The purpose of this study has been to address three key questions:
a) What combination of generation, transmission and storage technologies will achieve the least cost transition from existing infrastructure to a low carbon economy by 2050, and
b) How effectively will the National Energy Market (NEM) function under this new mix of technologies, and is a re-design on the market going to be required.
c) What is the value of distributed generation and storage technologies?
The project is a collaboration between the University of Melbourne and the University of New South Wales, along with the Australian Energy market Operator (AMEO), Bureau of Meteorology, Victorian Department of Treasury and Finance, GE and Market Reform. The project has been led by Dr Roger Dargaville and Prof Michael Brear from the University of Melbourne, and Assoc. Prof. Iain MacGill from UNSW.
A suite of modelling tools have been developed, with the first group simulating the function of the NEM, and second group examining the market behaviour of the NEM under the least cost generation mix determined by the first set of models. The key deliverables of this work have been several peer-reviewed papers in the academic literature along with the source code of the models being made publically available. The research team has made several public presentations of the project work. The papers contain significant detail describing the models and results, and the purpose of this report is to summarise the key outcomes.
The key outcomes of the study are:
a) Modelling tools designed to assess the performance of future energy systems under different levels of renewable energy penetration.
b) Descriptions of pathways to least cost combinations of generation and transmission technologies that achieve a range of carbon abatement trajectories
c) Improved collaboration between research community, government and renewable energy industry.
Three different technology mix models have been developed, and the code for each is available online. See below for more details. The models have been used to find least cost combinations of RE technologies, and the results have been written up in peer-reviewed articles. Engagement between the University researchers and government and industry have been enhanced during the course of the project with project meetings facilitating detailed discussions between the partners, and public events to present key results to a broad range of stakeholders.
The key results have been written up in the peer-reviewed literature (see list below) and are highlighted in a Powerpoint presentation available on the ARENA website. The results show that:
a) A low carbon energy system for the NEM is technically and economically feasible, and under certain circumstances may be cheaper than a business as usual scenario energy system with mostly fossil generation.
b) The model configuration and the cost assumptions that go into the simulations are very important and strongly affect the results. The results from Jeppesen et al. and Elliston et al. show that assuming the emission abatement target for 2050 is set at 80% of 2000 levels, the least cost combination of technologies is approximately 5% biomass, 20% onshore wind, 10% utility scale PV, 35% combined cycle gas turbines, 25% concentrating solar thermal, and 6% hydro power (Jeppesen, see figure 1), or 5% biomass, 60% onshore wind, 10% PV, 20% CCGT and 5% hydro (Elliston et al., figure 2).
c) The transition from today’s infrastructure to a low carbon energy system involves a rapid decline in capacity for black and brown coal over the period 2015-2030 while wind capacity ramps up, then an increase in combined cycle gas turbines and concentrating solar thermal from around 2030 onwards while the remaining coal capacity is retired.
d) A scenario with nuclear power and carbon capture and storage (CCS) included in the generation mix shows that at the Australian Energy Technology Assessment report (AETA, published by BREE) cost estimates, nuclear power is an attractive option, however CCS is not cost competitive. In none of the scenarios is geothermal power selected as cost competitive.
A key component of this study has been to make the source code for the models available. Each of the models, along with the documentation required to run the code has been made available.
Melbourne Uni Genetic Algorithm model:
Melbourne Uni Linear Programming version:
UNSW NEMO model:
Code and documentation: http://nemo.ozlabs.org