kazulagi / plantfem Goto Github PK

コメント、リクエストやバグの報告などがございましたらお気軽にissueをご活用ください。

Dear readers,

./src/PreprocessingClass has some small bugs which induce bug of boundary conditions. If you find the bug, please inform me in this issue.

Hello.
We start release cycles as performed in other software packages.
Our first version will be 21.10, scheduled for Oct. 2021.
Afterwards, our first Long-Time Support (LTS) version will also be released in Apr. 2022, served as plantFEM 22.04 (LTS).

Enjoy!

Haruka Tomobe
On behalf of the community

I suggest adding the topics agriculture, finite-element-method, fem in the About section.

The python grammar doesn't allow if condition without contents, please fix this by adding a pass or mute this if condition:

    # meshtype="None" should be put in the end
    def create(this,x_num,y_num,x_len,y_len,Le,Lh,Dr,thickness, \
        division,smooth,top,margin,inclineRate,shaperatio,master,\
        slave,x,y,z,dx,dy,dz,coordinate,meshtype="None",):   
        if meshtype.lower()=="Bar1D" :
            # initialize
            # this = Mesh()
            pass

2019/11/3追記

Dear readers,

We receive some requests about the GUI of SiCroF from researchers in the field of crop science and civil engineering. Although we know the importance of a nice GUI, it is difficult to have time and effort to create GUI. Now we would like to create API for Blender to use the Blender platform as a GUI of the SiCroF and some incomplete modules are put in ./src/BlenderClass. If you have some ideas or codes, please help us.

Then, we use the Jupyter-notebook and the Gmsh as GUIs until, at least, next January.

Cheers, Haruka Tomobe

Japanese: SiCroFについて、GUI操作が可能なようにしてほしいというリクエストをたくさん頂いたので、BlenderのGUIを間借りして実装しようとしたら、とても面倒くさいことが発覚した。外部ファイルを読み込むメソッドを実装することすら容易でない。すっかり疲弊したのでしばらく放置します。

SiCroFのGUIは当面のところ、これまで通りjupyter-notebookとGmshでやろうと思います。また、Windows対応についてはコストが大きいため、しばらくできなさそうです。向こう3ヶ月くらいはmacOSとLinux上でのみ動く感じになります。Windows版のデバッガー募集中です。

Dear all,

plantFEM 21.10 is released on 20th Oct. 2021, which is the first official release of this software.

plantFEM 21.10 is a CLI-based software to create and operate cyber research fields for Agriculture and Civil Engineering. You can conduct numerical simulations of deformation, diffusion, reactions in plants and soils based on some PDEs discretized by the FEM. You can use it FOR FREE for private/commercial uses under the MIT license, without any warranty of any kind.

plantFEM also contains pre-processing functions so you can create a cyber environment without any preparations. For example, you can create a single soybean plant with only 3 lines of code, and also can create 2000 ha of the cyber farm only by the 6 lines.

The software is not only for generating/operating cyber farms but also for big-data analysis of vegetation. For this kind of workflow, the software has similar functions and subroutines to NumPy, Pandas, MatLab, and R language. Although there are many things to be added, we are ready for the implementation. Feel free to contact us or write some Issues on Github. We will implement them as soon as possible. Also, questions, comments, and suggestions are welcome.

We are feeling sorry for our poor documentations. We recognize the importance of the documents and tutorials. We will hopefully add them in the next release, 22.04 on April 2022. There are also many things to be implemented, such as a GUI, APIs for Python, JavaScript, R, Matlab and simulators. We seriously need implementers of such fields.

We are financially supported by JSPS Grants-in-Aid for Scientific Research (KAKEN) grant number 17J02383, 20K22599, and Government Funding for Operating Expenses.

Haruka Tomobe, Ph.D.
(E-mail: tomobe.haruka.58m atmark gmail.com)
On behalf of the team

Please use type(FEMSolver_) instead of type(LinearSolver_) because type(LinearSolver_)%fix() is huge spaghetti code, requires unnecessary memory consumption, and has some critical bugs.

By contrast, type(FEMSolver_) is implemented to avoid the problem, which

• uses LAPACK for essential eigensolvers
• is based on more sophisticated frameworks and designs
• minimizes memory consumption
• uses OpenMP for acceleration.

Welcome to our issues.
Please feel free to ask any questions and comments.
Proposals and pull-requests are also welcomed.