Optimisation of collector locations and pipe networks for dish–Rankine power plants

The thermal efficiency of the collector field of a concentrating solar thermal power plant is strongly dependent on the geometry of the chosen collector. Paraboloidal dish collectors are realisations of the ideal reflective imaging concentrator geometry, and offer the highest possible optical efficiency of all collector geometries. The feasibility of multi-megawatt scale dish–Rankine power plants incorporating Big Dish collectors is investigated in this thesis by optimising collector placement, energy transport network pipe diameters, and network branching, accounting for shading, network heat loss, network pressure drop, network capital costs, and land costs. The thermal performance of the optimised collector field is assessed against respective fields comprised of linear Fresnel collectors, parabolic trough collectors, and central receiver systems. We find that the annual thermal energy delivered to the central power block by the optimised dish-Rankine field is approximately 120% of that delivered by a central receiver system with the same nameplate generation capacity, which is the technology that offers the next-highest field thermal efficiency. Field optimisation comprises optimisation of dish collector placement for minimal annual shading for a range of ground cover ratios. We find that collectors arranged in diamond-shaped layouts collect up to 1.4 percentage points more annual solar energy than those in rectangular layouts for ground cover ratios above 0.23. Using a layout informed by our shading optimisation work, we use a nested optimisation to optimise energy transport network pipe diameters for each network used in a genetic optimisation of network branching. Network exergy costs, pipe material costs, and installation costs are estimated using an exergoeconomic model. The optimal network tree is then found for a 10 MWe collector field and a 20 MWe network. Extra east-west pipe branches correspond to a lower cost for the 20 MWe field, despite the longer overall network pipe length. Exergy costs from heat loss and pressure drop account for approximately 60% of total network cost in both cases, where the remaining fraction is comprised of pipe material costs, installation costs, insulations costs, and pumping irreversibilities. Total annual network costs are, respectively, 8.9% and 9.5% of potential plant revenue for the 10 MWe and the 20 MWe networks. Annual simulations of power plant performance using the optimal 20 MWe network are performed for a range of collector separations. The rectangular layout corresponding to the highest annual plant income is relatively closely spaced in the north–south direction, and east–west separation is sensitive to land cost. The annual thermal performance of the optimised dish field is compared to values found in the literature for parabolic trough, linear fresnel, and central receiver systems. Thermal efficiencies associated with each collector type enable the comparison between the annual thermal energy delivered by equivalent power plants utilising the respective collectors.

Author dc:creator

• Cumpston, Jeffrey

• Solar Thermal Power
• Dish Collectors
• Big Dish
• Dish-Rankine Plants
• Collector Geometry
• Network Optimization
• Network Optimisation
• Branching Optimization
• Branching Optimisation
• Shading Minimization
• Shading Minimisation
• Ground Cover
• Thermal Efficiency
• Exergoeconomics
• Pipe Costs
• Heat Loss
• Pressure Drop
• Exergy Costs
• Receiver Systems
• Trough Collectors
• Fresnel Collectors
• Field Layout
• Plant Simulation

Language dc:language.iso

en

Handle dc:identifier.uri

https://hdl.handle.net/1885/733716314

OAI identifier oai:identifier

oai:openresearch-repository.anu.edu.au:1885/733716314