Sairaj Dhople

Sairaj Dhople

Auto-generated profile, based on publicly available information

Sairaj Dhople

University of Minnesota, Minneapolis, MN, USA

Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 October 2021 in IEEE Transactions on Signal Processing
Verified
Journal Article
Closed access
XZXinwei Zhang
Xinwei Zhang
MHMingyi Hong
Mingyi Hong
SDSairaj Dhople
Sairaj Dhople
WYWotao Yin
Wotao Yin
YLYang Liu
Yang Liu
Federated Learning (FL) is popular for communication-efficient learning from distributed data. To utilize data at different clients without moving them to the cloud, algorithms such as the Federated Averaging (FedAvg) have adopted a computation then aggregation model, in which multiple local updates are performed using local data before aggregation. These algorithms fail to work when faced with practical challenges, e.g., the local data being non-identically independently distributed. In this paper, we first characterize the behavior of the FedAvg algorithm, and show that without strong and unrealistic assumptions on the problem structure, it can behave erratically. Aiming at designing FL algorithms that are provably fast and require as few assumptions as possible, we propose a new algorithm design strategy from the primal-dual optimization perspective. Our strategy yields algorithms that can deal with non-convex objective functions, achieves the best possible optimization and communication complexity (in a well-defined sense), and accommodates full-batch and mini-batch local computation models. Importantly, the proposed algorithms are communication efficient , in that the communication effort can be reduced when the level of heterogeneity among the local data also reduces. In the extreme case where the local data becomes homogeneous, only $\mathcal {O}(1)$ communication is required among the agents. To the best of our knowledge, this is the first algorithmic framework for FL that achieves all the above properties.

18 August 2017 in IEEE Transactions on Power Systems
Verified
Journal Article
Closed access
EDEmiliano Dall'Anese
Emiliano Dall'Anese
SGSwaroop S. Guggilam
Swaroop S. Guggilam
ASAndrea Simonetto
Andrea Simonetto
YCYu Christine Chen
Yu Christine Chen
SDSairaj V. Dhople
Sairaj V. Dhople
This paper develops a real-time algorithmic framework for aggregations of distributed energy resources (DERs) in distribution networks to provide regulation services in response to transmission-level requests. Leveraging online primal-dual-type methods for time-varying optimization problems and suitable linearizations of the nonlinear AC power-flow equations, this work establishes a system-theoretic foundation to realize the vision of distribution-level virtual power plants. The optimization framework controls the output powers of dispatchable DERs such that, in aggregate, they respond to automatic generation control and/or regulation-services commands. This is achieved while concurrently regulating voltages within the feeder and maximizing customers' and utility's performance objectives. Convergence and tracking capabilities are analytically established under suitable modeling assumptions. Simulations are provided to validate the proposed approach.

25 January 2018 in IEEE Transactions on Power Systems
Verified
Journal Article
Closed access
SGSwaroop S. Guggilam
Swaroop S. Guggilam
CZChanghong Zhao
Changhong Zhao
EDEmiliano Dall'ANese
Emiliano Dall'ANese
YCYu Christine Chen
Yu Christine Chen
SDSairaj V. Dhople
Sairaj V. Dhople
This paper develops an approach to enable the optimal participation of distributed energy resources (DERs) in inertial- and primary-frequency response alongside conventional synchronous generators. Leveraging a reduced-order model description of frequency dynamics, DERs' synthetic inertias and droop coefficients are designed to meet time-domain performance objectives of frequency overshoot and steady-state regulation. Furthermore, an optimization-based method centered around classical economic dispatch is developed to ensure that DERs share the power injections for inertial- and primary-frequency response in proportion to their power ratings. Simulations for a modified New England test-case system composed of ten synchronous generators and six instances of the IEEE 37-node test feeder with frequency-responsive DERs validate the design strategy.

8 August 2023 in IEEE Transactions on Power Systems
Verified
Journal Article
Closed access
JLJose Daniel Lara
Jose Daniel Lara
RHRodrigo Henriquez-Auba
Rodrigo Henriquez-Auba
DRDeepak Ramasubramanian
Deepak Ramasubramanian
SDSairaj Dhople
Sairaj Dhople
DCDuncan Callaway
Duncan Callaway
SSSeth Sanders
Seth Sanders
The changing nature of power systems dynamics is challenging present practices related to modeling and study of system-level dynamic behavior. While developing new techniques and models to handle the new modeling requirements, it is also critical to review some of the terminology used to describe existing simulation approaches and the embedded assumptions. This article provides a first-principles review of the simplifications and transformations commonly used in the formulation of time-domain simulation models. It introduces a taxonomy and classification of time-domain simulation models depending on their frequency bandwidth, network representation, and software availability. Furthermore, it focuses on the fundamental aspects of averaging techniques, and model reduction approaches that result in modeling choices, and discusses the associated challenges and opportunities of applying these methods in systems with large shares of Inverter Based Resources (IBRs). The article concludes with an illustrative simulation that compares the trajectories of an IBR-dominated system.

25 March 2022 in IEEE Transactions on Power Electronics
Verified
Journal Article
Closed access
MLMinghui Lu
Minghui Lu
SDSairaj Dhople
Sairaj Dhople
BJBrian Johnson
Brian Johnson
Virtual oscillator control (VOC) is a time-domain strategy that leverages the dynamics of nonlinear oscillator circuits for synchronizing and regulating grid-forming inverters. In this article, we examine a class of second-order circuits composed of a harmonic oscillator and nonlinear state-dependent damping that has found extensive interest in the context of VOC. We center our analysis on the Van der Pol, Dead-zone, and Andronov-Hopf oscillators; these are characterized by several distinguishing attributes but they all share the common structure noted above. Analytical methods based on averaging and perturbation theory are outlined to derive several performance metrics related to harmonic and dynamic properties of these oscillators under a unified framework. Our results reveal that the Andronov-Hopf oscillator is well suited for grid-forming inverter applications since it can yield harmonics-free waveforms without compromising dynamic performance. Analytical results are validated with numerical simulations and experiments, and a multiinverter hardware setup is used to illustrate a practical use case.

5 September 2022 in IEEE Transactions on Energy Conversion
Verified
Journal Article
Closed access
NBNathan Baeckeland
Nathan Baeckeland
DVD. Venkatramanan
D. Venkatramanan
MKMichael Kleemann
Michael Kleemann
SDSairaj Dhople
Sairaj Dhople
Grid-forming (GFM) control offers promising performance features for inverter-based resources (IBRs) across scales. However, design, analysis, and benchmarking of GFM IBRs during unbalanced faults remains largely unexplored. In this paper, we outline a stationary-reference-frame nested-loop control architecture for GFM IBRs and integrate the same with novel current-limiting strategies. The architecture improves on virtual-impedance and current-reference-saturation limiting as well as state-of-the-art methods for control of voltage-source inverters. Electromagnetic-transient simulations for a modified IEEE 14-bus network validate salient features of the proposed control architectures. The proposed virtual-impedance limiter is shown to provide better voltage support during faults than the current-reference-saturation limiter (quantified via sequence voltages). On the other hand, the current-reference-saturation limiter offers better (and more accurate) fault-current contribution.