Leaderboard

Hi everybody, I recently tried my hands on integrating CAESES with fluent and i think its worth sharing for those who work with fluent and will like to use CAESES parametric models. Please find attached a copy of my project file. its a simple meta-surface elbow i designed with an ellipse curve.Functions for the "width and "height" of the curve are defined. you can have a look at how parameters are set from function curves. I used ICEMCFD as my meshing tool and Fluent as CFD solver. specific files such as the *.rpl and *.jou were used as input files( checkout the attachments). Absolute paths should be changed to relative paths via "getdesigndir()...." (not all paths are necessary!). A *.bat file, RunFluent.bat, was used to run both ICEMCFD and Fluent in batch mode. Snapshots, graphs etc from fluent can be included as results files for post processing. Details on post processing can be found in CAESES tutorials " getting started". You can have a look. Suggestions are welcome. Best regards, Richard "N.B: CAESES is one of the most powerful softwares for tight integration with CFD softwares for DoE and Optimization. All you need to know is to understand how your External CFD software handles its files to know exactly which input and result files to use. it must also have the capability of running in batch mode." Project1_elbow.fdbc

Hi together, please find attached a parametric model of a costa bulb and a feature definition. To recover the energy which gets lost by the hub vortex of the rotating propeller a costa bulb can help. I integrated the rudder bulb setup into the feature "Spade Rudder" which is shipped with CAESES/FFW. The bare hull to which I attached the appendages is also shipped with CAESES/FFW. All you need as input are some propeller parameters. Please find the feature here: baseline > abdy > appendages > feature:rudder Cheers Matthias (fdb file edited, 30.09.2014) containerVesselCosta.fdb

Hi Gabriel, Maybe using fv_all() command would be a better alternative. Please check the picture below; Basically, the fv_all command provides an objectlist. MyCurve.fv_all(0,myPoint:x) The command is applied to a curve. The first argument "0" refers to x-axis The second argument refers to the value on the selected axis. So result will be a list of points that have the same coordinate component value in the referred axis. As seen on the picture above, the curve would have two locations with the same x-coordinate value. Using "at(1)" I pick the second item within the objectlist (0 would be used to pick the first one). And finally I cast the entity to a FVector3 type object. Please let me know if you need further assistance. Cheers Ceyhan

Hi Adam, take a look at BReps -- they are the way to go when combining, uniting, substracting, etc. various parts of your geometry. For boudary conditions you can make use of colors (their names) which can be applied robustly even when topologiacal changes occur in the model during shape variation. You should be able to find everything to get started within the help menu of CAESES. Cheers, Heinrich

Hi Manoj, You can access the CAESES documentation in the Help section in the software CAESES. In the documentation you will find the "Fine Marine Pre-Processing" tutorial under Tutorials > Maritime > Software Connection > Fine Marine Pre-Processing. Cheers, Hedi

Hi Praveen, Please revise the tutorial for the ACT Connection. Within the mentioned document, it should clearly mention that, if SpaceClaim is your default geometry editor you beforehand have to open SpaceClaim and under options unselect the option "Use SpaceClaim Color Tones". Once this little procedure is completed, you can create and start the ACT procedure. Also, as mentioned within the tutorial, if SpaceClaim is your default geometry editor, please do not forget to assign a color to your BREp geometries. This will serve as named selection/s assigned to your domains. Cheers Ceyhan

Hi, you can select the option "Response Surface Optimization in the Optimization workspace. Afterwards, you can check the option "Use Result Pool" It will ask you to select the pools which you want to utilize for the surrogate based optimization after starting the design engine. Please note that if you do not have sufficient samples in your result pool, it will do additional samples before starting the optimization. It will then perform the optimization on the surrogate in the background without calling CFD. After that, a new design will show up in CAESES, which is then evaluated using the CFD to check if it matches the prediction from the surrogate.

Hi Furkan, If you set up a "Response Surface Optimization" and choose to "use result pool", a surroate will be created (if your pool is large enough, otherwise samples will be added first). An optimization on that surrogate is performed automatically in the background (a Genetic Algorithm is used here) and the optimal candidate (or multiple if you choose so for a multi-objective problem) is returned. The design you see created in CAESES is therefore either an additional sample or already the potential optimum. Check the dakota.out file in your design directory to see the details. Also, the finaldata file will show you the predicted performance of the optimum... If you keep this Algorithm running, the actual (CFD) result(s) of the optimum-candidate(s) will be added to the pool for the next iteration and the process repeats. Cheers, Heinrich

HI everyone， I'm having a hard time. I created the blades and tips. As shown in Fig. However, when I create the propeller with Model-propeller-propller, I only get incomplete blades. The tips were not created together. This is shown in the image below. Also, due to the language, I may have a problem with the description not being clear, so please forgive me. Looking forward to your answer. prop_para.cdbc

Hi Furkan, at the computation object, there is an option to set a list of constraints. If one of these constraints is violated, the computation will not be executed for that design. Cheers, Mattia

Hi Yukai, The force in the last line is the converged value from the solver. I would suggest to use a python code only for post-processing. For example, if you want to plot the convergence history of the forces over OpenFOAM iterations. By parsing various values, such as force in x-axis to parameters and establishing a robust integration CAESES with OpenFOAM, then you can continue with a Design of Experiments and/ or with a Optimization process. Best Regards, Andreas

Hi Carlos, In principle you are right. The SSH Resource Manager is part of the "entire functionality of CAESES". However, due to the low demand and the current revision of that piece of software, we only grant access on request. So please write your demand via email at license@friendship-systems.com with a short explanation on what you want to achieve with it. And please remind: "CAESES® must be installed on a personal PC/notebook rather than on University computers." "By granting you this free license we expect you to send us a summary report of the work that you have accomplished with CAESES® or a PDF copy of your thesis if applicable." Best, Carl

Hi Gustav, you could create a point through the console with: point p(curve.fv(2,zVal)) Instead of "curve" you'll have to choose your particular curve and zVal would be your z value. "2" corresponds to the z-axis (0=x, 1=y, 2=z).

Hi Yukai, you can set the sub folder by adding it to the file name of your template like this: If you have an allrun script which you execute for each design this would normally be executed in the same location as your STL file is exported to. Hence, typically referring to it by its name instead of full path works. Alternatively, you could add an entry an specify a path with getResultsDir() to have it adjust dynamically... Cheers, Heinrich

Hi CJ, I guess I would just create a planar BRep intersection (BRep-based curves) and from this get the area. See attached project... Cheers, Heinrich fishingVessel.cdb

Hi Carlos, the SSH Resource manager is an addon to CAESES which has to be purchased separately. After purchase, the SSH Resource Manager Server must be installed on a host in your network, e.g. the cluster main node. If you have installed this already, please ask your network manager where it is installed. Best regards, Hedi

Please insert a reasonable knotspacing value in the BRep and datareduction afterthe transformation. The transformation might otherwise rip the geometry apart, since it moves the control points of the vessels representation. In the flat of bottom area there might be only a few and that needs to be refined. Cheers Claus

Hey, I' ve created two feature definitions for importing and exporting of point data. Import: The file is read line by line. To create the point as objects you have to create the feature defintion via execute Defintion (right click). The point data in the csv or txt file has to be separated by white space or comma by default. You will find attached the project, the feature definitions and two example files as a zip file. Best regards Karsten ImportPointData.fdf writeCsv.fdb exampleFiles.zip ReadAndWriteFiles.fdb

The model is unfortunately very bad. There are big gaps between several surfaces. I will see if I can find a repaired one. Best Regards Claus

I think so ...

Hi David, You can display calculated values in the input dialog by defining an additional argument and editing the advanced settings of the feature argument. Here is a video tutorial on how to show result values in the input dialog. Drop down list for input values In addition, you can use String Options to choose an input value from a drop down list. Make sure you switch off the field "Allow Expression" when you use it. By using a "switch" you can define the method you would like to use for your calculations. FString chosenmethod(method) switch (chosenmethod) case "A" echo("using calculation method A") case "B" echo("using calculation method B") case "C" echo("using calculation method C") endswitch Here is the feature definition example that includes the result values in the input dialog and the switch cases with the string options for your reference. DynamicArguments_WithResultField_DropDownListOptions.fdf Have a nice day. Hedi

There is something, that seems to be related to the weight of the third point. But maybe it also comes from the curve intersetion point, there seems to be a discontinuity in the definition, see screenshot of control polygon. That is where the "folding" appears in the visualization.

Hi David, You are on the right track 🙂 In the "Advanced" Edit Menu of the input Arguments you can set the "Is Hidden Condition". You need to make sure that each time you insert a first input value, your editor gets refreshed. So for your first input argument you toggle the "refreshes editor" button (see screenshot). For the following dynamic input fields you can reference the first input value for example with this.getARGUMENTNAME() and add a condition to it. I will attach a simple example feature definition here for you to test. DynamicArguments.fdf I hope this helps. Cheers, Hedi

Hey, I am doing some hydrostatic calculations. I did not find a way to calculate the wetted surface of a boat in the hydrostatic tool. How do you calculate the wetted surface? Thanks.

Hi together, The software CAESES is a CAD and optimization platform. For students and PhD students there are free non-commercial licenses available. In addition, there are low-price offers for start-ups and smaller companies. CAESES can be used for 2D and 3D parametric modeling, see this link for some information about its CAD capabilities. Here are some screenshots: Compared to traditional CAD systems, CAESES is a bit different. It comes with a strict object-oriented approach, i.e. the user sets up dependencies between objects and these dependencies are then kept. This makes it easy to automate the geometry generation process. Here are some features of CAESES: Full 2D and 3D modeling capabilities (NURBS-based)Roughly 20 curve types and 15 surface typesStandard transformations (translation, rotation, scaling)Writing of custom features and functionsBoolean operationsTrimmingFillets between surfacesMorphing functionality for deformation of existing geometrySurface tessellation control through e.g. trimesh objects to create and fine-tune custom STL dataCommon import and export formats e.g. IGES, STEP, PARASOLID, STLIndustry-specific modules for blade and ship designBatch mode for non-GUI (hidden) geometry generation in the background Cheers Joerg LAST UPDATE: NOV 2017

Hi Nikolas, 1) I can suggest you a work-around; Please check the attached pictures and project; What I have performed is creating a BRep out of the entire exported geometry and then to each BRep I have added an "Remove Faces" operation where I kept only the specified colored surfaces. I have disabled the scope export line inside the fsc file and then included a few lines for the export of the specific surfaces. Please let me know if this resolves your problem. 2) Can you please be more specific about the "cracked" file? Do you refer to the main stl file within which exists the data of the other extracted stl files? Cheers Ceyhan axialfan_sample.cdb

Hi Nikolas, The naming for the "STL (Extract Colors)" export should be as follows; <base name provided for exported stl>+"_"+<color name>+".stl" In the example provided, I gave a name "fanExported.stl". So the base name would be "fanExported" For the blades, during the blade creation process I have assigned a user created color with a name "w_blades" As a result the stl file for the blades ends up being "fanExported_w_blades.stl" To conclude, if you would like to change the stl file names for the exported geometry, one has to modify the assigned color names. I think the Tutorial; Geometry Modelling > BRep and Solids refers to color assignment to operations. Please let me know if you have further questions. Cheers Ceyhan

Hi Nicolas, The Export type you require is "STL (Extract Colors)" In CAESES, please create a scope, let's say "02_Export". Then please locate the BRep/s you want to be exported. Please be sure that, the geometrical entities you want to be seperated do have separate colors assigned. Then select the folder, change the file type to "exportStlExtractColors" and provide a file name. Using this procedure, whenever you create an fsc file, the export information will be written automatically. Please let me know if you have further questions. Cheers Ceyhan

Hi Assiouras, it's quite simple: fdb is the file format of the CAESES4.x and older releases, while cdb is the current format of CAESES5.x. There is no backwards compatibility (you can open an fdb in CAESES5 and save it as cdb, but not the other way around), hence the different file type. In addition, anything with an additional "c" indicates a non-commercial license and therefore cannot be opened with a commercial license. Best regards, Heinrich

Hi Farzan, The operation "Solid From Intersections" tries to obtain a solid "water-tight" geometry from the combination of several provided geometries. The reason why the operation provided some unsatisfactory result maybe due to your provided inputs were not much intersecting but flush maybe? From the pictures it is not quite clear but is there even some fillet? I will recommend you to extend the blade geometries a bit so that there can be a proper intersection among the geometries and create the fillet afterwards as a separate operation. Please let me know if you need further assistance. Cheers Ceyhan

I have used the Hydrostatic tool with closed trimeshes. The sections displayed below the hull look broken like in the attached example. If I use an open trimesh, the sections look alright. In both cases the calculated values are the same but I am not sure what the broken sections want to tell me. In my current project I only got .stl-files with closed trimeshes of hulls so i can not use the surfaces to do the calculations. surface_trimesh_trimeshclosed.fdbc

Hi, Sometimes diagrams with important data are available only as graphics files (e.g. after scanning from a report or upon exporting from a pdf-file). CAESES / FRIENDSHIP-Framework can be quickly utilized to read off data from diagrams with high accuracy. To do so, import the graphics as an png-file within a "GL Picture Frame" (1st step). Upon setting the scales of abscissa (x-axis) and ordinate (y-axis) (2nd step) you can readily position a point in your diagram and get your x- and y-coordinates (3rd step). (You may want to use such a point to check the level of accuracy.) Furthermore, you can approximate a graph with a curve, say a B-spline curve (4th step) or interpolate it (5th step). Using the curve representation you can "inquire" the y-value for any given x-value (6th step). The attached fdb-project illustrates the work flow for an imaginary speed-power curve. In addition to using CAESES / FRIENDSHIP-Framework to extract data points from diagrams, you can follow the same approach to replicate a lines plan of a boat, yacht or ship, circumventing classic digitization. Offset data are thus produced effectively. Nice side effect is that you can adjust selected points, for instance to improve accuracy of lines remodeling and "repair" apparent outliers. Kind regards, Stefan ExampleDiagram.zip

My doubt is actually not hard, but with the actualization i lose a feature i used to use. Here is the situation: I need to create some points on this curve and each point on the top has its correspondent on the bottom exactly on the same elevation in X axis. To do this i need to use the parameter witch corresponds to the elevation required, but in X axis the curve has 2 possible coordinates so that once i insert the function related to the creation of the point only one coordinate (on the top) is selected. Previously i could use "PointIntersection2D" and select a index, witch allowed me to choose any intersection the function would find. Here is an Image of the situation: The question is: is there any function or any oder way to select another index on the curve? P.S.: I could split the curve in two new curves, and repeat the steps, but i'd like to know if there is a solution like the feature i used to use.

Hi together, If you want to change the rotational direction of a propeller blade (type FGenericBlade), then you can simply switch the orientation. See the attachment for more information. Cheers Joerg

The easiest way to flip the normal of a given surface is by reversing one of the surface directions u or v. In a NURBS/B-Spline surface this can be done with the command .reverse(1,0) or .reverse(0,1). In a Metasurface you can reverse start and end position or you create an image surface and in the u or v domain you enter [1,0] instead of [0,1].

Parametric models are typically built from various geometric or non-geometric entities, e.g. a projection curve depends on the curve that is going to be projected and the surface it is supposed to lay on. In most programs the user creates the desired object first (in this case the projection curve) - and is subsequently asked to supply the necessary objects (surface, curve and possible projection direction) until the configuration is complete. In CAESES/FFW missing information is indicated by a * next to the required attribute and you can set the relationship via drag and drop or typing. However, if you have selected a surface and a curve already when creating the projection curve, they will be automatically associated to the attributes. Note: Whenever the selection set fits to a creator called, the attributes will be set immediately. For every object you will find a list of available creators in the type documentation.

Hi, One of the famous hull forms found in literature is the so-called Wigley hull. It is mathematically defined, see attached formula, and used regularly for tests and validation work. By definition the Wigley hull is a (simple) fully parametric model with beam, draft and height as parameters to control the shape (often normalized by length). A realization of the Wigley hull via a MetaSurface that captures the mathematical formula is given in the attached CAESES project. In addition, some partially parametric modifications are shown, namely, Lackenby type swinging of sections that is realized via a DeltaShift. (Please note that a Generalized Lackenby variation would also be available but was not used here in order to keep the project light.) If you need the hull for your CFD validation work you can use the various exports for panels, offsets, STL, iges etc. More information about ship hull design can be found in the marine section of the CAESES website. Kind regards, Stefan standardWigley.fdb

Hi, please find attached an example for a so-called black box optimization problem. For optimization problems in technical applications it often happens that there is no information on the target function. The research area for the optimization is inside a "black box". So it is very difficult to find the optimum. Time is money. For this reason it is very important to investigate the unknown area quickly and effectively. In order to solve this kind of problem Friendship-Framework contains many optimization algorithms ranging from single-objective strategies for fast and simple studies to multi-objective techniques to investigate a non-linear design space with many local minima and maxima. In the attached file you can try out and compare different algorithms by yourself. Inside the black box (curtain) is a b-spline surface with some local minima and maxima. With the two design variables "usValue" and "vsValue" you can search for a minima by your own. The vertical transparent red surface represents a constraint. Feasible solution can be found only on the right side of the constraint. Sobol, TSearch and NelderMeadSimplex are already performed. Sort the column "objective" within the result table for the smallest value and check the first feasible design. Use the values of "usValue" and "vsValue" from the result table to see where the minimum is found on the b-spline surface (of course you can untoggle the visibility of the blackbox by clicking on the scope "curtain" to see the b-spline surface). Use the feature "optiVisualization" to visualize the way of the optimization algorithm on the surface. Please feel free to try out your own optimization setups. Maybe set the start values "usValue" and "vsValue" on a local minima and see what happens. File for CAESES versions below 4.0.3: optimization.fdb Project file for CAESES 4.1.x: optimization41x.fdb Cheers Matthias

Dear CAESES/FFW Users, I would like to use CAESES with OpenFOAM: - make very simple parametric geometry - use of snappyHexMesh (as mesh generation tool) - use of OpenFOAM (as flow solver) - modify some basic geometry parameters of the geometry - rerun CFD computations with the modified geometry - use of post-processing tools to compare CFD results between configurations I was wondering if someone has already done this kind of exercise and maybe could share it trough the CAESES/FFW forum ? Best regards, Stephane Sanchi.

Please ensure your export folder has the correct setting as shown in my first screenshot from monday. Furthermore, you could have a closer look what is exported by Caeses when updating a parameter in this file path:

Hi Rohan, yes this is possible. You can create a shortest distance line: Best regards Carsten

Dear Rohit, the design variables are used to describe/modify the functions given in 01_blade/functions. You can see how they affect their shape in 3D: You can find detailed information on their exact definitions in the documentations. To get there, just click the little blade icon next to the name of the blade: Best regards, Heinrich

If you created the other surfaces with "Coons Patch" by use of 4 boundaries, you could divide the current 2 lines into 4 lines with "image curves" and use them as boundaries for a "Coons Patch"

Hi Roopesh, find attached the sample model. best regards Carsten pistonbowl.fdb

Dear Matheus, you are absolutely right. Indeed, complex models can benefit a lot from parallel computing. Unfortunately, up to the latest CAESES 4.x no such option is available (except for the settings referring to multiple external tools running in parallel during an optimization that you mentioned). However, one of the mayor changes of our upcoming release CAESES 5, is in fact the full parallelization. If you are particularly interested in this feature you can stay tuned via our newsletter here: https://www.caeses.com/news/newsletter/ P.S: Often times it makes sense to take a closer look at what particular part of your geometry slows down the update -- you can use the profiler (go to help > start profiling, then trigger a model refresh, i.e. by changing a design variable and then go to help > stop profiling) to get statistics of the most time consuming objects in your model. Cheers, Heinrich

Hi Mlysyshyn, Please find attached the modified version. Some modifications I have performed; 1) Within the Runner for the "Local Application" I have selected the AllRun.bat which I have created. Please note that I am not using any arguments. 2) The RunAll.bat executable includes commands within the "C:\OpenFOAM\19.10\cygwin64\Cygwin.bat" executable (lines 4-9) and the path to my script file which is the AllRun.sh (line 9) 3) Finally made a little modification to the script file. The final configuration seems to be working but didn't pay attention to the OpenFoam setup. Cheers Ceyhan Sduct_with_openfoam_2_FSYS.fdbc

Hi Mlysyshyn, Can you please share your project file if it is not confidential? You can also send it to erdem@friendship-systems.com so that I can give a look at your SoftwareConnector setup. Cheers Ceyhan

Hi Mlysyshyn, I can see that the files are are moved to their related locations. Can you please share any console output? Or some OpenFoam logs where the problem can be tracked? Cheers Ceyhan

Hi folks please find attached a Feature Definition for a "Bulbous Bow Shape Analysis" and a Feature Definition which can be used in a CurveEngine for creating a MetaSurface. shapeAnalysis_Bulbous_Bow.fdf bulbous_bow_section.fdf [edited 05.10.14 - tangent analysis included] In order to replace a Bulbous-Bow-IGES-Import by a fully parametric MetaSurface you can use this Feature by execute the following steps: Create a Surface Group including all the IGES-surfacesCreate a new Feature Definition and reload the Analysis-Feature, apply and create the Feature by right click on the Feature Definition>Create FeaturePass the Surface Group to the Analysis-Feature and enter (if necessary) start- and end-position in x-direction of the bulbous bowAdditionally give a number of offsets (30-50 should be a good choice)Run the Feature This will create function curves which can be used to create a MetaSurface. The next steps should be: Re-model the function curves by e.g. FSplineCurvesCreate parameters for some values of the new curves, like start/end position, start/end tangent or area Create a CurveEngineCreate a new Feature Definition, reload the Section-Feature and applySelect the Bulbous-Bow-Section-Feature and pass the new function curves to the EngineAdditionally you can set start- end end-tangent valuesCreate a MetaSurface, select the Curve Engine and set the base positions according to the start- and end-position in x-directionNow you will have a single Metasurface, which is parametric and ready for a design study. Cheers Matthias

Hi folks, please find attached a fully parametric model of a WED - Wake Equalizing Duct. Such a device can increase the wake homogeneity and the hull propeller efficiency. In the project (baseline > abdy > appendages > duct) you can find three sets of parameters (in the scope 04Parameters): The first set controls the path of the duct which consists of three parts, an upper flat, an round mid-part and an lower flat. The parameters for the path are shown in the figure below. The second set of parameters controls the section of the duct. The standard parameters for a NACA 4 digit profile (chord length, maximum thickness, camber and camber position) are kept constant but can easily be replaced by functions. The angle of attack can be varied for three positions (shown in the figure below). The third set of parameters controls the global position of the duct. With xpos and zPos you can move the duct to the destination. With totalScale you can scale the duct. Taking into account the direction of the skeg surface (the red line shown in the figure below is a surface curve on the skeg at the height of the center of the duct) the duct aligns to the surface and to the surrounding flow. With the parameter addFlowAnlge you can apply an additional angle which results in a rotation around the z-Axis. The parameter AOATotal effects a rotation around the y-Axis. For external CFD computations you can find a watertight trimesh (imDuctMesh) and the parameter openEdges which checks the watertightness. This parameter can be used as a constraint for the computation. Cheers Matthias WED.fdb