what is the design point used by SAM in "CSP molten Salt Power Tower

Thanks Paul, after reviewing the three documents (DELSOL3 user's manual, “Simulation and Predictive Performance Modeling of Utility-Scale Central Receiver System Power Plants” and “Simulation of Direct Steam Power Tower Concentrated Solar Plant”) I still require some clarifications at this regard. I would be very grateful if you could help me.

First of all, I would like to know to which extent the models explained on these documents are consistent with the current version of SAM. Certain parameters included on these documents do not seem to be included in SAM. I mention some of them:

- Does SAM consider the seasonal and daily variation of sun positions and sunshape described on section III.B of DELSOL3’s manual? In such case, in which section? Is there any predetermined average factor to apply to the incident radiation?
- Where does SAM include the cosine effect, the shadowing and blocking of the heliostats and the tower shadow effect? How are they calculated?
- What are the input parameters in relation to the field layout? In section 2.1.2. (Wagner 2008), parameters seem to be the radial and azimuthal distances between heliostats, but SAM demands the number of heliostat per azimuthal and radial zone. Could you clarify this point please?
- How is the receiver spillage considered? Is there any average factor?
- What do the “hourly factors” on SAM’s section “Performance Adjustment” refer to?
- Thermal radiative and convective losses are considered as described on “Simulation of Direct Steam Power Tower Concentrated Solar Plant”? Is there input parameter at this regard or the model makes the calculations its own?

Thank you very much in advance!

Dear Paul,

In relation to the atmospheric attenuation, I still have one more question...

I have just read this article ( ases.conference-services.net/resources/252/2859/pdf/SOLAR2012_0765_full%20paper.pdf ) and it mentions an attenuation calculating method from visibility and on the conclusions it is said:

"Equation 7 provides a model for attenuation losses between
heliostat and receiver in a central receiver system that depends only
on DNI measurements. As DNI measurements are generally
available at any prospective CSP site or existing CSP plant such a
model can be readily used in CSP production modeling. The model
can be incorporated in available models such as the National
Renewable Energy Laboratory’s Solar Advisor Model (SAM)."

Can really SAM consider the equation 7 or it is only considering the expressions you talked me about?

Hello Juan,
Please see your questions below along with responses.

Best,
Mike Wagner
NREL


1. Does SAM consider the seasonal and daily variation of sun positions and sunshape described on section III.B of DELSOL3’s manual? In such case, in which section? Is there any predetermined average factor to apply to the incident radiation?
---
SAM and DELSOL consider the sun position in all calculations in a couple of different ways. During layout and optimization, DELSOL simulates 5 days in the half-year (assuming corresponding days in the other half of the year have symmetric performance), and the afternoon hours of the selected 5 days, again assuming symmetry between morning and afternoon performance. It bases the optimization on the results of this pseudo-annual simulation. The performance calculated during each of these optimization hours considers blocking/shadowing, cosine, attenuation, and image intercept optical losses based on the heliostat (zone) position, tower height, receiver size and orientation, and sun position. Radiation during each of the simulated hours is assumed to be clear sky, and the Meinel model (DELSOL variable ‘INSOL=0’) is used to estimate clear sky DNI for each hour/day of the simulation.

DELSOL is also used to generate an optical efficiency table that indicates total field optical efficiency (including all optical efficiency losses) as a function of sun position. This information is fed to SAM for annual hourly calculations, and the solar field optical efficiency is determined by interpolating this table. SAM determines sun position using algorithms documented elsewhere.

Finally, SAM simulates receiver performance that depends in part on the flux distribution provided by the heliostat field onto the receiver surface. DELSOL is also used to generate rough flux density maps for a number of hours of the year (see Wagner 2008 for more information). The flux maps, like the efficiency matrix, include all optical losses.

2. Where does SAM include the cosine effect, the shadowing and blocking of the heliostats and the tower shadow effect? How are they calculated?
---
See answer above. Basically, SAM relies on the optical efficiency calculations done by DELSOL and passed as an optical efficiency matrix as a function of sun position. All aspects of solar field optical performance are calculated by DELSOL.

3. What are the input parameters in relation to the field layout? In section 2.1.2. (Wagner 2008), parameters seem to be the radial and azimuthal distances between heliostats, but SAM demands the number of heliostat per azimuthal and radial zone. Could you clarify this point please?
---
DELSOL operates modeling the field as a small number of optical “zones”, as you mention. The zone is characterized by “heliostat density”, which is simply the total reflective area of the heliostats within the zone divided by the land area of the zone. Values that are displayed in SAM are calculated based on reported zonal density values from DELSOL. The user can change the number of heliostats in any zone using the heliostat field table, and these values are then converted into densities and passed back to DELSOL for use in optical efficiency/flux calculations. These performance calculations are done with DELSOL’s Field option ‘IUSERF=2’ and specified ‘DENSIT(K,L)’ values. See the DELSOL user manual pp. 179 & 181-182 for details.

4. How is the receiver spillage considered? Is there any average factor?
---
Spillage is calculated differently during a simulation depending on whether the field is being optimized or whether detailed optical efficiency/flux performance is being simulated. During layout simulations, the heliostats are all assumed to be aimed at the center point of the receiver to maximize flux absorption. During detailed performance simulations, the flux is spread out over the receiver surface to maintain peak flux below the allowable limit (specified by the user).

The intercept factor (1 – spillage) is determined by integrating the analytical flux density model over the receiver (more detail in DELSOL user manual, Dellin 1979 paper in DELSOL references) in 2 dimensions. The flux density model includes image shape effects from heliostat canting panels, focusing of the panels, optical errors specified by the user, heliostat position, sun shape, receiver position and orientation, and heliostat shape. The model is detailed and carefully formulated to include all significant effects, and has been validated against measured image data and modeled ray-tracing data successfully. It can take a while to understand how the analytical model was developed and how it should be used, but the references above should give you a starting point if you’re interested.

5. What do the “hourly factors” on SAM’s section “Performance Adjustment” refer to?
---
This table allows you to manually adjust the net electricity production during any hour of the year (by month). If you leave all values at 1.0, then the net annual electricity production will equal the production reported by the detailed performance model. If you alter the values, the output during each hour of the year will be multiplied by the corresponding factor to adjust the total annual output.

6. Thermal radiative and convective losses are considered as described on “Simulation of Direct Steam Power Tower Concentrated Solar Plant”? Is there input parameter at this regard or the model makes the calculations its own?
---
Yes. The losses are calculated based on convective and radiative behavior modeled from first principles and relevant semi-empirical correlations. You can adjust the heat loss from the receiver to match data with the “Heat loss factor” on the Tower and Receiver page. The factor multiplies both the convective and radiative heat loss coefficients calculated in the model.

7. Can really SAM consider the equation 7 or it is only considering the expressions you talked me about?
---
The model described in the Sengupta/Wagner paper is not currently included in SAM or DELSOL. Attenuation loss is modeled using the DELSOL3 built-in 25km visibility polynomial model.

Thanks a lot Mike for your responses.

Best regards.

does anyone have a sample design on a 10MW molten salt solar tower plant?

Dear Mark,

A basic starting point would be to use the default 100 MW system in SAM, and change the power cycle design turbine gross output to 12 MWe (for a 10 MW nameplate capacity), and also adjust the thermal storage "full load hours of TES" to whatever storage capacity you want to consider.

Then, on the Heliostat Field page, generate a new heliostat layout and optimize the solar field geometry and apply the optimized values to your inputs using the buttons provided. When you run a simulation, yoiu'll get results for a 10 MW system.

From that starting point, you can further refine the details of design as you consider different options.

Best regards,
Paul.