Optimizing thermal power plant efficiency through washout filter based proportional integral derivative controller.

Abstract

Enhancing the overall efficiency of thermal power plants has gained significant traction in contemporary society. While several methods have been developed for this purpose, a notable gap exists in the examination of how temperature and water flow rate control impacts the efficiency of thermal power generation. The initial step involves the utilization of demineralized water. The process commences by computing the deviation between the set point and the actual value, which is then fed into a WF (Washout Filter) - PID (Proportional Integral Derivative) controller. To fine-tune the parameters of the PID controller, the RL-SGLSTM (Reinforcement Learning based Scaled Gamma Long Short Term Memory) algorithm is employed. The output signals from the PID controller are subsequently employed by a PWM (Pulse Width Modulation) mechanism to produce the necessary voltage for activating water flow valves. Following the determination of the water flow rate, water is supplied to the power system to facilitate the generation of electrical power and the cooling of the condenser. During the power generation process, the generator's frequency disturbances are mitigated by the use of an SVD (Spring Vibration Damper) - BFC (modified Beta parameter-based Fuzzy Controller). Additionally, voltage sags occurring in the transmission lines are circumvented by employing a DV-UPFC (Dynamic Voltage Restorer-based Unified Power Flow Controller device). Finally, the cooling efficiency is assessed through the utilization of the MSC-KMA (Maximum Shifting Correlation Distance-based K-Means Algorithm) to gauge alterations in the power generation of the system. The simulation results conclusively demonstrate that the fine-tuning of PID parameters and the implementation of cooling efficiency-driven water flow control exert a substantial influence on the stability and performance of the thermal power plant system. Based on the observed results, the power generation efficiency of the proposed method attained a higher efficiency of 98 % which is relatively higher than existing methods.