Research

🚩 This is a wordy page 🚩

Research Interests

I work in the broad domain of power electronics for distributed renewable energy and storage systems. My general research interests span several facets of power engineering including power converter modeling, analysis & synthesis of novel converter control paradigms, simulations, power-electronics hardware design & development as well as experimental testing & validation. My past and present research efforts encompass all these aspects and are largely centered around the utilization of high-performance PWM power converters as inverter-based resources (IBRs) for the grid integration of solar photovoltaics (PV) and energy storage resources. Topics of recent interest and excitement include the design and operation of grid-forming (GFM) & grid-following (GFL) inverters with advanced functionalities to facilitate large-scale renewable integration into the grid.

Current Research

The power grid is undergoing a significant transformation and moving rapidly towards a decarbonized and sustainable/green form dominated by inverter-based resources (IBRs). These are power-electronics-based systems (interfacing PV, wind, batteries, etc. with the grid) with significantly different dynamic properties compared to the traditional grid assets i.e., synchronous generators. Hence, careful design, control, and coordination of inverters are critical to successfully realizing a sustainable grid.

As a Postdoctoral Associate at the University of Minnesota, I presently work on a Department of Energy (DoE) funded research project for Solar Energy Technologies Office (SETO) that aims to develop next-generation inverter technologies for future grids. I collaborate with a network of researchers cutting across specializations from---University of Washington (UW)-Seattle, University of Illinois-Urbana-Champaign (UIUC), KU Leuven-Belgium, Electric Power Research Institute (EPRI), and the National Renewable Energy Laboratory (NREL), as well as Enphase energy.

GFL Inverter SI model

GFL Inverter per-unit model

GFM Inverter SI model

GFM Inverter per-unit model

My current research lies in the space of IBRs at the intersection of power electronics and power systems, going beyond the realm of typical microgrids and with a focus on grid-forming inverter technology. I am exploring novel hardware- and control-design paradigms in GFM inverters to achieve interoperability, fault-tolerance, and grid-agnostic operation of inverter groups interconnected by a network. I am presently working on developing high-fidelity simulation models for GFL- and GFM-inverter networks that are scalable across ratings. For this, I have utilized the per-unit framework and developed a method grounded in system theory for normalization of dynamical system models---a notable departure from the convention. This method leverages similarity transformation (a classical tool in state-space algebra) for obtaining the per-unit transcription. When viewed from such a system-theoretic lens, the scalings of system parameters needed in the per-unit model are automatically teased out without requiring exhaustive base-value calculations and bookkeeping (see figure above). More importantly, several theoretical underpinnings and properties of the per-unit dynamical models, such as structure preservation, identicality of eigenvalues, uniqueness, change of base operation, etc., get easily unraveled, which were not evident thus far. This approach is noted to be constructive, rigorous, repeatable, and unified across timescales [see here for more details].

I have also been working (in collaboration with KU Leuven-Belgium,) on advanced GFM control architectures in stationary reference-frame for inverter fault current limiting, that offers enhanced performance under unbalanced conditions. Our observation and finding are that control realization in stationary-αβ reference frame is advantageous over conventional-dq reference frame when unbalanced conditions are encountered during inverter operation; this could be caused by poor quality grid or IBR line faults (LL or LG). Control performed in αβ reference frame yields a high-bandwidth dynamic performance even with unbalance (much better than the dq counterpart with +ve and -ve sequence control), and, allows for the incorporation of enhanced current limiting schemes enabling the IBR to inject sufficient (yet thermally safe levels of) fault current to minimize voltage imbalance seen across the grid network during asymmetrical line faults [see here].

In addition, I am also simultaneously involved on the experimentation front (via hardware test-bed development) to demonstrate the operation of clusters of grid-forming inverters interconnected by a distribution network.

Past Research

Doctoral Research

My doctoral work pertains to a next-generation grid-following (GFL) inverter architecture to contend with power outages and enhance PV utilization in residential and commercial applications. I worked extensively on solar PV-based power-electronic systems and proposed a novel GFL inverter architecture, called the reconfigurable grid-tied inverter (RGTI), that had the capability to access solar energy even during a power outage, unlike a conventional GFL inverter that ceases operation in the absence of the grid owing to the mandatory anti-islanding requirements. This was achieved by reconfiguring the GFL as a DC-DC converter, retrofitting it with (any) generic external UPS present in a given facility, and supporting it with PV energy during the backup period. The retrofitting philosophy was a notable thematic deviation from convention since it allowed upgrading the given facility with PV capability without requiring a change or replacement of the existing backup infrastructure, which was otherwise the case with the existing inverter solutions with integrated battery storage (e.g., dual-mode inverters). See the figure below for visualization.

A conventional GFL Inverter:

Dual-mode Inverter:

Reconfigurable Grid-tied Inverter (RGTI):

The scope of my research efforts encompassed many key aspects of power engineering including modeling & control, simulation as well as hardware development at scale for experimental validation. Specifically, I developed a dynamic-phasor GFL system model, an approach that facilitated analysis of unintentional islanding behavior of GFLs and mathematically established its stability and frequency characteristics [see here]. For the DC-DC operation when tied to the UPS, I proposed a comprehensive dynamic impedance model of the PV source and developed a reduced-order model of a PV-fed DC-DC converter for simplified analysis and control synthesis. I established limiting conditions for stable MPPT convergence and achieved improved dynamic performance [see here]. Next, I developed a supervisory system-level control scheme for the RGTI for performing power management and controlling mode changes; this ensures that the discharge burden on the physical UPS battery was always minimized under all operating conditions of PV and UPS. With the supervisory control in place, the RGTI concept was demonstrated practically by successful interfacing and operation with a commercial 6kVA online UPS from Schneider Electric, thus validating the features of the proposed next-generation GFL inverter [see here]. My research also involved considerable hardware (and infrastructure) design and development efforts for testing and validation of IBRs [see here, here, and here].

Additionally, in this stint, I have directly mentored and trained two junior graduate (master's) students and two project assistants in the power group at EE, IISc. Pertinent published works can be seen here, here, and here.

Ultracapacitor cells

Research Associateship

As a Senior Research Associate at IISc, Bangalore. I drove the effort on design of Ultracapacitor (UC) based energy storage systems (ESS) by mentoring a grad student. I worked on optimizing the size of the UC stack for a given contingency requirement. We developed an improved design procedure that not only optimizes the discharge ratio of the stack but also systematically engineers the stack parameters via an iterative selection process, which led to minimization of the overall system cost of the ESS [see here]. Subsequently, we uncovered the nonlinear behavior inherently exhibited by UCs, developed a framework for exact analysis & characterized the effective UC-stack capacitance, and finally, benchmarked the ESS design [see here].

Graduate Research

In my graduate student research, I worked on an integrated-magnetic PWM filter-transformer (IMFT) design for such a GFL inverter. The design achieved a higher-order LCL-filter action in an isolation transformer (typically needed for grid integration) by introducing a third transformer winding with a series-connected capacitor. Such an integrated approach to LCL filter design was a departure from the convention at that point, where, discrete magnetic components were almost always employed. The 3kVA filter-transformer was hand-wound in the lab, and, its theory was tested and validated via grid-connected operation of a single-phase 2L H-bridge inverter [see here].

IMFT hardware

Industrial R&D

In between my doctoral research and graduate degree, I gained four years of professional work experience in industrial R&D in two Fortune 500 companies, which are global leaders in power engineering and energy technologies today:

During this stint, I worked extensively on PWM converter design for two different power applications---uninterrupted power supplies (UPS) and Healthcare. In the UPS domain at Schneider, I was part of a complete product development cycle of an online double-conversion smart-UPS system—equipment that provides conditioned and regulated emergency power to mission-critical systems from a battery storage system via a controlled standalone PWM inverter. Here, I got the opportunity to work in a multi-converter system setting, with rectifier, inverter, and battery charger subsystems, all working in tandem to power the loads. I was also involved with the sizing, characterization, and management of battery energy-storage systems for meeting contingency requirements. The product that I worked on, is now available commercially [see here].

At GE, I worked on a BLDC motor drive for multi-axis motion control of patient-positioning platform in a cardiovascular X-ray imaging system. In the process, I gained insights into various critical electrical requirements demanded by medical-instrument standards.