Critical Infrasctructure

Smart Grid

Smart Grid Adoption and Conservation Behavioral Analysis: Critical to the success of the Smart Grid initiative is consumer acceptance of time-of-use pricing schedules. Although conventional economic wisdom suggests that consumers will adjust their behavior when opportunities to save money are presented to them, a number of pilot studies testing demand-side responses to dynamic pricing have shown that the savings available to many consumers is not sufficient to induce significant changes in their consumption habits. It is to this end that our research has turned to examining other factors, which may play a role in shaping the way utility customers use electricity. Specifically, we are investigating the influence that such factors as concern for the environment, concerns about national security, social learning, and competitive behavior might exert over how much, when, and where consumers are using electricity. It is our aim to use this information to develop usage feedback techniques in conjunction with smart metering technologies, which leverage these forces and influence consumers to embrace dynamic pricing and other emerging Smart Grid innovations. We have funding from NYSERDA to study energy conserving behavior of consumers in context of the Smart Grid.

Smart Grid Energy Modeling: Energy supply and demand need to match within a very narrow range for the power grid to stay stable. Since demand is constantly fluctuating, the supply has to be constantly adjusted to match the demand. The Smart Grid will facilitate integration of weather-dependent renewables (wind and solar) into the supply necessitating any imbalance in the supply due to weather variability to be balanced through use of fossil-fuel based generation or through demand management. In addition to the supply, demand is also impacted by user behavior, which adjusts based on weather. Finally, energy markets react to changes in supply and demand creating a complex set of interaction. The goal of the project is to model the variability in the supply-and-demand based on weather, user attitudes, and market behavior to improve forecasts for planners and operators.

Privacy in Smart Grid Data Analytics

The Smart Grid envisages integration of sensors and communication infrastructure into the existing power grid to enable control and management using operational intelligence that relies on real-time data collected from the grid. Data analytics has been widely integrated in Smart Grid applications. For instance, the current Smart Grid assumes that each node on the grid can produce or consume power. The goal is to balance supply and demand within a tight margin. If supply exceeds demand, there is a voltage spike; when the supply lags demand, the voltage sags. Both of these situations are detrimental to power grid operations and devices connected to the grid. Consumer metering data, including both generation and consumption data, is collected to facilitate the balance between supply and demand. The problem with creating this balance is variability both in the supply (from renewable sources such as wind and solar) and demand; however, forecasts can be made for both supply and demand via data analytics models that use a multitude of factors such as weather, usage history, and status of various fossil generators. The forecasts can be done at different time scales, short, medium and long. The balance between supply and demand can be achieved by two means: 1) increasing/decreasing supply through startup/shutdown (or partial load) fossil-fired generators that are quick to start; and 2) controlling user consumption through demand response via pre-signed contracts, user behavior modification based on price incentives or persuasive messages, or forced brownouts or blackouts. Each of these objectives relies heavily on the information collected from the grid, which in turn generates concerns related to data privacy.

These issues include grid state and operational information as well as consumer usage information. Breach of state and operational information can provide hackers and terrorists with the intelligence to exploit its vulnerabilities and destabilize the grid. This can be done in several ways such as injecting false operational data to force corrections that destabilize the grid or to create phantom anomalies that can misdirect the efforts of the repair crews. Consumers fear that fine-grained usage information, if exposed, could lead to revelation of their lifestyle making them vulnerable to discrimination, embarrassment, predatory marketing, and even federal and state investigations. Privacy issues related to the Smart Grid are very complex and broad involving the confluence of technology, human behavior, and computer science. The goal of the grid designers is to develop privacy-preserving schemes that would allow operational execution without compromising the privacy of the operational or personal information.

We specifically plan to create a series of privacy-preserving data analytics toolkits for various smart grid applications with measurable privacy guarantee. Such toolkits could advance energy management, customer relation management and quality of service in smart grid without compromising consumers' personal information and grid sensitive information. The proposed privacy-preserving schemes will significantly limit the privacy risks in different scenarios using measurable privacy notions and/or formal security analysis.


Impact of Security and Terrorism on Financial Markets

This stream of research involves studying the financial impact of security breaches incidents. The fundamental premise of this work is that financial markets are efficient and that their change reflects a fundamental impact to the economy. We investigated the impact of security breaches on the market valuation of the firms in the United States. Currently, we are extending the link between terrorist incidents and market valuations. This study transcends international boundaries and we want to examine how terrorism in a country can have financial impact on another country.


Resilient Transportation

The traffic project is an effort to implement principles of self-organization in traffic control. Instead of controlling traffic flow along the whole system through a centralized system, the goal is to promote optimal behavior at the local level that, in turn, will cascade into system-wide optimization. Work in this area has been funded by the University Transportation Research Center Region II, NYSERDA, and the James S. McDonnell Foundation. We are working with the City of Albany's Traffic Engineering Services, which is a part of the City of Albany Police Department to implement our self-organizing system onto several corridors in the City of Albany. The first use case, we are examining is the direct installation of our algorithms onto the traffic controller itself. The second and subsequent methods involve the incorporation of a secondary microprocessor, a RaspberryPi, onto the existing hardware controllers in order to provide additional analysis and decision-making power in order to apply our self-organizing algorithms.


Resilient Service Oriented Architeture

This work involves investigation of service-based architectures for executing complex transactions on the network. The architecture draws on concepts from grid computing as well as Web Services to create a distributed architecture on a diverse computational grid. This architecture consists of autonomous services that reside on the network and are discovered in real-time to create a federation of services. This federation then executes complex transactions, comprising of multiple tasks, with given precedence relationships. We compare service-based distributed architecture with the traditional client-server architectures for engineering analysis. We also investigate optimization models for selection of services in a distributed service-based architecture. Two specific problems are investigated, that is, a mortgage problem and a supply chain problem for turbine components. A directed acyclic graph is used to represent the transactions. Four separate problems are being investigated: 1) Service Provider Algorithm- finding the optimum set of services subject to the constraint of overall completion time. 2) Cover Problem- if the services are bundled by vendors in the supply chain; can we find a set of providers with bundling constraints such that just one vendor covers all the services? 3) Optimum Service Selection- when the services have additional associated constraints on trust, cost, quality, etc. 4) Service Location- if proximity of a service provider is one of the criteria for selection of services, can the optimum location of the service in a network grid be computed?
figura reserach

Figure 1. Diagram showing operation of the service broker and service provider.


Related Publications

  1. Goel, S., Bush, S. F. & Bakken, D. (Eds.). (2013). IEEE Vision for Smart Grid Communications: 2030 and Beyond. IEEE Press, pp. 1-390. (B)
  2. Goel, S., Talya, S., & Sobolewski, M. (2005). Service-Based P2P Overlay Network for Collaborative Problem Solving. Decision Support Systems. 45 (2), pp.
  3. Goel, S., & Pon, D. (2005). Distribution of Patches within Vulnerable Systems: A Distributed Model. In the Proceedings of the 6th IEEE Information Assurance Workshop, USMA, West Point, NY. (C)
  4. Goel, S., Talya, S.S., & Sobolewski, M. (2005). Preliminary Design Using Distributed Service-Based Computing. Accepted for publication May 2005 in the Proceedings of the 12th ISPE International Conference on Concurrent Engineering: Research and Applications, Fort Worth/Dallas, TX. (C)
  5. Goel, S., & Sobolewski, M. (December 2003). Trust and Security in Enterprise Grid Computing Environment, Proceedings of the IASTED Conference, New York City. (C)
  6. Goel, S., & Gangolly, J. (August 2003). Model for Trust Among Peers in Electronic Multiparty Transactions, Proceedings of the AMCIS Conference. (C)
  7. Rosenkrantz, D., Goel, S., Ravi, S.S., & Gangolly, J. (2005). Structure-Based Resilience Metrics for Service-Oriented Networks. Accepted for publication April 20-22 2005 in the Proceedings of the 5th European Dependable Computing Conference, Budapest, Hungary. (C)
  8. Goel, S., Belardo, S., & Iwan, L. (2004). A Resilient Network that Can Operate Under Duress: To Support Communication between Government Agencies during Crisis Situations, Hawaii International Conference on System Sciences, HW. (C)
  9. Sanjay Goel & Shobha Chengalur-Smith, An Innovative Approach to Security Policy Metric Development: A Foundation for Research in Security Policy Management, Soft-Wars December 11, 2005, Imperial Palace, Las Vegas, NV. (C)