Department of Electrical Engineering

Indian Institute of Technology Palakkad&nbsp;(IIT Palakkad or IIT PKD) is a public autonomous&nbsp;engineering&nbsp;and&nbsp;research&nbsp;institute located in&nbsp;Palakkad,&nbsp;Kerala. It was&nbsp;one of the five new&nbsp;IITs&nbsp;proposed in the&nbsp;2014 Union budget of India. The campus was inaugurated on 3 August 2015. The institute has 94 faculty, 978 students, and 63 non-teaching staffs.

IT Palakkad currently offers four-year Bachelor of Technology (B.Tech) and two years MS programmes in 8 different engineering disciplines including&nbsp;Computer Science, Electrical, Mechanical, Civil among others.

&nbsp;

IIT Palakkad

Due to the lack of a comprehensive multiobjective solid-state transformer (SST) design framework, SST designs are mostly obtained through trial/experience. In this article, a machine-learning (ML)-aided optimal SST design framework is proposed, which involves the objectives of maximizing efficiency ( η ) and power density ( ρ ). The challenges of computationally expensive magnetics design, coupled with the correlation between magnetics design and performance of semiconductor devices, are tackled by developing a hybrid local optimization algorithm. This local optimization is subsequently learned through ML techniques, using a limited number of optimal design data sets, and, thus, assists in the genesis of optimal SST design limits for several combinations of semiconductor devices and switching frequencies. The proposed framework is implemented for a cascaded matrix-based dual-active-bridge (CMB-DAB) SST comprising of SiC MOSFETs to demonstrate the optimization routine. The optimization results exhibit low-error fits of the selected ML models and the η - ρ limits in different categories of optimal SST designs. The SiC-based SST designs are also observed to offer better η - ρ optimal designs compared to Si-based SSTs. A laboratory-scale CMB-DAB prototype with experimental measurements is also presented to validate the proposed design optimization framework at a scaled-down level. © 2013 IEEE.

World University root node

IEEE Journal of Emerging and Selected Topics in Power Electronics

Machine-Learning-Aided Optimization Framework for Design of Medium-Voltage Grid-Connected Solid-State Transformers

Despite some recent efforts towards multi-objective design optimization of multilevel converters, design optimization of solid-state-transformers (SSTs) are not presented much in the literature mainly because of the lack of computationally feasible techniques. This paper is dedicated towards a computational feasibility study of multi-objective design optimization techniques for medium-voltage (MV) grid-connected SSTs. After defining the application and scope of SST design optimization problem, a brief description of the possible solution techniques are discussed which shows the merits of semi-numerical/hybrid design optimization techniques. Subsequently, a machine learning (ML) aided hybrid optimization technique is executed for a 15 kVA single-stage SiC-based SST design. Suitable component modelling is presented and a strong agreement is observed between theoretical optimization and experimental results. Finally, a comparative evaluation of the analytical, numerical, standalone hybrid and ML-aided hybrid optimization techniques (deployed for the same 15 kVA SiC-based SST design) reveals that the ML-aided hybrid strategy is best suited for SST design optimization as it requires feasible computational time for <5% error. © 2021 IEEE.

IECON Proceedings (Industrial Electronics Conference)

Computational Feasibility of Multi-objective Optimal Design Techniques for Grid-Connected Multi-cell Solid-State-Transformers

This paper proposes a multi-input linear control law based on the incremental passivity theory for cascaded H-bridge (CHB)-based battery energy storage systems (BESSs). The proposed control law guarantees large-signal stability and controls multiple state variables without assuming dynamic decoupling between the voltages and the currents. Assuming dynamic decoupling is generally not acceptable in low-capacitance converters. Furthermore, the broad range of operating conditions in a CHB-based BESS due to the varying state of the batteries and varying grid requirements, makes it important to consider the nonlinear nature of the system. This paper presents a passivity-based control that ensures the global stability despite changing operating conditions. Simulation results on a single-phase CHB-based BESS are shown to corroborate that the proposed control approach exhibits excellent performance. © 2021 IEEE.

2021 IEEE Energy Conversion Congress and Exposition, ECCE 2021 - Proceedings

Passivity Control in Modular Battery Energy Storage Systems

Single-stage inverters with a common ac and de ground are emerging topologies which effectively eliminate the high-frequency common mode voltage. However, most of the existing common-ground-type inverters are based on switched-capacitor (SC) circuits which are inherently subjected to several important issues that lead to many practical limitations. Disadvantages such as high current spikes, high capacitor voltage ripples, and distorted ac output voltage with de offset are some significant challenges in these inverters. A new common-ground-type inverter that discard the use of SCs is therefore proposed in this paper. It consists of a single-inductor and an ac output capacitor that are controlled by 8 power switches. Sinusoidal ac voltage generation in both buck and boost modes can be achieved by controlling the inductor with a sinusoidal duty-cycle. Simulations are presented to verify the operation of the proposed topology. © 2021 IEEE.

Proceedings of the Energy Conversion Congress and Exposition - Asia, ECCE Asia 2021

A Common-Ground-Type Single-Stage Buck-Boost Inverter with Sinusoidal Output Voltage

This paper presents a geometric finite-time inner-outer loop control strategy for slung payload transportation using a quadrotor. The underactuated nature of the quadrotor in conjunction with the cable connection makes it impossible to simultaneously and arbitrarily track the load position, cable direction and the quadrotor’s attitude. By exploiting the differential flatness of the system dynamics, it is possible to arbitrarily control the load position, with the cable direction and quadrotor attitude remaining dependent on the trajectory and controller dynamics. A two-loop feedback control strategy is proposed comprising an inner-loop control law addressing quadrotor attitude tracking within finite time, and an outer-loop controller addressing the stabilization of the cable direction and position tracking of the payload trajectory. The control laws are expressed in a coordinate-free setting and are singularity-free. We include the effects of unmodeled dynamics on the load system and design an adaptive estimator to estimate and reject such uncertainties. The performance and robustness of the feedback laws is demonstrated through numerical simulations and experimental results. © 2021, The Author(s), under exclusive licence to Springer Nature B.V.

Nonlinear Dynamics

Geometric finite-time inner-outer loop trajectory tracking control strategy for quadrotor slung-load transportation

Automated event detection and classification are vital to power system monitoring. This article proposes a novel event detection and classification scheme based on wide area frequency measurement system (WAFMS). Raw frequency measurements obtained from WAFMS are used as the only input to the event detection and classification algorithm (EDCA). Wavelet-based signal pre-processing is used to denoise the data. Afterward, the rate of change of frequency (ROCOF) is estimated from the frequency measurements using the Kalman filter (KF). In the same step, phase angle difference (PAD) across different stations is estimated using WAFMS. Thus, the overall algorithm uses three features such as frequency, ROCOF and PAD to detect and classify events in the power system. In the first step, an event is detected based on standard deviation (SD) of estimated ROCOF and PAD. In the second step, four types of events are classified using wide area frequency measurements. The suggested algorithms have been validated with real WAFMS data from the Indian Power System, recent 9th August 2019 U.K. blackout data collected from the U.K. power system and 20th July 2017 oscillation event data obtained from ISO New England (ISO-NE) power system. As a promising tool for power system monitoring, the suggested scheme requires less input measurement for decision making and is having a low computational complexity, which is suitable for practical application.

IEEE Transactions on Smart Grid

A Novel Event Detection and Classification Scheme Using Wide-Area Frequency Measurements

Effective management of multiple real-Time transient events and unstable low-frequency oscillations in the modern power system has been a matter of concern to the Transmission system operators. This manuscript unifies the idea of transient event detection and localization with the analysis of the oscillation mode that accompanies these impulsive events. The paper employs Discrete wavelet transform (DWT) to segregate transient events from oscillatory behavior that follows these events. The paper also proposes a novel indicator for the classification of transient events based on the most affected signals in due course of the event. Empirical Mode decomposition (EMD) and Hilbert spectral analysis (HSA) are utilized on the PMU signals severely affected by the event and further examine the oscillatory modes succeeding the real-Time events. Oscillatory modes deduced from the above adaptive transformations are further categorized into three frequency bands based on power system control applications, thus helping operators provide efficient control actions. The novel methodology discussed in the paper has been applied to the IEEE 39 bus system with a dataset generated using the RTDS power system simulator and GTNETx2 based PMUs. © 2021 IEEE.

2021 9th IEEE International Conference on Power Systems (ICPS)

A Novel Hybrid Algorithm for Event Detection, Localisation and Classification

The development of new smart grid technologies and the rising environmental concerns lead to increased penetration of green energy sources into the existing power grid. This increased penetration of renewable energy sources (RES) into the power system will affect system dynamics. This paper discusses the influence of integrating full converter-based renewable energy sources on different system stability indicators (SSIs) such as rate of change of frequency (ROCOF), frequency nadir, etc. Different test systems such as Kundur's two area model and IEEE-9 bus system are considered to evaluate the impact of different penetration levels of RES on power system stability indicators. The response of the system to common contingencies such as generator outage, load shedding, faults, etc. are studied using Dynamic Security Assessment Tools (DSA Tools) from Powertech Labs. It is observed that the post-disturbance dynamic behavior of a conventional synchronous generator-dominated system is significantly different from that of an inverter-dominated system and it is heavily dependent on penetration level, type of disturbance, type of RES, etc. © 2021 IEEE.

Impact of Different Penetration Level of Type-IV Renewable Energy Resources on Power System Dynamics

In this paper, we investigate the performance of the WiFi-based human presence detection (HPD) method using the two-dimensional convolutional neural network (2DCNN) and long-short term memory (LSTM) under generalized-room model and room-specific model learning environments with optimal 2DCNN and LSTM parameters. The HPD method is designed to classify the WIFi-data segment into a room with person (RWP) or a room without person (RWoP) using magnitude channel state information (CSI). The preprocessing is used to perform filtering and to split the recorded CSI data into blocks, and data preparation. We created the CSI database which consists of 33 hostel room datasets, 3 laboratory room datasets and 3 home room datasets for room specific models with support of 33 volunteers, by using the commercial WiFi device with single antenna. Based on the evaluation results under unseen datasets, cross-validation and room-specific models, we observe that the LSTM based HPD method had an average sensitivity, SE =91.70%, specificity, SP =89.60% and accuracy, ACC= 90.57% that outperforms the 2DCNN based HPD method (SE =89.93%, SP= 88.23% and ACC =88.96%). Further, the LSTM based HPD method had a processing time of 1.8 ms and model size of 934 KiloBytes whereas the 2DCNN based HPD method had a processing time of 2.9 ms and model size of 10470 KiloBytes. © 2021 IEEE.

Proceedings of the 13th International Conference on Electronics, Computers and Artificial Intelligence, ECAI 2021

Non-Collaborative Human Presence Detection Using Channel State Information of Wi-Fi Signal and Long-Short Term Memory Neural Network

This paper presents a hybrid method for detecting power quality events based on combined Hilbert transform and Fourier transform. The decision tree classifier has been implemented over the detection scheme to classify eight power quality disturbances. The proposed method extracts the envelope from the Hilbert transform (HT) and identifies the number of harmonic components present in the power quality (PQ) signal using the Fast Fourier transform (FFT) technique. To evaluate the proposed method, a standard database was created from the mathematical models and white Gaussian noise is added to the database with different noise levels. The relevant features, such as energy, kurtosis, minimum amplitude from the envelope, and the number of peaks from the Fourier magnitude spectrum are extracted. The features have been used to train the decision tree (DT) classifier and a classification accuracy of 99.30% is obtained. The robustness of the proposed algorithm is addressed on another synthetic database with different noise levels, and it achieves 99.27% overall classification accuracy. Finally, the proposed method is compared with recent techniques and achieves promising results in terms of classification accuracy. © 2021 IEEE.