Dataset for: Artificial Intelligence-Based Online Control Scheme for the Regulations of Interconnected Thermal Power Systems

Published: 22 August 2023| Version 2 | DOI: 10.17632/4x696z6xn4.2
Contributors:
,
,
,

Description

This dataset is an extension to our article titled " Artificial Intelligence-Based Online Control Scheme for the Regulations of Interconnected Thermal Power Systems." To cite the paper, use: @article{orka2023artificial, title={Artificial Intelligence-Based Online Control Scheme for the Regulations of Interconnected Thermal Power Systems}, author={Orka, Nabil Anan and Muhaimin, Sheikh Samit and Shahi, Md Nazmush Shakib and Ahmed, Ashik}, journal={Arabian Journal for Science and Engineering}, pages={1--24}, year={2023}, publisher={Springer} } The dataset comprises two CSV files: "linear_thermalthermal" contains optimal PID parameters corresponding to load disturbances in both regions of a two-area interconnected power system (IPS) with non-reheat thermal turbines. "nonlinear_hydrothermal" comprises optimum PID gains corresponding to load perturbations in both regions of a two-area hydrothermal IPS that includes nonlinearities like Generation Rate Constraint (GRC) and Governor Dead Band (GDB). The simulation and collection of data was conducted in MATLAB 2018b. To obtain the optimal controller parameters, Newton's method inspired Gradient-Based Optimizer (GBO), GBO’s updated algorithm Enhanced Gradient-Based Optimizer (EGBO), and nature-based Marine Predators Algorithm (MPA) are employed using an Integral Time-multiplied Absolute Error (ITAE) based objective function. A brief description of different features of the dataset is described below: Kp1: Proportional gain of the controller in area 1 Ki1: Integral gain of the controller in area 1 Kd1: Derivative gain of the controller in area 1 Kp2: Proportional gain of the controller in area 2 Ki2: Integral gain of the controller in area 2 Kd2: Derivative gain of the controller in area 2 w1: Step-change in load demands of area 1 w2: Step-change in load demands of area 2 The efficient utilization of this dataset can result in trained intelligent online controllers that can very precisely forecast the optimum controller gains in a practical setting.

Files

Steps to reproduce

Software: MATLAB 2018b Optimizer settings: iteration=500, pop_size=100

Institutions

Islamic University of Technology

Categories

Forecasting, Frequency Control in Power System, Optimal Control Theory

Licence