DATA-DRIVEN SUSTAINABLE ENERGY SYSTEMS:
From Data Collection to Information Technology for Sustainable Services
Prof. Marija Ilic (Carnegie Mellon University, Electrical & Computer Engineering and Engineering Public Policy Departments)
Prof. Richard Edward Schuler, Cornell University, Department of Economics/School of Civil & Environmental Engineering
Prof. Richard Edward Schuler, Cornell University
Dr. Bob Havemann, SRC
Dr. Krishna Kant, NSF
Prof. Jose Moura, ICTI
Leonard Hyman, Senior Advisor, Black & Veatch
Date: March 12, p.m.
Location: Carnegie Mellon University
Distance Learning Room
Hamerschlag Hall, Room 1107, CIT
Dates: March 13 and 14, 2012
Location: Carnegie Mellon University
Distance Learning Room
Hamerschlag Hall, Room 1107, CIT
This conference focuses on the role data might play in enabling future energy services. Recent advances in communications, sensing, computing and control, as well as the targeted investments toward deploying advanced meter infrastructures (AMIs) and synchrophasors have become drivers and sources of data previously unavailable in the electric power industry. This situation presents the community with major challenges and opportunities. The opportunities are to rely on just-in-time (JIT), just-in-place (JIP) and just-in-context (JIC) key information for flexible adaptation of available energy resources and users to the ever-changing hard-to-predict uncertainties. One of the key premises in this workshop is that IT-enabled flexible utilization of assets is key to sustainable energy services. However, the challenge is enormous as the industry paradigm shifts from the predictive, worst-case centralized data management to highly distributed interactive data and resource management. Invited papers by the academic, industry and government participants will be presented. They are intended to introduce fundamental approaches to transforming today's operating and planning industry practices, and the supporting policies, to manage such short-and long-term uncertain problems to ensure sustainable electricity services. Of particular interest are papers which offer frameworks for systematic use of IT and quantifiable performance enhancements in sustainable energy services.
The Carnegie Mellon Conference on the Electricity Industry is now in its eight year of offering networking on timely topics. This is the first conference without late Lester Lave, who together with Marija Ilic started it all. Semiconductor Research Corporation (SRC) as part of its Energy Research Initiative (ERI) outreach activities http://www.src.org/program/eri/ and the National Science Foundation (NSF) http://www.nsf.gov/ will be conference co-sponsors and co-organizers of the 8th CARNEGIE MELLON UNIVERSITY Electricity Conference. Invited academic speakers will be partially compensated for their travel expenses by the National Science Foundation. All presentations and white papers will be archived as part of the CARNEGIE MELLON UNIVERSITY Electricity Conference Series (see Archives).
Motivation for the Conference Theme:
The electric energy industry is at its crossroads. New unconventional resources are being deployed within the electric power grids which were not designed for accommodating such resources in order to meet the social objectives of deploying clean resources and reduce the environmental footprint in the years to come.
These resources are qualitatively different than the conventional power plants and require rethinking of today’s operating and planning practices. In addition, the overall complexity of the interconnected power systems does not lend itself well to the blue print standards and rules. Integration can be done in qualitatively different ways, and, depending on how it is done, the asset utilization, impact on customers and society will be drastically different. More near-term, it is essential to have business incentives which are aligned with the objectives which must be met and the valuation of candidate technologies. As in many other industries, tremendous progress in information technology is likely to change the way electricity is provided in previously unseen ways.
As unconventional technologies (many of them new) get connected to the electric power grid, and the system is driven by qualitatively new economic and environmental objectives, new phenomena will emerge. Some of these may lower cost and improve reliability, but some will challenge the stability and reliability of the electricity system. For example, if a large proportion of residential and commercial customers have solar panels that feed electricity into the grid when the sun is shining, but abruptly require power when a cloud passes over the sun, the dynamic stresses on the grid will increase, perhaps to unmanageable levels. A system based on minimizing costs with current rules for preserving stability and reliability could lead to unacceptable service. Before making these changes in technology and management practices, it would be prudent to understand the challenges that the current system faces, the reasons for its relatively good record of reliability, and how the system will be challenged by the changes.
Similarly, an effort toward planning for integrating many wind power plants, creates difficult challenge and opportunity. Tradeoffs between multiple objectives, such as business, economic and environmental impacts, on one side, and robust services, on the other, must be quantified, and supported by the right IT methods and protocols. The IT methods for managing huge uncertainties in a distributed way across energy providers, users and delivery providers and over multiple time horizons, ranging from near-real-time to very long-term horizons, must be formalized.
The idea for this conference is to provide a venue for structured brainstorming by the thought leaders in multiple communities. The underlying theme is the one of data-driven challenges and opportunities.
The Preliminary Program is organized this year by invitation only and it is targeted to answer several key questions as listed below. Other university, industry and government participants are invited to attend and participate in discussion. They are also welcome to send 1-2 page white papers which will, upon light review be posted as archival conference documents.
Pre-Conference Workshop: Engineering IT-Enabled Sustainable Electricity Services
Marija Ilic, Ed Schlesinger, and Jose Moura
Carnegie Mellon University (CMU), Semiconductor Research Corporation Energy Research Initiative (SRC ERI) and Information and Communication Technologies Institute (ICTI)
Location: Carnegie Mellon University - CIT/ECE (to be broadcast)
Date: March 12, 2012
Hamerschlag Hall 1107, Bombardier Smart Infrastructure Collaboration Center
Workshop Registration: $100
Free for SRC ERI members, ICTI members, and NSF-SRC Conference Presenters
This half day Pre-Conference Workshop will present the vision pursued under the umbrella of SRC Energy Research Initiative (SRC ERI). This is a very technical workshop which offers formal problem posing for modeling, analysis and decision making in support of sustainable IT-enabled electric energy services. This half-day workshop is organized with several objectives in mind. First, it will present the basic framework pursued within the SRC Smart Grid Research Center led by CMU. This framework is referred to as the Dynamic Monitoring and Decision Systems (DYMONDS) –based Approach to Smart Grids as enablers of sustainable electric energy services. As explained in [1,2], it is indeed possible to pose the objectives of designing smart grids so that these become the key enablers of implementing sustainable electric energy services as new technologies emerge.
The second objective is to demonstrate potential meaning of such framework for making island-type systems green without actually increasing the cost of services. These systems are particularly interesting because demonstrating different concepts is more transparent than in very large continental interconnected power systems. We have been fortunate enough through the ICTI activities to obtain the realistic data for two islands in the Azores Archipelago in Portugal and have spent many hours posing, simulating, analyzing end demonstrating the effects of DYMONDS-based framework using this data. Since we have obtained a permission to publicly post the data, we have prepared a monograph to be published by Springer which is organized to illustrate the basic operating and planning functions using this data, and to compare the outcomes based on systematic models and simulations. Part I of this workshop will summarize our models and findings. As such, this workshop is also a rollout event for our soon-to-appear Springer Monograph entitled: Engineering IT-Enabled Electricity Services: The Case of Low-Cost Green Azores Islands . As part of this workshop we will describe the basic Data Repository for Demonstrating Smart Grid Concepts.
Finally, since most of the co-authors of chapters in this monograph are at the same time active researchers in the SSRC SGRC, this will be an opportunity for Carnegie Mellon University and Semiconductor Research Corporation to present their activities. More than anything else, we offer this workshop as a venue for Carnegie Mellon University to demonstrate its approach to supporting and catalyzing industry innovation in the areas of smart grids and sustainable electric energy services. The emphasis will be on the problem formulations, their relevance, and demonstrations of using DYMONDS-based concepts in comparison with what is possible today. The detailed know-how algorithms are limited to current SRC ERI members. We wish to engage in close collaborative activities with the power utilities and pursue potential use of the algorithms under development for their systems. We also wish to draw attention of the on-going work to the technology developers in the area of smart grids.
Part I: The Case of Low-Cost Green Azores Islands
Part I of this workshop will summarize the results of the upcoming Springer Monograph co-edited by Marija Ilic and Le Xie. First the problem of predictions and look-ahead decision making for scheduling generation and demand in systems with large amounts of wind power will be posed. Simulations using realistic data for two Azores Islands will be presented to show a comparison between one’s ability to integrate wind power using today’s scheduling approaches and the proposed Dynamic Monitoring and Decision Making (DYMONDS)-based approach.
Second, the problem of balancing supply and demand within each scheduling interval so that frequency and voltage meet required industry standards is posed. We summarize an approach to enhanced Automatic Generation Control (E-AGC) and Enhanced Automatic Voltage Control (E-AVC) and illustrate how these approaches can be used to ensure no frequency or voltage problems despite persistent fast wind power fluctuations. The complexity of communications and control design, and potential use of synchrophasors for implementing E-AGC and E-AVC is illustrated using system and input data from the Azores Islands.
Third, the role of fast storage and power-electronically controlled FACTS in ensuring no stability problems during large sudden wind power gusts and/or system faults is illustrated. A comparison of system response to these sudden disturbances with an without fast automated storage control is illustrated for two Azores Islands. First insights regarding dynamics of power systems with all synchronous machines and with a mix of synchronous and induction type wind power plants is discussed. The role of electric vehicles, flywheels and other fast-responding storage in ensuring acceptable system performance during normal operations in systems with considerable wind power penetration is highlighted and illustrated using simulations.
Finally, a new approach to interactive multi-temporal planning and investments for selecting and deploying new technologies in the evolving future electric energy systems is posed and illustrated for the case of Azores Islands.
Part II: Lessons Learned for Transforming Continental T&D System Into an Enabler of Sustainable Electric Energy Systems
Part II of this workshop will summarize lessons learned through modeling, analyzing and designing decision making and automation algorithms for supporting penetration of clean power in technically and economically sustainable ways. Next, fundamental principles of today’s approach to operations and planning of electric power systems are described with the emphasis on assumptions which no longer hold and which require evolution of today’s methods in order for the T&D systems to enable integration, and delivery of large-scale wind power across large electric distances. Finally, a possible DYMONDS-based approach to implementing next generation SCADA in support of operating and planning deployment of new technologies at value is outlined. The importance of considering this framework is stressed by discussing possible outcomes with and without enhanced operating and planning framework for continental large electric power interconnections, such as Eastern and Western Interconnection in the United States.