Temperature control method in ceramic production process

Abstract

The invention discloses a temperature control method in a ceramic production process, which comprises the following steps of: establishing a heat transfer model of a ceramic shuttle kiln, and fitting an optimal temperature rise curve; setting parameters of a PID algorithm by using a deep belief neural network so as to ensure that the actual temperature in the ceramic production process follows the optimal temperature rise curve; designing a PID control method based on the fuzzy deep belief neural network by combining a fuzzy control method and a deep learning method; an improved Levenberg-Marquardt algorithm is applied to carry out parameter estimation on the fuzzy deep belief neural network so as to optimize a PID control method; based on the optimized PID control method, the temperature in the ceramic production process is controlled. Through accurate temperature control, the production efficiency and the consistency and quality of ceramic products are improved; meanwhile, automation and intelligentization of temperature control of the ceramic shuttle kiln are achieved, and flexibility and reliability of operation are improved.

Description

Technical Field

[0001] The invention belongs to the technical field of temperature control in a ceramic production process, and in particular relates to a temperature control method in a ceramic production process. Background Art

[0002] The color and molding effect of ceramics are closely related to the temperature changes during the ceramic firing process. Famous ceramics have very precise temperature control during the firing process. Therefore, temperature control in the ceramic production process is a key link, which directly affects the quality and performance of ceramic products.

[0003] Ceramic materials have specific temperature requirements. Too high or too low a temperature may lead to firing failure. Especially in ceramic materials fired at high temperatures, such as porcelain and advanced ceramics, temperature control is particularly important. Currently available temperature control systems may lack sufficient accuracy, resulting in temperature fluctuations and affecting product quality.

[0004] Controlling the temperature rise and fall speed is also an important consideration. Too fast a temperature rise and fall speed may cause excessive stress inside the ceramic material, causing cracking; conversely, too slow a temperature rise and fall speed will extend the firing cycle and increase energy consumption. Existing control systems may not be able to reasonably control the temperature rise and fall speed according to the specific requirements of ceramic materials and products.

[0005] At present, the existing ceramic kiln temperature control mostly adopts traditional PID control. This control strategy is relatively simple and difficult to meet the temperature control requirements of building ceramic kiln firing. Moreover, the setting of PID control parameters usually relies on manual experience and lacks adaptive and self-tuning functions, which limits the flexibility and response speed of the control strategy. Summary of the invention

[0006] In order to solve the above technical problems, the present invention proposes a temperature control method in a ceramic production process to solve the above problems existing in the prior art.

[0007] To achieve the above object, the present invention provides a temperature control method in a ceramic production process, comprising:

[0008] Establish a heat transfer model for ceramic shuttle kiln and fit the optimal heating curve;

[0009] Using a deep belief neural network to adjust the parameters of the PID algorithm to ensure that the actual temperature in the ceramic production process follows the optimal heating curve;

[0010] Combining fuzzy control methods and deep learning methods, a PID control method based on fuzzy deep belief neural network is designed;

[0011] The improved Levenberg-Marquardt algorithm is applied to the parameter estimation of fuzzy deep belief neural network to optimize the PID control method;

[0012] Based on the optimized PID control method, the temperature in the ceramic production process is controlled.

[0013] Preferably, the heat transfer model includes a dynamic simulation of temperature distribution in a ceramic shuttle kiln.

[0014] Preferably, the process of adjusting the parameters of the PID algorithm using a deep belief neural network includes:

[0015] Obtaining kiln operation data during ceramic production;

[0016] The operation data is trained using a deep belief neural network to determine optimal values ​​of PID parameters; wherein the PID parameters include: a proportional parameter, an integral parameter, and a differential parameter.

[0017] Preferably, the process of designing a PID control method based on a fuzzy deep belief neural network includes:

[0018] Design a fuzzy rule base to describe various operations in temperature control of ceramic production;

[0019] The fuzzy rule base is optimized using a deep learning method to achieve target temperature control.

[0020] Preferably, the fuzzy control method is used to deal with uncertainty and nonlinear problems in temperature control in ceramic production; and the deep learning method is used to optimize fuzzy rules and parameters.

[0021] Preferably, the improved Levenberg-Marquardt algorithm comprises:

[0022] Define a cost function to evaluate the difference between the network output temperature and the actual output temperature;

[0023] The cost function is iteratively optimized and the network weights are adjusted to achieve temperature control performance.

[0024] The temperature control method also includes: an adaptive adjustment step for automatically adjusting PID parameters according to real-time temperature data during the ceramic production process.

[0025] Preferably, in the adaptive adjustment step, the adjustment of PID parameters is based on an online learning algorithm to achieve real-time optimization control of the kiln temperature.

[0026] In order to achieve the above technical objectives, the present invention also provides a temperature control system in a ceramic production process, which is used to implement the temperature control method in the ceramic production process, and the system includes:

[0027] Heat transfer model building module, used to build the heat transfer model of ceramic shuttle kiln and fit the optimal heating curve;

[0028] A deep learning module, for adjusting the parameters of the PID algorithm using a deep confidence neural network to ensure that the actual temperature in the ceramic production process follows the optimal heating curve;

[0029] The fuzzy control module is used to combine the fuzzy control method and the deep learning method to design a PID control method based on the fuzzy deep belief neural network;

[0030] Parameter estimation module, used to apply the improved Levenberg-Marquardt algorithm to perform parameter estimation of fuzzy deep belief neural network to optimize PID control method;

[0031] The temperature control module is used to control the temperature in the ceramic production process based on the optimized PID control method.

[0032] Compared with the prior art, the present invention has the following advantages and technical effects:

[0033] The present invention provides a temperature control method in a ceramic production process. Firstly, by establishing a heat transfer model of a ceramic shuttle kiln, the temperature distribution in the kiln can be simulated and predicted more accurately, thereby improving the accuracy of temperature control. Secondly, by using a deep confidence neural network to adjust the parameters of a PID algorithm, the proportional parameter, the integral parameter and the differential parameter can be automatically adjusted to adapt to different firing processes and kiln dynamics, thereby improving the response speed and stability of the control system. Then, by combining a fuzzy control method with deep learning, the uncertainty and nonlinear problems in temperature control can be handled, so that the control system can remain stable and effective in the face of a complex and changeable firing environment. Furthermore, by applying an improved Levenberg-Marquardt algorithm for parameter estimation, the computational complexity and storage space requirements can be reduced, thereby making the control system more efficient. The improved Levenberg-Marquardt algorithm can also reduce the computational complexity and training time of the neural network, thereby accelerating the training and deployment process of the model.

[0034] The present invention can reduce energy consumption and scrap rate in the ceramic firing cycle through precise temperature control, improve production efficiency and consistency and quality of ceramic products. At the same time, it realizes automation and intelligence of temperature control of ceramic shuttle kiln, reduces dependence on manual operation, reduces operation difficulty, and improves operation flexibility and reliability.

[0035] The control method provided by the present invention can adapt to different ceramic materials and firing requirements, and has good adaptability and versatility. In addition, the combination of deep learning model and fuzzy control strategy makes the system easy to maintain and upgrade to adapt to new production requirements and technological development. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0037] Figure 1 The figure is a flow chart of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0038] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0039] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0040] Embodiment 1

[0041] like Figure 1 As shown, this embodiment provides a temperature control method in a ceramic production process, comprising:

[0042] Establish a heat transfer model for ceramic shuttle kiln and fit the optimal heating curve;

[0043] Using a deep belief neural network to adjust the parameters of the PID algorithm to ensure that the actual temperature in the ceramic production process follows the optimal heating curve;

[0044] Combining fuzzy control methods and deep learning methods, a PID control method based on fuzzy deep belief neural network is designed;

[0045] The improved Levenberg-Marquardt algorithm is applied to the parameter estimation of fuzzy deep belief neural network to optimize the PID control method;

[0046] Based on the optimized PID control method, the temperature in the ceramic production process is controlled.

[0047] Furthermore, the heat transfer model includes a dynamic simulation of the temperature distribution in the ceramic shuttle kiln.

[0048] Specifically, by establishing a heat transfer model for a ceramic shuttle kiln and fitting the optimal heating curve, an accurate temperature change basis is provided for PID control. This step is the basis of the entire control process and ensures the accuracy and effectiveness of subsequent control strategies.

[0049] Furthermore, the process of tuning the parameters of the PID algorithm using the deep belief neural network includes:

[0050] Obtaining kiln operation data during ceramic production;

[0051] The operation data is trained using a deep belief neural network to determine optimal values ​​of PID parameters; wherein the PID parameters include: a proportional parameter, an integral parameter, and a differential parameter.

[0052] Specifically, the powerful data characterization capability of the Deep Belief Neural Network (DBN) is used to adjust the parameters of the PID algorithm, including the proportional (Kp), integral (Ki) and differential (Kd) parameters. By learning a large amount of data, the Deep Belief Neural Network can automatically adjust these parameters to adapt to the complexity of temperature changes during the ceramic firing process.

[0053] Furthermore, the process of designing a PID control method based on a fuzzy deep belief neural network includes:

[0054] Design a fuzzy rule base to describe various operations in temperature control of ceramic production;

[0055] The fuzzy rule base is optimized using a deep learning method to achieve target temperature control.

[0056] Furthermore, the fuzzy control method is used to deal with uncertainty and nonlinear problems in temperature control of ceramic production.

[0057] This embodiment, through the combination of fuzzy control method and deep learning method, is conducive to improving the performance of ceramic production temperature control system.

[0058] Further, the improved Levenberg-Marquardt algorithm includes:

[0059] Define a cost function to evaluate the difference between the network output temperature and the actual output temperature;

[0060] The cost function is iteratively optimized and the network weights are adjusted to achieve temperature control performance.

[0061] In this embodiment, by optimizing parameter estimation, the computational complexity and storage space can be reduced, the computational workload and training time of the neural network can be reduced, and the generalization ability of the network can be improved.

[0062] Furthermore, the temperature control method also includes: an adaptive adjustment step for automatically adjusting PID parameters according to real-time temperature data during the ceramic production process.

[0063] Furthermore, in the adaptive adjustment step, the adjustment of PID parameters is based on an online learning algorithm to achieve real-time optimization control of the kiln temperature.

[0064] Specifically, the specific process of the adaptive adjustment step includes:

[0065] S1, real-time monitoring. The system continuously monitors the real-time temperature data of the ceramic kiln, which serves as the basis for the PID controller to adjust its parameters.

[0066] S2, system modeling and parameter estimation. Establish a mathematical model of the system and use online identification techniques, such as the least squares algorithm, to estimate the current state and dynamic characteristics of the system.

[0067] S3, adaptive mechanism. Adaptive adjustment of PID parameters (Kp, Ki, Kd) according to the changes in real-time performance or characteristics. This mechanism ensures that the control system maintains optimal performance even when the system dynamics or external conditions change.

[0068] S4, optimization algorithm. Various optimization algorithms, such as gradient descent, genetic algorithm, particle swarm algorithm, etc., are used to adjust PID parameters. These algorithms search for optimal parameters by calculating error gradients or simulating natural evolution processes.

[0069] S5, controller output. The adaptive PID controller calculates an output signal based on the adjusted parameters, which is used to adjust the input of the heating or cooling system to control the temperature of the kiln.

[0070] S6, performance feedback. The system compares the controller output with the expected performance indicators and uses this as feedback to further adjust the PID parameters, forming a closed-loop control process.

[0071] Adaptive adjustment ensures that the control system can remain stable and achieve the preset performance indicators even when encountering various uncertain factors in the ceramic production process, such as changes in raw materials, changes in the internal environment of the kiln, etc. In this embodiment, through this adaptive adjustment step, the PID controller can achieve precise control of the temperature in the ceramic production process, thereby improving product quality and production efficiency. This method is particularly suitable for ceramic firing processes that have extremely high requirements for temperature control, and can significantly reduce the scrap rate caused by improper temperature control.

[0072] This embodiment integrates the control method of the optimal heating curve, deep confidence neural network, fuzzy control and improved Levenberg-Marquardt algorithm, fully utilizes the advantages of deep learning in data processing and parameter optimization, improves the automation and intelligence level of the ceramic firing process, and further improves the kiln temperature control effect and ceramic product quality.

[0073] Embodiment 2

[0074] This embodiment provides a temperature control system in a ceramic production process, which is used to implement the temperature control method in a ceramic production process described in Embodiment 1. The system includes:

[0075] Heat transfer model building module, used to build the heat transfer model of ceramic shuttle kiln and fit the optimal heating curve;

[0076] A deep learning module, for adjusting the parameters of the PID algorithm using a deep confidence neural network to ensure that the actual temperature in the ceramic production process follows the optimal heating curve;

[0077] The fuzzy control module is used to combine the fuzzy control method and the deep learning method to design a PID control method based on the fuzzy deep belief neural network;

[0078] Parameter estimation module, used to apply the improved Levenberg-Marquardt algorithm to perform parameter estimation of fuzzy deep belief neural network to optimize PID control method;

[0079] The temperature control module is used to control the temperature in the ceramic production process based on the optimized PID control method.

[0080] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A temperature control method in a ceramic production process, characterized in that: The following steps are involved: Establish a heat transfer model for ceramic shuttle kiln and fit the optimal heating curve; Using a deep belief neural network to adjust the parameters of the PID algorithm to ensure that the actual temperature in the ceramic production process follows the optimal heating curve; Combining fuzzy control methods and deep learning methods, a PID control method based on fuzzy deep belief neural network is designed; The improved Levenberg-Marquardt algorithm is applied to the parameter estimation of fuzzy deep belief neural network to optimize the PID control method; Based on the optimized PID control method, the temperature in the ceramic production process is controlled.

2. The temperature control method in the ceramic production process according to claim 1, characterized in that: The heat transfer model includes a dynamic simulation of the temperature distribution in a ceramic shuttle kiln.

3. The temperature control method in the ceramic production process according to claim 1, characterized in that: The process of tuning the parameters of the PID algorithm using a deep belief neural network includes: Obtaining kiln operation data during ceramic production; The operation data is trained using a deep belief neural network to determine optimal values ​​of PID parameters; wherein the PID parameters include: a proportional parameter, an integral parameter, and a differential parameter.

4. The temperature control method in the ceramic production process according to claim 1, characterized in that: The process of designing a PID control method based on a fuzzy deep belief neural network includes: Design a fuzzy rule base to describe various operations in temperature control of ceramic production; The fuzzy rule base is optimized using a deep learning method to achieve target temperature control.

5. The temperature control method in the ceramic production process according to claim 1, characterized in that: The fuzzy control method is used to deal with the uncertainty and nonlinearity problems in temperature control of ceramic production.

6. The temperature control method in the ceramic production process according to claim 1, characterized in that: The improved Levenberg-Marquardt algorithm includes: Define a cost function to evaluate the difference between the network output temperature and the actual output temperature; The cost function is iteratively optimized and the network weights are adjusted to achieve temperature control performance.

7. The temperature control method in the ceramic production process according to claim 1, characterized in that: Also includes: The adaptive adjustment step is used to automatically adjust the PID parameters according to the real-time data of temperature during the ceramic production process.

8. The temperature control method in the ceramic production process according to claim 7, characterized in that: In the adaptive adjustment step, the adjustment of PID parameters is based on an online learning algorithm to achieve real-time optimization control of the kiln temperature.

9. A temperature control system in a ceramic production process, characterized in that: For implementing the temperature control method in a ceramic production process according to any one of claims 1 to 8, the system comprises: Heat transfer model building module, used to build the heat transfer model of ceramic shuttle kiln and fit the optimal heating curve; A deep learning module, for adjusting the parameters of the PID algorithm using a deep confidence neural network to ensure that the actual temperature in the ceramic production process follows the optimal heating curve; The fuzzy control module is used to combine the fuzzy control method and the deep learning method to design a PID control method based on the fuzzy deep belief neural network; Parameter estimation module, used to apply the improved Levenberg-Marquardt algorithm to perform parameter estimation of fuzzy deep belief neural network to optimize PID control method; The temperature control module is used to control the temperature in the ceramic production process based on the optimized PID control method.

Images