The Thing About Control
The Thing About Control
Volume 8 Issue 5 News & Resources | May 2015
The urge to control the environment around us is very strong that we will feel very uneasy when placed in an unknown environment. This urge may be due to the fight or flight instinct or it is just an innate urge to control others. We do it to all inanimate object and living beings. We are the inadvertent and willing master of the universe. Well, if we can arrange things according to our whims; we would like very much to be king, and make others subservient to us. Alas, the world is not totally cooperative to this desire. We are inherently weak though we have the power of intellect which allows human to have a sense of control to the world. Through this natural equilibrium, human beings have started to appreciate his rightful position in the real world, and have grown to adapt his needs to the actual thing that he can acquire. He is now more pragmatic from the control point of view. We may not be able to acquire everything, but surely we can acquire something, and, vice versa. From here, the philosophy is; we may not be able to control everything, but we can surely control some of them. Hence, the main aim is to be able to control the most important parameters. In this, the control urge can somewhat be realized and fulfilled.
Any system can be represented into a control system block. Unless, it is an open loop system, there will always be a feedback loop which will link the output stage to the desired input stage. A system or plant will be one major block in either open or closed-loop control system representation. The parameters of the plant can be either fixed or dynamic, i.e. varies with time. This time variant or invariant system will be crucial in understanding the behavior of the system to input response. The plant can then be represented in a mathematical model, either in the transfer function mode or state-space representations. The number of input and output will also affect the way the particular system is analyzed (SISO, MIMO, SIMO & MISO). A sensor block in the feedback loop is represented by an array of internal and external sensor modules. The accuracy and sensitivity of the sensors are critical in ensuring the system time response is acceptable. This time factor is also termed as sampling time. Another important block is the controller block which will handle the error (difference between desired and output values) and ensuring that the input to the system will bring the output nearer to the desired value or so that they will converge. Various control approaches have been proposed and will continue to be developed in order to handle external disturbance and non-linearity components of the plant. Most practical system is inherently non-linear in and the disturbances are mostly non-deterministic. The accuracy of the plant and variable input models will be pertinent to the reliability of the chosen control approach.
Control designer is always obsessed about accurate modeling of the real system being controlled. They strive very hard to understand all the constraint and limitations of the current system model. The more information they have, the more controllable is the system. Real appreciation of the external disturbances is also important and many methods have been proposed to ensure desired performances are achieved. The important aim is to arrive at the best control law that will reduce the error to zero. It may be very taxing and time consuming to design, analysis and test the control approach on real or actual system as the actual system complexity increases. Hence, mathematical and simulation approaches in controller design will be crucial and cost-effective. Because of this dependence on mathematical representations or model of the system, the methods for extracting model must be considered carefully. Various methods to implement the developed control law, i.e. by injecting input triggers and disturbance are conducted. These are done in order to test the controller robustness and frequency response. At the same time, new control theories are continuously being developed while the more implementable control methods are realized. Control laws in living organisms are continuously being emulated to man-made system. Some hybrid control methods have also been investigated. Three main aims of control system are for stability, tracking and regulating applications. These aims cover most of the man-made applications. And, as long as new machines and system are invented, new and more innovative control methods will always be developed.
The progress of the modern civilization has boosted the need for newer and more reliable control algorithms. As the system complexity increases and more unique application are invented, unique and more cost-effective controllers are required. A complex control system normally requires very complex hardware implementations and such generalization is not desirable since the set-up will not be sustainable. The challenge is to be able to utilize simple control laws for complex control applications, and with simple hardware implementations. This is an utopian aim, but one which will boost and encourage more research work in control theories and control algorithms. Probably a fundamental shift of how we look at control methods is required. Our approach into novel control design is strictly bounded by the accepted control theories and its accepted constraints. Hopefully one day these unquestioned boundaries will be reconsidered and new control laws will emerge. And, one thing for sure, the variances in human needs are unlimited, and so does the control requirements.
Written by :
Assc. Prof. Dr. Mohd Rizal Arshad
School of Electrical & Electronic Engineering