We present a stochastic programming model for investments in thermal generation capacity to study the impact of demand response (DR) at high wind penetration levels. The investment model combines continuous operational constraints and wind scenarios to represent the implications of wind variability and uncertainty at the operational level. DR is represented in terms of linear price-responsive demand functions. A numerical case study based on load and wind profiles of Illinois is constructed with 20 candidate generating units of various types. Numerical results show the impact of DR on both investment and operational decisions. We also propose a model in which DR provides operating reserves and discuss its impact on lowering the total capacity needed in the system. We observe that a relatively small amount of DR capacity is sufficient to enhance the system reliability. When compared to the case with no DR, a modest level of DR results in less wind curtailment and better satisfaction of reserve requirements, as well as improvements in both the social surplus and generator utilization, as measured by capacity factors.
Available at: http://works.bepress.com/sarah_m_ryan/4/