John (Alex) Magerko with adviser P. Krein
Clouds, birds, leaves, snow, temperature, and other factors result in photovoltaic (PV) power output that can vary from hour to hour, minute to minute, or even on time scales less than one second. In order for significant PV resources to be compatible with the electric grid, their variability must be mitigated or compensated so that utilities can plan on “smoother,” more predictable PV power. Otherwise, large penetrations of solar PV power pose stability and voltage regulation concerns. This research initially analyzed 100 days’ worth of high-frequency PV voltage, current, and power data taken semi-randomly from nearly 500 total days’ worth of data. Variability was then quantified in terms of energy capture reduction as a result of discrete maximum power point (MPP) update rates, as shown in Figure 33.
The effects of two proposed methods were modeled to determine their potential PV variability mitigation capabilities. The first method implemented dynamic controls on heating ventilation and air conditioning (HVAC) fan drives. By perturbing HVAC controls, variable-speed fan drives could consume additional power during instances of solar energy spikes and reduce power consumption during periods of solar irradiation dips. This filtering capability relies on building thermal inertia to maintain occupant comfort. Acoustic and thermal constraints were investigated and their effect on the energy-filtering potential considered. The second method consists of curtailing maximum PV power output. Doing so permits available power overhead, or reserve, to compensate for some rapid increases/decreases in PV power with near instantaneous response time. The effect is to reduce rapid grid demands typically arising from the difference between the raw solar power and a smoothed, more predictable solar output. The optimal economic trade-off between energy sacrifice and variability mitigation is currently being investigated, and initial results show that energy associated with high-speed variability can be reduced by up to 27 percent as a fraction of energy sacrificed by curtailment. This research is supported by the Grainger Center for Electric Machinery and Electromechanics and the Siebel Energy Institute.