Thursday 1 February 2018

ADAPTIVE CONTROL - AN INTRODUCTION

Adaptive Control
Adaptive control detects the changes in the characteristics of the process and adjusts the controller parameters automatically to compensate for the changing conditions of the process and in turn to optimize the loop response.

Block Diagram of an Adaptive Control System
Block Diagram of an Adaptive Control System
An adaptive control system consists of a normal feedback loop and a parameter adjustment loop. The normal loop includes the controller and the process. The parameter adjustment loop consists of the parameter adjustment mechanism and a controller with adjustable parameters. The parameter adjustment loop is usually slower than the normal loop. The different types of adaptive control differ only by the mechanism it used for parameter adjustment.
The adaptive control differs from all other control schemes. The adaptive control adjust the controller parameters whereas in other strategies a process variable is adjusted.

Need for Adaptive Control
1. The key reason is that most of the processes are nonlinear. The control loops are generally designed to maintain the controlled variable at its setpoint by compensating for all disturbance occurring in the process. The controller performance is optimum only for a particular range in which the process is linearized. Once the process starts to operate beyond the linearized range, the controller fails to produce desired performance. It is because of the fact that the parameters of the controller is not suitable for the current operating conditions.
2. The changes in transfer function of process which occurs due to parameter variations or variation in the coefficients or wear and tear of important components.
3. The nature and magnitude of disturbances vary with time. There may be an occurrence of an unpredictable and unknown disturbance in the process.
4.There may be a change in nature of inputs to the process and the properties of raw materials.
In all the above cases, a conventional controller cannot perform at a satisfactory level. This demands the need for a special type of controller that adapt in accordance with the uncertainties in the process and in turn Adaptive Control

Types of Adaptive Control
There are two major approaches to determine the controller parameters adaptation. They are,
1. Programmed or Gain Scheduled Adaptive Control
2. Self-Adaptive Control
The Programmed Adaptive Control is compared to feedforward compensation because it adjust the controller parameters based on the measurement of an auxiliary variable and the knowledge of operating conditions of process. Also, there is no feedback to check the correctness of adaptation.
The Self-Adaptive Control is comparable to feedback compensation because the adaptation of controller parameters is based on the measurement of closed loop performance and aim to optimize it.  Model Reference Adaptive Control (MRAC) and Self Tuning Regulator (STR) are two mechanisms that come under self-adaptive control.

Applications of Adaptive Control
1. Adaptive control is used in the robotic manipulators of robotic systems which demands high positioning accuracy.
2. Adaptive control is used for altitude control of satellites. The observation satellites should be operated at lower altitudes where the air-drag makes the quick reorientation of satellite is necessary to increase the observation time.
3. Adaptive control is used in the autopilot of air crafts and steering control of ships.
4. Adaptive control is used in the control of strip temperature for the continuous annealing and processing in metallurgical processes.
5. Adaptive control is used in distillation columns to provide high product quality and a considerable reduction of thermal energy usage. 
6. Adaptive control is used to stabilize PID based pH control system in chemical industries. Without adaptive control, the process gain (Kp) increases as the pH value becomes neutral and it leads to change in total loop gain and finally to instability of loop. The adaptive control keeps the total loop gain at the desired value (usually 0.5) by lowering the controller gain (Kc).



NEXT TOPIC: Programmed or Gain Scheduled Adaptive Control

Also read other control schemes









No comments:

Post a Comment

Share your reading experience here

Popular Posts

Tell us your feedback about INSTRUMENTATION IN A NUTSHELL

Name

Email *

Message *