Towards a Resilient Grid: A Risk-Based Decision Analysis Incorporating the Impacts of Severe Weather-Induced Power Outages

Sayanti Mukhopadhyay, Purdue University

Abstract

Severe weather power outages cause devastating damages affecting tens of millions of people and costing tens of billions of dollars of economic loss annually. It is estimated that these outages have cost the U.S. economy an inflation-adjusted annual average of 20 billion USD to 55 billion USD during the period of 2003–2012. The National Association of Regulatory Utility Commissioners (NARUC) have recently highlighted the importance of resilience building in the electricity vis-a-vis minimizing the risk of extreme events on the electricity sector, and thereby enhancing electricity service security and long term economic benefits. However, the federal and state-level risk and reliability metrics and standards for both electricity transmission and distribution networks do not include the impact of extreme events on electric power service, and have no effective mechanisms for assessing and regulating the levels of preparedness and response of the utilities impacted by extreme events. In other words, the existing framework to assess the benefits and costs of the investments in the electricity sector grossly undervalues the economic and societal costs associated with severe weather power outages. To address this gap, this research developed a risk-based decision support framework to assist regional and local planners and reliability and regulatory bodies to foster utility capital investments that minimize severe weather risks on electricity sector. The major objectives of this research are: (1) identify risk factors to severe weather-induced power outages affecting end-use consumers; (2) probabilistic estimation of economic loss due to a 100-year major power outage event leveraging the value of load loss concept; and (3) a framework for investment strategy selection based on reduced economic loss in the post-strategy implementation scenario compared to the traditional practice. First, the study conducted an exploratory analysis of the nationwide major power outage data during 2000–2015 to understand the patterns of the major power outages that occurred in the past. The results obtained from this analysis concluded that the U.S. electricity sector is vulnerable to failure in face of severe weather events with 52.9 percent of the major power outages caused owing to the severe weather and climate events. The exploratory analysis was followed by investigating the key predictors that increases the risks of severe weather-induced sustained major power outages using a multi-hazard approach. A two-stage hybrid risk estimation model was developed leveraging algorithm-based advanced data-mining techniques. Publicly available historical data on major power outages, socio-economic data, state-level climatological information, electricity consumption patterns and land-use data was used to train the models. The model analyzed state-level power outage data and considered all the states in the continental U.S. for identifying the major risk factors. Finally, the risk-based decision support system (RDSS) for strategy selection at the state-level was developed using the state of Texas (TX) as a case study. The different steps involved include: (i) identify the key risk predictors and their influence on the state electricity sector using the two-stage hybrid risk estimation model; (ii) identify strategies that will reduce the risks of the severe weather power outages; (iii) probabilistically estimate the risk of a major power outage that has a 1.0 percent chance to occur in the next year, i.e., in other words, estimate the size of a 100-year major power outage event under all the pre- and post- strategy implementation scenarios; (iv) assess the economic loss considering the value of load loss in face of the 100-year event and calculate the reduced loss due to implementation of the strategies. Thus, the RDSS will provide a platform to the state regulators to select strategies that would minimize the major power outage risk and reduce the economic loss in face of a major power outage event. This novel approach of considering the risks of severe weather induced power outages and proposing a method to include the economic costs of such outages in the regulatory decision-making process will help in the resilience enhancement of the grid.

Degree

Ph.D.

Advisors

Hastak, Purdue University.

Subject Area

Civil engineering|Energy|Operations research

COinS