Arghya Chakravarty

Actuator fault tolerant control scheme for nonlinear uncertain systems using backstepping based sliding mode

A fault tolerant control (FTC) scheme for tracking control is proposed for nonlinear uncertain systems affected by actuator faults and matched uncertainties with unknown upper bound. The proposed controller utilizes backstepping control which is integrated with integral sliding mode control method. For designing the integral sliding mode controller, an adaptive law is used to estimate the unknown upper bounds of faults and uncertainties. The proposed FTC ensures satisfactory transient and steady state performances with a chattering free control input. Asymptotic stability in presence of matched uncertainties and partial loss of actuator effectiveness is guaranteed in the proposed control system. The efficiency of the proposed fault tolerant control scheme is demonstrated by simulation results.

Finite time current observer based adaptive backstepping control of buck converters

A finite time current observer based adaptive backstepping control strategy is proposed for the output voltage regulation of buck type DC-DC converters. The proposed current observer is designed by utilizing only the output voltage information to reconstruct the inductor current profile while achieving an adaptive control by means of backstepping procedure. This method eliminates the usage of extra sensor involved in sensing the inductor current, thereby reducing the cost of control besides overcoming the problems of measurement noise encountered while sensing. Simulations have been performed on a buck converter using Matlab software under both continuous and discontinuous conduction modes. Further, the usefulness of proposed scheme is also examined by subjecting the buck converter system to sudden changes in load resistance. The results obtained reveal that the proposed observer is successful in not only estimating the nominal inductor current but also estimates the perturbed level of inductor current under load disturbances in finite time. Satisfactorily transient and steady state response in the output voltage are ensured by using the proposed method.

Real time implementation of an adaptive backstepping control of buck converter PMDC-motor combinations

This article presents an experimental realization of adaptive backstepping control methodology on a cascaded buck converter permanent magnet dc (PMDC)-motor combination for angular velocity control. The experiment aims at illustrating the practical applicability of adaptive control to power converters fed with a DC motor load. The systematic design procedure of conventional backstepping control design is enhanced by incorporating an online adaptive control mechanism to estimate the unknown non-linear load torque. Asymptotic stability of the closed loop system under the action of proposed control law is ensured and update law is derived satisfying Lyapunov stability criterion. The experimental investigation is conducted using dSPACE, Control Desk DS1103 setup with an embedded TM320F240 Digital Signal Processor. The buck dc-dc converter fed PMDC motor system is subjected to a wide variation in load torque and set point angular velocity tracking. The results obtained through adaptive backstepping control scheme have been evaluated against the conventional backstepping control mechanism. Results highlight a superior performance using adaptive backstepping control by producing an accurate and time bound estimation of unknown load torque, under both nominal and perturbed conditions, thereby improving the transient and steady state response of desired angular velocity.

Backstepping enhanced adaptive second order sliding mode controller to compensate actuator failures

In this paper, a failure compensation scheme is designed for affine nonlinear systems prone to actuator stuck failures unknown in time, magnitude and pattern. The design scheme utilizes an adaptive second order sliding mode control developed in a backstepping framework exploiting the advantages of both the methodologies. An adaptive law is used to estimate the unknown upper bounds of uncertainties introduced due to the occurrence of actuator failures and guarantees a globally bounded estimation. The proposed failure compensation scheme ensures effective failure accommodation while providing excellent transient and steady state performances compared to the basic scheme based on backstepping, with chattering free control inputs. Stability and asymptotic tracking in presence of unknown actuator failures at unknown time instants is proved for the proposed control scheme using Lyapunov criterion. Simulation results illustrate the effectiveness of the proposed methodology.

Analysis and experimental investigation into a finite time current observer based adaptive backstepping control of buck converters

A finite time current observer based adaptive backstepping control strategy is proposed for the output voltage regulation of buck type DC-DC converters. The proposed current observer is designed by utilizing only the output voltage information to reconstruct the inductor current profile while achieving an adaptive control by means of backstepping procedure. This method eliminates the usage of extra sensor involved in sensing the inductor current, thereby reducing the cost of control besides overcoming the problems of measurement noise encountered while sensing. Simulations have been performed on a buck converter using Matlab software under both continuous and discontinuous conduction modes. Further, the usefulness of proposed scheme is also examined by subjecting the buck converter system to sudden changes in load resistance. The results obtained reveal that the proposed observer is successful in not only estimating the nominal inductor current but also estimates the perturbed level of inductor current under load disturbances in finite time. Satisfactorily transient and steady state response in the output voltage are ensured by using the proposed method.

Relay approach for parameter extraction of li-ion battery and SOC estimation using finite time observer

This paper proposes a novel integrated robust system identification and state of charge estimation of Li-ion battery. A finite time extended state observer is used for state of charge (SOC) estimation of Li-ion battery. A relay feedback test has been adopted to yield accurate battery parameters in noisy environment. New identification method using state space based relay feedback approach is employed with a disturbance nullifier to extract the parameters of battery model. Thereafter, the objective of SOC estimation is carried out by designing a novel finite time extended state observer based on higher order sliding modes. The unknown battery model parameters are determined first, followed by the estimation of SOC next, under both charging and discharging conditions. This investigation finds a high amount of accuracy in the proposed battery model identification method, besides an accurate and time bound estimation of SOC using the proposed observer scheme.

Compensating actuator failures in near space vehicles using adaptive finite time disturbance observer based backstepping controller

In this paper, a novel adaptive failure compensation scheme is designed for near space vehicles (NSVs) which belong to a class of affine nonlinear MIMO systems, prone to actuator float free, stuck and oscillatory failures, unknown in time, pattern and magnitude. The proposed control methodology utilizes a backstepping procedure based on an adaptive finite time disturbance observer to estimate failure induced uncertainties and external disturbances. The proposed failure compensation scheme ensures effective failure accommodation in finite time while providing excellent transient and steady state performances even when three input failures occur simultaneously. Stability and asymptotic tracking in presence of unknown actuator failures at unknown time instants using the proposed control scheme as well as finite time stability of the adaptive finite time disturbance observer is proved using Lyapunov criterion. Simulation results illustrate the effectiveness of the proposed methodology in application to NSVs.