Master of Philosophy
School of Electrical, Computer and Telecommunications Engineering
Li, Bosui, Evaluation of building demand and PV system output in order to devise an improved tariff and operational strategy for sharing PV power, Master of Philosophy thesis, School of Electrical, Computer and Telecommunications Engineering, University of Wollongong, 2017. https://ro.uow.edu.au/theses1/284
Governments in a number of countries have developed policies to support the utilisation of solar energy. These policies drive the adoption of new electricity tariffs such as: Feed-in tariff (FiT), net metering (NM), and virtual net metering (VNM). In recent years, VNM has attracted greater attention due to its improved performance for extending the utilisation of PV power. In the U.S., VNM has been widely applied to operate community solar projects and a number of studies have been done to evaluate the policy issues of VNM. However, studies focusing on policy and regulation are not enough to develop a comprehensive understanding of a solar tariffs because other contributing factors, such as load behaviour, PV system output and network service charges, have potential impact on the performance of solar schemes. The motivations of this thesis are to increase the understanding of VNM tariff by considering the above factors, and subsequently propose potential improvements. Furthermore, this thesis proposes an alternative to solar tariffs in order to share PV power where such solar tariffs are unavailable.
In this thesis an evaluation of VNM, FiT and NM was first developed using microeconomic models for the general comparison of the three widely applied solar tariffs in an Australian context. The results indicate that FiT generally provides the worst performance and VNM illustrates potential to be the best option, depending on the ability to minimize surplus credits.
To evaluate the impact of variations in building demand and PV system output on solar tariffs, this thesis developed several characteristic profiles of solar generation and load consumption using MATLAB simulations and analysis of real data. For the preliminary work, two types of Australian utility retail tariffs, time of use (TOU) and fixed-rate, were introduced to link the profiles to financial performance. The developed profiles of load demand and PV system output are also utilised for the financial evaluation of other solar tariffs such as VNM.
The research presented here on a credit based VNM tariff demonstrates that customer profiles with high peak load tend to provide a shorter payback period, and the percentage of VNM tariff participants also has significant impact on payback period. It is also shown that the value of VNM declines with increasing surplus (un-allocated) credits. Thus minimizing surplus credits and adjusting participant composition could be two approaches to improve the financial performance of VNM tariff. In addition, the shared PV system of a VNM program is generally only sized by referring to the capacity of overall loads. This study develops several sizes of PV solar system to illustrate the impact of sizing the PV system has on a VNM project.
Service charges of transmission and distribution utilities associated with solar sharing, also called wheeling costs, are studied using an approach of tracing power flow and MW-mile method. The results indicate that the percentage of power shared is more significant to the value of wheeling charges than the distance the shared power passes through and the amount of shared power. It is also demonstrated how the value of wheeling costs are influenced by the installation location of the shared PV system, the methodology for allocating wheeling costs to end users and the profile of load demand.
The last component of this study proposes a conceptual network as an alternative to solar tariff for sharing solar in a small-scale installation. The simulated results indicate that the network can implement solar sharing without the support of utility tariffs and energy storage, but the irregularity and variation of daily PV system output significantly challenge the network operation. Consequently, a probability-based methodology is proposed to minimise the impact from PV system output variation.
The profiles of load demand and PV system output are identified as the most critical components to successful PV sharing schemes, however, further research into the behaviour of load demand and PV system output is required. In addition, the evaluation of wheeling costs associated with solar sharing is at the initial stage, so the verification of relevant achievements still needs further research.