Abstract: Traditional control methods struggle to meet the control requirements of temperature control systems in industrial glass tempering furnaces, which exhibit nonlinearity, time-varying time delays, severe inter-variable correlations, complex disturbances, and difficulty in modeling the controlled object. For such complex control systems, fuzzy control algorithms can overcome the shortcomings and deficiencies of traditional control methods, demonstrating superior performance. Simulation results using MATLAB show that fuzzy control algorithms can effectively adapt to uncertain systems with nonlinear and time-varying characteristics, exhibiting excellent control performance for these systems.
Keywords: Temperature control of glass tempering furnace; Fuzzy control algorithm; Nonlinearity; Time-varying parameters; Simulink simulation
0 Introduction
Tempered glass, as a safety glass, has the advantages of being impact-resistant, bending-resistant, having good thermal stability, and breaking into small pieces when broken, which is less likely to cause harm to the human body[1]. It is widely used in high-rise buildings, glass railings, train windows and other places where glass safety and quality requirements are high.
The tempering furnace is the most important piece of equipment for tempering glass. The cooling stage in the tempering process is a crucial step, and controlling the cooling temperature directly affects the quality of the tempered glass; improper control can even lead to glass breakage. Therefore, the cooling temperature is a critical control parameter during the tempering process.
1. Research Background
The furnace temperature regulation during the tempering furnace cooling stage adopts air cooling technology, that is, cooling the tempered glass by sending cooling air through a fan. Due to the many uncertainties in the cooling stage, such as the temperature of the cooling air changing with the outside world and the glass thickness changing with different stages of tempering, the temperature control during the cooling stage mainly adopts a combination of manual operation and PID. Although manual adjustment can control the temperature within a reasonable range, it is greatly affected by human factors and it is difficult to achieve high-standard control. Although PID control is simple and easy to implement, it is highly dependent on the model of the controlled object and can only meet the control requirements when the parameters of the system remain basically unchanged [2]. However, the actual system is an uncertain system with parameter following and parameter coupling, so it is difficult for PID to achieve precise control. Fuzzy control is a branch of intelligent control. Due to its insensitivity to changes in the parameters and structure of the controlled object, the lack of need to establish a mathematical model of the controlled object, and good robustness, it plays a very important role in the control field. It uses the operator's experience, knowledge, or operational data to control the controlled object, and has the advantages of fast dynamic response and small overshoot. It has a particularly good control advantage for complex systems such as dynamic time-varying systems, nonlinear systems, and large time-delay systems.
2. Fuzzy Control System
Fuzzy control is an intelligent control method based on fuzzy set theory, fuzzy linguistic variables and fuzzy logic reasoning[3]. It continues the traditional control idea and consists of the controlled object, fuzzy controller, actuator and feedback loop[4].
2.1 Fuzzy Controller
The fuzzy controller is the core of the fuzzy control system. It replaces the position of the controller in the traditional control system. The fuzzy control quantity u is obtained by fuzzification of the input quantity, fuzzy inference, and defuzzification of the output. According to the number of input variables, the fuzzy controller can be divided into one-dimensional fuzzy controller, two-dimensional fuzzy controller and multi-dimensional fuzzy controller [5]. In engineering, a two-dimensional fuzzy controller structure with the deviation e and the rate of change of deviation ec as the input of the fuzzy controller is generally adopted.
Figure 2. Structure of the two-dimensional fuzzy controller
The relevant operational experience can be summarized as follows:
Based on the above operational experience and the division of fuzzy subsets, the control rules adopt the fuzzy form of If e is NB and ec is NB, then u is PB-Mamdani, i.e., segment bc, which yields 49 control rules:
2.3 Setting of membership function and defuzzification
In engineering applications, the most commonly used membership functions are triangular, trapezoidal, bell-shaped, Gaussian, S-shaped, and Z-shaped. For fuzzy subsets in the middle of the universe of discourse, symmetrical membership functions are often chosen, while membership functions for the left and right boundaries are often chosen with slowly changing boundaries. In the design, triangular functions are used for the membership functions of both the input and output fuzzy subsets.
3. MATLAB Simulation of Fuzzy Control Systems
The glass tempering furnace is a relatively complex industrial control object. During the cooling process, the temperature is regulated by the introduction of cooling air, which is affected by factors such as time lag, nonlinearity, inertia, and time-varying characteristics. Therefore, it is difficult to establish an accurate mathematical model for it; a simplified mathematical model is generally used.
The simulation of the system mainly involves observing the changes in the input signal before and after passing through the system in order to understand the control performance of the control system. A step signal is used as the system input, the inference algorithm adopts the system's default Max-Min synthesis method, the defuzzification method adopts the area center method, and the simulation adopts a fixed step size.
Simulation results show that the fuzzy control algorithm with reasonable parameter settings not only has a good dynamic response curve, but also has the advantages of small overshoot, fast response speed and high steady-state accuracy. In addition, it can overcome the influence of noise disturbance when encountering disturbances, and has high control quality and good robustness.
4. Conclusion
For complex systems with nonlinear and time-delay characteristics, fuzzy control has the advantages of small overshoot, fast response, and high steady-state accuracy. It also has good robustness and adaptability to disturbances and can maintain high control quality, thus having significant theoretical and practical application value.
References
[1] Chen, Lisong. Research on horizontal roller conveyor glass tempering furnace and its temperature control system [J]. Computer and Modernization, 2011(6):112-116
[2] Yu Xiaocao, Chen Jian, Ding Zhuyu, et al. Fuzzy controller for temperature of pepper drying room based on MATLAB [J]. Agricultural Mechanization Research, 2009, 10: 184-186.
[3] Liu Jinkun. Intelligent Control [M]. Beijing: Electronic Industry Press, 2005.
[4] Hu Shousong. Principles of Automatic Control [M]. Beijing: Tsinghua University Press, 2008.
[5] Sun Gengshan, Lan Xizhu. Engineering Fuzzy Control [M]. Beijing: Machinery Industry Press, 1995.7.
[4] Song Rui, Lin Qingjia, et al. Simulation of fuzzy control system using MATLAB SIMULINK [J]. Shandong Metallurgy, 2003, 25(3):54-56.
[5] Shi Xinmin, Hao Zhengqing. Fuzzy Control and Its MATLAB Simulation [M]. Beijing: Tsinghua University Press; Beijing Jiaotong University Press, 2008.2
[6] Ma Yong, Li Gang. Research on temperature control system based on fuzzy technology [J]. Journal of Northwest A&F University, 2005, 33(10):120-124
[7] Xing Chungui, He Xiwen. MATLAB simulation and application of fuzzy control systems [J]. Microcomputer Processing, 2004, 1: 34-36
[8] Wang Zhenglin, Guo Yangkuan. MATLAB/Simulink and Process Control System Simulation [M]. Beijing: Electronic Industry Press, 2004.9
[9] Liu, Jinkun. Advanced PID Control Simulation in MATLAB [M]. Beijing: Electronic Industry Press, 2012.9
[10] Guo Guanzhou. Simulation and application of furnace temperature control algorithm based on fuzzy controller [J]. Automation Instrumentation, 2011, 32(5):42-45
[11] Yang Shiyong, Xu Guolin. Comparison of fuzzy control and PID control and their composite control [J]. Automation Technology and Application, 2011, 30(11):21-25
[12] Deng Wei, Zhang Baoping. Design and Matlab Simulation of Fuzzy Temperature Controller [J]. Journal of Zhengzhou University of Light Industry, 2009, 24(2):50-52
