Design and application of controller in continuous casting mold liquid level control system at Baotou Steel
mold liquid level control is a very important link in metallurgical continuous casting system. Because the mold liquid level system has time-varying and nonlinear characteristics, and there are many uncertain disturbance factors, it is impossible to establish an accurate model, and the conventional control method can not meet the requirements, This paper introduces the design and application of Fuzzy-PID Compound Adaptive Control System in the mold level control of Baotou steel continuous casting system. The design and application of Fuzzy-PID in the crystallizer level control system of Baotou Steel Continuous Casting Project 1.1 overview of control system crystallizer level control is a very important link in continuous casting production, which has a decisive impact on the quality of billets, the smooth progress of casting, and the reduction of workers' labor intensity. The block diagram of the control system is shown in Figure 1-1. This system uses stopper to control the flow of molten steel into the crystallizer. The characteristic relationship between stopper position (opening) and molten steel flow is an important basis for designing mold level control. During the casting process, the stopper will be washed and corroded by the molten steel, causing the head of the stopper to deform, resulting in changes in the characteristic relationship between the stopper position and the molten steel flow, as shown in Figure 1-2. In the figure, curve 1 is the characteristic relationship under normal conditions; Curve 2 is the characteristic relationship after the plug head is scoured. It can be seen from the figure that the characteristic curve becomes steeper with the scouring of the stopper head. During continuous casting, the casting time is very long, and the plug head is seriously scoured, that is, with the extension of casting time, the characteristic relationship between the plug position and molten steel flow will change greatly, and the change of this characteristic relationship cannot be described by a certain mathematical relationship. In this case, it is difficult to achieve accurate control of liquid level only by using conventional control 8. Power supply voltage: ~ 220V ± 10%. This system adopts the compound adaptive control system of fuzzy control PID control to solve this problem. 1.2 hardware configuration among the professional service manufacturers of Jinan testing machine factory of this system, the PLC adopts Siemens s5-135u and the CPU is 928B. In order to improve the response speed to the change of liquid level system, double CPUs are used in PLC. One CPU is used for main system control, and the other CPU is specially used for crystallizer level control. 928B is a high-performance CPU, and its response time can reach ms level. 1.3 software design 1.3.1 mould liquid level control principle the control principle is shown in Figure 1-3: when the automatic pouring is started or the liquid level is lower than 20%, the change-over switch (software) k is switched to the "1" position, and the amplification factor KP is 1 * k (k is the adjustment factor of KP). At this time, the fuzzy controller does not work. During normal control, K is switched to "2", that is, the output of the fuzzy controller is used as the amplification factor KP of the PID controller. This design is to quickly increase the opening of the water port when the liquid level is very low, so that the liquid level can quickly reach the set liquid level value. When the actual liquid level exceeds 40%, switch K is switched to "2" position to enter normal control. 1.3.2 design of fuzzy controller 1 (fuzzy1). The input variable of fuzzy1 is the sum of the deviation of liquid level and the change of casting speed. It is a fuzzy controller with one-dimensional input and one-dimensional output. 220.127.116.11 fuzziness of input variables set the domain of input variables as [- 10000, + 10000]. On this domain, set five fuzzy subsets, which are Nb (negative large), NS (negative small), Z (zero recovery of the international market), PS (positive small), Pb (positive large). Its shape distribution is shown in Figure 1-4. 18.104.22.168 design of output variables set the domain of output variables as . Three fuzzy subsets are set on this universe. The three fuzzy subsets are s (small), m (medium), and B (large). Its shape distribution is shown in Figure 1-5. 22.214.171.124 control rules 126.96.36.199 control decision according to the control rules, the center of gravity method in fuzzy control decision is used to obtain the control decision curve, as shown in Figure 1-6.