Geographically diversified VRE portfolios can be used to smooth out the volatility of VRE output ramps. This study, using Taiwan as case study, developed efficient frontiers of optimal VRE portfolios to minimize the volatility for each possible level of total installed VRE capacity that can meet 10%, 20% and 30% of electricity demand. Our analysis shows that optimal portfolios are also beneficial to reduce the magnitude of extreme downward ramp events, which are sudden losses in VRE power outputs. We specifically investigated hourly extreme ramps that are expected to occur on average once-every-three-year. They are 13-30 % of each unit of installed VRE capacity for optimal VRE portfolios, which are significantly smaller than that (20-64%) for most individual VRE assets. This result helps to manage risks associated with extreme ramp events in power system operation. To capture the benefits associated with optimal portfolios, it is recommended for policy-makers to coordinate the investment and development of VRE assets across multiple locations.
Keywords variable renewable electricity, mean-variance portfolio, extreme value analysis, generalized extreme value distribution, extreme ramp