Vaishnavi P.

For pressure control in industries, an intelligent network controllers like neural network predictive controller, Fuzzy logic controller has been developed and compared with ZN tuning PID controller. The errors of ISE, IAE and IATE are used for the primary comparison of the results acquired in these controllers.

The additive and multiplicative faults in sensor and actuator used in the quadruple tank system has been detected using Kalman filter based estimation technique. The residual signal is used as a performance metric and it is found as the difference between the plant level and level that obtained from the model. The zero residual signal indicates no fault in the system and the deviation of residual signal from zero indicates that fault occurs in the system components.

An evaporator is a nonlinear Multi Input and Multi Output (MIMO) system. To achieve the required concentration and feed flow of the product, the states of the system must be known. The Unscented Kalman Filter (UKF) algorithm is applied to estimate both states in the existence of random noise in measurements. The estimation accuracy of this method is analysed by comparing the Mean Square Error (MSE) of UKF with EKF.

Unstable system is difficult to control when compared with stable system. The cascade control method is the successful and best method for controlling the unstable system than the single feedback control. In this work we suggest the simple method of tuning the PID controller using Genetic Algorithm (GA) for cascade unstable first order plus time delay process. The performance of the above mentioned algorithm has been evaluated by comparing the Integral Time Absolute Error (ITAE).