Australian electricity generation technology cost projections: application of a global and local learning model

[摘要] The Bureau of Resource and Energy Economics (BREE) is conducting an Australian Energy Technology Assessment. Worley Parsons is the main organisation responsible for providing projections of levelised costs of electricity (LCOE) for a wide variety of technologies up to the year 2050 on behalf of BREE. An important component of the projected LCOE is the future capital cost of each technology. CSIRO is assisting Worley Parsons to develop these projections. The initial technology capital costs, prepared by Worley Parsons, have been incorporated into CSIRO's capital cost projection methodology. CSIRO’s methodology provides projections of the rate of change in technology costs over time based on a global and local technology learning and adoption model.Different approaches can be used for projecting electricity generation technology capital costs and each has its own advantages and disadvantages. The approaches include bottom-up engineering and materials cost analysis, economic modelling with learning curves, recent quotes and price estimates, Delphi/subject matter groups and surveys or a combination of any of these. In this exercise, we combined bottom-up engineering, economic modelling with learning curves and review by a stakeholder group.The bottom-up engineering cost estimates were prepared by Worley Parsons. They are typically a good indicator of current costs of a technology and can be used to identify potential cost reductions in terms of materials or the technology itself. CSIRO used learning curves within the framework of an economic model to project the future costs. Learning curves, when used in an economic model that projects uptake of technologies, are a useful indicator of future costs. BREE convened a stakeholder reference group to review preliminary outputs from Worley Parsons and CSIRO.The economic model developed and used by CSIRO is the Global and Local Learning Model (GALLM). It features endogenous technology learning, nine regions, and twenty different electricity generation technologies. Projections are made out to the year 2050, both of uptake and capital cost, per technology. It includes a methodology for projecting cost increases due to market forces and global and local learning curves for technologies where data is available.Worley Parson's projected technology availability dates and initial capital costs were used as starting points for the model. The model was then run and the changes in capital costs projected.The resultant cost projections show decreases in costs up to the year 2050 for all technologies that have some uptake in the model. Technological change occurs more rapidly in the next two decades and slows toward the end of the projection period. Larger cost decreases occurred for emerging technologies such as wave, solar and carbon capture and storage (CCS) technologies. Hot fractured rocks, while emerging, has reduced potential for cost reduction due to limited global economically-recoverable resources. Consequently, most learning must be driven locally whereas other technologies can free ride on cost savings driven by deployment globally. Drilling is also a high percentage of total capital costs.Since the initial capital costs were taken from Worley Parson and not CSIRO, the resulting costs after learning effects were different to previous CSIRO publications. Comparing capital costs projected by this study in the year 2030 and those from previous studies (from 2009-2011), reveals that the projected costs of the majority of the coal-based technologies, biomass and hot fractured rocks are now higher than they were previously. The coal-based technologies and biomass can have large plant which would be affected by increasing commodity prices. Hot fractured rocks has a higher cost for the reasons mentioned above. The solar technologies, large scale PV in particular, are projected to be lower in cost than they were previously. This is due to the higher than expected glo...

[发布机构] CSIRO

[学科分类] 地球科学(综合)