Comparing Photovoltaic Capacity Value Metrics: A Case Study for the City of Toronto

Abstract

Hourly electric power demand data in Toronto from 2000 to 2006 was analyzed along with coincident, simulated hourly photovoltaic (PV) power generation to quantify PV capacity value on a year-round basis. Three different methods commonly employed by electric utilities were used to assess PV capacity value, and their results were compared. The first method is the Garver approximation to effective load carrying capability (ELCC), which served as a benchmark for capacity value. The other two methods equate PV capacity value with the capacity factor during ''peak demand intervals'': for method 2 the interval includes all hours with loads within a given per cent deviation from the peak load; for method 3, a fixed ''on-peak'' interval of 11-17 h in June-August is used. Methods 2 and 3 yielded PV capacity values of about 40%, in agreement with the results of the Garver approximation at low grid penetration. This is considerably higher than the yearly PV capacity factor of about 12%, and is in good agreement with previous studies. Capacity value varies significantly from year to year: for instance, values from method 1 at low grid penetration levels range from 30% (year 2000) to 44% (year 2006). Yearly variations in capacity value appear correlated with variations in the demand summer to winter peak ratio, reflecting the fact that PV capacity value is strongly linked to its capacity to reduce peak demand (''peak shaving'') during the summer.

Summary of Time Period-Based and Other Approximation Methods for Determining the Capacity Value of Wind and Solar in the United States: September 2010 - February 2012

“…One of the methods adopted an interval that includes all hours with loads within a 10% deviation from peak load, while the other method adopted a fixed interval for on-peak, using June to August 11 a.m. to 5 p.m. Both of these methods found the solar PV capacity value to be around 40%, which is close to the Garver method results at low grid-penetration levels [12].…”

The Role of Plug-In Electric Vehicles with Renewable Resources in Electricity Systems

“…Wind's marginal capacity value drops quickly as more is added to the system, however. Analyses of solar conducted by Madaeni et al (2012Madaeni et al ( , 2013; Pelland and Abboud (2008); Perez et al (2006Perez et al ( , 1993Xcel (2009) show higher capacity values than wind, due to greater coincidence between solar availability and electricity demand in many power systems. Underestimating the capacity contribution of renewables can result in higher ratepayer costs, since excess generating capacity must be built.…”

Comparison of Capacity Value Methods for Photovoltaics in the Western United States

“…These techniques have been applied to wind [24][25] and PV [9] and compared with reliability-based methods to assess their accuracy. Milligan and Parsons [21] introduce three different approximation methods, which differ based on the set of hours examined.…”

“…9 This analysis uses the entire WECC footprint to determine system loads and LOLPs. Because this assumption keeps the underlying system fixed, differences in the capacity value of PV at different locations can be attributed entirely to differences in solar resource, without system characteristics confounding the results.…”

“…Therefore, assessing the adequacy of renewable generation technologies and consequently estimating their capacity value is crucial for accurate reliability and planning of power systems [2]. Previous analyses have considered the capacity value of wind [1,[3][4][5][6][7][8], photovoltaic (PV) solar [9][10][11][12][13][14], and concentrating solar power (CSP) plants [15]. Partially due its maturity, the capacity value of wind has been more widely studied than solar technologies.…”