Figure 1 - uploaded by Paul Bannister
Content may be subject to copyright.
6: A view of the White Cliffs solar thermal power plant collector array with all collectors tracking the sun.
Similar publications
Citations
... Skocypec and Romero [2] modelled and tested a molten salt cavity receiver, and found that convective and radiative losses were of similar magnitude. Bannister modelled cavity convective and radiative losses, advancing several concepts for direct-steam receivers with circum-aperture absorbing surfaces [3], (or see Figure 2), concepts later adopted in other designs [4,5]. Steinfeld and Schnubell [6] considered simple isothermal cavity receivers and established optimal cavity diameters and operating temperatures for a specified flux profile. ...
... Initially, simple receiver shapes described by just four parameters were examined [17] (Figure 1). Analysis was then expanded to receivers with an arbitrary number of frusta (as Bannister [3]), including handling of adiabatic wall sections such as at the back of the cavity. ...
... The stochastic optimisation routine (above) was used to identify best-performing configurations, but a manual process was also used in parallel, to incorporate a range of additional design considerations considered to be important but which were not straightforward to include in the model. Some of the adjustments made in response to those additional consideration were as follows: (1) avoided excessively long overall pipe lengths, and respect size limits imposed by the SG4 receiver frame; (2) adjusted diameter and cone-angle of the circum-aperture region to minimise spillage, and provide tolerance for potential degradation in dish optical quality over time; (3) increased downstream pipe diameter in the dry-out and superheated regions, to reduce pressure drops; (4) adjusted angle of internal cavity walls to push the dry-out region beyond the highest-flux part of the cavity; (5) smoothed the cross-sectional profile in the vicinity of the main cavity aperture, to smooth the flux profile in the transition from the circum-aperture to the cavity; (6) smoothed-out the facetted cross-sectional profile to achieve more uniform flux profile, approaching the pear-shaped configuration concept as proposed by Shaui et al [9]; (7) Shape and materials for the passive section covering the centre of the back of the cavity were refined; (8) effects of pipe material, conductivity and surface properties were considered; (9) the option of using the Pyromark® coating inside the cavity was considered. Based on peak operating conditions for the receiver (600°C outlet, 160 bar), 253MA pipe material was selected (with Pyromark coating), and the 10,000 hour creep rupture strength was able to be achieved; creep rupture was the critical design criterion, when following the relevant Australian Standards. ...
An integrated model for an axisymmetric helical-coil tubular cavity receiver is presented, incorporating optical ray-tracing for incident solar flux, radiosity analysis for thermal emissions, computational fluid dynamics for external convection, and a one-dimensional hydrodynamic model for internal flow-boiling of water. A receiver efficiency of 98.7% is calculated, for an inlet/outlet temperature range of 60–500°C, which is the ratio of fluid heating to receiver incident irradiance. The high-efficiency design makes effective use of non-uniform flux in its non-isothermal layout, matching lower temperature regions to areas of lower flux. Full-scale testing of the design will occur in late 2015.
... The increased collector size corresponds to reduced number of collectors, lower network surface area, and lower network pipe material volume. This may mitigate the major problems with slow start-up times and flow control observed in SP1 and STEP (Bannister, 1991). To quantify this advantage, network infrastructure must be characterised to allow for performance simulations to be conducted. ...
We optimise steam network trees that connect Big Dish paraboloidal collectors to a central power block. Exergy costs, pipe material costs, and installation costs are estimated using an exergoeconomic model and used to optimise pipe links in network trees. The optimal network tree is then found for a 10MWe collector field using a genetic algorithm. An optimised tree is found for a 20MWe network. The optimised 20MWe network has additional east-west branches not seen in the optimal 10MWe network, that reduce steam transport costs from network extremities. Exergy costs from heat loss and pressure drop account for approximately 60% of total network cost in both cases. Total network costs are, respectively, 8.9% and 9.5% of potential plant revenue for the 10MWe and the 20MWe networks.
... Siangsukone calculated radiative losses to be approximately 3 kW for operating receiver outlet temperatures between 380 and 450 °C. An emissivity for steel of 0.87 = ε as proposed by Bannister [10] was used. The overall heat transfer coefficient in Siangsukone's data was approximately U=21 W/m²/K. ...
... This profile not only considered the direct radiation from the field but also the effect of multiple reflections inside the cavity. The cavity interior has an absorptivity of 89% and reflectivity of 11% [7], so that after three reflections, only 0.13% of the incident radiation was left unabsorbed. Therefore, the cavity as a whole may be assumed as a black body for practical purposes. ...
... Finally, for the view factor between the top and bottom surfaces, a relation presented by Howell et al [12] was utilized. The emissivity was taken as 0.89 for all surfaces as was proposed by Bannister [7]. ...
A study into the integration of a molten salt based thermal storage system with the ANU SG4 500 m 2 dish solar concentrator was performed. Specifically, the objective was to research the behaviour of molten salt as a heat transfer fluid for the SG4 dish solar concentrator, including its receiver, auxiliary piping and accessories. A numerical model was developed to analyse the heat transfer at the receiver and also to obtain the hydraulic pressure losses and demand curve for the entire system. This case study was restricted to a receiver position of 90º, or facing downwards, as this position requires the highest molten salt flow, and hence gives a maximum sizing for system components. The results indicate that the molten salt temperature profile inside the receiver is dominated by the solar irradiance profile over the cavity surface; with the heat exchange by radiation, conduction and natural convection having a lesser effect. Existing pipework, receiver tubes and rotary joints from the SG4 dish would need to be resized for reduced pressure drops. However, this initial investigation suggests that the use of molten salt as a heat transfer fluid for the ANU 500 m 2 dish is feasible, and warrants further investigation.
Convective air flows are a significant source of thermal loss from tubular cavity receivers in concentrating solar-thermal power (CSP) applications. Reduction in these losses is traditionally achieved by tailoring the cavity geometry, but the potential of this method is limited by the aperture size. The use of active airflow control, in the form of an air curtain, is an established practice to prevent infiltration of cold air through building doorways. Its application in reducing solar receiver convective heat loss is new. In this study, computational fluid dynamics (CFD) simulations are presented for the zero wind case, demonstrating that an optimised air curtain can readily reduce convective losses by more than 45%. A parametric investigation of jet direction and speed indicates that two distinct optimal air curtain flow structures exist. In the first, the jet reduces the size of the convective zone within the cavity by partially sealing the aperture. The optimum velocity range for this case occurs with a low strength jet. At higher jet speeds, the losses are generally set by the flow induced in the cavity and entrainment into the jet. However, a second optimal configuration is discovered for a narrow range of jet parameters, where the entrainment is reduced due to a shift in the stack neutral pressure level, allowing the jet to fully seal the cavity. A physical model is developed, based on the fluid physics of a jet and the 'deflection modulus' concept typically used to characterise air curtains in building heating and ventilation applications. The model has been applied to the solar thermal cavity case, and shows good agreement with the computational results.
The ammonia dissociation reaction is one of a number of reactions which has been investigated for use in closed loop solar thermochemical energy storage systems, over a period of nearly two decades. A recent series of experiments with an electrically heated high pressure ammonia dissociation reactor has validated a two dimensional pseudo-homogenous theoretical reactor model, established rate parameters for the catalyst used, paved the way for a closed loop demonstration and simulated operation of a receiver/reactor under solar operation. The model has subsequently been used to investigate full sized receiver/reactor options for a 20 m2 paraboloidal dish. Technically feasible designs based on: directly irradiated catalyst filled tubes, sodium reflux heat transfer to catalyst tubes and direct absorption of radiation using a windowed pressure vessel, have been identified.
The principal features of a computer simulation used as a design tool for solar thermal receivers for paraboloidal dish collectors are described. Performance predictions derived from this model are outlined and compared with experiments on two receiver designs testing the overall thermal performance and the flux distribution. It is shown that predicted performance is in good agreement and that the model is suitable for design purposes.
The failure of receivers has been one of the main operating problems at the White Cliffs solar thermal power plant. This Technical Note reports the results of an initial investigation that identifies the cause as having been their thermal fatiguing of the tube walls. The fatigue appears to be caused by unstable heat transfer at vapor qualities below the point where critical heat flux is generally exceeded. Methods for avoiding this problem are tested.
The Australian National University has a 400m 2 Paraboloidal dish solar concentrator system, informally named "the Big Dish" that produces superheated steam via a receiver mounted monotube boiler. The system, (shown in Figure 1) which is the worlds largest was constructed in 1994. This paper describes the results of the latest experimental tests plus associated system performance modelling. The system was modelled in the context of feasibility study and performance assessment for multiple dishes, central generation Rankine cycle power plants using the transient simulation pack age TRNSYS. The new custom components of the TRNSYS deck file constructed for this study are the Paraboloidal dish, the steam generating cavity receiver, steam line and steam engine. These component models are based on transient model using the energy balance equation and the empirical derived formulation. Validation test were performed by comparing with the latest experimental results measured with a 1-minute time step.
