Design of Dual Fuzzy Temperature Controller Based on AT89S52 Single Chip Microcomputer

In modern industrial control, temperature control is very important and increasingly complex. Due to the characteristics of temperature control such as nonlinearity, large lag, time-varying, and unidirectional temperature rise, it is difficult to establish an accurate mathematical model in practical applications, and it is impossible to use classical control theory and modern control theory to solve the temperature control effect. At present, fuzzy control in modern intelligent control that does not rely on the mathematical model of the object and can effectively control time-varying and nonlinear systems has been widely used in industrial temperature control. Through basic fuzzy control of temperature, a good control effect can be achieved, but there is a certain steady-state error, and it is difficult to achieve a high control accuracy. If different temperature fuzzy controls are used according to different working states of the system, that is, dual fuzzy control, the steady-state error can be greatly improved and the control accuracy can be improved. At the same time, using a single-chip microcomputer as the main control chip can effectively complete temperature fuzzy processing and real-time temperature control with high reliability.

1 Controller Function and Hardware Design

The dual fuzzy temperature controller mainly uses a single-chip microcomputer as the main control chip, which is mainly responsible for the fuzzy processing of temperature, the implementation of fuzzy control algorithm, and the size control processing of output temperature. The industrial field temperature is collected by a temperature sensor, and the sensor output signal is converted into a standard signal of 0-5 V by a transmitter. After A/D conversion, it is compared with the temperature setting value to obtain the temperature error signal e and the temperature change rate ec. In the initial stage and steady state stage of the system, both are sent to different single fuzzy controllers for fuzzy processing to obtain the output control quantity u. After isolation and amplification, the power thyristor is controlled to change the power of the heating element, thereby completing the temperature adjustment. The principle block diagram of its temperature control system is shown in Figure 1.

The main control chip of the temperature controller uses the AT89S52 microcontroller from ATMEL, which is a low-power, high-performance CMOS 8-bit microcontroller with 8 kB in-system programmable Flash memory and 256B RAM, 32-bit I/O lines, 3 16-bit timers, 6 interrupt sources and a watchdog timer, etc., which can meet the basic design and extended design requirements of the temperature controller.

The temperature sensor uses nickel-chromium/nickel-silicon thermocouples, which have the characteristics of good linearity, large thermoelectric potential, high sensitivity, good stability and uniformity. Its operating temperature is 0~1300℃, and the corresponding output is 0~52.37 mV[1]. The temperature transmitter uses DBW type to convert the millivolt signal output by the thermocouple into a 0~5 V standard analog signal. The signal is input through the IN0 channel of the A/D converter ADC0809 and converted into a digital signal. The AT89S52 microcontroller uses the control program to input its internal RAM unit from the P0 port to compare with the temperature setting value[2]. The fuzzy controller outputs the control quantity through P1.0 output, and drives the power thyristor through the photoelectric bidirectional thyristor driver MOC3051, thereby changing the heating power of the heating element to achieve the purpose of temperature regulation.

Considering the human-computer interaction of the system, the parallel interface chip 8155 is used to expand the I/O port to complete the design of keyboard input and display output. The keyboard adopts a matrix keyboard, which is responsible for parameter setting and some switch input, such as: start, stop, reset, temperature setting, set value modification, temperature numeric keys, etc.; and the display adopts an LED display to display the system set temperature and actual temperature at the same time.

2 Dual fuzzy control algorithm design

2.1 Dual fuzzy controller structure design

This fuzzy controller adopts a dual fuzzy control structure and a typical dual-input, single-output method, as shown in Figure 2. The error e and the rate of change ec between the temperature set value and the temperature feedback value are used as input quantities, and the temperature control quantity u is used as the output quantity. Since the system has errors of different sizes under different control states, if a single fuzzy controller is considered, there will be a contradiction between the system's rapid response and control accuracy [3], and the two cannot be taken into account. For this reason, a dual fuzzy controller is designed, and an error critical value is artificially set to complete the dual-mode control switching [4]. In the initial stage of the system, the system error is large, and a fuzzy controller with relatively small system factors Kec and Ku (such as Kec1, Ku1) is used to achieve rapid response and eliminate the purpose of error; in the steady-state stage of the system, the system error is small, and a fuzzy controller with appropriately increased system factors Kec and Ku (such as Kec2, Ku2) is used to improve the steady-state performance of the system.

[page]

2.2 Dual fuzzy control strategy

Considering the characteristics of temperature control, the domain of error e, change rate ec and output u is set to [-6, 6], which is quantified into 13 levels, and 7 language values ​​are selected for error e, change rate ec and output u, namely {NL, NM, NS, ZO, PS, PM, PL}. The membership functions of the three all adopt trapezoidal distribution [5], as shown in Figure 3. Based on the experience summary of industrial process control, the corresponding fuzzy control rule table is formulated as shown in Table 1.

In order to improve the real-time response speed of the system, the fuzzy control master table is calculated offline in advance according to the fuzzy control rule table and the language variable assignment table as shown in Table 2. The table is stored in the program memory of the single-chip computer after strict practical testing and repeated modifications. Then, according to the actual change range and domain of the input quantities e and ec in different working states, the quantization factors Ke1, Kec1 and Ke2, Kec2 are calculated, and the proportional factors Ku1 and Ku2 are determined. In actual control, the fuzzy controller multiplies the input quantities e and ec of the system in different working states by the corresponding Ke and Kec respectively, and quantizes them into the language variable domain of the input quantity. Then, according to the quantization result, it is compared with the fuzzy control master table, and the required output quantity U is obtained through the table lookup program. Finally, it is multiplied by Ku to obtain the actual output control quantity u of the system in different working states.

3 Controller Software Design

The software of the dual fuzzy temperature controller adopts a modular design concept, which mainly includes the main program, temperature acquisition program, keyboard/display control program, fuzzy control algorithm program, etc. The following mainly gives the flow charts of the main program and fuzzy control algorithm program, as shown in Figure 4 and Figure 5 respectively [6].

4 Conclusion

The dual fuzzy temperature controller with single chip microcomputer as the control core proposed in this paper is based on the proportional factor self-tuning fuzzy control theory. It uses two fuzzy controllers with different system factors to control the temperature according to the size of the system error. It has a simple structure, is easy to implement, has strong adaptability, can greatly improve the steady-state accuracy, can achieve good control effect for temperature control, and can be widely used in industrial production.