IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (IEEE SmartGridComm 2020)

Session Opening-remarks-3: Opening Remarks Session Awards-Ceremony: Best Paper Awards Ceremony Session Awards-Ceremony2: Technical Achievements Awards Ceremony Session Awards-Ceremony3: Announcement of SGC2021 Session SIFI: Student Industry Faculty Interaction
Session A4: Analytics 4: Load and Price Forecasting Session C3: Control 3: Energy Markets Session S4: Security 4: Privacy, Attacks and Mitigation Session A5: Analytics 5: Data Analytics for Grid II Session C4: Control 4: Storage and attacks Session IS-5: Invited 5: Distribution Systems and DER Session IS-6: Invited 6: Grid Optimization under Emerging Constraints Session S5: Security 5: System Security and Modeling Session WMLPDS: Workshop on Machine Learning and Big Data Analytics in Power Distribution Systems Session WEC: Workshop on Edge Computing for Smart Grids Session WMLPDS2: Workshop on Machine Learning and Big Data Analytics in Power Distribution Systems
Session Tutorial-3: Accelerating AI on the Grid: A Hands On Tutorial on PMU Data Analysis
Session Industry-Panel: Industry Panel

Session Opening-remarks-3

Opening Remarks

Conference
2:00 PM — 2:10 PM UTC
Local
Nov 13 Fri, 8:00 AM — 8:10 AM CST
Session Chair

Lalitha Sankar (ASU) & Anuradha Annaswamy (MIT) <br> Zoom Room Host(s): Melissa Torres (IEEE), Andrea Pincetti (ASU)

Session Awards-Ceremony

Best Paper Awards Ceremony

Conference
2:10 PM — 2:15 PM UTC
Local
Nov 13 Fri, 8:10 AM — 8:15 AM CST
Session Chair

György Dán (KTH) <br> Zoom Room Host(s): Melissa Torres (IEEE), Andrea Pincetti (ASU)

Session Awards-Ceremony2

Technical Achievements Awards Ceremony

Conference
2:15 PM — 2:25 PM UTC
Local
Nov 13 Fri, 8:15 AM — 8:25 AM CST
Session Chair

Angela Yingjun Zhang (CUHK) <br> Zoom Room Host(s): Melissa Torres (IEEE), Andrea Pincetti (ASU)

Session Awards-Ceremony3

Announcement of SGC2021

Conference
2:25 PM — 2:30 PM UTC
Local
Nov 13 Fri, 8:25 AM — 8:30 AM CST
Session Chair

Anna Scaglione (ASU) & Antonello Monti (RWTH Aachen) <br> Zoom Room Host(s): Melissa Torres (IEEE), Andrea Pincetti (ASU)

Session A4

Analytics 4: Load and Price Forecasting

Conference
2:30 PM — 3:20 PM UTC
Local
Nov 13 Fri, 8:30 AM — 9:20 AM CST

Real-time Locational Marginal Price Forecasting Using Generative Adversarial Network

Zhongxia Zhang and Meng Wu (Arizona State University, USA)

2
In this paper, we propose a model-free unsupervised learning approach to forecast real-time locational marginal prices (RTLMPs) in wholesale electricity markets. By organizing system-wide hourly RTLMP data into a 3-dimensional (3D) tensor consisting of a series of time-indexed matrices, we formulate the RTLMP forecasting problem as a problem of generating the next matrix with forecasted RTLMPs given the historical RTLMP tensor, and propose a generative adversarial network (GAN) model to forecast RTLMPs. The proposed formulation preserves the spatio-temporal correlations among system-wide RTLMPs in the format of historical RTLMP tensor. The proposed GAN model learns the spatio-temporal correlations using the historical RTLMP tensors and generate RTLMPs that are statistically similar and temporally coherent to the historical RTLMP tensor. The proposed approach forecasts system-wide RTLMPs using only publicly available historical price data, without involving confidential information of system model, such as system parameters, topology, or operating conditions. The effectiveness of the proposed approach is verified through case studies using historical RTLMP data in Southwest Power Pool (SPP).

Electricity Load Forecasting with Collective Echo State Networks

Siwu Liu and Chenxiao Xu (Stony Brook University, USA); Ying Liu and Dimitrios Katramatos (Brookhaven National Laboratory, USA); Shinjae Yoo (Brookhaven National Lab, USA)

0
Short-term load forecasting (STLF) is one of the most important tasks in power grid systems. Despite thorough research of STLF on community scales, predicting individual power consumption remains challenging. With the deployment of advanced metering infrastructure (AMI) technology, it is more convenient to access and operate load usage data on an individual scale. This work first investigates a novel deep learning method, Collective Long Short-term Memory (LSTM), which can fully utilize correlated meter information from AMI. Then, Collective Echo State Networks (ESN) are proposed as a simplified form of Collective LSTM using the reservoir method. Experimental results show that Collective ESN can achieve state-of-the-art prediction accuracy with extremely fast speed and fewer computational resources.

Coordinated Demand Response By Data Centers Using Inverse Optimization

Athanasios Tsiligkaridis, Ioannis Paschalidis and Ayse Coskun (Boston University, USA)

2
Demand Response (DR) policies define the interactions between an energy supplier and its consumers and allow for customer energy regulation given a supplier request. Given the high flexibility and controllability of Data Centers (DC), they are promising candidates to participate in DR for power grid stabilization. In this work, we consider the setting where an energy supply deficit event occurs and must be addressed to avoid grid strain. We present two novel frameworks for DR, where a load aggregator offers price incentives to a set of consumer DCs so they can dynamically adjust their electricity consumption and provide DR to the grid via server usage reductions. Modeling DCs using realistic cost functions based on Quality of Service (QoS) requirements of the DC workloads, we present a data-driven inverse optimization method to estimate DC cost function parameters for precise and efficient pricing and provide an algorithm for solving the inverse problem. Experimental results on two test cases demonstrate the benefits of our proposed DR mechanisms for energy control.

Load Approximation for Uncertain Topologies in the Low-Voltage Grid

Ludovic Mouline, Maxime Cordy and Yves Le Traon (University of Luxembourg, Luxembourg)

0
Smart grids allow operators to monitor the grid continuously, detect occurring incidents, and trigger corrective actions. To perform that, they require a deep understanding of the effective situation within the grid. However, some parameters of the grid may not be known with absolute confidence. Reasoning over the grid despite uncertainty needs the consideration of all possible states. In this paper, we propose an approach to enumerate only valid potential grid states. Thereby, we allow discarding invalid assumptions that poison the results of a given computation procedure. We validate our approach based on a real-world topology from the power grid in Luxembourg. We show that the estimation of cable load is negatively affected by invalid fuse state combinations, in terms of computation time and accuracy.
Session Chair

Chen Chen (Xi'an Jiaotong University) <br> Zoom Room Host(s): Raksha Ramakrishna (KTH), Boya Hou (UIUC)

Session C3

Control 3: Energy Markets

Conference
2:30 PM — 3:20 PM UTC
Local
Nov 13 Fri, 8:30 AM — 9:20 AM CST

Combflex: a Linear Combinatorial Auction for Local Energy Markets

Diego Kiedanski (Telecom ParisTech, France); Ariel Orda (Technion, Israel); Daniel Kofman (Telecom Paristech, France)

1
Local energy markets, platforms in which prosumers in the same Low Voltage network can trade energy among themselves, offer a great opportunity to incentivize the consumption of locally generated energy.
Unfortunately, traditionally proposed implementations of local energy markets such as simple double auctions and peer to peer exchanges do not fully exploit the available flexibility in these systems.

We design a market mechanism that exploits the characteristics of the players, providing them with expressive bids to represent their flexibility, which we assume is due to energy storage.
The proposed market is not obviously manipulable and can be cleared by solving a linear programming problem that grows linearly in the number of participants.

Using realistic data, we benchmark the proposed mechanism against sequential auctions and peer to peer exchanges often used in the literature.
Our numerical results show that the proposed mechanism outperforms traditional implementations.

Misalignments of Objectives in Demand Response Programs: A Look at Local Energy Markets

Diego Kiedanski (Telecom ParisTech, France); Daniel Kofman (Telecom Paristech, France); Patrick Maillé (IMT Atlantique, France); Jose Horta (Telecom ParisTech, France)

1
Local energy markets (LEMs) have been proposed to mitigate the variability introduced in power systems by distributed renewable energy resources such as photo-voltaic energy.
During the progressive release of LEMs, the decision problem faced by prosumers (consumers that might also produce energy), will differ from the wholesale electricity market's one because there is always the alternative to buy from or sell to the utility company.
In this setting, guaranteeing that the aggregated energy consumption will be well behaved depends on the properties of the mechanisms used to implement the market, the alternative tariff offered to participants by their utility and how prosumers interact among themselves.
We present a pathological example of a LEM in which the best strategy for the agents results in unnecessary peaks of demand.
A decision model for players participating in LEMs is developed to study the existence of undesirable behaviour while using realistic data and number of participants.
Through numerical experiments, we identify the key aspects of the player's behaviour, strategy and environment that lead to the aforementioned peaks, all under reasonable circumstances.
Simple fixes are discussed to overcome the pitfalls of such markets.

DER Information Unaware Coordination via Day-ahead Dynamic Power Bounds

Thomas Navidi, Chloe Leblanc, Abbas El Gamal and Ram Rajagopal (Stanford University, USA)

0
Reliability and voltage quality in distribution networks have been achieved via a combination of transformer power rating satisfaction and voltage management asset control. To maintain reliable operation under this paradigm, however, future grids with deep DER penetrations would require costly equipment upgrades. These upgrades can be mitigated via judicious coordination of DER operation. Earlier work has assumed a hierarchical control architecture in which a global controller (GC) uses detailed power injection and DER data and knowledge of DER owners' objectives to determine setpoints that local controllers should follow in order to achieve reliable and cost effective grid operation. Having such detailed data and assuming knowledge of DER owners' objectives, however, are often not desirable or possible. In an earlier work, a 2- layer DER coordination architecture was shown to achieve close to optimal performance despite infrequent (e.g., once per day) communication to a global controller. Motivated by this work, this paper proposes a day-ahead coordination scheme that uses forecasted power profile ranges to generate day-ahead dynamic power rating bounds at each transformer. Novel features of this scheme include: (i) the GC knows only past node power injection data and does not impose or know DER owner objectives, (ii) we use bounds that ensure reliable operation to guide the local controllers rather than setpoint tracking, and (iii) we consider electric vehicle (EV) charging in addition to storage. Simulations using the IEEE 123-bus network show that with random placements of 50% solar, 50% EVs and only 10% storage penetrations, the uncoordinated approach incurs rating violations at nearly all 86 transformers and results in 10 times higher voltage deviation, while our approach incurs only 12 rating violations and maintains almost the same voltage deviations as before the addition of solar and EVs.

An Incentive Compatible Market Mechanism for Integrating Demand Response into Power Systems

Ce Xu, Hossein Khazaei and Yue Zhao (Stony Brook University, USA)

0
In this paper, we propose a market mechanism that allows demand response providers (DRPs) to participate in a two-settlement electricity market as power suppliers alongside conventional generators. Each DRP bids her demand reduction capacity and cost rate, and the independent system operator (ISO) schedules the power dispatch to minimize the overall system cost. We show that, with the proposed mechanism, truthful bidding by the DRPs is achieved at a Nash equilibrium (NE), and as a result, the social welfare is maximized. The theoretical results are corroborated by simulation studies.
Session Chair

Yue Zhao (Stony Brook Univ.) <br> Zoom Room Host(s): Rajasekhar Anguluri, Abrar Zahit (ASU)

Session S4

Security 4: Privacy, Attacks and Mitigation

Conference
2:30 PM — 3:20 PM UTC
Local
Nov 13 Fri, 8:30 AM — 9:20 AM CST

Towards Privacy-Preserving Anomaly-based Attack Detection against Data Falsification in Smart Grid

Yu Ishimaki (Waseda University, Japan); Shameek Bhattacharjee (Western Michigan University, USA); Hayato Yamana (Waseda University, Japan); Sajal K. Das (Missouri University of Science and Technology, USA)

1
In this paper, we present a novel framework for privacy preserving anomaly based data falsification attack detection in smart grid. Specifically, we propose an anomaly detector over homomorphically encrypted (HE) ciphertext data. Unlike existing privacy-preserving anomaly detectors, our framework not only identifies energy theft (i.e.,deductive attack), but also detects more advanced data integrity attacks (i.e.additive and camouflage attacks). We optimize the anomaly detection algorithm for computational efficiency, thus making it practical for resource-constrained devices like smart meters, achieving 40x speed-up over the naive method. We also validate the proposed framework with real data sets from smart metering infrastructures, and demonstrate that it is possible to detect data integrity attacks with high sensitivity, yet without sacrificing user privacy. Experimental results with a real data set from Texas for 200 houses, in a grid of hourly meter resolution showed that the detection sensitivity of the plaintext algorithm is not lost due to the use of HE.

Optimal Cyber Defense Strategy of High-Voltage DC Systems for Frequency Deviation Mitigation

Jiazuo Hou (The University of Hong Kong, China); Shunbo Lei (University of Michigan, USA); Wenqian Yin (The University of Hong Kong, China); Chaoyi Peng (China Southern Power Grid, China); Yunhe Hou (University of Hong Kong, China)

1
As a critical infrastructure for bulk power transmission in power system, the high-voltage DC (HVDC) transmission system faces the problem of potential cyberthreats. The controllability of the HVDC system heavily relies on the measurements and corresponding communication systems, which makes the HVDC system vulnerable to cyberattacks. This paper proposes an optimal sequence of cyber defense strategy for HVDC transmission system to minimize the frequency deviation caused by cyberattacks on measurements spoofing. The whole sequence of the cyberattack and cyber defense strategies are modeled as a multi-stage attack-defense decision problem, which is transferred into a single-level mixed integer quadratic programming (MIQP) problem with the assumption of the simple if-else cyberattack strategy. The effectiveness of the optimal cyber defense strategy is verified in a two-terminal HVDC transmission system.

An Attack-Trace Generating Toolchain for Cybersecurity Study of IEC 61850 based Substations

Partha P. Biswas, Yuan Li, Heng Chuan Tan and Daisuke Mashima (Advanced Digital Sciences Center, Singapore); Binbin Chen (Singapore University of Technology and Design, Singapore)

2
The digitisation of modern power grid substations
allows them to better support various advanced operations.
However, it also poses greater risk of cyber-related attacks. There
is an array of cyber-security solutions (e.g., intrusion detection systems) available in the market to prevent, detect, or respond to cyberattacks. This calls for the creation of datasets and test cases for the validation of those cyber-security solutions. In our recent work, we have generated a synthesized dataset for testing of cyber-security solutions of IEC 61850 based substations. Our
dataset contained traces for some typical attack-free disturbance
scenarios and cyberattack scenarios in a substation. In this
work, we present the toolchain we have developed to allow easy
generation of such traces. We discuss the design considerations
behind our toolchain and provide step-by-step guide to potential
users on how to create customized trace files for specific scenarios
using our toolchain. By open-sourcing the project for the broad
community, we hope our toolchain will enrich the body of testing
datasets for substation cyber security solutions.

A Cyber-Resilient Privacy Framework for the Smart Grid with Dynamic Billing Capabilities

Gaurav S. Wagh and Sumita Mishra (Rochester Institute of Technology, USA)

1
The desired features for the smart grid include dynamic billing capabilities along with consumer privacy protection. Existing aggregation-based privacy frameworks have limitations such as centralized designs prone to single points of failure and/or a high computational overload on the smart meters due to in-network aggregation or complex algorithmic operations. Additionally, these existing schemes do not consider how dynamic billing can be implemented while consumer privacy is preserved. In this paper, a cyber-resilient framework that enables dynamic billing while focusing on consumer privacy preservation is proposed. The distributed design provides a framework for spatio-temporal aggregation and keeps the process lightweight for the smart meters. The comparative analysis of our proposed work with existing work shows a significant improvement in terms of the spatial aggregation overhead, overhead on smart meters and scalability. The paper also discusses the resilience of our framework against privacy attacks.
Session Chair

Samir M. Perlaza (INRIA, France) <br> Zoom Room Host(s): Ezzeldin Shereen, Lamia Varawala (KTH)

Session A5

Analytics 5: Data Analytics for Grid II

Conference
3:30 PM — 4:20 PM UTC
Local
Nov 13 Fri, 9:30 AM — 10:20 AM CST

Household Level Electricity Load Forecasting using Echo State Network

Debneil Saha Roy (Stony Brook University, USA)

1
Load forecasting at the household level is challenging because the electricity consumption behavior can be much more variable than those at aggregate levels. The introduction of Advanced Metering Infrastructure (AMI) systems has helped to better forecast the load of an individual household. Since smart meter data is streaming data and there is a need to deal with a massive amount of such data in a real-time fashion, an efficient and fast framework to handle this challenge is required. Deep neural networks like Long Short Term Memory (LSTM) can be used for this purpose but they take a long time to train. In this paper a novel k-nearest meter based Echo State Network (ESN) is proposed and experimental results demonstrate that it is a more suitable candidate for load forecasting, since it is much easier to train and has a great deal of accuracy. The model is compared with other time series models like Persistent (PM) and Vector Autoregression (VAR), as well as deep learning models like multi-layer perceptron (MLP), LSTM and a combination of convolutional neural network (CNN) and LSTM (CNN-LSTM). The results show that the proposed model has a significant improvement over all other models on a dataset spanning 4 months, along with a significant reduction in training time compared primarily to deep learning models.

Transfer Learning for Operational Planning of Batteries in Commercial Buildings

Brida Mbuwir (KU Leuven & EnergyVille, Belgium); Kaveh Paridari (KTH Royal Institute of Technology, Sweden); Fred Spiessens (VITO, Belgium); Lars M Nordström (Royal Institute of Technology, KTH, Sweden); Geert Deconinck (KU Leuven, Belgium)

1
Recently, building owners are investing in rooftop photovoltaic (PV) installations and batteries in order to meet the (facility) load in their buildings. As a consequence, several commercial and research solutions have emerged for battery energy management in such buildings. Most of these solutions rely on sufficiently accurate system models and are tailor-made for those systems. This work proposes the use of transfer learning in model-free reinforcement learning (RL) to control the operation of batteries in buildings. This enables knowledge from the control of a battery in one building to be used by a RL algorithm to control a battery in another building with similar characteristics. In this paper, the K-shape clustering algorithm is used to group buildings with similar characteristics - based on their energy consumption patterns. To plan the operation of the batteries, we use fitted Q-iteration, a RL algorithm. Simulation results using real-world data show that by including forecast information on energy consumption and PV generation in the feature space of the control algorithm, RL competes with mixed integer linear programming - which assumes perfect knowledge of the system. We also investigate through simulation, the effect of transferring a policy learned with data from one building to another building - all buildings belonging to the same cluster. Simulation results show a faster convergence - convergence achieved with fewer training samples required - to a near optimal policy.

Exploiting Satellite Data for Solar Performance Modeling

Akansha Singh Bansal (University of Massachusetts Amherst, USA); David Irwin (University of Massachusetts, Amherst, USA)

1
Developing accurate solar performance models, which infer solar power output in real time based on the current environmental conditions, are an important prerequisite for many advanced energy analytics. Recent work has developed sophisticated data-driven techniques that generate customized models for complex rooftop solar sites by combining well-known physical models with both system and public weather station data. However, inferring solar generation from public weather station data has two drawbacks: not all solar sites are near a public weather station, and public weather data generally quantifies cloud cover---the most significant weather metric that affects solar---using highly coarse and imprecise measurements.

In this paper, we develop and evaluate solar performance models that use satellite-based estimates of downward shortwave (solar) radiation (DSR) at the Earth's surface, which NOAA began publicly releasing after the launch of the GOES-R geostationary satellites in 2017. Unlike public weather data, DSR estimates are available for every 0.5km 2 area. As we show, the accuracy of solar performance modeling using satellite data and public weather station data depends on the cloud conditions, with DSR-based modeling being more accurate under clear skies and station-based modeling being more accurate under overcast skies. Surprisingly, our results show that, overall, pure satellite-based modeling yields similar accuracy as pure station-based modeling, although the relationship is a function of conditions and the local climate. We also show that a hybrid approach that combines the best of both approaches can also modestly improve accuracy.

Predictive Maintenance for Increasing EV Charging Load in Distribution Power System

Salman Shuvo and Yasin Yilmaz (University of South Florida, USA)

1
The increasing number of electric vehicles (EVs)
introduces a high intensity charging load to the power system.
The distribution systems are not well prepared to cope with this
high variance load. To handle such EV charging load, utility
companies need a predictive maintenance approach for the distribution transformers. We propose a deep reinforcement learning
(RL) based policy to timely replace the distribution transformers
by similar or higher capacity ones under a budgetary constraint
of selecting at most one transformer for replacement per time
step. Our policy outperforms the myopic policies which replace
transformers based on load, age, and failure in terms of both
economic cost and power outage.

SoDa: An Irradiance-Based Synthetic Solar Data Generation Tool

Ignacio Losada Carreno (ASU, USA); Raksha Ramakrishna and Anna Scaglione (Arizona State University, USA); Daniel Arnold (Lawrence Berkeley National Laboratory, USA); Ciaran Roberts (Lawrence Berkeley National Lab & UC Berkeley, USA); Sy-Toan Ngo and Sean Peisert (Lawrence Berkeley National Laboratory, USA); David Pinney (National Rural Electric Cooperative Association, USA)

1
In this paper, we present SoDa, an irradiance-based synthetic Solar Data generation tool to generate realistic sub-minute solar photo-voltaic (PV) output power time series, that emulate the weather pattern for a certain geographical location. Our tool relies on the National Solar Radiation Database (NSRDB) to obtain irradiance and weather data patterns for the site. Irradiance is mapped onto a PV model estimate of a solar plant's 30-min power output, based on the configuration of the panel. The working hypothesis to generate high-resolution (e.g. 1 second) solar data is that the conditional distribution of the time series of solar power output given the cloud density is the same for different locations. We therefore propose a stochastic model with a switching behavior due to different weather regimes as provided by the cloud type label in the NSRDB, and train our stochastic model parameters for the cloudy states on the high-resolution solar power measurements from a Phasor Measurement Unit (PMU). In the paper we introduce the stochastic model, and the methodology used for the training of its parameters. The numerical results show that our tool creates synthetic solar time series at high resolutions that are statistically representative of the measured solar power and illustrate how to make use of the tool to create synthetic data for arbitrary sites in the footprint covered by the NSRDB.
Session Chair

Chen Chen (Xi'an Jiaotong University) <br> Zoom Room Host(s): Nima Taghizhpourbazargani (ASU)

Session C4

Control 4: Storage and attacks

Conference
3:30 PM — 4:20 PM UTC
Local
Nov 13 Fri, 9:30 AM — 10:20 AM CST

Time-of-Use and Demand Charge Battery Controller using Stochastic Model Predictive Control

Michael Blonsky (National Renewable Energy Laboratory & Colorado School of Mines, USA); Killian McKenna (National Renewable Energy Laboratory, USA); Tyrone Vincent (Colorado School of Mines, USA); Adarsh Nagarajan (National Renewable Energy Laboratory, USA)

2
Stationary batteries in residential and commercial buildings are often used to smooth customer load profiles and to lower customer electricity bills. Controllers for these battery systems should account for customer energy consumption, rate structures, and high internal battery temperatures, which can lead to reduced performance over the battery lifetime. It is important to consider the uncertainty in forecasting energy consumption and temperature, especially for customers with highly variable and uncertain loads. We propose a novel battery controller using stochastic model predictive control that accounts for these uncertainties and can handle complex rate structures, including demand charges. We show that the controller performs better than standard model predictive control when there is significant uncertainty in the forecast. We also show improvements in the performance with more accurate forecasts and with a more aggressive control strategy.

Impact Analysis of EV Preconditioning on the Residential Distribution Network

Joseph Antoun, Mohammad Ekramul Kabir, Ribal Atallah, Bassam Moussa, Mohsen Ghafouri and Chadi Assi (Concordia University, Canada)

1
Electric Vehicles (EV) are coming with a stupendous load demand that raises enough concerns for the power sector. The backlash of such increased demand is notable at the distribution side with different aspects of EV usage. During winter, EV users favor preconditioning their vehicles before leaving their houses, such as heating the cabin and battery compartment to make the operation of EVs more comfortable. Consequently, such behavior along with a higher penetration of level 2 smart chargers prompt the presence of a new peak in the residential load profile. This new unexpected peak that operators have to face can disturb the performance of the network. To forsee the impact of preconditioning, we simulate multiple scenarios to assess the network's quality metrics (voltage level and power losses). We expose that preconditioning poses risks on the network in its current state. Furthermore, we evaluate the competencies of network reconfiguration to handle the new imposed preconditioning demand.
We find out that reconfiguration will be able to aid the performance of the network to an average EV penetration rate.

Grid-Coupled Dynamic Response of Battery-Driven Voltage Source Converters

Ciaran Roberts (Lawrence Berkeley National Lab & UC Berkeley, USA); Jose Daniel Lara and Rodrigo Henriquez-Auba (University of California, Berkeley, USA); Bala K Poolla (National Renewable Energy Laboratory, USA); Duncan Callaway (UC Berkeley, USA)

0
With the increasing interest in converter-fed islanded microgrids, particularly for resilience, it is becoming more critical to understand the dynamical behavior of these systems.This paper takes a holistic view of grid-forming converters and considers control approaches for both modeling and regulating the DC-link voltage when the DC-source is a battery energy storage system. We are specifically interested in understanding the performance of these controllers, subject to large load changes, for decreasing values of the DC-side capacitance. We consider a fourth, second, and zero-order model of the battery;and establish that the zero-order model captures the dynamics of interest for the timescales considered for disturbances examined.Additionally, we adapt a grid search for optimizing the controller parameters of the DC/DC controller and show how the inclusion of AC side measurements into the DC/DC controller can improve its dynamic performance. This improvement in performance offers the opportunity to reduce the DC-side capacitance given an admissible DC voltage transient deviation, thereby, potentially allowing for more reliable capacitor technology to be deployed.

Model-Agnostic Algorithm for Real-Time Attack Identification in Power Grid using Koopman Modes

Sai Pushpak Nandanoori, Soumya Kundu, Seemita Pal, Khushbu Agarwal and Sutanay Choudhury (Pacific Northwest National Laboratory, USA)

1
Malicious activities on measurements from sensors like Phasor Measurement Units (PMUs) can mislead the control center operator into taking wrong control actions resulting in disruption of operation, financial losses, and equipment damage. In particular, false data attacks initiated during power systems transients caused due to abrupt changes in load and generation can fool the conventional model-based detection methods relying on thresholds comparison to trigger an anomaly. In this paper, we propose a Koopman mode decomposition (KMD) based algorithm to detect and identify false data attacks in real-time. The Koopman modes (KMs) are capable of capturing the nonlinear modes of oscillation in the transient dynamics of the power networks and reveal the spatial embedding of both natural and anomalous modes of oscillations in the sensor measurements. The Koopman-based spatio-temporal nonlinear modal analysis is used to filter out the false data injected by an attacker. The performance of the algorithm is illustrated on the IEEE 68-bus test system using synthetic attack scenarios generated on GridSTAGE, a recently developed multivariate spatio-temporal data generation framework for simulation of adversarial scenarios in cyber-physical power systems.
Session Chair

Adarsh Nagarajan (NREL) <br> Zoom Room Host(s): Arka Sanka (UT Austin)

Session IS-5

Invited 5: Distribution Systems and DER

Conference
3:30 PM — 4:20 PM UTC
Local
Nov 13 Fri, 9:30 AM — 10:20 AM CST

Voltage Regulation and Protection for Power Distribution Systems using Reinforcement Learning

Dileep Kalathil (TAMU)

0
The complexity of power system protection is rapidly increasing as a result of increasing penetration of power electronic interfaced energy resources and loads. We propose a Deep-Reinforcement Learning based approach that is highly accurate, robust to factors including high fault impedance and distributed generation, and has a very fast response time. A sequential training algorithm is developed to enable communication-free coordination between multiple protection devices. Intuitive results on the convergence and optimality of the sequential training algorithm is also given from the aspect of optimal control.

Optimal Coordination of High and Low Voltage Systems to Leverage DERs

Lindsay Anderson (Cornell)

0
With recent and predicted future growth in demand-side and distributed resources, it is becoming increasingly valuable to effectively leverage these resources in support of the changing generation mix. Demand response programs have long been seen as a key mechanism to compensate for the increasing variability and uncertainty associated with renewable resource development. However, demand response programs are not well characterized within the operations of transmission systems, where we expect significant growth in renewable resources. Co-optimization of transmission and distribution systems is a potentially valuable mechanism for coordinating demand-side flexibility to manage uncertainty and diversify variabilities. This work explores the co-optimization of transmission and distribution systems through a bi-level framework. This formulation provides the necessary information sharing, wherein the upper-level problem represents the high voltage transmission system, and the lower level accounts for the lower voltage distribution systems. A single level reformulation strategy is implemented for computational tractability and shows the potential to coordinate multiple distribution systems within a single transmission system. This presentation will provide a brief overview of the bi-level formulation and compare this approach to a representative traditional centralized optimization approach. Some preliminary results will highlight the impact and potential of co-optimization for leveraging the flexibility of demand-side resources in supporting the integration of large-scale renewable generation.
Session Chair

Line Roald (UW Madison) <br> Zoom Room Host(s): Abrar Zahit (ASU)

Session IS-6

Invited 6: Grid Optimization under Emerging Constraints

Conference
4:30 PM — 5:20 PM UTC
Local
Nov 13 Fri, 10:30 AM — 11:20 AM CST

Balancing Wildfire Risk and Power Outages through Optimized Power Shut-offs

Line Roald (UW-Madison)

0
Electric grid faults can be the source of catastrophic wildfires, particularly in regions with high winds and low humidity. In short-term operations, electric utilities have few options to mitigate the risk of wildfire ignitions, leading to use of disruptive measures such as proactive de-energization of equipment, frequently referred to as public safety power shut-offs. Decisions of how to operate the grid in situations with high wildfire risk has significant impacts on customers, who may loose access to electricity in an attempt to protect them from fires. This talk discusses our proposed optimal power shut-off problem, an optimization model to support operational decision making in the context of extreme wildfire risk. Specifically, the model optimizes grid operation to maximize the amount of power that can be delivered, while proactively minimizing the risk of wildfire ignitions by selectively de-energizing components in the grid. This is the first optimization model to consider optimization of preventive wildfire risk measures in a short-term, operational time-frame. The effectiveness of the method is demonstrated on an augmented version of the IEEE-RTS GMLC test case, and compared against two simpler approaches. We observe that the optimization-based model reduces both wildfire risk and lost load shed relative to the benchmarks.

Electricity and Water Do Mix: Interdependent Electric and Water Infrastructure Modeling, Optimization and Control

Vijay Vittal (ASU)

0
The phrase water-energy nexus is commonly used to describe the inherent and critical interdependencies between the electric power system (EPS) and the water distribution system (WDS). In this paper, the analytical framework capturing the interactions between these two critical infrastructures is examined and a mathematical model to describe the associated dynamics is developed. Based on the time-scale of these associated dynamics, the EPS simulation is conducted using time-series power flows following unit commitment and optimal power flow solutions. The WDS control optimization-simulation model formulated here is solved using a genetic algorithm solution technique interfaced with EPANET. An integrated simulation engine of the interdependent infrastructure systems was created to conduct long-term simulations. The simulation engine was applied using representative WDS and EPS networks. The implemented control optimization benefits both systems by reducing the effect of severe contingencies. The results of the simulations conducted prove the applicability of the proposed methodology for long – term, water-energy nexus contingency simulations having both power outages and droughts.
Session Chair

Dileep Kalathil (TAMU) <br> Zoom Room Host(s): Jiaqi Wu (ASU)

Session S5

Security 5: System Security and Modeling

Conference
4:30 PM — 5:20 PM UTC
Local
Nov 13 Fri, 10:30 AM — 11:20 AM CST

Benefits and Cyber-Vulnerability of Demand Response System in Real-Time Grid Operations

Mingjian Tuo, Arun Venkatesh Ramesh and Xingpeng Li (University of Houston, USA)

0
With improvement in smart grids through two-way communication, demand response (DR) has gained significant attention due to the inherent flexibility provided by shifting non-critical loads from peak periods to off-peak periods, which can greatly improve grid reliability and reduce cost of energy. Operators utilize DR to enhance operational flexibility and alleviate network congestion. However, the intelligent two-way communication is susceptible to cyber-attacks. This paper studies the benefits of DR in security-constrained economic dispatch (SCED) and then the vulnerability of the system to line overloads when cyber-attack targets DR signals. This paper proposes a false demand response signal and load measurement injection (FSMI) cyber-attack model that sends erroneous DR signals while hacking measurements to make the attack undetectable. Simulation results on the IEEE 24-bus system (i) demonstrate the cost-saving benefits of demand response, and (ii) show significant line overloads when the demand response signals are altered under an FSMI attack.

Making Renewable Energy Certificates Efficient, Trustworthy, and Anonymous

Dimcho Karakashev and Sergey Gorbunov (University of Waterloo, Canada); Srinivasan Keshav (University of Cambridge, United Kingdom (Great Britain))

1
Although renewable energy costs are declining rapidly, producers still rely on additional incentives, such as Renewable Energy Certificates (RECs), when making an investment decision. An REC is a proof that a certain amount of energy was generated from a renewable resource. It can be traded for cash in an REC market. Unfortunately, existing mechanisms to ensure that RECs are trustworthy—not fraudulently generated and from a universally-agreed renewable energy source—require periodic audits of the generation plant, which adds costly administrative overheads and locks out small producers. Although prior work has attempted to address these issues, existing solutions lack privacy and are vulnerable to tampering. In this work, we design, implement, and evaluate a system that is efficient, trustworthy, and anonymous, thus opening the REC market to small-scale energy producers. Our implementation is based on the commercially-available Azure Sphere micro-controller unit and the Algorand public blockchain.

Online Reasoning about the Root Causes of Software Rollout Failures in the Smart Grid

Ewa Piatkowska, Catalin Gavriluta, Paul Smith and Filip Pröstl Andren (AIT Austrian Institute of Technology, Austria)

0
An essential ingredient of the smart grid is software-based services. Increasingly, software is used to support control strategies and services that are critical to the grid's operation. Therefore, its correct operation is essential. For various reasons, software and its configuration needs to be updated. This update process represents a significant overhead for smart grid operators and failures can result in financial losses and grid instabilities. In this paper, we present a framework for determining the root causes of software rollout failures in the smart grid. It uses distributed sensors that indicate potential issues, such as anomalous grid states and cyber-attacks, and a causal inference engine based on a formalism called evidential networks. The aim of the framework is to support an adaptive approach to software rollouts, ensuring that a campaign completes in a timely and secure manner. The framework is evaluated for a software rollout use-case in a low voltage distribution grid. Experimental results indicate it can successfully discriminate between different root causes of failure, supporting an adaptive rollout strategy.

Stacked Metamodels for Sensitivity Analysis and Uncertainty Quantification of AMI Models

Michael Rausch (University of Illinois at Urbana-Champaign, USA); William Sanders (Carnegie Mellon University, USA)

0
Models can help architects design effective and secure advanced metering infrastructure (AMI) deployments. Because of the complex interactions among the numerous smart meters, smart home devices, customers, the utility, and potential adversaries, the models are often complex and have long execution times. In addition, the models often contain a large number of uncertain input variables. Modelers seek to understand the impact of uncertain input variables on the model through the use of sensitivity analysis (SA) and uncertainty quantification (UQ). However, long-running models are not amenable to such techniques, since they require that the model be run many times. One approach to help overcome this challenge is to build a metamodel (a model of the model) that accurately emulates the original model but is much faster.
Session Chair

Katherine Davis (Texas A&M University <br> Zoom Room Host(s): Rajasekhar Anguluri (ASU)

Session WMLPDS

Workshop on Machine Learning and Big Data Analytics in Power Distribution Systems

Conference
4:30 PM — 5:20 PM UTC
Local
Nov 13 Fri, 10:30 AM — 11:20 AM CST

Machine Learning for Quick Stability Decisions Using Dispatch as Inputs (Invited talk)

Yilu Liu (University of Tennessee, USA)

2
This talk introduce a fast way to assess transient stability, frequency stability and small signal stability of a system from its dispatch as inputs. Either simulation or operation data could be used as training data set to teach the ML package how to associate any dispatch to its stability margins. The ML tool will allow engineers to quickly map any dispatch to a stability zone both online or offline without actually running the simulations. The ML tool will reduce the time for both planning and operation required simulations.

Dynamic State Estimation Based Monitoring of High Frequency Transformer

Boqi Xie (Georgia Institute of Technology, USA); Dongbo Zhao (Argonne National Laboratory, USA); Tianqi Hong (NYU Polytechnic School of Engineering, USA); Alex Huang (University of Texas at Austin, USA); Zhicheng Guo (UT Austin, USA); Yuzhang Lin (University of Massachusetts, Lowell, USA)

2
Solid-state transformer (SST) is considered as a promising option for interfacing between different voltage levels in power systems. High frequency transformer (HFT), as the core of SST, needs to be continuously monitored to ensure its operating health. This paper presents a dynamic state estimation based approach that monitors the health status such as winding deformation of HFT. The proposed method is constructed upon the detailed high-fidelity HFT device model. With the device model and given measurements, an overall measurement model of the monitored HFT is formulated. Then, a dynamic state estimation algorithm is performed directly on the measurement model, and a health status observer is designed to check the consistency between the measurements and the HFT device model. A use case is applied in this paper for demonstration and it substantiates the proposed method by checking the health status of the monitored HFT.

Learning to Optimize Power Distribution Grids using Sensitivity-Informed Deep Neural Networks

Manish Kumar Singh, Sarthak Gupta and Vassilis Kekatos (Virginia Tech, USA); Guido Cavraro and Andrey Bernstein (National Renewable Energy Laboratory, USA)

1
Deep learning for distribution grid optimization can be advocated as a promising solution for near-optimal yet timely inverter dispatch. The principle is to train a deep neural network (DNN) to predict the solutions of an optimal power flow (OPF), thus shifting the computational effort from real-time to offline. Nonetheless, before training this DNN, one has to solve a large number of OPFs to create a labeled dataset. Granted the latter step can still be prohibitive in time-critical applications, this work puts forth an original technique for improving the prediction accuracy of DNNs by taking into account the sensitivities of the OPF minimizers with respect to the OPF parameters. By expanding on multiparametric programming, it is shown that although inverter control problems may exhibit dual degeneracy, the required sensitivities do exist in general and can be computed readily using the output of any standard quadratic program (QP) solver. Numerical tests showcase that sensitivity-informed deep learning can enhance prediction accuracy in terms of mean square error (MSE) by 2-3 orders of magnitude at minimal computational overhead. Improvements are more significant in the small-data regime, where a DNN has to learn to optimize using a few examples. Beyond multiparametric QPs, the approach is currently being generalized to parametric (non)-convex optimization problems.

Restoring Distribution System Under Renewable Uncertainty Using Reinforcement Learning

Xiangyu Zhang, Abinet Tesfaye Eseye, Bernard Knueven and Wesley Jones (National Renewable Energy Laboratory, USA)

1
Distributed energy resources (DERs) in distribution systems, including renewable generation, micro-turbine, and energy storage, can be used to restore critical loads following extreme events to increase grid resiliency. However, properly coordinating multiple DERs in the system for multi-step restoration process under renewable uncertainty and fuel availability is a complicated sequential optimal control problem. Due to its capability to handle system non-linearity and uncertainty, reinforcement learning (RL) stands out as a potentially powerful candidate in solving complex sequential control problems. Moreover, the offline training of RL provides excellent action readiness during online operation, making it suitable to problems such as load restoration, where in-time, correct and coordinated actions are needed. In this study, a distribution system prioritized load restoration based on a simplified single-bus system is studied: with imperfect renewable generation forecast, the performance of an RL controller is compared with that of a deterministic model predictive control (MPC). Our experiment results show that the RL controller is able to learn from experience, adapt to the imperfect forecast information and provide a more reliable restoration process when compared with the baseline controller.
Session Chair

Deepjyoti Deka (LANL), Yu Zhang (UC Santa Cruz) <br> Zoom Room Host(s): Shammya Saha (ASU), Amir Gholani (WSU)

Session Industry-Panel

Industry Panel

Conference
5:30 PM — 6:20 PM UTC
Local
Nov 13 Fri, 11:30 AM — 12:20 PM CST

Grid control in the presence of high levels of inverter-based resources

Dr. Evangelos Farantatos (EPRI)

2

Introduction

Anamitra Pal (ASU)

0
This talk does not have an abstract.

Challenges and solutions for understanding behind-the-meter energy resources

Dr. YingChen Zhang (NREL)

2

Emerging reliability issues for the bulk power system

Dr. Ryan Quint (NERC)

2

Spectral Analysis of Long Term Synchrophasor Data

Dr. Chetan Mishra (Dominion Energy)

2
Session Chair

Anamitra Pal (ASU) <br> Zoom Room Host(s): Jiaqi Wu (ASU)

Session WEC

Workshop on Edge Computing for Smart Grids

Conference
5:30 PM — 6:00 PM UTC
Local
Nov 13 Fri, 11:30 AM — 12:00 PM CST

Edge Layer Design and Optimization for Smart Grids

Adetola Adeniran and Md Abul Hasnat (University of South Florida, USA); Minoo Hosseinzadeh and Hana Khamfroush (University of Kentucky, USA); Mahshid Rahnamay-Naeini (University of South Florida, USA)

0
The emergence of modern monitoring, communication, computation, and control equipment into power systems has made them evolve into smart grids that can be thought of as the electric grid of things. This evolution has enhanced the efficiency of the power systems through the availability of a large volume of system data that can help with system functions nevertheless, it has intensified the communication and computation burden on these systems. While many such computations were traditionally deployed in central servers, new technologies such as edge computing can provide unique opportunities to address some of the computational challenges and improve the responsiveness of the system by processing data locally. In this paper, an edge enabled smart grid architecture is presented. The edge layer for the smart grid is designed through various optimization formulations to identify the placement of edge servers and their connectivity structure to the Phasor Measurement Units in the system. Various factors affecting the design, such as the geographical and resource constraints as well as the communication technology considerations have been incorporated in the formulations and evaluated using the IEEE 118 bus system.

Distributed Anomaly Detection and PMU Data Recovery in a Fog-computing-WAMS Paradigm

Kaustav Chatterjee (The Pennsylvania State University, USA); Nilanjan Chaudhuri (Pennsylvania State University, USA)

1
Large volumes of monitoring data from massive deployment of Phasor Measurement Units (PMUs) are expected to overwhelm the data pre-processors at a centralized computing facility. This coupled with the requirements of lower latency and increased resilience to anomalies advocates for distributed architectures for data conditioning. To that end, in this paper, we present a fog-computing-based hierarchical approach for distributed detection and correction of anomalies in PMU data. In our proposed approach, each fog node responsible for real-time data pre-processing, is dynamically assigned a smaller group of PMU signals with similar modal observabilities using software-defined-networking (SDN). The SDN controller residing at a central node feeds on the mode-shapes estimated from the recovered signals at each fog node for running the PMU-grouping algorithm. Grouping ensures adequate denseness of each signal set and guarantees data recovery under corruption. Also, the grouping is soft-real-time, infrequent, and triggered only upon a change in operating condition, and therefore heavily relieves the computational burden off the central node. The effectiveness of the proposed approach is demonstrated using simulated data from the IEEE 5-area 16-machine test system.

Optimal Smart Grid Operation and Control Enhancement by Edge Computing

Yuan Liao and Jiangbiao He (University of Kentucky, USA)

0
The national electric power grid is being transformed into a smart grid through deploying a huge number of distributed sensors across the network, a two-way communication system, intelligent control and optimization algorithms, and advanced hardware components. An increasing volume of data are collected by the sensing system, and transfer of these data to a central location for centralized processing poses burden to the communication system and the central computing system. Edge computing, a distributed computing paradigm, processes data and makes proper decisions locally, stores data locally and provides selected, processed data to a higher level, and thus may significantly relieve communication burden and reduce response time of certain control applications. This paper explores possible applications of edge computing to enhance distributed optimization and control of smart grid, including power system asset management, distributed charging scheme and microgrid protection.
Session Chair

Mashid Naeini (USF), Hana Khamfroush (Univ. of Kentucky) <br> Zoom Room Host(s): Andrea Pinceti (ASU)

Session WMLPDS2

Workshop on Machine Learning and Big Data Analytics in Power Distribution Systems

Conference
5:30 PM — 6:00 PM UTC
Local
Nov 13 Fri, 11:30 AM — 12:00 PM CST

Control Design for DER Smart Inverter Cybersecurity

Daniel Arnold (Lawrence Berkeley National Laboratory, USA)

1
The proliferation of Distributed Energy Resources (DER), such as solar photovoltaic inverters, in the electric grid constitutes a potential vulnerability that can be exploited to disrupt normal system operations during a cyber-attack. Specifically, systems designed to update settings in DER via internet or cellular connection can be exploited to introduce large oscillations in system voltages, or create large voltage imbalances. This talk outlines a line of research under study at Lawrence Berkeley Lab, in which we leverage adaptive control and reinforcement learning, to address this problem. In the event that the remote update capability for subsets of DER (perhaps from the same manufacturer) is compromised, our work specifically focuses on adapting the behavior of non-compromised DER smart inverters to mitigate the effect of certain cyber attacks in real time.

Deep Reinforcement Learning for DER Cyber-Attack Mitigation

Ciaran Roberts (Lawrence Berkeley National Lab & UC Berkeley, USA); Sy-Toan Ngo (Lawrence Berkeley National Laboratory, USA); Alexandre Milesi (Lawrence Berkeley National Lab, USA); Sean Peisert and Daniel Arnold (Lawrence Berkeley National Laboratory, USA); Shammya S Saha, Anna Scaglione and Nathan G Johnson (Arizona State University, USA); Anton Kocheturov and Dmitriy Fradkin (Siemens Corporation Corporate Technology, USA)

2
The increasing penetration of DER with smart-inverter functionality is set to transform the electrical distribution network from a passive system, with fixed injection/consumption, to an active network with hundreds of distributed controllers dynamically modulating their operating setpoints as a function of system conditions. This transition is being achieved through standardization of functionality through grid codes and/or international standards. DER, however, are unique in that they are typically neither owned nor operated by distribution utilities and, therefore, represent a new emerging attack vector for cyber-physical attacks. Within this work we consider deep reinforcement learning as a tool to learn the optimal desired behavior for a set of uncompromised DER units to actively mitigate the effects of a cyber-attack on a subset of network DER.
Session Chair

Deepjyoti Deka (LANL), Yu Zhang (UC Santa Cruz) <br> Zoom Room Host(s): Nima Taghizhpourbazargani (ASU)

Session Tutorial-3

Accelerating AI on the Grid: A Hands On Tutorial on PMU Data Analysis

Conference
6:00 PM — 6:30 PM UTC
Local
Nov 13 Fri, 12:00 PM — 12:30 PM CST

Accelerating AI on the Grid: A Hands On Tutorial on PMU Data Analysis

Alexandra von Meier (UC Berkeley), Kevin Jones (Dominion Energy), Laurel Dunn (UC Berkeley), Mohini Bariya (UC Berkeley), Miles Rusch (UC Berkeley), Sean Murphy (PingThings)

2
This tutorial will prepare attendees to analyze PMU (synchrophasor) data for research and practical applications. The tutorial will provide hands-on experience with state-of-the-art tools for digesting and visualizing high-frequency time series data, and for exploring novel applications. PMU data give empirical evidence of physical phenomena that happen on time scales unobservable to conventional sensor networks. Effective use of PMU measurement data can unlock novel opportunities for using Artificial Intelligence (AI) to extract insights into the condition of the grid. Making these insights accessible to decision makers in real-time has already begun to radically change best practices in grid operations, maintenance, and planning. This tutorial will provide attendees with the context and skills they need to leverage AI methods to begin using PMU and other high frequency data in their own work. The course will begin by teaching fundamental concepts from power systems engineering, and their relation to PMU measurement data. This talk will provide context necessary for both newcomers and domain experts to begin analyzing and interpreting PMU data. The course will go on to describe the data analytics program at Dominion, where streamlined access to PMU data has unlocked unexpected opportunities to improve decision-making processes related to grid operations, maintenance, and planning. Finally, we describe how companies that are successful at leveraging data and AI have radically changed the way they do business. We discuss examples from other sectors, such as Amazon and Google, and will share an outlook for similar transitions in the energy sector. Attendees will gain hands-on experience working with PMU data and state-of-the-art computational tools designed to facilitate the analysis and interpretation of big data. The hands-on portion of the session will use the National Infrastructure for AI on the Grid (NI4AI) powered by PingThings’ PredictiveGrid Platform and will provide real-time support for participants to gain API access to PMU and other data that are publicly hosted on the platform. Talks will include an interactive exercise for participants to familiarize themselves with the visualization capabilities of the platform, and to access the data on their personal computers using the Python API. Talks will include live coding demonstrations of two use cases for PMU data, to examine voltage sag events and explore the relationship between solar generation and voltage or frequency on the grid. Participants are requested to bring personal computers that they may follow along and replicate analytics on their own devices.
Session Chair

Zoom Room Host(s): Ezzeldin Shereen (KTH)

Session SIFI

Student Industry Faculty Interaction

Conference
6:30 PM — 7:20 PM UTC
Local
Nov 13 Fri, 12:30 PM — 1:20 PM CST
Session Chair

Anamitra Pal (ASU) <br> Zoom Room Host(s): Andrea Pinceti, Nima Taghizhpourbazargani, Rajasekhar Anguluri, Antos Varghese (ASU)

Made with in Toronto · Privacy Policy · © 2022 Duetone Corp.