How to input parameters in molten salt CSP model

Dear,

Recently I'm trying to reproduce one published paper's work, but i failed after I tried days. Can I ask how should i decide the tower height, receiver height and receiver diameter? These three really important parameters would influence the layout of heliostats. 
When I tried to repeat the author's work, I tried to 'optimize the layout' from initial value (203.33m,20.41m,17.67m) which is quite different from his work.
So I tried to 'generate a layout' by inputting his results (tower height, receiver height and receiver diameter), then I got the same calculated land area, heliostats numbers and reflective area which I successfully repeated his work. But it is difficult to think why he choose these (tower height, receiver height and receiver diameter) values. 
So can I quickly ask how to decide the parameters of these three parameters. Are they related to the solar multiple and receiver thermal power capacity? Cause i read a thesis it's related but the correlation couldn't get these three parameters. Is there a guidance to  approach?
Any suggestions would be really helpful to me. THANK YOU so much!
(By the way, there is no power cycle in my model, so my 'Estimated gross to net conversion factor' and 'Rated cycle conversion efficiency' is 1)

Best regards,
Francesca 

Hi Francesca,

For the Power Tower models, there are a few options on the Heliostat Field page that determine the value of tower height, receiver height, and receiver diameter.

If you know the values of those parameters and the heliostat positions, you can input the dimensions on the Tower and Receiver input page, and import a heliostat position table on the Heliostat Field page. Do not check either Generate heliostat layout using tower dimensions or Optimize heliostat layout and tower dimensions, and do not click either Calculate button next to those check boxes. When you run a simulation, the results will be based on the inputs you provided.

If you know the tower height, receiver height, and receiver diameter, but do not know the number of heliostats or their positions, you can enter the dimensions on the Tower and Receiver page, and check Generate heliostat layout using tower dimensions, but not the Optimize... button. When you click Simulate, SAM will calculate the number of heliostats and their positions, and populate the heliostat position table before it starts the simulation. You can also click the Calculate button next to the check box to calculate the heliostat positions without running a simulation.

Finally, if you want SAM to automatically calculate the tower height, receiver height, and receiver diameter, and also calculate the number of heliostats and their positions, then you should check Optimize heliostat layout and tower dimensions (or click the Calculate button for that option). SAM will use tower and receiver cost inputs on the System Costs page to calculate tower and receiver dimensions and a field layout that achieves the solar multiple, design turbine gross output given the other assumptions on the System Design page and minimizes the cost of those components. SAM will show you the values it calculates and ask if you want to replace your inputs with the optimized values before it runs the simulation.

To replicate someone else's analysis, you really need to know what all of their assumptions are. In addition to the inputs I mention above, design assumptions like the Design point DNI, Solar multiple, Full load hours of storage, and the location, heliostat dimensions, etc. will all affect the tower and field optimization results.

Best regards,
Paul.

Thank you so much.

Dear Paul,

For the CSP power tower molten salt case, if I want to optimize the heliostats layout, do I need to optimize the layout after I already optimize the layout. The operation is : after i finish all the inputs, I CLICK 'optimize solar field geometry', I get a result. Then I CLICK 'optimize solar field geometry' again, a new layout is generated. For some initial values, there is quite difference between the first and second even third result. So do we need to optimize the layout several times? Or we only need to input the initial value and optimize the layout once? 

Best regards,
Francesca

Hi Francesca,

The optimizationalgorithm does not follow a deterministic path because it uses randomization in the step direction as it proceeds. The algorithm also has a relatively flat response surface. Both of these factors can cause the algorithm to return different results given the same set of inputs. A reasonable approach is to run the algorithm several times and choose the best result. This is discussed in SAM's Help topic "Heliostat Field, SolarPILOT Field Layout and Geometry Optimization."

Best regards,
Paul.

Dear Paul,

Sorry about that I didn't point out that I am using 2015.1.30 version. In this version, I choose BOBYQA which I thought it is a deterministic optimization method. So if I am using this deterministic optimization method, for example,
My initial values are : Tower height = 110.67m, Receiver height  = 10.53m, Receiver diameter = 5.92m.
I optimized the layout of heliostats, New values are: Tower height = 90.2963m, Receiver height  = 11.4311m, Receiver diameter = 6.58562m, Heliostats reflective area = 198651 m² . Then i accepted the updates of this layout.
In the same file, based on this updated layout (The number of heliostats and tower height, receiver height etc are already changed), I choose to optimize the layout again: Then the value became: Tower height = 89.8448m, Receiver height  = 11.9455m, Receiver diameter = 6.88198m, Heliostats reflective area = 198203 m2 . 
I'm confused that whether I need to update the layout again after I already optimized it one time when using deterministic method. Is only the result for the first time optimization acceptable or I can optimized the helioastats layout based on the former optimized layout several times?

Hope you are very well.

Best regards,
Francesca