Publication

Abstract : The primary objective of this research is to improve the efficiency and stability of Boiler Turbine Systems (BTS) in modern thermal power plants. The study aims to address the challenges posed by the non-linear dynamics of BTS and the need for responsive, efficient energy conversion to meet rising global energy demands while adhering to strict environmental regulations. This research utilizes a dynamic model of the BTS, as originally proposed by Bell and Astrom. To facilitate analysis, the inherently non-linear BTS model is linearized around a chosen operating point using a Taylor series expansion. Following this, an interaction analysis is conducted to understand the relationship between inputs and outputs. This allows the complex Multi Input Multi Output (MIMO) system to be decomposed into three simpler Single Input Single Output (SISO) systems for further study: (i) Fuel flow rate vs drum pressure; (ii) Steam flow to the turbine vs electric power; (iii) feedwater flow vs drum water level. This research utilizes several controllers, such as Proportional-Integral-Derivative (PID), Internal model control (IMC-PID), Linear quadratic regulator (LQR), LQR+PI, and Robust LQR controller, to ensure the efficient performance of the BTS. Due to the nonlinear dynamics of the BTS, controlling the drum pressure, electric power, and drum water level is difficult with traditional control methods. Hence in order to improve the performance and stability in a complex MIMO environment, LQR and Robust LQR techniques are required to enhance transient response, overshoot, and settling time. These controllers are computationally simple and robust, which provide stability and optimize performance measures. A comparative analysis is performed on the proposed controllers for BTS, thus examining their performance and error metrics for drum pressure, electric power and drum water level. The stability analysis of the closed-loop system is conducted, taking into account the performance measures such as rise time, settling time, overshoot, peak response, and error metrics such as Integral Time Absolute Error (ITAE), Integral Absolute Error (IAE), and Integral Square Error (ISE). LQR+PI and Robust LQR control strategies demonstrate superior performance in regulating the system compared to PID, IMC-PID, and LQR controllers. Specifically, LQR and Robust LQR achieve significantly better rise times, with reductions of 98.5 %, 97.77 %, and 96.4 % than PID, IMC-PID and LQR+PI controllers respectively. This rapid response of LQR and Robust LQR leads to considerably faster settling time, outperforming the PID, IMC-PID and LQR+PI controllers by a margin of 98.9 %, 97.11 %, and 89.1 %. Furthermore, LQR and Robust LQR excel in minimizing overshoot, achieving a reduction of 98.6 %, 99.9 %, and 97.7 % than PID, IMC-PID and LQR+PI controller respectively. The superior performance of LQR+PI and Robust LQR highlights their ability to achieve the desired system state rapidly, with minimal oscillations, and enhanced stability. The simulation studies show that the proposed LQR+PI and Robust LQR control scheme effectively handles set point tracking of BTS and meets the required electrical demand.