PNNL

Abstract

Transitioning to net-zero electricity requires extensive renewable energy resources to be integrated with the grid. It might cause unexpected voltage and frequency violation problems that can damage the existing power system configurations. Identifying the zones containing more critical violations in a bulk system under different contingencies is crucial to securing the power system operation and infrastructure safety. This paper presents an advanced critical zone identification framework that can retrieves information from extensive data yielded by dynamic contingency simulations from a bulk interconnection power grid. A machine learning (ML) based clustering approach is applied to identify the critical zone under 213 scenarios. The developed framework is tested in the 2028 western electricity coordinating council (WECC) system, and the detailed results are discussed in this paper.