasting Non-Oil and Gas Exports in Indonesia Using Double and Triple Exponential Smoothing Methods |
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 |