Introduction to z-transforms and Development of Grey-box models
5
Introduction to Stability Analysis and Development of Output Error Models
6
Introduction to Stochastic Processes
7
Introduction to Stochastic Processes (Contd.)
8
Development of ARX models
9
Statistical Properties of ARX models and Development of ARMAX models
10
Development of ARMAX models (contd.) and Issues in Model Development
11
Model Structure Selection and Issues in Model Development (contd.)
12
Issues in Model Development (contd.) and State Realizations of Transfer Function Models
13
Stability Analysis of Discrete Time Systems
14
Lyapunov Functions and Interaction Analysis and Multi-loop Control
15
Interaction Analysis and Multi-loop Control (contd.)
16
Multivariable Decoupling Control and Soft Sensing and State Estimation
17
Development of Luenberger Observer
18
Development of Luenberger Observer (contd.) and Introduction to Kalman Filtering
19
Kalman Filtering
20
Kalman Filtering (contd.)
21
Kalman Filtering (contd.)
22
Pole Placement State Feedback Control Design and Introduction to Linear Quadratic Gaussian Control
23
Linear Quadratic Gaussian (LQG) Regulator Design
24
Linear Quadratic Gaussian (LQG) Controller Design
25
Model Predictive Control (MPC)
26
Model Predictive Control (contd.)
Description:
Instructor: Prof. Sachin Patwardhan, Department of Chemical Engineering, IIT Bombay.
This course has been designed to introduce concepts of multivariable state feedback controller synthesis using discrete time state space models. Development of control relevant dynamic models is viewed as an integral part of the process of controller synthesis. Thus, the course begins with the development of continuous time and discrete time linear perturbation models (state space and transfer functions) starting from mechanistic models commonly used in engineering. However, in practice, a mechanistic dynamic model may not be available for a system. In such a situation, control relevant discrete dynamic black-box models can be developed using perturbation test data. Development of output error, ARX and ARMAX models from time series data and constructing state realizations of the identified models is dealt next.