Multiagent Stochastic Simulation of Minute-to-Minute Grid Operations and Control to Integrate Wind Generation Under AC Power Flow Constraints

Abstract

The variability and uncertainty inherent to wind generation, which is rapidly increasing, could significantly impact operations efficiency in the future, particularly frequency regulation reserves. This paper addresses these issues from both analytical and curative standpoints through operational impact studies which combine a transmission grid representation with a distributed agent-based control architecture that mimics industry organization charts and follows NERC reliability management rules. As simulating a control area over many years of recorded historical operating conditions is a massive computation problem, the new scheme uses distributed computing with 228 computing nodes to maintain a reasonable simulation time. The simulator-derived automatic generation control and load following generated with a data set characterizing 3000 MW of wind generation integrated in the Quebec interconnection were compared with statistical analysis-based results. The results of the simulation of a full year of minute-by-minute operations suggest that wind integration will increase the number of generating unit start-ups and shut-downs by approximately 5%. The simulator was also able to estimate import/export opportunity losses, as well as several impacts related to voltages and reactive power attributable to increased wind penetration.