VKVivek Kumar

Vivek Kumar

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Vivek Kumar has a total of 23 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Robust Global Adaptive Asymptotic Performance in Power Systems with Indefinite Control Parameters

    by Vivek Kumar, Soumya R. Mohanty, Nand Kishor, Udit Prasad

    1. 1.1 Modern power systems are increasingly evolving into large-scale, highly interconnected networks as renewable energy sources and distributed generation penetrate. While this transformation en...

  • Resilient Event Triggering-Based Agent Uncertainty Compensation in Grid-Forming Autonomous Network

    by Vivek Kumar, Padmini Singh, Sanjay Tolani

    Misbehaving agents in microgrids, caused by sensor faults, communication failures, malicious actions, or algorithmic deviations, can severely affect system stability and efficiency. This paper presents a dynamic event-triggered secondary control strategy, coordinated with primary control, to enhance performance and resilience. An event-based adaptive law for neural network (NN) observers is proposed to detect and mitigate agent deviations in distributed generation (DG) systems within an autonomous network. Event-triggering rules for updating NN weights and secondary control actions are derived to minimize communication and control efforts. A finite-time convergent primary control, based on dual sliding-surface formulations, complements the secondary layer by estimating power profile deviations under droop control during disturbances. Simulation results in MATLAB/Simulink confirm that the proposed framework improves stability and resiliency against agent deviations in leader-follower DG networks.


  • Open Access

    Deep Learning Model for Analyzing EEG Signal Analysis

    by Varun Gupta, Vivek Kumar, Prince, Saurabh Singh, Young-Seok Lee, In-Ho Ra

    To analyze the physiological information within the acquired EEG signal is very cumbersome due to the possibility of several factors, viz. noise and artifacts, complexity of brain dynamics, and inter-subject variability. To address these issues, this paper compares a U-shaped encoder-decoder network (UNET) and Bat-based UNET signal analysis (BUSA) techniques to classify depression rates in the Electroencephalogram (EEG) datasets. The main objective of including these two techniques is to reveal their effectiveness. It comprises pre-processing, feature extraction, feature selection, and classification stages. The framework excels at noise reduction during pre-processing, enhancing dataset integrity. Feature extraction leverages band power and correlation dimension to extract crucial features. Furthermore, feature selection optimizes classification accuracy by refining the fitness function of bats in the classification layer. The performance of UNET and BUSA are compared based on the following performance evaluating parameters viz. accuracy (Acc), Area Under the Curve (AUC), precision (P), and recall (R) (or sensitivity (Se)). The results indicated that the BUSA technique outperforms the UNET technique.


  • Two-Inductor CPM Control in Interleaved Bidirectional High-Gain DC–DC Converter

    by Amit Pratap Singh, Sonu Kumar, Sanjay Tolani, Vivek Kumar

    This article proposes an improved control strategy for an existing interleaved bidirectional high-gain dc–dc converter. The converter provides excellent steady-state performance. It uses an intermediate voltage-lift capacitor to achieve a high-voltage conversion ratio, whereas interleaving between its two inductor currents results in low ripple in low-voltage side high current. However, it suffers from poor transient performance in its step-up mode due to the fourth-order duty-to-output voltage transfer function with one right-half-plane zero (RHPZ). The article proposes a two-inductor current program mode (CPM) control in the step-up mode of the converter, which modifies the control-to-output voltage transfer function to a second-order system. It enhances the controller's bandwidth and results in a faster dynamic response. The proposed two-inductor CPM control relies on the feedback of both the inductor currents. A method is developed to estimate the average inductor currents from the intermediate capacitor current, reducing the required number of current sensors to one. A 200 W, 30/300 V, and 50 kHz laboratory prototype is developed to validate all the theoretical inferences and demonstrate the enhanced dynamic response resultant from the proposed two-inductor CPM control.


  • Entirely Coupled Recurrent Neural Network-Based Backstepping Control for Global Stability of Power System Networks

    by Vivek Kumar, Udit Prasad, Soumya R. Mohanty

    In an interconnected power system network, global performance is affected by various unknown power system parameters subjected to system uncertainties. These unknown parameters in the control expression deteriorate the performance of the power system network. Therefore, these terms associated with the control expression must be estimated precisely. This paper proposes an adaptive backstepping scheme using an entirely coupled recurrent neural network (ECRNN) to estimate these control terms. Three continuous differentiable functions are used to design control input, virtual signals, and adaptive control laws to achieve global performance. The objective of ECRNN is to reduce the control complexity by estimating the associated nonlinear term in the control expression. It will facilitate the complex formulation of the controller and improve system performance. The robust functional estimation ability of ECRNN will make the system immune to uncertainties that may appear due to external disturbances. In ECRNN, feedback loops are added to each neuron layer to achieve additional estimation accuracy. The proposed adaptive law enhances the online weight update speed and accuracy. Verification of the proposed scheme is carried out in MATLAB/Simulink. It is also verified in the real-time digital simulator (RTDS) platform using the IEEE standard New England 39-bus, 10-machine power system model. Note to Practitioners —This paper is motivated by the existing limitations in the global performance of a multimachine power system network due to uncertainty in the system parameters. Uncertainties in the power system network primarily appeared due to penetration of renewable power sources, load uncertainty, or occurrence of an uncertain fault in the network. As power system networks are complex nonlinear interconnected systems, these perturbations in the generator states may cause economic losses, load demand uncertainty, or electric scarcity. Thus, a control scheme is required to address the global performance of such networks. This article proposes uniform ultimate bounded stability of synchronous generators in a multimachine power system network with a robust ECRNN-based backstepping control design. The other available technique has not adequately addressed the global performance under uncertainties. The efficacy of the proposed scheme is verified in a real-time environment via a real-time digital simulator which may be a feasible solution for designing excitation control for the synchronous machine in an extensive industrial network.


  • Resilient Optimal Gain Control and Continuous Twisting Observer for Enhanced Power System Performance Under Uncertainties

    by Vivek Kumar, Soumya R. Mohanty

    In this article, resilient optimal gain control is presented for a multimachine power system for mismatched uncertainty compensation and to damp out the interarea oscillations. The proposed control scheme is analyzed in two substeps. In the first step, a power rate reaching law-based virtual control is designed, whereas the main control input in the second step is formulated using optimally tuned gains. Thus, this method does not incorporate multiple differentiation, unlike existing backstepping methods, hence eliminating magnitude limiter, rate limiter, and command filter dependence. The controller is designed by separating power system nonlinearities to eliminate complex expression from the control input that reduces the control computation. The proposed method has two steps procedure in control designing that suppresses the recursive step error in multistep virtual control formulation of backstepping methods. Furthermore, a nonlinear observer is designed using continuous twisting law with the integral surface design to counter the mismatched uncertainties along with the disturbance rejection. The proposed scheme is validated in MATLAB/Simulink and in real-time digital simulator platform of IEEE Standard New England 39-bus, 10-machine power system and compared with the existing methods.


  • Fractional Order Control with Super Twisting Observer for AC Microgrid

    by Vivek Kumar, Soumya R. Mohanty, Mahitosh Banafer

    In this paper, a fractional-order continuous control with a modified super twisting disturbance observer (STDO) for a microgrid (MG) is proposed. Transient accuracy under finite-time control is a challenging issue in Islanded MG with strong disturbance rejection. sliding mode control (SMC) with conventional Sliding surface design exhibits asymptotic response with chattering contamination in the control signal. If the terminal surface is used for finite-time control, singularity and chattering problem may occur. In order to achieve chattering free finite time response with no singularity problem, fractional second-order continuous control (FSCC) is proposed with STDO. In this paper, FSCC is primarily designed to improve transient response with finite-time control. Next, STDO is utilized to achieve strong robustness against disturbance with comprehensive chattering elimination by adding a fractional term. To capitalize on optimal control effort, event-triggering control is implemented. Stability proof and event trigger conditions are derived using rigorous Lyapunov based analysis. The efficacy of the proposed controller is tested under MATLAB/Simulink.