NINTH ANNUAL CARNEGIE MELLON CONFERENCE ON THE ELECTRICITY INDUSTRY

Role of Distributed Coordination in Resilient
& Fine-Grain

Control of Power Grids

Control of Power Grids

Conference Dates: Monday-Wednesday, February 3-5, 2014

Location: Carnegie Mellon University

Roberts Engineering Hall (REH) - Singleton Room

OPENING RECEPTION - Monday, February 3, 2014

POSTER SESSION / DINNER, - Tuesday, February 4, 2014

**2014 Abstracts:**

**1) Title: A Framework for Actuator Placement in Large Scale Power Systems: Minimal Strong Structural Controllability**

Presenter: Sergio Pequito, <spequito@andrew.cmu.edu>, Nipun Popli, <nipun@cmu.edu>

Advisor: Soummya Kar, <soummyak@andrew.cmu.edu>, A. Pedro Aguiar,

Marija Ilic, <milic@ece.cmu.edu>

Abstract: We address the problem of minimal placement of actuators in large-scale linear time invariant (LTI) systems, such as large-scale power systems, for dynamic controller design. A novel sufficient and necessary condition to ensure a strong structurally controllable (SSC) system is proposed. Specifically, the paper addresses the problem of obtaining the minimal number of dedicated inputs, i.e., inputs which actuate only a single state variable, and the respective state variables they should be assigned to, such that the LTI system is SSC. In addition, an efficient and scalable algorithm, with polynomial implementation complexity, to achieve such minimal placement of dedicated inputs is proposed. An illustration of the proposed design methodology is provided on the IEEE 5-bus test system, thereby identifying the minimal number of physical state variables to be actuated for ensuring strong structural controllability.

**2) Title: Cooperative Control of Distributed Energy Storage Systems in a Microgrid **

Presenter: Yinliang Xu, <danielxu@andrew.cmu.edu>

Abstract— Energy storage systems have been widely utilized for frequency support in microgrid systems. Due to the intermittency of the renewable generation and constant changing load demand, the charging/discharging of various energy storage systems in an autonomous microgrid need to be properly coordinated to ensure the supply-demand balance. Recent researches find out the charging/discharging efficiency of the energy storage systems has remarkable dependence on the charging/discharging rate and state of charge. This paper proposes a distributed cooperative control strategy for coordinating the energy storage systems to maintain the supply demand balance and minimize the total power loss associated with charging/discharging efficiency. The effectiveness of the proposed approach is validated with simulation results.

**3) Title: Designing Decentralized Control Systems without Structural Fixed Modes: A Multilayer Approach**

Presenter: Sergio Pequito, <spequito@andrew.cmu.edu>, Clarance Agbi <cagbi@ece.cmu.edu>, Nipun Popli, <nipun@cmu.edu>

Advisor: Soummya Kar, <soummyak@andrew.cmu.edu>, A. Pedro Aguiar,

Marija Ilic, <milic@ece.cmu.edu>

Abstract: We propose a methodology to design decentralized controllers applied to large-scale systems. The key idea is to split the design into two control layers. The first layer, consists mainly in a pre-processing step, where an optimal subset of inputs and outputs are used for feedback to close the loop, such that the resulting associated state matrix of the closed-loop system has two desirable structural properties: it has no structural fixed modes and it is structurally controllable and observable thru any single state variable. After this first layer, we can select an arbitrary decentralization control scheme to achieve some specified performance, which is the second layer of the proposed approach. We illustrate this methodology in frequency regulation of a 3-bus system.

**4) Title: Distributed Control for Secondary Frequency Control**

Presenter: Chenye Wu, <chenyewu@andrew.cmu.edu>,

Advisor: Gabriela Hug, <ghug@ece.cmu.edu>, CMU

**5) Title: Distributed Power Flow for and Distributed Wide-Area Situational Awareness (D-WASA) to Ensure Feasible Power Delivery**

Presenter: Andrew Hsu, <andrewhsu@cmu.edu>

Advisor: Marija Ilic, <milic@ece.cmu.edu>

**6) Title: Intelligent Partitioning in Distributed Optimization of Electric Power Systems**

Presenter: Junyao Guo, < junyaog@andrew.cmu.edu >

Advisor: Gabriela Hug, <ghug@ece.cmu.edu>, Ozan Tonguz, <tonguz@ece.cmu.edu>

Abstract: Methods for distributed optimization are important enablers and indispensable tools to handle the increased control complexity in the future electric power systems. However, the performance of such methods depends heavily on the chosen partitioning of the system into subsystems, especially on the number of subsystems and which bus should be assigned to which subsystem. There has only been little effort on how to optimally partition the grid, which is an important decision for the implementation of distributed optimization.

This poster presents an intelligent partitioning method for electric power systems that significantly reduces the convergence time of decomposition algorithms implemented on the resulting partitioned system to solve a distributed optimal DC power flow problem. First, we define multidimensional criteria to systematically evaluate the performance of distributed methods considering metrics such as convergence time, amount of required information exchange, and robustness. Then focusing on one of the metrics-convergence time, we show that the convergence time of decomposition algorithms heavily depends on the chosen partitioning of the system. Finally, an intelligent partitioning method based on spectral clustering is proposed which computes the optimal partition of the system that enables the decomposition algorithms to converge in fewest iteration steps.

**7) Title: Modeling and Simulation of Heterogeneous Smart-Grid Platforms in Ptolemy II**

Presenter: Ilge Akkaya <ilgea@eecs.berkeley.edu>

Advisor: Edward A. Lee

Abstract: Power grids are being integrated with sensors that provide measurements in high rates and resolution. The abundance of sensor measurements, as well as the elevated complexity of required applications generate a demand for pre-deployment system simulation to evaluate the characteristics of appropriate network fabrics, realistic timing profiles and distributed application schemes for power applications. Although simulation aids in the pre-deployment decision making process, even the system models for these complicated cyber-physical energy systems can easily diverge in complexity and become unusable for the purposes of specialized evaluation of aspects. Moreover, the interrelation between physical components, controllers and clock synchronization aspects need to be studied for the system hierarchy. We present a modeling formalism in Ptolemy II that aims at presenting solutions to fundamental modeling problems for heterogeneous composition of cyber-physical smart grid components. We present the aspect-oriented modeling and simulation paradigm that focuses to address this challenge by separating functional models from design-specific concerns such as network topology, latency profiles, QoS requirements and capacity and provide an analysis of how the abstraction can be used for evaluation and adaptation of distributed power applications, with the prominent example of distributed state estimation.

Bio: Ilge Akkaya received her B.S. degree in electrical engineering from Bilkent University in 2010 and is currently a Ph.D Candidate at University of California at Berkeley, supervised by Edward A. Lee. Her research focuses on aspect-oriented modeling and simulation of cyber-physical energy systems and fault modeling for safety-critical heterogeneous system design.

**8) Title: MDP Methods for Valuation of Multi-Purpose Battery Energy Storage under Uncertainty**

Presenter: Jonathan Donadee, <jdonadee@gmail.com>

Collaborator: Jianhui Wang, Computational Engineer, Argonne National Laboratory

Advisor: Marija Ilic, <milic@ece.cmu.edu>

Abstract:
In this work we propose using the modeling and computational techniques of the Markov decision problem (MDP) framework to estimate the net present value (NPV) of revenues earned by a grid-scale battery energy storage system (BESS) in a liberalized market setting. First, we propose an MDP to optimize hourly operational decisions under uncertainty and also while considering battery capacity degradation. The solution of this MDP is then used to estimate the rate at which the BESS earns profits and the rate at which the maximum energy storage capacity of the BESS degrades. Finally, we use our proposed methods to estimate the NPV of a BESS’s revenues.

**9) Title: Automated Modeling of Power System Dynamics for Nonlinear Control Using Flywheels during Large Disturbances**

Presenter: Kevin Bachovchin, <kbachovc@andrew.cmu.edu>

Advisor: Marija Ilic, <milic@ece.cmu.edu>

Abstract: Flywheel energy storage systems have great potential to be used in electric power systems for transient stabilization during large disturbances or faults. Since flywheels are best used for small-scale transient applications, it is necessary to first obtain the dynamic equations of the power system, so that control can be applied and tested. Non-linear dynamics and non-linear control are needed because with large wind disturbances, the system can be displaced far enough away from the equilibrium where linearized models are no longer accurate. An automated method was implemented for symbolically deriving the dynamic equations of general power systems using the Lagrangian formulation from classical mechanics, given user input about the power system topology. To improve the computational efficiency, a modular automated procedure was also implemented. Using this approach, the system is divided into modules, the state space model for each module is determined separately, and then the state space models from each module are combined in an automated procedure. To stabilize power systems against large wind disturbances using flywheels, sliding mode control is used. Sliding mode control is a non-linear control method where the feedback control law switches from one continuous function to another. Finally it is demonstrated, using the real-life Flores Island power system, that sliding mode control with flywheels can be used for frequency stabilization against large wind disturbances.

**10) Title: Module-based Interactive Modeling and Control of Nonlinear Power
System Dynamics; The Case of Marcy Station, NY**

Presenter: Milos Cvetkovic, <mcvetkov@andrew.cmu.edu>

Advisor: Marija Ilic, <milic@ece.cmu.edu>

Abstract: Many real-world installations of Flexible Alternating Current Transmission Systems (FACTS) are still being used as static compensation devices. In rear cases in which they are controlled dynamically, their controllers are often tuned to improve dynamic response of the local states. Their impact on stabilization of interconnected power systems is still insufficiently explored. In this work, we study the effect of a multi-configuration FACTS device (such as the one at Marcy station, NY) on the stability of a power system under large disturbances. We use a module-based framework to model the system and design the controllers with provable performance.

**11) Title: Optimal Storage Sizing using Two-Stage Stochastic Optimization**

Presenter: Kyri Baker, <kabaker@andrew.cmu.edu>

Advisor: Gabriela Hug, <ghug@ece.cmu.edu>

**12) Title: Regression-based corrective power flow control for system risk minimization**

Presenter: Rui Yang, <ruiy@andrew.cmu.edu>

Advisor: Gabriela Hug, <ghug@ece.cmu.edu>

**13) Title: Robust Data-Driven State Estimation in Smart Grids**

Presenter: Yang Weng, <yangweng@andrew.cmu.edu>

Advisor: Marija Ilic, <milic@ece.cmu.edu>, Rohit Negi, <negi@ece.cmu.edu>

**14) Title: Rural Microgrids**

Presenter: Janak Gahlot, <jgahlot@andrew.cmu.edu>,

Sidharth Choudhary, <sidhartc@andrew.cmu.edu>

Advisor: Marija Ilic, <milic@ece.cmu.edu>

**15) Title: Toward Reconfigrable Smart Distribution Systems for Differentiated Reliability of Service**

Presenter: Siripha Junlakarn <sjunlaka@andrew.cmu.edu>

Advisor: Marija Ilic, <milic@ece.cmu.edu>

**16) Title: How Does Wind Power Distort a Electricity Market if The Dispatch Protocol and Transmission Policies Are not Correspondingly Reformed?**

Presenter: Yang Yu <yangyu1@stanford.edu>, Baosen Zhang, Ram Rajagopal <ramr@stanford.edu>. Stanford Sustainable System Lab

**18) Title: Role of Communication on the Convergence Rate of Fully Distributed DC Optimal Power Flow**

Presenter: Javad Mohammadi <jmohamma@andrew.cmu.edu>

Advisor: Gabriela Hug <ghug@ece.cmu.edu>, Soummya Kar <soummyak@andrew.cmu.edu>

**19) The title of the poster is: "Soft Magnetic Nanocomposites for High Frequency Energy Conversion."**

Presenter: Vincent DeGeorge

Abstract/Bio: Vincent DeGeorge is a doctoral candiate in Materials Science and Engineering whose research in magnetism focusses on novel magnetic materials for use in power grid and electric motor applications. By optimizing these materials' unique magnetic, electrical, and structural properties and limiting high frequency power losses more efficient and more energy dense power converters and electric machines can be integrated into the power grid and electric/hybrid vehicles. Vincent is from Morgantown, WV and received his bachelor's degree in Physics from John Carroll University, Cleveland, OH.

**20) Title: Decomposition Methods for Stochastic Optimal Coordination of Energy Storage and Generation??**

Presenter: Dinghuan Zhu <dinghuan@cmu.edu>

Advisor: Gabriela Hug.<ghug@ece.cmu.edu>