Intelligent control method and system for microwave sintering process of ceramic material and storage medium

Abstract

The application discloses a kind of ceramic material microwave sintering process intelligent control method, system and storable medium, wherein method includes: obtaining the working state parameter and adjustment parameter of ceramic material at the same time, and data preprocessing is carried out;Multiple classifier models are constructed based on machine learning algorithm, the working state parameter and adjustment parameter after preprocessing are formed training sample, and are input to multiple classifier models for training, to obtain the mapping relationship corresponding to multiple working state parameters and adjustment parameters;A decision machine model is constructed, multiple mapping relationships are input to the decision machine model, to obtain the data model of working state parameter and adjustment parameter, realize the sintering process of ceramic material is controlled;The present application can realize microwave sintering process to realize fast, accurate control.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of process control, in particular to a ceramic material microwave sintering process intelligent control method, system and storage medium. BACKGROUND

[0002] At present, microwave sintering technology is known as "21st century new generation sintering technology", compared with conventional sintering technology, it has the characteristics of fast heating speed, high energy utilization rate, high heating efficiency, safety, health and no pollution, and can improve the uniformity and yield of products, and improve the microstructure and performance of the sintered material.

[0003] However, with the increasing requirements of blast furnace production on sinter quality and output, the target of sintering process control is also getting higher and higher, and the traditional local basic automatic control has been unable to meet the modern production requirements. The traditional PID (proportion-integral-derivative) control generally cannot be changed at will after setting the parameters, however, in the sintering process of a certain material, the running state change law of the reaction cavity is not fixed, and the preset parameters are not always the best control state, if it is always kept unchanged, the system control precision will be reduced.

[0004] Therefore, how to provide a ceramic material sintering process control method which can solve the above problems is a problem that those skilled in the art need to solve. SUMMARY

[0005] Therefore, the present application provides a ceramic material microwave sintering process intelligent control method, system and storage medium, which trains a large amount of monitoring data by machine learning to obtain a data model, and then uses the model to realize fast and accurate control of the ceramic microwave sintering process, so as to improve the sintering control precision and efficiency.

[0006] In order to achieve the above purpose, the present application adopts the following technical scheme:

[0007] A ceramic material microwave sintering process intelligent control method, comprising:

[0008] Obtaining the working state parameters and adjusting parameters of the ceramic material at the same time, and performing data preprocessing;

[0009] Building a plurality of classifier models, the working state parameters and adjusting parameters after preprocessing are used to form training samples, and are input into a plurality of classifier models for training, to obtain a plurality of mapping relationships corresponding to the working state parameters and adjusting parameters;

[0010] Building a decision machine model, inputting a plurality of mapping relationships into the decision machine model, obtaining a data model of working state parameters and adjusting parameters, and realizing control of the sintering process of the ceramic material.

[0011] Preferably, the specific process of obtaining the data model of the working state parameter and the adjustment parameter comprises:

[0012] The mapping relationship generated by the plurality of classifier models is integrated to form a decision database;

[0013] The decision machine model sets the weight of the mapping relationship to generate the data model of the working state parameter and the adjustment parameter.

[0014] Preferably, the classifier model adopts any one or any combination of a clustering algorithm, a support vector machine algorithm and a deep learning algorithm.

[0015] Preferably, the preprocessing process comprises any one or any combination of redundant data elimination, missing data supplement and data normalization operation.

[0016] Preferably, the working state parameter comprises temperature, humidity, oxygen partial pressure, total gas pressure and reflected power of the reaction cavity, and the adjustment parameter comprises temperature rising rate and tray rotating speed.

[0017] Further, the present application also provides a ceramic material microwave sintering process intelligent control system, comprising:

[0018] A parameter acquisition module is configured to acquire working state parameters and adjustment parameters of a ceramic material in a sintering process;

[0019] A construction module is configured to construct a classifier model and a decision machine model for subsequent data processing;

[0020] A processing module is configured to preprocess the working state parameters and the adjustment parameters by using the classifier model and the decision machine model, and output corresponding data models;

[0021] A control module is configured to realize parameter control of the ceramic material in the sintering process according to the data models.

[0022] Preferably, the parameter acquisition module comprises any one or any combination of an infrared temperature measuring instrument, a humidity measuring instrument, an oxygen partial pressure measuring instrument, a micro-pressure gauge and a reflected power monitor.

[0023] Further, the present application also provides a computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, and the computer program is executed by a processor to implement the processing method according to any one of the above.

[0024] As can be seen from the above technical solutions, compared with the prior art, the present invention discloses an intelligent control method, system and storage medium for the microwave sintering process of ceramic materials. It uses machine learning technology to train the monitoring data, generate a data model, and monitor the operating status of the reaction chamber in real time to achieve real-time control of microwave sintering. The invention effectively improves the process control accuracy and sintering efficiency of microwave sintering and ensures the quality of the sintered materials. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0026] Figure 1 A schematic diagram illustrating the data model construction process of an intelligent control method for microwave sintering of ceramic materials provided by this invention;

[0027] Figure 2 The present invention provides a structural principle block diagram of an intelligent control system for the microwave sintering process of ceramic materials. Detailed Implementation

[0028] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] See appendix Figure 1 As shown in the figure, an embodiment of the present invention discloses an intelligent control method for the microwave sintering process of ceramic materials, comprising:

[0030] Obtain the set of working state parameters of ceramic materials at the same moment. and the set of adjustment parameters And perform data preprocessing;

[0031] Where n is the total number of working state parameter categories of the reaction chamber during microwave sintering of ceramic materials, m is the total number of adjustment parameter categories, and t is a time marker;

[0032] For example, the ceramic material can be selected as a chromium oxide ceramic material, and when the partial operating state parameters of the reaction chamber, such as temperature x1, humidity x2 and reflected power x3, of the chromium oxide ceramic material during microwave sintering are input, the input vector is {x1, x2, x3}, t is a time identifier, the heating rate y1 and the tray rotation speed y2 in the reaction chamber that affect the sintering quality of the chromium oxide ceramic material are selected as the output vector, and the output vector is {y1, y2};

[0033] The s classifier models are constructed, the preprocessed working state parameters and the adjustment parameters are combined into p training samples (p is a positive integer), and are input into the s classifier models for training to obtain the mapping relationship y of the s working state parameters and the adjustment parameters t ; t

[0034] For example, the microwave sintering process control of the chromium oxide ceramic material is abstracted into a multi-classification problem for various operating state parameters x i , the binary tuples between the input vector and the output vector are constructed according to the microwave sintering history statistics of the chromium oxide ceramic material as a training sample set of a support vector machine (SVM), the training sample set takes the time label t as the basis for dividing the samples, each sample records the statistical values of the input vector and the output vector at t, and the recording format of the sample is The s classifiers all use the support vector machine algorithm.

[0035] The multi-classification problem is solved, as shown in formulas (1)-(2), wherein y i is the adjustment parameter, φ(X) is the multi-classification problem of the operating state parameter, that is, the kernel function, X is the operating state parameter, w and b are the multi-classification problem parameters that can be solved by the support vector machine training, that is, the weight and the threshold value, X1, X2,..., X m are m-dimensional categories of the operating state parameter X. The multi-classification problem belongs to a nonlinear problem, and therefore the kernel function φ(X) is solved by using a Gaussian radial function, as shown in formula (3), wherein z is the center of the kernel function, γ is the width parameter of the kernel function, controls the radial range of the function, and the function can quickly realize the inner product transformation of the support vector machine, so as to realize the rapid control of the sintering process of the ceramic material.

[0036]

[0037]

[0038] ​A decision-making machine model is constructed, and multiple mapping relationships are input into the decision-making machine model. The data model of working state parameters and adjustment parameters is obtained through the decision database to realize the control of the sintering process of ceramic materials.

[0039] In a specific embodiment, the process of obtaining the data model of the operating state parameters and adjustment parameters includes:

[0040] By combining the mapping relationships generated by the s classifier models, a decision database is formed.

[0041] The decision-making machine model sets the weights {q1,q2,...,q} of the mapping relationship. s}, generate a data model of working status parameters and adjustment parameters.

[0042] In one specific embodiment, the classifier model employs any one or more of the following: clustering algorithm, support vector machine algorithm, and deep learning algorithm.

[0043] Among them, the s classifiers can choose the same or different training parameter optimization algorithms, and the training parameter optimization algorithms include: genetic algorithm, particle swarm optimization algorithm, ant colony optimization algorithm and fish swarm optimization algorithm, etc.

[0044] The weights {q1,q2,...,q} of the s classifiers are set using a decision machine. s The data model of the operating state parameters and adjustment parameters of the reaction chamber during microwave sintering of chromium oxide ceramic materials is obtained. Each classifier has its own exclusive function mapping relationship and uses the function mapping relationship as its own decision rule. The decision machine integrates the decision rules of each classifier into a decision rule library and assigns decision weights to each classifier. The decision machine performs weighted calculation on the decision results of each classifier.

[0045] In one specific embodiment, the preprocessing process includes any one or more of the following: redundant data removal, missing data supplementation, and data normalization operations.

[0046] In one specific embodiment, the operating status parameters include: temperature, humidity, oxygen partial pressure, total gas pressure, and reflected power of the reaction chamber, and the adjustment parameters include: heating rate and tray rotation speed.

[0047] See appendix Figure 2 As shown, an embodiment of the present invention provides an intelligent control system for the microwave sintering process of ceramic materials, comprising:

[0048] The parameter acquisition module is used to collect the working status parameters and adjustment parameters of ceramic materials during the sintering process.

[0049] The building module is used to construct the classifier model and decision machine model for subsequent data processing;

[0050] The processing module is used to process the working state parameters and adjustment parameters using the classifier model and decision machine model, and output the corresponding data model.

[0051] The control module is used to control the parameters of the ceramic material during the sintering process based on the data model.

[0052] In one specific embodiment, the parameter acquisition module includes any one or more of the following: an infrared thermometer, a humidity meter, an oxygen partial pressure meter, a micromanometer, and a reflection power monitor.

[0053] Furthermore, embodiments of the present invention also provide a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the processing method as described in any one of the above embodiments.

[0054] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0055] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for intelligent control of microwave sintering process of ceramic materials, characterized in that, include: The working state parameters and adjustment parameters of the ceramic material at the same moment are obtained, and the data is preprocessed. Multiple classifier models are constructed, and the preprocessed working state parameters and adjustment parameters are used to form training samples, which are then input into the multiple classifier models for training to obtain the mapping relationship between the multiple working state parameters and the adjustment parameters. A decision-making machine model is constructed, and multiple mapping relationships are input into the decision-making machine model to obtain a data model of working state parameters and adjustment parameters, thereby realizing the control of the sintering process of ceramic materials; The specific process of obtaining the data model of the operating state parameters and adjustment parameters includes: By combining the mapping relationships generated by multiple classifier models, a decision database is formed. The decision-making machine model sets the weights of the mapping relationship and generates a data model of working state parameters and adjustment parameters; The control of microwave sintering process of ceramic materials is abstracted into a multi-classification problem for various operating state parameters. Based on the historical statistics of microwave sintering of ceramic materials, a pair between the input vector and the output vector is constructed as the training sample set of support vector machine. The training sample set is divided by time index. Each sample records the statistical values ​​of the input vector and the output vector at time. Multiple classifiers adopt the support vector machine algorithm. This multi-classification problem is solved using a kernel function, which is calculated using a Gaussian radial function, thereby enabling rapid control of the sintering process of ceramic materials. Solve this multi-class classification problem, where, It's about adjusting parameters. It's a multi-class classification problem involving runtime state parameters, which is related to kernel functions. These are runtime status parameters. and The parameters, namely weights and thresholds, are obtained by training a support vector machine to solve a multi-class classification problem. These are running status parameters. The m-dimensional categories; this multi-class classification problem is a non-linear problem, therefore the kernel function... The solution uses the Gaussian radial function, where, As the center of the kernel function, The width parameter of the kernel function controls the radial range of the function. The specific formula is as follows: A decision-making machine model is constructed, and multiple mapping relationships are input into the decision-making machine model. A data model of working state parameters and adjustment parameters is obtained through a decision database to realize the control of the sintering process of ceramic materials. The operating parameters include: temperature, humidity, oxygen partial pressure, total gas pressure, and reflected power of the reaction chamber; the adjustment parameters include: heating rate and tray rotation speed.

2. The intelligent control method for the microwave sintering process of ceramic materials according to claim 1, characterized in that, The preprocessing process includes any one or more of the following: redundant data removal, missing data supplementation, and data normalization.

3. A system utilizing the intelligent control method for the microwave sintering process of ceramic materials according to any one of claims 1-2, characterized in that, include: The parameter acquisition module is used to collect the working status parameters and adjustment parameters of ceramic materials during the sintering process. The building module is used to construct classifier models and decision machine models for subsequent data processing. The processing module is used to preprocess the working state parameters and adjustment parameters using the classifier model and decision machine model, and output the corresponding data model. The control module is used to control the parameters of the ceramic material during the sintering process based on the data model.

4. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, which, when executed by a processor, implements the method as described in any one of claims 1 to 2.

Images