GLORA, a turn-key HPC Solution for the Locational Reliability Assessment of the Electrical Grid

Statement of the Problem or Situation that is Being Addressed. This project focuses primarily on resource adequacy, a critical component of power system reliability which sets design and long-term planning criteria for the type, level and location of system reserves to ensure the feasibility of reliable operation of the power system. Resource adequacy is a critical driver of power system planning processes, and is essential to the economic and social well-being of the economy. Existing methods and tools address these problems at a broad regional level and are incapable of providing precise reliability information at the locational level resulting in inefficient system expansion and investment decisions. Statement of How this Problem or Situation is Being Addressed The proposed project will implement GLORA, a cloud-based hybrid CPU/GPU software tool for locational reliability assessment of large bulk electrical systems such as North American Interconnections. This system will provide a turn-key solution for assessing the current and future reliability of the electrical grid at an unparalleled level of detail. GLORA will compute locational reliability indices through a Monte-Carlo based probabilistic modeling of the power system by simulating random outages of generating units and transmission branches, fuel supply interruptions, stochastic behavior of electricity consumption, performance of demand response resources, and of variable renewable generators such as wind and solar. What is to be Done in Phase I During Phase I, Newton Energy Group will prototype GLORA on basis of the pCloudAnalytics platform, implemented on the Amazon cloud and developed by Newton Energy Group in collaboration with Polaris Systems Optimization, Inc. GLORA will be capable of performing a locational reliability assessment of a large power system (with 10,000 15,000 nodes) within 1-2 hr time from start to finish. GLORA will be implemented on a commercial cloud incorporating hybrid GPU/CPU machine instances. Phase I efforts will focus on the development of novel algorithms and modifications of the pCloudAnalytics architecture suitable for their implementation. Commercial Applications and Other Benefits Development of GLORA will create significant technical, economic and social benefits to the power industry, to all of industry stakeholders and to the general public as everyone consumes electricity. GLORA users will include System Operators and Regional Transmission Organizations, transmission and generation owners and developers, engineering and consulting companies and federal and state government and regulatory agencies. Users would be able to precisely and unambiguously identify the least reliable elements of power systems they operate and obtain locational signals on where investments are needed. GLORA will make it possible to explicitly address one of the thorniest issues of today: how to adequately compare resource-based and transmission-based solutions to resource adequacy. Key words. Power systems, resource adequacy, Monte-Carlo analysis, high performance computing, high scalability computing, cloud computing Summary for Member of Congress. The project will use cloud computing to provide locational assessment of the reliability of the electrical grid. This will help to design and expand generation and transmission components of power systems with greater efficiency avoiding unnecessary investments and reducing consumer costs.