Optimisation of a residential energy system with an embedded PV source

Date
2016-03
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH ABSTRACT: The economic viability of PhotoVoltaic (PV) systems for the residential sector remains one of the greatest barriers to PV adoption. Economic viability of PV systems can be expressed in terms of a wide array of financial indicators. The economic viability of a PV system can be difficult to interpret for potential PV system owners, due to the fact that financial indicators for PV systems can involve concepts such as inflation, changing electricity tariffs, time-of-use tariffs and feed-in tariffs. This project focuses on simple payback time and determines the effect of tariff structures, load schedule optimisation and battery storage on the payback time of PV systems. The project goes on to determine whether an optimal PV system rating exists for which the payback time is minimum. For this purpose, a mathematical model is developed for a residential energy system. This mathematical model includes the subsystems present in a smart residential energy system, namely the non-controllable loads, controllable loads, battery storage, a PV system and the grid. The grid is associated with electricity tariffs, allowing for time-of-use tariffs as well as feed-in tariffs. The mathematical model can model the energy flow between subsystems. It provides a method of calculating energy cost for the residential energy system. A software application that implements the above mathematical model is developed to explore the payback time of residential PV systems. The application takes as input a load profile, solar profile and grid connection tariff structure. It calculates the PV system payback time as a function of PV systems rating. An optimisation is implemented to identify the PV system rating with the minimum payback time. Financial performance and optimisation results are presented for two sets of case studies. The first set of case studies is exploratory. Using simple input parameters, cause-and-effect relationships between input parameters and results established. The second set of case studies use representative input parameters to confirm that the observed cause-and-effect relationships are present in practical residential energy systems. The project identifies important mathematical factors that determine PV system payback time depending on the use of tariff structure, the inclusion of load schedule optimisation and/or the inclusion of battery storage. It is concluded that for each residential energy system, an optimal PV system rating with a minimum payback time exists.
AFRIKAANSE OPSOMMING: Die ekonomiese vatbaarheid van Photovoltaiese (PV) stelsels vir die residensiële sektor bly een van die grootste verhinderings tot PV installasies. Die begrip van ekonomiese vatbaarheid van PV stelsels word bemoeilik vir potensiële PV stelsel eienaars deurdat finansiële aanwysers vir PV stelsels konsepte insluit soos inflasie, veranderende electrisiteitspryse en tyd-afhanklike elektrisiteitaankoop en -verkoop tariewe. Die projek fokus op terugbetalingstydperk en bepaal hoe elektrisiteit-tariefstrukture, las-skedule optimering en batterykrag die terugbetaaltyd van PV stelsels beïnvloed. Die projek stel verder ondersoek in om te bepaal of ‘n optimale PV stelsel grootte met minimum terugbetalingstydperk bestaan. Ten einde die doel te bereik is ‘n wiskudige model vir ‘n residensiële kragstelsel ontwikkel. Die wiskundige model sluit onderliggende stelsels in ‘n intelligente huiskragstelsel in, naamlik die nie-beheerbare laste, die beheerbare laste, batterykrag, die PV stelsel en die kragnetwerk. Die kragnetwerk word geassosieer met ‘n elektrisiteits-tariefstruktuur, wat toelaat vir tyd-van-die-dag verbruikstariewe en invoer tariewe. Die wiskundige model modelleer die vloei van energie tussen die onderliggende stelsels. Dit bied die geleentheid om koste aangaande die residensiële kragstelsels te bereken. ‘n Sagteware program wat die bogenoemde wiskundige model implementeer is ontwikkel om verkenningswerk te doen aangaande die terugbetaaltydperk van PV stelsels. Die program se invoer is ‘n lasprofiel, sonkragprofiel en die tariefstruktuure van die kragnetwerk. Die program bereken die terugbetalingstydperk as ‘n funksie van die PV stelsel grootte. ‘n Optimering identifiseer die PV stelsel te identifiseer met die minimum tergbetalingstydperk. Finansiële prestasie- en optimeringsresultate word dan aangebied vir twee stelle gevallestudies. Die eerste stel is verkennend. Eenvoudige invoer parameters word gebruik om oorsaak-en-gevolg verhoudings tussen invoer parameters en resulate te verken. Die tweede stel gevallestudies gebruik verteenwoordigende invoer parameters om te bevestig dat die oorsaak-en-gevolg verhoudings wel teenwoordig is in praktiese residensiële kragstelsels. Die projek identifiseer belangrike wiskundige faktore ten opsigte van PV stelsels se terugbetaaltyd na aanleiding van die betrokke tariefstruktuur, die gebruik van lasprofiel optimering en/of die insluiting van batterystelsels. Dit word bevind dat ‘n optimale PV stelsel grootte met minimum terugbetalings tydperk vir ‘n residensiële kragstelsel bestaan.
Description
Thesis (MEng)--Stellenbosch University, 2016.
Keywords
UCTD, Photovoltaic power system, Dwellings -- Power supply, Electric power systems -- Mathematical models, Electric power systems -- Management, Electric power systems -- Economic aspects
Citation