The expanding share of the fluctuating and less predictable wind power generation can introduce complexities in power system reliability evaluation and management. This entails a need for the system operator to assess the system status more accurately for securing real-time balancing. The existing reliability evaluation techniques for power systems are well developed. These techniques are more focused on steady-state (time-independent) reliability evaluation and have been successfully applied in power system planning and expansion. In the operational phase, however, they may be too rough an approximation of the time-varying behavior of power systems with high penetration of wind power. This paper proposes a time-varying reliability assessment technique. Time-varying reliability models for wind farms, conventional generating units, and rapid start-up generating units are developed and represented as the corresponding universal generating functions (UGFs), respectively. A multistate model for a hybrid generation and reserve provider is also proposed based on the developed UGF representations of wind farms, conventional generating units, and rapid start-up generating units. The proposed technique provides a useful tool for the system operator to evaluate the reliability and arrange reserve for maintaining secure system operation in the short- as well as medium-terms.
Ieee Transactions on Sustainable Energy, 2014, Vol 5, Issue 3, p. 896-906
Power, Energy and Industry Applications; Markov processes; Medium-term; Power system reliability; random process; reliability; short-term; universal generating function (UGF); Wind farms; wind power; Wind power generation; Wind speed