Transforming power grid operations via high performance computing

Abstract

Past power grid blackout events revealed the inadequacy of grid operations software when handling adverse situations, to a large degree due to its low computational efficiency. High performance computing (HPC) promises a solution to this problem. In addition, the HPC technology becomes necessary to take advantage of parallel computing platforms as the computer industry is undergoing a significant change from the traditional single-core processors to multi-core based computing platforms. HPC can improve todaypsilas key grid operation functions like state estimation and contingency analysis and reduce the solution time from minutes to seconds, comparable to SCADA measurement cycles. HPC also enables the integration of dynamic analysis into real-time grid operations. Dynamic state estimation, look-ahead dynamic simulation and real-time dynamic contingency analysis would be three crucial dynamic functions in future control centers. HPC provides opportunities for developing better decision support tools, which also need HPC support to handle large volume of data and large number of cases. Given the complexity of the grid and the sheer number of possible configurations, we argue that HPC will be an indispensible element in the next generation control centers.