CN113312846B - Intelligent detection method and system based on mixed technological process - Google Patents
Intelligent detection method and system based on mixed technological process Download PDFInfo
- Publication number
- CN113312846B CN113312846B CN202110617621.4A CN202110617621A CN113312846B CN 113312846 B CN113312846 B CN 113312846B CN 202110617621 A CN202110617621 A CN 202110617621A CN 113312846 B CN113312846 B CN 113312846B
- Authority
- CN
- China
- Prior art keywords
- mixing
- obtaining
- grade
- scheme
- sampling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/211—Schema design and management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses an intelligent detection method and system based on a mixed process, which are applied to a mixed process control system, wherein the method comprises the following steps: extracting characteristics of the first material mixing scheme to obtain first characteristic information, and constructing a first characteristic database; inputting the first material mixing scheme into a first process analysis model by taking the first characteristic database as training data to obtain a first pre-estimated mixing grade, and determining a first uniformity detection scheme comprising a first sampling point and first sampling time; generating a first sampling instruction to perform sampling operation, and evaluating a sample to obtain a first mixing uniformity evaluation grade; and marking and storing the first mixing uniformity evaluation grade and the first material mixing scheme. The technical problem of among the prior art mixing process in-process, carry out the sampling point and the sampling time of degree of consistency sample to the mixture and formulate the inaccuracy to influence the degree of consistency assessment result is solved.
Description
Technical Field
The invention relates to the field of mixed processes, in particular to an intelligent detection method and system based on a mixed process.
Background
The mixing machine can mix various materials into a uniform mixture, such as cement, sand, broken stone and water are mixed into a concrete wet material; it is also possible to increase the material contact surface area. To facilitate the chemical reaction; physical changes can also be accelerated, for example, addition of a particulate solute to a solvent can accelerate dissolution and mixing by the action of a mixing mechanism. The method for measuring the mixing uniformity comprises the following steps: samples are drawn at the mixer discharge at even intervals and, if sampled in the mixer, randomly at different depth positions along the length or circumference of the aliquot mixing chamber. The selection of sampling points and the interval of sampling time become important factors for determining the mixing uniformity.
In the process of implementing the technical scheme of the invention in the embodiment of the present application, the inventor of the present application finds that the above-mentioned technology has at least the following technical problems:
in the process of the mixing process, the sampling point and the sampling time for carrying out uniformity sampling on the mixed substance are not established accurately, so that the uniformity evaluation result is influenced.
Disclosure of Invention
The embodiment of the application provides an intelligent detection method and system based on a mixed process, and solves the technical problems that in the prior art, the mixed process is inaccurate in making sampling points and sampling time for sampling uniformity of mixed substances, so that uniformity evaluation results are affected, a training database is established, the mixing grade of the mixed substances is predicted based on a neural network model, a sampling scheme is made according to a prediction result, and the sampling scheme is made to be more accurate and effective.
In view of the foregoing problems, embodiments of the present application provide an intelligent detection method and system based on a hybrid process.
In a first aspect, the present application provides a method for intelligent detection based on a hybrid process, wherein the method is applied to a hybrid process control system, and the method includes: obtaining a first material mixing scheme; performing feature extraction on the first material mixing scheme to obtain first feature information; constructing a first characteristic database according to the first characteristic information; inputting the first material mixing scheme into a first process analysis model by taking the first characteristic database as training data to obtain a first pre-estimated mixing grade; obtaining a first evenness degree detection scheme according to the first pre-estimated mixing grade, wherein the first evenness degree detection scheme comprises a first sampling point and a first sampling time; generating a first sampling instruction according to the first sampling point and the first sampling time; sending the first sampling instruction to the mixed process control system for sampling operation, and obtaining a first sample evaluation result; obtaining a first mixing uniformity evaluation grade according to the first evaluation result; and marking and storing the first mixing uniformity evaluation grade and the first material mixing scheme.
On the other hand, the application also provides an intelligent detection system based on the hybrid process, wherein the system comprises: a first obtaining unit for obtaining a first material mixing scheme; the second obtaining unit is used for extracting the characteristics of the first material mixing scheme to obtain first characteristic information; a third obtaining unit, configured to construct a first feature database according to the first feature information; the first input unit is used for inputting the first material mixing scheme into a first process analysis model by taking the first characteristic database as training data to obtain a first pre-estimated mixing grade; a fourth obtaining unit, configured to obtain a first uniformity detection scheme according to the first pre-estimated mixing level, where the first uniformity detection scheme includes a first sampling point and a first sampling time; a fifth obtaining unit, configured to generate a first sampling instruction according to the first sampling point and the first sampling time; the first sending unit is used for sending the first sampling instruction to the hybrid process control system for sampling operation and obtaining a first sample evaluation result; a sixth obtaining unit configured to obtain a first blending uniformity evaluation level from the first evaluation result; the first storage unit is used for marking and storing the first mixing evenness evaluation grade and the first material mixing scheme.
On the other hand, an embodiment of the present application further provides an intelligent detection system based on a hybrid process, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor implements the steps of the method according to the first aspect when executing the program.
One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:
the scheme of obtaining the first material mixing is adopted; performing feature extraction on the first material mixing scheme to obtain first feature information; constructing a first characteristic database according to the first characteristic information; inputting the first material mixing scheme into a first process analysis model by taking the first characteristic database as training data to obtain a first pre-estimated mixing grade; obtaining a first evenness degree detection scheme according to the first pre-estimated mixing grade, wherein the first evenness degree detection scheme comprises a first sampling point and a first sampling time; generating a first sampling instruction according to the first sampling point and the first sampling time; sending the first sampling instruction to the mixed process control system for sampling operation, and obtaining a first sample evaluation result; obtaining a first mixing uniformity evaluation grade according to the first evaluation result; and marking and storing the first mixing uniformity evaluation grade and the first material mixing scheme. The technical problem of among the prior art mixing process is in process, carry out the sampling point and the sampling time of degree of consistency sample to the mixture and formulate the inaccuracy to influence the degree of consistency assessment result is solved, realized through establishing the training database, and predict the mixture grade of mixture based on neural network model, thereby formulate the sampling scheme according to the prediction result, make the sampling scheme formulate more accurate, effectual technical purpose.
The foregoing is a summary of the present disclosure, and embodiments of the present disclosure are described below to make the technical means of the present disclosure more clearly understood.
Drawings
FIG. 1 is a schematic flow chart of an intelligent detection method based on a hybrid process according to an embodiment of the present application;
FIG. 2 is a schematic structural diagram of an intelligent detection system based on a hybrid process according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of an exemplary electronic device according to an embodiment of the present application.
Description of reference numerals: a first obtaining unit 11, a second obtaining unit 12, a third obtaining unit 13, a first input unit 14, a fourth obtaining unit 15, a fifth obtaining unit 16, a first sending unit 17, a sixth obtaining unit 18, a first storage unit 19, a bus 300, a receiver 301, a processor 302, a sender 303, a memory 304, a bus interface 305.
Detailed Description
The embodiment of the application provides an intelligent detection method and system based on a mixed process, and solves the technical problems that in the prior art, the mixed process is inaccurate in making sampling points and sampling time for sampling uniformity of mixed substances, so that uniformity evaluation results are affected, a training database is established, the mixing grade of the mixed substances is predicted based on a neural network model, a sampling scheme is made according to a prediction result, and the sampling scheme is made to be more accurate and effective.
Hereinafter, example embodiments of the present application will be described in detail with reference to the accompanying drawings, and it is apparent that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and it is to be understood that the present application is not limited by the example embodiments described herein.
Summary of the application
The method for measuring the mixing uniformity comprises the following steps: samples are drawn at the mixer discharge at even intervals and, if sampled in the mixer, randomly at different depth positions along the length or circumference of the aliquot mixing chamber. The selection of sampling points and the interval of sampling time become important factors for determining the mixing uniformity. In the prior art, the technical problem that the uniformity evaluation result is influenced by inaccurate establishment of sampling points and sampling time for uniformly sampling mixed substances in the process of a mixing process exists.
In view of the above technical problems, the technical solution provided by the present application has the following general idea:
the application provides an intelligent detection method based on a hybrid process, wherein the method is applied to a hybrid process control system and comprises the following steps: obtaining a first material mixing scheme; performing feature extraction on the first material mixing scheme to obtain first feature information; constructing a first characteristic database according to the first characteristic information; inputting the first material mixing scheme into a first process analysis model by taking the first characteristic database as training data to obtain a first pre-estimated mixing grade; obtaining a first evenness degree detection scheme according to the first pre-estimated mixing grade, wherein the first evenness degree detection scheme comprises a first sampling point and a first sampling time; generating a first sampling instruction according to the first sampling point and the first sampling time; sending the first sampling instruction to the mixed process control system for sampling operation, and obtaining a first sample evaluation result; obtaining a first mixing uniformity evaluation grade according to the first evaluation result; and marking and storing the first mixing uniformity evaluation grade and the first material mixing scheme.
Having thus described the general principles of the present application, various non-limiting embodiments thereof will now be described in detail with reference to the accompanying drawings.
Example one
As shown in fig. 1, an embodiment of the present application provides an intelligent detection method based on a hybrid process, which is applied to a hybrid process control system, wherein the method includes:
step S100: obtaining a first material mixing scheme;
specifically, the mixing device can mix various materials into a uniform mixture, such as cement, sand, gravel and water into a concrete wet material; it is also possible to increase the material contact surface area. To facilitate the chemical reaction; physical changes can also be accelerated, for example, addition of a particulate solute to a solvent can accelerate dissolution and mixing by the action of a mixing mechanism. The first material mixing scheme comprises specific schemes of a material mixing process, such as a mixing equipment type, a material type, equipment running time, a mixing process monitoring index, material pretreatment, a material input mode and the like. And inputting the first material mixing scheme into the mixed process control system, and intelligently extracting and executing process information in the first material mixing scheme by the mixed process control system. The first material mixing scheme can be an electronic version or a paper version, and the mixed process control system performs corresponding scanning or identification processing according to the file type.
Step S200: performing feature extraction on the first material mixing scheme to obtain first feature information;
step S300: constructing a first characteristic database according to the first characteristic information;
specifically, after preprocessing the first material mixing scheme in the system, the mixed process control system identifies the first material mixing scheme and extracts text information therein based on an image identification technology, and then performs feature extraction on the text information based on a semantic identification technology, thereby obtaining the first feature information. The first characteristic information comprises characteristics of mixture type, mixing proportion, mixture content, mixing equipment model, mixing time and the like. Since the mixing uniformity is related to the addition ratio, content, equipment type and the like of the mixed substances, the first characteristic database is constructed by taking each characteristic information in the first characteristic information as a data acquisition category and based on a big data information processing technology. The first feature database includes various blending process recipes for various feature information and their corresponding blending uniformity levels. And establishing the first characteristic database to lay a foundation for accurately estimating the mixing uniformity of the first material mixing scheme.
Step S400: inputting the first material mixing scheme into a first process analysis model by taking the first characteristic database as training data to obtain a first pre-estimated mixing grade;
specifically, the first process analysis model is a Neural network model (NN), which is a complex network system formed by a large number of simple processing units (called neurons) widely connected to each other, reflects many basic features of human brain functions, and is a highly complex nonlinear dynamical learning system. The first process analysis model takes the first characteristic database as training data for training and learning, so that the first material mixing scheme is input into the first process analysis model as input data, and the accurate first pre-estimated mixing grade can be obtained. The first estimated mixing grade is a mixing grade threshold value divided according to the mixing uniformity, and the higher the mixing uniformity is, the higher the first estimated mixing grade is. The first process analysis model has the characteristics of continuously learning and acquiring experience to process data, so that the output result is more accurate.
Step S500: obtaining a first evenness degree detection scheme according to the first pre-estimated mixing grade, wherein the first evenness degree detection scheme comprises a first sampling point and a first sampling time;
step S600: generating a first sampling instruction according to the first sampling point and the first sampling time;
specifically, a threshold value of uniformity of the material obtained by executing the first material mixing scheme may be obtained according to the first estimated mixing level. Therefore, a scheme required for detecting the uniformity of the material is obtained according to the material uniformity pre-estimated value, wherein the scheme comprises the first sampling point and the first sampling time. The first sampling point is a sampling position obtained when the uniformity detection is carried out by uniformly dividing the mixing chamber according to the length and the depth of the mixing chamber of the mixing device. When the uniformity detection is carried out, sampling is carried out on different sampling points, and sampling is carried out at different time thresholds so as to ensure the effectiveness of data acquisition. The first pre-estimated mixing grade is different, and the first evenness degree detection scheme required to be set is also different. For example, the higher the first estimated mixing level is, the higher the predicted value of the uniformity is, the fewer sampling points need to be set, and the longer the sampling time interval is, otherwise, the more sampling points are, the shorter the sampling time interval is.
Step S700: sending the first sampling instruction to the mixed process control system for sampling operation, and obtaining a first sample evaluation result;
specifically, after the first uniformity detection scheme is obtained, the first uniformity detection scheme is transmitted to the hybrid process control system, after the system detects that the hybrid equipment starts to operate, the materials in the mixing chamber are sampled and the samples are conveyed to the detection equipment according to the equipment operation time and the first uniformity detection scheme, the detection equipment feeds back the detection results to the hybrid process control system, the mixing uniformity of each sample is detected respectively, and comprehensive evaluation is performed to obtain the first sample evaluation result. Based on the technology of the Internet of things, the obtaining efficiency of the sample evaluation result is improved.
Step S800: obtaining a first mixing uniformity evaluation grade according to the first evaluation result;
step S900: and marking and storing the first mixing uniformity evaluation grade and the first material mixing scheme.
Specifically, after the first sample evaluation result is obtained, the overall uniformity of the mixed material is analyzed according to the uniformity evaluation result of the sample, and the first mixed uniformity evaluation grade is obtained, wherein the first mixed uniformity evaluation grade is the actual uniformity of the mixed material obtained by implementing the mixing process according to the first material mixing scheme. Because first mixed homogeneity aassessment grade with first material mixing scheme is the corresponding relation, consequently will first mixed homogeneity aassessment grade with first material mixing scheme marks the storage, and save to mixed process control system, and transmit to first characteristic database for the construction of characteristic database is more accurate, has also realized through constructing the training database to predict the mixed grade of mixed material based on neural network model, thereby formulate the sampling scheme according to the prediction result, make the sampling scheme formulate more accurate, effectual technical purpose.
Further, step S300 in the embodiment of the present application further includes:
step S310: obtaining a first feature data set from the first feature database;
step S320: performing centralized processing on the first characteristic data set to obtain a second characteristic data set;
step S330: obtaining a first covariance matrix of the second feature data set;
step S340: performing matrix operation on the first covariance matrix to obtain a first eigenvalue and a first eigenvector of the first covariance matrix;
step S350: projecting the first feature data set to the first feature vector to obtain a first dimension reduction data set, wherein the first dimension reduction data set is a feature data set obtained after dimension reduction of the first feature data set;
step S360: a second feature database is obtained from the first reduced dimensional dataset.
Specifically, the feature data extracted from the first feature database is subjected to digitization processing, a feature data set matrix is constructed, and the first feature data set is obtained. And then carrying out centralization processing on each feature data in the first feature data set, firstly solving an average value of each feature in the first feature data set, then subtracting the average value of each feature from each feature for all samples, and then obtaining a new feature value, wherein the second feature data set is formed by the new feature data set, and is a data matrix. By the covariance formula:
and operating the second characteristic data set to obtain the first covariance matrix of the second characteristic data set. Wherein the content of the first and second substances,
characteristic data in the second characteristic data set;
is the average value of the characteristic data; and M is the total amount of sample data in the second characteristic data set. Then, through matrix operation, the eigenvalue and the eigenvector of the first covariance matrix are solved, and each eigenvalue corresponds to one eigenvector. And selecting the largest first K characteristic values and the corresponding characteristic vectors from the obtained first characteristic vectors, and projecting the original characteristics in the first characteristic data set onto the selected characteristic vectors to obtain the first characteristic data set after dimension reduction. The K characteristic values can be obtained through training of the neural network model, and the accuracy of the data volume is guaranteed through the value of the K value. The feature data in the database are subjected to dimensionality reduction processing through a principal component analysis method, and redundant data are removed on the premise of ensuring the information quantity, so that the sample quantity of the feature data in the database is reduced, the loss of the information quantity after dimensionality reduction is minimum, and the operation speed of a training model on the data is accelerated.
Further, step S360 in the embodiment of the present application further includes:
step S361: inputting the first material mixing scheme into a first process analysis model by taking the second characteristic database as training data to obtain a second pre-estimated mixing grade;
step S362: analyzing defect data of the first pre-estimated mixed grade and the second pre-estimated mixed grade to obtain first missing data;
step S363: and inputting the first missing data into the first process analysis model for incremental learning to obtain a second process analysis model, wherein the second process analysis model is a new model obtained by incremental learning.
Specifically, the second feature data set is the first feature data set after dimensionality reduction, and has the characteristics of less sample size and capability of eliminating mutual influence among original data components, so that the second feature data base is used as training data, the first material mixing scheme is input to the first process analysis model, and a new output result, namely the second pre-estimated mixing grade, can be obtained. And analyzing the defect data of the first pre-estimated mixed grade and the second pre-estimated mixed grade, inputting the missing data into the first process analysis model for incremental learning, and acquiring the second process analysis model through the incremental learning, so that the accuracy of the model is improved, and the response efficiency of the model is improved.
Further, step S363 in this embodiment of the present application further includes:
step S3631: obtaining a first difference parameter of the first pre-estimated mixing grade and the second pre-estimated mixing grade;
step S3632: obtaining a first model plasticity of the first process analysis model according to the first difference parameter;
step S3633: judging whether the plasticity of the first model reaches a preset model plasticity threshold value;
step S3634: if the plasticity of the first model reaches the preset model plasticity threshold value, obtaining a first input instruction;
step S3635: inputting the first loss data into the first process analysis model according to the first input instruction.
Specifically, the first difference parameter is difference loss data obtained by performing comparative analysis on the first pre-estimated mixing grade and the second pre-estimated mixing grade obtained based on the first process analysis model, and further, the lower the first difference parameter is, the higher the plasticity of the first model is, the inverse relation is formed, wherein the preset model plasticity is the learnability of the constructed action recognition model capable of performing incremental learning, when the plasticity of the first model is higher, the higher the training performance and stability of the current model are indicated, and when the plasticity of the first model reaches a certain target plasticity, the incremental learning is performed according to the loss data to train and obtain the second process analysis model.
Further, step S800 in the embodiment of the present application further includes:
step S810: judging whether the first mixing uniformity evaluation grade is the same as the first pre-estimated mixing grade or not;
step S820: if the first mixing uniformity evaluation grade is different from the first pre-estimated mixing grade, obtaining a first grade difference value;
step S830: obtaining a first accuracy impact coefficient from the first level difference;
step S840: and obtaining first adjustment information according to the first accuracy influence coefficient, and adjusting the first uniformity detection scheme according to the first adjustment information to obtain a second uniformity adjustment scheme.
Specifically, after the evaluation of the actual mixing uniformity of the first material mixing scheme is completed through sampling analysis, whether the first mixing uniformity evaluation level is the same as the first estimated mixing level obtained by the neural network model or not is judged, and if the first mixing uniformity evaluation level is the same as the first estimated mixing level, the first mixing uniformity evaluation level is higher in accuracy and can be used as a final evaluation result; if the difference is different, the first grade difference is obtained, namely the difference between the estimated mixing grade and the actual mixing grade is obtained, the first accuracy influence coefficient is obtained according to the first grade difference, and the larger the first grade difference is, the larger the first accuracy influence coefficient is, the lower the accuracy of the first mixing uniformity evaluation grade is. The first uniformity detection scheme needs to be adjusted according to the first accuracy influence coefficient, and more accurate detection is performed on the material mixing uniformity, so that the accuracy of the first mixing uniformity evaluation grade is further improved.
Further, step S840 according to this embodiment of the present application further includes:
step S841: obtaining a second sampling instruction according to the second uniformity adjusting scheme;
step S842: sending the second sampling instruction to the mixed process control system for sampling operation, and obtaining a second sample evaluation result;
step S843: obtaining a second mixing uniformity evaluation grade according to the second evaluation result;
step S844: combining the first material mixing scheme with the second mixing uniformity evaluation grade to obtain first combined data;
step S845: updating the first combined data to the first feature database.
Specifically, after the second uniformity adjustment scheme is obtained, the second sampling instruction is generated and executed according to a new sampling point and a new sampling time, a new sample evaluation result is obtained by evaluating a sample, and the second mixing uniformity evaluation level is obtained. And the second mixing uniformity evaluation grade is a more accurate uniformity evaluation result, so that the first material mixing scheme and the second mixing uniformity evaluation grade are combined and stored, the combined data information is updated to the first characteristic database, and the data of the estimated mixing grade can be acquired more accurately through the neural network model by updating the database.
Further, step S400 in the embodiment of the present application further includes:
step S410: inputting the first material mixing scheme into a first process analysis model, wherein the first process analysis model is obtained by training a plurality of groups of training data, and each group of training data in the plurality of groups comprises: the first material mixing scheme and identification information used for identifying the first pre-estimated mixing grade;
step S420: and obtaining a first output result of the first process analysis model, wherein the first output result is a first pre-estimated mixing grade.
Specifically, the first process analysis model is a neural network model obtained by training a plurality of sets of training data, and the process of training the neural network model by the training data is essentially a supervised learning process. Each group of training data in the multiple groups of training data comprises the first material mixing scheme and identification information for identifying the first pre-estimated mixing grade; building a plurality of groups of training data by using the first pre-estimated mixing grade and identification information for identifying the first pre-estimated mixing grade, under the condition of obtaining the first material mixing scheme, outputting the identification information of the first pre-estimated mixing grade by a neural network model to verify the first pre-estimated mixing grade output by the neural network model, and if the output first pre-estimated mixing grade is consistent with the first pre-estimated mixing grade identified, finishing the data supervised learning and then carrying out the next group of data supervised learning; and if the output first pre-estimated mixing grade is not consistent with the first pre-estimated mixing grade of the identifier, adjusting the neural network model by the neural network model, and performing supervised learning on the next group of data after the neural network model reaches the expected accuracy. The neural network model is continuously corrected and optimized through training data, the accuracy of the neural network model for processing the data is improved through a supervised learning process, and the first pre-estimated mixing grade is more accurate.
In summary, the intelligent detection method based on the hybrid process provided by the embodiment of the present application has the following technical effects:
1. the scheme of obtaining the first material mixing is adopted; performing feature extraction on the first material mixing scheme to obtain first feature information; constructing a first characteristic database according to the first characteristic information; inputting the first material mixing scheme into a first process analysis model by taking the first characteristic database as training data to obtain a first pre-estimated mixing grade; obtaining a first evenness degree detection scheme according to the first pre-estimated mixing grade, wherein the first evenness degree detection scheme comprises a first sampling point and a first sampling time; generating a first sampling instruction according to the first sampling point and the first sampling time; sending the first sampling instruction to the mixed process control system for sampling operation, and obtaining a first sample evaluation result; obtaining a first mixing uniformity evaluation grade according to the first evaluation result; and marking and storing the first mixing uniformity evaluation grade and the first material mixing scheme. The technical problem of among the prior art mixing process is in process, carry out the sampling point and the sampling time of degree of consistency sample to the mixture and formulate the inaccuracy to influence the degree of consistency assessment result is solved, realized through establishing the training database, and predict the mixture grade of mixture based on neural network model, thereby formulate the sampling scheme according to the prediction result, make the sampling scheme formulate more accurate, effectual technical purpose.
2. The neural network model is adopted for training and learning, the first material mixing scheme is input into the first process analysis model, the first pre-estimated mixing grade is obtained, the data processing characteristics can be continuously learned and obtained based on the neural network model, the first pre-estimated mixing grade is more accurately obtained through training data, and therefore the first uniformity detection scheme is more accurately formulated.
3. Because the feature data in the database are subjected to dimensionality reduction by a principal component analysis method, redundant data are removed on the premise of ensuring the information quantity, so that the sample quantity of the feature data in the database is reduced, the loss of the information quantity after dimensionality reduction is minimum, and the operation speed of a training model on the data is accelerated. And the model is optimized through incremental learning, and the technical effect of ensuring the stability and the accuracy of the output performance of the model is achieved.
Example two
Based on the same inventive concept as the intelligent detection method based on the hybrid process in the foregoing embodiment, the present invention further provides an intelligent detection system based on the hybrid process, as shown in fig. 2, the system includes:
a first obtaining unit 11, wherein the first obtaining unit 11 is used for obtaining a first material mixing scheme;
a second obtaining unit 12, where the second obtaining unit 12 is configured to perform feature extraction on the first material mixing scheme to obtain first feature information;
a third obtaining unit 13, where the third obtaining unit 13 is configured to construct a first feature database according to the first feature information;
the first input unit 14 is configured to input the first material mixing scheme to a first process analysis model by using the first feature database as training data to obtain a first pre-estimated mixing grade;
a fourth obtaining unit 15, where the fourth obtaining unit 15 is configured to obtain a first uniformity detection scheme according to the first estimated mixture level, where the first uniformity detection scheme includes a first sampling point and a first sampling time;
a fifth obtaining unit 16, wherein the fifth obtaining unit 16 is configured to generate a first sampling instruction according to the first sampling point and the first sampling time;
a first sending unit 17, where the first sending unit 17 is configured to send the first sampling instruction to the hybrid process control system for sampling operation, and obtain a first sample evaluation result;
a sixth obtaining unit 18, wherein the sixth obtaining unit 18 is configured to obtain a first blending uniformity evaluation level from the first evaluation result;
a first storage unit 19, wherein the first storage unit 19 is used for storing the first mixing uniformity evaluation grade and the first material mixing scheme in a marking mode.
Further, the system further comprises:
a seventh obtaining unit configured to obtain a first feature data set from the first feature database;
an eighth obtaining unit, configured to obtain a second feature data set by performing centralized processing on the first feature data set;
a ninth obtaining unit for obtaining a first covariance matrix of the second feature data set;
a tenth obtaining unit, configured to perform matrix operation on the first covariance matrix to obtain a first eigenvalue and a first eigenvector of the first covariance matrix;
an eleventh obtaining unit, configured to project the first feature data set to the first feature vector to obtain a first dimension-reduced data set, where the first dimension-reduced data set is a feature data set obtained after dimension reduction of the first feature data set;
a twelfth obtaining unit for obtaining a second feature database from the first dimension-reduced dataset.
Further, the system further comprises:
the second input unit is used for inputting the first material mixing scheme into a first process analysis model by taking the second characteristic database as training data to obtain a second pre-estimated mixing grade;
a thirteenth obtaining unit, configured to perform defect data analysis on the first pre-estimated mixed level and the second pre-estimated mixed level to obtain first missing data;
a third input unit, configured to input the first missing data to the first process analysis model for incremental learning, so as to obtain a second process analysis model, where the second process analysis model is a new model obtained by incremental learning.
Further, the system further comprises:
a fourteenth obtaining unit, configured to obtain a first difference parameter between the first pre-estimated mixing level and the second pre-estimated mixing level;
a fifteenth obtaining unit, configured to obtain a first model plasticity of the first process analysis model according to the first difference parameter;
the first judging unit is used for judging whether the plasticity of the first model reaches a preset model plasticity threshold value;
a sixteenth obtaining unit, configured to obtain a first input instruction if the first model plasticity reaches the preset model plasticity threshold;
a fourth input unit to input the first loss data into the first process analysis model according to the first input instruction.
Further, the system further comprises:
a second judging unit, configured to judge whether the first blending uniformity evaluation level is the same as the first pre-estimated blending level;
a seventeenth obtaining unit, configured to obtain a first level difference if the first blending uniformity evaluation level is different from the first pre-estimated blending level;
an eighteenth obtaining unit configured to obtain a first accuracy influence coefficient from the first level difference;
and the first adjusting unit is used for obtaining first adjusting information according to the first accuracy influence coefficient, and adjusting the first evenness degree detection scheme according to the first adjusting information to obtain a second evenness degree adjusting scheme.
Further, the system further comprises:
a nineteenth obtaining unit, configured to obtain a second sampling instruction according to the second uniformity adjustment scheme;
a twentieth obtaining unit, configured to send the second sampling instruction to the hybrid process control system for sampling operation, and obtain a second sample evaluation result;
a twenty-first obtaining unit configured to obtain a second blending uniformity evaluation level from the second evaluation result;
a twenty-second obtaining unit, configured to combine the first material mixing scheme with the second mixing uniformity evaluation level to obtain first combined data;
a twenty-third obtaining unit for updating the first combined data to the first feature database.
Further, the system further comprises:
a fifth input unit, configured to input the first material mixing scheme to a first process analysis model, where the first process analysis model is obtained through training of multiple sets of training data, and each set of training data in the multiple sets includes: the first material mixing scheme and identification information used for identifying the first pre-estimated mixing grade;
a twenty-fourth obtaining unit, configured to obtain a first output result of the first process analysis model, where the first output result is a first pre-estimated mixing level.
The foregoing intelligent detection method based on hybrid process in the first embodiment of fig. 1 and the specific examples are also applicable to the intelligent detection system based on hybrid process in the present embodiment, and a person skilled in the art can clearly know the intelligent detection system based on hybrid process in the present embodiment through the foregoing detailed description of the intelligent detection method based on hybrid process, so for the brevity of the description, detailed descriptions are omitted here.
Exemplary electronic device
The electronic device of the embodiment of the present application is described below with reference to fig. 3.
Fig. 3 illustrates a schematic structural diagram of an electronic device according to an embodiment of the present application.
Based on the inventive concept of the hybrid process based intelligent detection method, the invention further provides a hybrid process based intelligent detection system, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the steps of any one of the methods of the hybrid process based intelligent detection method.
Where in fig. 3 a bus architecture (represented by bus 300), bus 300 may include any number of interconnected buses and bridges, bus 300 linking together various circuits including one or more processors, represented by processor 302, and memory, represented by memory 304. The bus 300 may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. A bus interface 305 provides an interface between the bus 300 and the receiver 301 and transmitter 303. The receiver 301 and the transmitter 303 may be the same element, i.e., a transceiver, providing a means for communicating with various other apparatus over a transmission medium.
The processor 302 is responsible for managing the bus 300 and general processing, and the memory 304 may be used for storing data used by the processor 302 in performing operations.
The application provides an intelligent detection method based on a hybrid process, wherein the method is applied to a hybrid process control system and comprises the following steps: obtaining a first material mixing scheme; performing feature extraction on the first material mixing scheme to obtain first feature information; constructing a first characteristic database according to the first characteristic information; inputting the first material mixing scheme into a first process analysis model by taking the first characteristic database as training data to obtain a first pre-estimated mixing grade; obtaining a first evenness degree detection scheme according to the first pre-estimated mixing grade, wherein the first evenness degree detection scheme comprises a first sampling point and a first sampling time; generating a first sampling instruction according to the first sampling point and the first sampling time; sending the first sampling instruction to the mixed process control system for sampling operation, and obtaining a first sample evaluation result; obtaining a first mixing uniformity evaluation grade according to the first evaluation result; and marking and storing the first mixing uniformity evaluation grade and the first material mixing scheme. The technical problem of among the prior art mixing process is in process, carry out the sampling point and the sampling time of degree of consistency sample to the mixture and formulate the inaccuracy to influence the degree of consistency assessment result is solved, realized through establishing the training database, and predict the mixture grade of mixture based on neural network model, thereby formulate the sampling scheme according to the prediction result, make the sampling scheme formulate more accurate, effectual technical purpose.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a system for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction system which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks. While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (9)
1. An intelligent detection method based on a hybrid process, wherein the method is applied to a hybrid process control system, and comprises the following steps:
obtaining a first material mixing scheme;
performing feature extraction on the first material mixing scheme to obtain first feature information;
constructing a first characteristic database according to the first characteristic information;
inputting the first material mixing scheme into a first process analysis model by taking the first characteristic database as training data to obtain a first pre-estimated mixing grade;
obtaining a first evenness degree detection scheme according to the first pre-estimated mixing grade, wherein the first evenness degree detection scheme comprises a first sampling point and a first sampling time;
generating a first sampling instruction according to the first sampling point and the first sampling time;
sending the first sampling instruction to the mixed process control system for sampling operation, and obtaining a first sample evaluation result;
obtaining a first mixing uniformity evaluation grade from the first sample evaluation result;
and marking and storing the first mixing uniformity evaluation grade and the first material mixing scheme.
2. The method of claim 1, wherein said building a first feature database from said first feature information further comprises:
obtaining a first feature data set from the first feature database;
performing centralized processing on the first characteristic data set to obtain a second characteristic data set;
obtaining a first covariance matrix of the second feature data set;
performing matrix operation on the first covariance matrix to obtain a first eigenvalue and a first eigenvector of the first covariance matrix;
projecting the first feature data set to the first feature vector to obtain a first dimension reduction data set, wherein the first dimension reduction data set is a feature data set obtained after dimension reduction of the first feature data set;
a second feature database is obtained from the first reduced dimensional dataset.
3. The method of claim 2, wherein the method comprises:
inputting the first material mixing scheme into a first process analysis model by taking the second characteristic database as training data to obtain a second pre-estimated mixing grade;
analyzing defect data of the first pre-estimated mixed grade and the second pre-estimated mixed grade to obtain first missing data;
and inputting the first missing data into the first process analysis model for incremental learning to obtain a second process analysis model, wherein the second process analysis model is a new model obtained by incremental learning.
4. The method of claim 3, wherein the method comprises:
obtaining a first difference parameter of the first pre-estimated mixing grade and the second pre-estimated mixing grade;
obtaining a first model plasticity of the first process analysis model according to the first difference parameter;
judging whether the plasticity of the first model reaches a preset model plasticity threshold value;
if the plasticity of the first model reaches the preset model plasticity threshold value, obtaining a first input instruction;
inputting the first missing data into the first process analysis model according to the first input instruction.
5. The method of claim 1, wherein the method comprises:
judging whether the first mixing uniformity evaluation grade is the same as the first pre-estimated mixing grade or not;
if the first mixing uniformity evaluation grade is different from the first pre-estimated mixing grade, obtaining a first grade difference value;
obtaining a first accuracy impact coefficient from the first level difference;
and obtaining first adjustment information according to the first accuracy influence coefficient, and adjusting the first uniformity detection scheme according to the first adjustment information to obtain a second uniformity adjustment scheme.
6. The method of claim 5, wherein the method comprises:
obtaining a second sampling instruction according to the second uniformity adjusting scheme;
sending the second sampling instruction to the mixed process control system for sampling operation, and obtaining a second sample evaluation result;
obtaining a second mixing uniformity evaluation grade according to the second sample evaluation result;
combining the first material mixing scheme with the second mixing uniformity evaluation grade to obtain first combined data;
updating the first combined data to the first feature database.
7. The method of claim 1, wherein the method comprises:
inputting the first material mixing scheme into a first process analysis model, wherein the first process analysis model is obtained by training a plurality of groups of training data, and each group of training data in the plurality of groups comprises: the first material mixing scheme and identification information used for identifying the first pre-estimated mixing grade;
and obtaining a first output result of the first process analysis model, wherein the first output result is a first pre-estimated mixing grade.
8. An intelligent detection system based on a hybrid process, wherein the system comprises:
a first obtaining unit for obtaining a first material mixing scheme;
the second obtaining unit is used for extracting the characteristics of the first material mixing scheme to obtain first characteristic information;
a third obtaining unit, configured to construct a first feature database according to the first feature information;
the first input unit is used for inputting the first material mixing scheme into a first process analysis model by taking the first characteristic database as training data to obtain a first pre-estimated mixing grade;
a fourth obtaining unit, configured to obtain a first uniformity detection scheme according to the first pre-estimated mixing level, where the first uniformity detection scheme includes a first sampling point and a first sampling time;
a fifth obtaining unit, configured to generate a first sampling instruction according to the first sampling point and the first sampling time;
the first sending unit is used for sending the first sampling instruction to the hybrid process control system for sampling operation and obtaining a first sample evaluation result;
a sixth obtaining unit for obtaining a first blending uniformity evaluation level from the first sample evaluation result;
the first storage unit is used for marking and storing the first mixing evenness evaluation grade and the first material mixing scheme.
9. A hybrid process based intelligent detection system comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method of any one of claims 1 to 7 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110617621.4A CN113312846B (en) | 2021-06-02 | 2021-06-02 | Intelligent detection method and system based on mixed technological process |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110617621.4A CN113312846B (en) | 2021-06-02 | 2021-06-02 | Intelligent detection method and system based on mixed technological process |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113312846A CN113312846A (en) | 2021-08-27 |
| CN113312846B true CN113312846B (en) | 2022-02-08 |
Family
ID=77377201
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110617621.4A Active CN113312846B (en) | 2021-06-02 | 2021-06-02 | Intelligent detection method and system based on mixed technological process |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113312846B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114612007B (en) * | 2022-04-11 | 2023-05-30 | 宁夏全通枸杞产业有限公司 | Combined production regulation method and system for wolfberry brandy and wolfberry health vinegar |
| CN117359812B (en) * | 2023-12-08 | 2024-02-06 | 天津滨海津丽电子材料股份有限公司 | Intelligent preparation method and system for epoxy resin sealing glue block |
| CN117970896B (en) * | 2024-04-02 | 2024-06-18 | 安徽伟宏食品有限公司 | Intelligent mixing method and device for milk tea powder, electronic equipment and medium |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102661911A (en) * | 2012-04-26 | 2012-09-12 | 西安理工大学 | Method for rapidly detecting mixing uniformity of powdery material |
| CN105893715A (en) * | 2016-06-01 | 2016-08-24 | 江苏科技大学 | Process information intelligent matching method based on machining characteristics |
| JP7280705B2 (en) * | 2019-02-07 | 2023-05-24 | 株式会社日本総合研究所 | Machine learning device, program and machine learning method |
| CN112003901B (en) * | 2020-07-29 | 2023-05-05 | 广东溢达纺织有限公司 | Leasing process information importing method and device, computer equipment and storage medium |
-
2021
- 2021-06-02 CN CN202110617621.4A patent/CN113312846B/en active Active
Non-Patent Citations (1)
| Title |
|---|
| 环保智能型气流粉碎混合系统在粉体农药生产中的推广运用;吴建明等;《现代农药》;20101231(第06期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113312846A (en) | 2021-08-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113312846B (en) | Intelligent detection method and system based on mixed technological process | |
| CN110352660B (en) | Method and device for processing information of rapid nondestructive testing of vitality of delinted cotton seeds | |
| Reynolds | Estimating the error in model predictions | |
| CN113267607B (en) | Characteristic parameter identification system for field organic pollutant migration process | |
| Mazer et al. | Fuzzy logic for estimating chloride diffusion in concrete | |
| CN111199493A (en) | Arrearage risk identification method based on customer payment information and credit investigation information | |
| CN113393910A (en) | Preparation method and system of high-toughness steel slag micro-surfacing material | |
| Alavi et al. | Development of a machine learning model for on-site evaluation of concrete compressive strength by SonReb | |
| AU2021101715A4 (en) | A transfer learning based model for fatigue crack initiation sites detection | |
| CN113255941A (en) | Method and device for treating bridge construction waste | |
| Sbartaï et al. | Combining the bi-objective approach and conditional coring for a reliable estimation of on-site concrete strength variability | |
| Elhishi et al. | Unboxing machine learning models for concrete strength prediction using XAI | |
| CN111901330A (en) | Ensemble learning model construction method, ensemble learning model identification device, server and medium | |
| CN109359047B (en) | Software testing sufficiency measuring method based on system characteristic state | |
| CN110855519A (en) | Network flow prediction method | |
| Ballesteros-Pérez et al. | Standard deviation of bids for construction contract auctions | |
| Castelli et al. | An evolutionary system for the prediction of high performance concrete strength based on semantic genetic programming | |
| CN113533638B (en) | Finished product detection method and system for special chemicals | |
| CN114782099A (en) | Budget estimation and analysis method for construction cost | |
| CN115440299A (en) | Method, apparatus, medium and program product for determining background microorganisms | |
| CN115127856A (en) | Method and device for sampling and identifying concrete test block compression test robot | |
| Marsili et al. | A Bayesian network for the definition of probability models for compressive strength of concrete homogeneous population | |
| JP2006313428A (en) | Database evaluation device | |
| CN113567369A (en) | Forest environment monitoring method and system based on multispectral remote sensing | |
| Silva et al. | A comparative study of machine learning methods for compressive strength of concrete |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |