基于模糊神经网络PID算法的电阻炉温度控制系统的研究

发布时间：2018-04-01 16:45

  本文选题：电阻炉　切入点：PID　出处：《河南理工大学》2014年硕士论文

【摘要】：电阻炉是一种重要的热加工设备,广泛地应用于工业生产、科研教学等场合,用作金属加热、元素分析测定和钢件的热处理等操作。电阻炉温度控制效果的优劣直接影响产品的品质和工作效率。随着科学技术的不断发展,对电阻炉温度控制系统提出了更高的要求,即要求电阻炉温度控制系统具有反应迅速、温度控制准确和较高的温控精度等性能。由于电阻炉的温度是一个大惯性、大滞后、时变、且非线性的参数,采用传统PID控制不能解决系统的非线性、时变和PID参数的在线整定难等问题,因此电阻炉温度控制系统当采用PID控制算法时,不能达到较好的控制效果。本文针对电阻炉温度的特性,采用了一种新的控制算法,即模糊神经网络PID算法。可根据电阻炉的温度的偏差及其变化实时对PID的3个参数进行优化,达到具有最佳组合的PID控制,从而实现PID控制的自适应和智能化性能。本文以SX2-4-10高温箱式电阻炉为研究对象,分析电阻炉的工作特性,采用机理分析法对电阻炉温度对象进行分析,从理论上建立电阻炉被控对象的数学模型,并采用理论和实验相结合的求取电阻炉温度的传递函数。使用Matlab的simulink仿真,通过传统PID与模糊神经网络PID阶跃响应曲线的比较,表明系统采用模糊神经网络PID算法具有更好的动、静态特性和自适应性,对突加的外部的扰动具有良好的抗扰动能力。本文以单片机ATmega16为核心控制器,以K型热电偶为温度传感器,设计了系统的硬件系统,完成了温度的检测、控制、报警、显示等功能,温度的控制是通过调压电路来实现的。硬件系统通过串行口与上位机相连。本文设计了系统的软件,包括上位机和下位机软件,上位机主要功能是通过采用图形化编程软件LABVIEW来对炉温进行监控界面设计,下位机则是通过对单片机进行软件编程来实现炉温的控制功能。最后,本文对系统进行了运行检验,最后得到一个较理想的电阻炉温度曲线,证明本文设计的电阻炉温度控制系统能在较短时间内把炉温升到设定值,响应速率快,炉温稳定时间大约在200s左右。实际运行时,超调量和稳定性没有仿真效果那么理想,但系统的最大炉温超调很小,炉温波动保持在目标值±1℃范围以内。
[Abstract]:Resistance furnace is an important hot processing equipment, widely used in industrial production, scientific research teaching and other occasions, used as metal heating, With the development of science and technology, the temperature control system of resistive furnace is required to be higher, and the quality of the temperature control system of resistor furnace is greatly improved with the development of science and technology, such as the operation of element analysis and heat treatment of steel parts, and the effect of temperature control on resistance furnace directly affects the product quality and working efficiency. That is to say, the resistance furnace temperature control system is required to have the characteristics of rapid reaction, accurate temperature control and high temperature control accuracy. Because the temperature of resistance furnace is a large inertia, large lag, time-varying and nonlinear parameters, Traditional PID control can not solve the problems of nonlinear, time-varying and on-line tuning of PID parameters. Therefore, when the temperature control system of resistor furnace adopts PID control algorithm, In this paper, a new control algorithm is used to control the temperature of resistance furnace. That is, fuzzy neural network PID algorithm. According to the temperature deviation of resistor furnace and its variation, the three parameters of PID can be optimized in real time, and the PID control with the best combination can be achieved. In order to realize the adaptive and intelligent performance of PID control, this paper takes the SX2-4-10 high-temperature box resistor furnace as the research object, analyzes the working characteristics of the resistor furnace, and analyzes the temperature object of the resistor furnace by using the mechanism analysis method. The mathematical model of the controlled object of resistor furnace is established theoretically, and the transfer function of resistance furnace temperature is obtained by combining theory and experiment. The simulink simulation of Matlab is used to compare the step response curve of traditional PID with that of PID based on fuzzy neural network. It shows that the fuzzy neural network PID algorithm has better dynamic, static and adaptive characteristics, and has a good anti-disturbance capability to the sudden external disturbance. The core controller of this system is single chip microcomputer (ATmega16). Taking K type thermocouple as temperature sensor, the hardware system of the system is designed, and the functions of temperature detection, control, alarm, display and so on are completed. The temperature control is realized by voltage regulation circuit. The hardware system is connected to the host computer by serial port. The software of the system is designed in this paper, including the upper computer and the lower computer software. The main function of the upper computer is to design the furnace temperature monitoring interface by using the graphical programming software LABVIEW, and the lower computer to realize the control function of the furnace temperature by the software programming of the single chip microcomputer. Finally, an ideal resistance furnace temperature curve is obtained, which proves that the temperature control system designed in this paper can raise the furnace temperature to a set value in a short time, the response rate is fast, and the furnace temperature stability time is about 200 s. The overshoot and stability are not as ideal as the simulation results, but the maximum furnace temperature overshoot of the system is very small, and the furnace temperature fluctuation is kept in the range of 卤1 鈩，

本文编号：1696409