Spinning reserve quantification by a stochastic–probabilistic scheme for smart power systems with high wind penetration

• A stochastic–probabilistic approach is proposed for spinning reserve quantification.
• A new linearized formulation integrating reliability metrics is presented.
• The framework manages the reserve provided by responsive loads and storage systems.
• The proposed method is capable of detaching the spinning reserve for different uses.

This paper introduces a novel spinning reserve quantification scheme based on a hybrid stochastic–probabilistic approach for smart power systems including high penetration of wind generation. In this research the required spinning reserve is detached into two main parts. The first part of the reserve is procured to overcome imbalances between load and generation in the system. The second part of the required spinning reserve is scheduled according to the probability of unit outages. In order to overcome uncertainties caused by wind generation and load forecasting errors different scenarios of wind generation and load uncertainties are generated. For each scenario the reserve deployed by different components are taken account as the first part of the required reserve which is used to overcome imbalances. The second part of the required reserve is based on reliability constraints. The total expected energy not supplied (TEENS) is the reliability criterion which determines the second part of the required spinning reserve to overcome unit outage possibilities. This formulation permits the independent system operator to purchase the two different types of reserve with different prices. The introduced formulation for reserve quantification is also capable of managing and detaching the reserve provided by responsive loads and energy storage devices. The problem is formulated as a mixed integer linear programming (MILP) problem including linearized formulations for reliability metrics. Obtained results show the efficiency of the proposed approach compared with the conventional stochastic and deterministic approach.