Ceramic sintering process thermal deformation intelligent prediction and adaptive temperature precise regulation and control system

Abstract

The invention relates to the technical field of ceramic intelligent manufacturing, in particular to a ceramic firing process thermal deformation intelligent prediction and self-adaptive temperature precise regulation and control system, which comprises a multi-source sensing module (a thermal infrared camera array, a temperature sensor, a laser displacement meter and a powder characteristic analyzer); a hybrid prediction model (a space-time convolutional neural network constrained by a physical mechanism); and the self-adaptive temperature control actuator comprises a partitioned burner group, a quenching / slow cooling pipeline and a gas-air volume decoupling controller. According to the scheme, through a multi-mode perception-hybrid modeling-dynamic optimization full-process closed-loop architecture, the single-point technical limitation of existing patents is broken through, and the zero-defect precise manufacturing target of ceramic firing is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ceramic intelligent manufacturing, in particular to a ceramic firing process thermal deformation intelligent prediction and self-adaptive temperature precise regulation and control system. BACKGROUND

[0002] There is a deformation prediction method based on random forest (patent CN119207602A), specifically, a building ceramic firing deformation prediction method based on machine learning is disclosed, which is based on the actual production data of a certain enterprise, obtains the press top powder moisture, particle size, firing kiln temperature and firing deformation detection data, matches the collected data, processes abnormal values, selects features and standardizes data. The processed data is divided into training set and test set according to 4:1, and random forest and support vector regression algorithm are selected to build prediction model. The training set is used for model training, and the test set data is used for model verification;

[0003] The disclosed partition temperature difference control technology (patent CN108046763A) specifically discloses a dry porcelain sleeve firing adjustment method at high temperature, which controls the upper and lower part of the porcelain sleeve to be low in temperature and the middle part to be high in temperature during the reduction stage, and the temperature is increased from 970 DEG C to the fire stopping temperature. By reducing the temperature in a certain range of the local area of the kiln at high temperature, the generation of deformation defects is effectively prevented; the firing adjustment method is improved by reducing the temperature in a certain range of the local area of the kiln at high temperature, and is suitable for the firing of a million-volt hollow porcelain sleeve blank, forming a unified specification, and effectively preventing the generation of deformation;

[0004] The authorized hybrid driving temperature prediction model (patent CN113627064B) specifically discloses a roller kiln firing zone temperature prediction method based on mechanism and data hybrid driving, which comprises the following steps: S1: modeling the mechanism process of the roller kiln firing zone by the law of conservation of mass and the law of conservation of energy, and solving by the finite difference method; S2: establishing an error compensation model based on data driving based on the instant moving window and the local weighted kernel principal component regression; S3: combining the mechanism model obtained in S1 and the error compensation model obtained in S2 to obtain a roller kiln firing zone temperature hybrid prediction model; S4: predicting the roller kiln firing zone temperature by using the roller kiln firing zone temperature hybrid prediction model. The present application establishes a mechanism model of the ceramic roller kiln firing zone temperature, and simultaneously uses a data modeling method to compensate the error of the mechanism model result, and combines the hybrid modeling strategy of mechanism modeling and data modeling to ensure the effectiveness of the model when the working condition changes.

[0005] The thermal deformation in the ceramic firing process is mainly caused by non-uniform shrinkage of the material, temperature gradient stress and phase change. The traditional solution has three defects:

[0006] Prediction hysteresis: relying on artificial offline sampling of deformation data, unable to monitor in real time;

[0007] Model limitations: single mechanism model is difficult to handle complex reactions (such as green body creep, sintering stress), and pure data-driven model lacks physical constraints, resulting in poor working condition adaptability;

[0008] Control roughness: the existing temperature control system only adjusts the gas valve through PID, and cannot accurately suppress local deformation;

[0009] Therefore, it is imperative to break through the single-point technical limitation by constructing a "multi-modal perception-hybrid model prediction-multi-objective optimization control" closed-loop system. For this purpose, we propose an intelligent prediction and adaptive temperature precise control system for thermal deformation in the ceramic sintering process. SUMMARY

[0010] The purpose of the present application is to provide an intelligent prediction and adaptive temperature precise control system for thermal deformation in the ceramic sintering process.

[0011] In order to achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0012] The intelligent prediction and adaptive temperature precise control system for thermal deformation in the ceramic sintering process comprises:

[0013] Multi-source perception module: thermal infrared camera array, temperature sensor, laser displacement meter, powder property analyzer;

[0014] Hybrid prediction model: spatiotemporal convolutional neural network with physical mechanism constraints;

[0015] Adaptive temperature control actuator: zoned burner group, rapid / annealing cooling pipeline, gas-air volume decoupling controller.

[0016] Preferably, the physical mechanism module integrates the sintering strain rate equation as

[0017] Preferably, the burner control strategy adopts a combination of long and short combustion diverters:

[0018] The long diverter has a flame length of ≥300mm, covering the center area of the green body;

[0019] The short diverter has a flame length of ≤150mm, acting on the edge area.

[0020] Preferably, the digital twin platform reconstructs the three-dimensional deformation of the green body based on the VTK algorithm, and outputs the linear shrinkage rate-temperature curve and the density cloud chart in real time.

[0021] The present application has at least the following beneficial effects:

[0022] Prediction accuracy improvement: hybrid model error reduction of more than 40% compared to single data-driven model;

[0023] Energy consumption optimization: gas utilization rate increased by 15% to 20%;

[0024] Universal expansion: suitable for building ceramics, aerospace ceramic cores, high-voltage porcelain sleeves and other multi-element scenarios.

[0025] The scheme avoids the single-point technical limitations of existing patents through a multi-modal perception-hybrid modeling-dynamic optimization closed-loop architecture, achieving "zero defect" precision manufacturing of ceramic firing. BRIEF DESCRIPTION OF DRAWINGS

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings described below are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0027] Figure 1 The schematic diagram of the present application. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical scheme and advantages of the present application more clear, the following will further describe the present application in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0029] Referring to Figure 1 , the intelligent prediction and adaptive temperature precision control system of ceramic firing process thermal deformation is composed of a closed-loop control chain of three modules:

[0030] Multi-source perception layer: deploy multispectral thermal infrared camera (collecting green body surface temperature field), embedded temperature sensor array (monitoring kiln environment temperature), laser displacement sensor (real-time deformation detection), and powder characteristics online analyzer (particle size, moisture);

[0031] Intelligent decision-making layer: hybrid prediction model is constructed by combining physical mechanism and deep learning to generate dynamic temperature control instructions;

[0032] Execution control layer: distributed burner group based on gas / air volume double-loop regulation, realizing kiln cross-section temperature difference ≤5.

[0033] Thermal deformation hybrid prediction model

[0034] Physical mechanism module:

[0035] Build thermal-mechanical-chemical coupling equation to quantify sintering strain rate:

[0036]

[0037] wherein, in formula γ s is the surface energy, R is the particle radius, K v is the bulk viscosity;

[0038] Data-driven module:

[0039] The spatio-temporal convolutional neural network (ST-CNN) is used to process multi-source time-series data:

[0040] Input: 42-dimensional temperature data (each partition of the kiln), powder particle size distribution (60 mesh / 100 mesh / 200 mesh residue), moisture content;

[0041] Feature extraction:

[0042] The spatial attention convolution extracts the local deformation features of the thermal infrared image;

[0043] The 3D convolution fuses the time dimension thermal distribution evolution law;

[0044] Output: deformation probability distribution and key deformation area (such as the difference in shrinkage rate between the edge and center of the body).

[0045] Model fusion mechanism:

[0046] The output of the physical model is used as a constraint condition for the deep learning model, and the data bias is corrected through residual adversarial training, with a prediction error of ≤3% (40% lower than a single model).

[0047] Adaptive temperature regulation strategy

[0048] Multi-objective optimization function: min J = α|ΔT desired -ΔT actual |+β|δ shape |+γ·E

[0049] wherein, ΔT desired is the desired temperature difference, δ shape is the deformation prediction value, and the weight coefficients α, β, γ are adjusted online.

[0050] The sintering strain rate equation is integrated into the physical mechanism module as

[0051] The long-short combustion diverter combination is used for the burner control strategy:

[0052] The long diverter has a flame length ≥300mm, covering the center area of the body;

[0053] The short diverter has a flame length ≤150mm, acting on the edge area.

[0054] Dynamic matching of actuators:

[0055] Staged control of the burner:

[0056] Long combustion splitter adjusts center zone temperature, short splitter controls edge zone;

[0057] Quenching / annealing synergy: based on deformation prediction to trigger gradient cooling (cooling rate 10-20℃ / h);

[0058] Gas flow and air volume decoupling control: air / gas ratio is accurately maintained at (12.5-13.5):15.

[0059] Digital twin platform reconstructs the three-dimensional deformation of the blank based on VTK algorithm, and outputs the linear shrinkage-temperature curve and density cloud map in real time.

[0060] Take the sintering of an aero ceramic core as an example:

[0061] Data acquisition:

[0062] Final sintering temperature 1280℃, heating rate 50℃ / h, powder particle size D50=15μm;

[0063] Embed process parameters into the RGB channel of the green body slice.

[0064] Deformation prediction:

[0065] ST-CNN model outputs edge shrinkage rate exceeding threshold (predicted value 0.8%, threshold 0.5%);

[0066] Physical module verification shows that the sudden change in creep strain rate is the cause.

[0067] Temperature control execution:

[0068] Reduce the edge zone burner gas flow from 28→25m 3 / h, temperature difference control is ΔT=10℃;

[0069] Extend the holding time by 30min to make the deformation fall back to the safe interval.

[0070] Effect verification:

[0071] Finished product straightness deviation ≤0.1mm (compared to the uncontrolled system, improved by 80%), gas loss reduced by 12%.

[0072]

[0073] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above-mentioned embodiments, and the above-mentioned embodiments and descriptions in the specification are only the principles of the present application. Various changes and improvements can be made without departing from the spirit and scope of the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A system for intelligent prediction of thermal distortion and adaptive precise temperature control during ceramic firing process, characterized in that, Comprise: Multi-source perception module: thermal infrared camera array, temperature sensor, laser displacement meter, powder property analyzer; Hybrid prediction model: physically mechanism-constrained spatio-temporal convolutional neural network; Adaptive temperature control actuator: zoned burner group, quenching / annealing pipeline, gas-air volume decoupling controller.

2. The intelligent prediction and adaptive temperature control system for ceramic sintering process according to claim 1, characterized in that: The physical mechanism module integrates the sintering strain rate equation as 3.The intelligent prediction and self-adaptive temperature precise control system for thermal deformation in ceramic sintering process according to claim 1, wherein, The burner control strategy adopts a combination of long and short combustion diverters: The long diverter has a flame length of ≥300mm, covering the center area of the body; The short diverter has a flame length of ≤150mm, acting on the edge area. 4.The intelligent prediction and self-adaptive temperature precise control system for thermal deformation in ceramic sintering process according to claim 1, characterized in that, The digital twin platform reconstructs the 3D deformation of the body based on the VTK algorithm, and outputs the linear shrinkage-temperature curve and the density cloud chart in real time.

Images