Effects of time - scale on householder PV economic analyses: over - estimation of self - consumption
The economic analysis of when solar roof-top photovoltaic (PV) systems become viable is of great interest to many householders. It is well known that buy-back rates for solar energy are low and in general a PV system only becomes financially attractive if households consume a high proportion of the energy generated, in order to benefit from avoiding the cost of buying electricity at retail rates. Battery storage for most householders remains uneconomic (although prices are trending downwards) making power consumption patterns critical to determining financial viability. Currently householders have access to their power consumption from their retailer in half-hour time periods. If this information is combined with PV generation at the householder’s location, a good estimate of the financial viability can be determined. However, using a half-hour analysis period implies that the power generated and load consumed in the half hour period is constant. The reality can be very different though, as both loads and generation can spike or dip and it is the instantaneous load and generation that determines when power is exported or imported from the grid. To determine the effect of time-scale on self-consumption, values were calculated from one-minute generation and load data for a number of New Zealand installations and compared with calculations for half-hour and hourly data. All households analysed showed some over-estimation of self-consumption when compared with the one-minute baseline data. For a typical 3.5kW PV system, the over-estimation varied between 1% and 8%, depending on the base level of self-consumption and characteristics of the load profile. In addition, the impact of using median load profiles, such as those used by the EECA EnergywiseTM Solar Calculator on self-consumption values was investigated. The process of creating median load profiles further smooths the load, similar to the effect of using longer-time-scales. When using median load profiles, the self-consumption over-estimation was found to be in the order of 5% for the majority of householder profiles for the various regions specified in the Solar Calculator.