Prediction-Based Sliding Mode Load Frequency Control with Unknown I/O Time Delays using Delay and Disturbance Estimation

The paper investigates prediction-based sliding mode load frequency control (LFC) with unknown input and output (I/O) time delays using delay estimation. The control architecture consists of a prediction-based super twisting control (STC), higher order sliding mode observer (HOSMO), disturbance observer and a delay estimator. The unknown round trip delay (algebraic sum of I/O delays) is estimated using a gradient descent algorithm. Single area power system model is considered to validate the effectiveness of the proposed architecture. The sliding surface is designed based on predictor states and estimated disturbance. The efficacy of the control scheme lies in finite time convergence of frequency under load disturbance with I/O time delays. Furthermore, frequency deviation in hybrid power system is also observed with an integration of wind and solar energies. Asymptotic stability is theoretically proved using Lyapunov functional candidate and verified using simulation.