The indoor temperature variations in a commercial building conditioned by a solar thermally driven desiccant cooling system without a backup heat source were investigated by White et al。 [23]。 They found that, because their system could provide some cooling without a solar heat input using evaporative cooling, the indoor temperature variations were minimal in climates amenable to evaporative cooling。 However, a small amount of electrical power was still required to operate this system。 For the case of a totally off- grid PV driven vapour-compression system, it appears that there has been minimal investigation of the varying indoor conditions to date。

For totally off-grid PV air-conditioning, the question of building thermal design becomes one of minimising the required capacity or size of the system required to obtain a given level of comfort, sense no building-occupant combination can place a load on the electricity infrastructure, no matter how poor the building thermal

178 M。J。 Goldsworthy / Energy and Buildings 135 (2017)  176–186

design。 That is, the best building design is the one that requires the smaller investment in off-grid cooling system; importantly though any existing building can still become a ‘high efficiency’ building (at least in terms of air-conditioning) without relying on energy conscious occupants。 Indeed occupant behaviour can be somewhat removed from the design problem, since the system is effectively free to operate, the control can be autonomous and the use of mechanical air-conditioning can be made entirely on the basis of comfort alone (not cost), which should lead to overall improved occupant comfort throughout the  year。

Thus here we investigate the ability of an off-grid PV-battery driven vapour compression system to create comfortable condi- tions in different Australian climates and buildings with minimal PV panels and storage。 In order to simplify the analysis, condi- tioning of a single zone by a typical high efficiency split-system air-conditioner is considered and the influence of a range of build- ing envelope thermal parameters on the overall comfort levels is investigated。

The simulation model used is described in Section 2 and the building model validation in Section 3。 Because of the large number of possible parameter combinations with each involving a detailed annual simulation, a statistical approach has been adopted and is described in Section 4。 Results are then presented in terms of the influence of different system sizes and building design parameters, including their second order interactions, on the building comfort conditions。 For comparison purposes, the influence of the same building design parameters on the comfort if the building were unconditioned building is also investigated。 Finally these results are discussed in the context of the suitability of a modest size off-grid PV vapour compression system for Australian  climates。

2。 Simulation model

The simulation model is composed of sub-models for the vapour-compression air-conditioner and control, photovoltaic and battery storage system, weather data processing and building as described below。

2。1。 Vapour-compression air-conditioner

A simple vapour compression air conditioner cooling perfor- mance model was implemented using an input lookup data table to interpolate values of the ratio of the total cooling capacity Qt , sensi- ble capacity Qs, and electrical power consumption Pac at off-design conditions to the values of these parameters at the rated condition (Q ∗, Q ∗ and Pac ∗)。

Fig。 1。 Lookup table data for the variation of air-conditioner part load total cool- ing ratio and power consumption ratio as a function of ambient temperature and building wet-bulb temperature。