Temperature control of extrusion machine

In this postgraduate thesis, a practical study on the temperature control of the extrusion machine, which is a the frequently used thermal system in the industrial automation sector, is presented. For this purpose, a relevant system in the laboratory environment is controlled using PWM signals over a solid state relay via Siemens Simatic S7-1500 PLC. Industrial automation is composed of many sub-branch processes and most of these include a thermal system. For example, there are many vessels in an oil factory and these vessels need a certain temperature to process the relevant material. In order to obtain this temperature, resistances, burners and other equipments are used. These thermal processes are modeled using various procedures in control theory. However, first order transfer functions with time delay are the most preferred models in the literature because of their simple structures and well-known characteristics.The design stage of the controller is also easier with first order transfer functions with time delay. Whether a first or higher order the model is chosen, the most preferred controller is the PID controller in all industry. Besides, the most commonly used device to control the processes is PLC because of its reliability, robustness and ease of use. Micro controllers rank number two in this area. Extrusion machine is a thermal device mostly used for plastic or metal forming in industry. It works like an ordinary heater and starts to warm up by passing the electric current through the resistors in it. Although it differs according to the extrusion machine used, the system can warms up to about 800 ° C in approximately a 1 hour period. The extrusion machine used in the thesis have 6 resistances and solid state relay is used to control voltage. The system can be hold at desired temperatures with PWM signals. Siemens Simatic S7-1500 PLC is used to supply PWM signals to the system. Siemens Simatic S7-1500 PLC is used to supply PWM signals to the system. This PLC is the new controller family of Siemens and has many useful properties to control any systems. In this thesis project, trend view, auto tuning of PID parameters and some other features are used. Apart from extrusion machine and PLC information, connection diagrams and some photographs of these two devices are given in detail and it is aimed to have an idea about the extrusion machine system. Firstly, we tried to find the mathematical model of the extrusion machine. For this purpose, PWM signals ranging from %5 to %20 are applied to the system at temperatures beginning from room temperature to 170°C. Twenty-five different models are obtained for the heating and cooling phases of the machine. Later, for each heating model, a PI controller is tried to be designed using the "PID Tuner" tool in MATLAB. The parameters of the controller are tested on the real system and fine-tuning is made for the optimum solution. A direct design could not be made due to the system limits for cooling models, instead the relevant heating controllers were utilized and experimental parameters were obtained around the operating point. Moreover, as an alternative design, a PID auto tuning software is used for the control of the heating and cooling phases of the extrusion machine. For this purpose, Siemens TIA Portal which is the programming software for Siemens PLC's is utilized. This auto tuning software has the ability to design two degrees of PI and PID controllers according to user decision. PI and PID controller parameters are obtained for certain temperatures and the results are recorded. According to the results obtained, PI controller responses are found to be better than the responses obtained using PID with respect to some time-domain criteria. It is observed that the PID controller causes higher overshoot and longer settling time when compared with its PI counterpart. After designing and testing the controllers obtained using the mathematical model and auto tuning software, all the controllers have been integrated into the system and for this purpose, a function block is written in SCL which is similar to ST (Structured Text) defined in the standard IEC 61131-3.The function block has two important task in the process. The first task is the automatic entry of the controller coefficients into the PID block when the desired temperature value changed. Another task of the block is to apply a softening operation to control output during the parameter change to avoid pulse responses. A lead-lag compensator in TIA Portal library is used for this operation. Necessary error information and some other features are also added to function block. A SCADA project that communicates with the PLC is also created in the thesis. For this purpose, "WinCC Comfort Advanced" which is a monitoring and controlling application of Siemens for the automation projects is installed to the computer and necessary adjustments are made. "WinCC Comfort Advanced" enables to monitor the set point, process value and PID parameters used in the thesis project. Besides, a trend screen which shows the change of set point, process value and controller output for two minutes is presented below the page. Finally, a short heating and cooling scenario is considered and the controllers obtained using the auto tuning software and the mathematical model are compared with each other. It is observed that both PI controllers obtained using different techniques present sufficient results for the control of the extrusion machine. However, the PI controllers obtained using auto tuning software and mathematical model have shorter settling time and lower overshoot compared with the PID controllers obtained. As it is well known, PID controllers in thermal processes cannot give good results because of high overshoots and settling times. The most important benefit of the controllers designed by using the mathematical model from the controllers obtained using the auto tuning software is that these controllers provide a flexible choice since the user has a sufficient knowledge about the system. In addition, auto tuning software in industry lead to dangerous situation sometimes because of the oscillation of the system at operating point. Beside this, the relevant system must be observed continuously since the function block of PID and auto tuning software do not work properly seldom. The thesis is divided into 5 sections. In the first section, we focus on the thermal processes, their mathematical models, the preferred control methods and the devices that can be implemented. In the second section, the extrusion machine and its working principles are introduced. The mathematical model derivation and relevant controller design are explained in the third section. In the fourth section, the controllers obtained are compared with each other on a heating and cooling scenario. Conclusion takes place in the fifth section.