Hybrid power generation systems that combine more than one renewable resource type are a potential option to improve the capability of renewable power generation systems to meet network demands reliably. This increases the system complexity and cost, so this must be compensated by achieving correspondingly higher capacity factors to achieve similar financial performance to simpler systems. In the analysis undertaken, the hybrid systems developed were limited to combinations of biomass combustion and concentrated solar thermal technology for production of steam to feed a Rankine cycle turbine system. To ensure that resource availability was realistic in the study, biomass availability was based on 5 years of historical data for an existing biomass power generation site in Australia that currently has limited seasonal operations and matching solar data for the same location. A technoeconomic assessment was undertaken in parallel with optimization of plant configurations by inclusion of additional plant components and varying sizing. This included plant designs with different storage capacities, both thermal storage for solar energy and torrefaction char from short-term surpluses of biomass. Several system options were identified where financial performance matched the simple biomass combustion system, but with significant increases in capacity factor through hybridization.

In recent years, there have been significant changes in the dominant methods used for electricity generation worldwide with a higher increase in new generation from renewable technologies in comparison to a lower increase in traditional fossil or nuclear technologies.1 The rate and extent of these changes will vary from country to country, and even between different regions within countries, due to the varying quantity and quality of resources available, but the trend of increasing renewable generation is expected to continue into the future internationally.2 The change in electricity generation types can result in reductions in grid stability and supply reliability due to the increase in unscheduled generation, such as wind and solar photovoltaics, and the retirement of older baseload generators. In Australia, this has resulted in bodies such as the Australian Energy Council (AEC),3 the Australian Electricity Market Operator (AEMO),4 and the Australian Electricity Market Commission (AEMC),5 discussing the need for changes in the market to improve the reliability of supply.

The distribution of Australia's electricity networks and generators is shown in Fig. 1, indicating that there are several separated networks that do not interconnect, and that coverage does not extend into much of the inland area. Current electricity generators are also shown, with the type of generator varying markedly by region due to the available local resources and the demand for electricity, which varies noticeably with region. It needs to be noted that the data presented come from a variety of sources and are unlikely to be comprehensive. In particular, small renewable plants are shown when natural gas or diesel generators of similar size are not due to the broad distribution of such plants in roles such as emergency backup. Location also influences the identification of power generators as significant, with small units within the major distribution networks being less significant than units in isolated areas. Some general trends are evident in the locations of different types of generator:

  • Natural gas and distillate power plants supply much of the inland areas that are not serviced by the major electricity networks.

  • Large coal- and natural gas-fired generators are typically close to major population centers and are dominant nationally by capacity and generation.

  • Renewable energy sources provide a smaller portion of the total generation and the types used vary by region:

    • Hydroelectric power stations are predominately in the south-eastern areas of the National Electricity Network (NEM), in particular, in Tasmania.

    • Wind turbine installations have a high density along the southern coastline or along the mountain ranges in the south-eastern region.

    • Solar photovoltaic generation has numerous smaller installations in the more heavily populated areas along the coastline but has experienced growth in larger inland installations (100–150 MWe) in recent years where network connection is possible.

    • Biomass generation includes a range of technologies, such as biogas and agricultural waste combustion, with the installations typically being associated with either population centers or large-scale primary industries, such as timber and sugar cane.

FIG. 1.

Australian electricity network distribution and generation types.

FIG. 1.

Australian electricity network distribution and generation types.

Close modal

The different types of electricity generation differ in performance characteristics, such as reliability and the ability to dispatch electricity on demand. Wind and solar photovoltaic technologies have an increasing market share but offer limited control over output and require implementation of separate storage technologies, such as batteries or pumped storage, with additional costs and losses to provide any ability to match demand. Hydroelectric systems are typically capable of providing rapid demand response, at least to the extent that rainfall in the dam catchment is reasonable or through use of pumped storage techniques, but there are limited opportunities for new or expanded plants. Biomass systems offer considerable flexibility in operation, particularly if either solid biomass or stabilized chars can be stored for later use for periods in the range of months, but limitations in the availability of suitable biomass hinder increased usage. Another renewable energy technology that is not currently implemented at significant scale in Australia is concentrated solar thermal (CST), which can provide better capability of meeting demand than solar photovoltaics through use of large-scale thermal storage.

In a previous paper,6 an assessment of CST systems with different specifications indicated that, even at very high solar availability sites in inland Australia, it is not practical to achieve capacity factors higher than 80%. It is also evident that large CST systems have improved financial performance, which provides a challenge for connecting plants in the inland areas where electricity transmission networks are limited. These issues provide distinct challenges for implementation of the technology at significant scale in Australia. In contrast, biomass technology appears to offer the potential for reliable high capacity factor electricity as a standalone technology due to the ability to store biomass or char for extended periods, but the limited availability of biomass in many areas provides a challenge to increasing technology uptake. RIRDC7 reviews the potential for increased usage of biomass and highlights that relatively small-scale systems are favored due to the cost in transporting large quantities of material for long distances and that seasonal variability in the supply for many sources results in inconsistent output. This combination of issues for CST and biomass technologies provides an unusual circumstance as both are thermal technologies that currently use steam turbine systems for power generation and there are likely to be locations where hybridization of the two technologies can combine increased plant sizes for improved financial performance with improved ability to meet network electricity demand. However, previous assessments of the potential for hybrid CST and biomass plants in Australia8,9 have considered the general distribution of both resources on an annual basis without detailed consideration of the resource variability and the impact this will have on hybrid plant operations.

Hybridization of CST and biomass systems is not a new concept and the first commercial CST-biomass hybrid plant, Termosolar Borges in Spain, commenced operation in 2012. This plant has a nameplate capacity of 22.5 MWe and consists of a parabolic trough solar field and external biomass-fired (wood chips) boilers. The solar field generates saturated steam at 4 MPa, and the biomass boilers superheat this steam to 520 °C.10 The approach taken is not dissimilar to the use of CST in feedwater heating at coal-fired power stations, but is perhaps more beneficial in a biomass plant where the available quantity of biomass can be limited, particularly on a seasonal basis, and reduction in daytime usage would extend the output of the plant through the year. There have been other suggested CST-biomass hybridization scenarios, but it appears that none have been practically implemented. Some suggestions have been to integrate a biomass-fired boiler into the CST plant's water-steam cycle, incorporating a biomass heater into thermal oil or molten salt cycles, using an air receiver to preheat the gas turbine air after the compressor and then using this heated air to combust biomass-derived syngas, and supplying steam to a joint turbine by independent biomass and CST systems.10 This last approach has some advantages as it allows the CST and biomass systems to operate largely independently, potentially using existing equipment in an upgraded system, and for the steam turbine to be operated at part load when only one steam supply is available. Biomass gasification systems have also been proposed for hybridization with solar input, for example,11 proposes that the exhaust from the gas turbine using gasifier product be a combined source of steam with the solar input. Torrefaction is an additional technology that has been proposed as being beneficial to ensuring that biomass systems can operate reliably and could potentially be included within the scope of a hybrid process.

In this paper, we will assess the biomass and CST resource availability in Australia and evaluate the likely performance of a hybrid system incorporating biomass combustion and CST for electricity generation, including consideration of biomass torrefaction, at a site that has both resources available in suitable quantities.

Biomass use for electricity generation utilizes a variety of feedstocks and technologies, with the distribution of current electricity generation plants by feedstock in Australia provided in Fig. 2. Solar resources are more broadly distributed, with the general availability also indicated. Bagasse, woody material, and other wastes categories in these data all currently utilize technologies where the biomass is combusted to produce steam for use in a steam turbine power block. Landfill and sewage gas are typically used in gas engines if operating as a standalone technology but could also be used as supplementary fuel in other thermal electricity generation technologies, such as CST. Bagasse dominates the generation capacity from biomass for Australia, with sugar mills ranging from northern NSW to north Queensland. One sugar mill is also indicated at Ord River in northern WA, but sugar cane production has currently been abandoned in the region and this is no longer in use. The high generation capacity for bagasse arises from the sugar cane being transported to central mills for processing, resulting in a large quantity of waste bagasse at the mills that requires disposal. Traditionally, this has been used for steam and electricity generation for use within the mills, but with increases in plant efficiency and a demand for renewable electricity in the market, there has been an increase in the generation of surplus electricity for sale from the mills. It should be noted that sugar milling is a seasonal activity, with Australian mills typically operating from May to December, and in most cases electricity generation for export will only occur during this period. Stockpiling of bagasse for off-season use is limited and has historically occurred only at mills that also refine sugar but has been introduced at a limited number of mills to extend electricity production.12 Torrefaction is a potential method for allowing longer term storage of bagasse as char as the char is more chemically stable, does not spontaneously combust, and is hydrophobic, so it can be stored for long periods without the degradation impact experienced in bagasse stockpiles.13 

FIG. 2.

Distribution of biomass generation plants by type in comparison with solar availability.

FIG. 2.

Distribution of biomass generation plants by type in comparison with solar availability.

Close modal

Figure 2 also indicates that bagasse electricity generation plants in northern Queensland are also in an area of relatively high solar availability, so they provide an interesting opportunity for assessment of potential hybrid opportunities with solar input. A detailed breakdown of the locations and generation capacities for sugar mills in northern Queensland is provided in Fig. 3, with the capacities being the total generation from all power systems at the mill. This is a significant distinction as most mills have multiple boilers and small turbines that supply steam and electricity for use in the mill and will typically only have a single turbine that can export electricity. Hybridization of solar with the systems supplying the mill would result in complex issues due to interaction with the mill operations and the offtake of various pressures of steam to supply different equipment in the mill, so it is not ideal for assessment. Two mills in the southern part of the map, Invicta and Pioneer, are notable in the relatively large capacity of power generation and will be considered in more detail. Both are currently owned by Wilmar Corporation, who also own the nearby Kalamia and Inkerman mills with the benefit that relatively large power block capacities can be operated by transfer of surplus bagasse between the mills. Pioneer mill was subject to an extensive upgrade in 2005 which included a new boiler and steam turbine that was dedicated to exporting up to 37 MWe of electricity to the NEM.12,14 This turbine uses the relatively advanced, for biomass plant, steam conditions of 6.5 MPa/483 °C and operates from late December to March on stored bagasse after the normal sugar milling season has ended. This is facilitated by the transfer of approximately 110 000 tonnes of bagasse from other mills to Pioneer when they are operating and a storage system that requires that the oldest bagasse is used first.

FIG. 3.

Locations and generation capacities for sugar mills in northern Queensland.

FIG. 3.

Locations and generation capacities for sugar mills in northern Queensland.

Close modal

Invicta mill has a 39 MWe turbine system specifically for electricity export that was installed in 1995 and upgraded in 1997 to operate at steam conditions of 4.3 MPa/450 °C.15 This system typically operates and exports electricity to the network only when the mill is processing without significant bagasse storage, so data on electricity export can be relatively simply translated into estimates of the bagasse availability throughout the year. Data on the export of electricity to the NEM from individual generators is available from the NEMweb online system16 and is shown for 5 years of operation for Invicta in Fig. 4. These data are provided in 5-min increments and has been condensed to daily totals for simplification, with the selection of the range of years from 2011 to 2015 in order to align with other data availability. It is notable in the data that 2013 has considerably lower electricity generation than other years, with Australian Canegrower17 providing an explanation that exceptionally dry conditions in the area resulted in a very short harvesting and processing season. The average data show that bagasse availability is less predictable in early and late stages of the processing season, with only mid-season operations being consistently near the full capacity of the plant.

FIG. 4.

Exported electricity from Invicta sugar mill to NEM for 2011–2015.16 

FIG. 4.

Exported electricity from Invicta sugar mill to NEM for 2011–2015.16 

Close modal

Solar data for Australia is readily available from the Australian Renewable Energy Mapping Infrastructure (AREMI)18 as hourly data for multiple years based on satellite data. Direct Normal Irradiance for the area around Invicta Mill for the years 2011 to 2015, matching the biomass power generation period, is shown in Fig. 5 as daily totals. Due to the tropical location, there is relatively minor variation in the data with season. However, the solar availability is slightly lower during the first half of the year than during the sugar processing season in the mid to late part of the year. This is not ideal for a hybrid biomass-CST plant where it would be beneficial if availability was out of phase so total output could be consistent throughout the year, so approaches that can assist in levelising the output would be beneficial.

FIG. 5.

Daily solar availability at Invicta mill for 2011–2015.18 

FIG. 5.

Daily solar availability at Invicta mill for 2011–2015.18 

Close modal

A number of different hybridization options for bagasse and CST systems are available and the selection of an optimal approach is likely to be influenced by a combination of theoretical and practical factors.19–22 Based on the selection of Invicta mill as a potential site, there is an existing biomass power plant and data are available on some of the historical operations to the extent that the multi-year availability of bagasse for electricity generation can be estimated for small time steps (15 min). The existing boiler and steam turbine at the plant are not current state-of-the-art technology, so a new build would be likely to utilize plant options similar to those proposed by Tully Sugar Ltd23 with turbine inlet steam conditions approximately 7.9 MPa/522 °C. This would both increase the efficiency of bagasse utilization, thus increasing generation, and be compatible with the typical specifications for conventional CST power blocks suiting molten salt central receiver tower technologies. Bagasse boiler efficiency in Ref. 23 is 66.9% (HHV basis), which is mid-range of the 65.5 to 69.5% results found experimentally for six different boilers,24 so it was used as the default value for the analysis. This was also assumed to be constant across the part-load boiler operations based on an experimental and modeling assessment of bagasse boiler performance25 that identified only minor variation. Gasification of biomass for a hybrid system as proposed in Ref. 11 offers a potentially higher performance system, but bagasse is a challenging feed for gasifiers due to low bulk and energy densities. A review of the available technologies for prepared bagasse for gasification technologies26 reached the conclusion that this is not currently feasible at commercial scales.

Central receiver systems have been the selected CST technology in most international research programs, such as SunShot in the U.S and ASTRI in Australia, aimed at developing more cost competitive CST plants. In addition, as suggested by Burin et al.,20 the most promising hybrid CST-bagasse power plant consists of separate biomass and solar tower systems where both supply steam independently to the joint turbine. This is mainly due to the capability to operate in a solar-only mode that provides steam at appropriate conditions for operation of the steam turbine system during the part of the year when fresh bagasse supplies are not available. Central receiver tower systems are the preferred solar technology due to the required steam temperatures for efficient power blocks and the expected scale of operations, so this will be the only solar technology considered. The variant of tower technology utilizing molten salt, a mixture of sodium and potassium nitrates as both heat transfer fluid and storage have also been selected. It may seem superfluous to include a thermal storage system given that bagasse or torrefied char can be stored, but the scale of bagasse or char storage required to supply the system for the whole year is excessive if continuous or near continuous plant operation is desired.

Utilizing the system design approach and modeling methods described in Ref. 6, a 50 MWe central receiver tower system utilizing a two-tank molten salt storage with the capacity of 10 h is predicted to have the output shown in Fig. 6 for the period 2011 to 2015. Additional systems with 4 and 16 h of storage were also developed and modeled for use in assessments of the sensitivity to the storage capacity of the CST system. It should be noted that the power block for these CST systems is comparable in size to the current one at Invicta mill but is not identical, and to meet this performance, a new slightly larger unit would have to be installed or the solar plant scale reduced slightly to match the existing power block.

FIG. 6.

Predicted electricity production for a standalone CST system with 10 h of storage and 50 MWe power block for 2011–2015.

FIG. 6.

Predicted electricity production for a standalone CST system with 10 h of storage and 50 MWe power block for 2011–2015.

Close modal

There are multiple options for integration of the solar thermal plant with the bagasse system, which each likely to have different operation benefits for specific operational scenarios. The process shown conceptually in Fig. 7 was selected for this study based on a integrating a significant quantity of existing biomass plant with a new solar plant and high efficiency power block. The existing superheated steam from the biomass boiler is supplemented by steam generated from hot molten salt supplied by the solar plant. A major advantage of using the molten salt storage system is that this can be located a moderate distance from the solar field and close to the power block, so control of steam supply to the turbine can have minimal lag issues. As a guide to reasonable transfer distances, an engineering assessment by Abengoa Solar27 found that molten salt could be transferred up to 1.5 km with minimal parasitic losses. The steam turbine selected has a single reheat stage, which improves efficiency over the power blocks typically used with bagasse systems, and the reheater only uses molten salt. This eliminates the need to modify the existing bagasse boiler to extract more high-grade heat, which may not be practical, and instead uses the more readily accessed molten salt system. Operationally this will require that sufficient hot molten salt is always retained to operate the reheater even when other solar plant components are not in operation. This is in opposition to the decision to keep the bagasse and solar steam generation systems independent but is seen as warranted to achieve a cost-effective design with a high efficiency power block.

FIG. 7.

Hybrid Biomass-CST system utilizing a common steam turbine system.

FIG. 7.

Hybrid Biomass-CST system utilizing a common steam turbine system.

Close modal

The option of including a torrefaction plant has also been included in the process, using molten salt and recycled flue gas to produce char from surplus bagasse during periods of operation when the combined bagasse and solar availability exceeds the system requirements. Numerous different torrefaction technology variants have been proposed and tested,28 but in broad terms the process involves heating the bagasse to 230–300 °C in an inert or non-oxidising gas.29 This results in the volatiles being released from the biomass and the residual char having a higher carbon content and energy density. Volatiles from the torrefaction process are fed to the bagasse boiler for disposal, and the torrefied char is stored to supplement solar availability during periods of the year where bagasse is not available. Through the combination of combustion of the released volatiles, efficient heat recovery systems in the torrefaction plant, and improvements in the char chemistry for subsequent combustion, there is minimal energy loss in the process.30 The design of the torrefaction plant is not a major objective of this study, and the approach taken by Ref. 30 has been adopted with the minor modification that the hot gas used has been heated with molten salt, rather than combustion of the volatiles. This is aimed at simplifying the plant by elimination of the additional combustion section but is taken to have no impact on performance or cost. The improvements in energy density and low moisture content in the char should also be expected to improve combustion efficiency when the char is used as feed to boilers, but this does not appear to have been proven at industrial scale and has been neglected in the analysis.

The approach taken to assessing the optimal design and sizing of a hybrid system involved independently increasing the size of the concentrated solar plant and the power block with three different sizes (4, 10, and 16 h relative to the solar plant capacity) of molten salt storage and optional inclusion of a torrefaction plant. Operation of these plant options was simulated in parallel with the historical operating data for the biomass plant using the matching solar data for the site over the same five years. Using matching data is an important factor in assessing renewable technologies as it is likely that weather events such as rainfall that are likely to influence the production rates for biomass will also influence the performance of a co-located solar plant. The time resolution for the simulation of the hybrid plant is one hour as this matches the resolution of the solar data available from AREMI and the use of NREL's SAM package for simulation of the solar plant components. Biomass data, based on the historical power generation data from the existing power plant obtained from NEMWeb, was condensed to matching these hourly intervals. Simulation of the hybrid system involves assessing the potential total steam production from the combined biomass and solar plants in a given period. If this exceeds the requirements of a specified turbine size, then the decision logic as shown in Fig. 8 is used to reduce steam production by either diverting biomass to short-term storage, diverting molten salt to storage, or diverting both to a torrefaction plant to produce char for long-term storage. Optimization of the system design and operation was conducted for the three different CST storage capacities to determine the size of CST system and size of the power block that achieves the lowest average LCOE over the 5 years of operation, both with and without a torrefaction plant.

FIG. 8.

Logic diagram indicating decisions on usage and storage of different streams.

FIG. 8.

Logic diagram indicating decisions on usage and storage of different streams.

Close modal

The methodology explained in6 has been used to assess the economic performance of the system. Levelized cost of electricity (LCOE), as defined by Eq. (1), has been the main metric for economic evaluation of the studied system,

LCOE=t=1nCapExt+OpExt+Ft(1+r)tt=1nEt(1+r)t,
(1)

where CapEx is total capital cost ($), OpEx is the operational and maintenance cost ($/y), F is Fuel cost ($/y), n is Life of project (years), r is the discount rate, E is the electricity output (kWh/y), t is the year of the project, and LCOE is the levelised cost of electricity in c/kWh. Central receiver costs are based on the cost data provided in Ref. 6 with minor updating to account for observed changes in plant costing based on the Chemical Engineering Plant Cost Index (CEPCI).24 Unit costs for CST plants are variable with location, system size, and storage capacity, so the approach taken in Ref. 6 was used to produce a series of plant designs and costs with different storage capacities at the target site with NREL's System Advisor Model (Version 2018.11.11). The simplified costing expressions given in Table I for CST with the three different storage capacities were generated from the model predictions to enable resizing of the CST component during hybrid design optimization for minimizing LCOE. The additional expressions for biomass boiler, torrefaction, and power block components were produced based on Ref.23 and Ref. 30 with all component costs updated using the CEPCI31 and including provision for related balance of plant items and a combined EPC, Owners Costs and Contingency provision of 18%. Component efficiencies were assumed to be as provided in the reference publications.

TABLE I.

Costing expressions for CST, biomass, torrefaction, and power block plant components for use in economic assessment. Ms = design steam rate (MWt); Me = design gross electrical output (MWe); and Mb = design bagasse torrefaction rate (t/y).

ComponentCapital cost, $O&M cost, $/yBasis
CST (4 h, SM = 2.3) 3 330 000 Ms0.858 427 319 Ms0.458 6  
CST (10 h, SM = 3.3) 4 541 300 Ms0.869 392 727 Ms0.500 6  
CST (16 h, SM = 4.1) 5 015 000 Ms0.897 342 678 Ms0.547 6  
Biomass boiler 824 504 Ms 3 485 Ms 23  
Torrefaction 2 193 Mb0.800 84.44 Mb0.871 30  
Power block 1 306 224 Me 32 656 Me 23  
ComponentCapital cost, $O&M cost, $/yBasis
CST (4 h, SM = 2.3) 3 330 000 Ms0.858 427 319 Ms0.458 6  
CST (10 h, SM = 3.3) 4 541 300 Ms0.869 392 727 Ms0.500 6  
CST (16 h, SM = 4.1) 5 015 000 Ms0.897 342 678 Ms0.547 6  
Biomass boiler 824 504 Ms 3 485 Ms 23  
Torrefaction 2 193 Mb0.800 84.44 Mb0.871 30  
Power block 1 306 224 Me 32 656 Me 23  

Despite the decision to base the biomass plant on historical biomass supply data without considering potential future variability, there are multiple variables that can be modified to produce many system designs for potential hybrid systems. Variations include the size of the power block, the size of the CST system and associated storage, and the inclusion of a torrefaction system to allow char storage. In Fig. 9, a broad selection of these options has been optimized to minimize LCOE for power block sizes ranging from 40 to 100 MWe. Adoption of a more advanced boiler and steam power block with the existing biomass supply improves the plant efficiency to the extent that a power block of approximately 50 MWe capacity can be used, compared to the lower efficiency 39 MWe power block used in the current plant, and offers the lowest LCOE for a biomass-only system. This is a considerable gain in output that requires successful implementation of a reheater, which is uncommon in bagasse-fired power plants, but allows direct comparison with the higher efficiency power blocks typically used in CST plants. Hybridizing with a CST plant with 4 h of thermal storage and torrefaction char facilities offers a similar LCOE, but, in general, hybrid systems benefit from increased power block size and larger storage capacities. As the biomass input is fixed, using larger power blocks would require increasing the scale of the CST plants to supply the steam demand and increasing beyond 100 MWe scale would effectively make the plants increasingly CST-biased with biomass input becoming less significant. The decision to limit power block size to 100 MWe was based on the site not being an ideal location for a CST plant as sites further inland in Australia can have considerably higher solar availability and would be selected if a large CST installation was the target. The assessment of hybrid plant designs is, therefore, essentially bounded by the comparison of options at 50 MWe and 100 MWe, where different options appear to provide opportunities with minimized LCOE. It is apparent from the graph that plants without torrefaction and char storage must be larger to be competitive with those with char facilities.

FIG. 9.

Predicted LCOE with variations in system design with size of the CST component optimized for each case.

FIG. 9.

Predicted LCOE with variations in system design with size of the CST component optimized for each case.

Close modal

The predicted capital cost and capacity for 50 MWe plant designs are shown in Fig. 10 for a range of biomass, CST, and hybrid with or without torrefaction capabilities for char storage plants, all utilizing the same 50 MWe power block. While not explicitly shown on the figure, it is implicit that the higher cost of the hybrid systems with char compared to the biomass-only plant is compensated by a corresponding rise in generation to maintain a similar LCOE for the systems. The CST-only systems shown for comparison have considerably higher cost than the biomass system, so there is an obvious advantage to CST in hybridizing with biomass at this location. The similarity in output and total cost for the cases that include torrefaction implies that the optimization process is identifying the CST storage and torrefaction as being approximately equivalent and adjusting the plant sizing to correct toward an optimal overall storage strategy regardless of the CST storage capacity selected.

FIG. 10.

Capital cost breakdown and capacity factor for 50 MWe systems.

FIG. 10.

Capital cost breakdown and capacity factor for 50 MWe systems.

Close modal

The predicted capital cost and capacity for 100 MWe plant designs are shown in Fig. 11 for a range of biomass, CST, and hybrid with or without torrefaction capabilities for char storage plants, all utilizing the same 100 MWe power block. It is evident, in comparison with the 50 MWe power block system, that the biomass supply alone is not sufficient for efficient operation of the 100 MWe power block. It is apparent that implementing solar plant into a hybrid with biomass will increase the total capital cost considerably, with the solar components typically contributing more than twice the cost of the biomass components. However, the hybrid systems can result in major increases in capacity factor over the biomass-only system.

FIG. 11.

Capital cost breakdown and capacity factor for 100 MWe systems.

FIG. 11.

Capital cost breakdown and capacity factor for 100 MWe systems.

Close modal

A summary of the systems with the lowest LCOE for each system type at both 50 and 100 MWe scale is provided in Table II, including the capital cost, generation, LCOE, and capacity factor achieved by each system. The biomass available is constant for all cases, but the optimum solar plant size for hybrid options was determined separately for each system configuration and ranges from 30 to 80 MWe. This is significant for parts of the year where solar alone must supply the power block as the power block efficiency will reduce at low loads with an impact on the benefits of the system. In addition to the integrated hybrid systems, an additional option is shown that combines standalone 50 MWe biomass and CST-4 h systems for comparison. The 50 MWe biomass only system is essentially a base case that represents a conventional approach, albeit with a more advanced power block than is currently used. Despite the higher capital cost of the large hybrid systems, the increased output is shown to improve the financial performance over the 50 MWe biomass system in two cases. These options are 100 MWe systems with either 10 h of storage with torrefaction char capability or 16 h of storage without torrefaction plant. These two options have similar capacity factors of approximately 70% and, along with the larger turbine size, this results in greater than fourfold increase in electricity output compared to the 50 MWe biomass only plant. Generation duration curves based on the predicted output over the five years of data are shown in Fig. 12 for the biomass, CST and hybrid systems with lowest LCOE at the two power block sizes. These show a decline in ability of the hybrid systems to maintain full generation when either resource is unavailable, with the biomass input having a reduced influence on the larger power block plants due to the higher solar input.

TABLE II.

Comparison of cost and performance characteristics for selected systems.

System designPower blockSolar plantaTotal capital costAverage generation MWh/yLCOE c/kWhCapacity factor
Biomass only 50 MWe N/A $184.3 m 132 335 15.7 31.5% 
Biomass+CST-4 h 50 MWe 40 MWe $381.7 m 235 070 18.0 53.7% 
Biomass+char+CST-4 h 50 MWe 30 MWe $337.6 m 250 847 15.7 57.3% 
CST-16 h only 50 MWe 50 MWe $502.1 m 311 021 17.4 71.0% 
Separate biomass & CST-4 h 50 + 50 MWe 50 MWe $566.0 m 372 861 17.2 42.6% 
Biomass only 100 MWe N/A $249.6 m 120 378 24.8 13.7% 
Biomass+CST-16 h 100 MWe 80 MWe $914.9 m 622 160 15.5 71.0% 
Biomass+Char+CST-10 h 100 MWe 80 MWe $790.9 m 611 987 14.1 69.9% 
CST-16 h only 100 MWe 100 MWe $943.9 m 623 295 16.0 71.2% 
System designPower blockSolar plantaTotal capital costAverage generation MWh/yLCOE c/kWhCapacity factor
Biomass only 50 MWe N/A $184.3 m 132 335 15.7 31.5% 
Biomass+CST-4 h 50 MWe 40 MWe $381.7 m 235 070 18.0 53.7% 
Biomass+char+CST-4 h 50 MWe 30 MWe $337.6 m 250 847 15.7 57.3% 
CST-16 h only 50 MWe 50 MWe $502.1 m 311 021 17.4 71.0% 
Separate biomass & CST-4 h 50 + 50 MWe 50 MWe $566.0 m 372 861 17.2 42.6% 
Biomass only 100 MWe N/A $249.6 m 120 378 24.8 13.7% 
Biomass+CST-16 h 100 MWe 80 MWe $914.9 m 622 160 15.5 71.0% 
Biomass+Char+CST-10 h 100 MWe 80 MWe $790.9 m 611 987 14.1 69.9% 
CST-16 h only 100 MWe 100 MWe $943.9 m 623 295 16.0 71.2% 
a

Equivalent power block size if developed as a standalone solar plant.

FIG. 12.

Generation duration curves for selected biomass, CST and hybrid systems with the best LCOE at 50 and 100 MWe scales.

FIG. 12.

Generation duration curves for selected biomass, CST and hybrid systems with the best LCOE at 50 and 100 MWe scales.

Close modal

The system with the lowest LCOE, a hybrid 100 MWe system utilizing CST with 10 h of storage and torrefaction char systems, provides a capacity factor of 69.9% compared to the 31.5% achieved by the 50 MWe biomass system. The predicted power generation over 5 years of operation for this hybrid system is presented in Fig. 13. Another hybrid option of interest also utilizes a 100 MWe power block but with the CST system including 16 h of storage without torrefaction char systems, with predicted power generation over 5 years of operations shown in Fig. 14. This system has only a slightly lower LCOE than the biomass only system at 15.5 c/kWh, but a high capacity factor at 71.0% and has less complexity than the other proposed hybrid system due to the lack of a torrefaction system.

FIG. 13.

Predicted electricity generation for a combined biomass and CST system with 10 h of storage, torrefaction char facilities and a 100 MWe power block for 2011–2015.

FIG. 13.

Predicted electricity generation for a combined biomass and CST system with 10 h of storage, torrefaction char facilities and a 100 MWe power block for 2011–2015.

Close modal
FIG. 14.

Predicted electricity generation for a combined biomass and CST system with 16 h of storage and a 100 MWe power block for 2011–2015.

FIG. 14.

Predicted electricity generation for a combined biomass and CST system with 16 h of storage and a 100 MWe power block for 2011–2015.

Close modal

Integration costs were neglected in the optimization analysis and precise determination would require a more detailed engineering assessment of the specific plant layout. However, based on a separation between the solar plant and the existing sugar mill of 1.5 km, the costing provided in Ref. 27 provides an estimate that transfer systems for molten salt would add approximately $15 m to the capital cost or from Ref. 23 that the trucking of bagasse would add approximately $1.7 m to the annual operating cost. These correspond to increases of approximately $0.23 and $0.28 to the LCOE in the cases and would allow more flexibility in locating the solar plant without major changes to the system viability. The area around the sugar mill is largely sugar cane cropping land, so if the solar plant was located 10 km away to avoid farmed land, the trucking cost for bagasse would increase the LCOE by approximately $0.45, which would have a more significant impact on viability.

A summary of the predictions for the three technology variants of interest is shown in Fig. 15 in terms of average daily output for months derived from the detailed predictions. This highlights the problems with reliability experienced with the biomass only system, which varies significantly in output for most months excepting August and September. The addition of a hybrid solar component reduces this variability in performance although the capacity factors are clearly reduced in months with low biomass availability. The system incorporating torrefaction char still has a noticeable bias toward improved performance when biomass is available, with the higher storage capacity of the system without char providing greater benefit to output during periods without fresh biomass availability. This generally higher load operation will benefit the efficiency of power block operation and, despite the slightly higher LCOE, is likely to be beneficial to perceptions of reliability for network supply.

FIG. 15.

Predicted monthly output ranges for three technology variants for 2011–2015 with average values indicated.

FIG. 15.

Predicted monthly output ranges for three technology variants for 2011–2015 with average values indicated.

Close modal

There are several reasons why hybrid power generation plants may be attractive prospects for increasing the renewable energy provision to electricity networks. For example, combining more than one source of energy input in an appropriate manner could be a method for improving the reliability and capacity factor of renewable electricity generation. Different approaches to this could include introduction of a second input into an existing plant to increase the usage of smaller generation equipment or construction of a larger new plant using the combined resource to gain improvements in efficiency. The major penalty in adopting a hybrid designs can be summarized as an increase in complexity. The approach taken in developing the hybrid systems in this study has been to maintain as much independence between the systems as is practical, with biomass and CST systems both being capable of supplying superheated steam at the appropriate conditions to the turbine. However, it is not practical to split load of steam reheating and that makes the system reliant on the CST storage always being available. This may mean that a suitable backup, such as natural gas or propane, would be appropriate for the steam reheat duty to ensure system reliability. Similarly, the torrefaction plant requires both flue gas from the biomass furnace and molten salt from the CST plant to operate in the hybrid systems including it. A reliable backup heat source may also be warranted if a plant including torrefaction was constructed.

Torrefaction was included as an option in the hybrid plants due to the large number of research publications that have found significant benefits can be delivered in the flexibility of biomass with minimal cost and efficiency penalties. However, these findings have primarily been through small-scale experimentation and the technology has not been widely commercialized at larger scales. A review of progress in commercial implementation of torrefaction32 provides some succinct views on the reasons for this, with some key issues being that fluctuations in performance, the generation of toxic or explosive gases, a need for flushing with inert gases and auto-ignition of hot char make safe operations at large scales difficult to achieve. Some of these issues may be remedied by the novel approach in the hybrid operation where the generated gases can be disposed of rapidly in a nearby biomass boiler and heating for the torrefaction can utilize a readily controlled and non-flammable fluid (e.g., molten salt or steam), but it is notable that this application would be the biggest torrefaction facility built and would require an extensive engineering and safety assessment before construction. The impact of omitting torrefaction as an option in the hybrid systems is that the solar plant would require a larger storage capacity (16 h) than the plant incorporating torrefaction (10 h) and would have higher LCOE, although this would be similar LCOE to that of the biomass-only plant. The impact of plant specification on financial risk has not been quantitively assessed, but it would appear unlikely that torrefaction would not be favored until larger units have been successfully demonstrated on similar feed materials.

An alternative to torrefaction would be to utilize another biomass stream in the period of the year when bagasse is unavailable. This has been proposed at other sites, but is problematic in this region, where sugar cane is dominant to the extent that crops of other types, including timber, are minimal. In the assessment, it was shown that the cost of bagasse transport over 10 km starts adversely affecting the economics of the process, even with the bagasse considered a free resource. Higher energy density biomass can be transported more efficiently but tends to have a cost of purchase or preparation that, with the transport distances from outside the region being considerable, will make this unreasonable.

The predicted behavior of the different plant configurations assessed in this study suggests that there can be more than one distinctly different option that can provide potentially viable financial performance, even taking the biomass input as non-adjustable. In order to reduce computational intensity, the modeling applied in this study was restricted to stepwise increments in solar plant size, thermal storage capacity and power block, so the different hybrid system designs found to match or better the LCOE of a standalone biomass plant could continue to be optimized in more detailed and specific analyses. It is notable that the performance of the systems during periods where only one energy source is available often results in the power block operating at part load with reduced efficiency for long periods of time. If a specific part load operation is likely to be used for extended periods, then this can be used in specifying the power block characteristics to ensure that the overall system efficiency on an annual basis is optimized. In simple terms, the findings of the study indicate that expenditure on new solar thermal components can be optimized to increase capacity factor over a biomass-only operation to an extent sufficient to warrant the investment and, in the extreme of a large solar plant with large storage capacity, may warrant an increase in the size of the power block. It is likely that the total cost of the system, rather than the LCOE, will provide a limit to the types of system that are acceptable to a commercial operator and this can favor the adoption of smaller plant additions, despite higher LCOE, due to the costs of financing and risk assessments. This would suggest that the configurations that continue to utilize a smaller power block and have relatively incremental improvements in capacity factor without significant improvement in LCOE are more likely to be applied than the implementation of very large solar plants with larger power blocks.

The study has been primarily a modeling activity with only minimal consideration of practical engineering issues, which will undoubtably be numerous if a real application of the technology was to be attempted. Some of these issues are relatively self-evident, such as the need to transfer either biomass, steam or heat transfer fluid to a common power plant location, balancing of the supply steam pressures from different sources and identifying available land of sufficient area for the solar field. Sugar mills are typically closely surrounded by extensive sugar cane fields that are near-level and could suit a large heliostat field, if the economic case supports transfer in usage. In the case of Invicta Mill, a river and rail system are located to the east and the small town of Giru just to the north of the mill. However, extensive flat areas of land to the south-west of the mill that are currently used for sugar cane crops could have potential to support a solar field development and be a more suitable location for reduced glare for the town. It is clear that verifying this would require considerable additional analysis and consultation as even the smallest option of a 30 MWe with 4 h storage solar plant would require an area of almost 2 km2 with a 119 m tower height. While this could fit on land close to Invicta Mill without interfering with existing structures, roads and waterways, the visual impact of the receiver is likely to be clearly visible from the Bruce Highway, a major national road, approximately 2 km to the south of the mill. The larger solar plants identified as of interest, at 80 MWe scale, require land areas in the range of 8–10 km2 with towers exceeding 200 m height and are almost certainly inappropriate for this site. In addition, the area of the mill is prone to flooding in a high proportion of years with 14 major flood events identified in the town of Giru since 1978.33 This flood risk is likely to increase the cost of installation for the heliostat field or make it impractical in this location.

Renewable sources of electricity are often of limited size and availability, so it is of interest to examine methods for increasing both the size and capacity factor while maintaining or improving the financial viability. Adoption of hybrid systems, where different types of renewable input are used in a complementary manner, potentially offers a method for ensuring more reliable output in a cost-effective manner. However, this requires the identification of sites where the different renewable resources exist in sufficient concentration to be economically viable. This is apparently the case for some areas within the Australian sugar cane processing regions, where existing industrial sites already use the bagasse by-product for power generation and solar resources are sufficient to be of interest to solar thermal operations. Combining the two resources adds to the complexity of power plant design and performance prediction, and there is not an ideal match in availability, with optimum solar availability tending to align with the approximately 6-month sugar mill operating period. However, it is possible to design solar thermal plants to integrate with the existing biomass power plants and, in combination with the use of both thermal storage and torrefaction of biomass for storage, to shift resources from periods of high availability to periods of low availability. This can significantly increase the capacity factor compared to biomass-only operation or allow for the use of a larger power block without increasing the LCOE over that of a biomass-only operation. It is unlikely that implementation of the findings of this study at the target site would be simple due to local environment factors, such as land constraints, visual impact of the receiver to motorists, and the flood prone nature of the site. However, the methodology developed provides a valuable tool for assessing other sites, and the successful identification of potentially viable hybrid biomass-solar system designs will speed the process of short-listing sites for future studies.

The Australian Solar Thermal Research Institute (ASTRI) program is supported by the Australian Government through the Australian Renewable Energy Agency (ARENA).

The authors have no conflicts to disclose.

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
IEA (the International Energy Agency)
, see https://www.iea.org/reports/global-energy-co2-status-report-2019 for “
Global Energy and CO2 Status Report 2019
2.
IEA (the International Energy Agency), see
https://www.iea.org/reports/world-energy-outlook-2019 for “
World Energy Outlook 2019
.”
3.
AEC (the Australian Energy Council)
, see https://www.energycouncil.com.au/analysis/where-do-we-need-a-new-dispatchable-power-station/ for “
Where Do We Need a New Dispatchable Power Station?
(
2017
).”
4.
AEMO (Australian Electricity Market Operator), see
https://www.aemo.com.au/-/media/Files/Media_Centre/2018/AEMO-observations_operational-and-market-challenges-to-reliability-and-security-in-the-NEM.pdf for “
AEMO observations: Operational and market challenges to reliability and security in the NEM
(
2018
).”
5.
AEMC (the Australian Electricity Market Commission), see
https://www.aemc.gov.au/energy-system/electricity/electricity-system/reliability for “
Reliability
(
2018
).”
6.
M. A.
Meybodi
and
A. C.
Beath
, “
Impact of cost uncertainties and solar data variations on the economics of central receiver solar power plants: An Australian case study
,”
Renewable Energy
93
,
510
524
(
2016
).
7.
RIRDC (Rural Industries Research and Development Corporation)
, see https://www.agrifutures.com.au/wp-content/uploads/publications/10-078.pdf for “
Overview of Bioenergy in Australia
(
2010
).”
8.
J. H.
Peterseim
,
A.
Herr
,
S.
Miller
,
S.
White
, and
D. A.
O'Connell
, “
Concentrating solar power/alternative fuel hybrid plants: Annual electricity potential and ideal areas in Australia
,”
Energy
68
,
698
711
(
2014
).
9.
E.
Middlehoff
,
L. A.
Furtado
,
J. H.
Peterseim
,
B.
Madden
,
F.
Ximenes
, and
N.
Florin
, “
Hybrid concentrated solar biomass (HCSB) plant for electricity generation in Australia: Design and evaluation of techno-economic and environmental performance
,”
Energy Convers. Manage.
240
,
114244
(
2021
).
10.
J. H.
Peterseim
,
U.
Hellwig
,
A.
Tadros
, and
S.
White
, “
Hybridisation optimization of concentrating solar thermal and biomass power generation facilities
,”
Sol. Energy
99
,
203
214
(
2014
).
11.
Y.
Tanaka
,
S.
Mesfun
,
K.
Umeki
,
A.
Toffolo
,
Y.
Tamaura
, and
K.
Yoshikawa
, “
Thermodynamic performance of a hybrid power generation system using biomass gasification and concentrated solar thermal processes
,”
Appl. Energy
160
,
664
672
(
2015
).
12.
P.
Trayner
, “
Bagasse transport and storage for the Pioneer cogeneration project
,” in
Proceedings of the 2008 Conference of the Australian Society of Sugar Cane Technologists
(
Australian Society of Sugar Cane Technologists
,
2008
) pp.
437
448
.
13.
B. S.
Purchase
,
S.
Rosettenstein
, and
D. V.
Bezuidenhoudt
, “
Challenges and potential solutions for storage of large quantities of bagasse for power generation
,” in
Proceedings of the South African Sugar Technologists Association
(
South African Sugar Technologists Association
,
2013
), Vol.
86
, pp.
495
513
.
14.
B. P.
Lavarack
,
J. J.
Hodgson
,
R.
Broadfoot
,
S.
Vigh
, and
J.
Venning
, “
Improving the energy efficiency of sugar factories: Case study for Pioneer mill
,”
Int. Sugar J.
106
(
1266
),
337
342
(
2004
).
15.
D.
Burns
and
G.
O'Malley
, “
Design modifications for the new economiser installation at Invicta Mill
,”
in Proceedings of 2015 Conference of the Australian Society of Sugar Cane Technologists
(
Australian Society of Sugar Cane Technologists
,
2015
), pp.
446
454
.
16.
NEMweb
, see www.nemweb.com.au for “
Market Data NEMweb
(
2019
).”
17.
See http://www.canegrowers.com.au/page/magazine/magazine/2013 for “
Australian Canegrower
(last accessed October 28,
2013
).”
18.
AREMI (Australian Renewable Energy Mapping Infrastructure)
, see nationalmap.gov.au/renewables/ for “
NationalMap is an online map-based tool to allow easy access to spatial data from Australian government agencies, including AREMI project
(
2019
).”
19.
E. K.
Burin
,
L.
Buranello
,
P. L.
Giudice
,
T.
Vogel
,
K.
Görner
, and
E.
Bazzo
, “
Boosting power output of a sugarcane bagasse cogeneration plant using parabolic trough collectors in a feedwater heating scheme
,”
Appl. Energy
154
,
232
241
(
2015
).
20.
E. K.
Burin
,
T.
Vogel
,
S.
Multhaupt
,
A.
Thelen
,
G.
Oeljeklaus
,
K.
Görner
, and
E.
Bazzo
, “
Thermodynamic and economic evaluation of a solar aided sugarcane bagasse cogeneration power plant
,”
Energy
117
,
416
428
(
2016
).
21.
M. S.
Bhatt
, “
Optimisation of hybrid cane-based combined heat and power-concentrating solar thermal power plants of 30 MW
,”
Int. J. Sustainable Energy
33
,
904
920
(
2014
).
22.
J. C.
Lopez
,
A.
Restrepo
, and
E.
Bazzo
, “
Exergy analysis of the annual operation of a sugarcane cogeneration power plant assisted by linear Fresnel solar collectors
,”
J. Sol. Energy Eng.
140
,
061004
(
2018
).
23.
Tully Sugar Ltd
, see www.tullysugar.com.au/pdf/Tully_Cogeneration_Report_July_2009.pdf for “
Cogeneration project investigation
(
2009
).”
24.
E. F.
Cortes-Rodriguez
,
S. A.
Nebra
, and
J. H.
Sosa-Arnao
, “
Experimental efficiency analysis of sugarcane bagasse boilers based on the first law of thermodynamics
,”
J Braz. Soc. Mech. Sci. Eng.
39
,
1033
1044
(
2017
).
25.
J.
Barroso
,
F.
Barreras
, and
H.
Amaveda
, “
On the optimization of boiler efficiency using bagasse as fuel
,”
Fuel
82
(
12
),
1451
1463
(
2003
).
26.
A.
Anukam
,
S.
Mamphweli
,
P.
Reddy
,
E.
Meyer
, and
O.
Okoh
, “
Pre-processing of sugarcane bagasse for gasification in a downdraft biomass Gasifier system: A comprehensive review
,”
Renewable Sustainable Energy Rev.
66
,
775
801
(
2016
).
27.
B.
Kelly
, “
Advanced thermal storage for central receivers with supercritical coolants
,” Technical Report No. DE-FG36-08GO18149 (
Abengoa Solar Inc
,
2010
).
28.
B.
Acharya
,
I.
Sule
, and
A.
Dutta
, “
A review on advances of torrefaction technologies for biomass processing
,”
Biomass Convers. Biorefin.
2
,
349
369
(
2012
).
29.
Y.
Joshi
,
M. D.
Marcello
,
E.
Krishnamurthy
, and
W.
de Jong
, “
Packed-bed torrefaction of bagasse under inert and oxygenated atmospheres
,”
Energy Fuels
29
,
5078
5087
(
2015
).
30.
P. A.
Hobson
, “
Torrefaction and gasification for high efficiency second-generation biofuel production
,” in
Proceedings of 2009 Conference of the Australian Society of Sugar Cane Technologists
(
Australian Society of Sugar Cane Technologists
,
2009
), pp.
389
399
.
31.
CEPCI (Chemical Engineering Plant Cost Index
), see https://www.chemengonline.com/2019-cepci-updates-january-prelim-and-december-2018-final/for
2018 Annual Value
(
2019
).”
32.
T. P.
Causer
,
Why Hasn't Torrefaction Taken off?
(
Biomass Magazine
,
2019
).
33.
BOM (Australian Bureau of Meteorology)
, see http://www.bom.gov.au/qld/flood/brochures/haughton/haughton.shtml
Flood Warning System for the Haughton River
(
2019
).”