A unified approach to modelling of time-varying delay system for electric vehicles

The article uses an identification approach for reducing the time-varying delay systems of electric vehicles in the unified delta operator framework. Constrained hybrid grey wolf optimizer based algorithms are applied to reduce the Padé approximated high-order model generated from a time-delay system to lower-order representation to develop its control scheme. Five new hybrid propositions with grey wolf optimizer and chaotic particle swarm optimization as its constituents have been utilized in this paper for system modeling of time-delay systems. To estimate the reduced model parameters, step response matching is typically used. As step input is a biased signal, the parameters identification process in linear time-invariant systems is not a reasonable yardstick since quicker transients are not visible clearly. Therefore, this paper uses the Pseudo-Random Binary Sequence as the input signal which exhibits behavior similar to white noise, having a crest factor on the lower side, the frequency of which can be changed if the rate of sampling is varied. A test system is employed to show the efficiency of the proposed procedure. The suggested methods ultimately display their superiority over the current and comparable algorithms.