Dynamic uncertainty set optimization method for transmission expansion planning considering the prediction accuracy of wind power generation

Network planning operations involving systems with a high penetration of renewable power generation have widely applied robust transmission expansion planning (RTEP) to address the inherent uncertainty of these systems. However, the construction of uncertainty sets in RTEP is the core factor determining the quality of the resulting planning schemes, and existing construction methods tend to obtain suboptimal results by treating the boundaries of uncertainty sets as a static preset parameter, regardless of the accuracy of renewable power generation prediction that can be obtained at a given planning stage. The present work addresses these issues by proposing a risk-based method that treats the boundaries of the uncertainty set as a decision variable that is optimized in the solution process dynamically and objectively based on the prediction accuracy conditions of wind power generation observed at different transmission expansion planning stages. The proposed complex model, which includes variable-limit integral risk calculation terms, is solved by first applying an iterative algorithm with the least squares approximation technique to transform the model into a convex mixed integer non-linear programming problem that can be effectively solved by commercial solvers. The good performance of the proposed RTEP method is verified in terms of planning costs and operation efficiency by comparing the results obtained for a Garver 6-bus system and an IEEE RTS-79 system with those obtained using currently available RTEP methods.