export salome mesh to FeNICS mesh

Posted on September 24, 2018 by news

Salome installation

download

Salome platform 8.5 binary for Ubuntu 16.04 (64bit)

Salome-meca is the combination of salome and code_aster (FEA), it is also possible to manually add the Code_saturne plugin for CFD

installation

extract this compressed file to a position like /opt, it has its own python2.7 and most not-so-common dependent packages Do read the readme to check if you have installed all the essential packages.

run the program

run the program in a terminal (to catch output/error message) /path_to/salome.py which set env var.


Start SALOME, elapsed time :   2.1 seconds

Fatal Python error: Py_Initialize: Unable to get the locale encoding

  File "/opt/SALOME-8.5.0-UB16.04-SRC/BINARIES-UB16.04/Python/lib/python2.7/encodings/__init__.py", line 123

    raise CodecRegistryError,\

Start SALOME, elapsed time : 2.1 seconds

Fatal Python error: Py_Initialize: Unable to get the locale encoding

File "/opt/SALOME-8.5.0-UB16.04-SRC/BINARIES-UB16.04/Python/lib/python2.7/encodings/__init__.py", line 123

raise CodecRegistryError,\

to fix this error on my machine, I need unset PYTHONPATH in my “~/.bashrc”

BTW, 3D The hardware acceleration is disable in my virtualbox,

“~/.bashrc”

#3D window is not working if 3D accel is turned on for vbox

export LIBGL_ALWAYS_SOFTWARE=1

macro recording function

The operation within GUI can be recorded into python script by “File-> dump study”, this saved python script can be loaded later, or adapted for used in batch mode.

NoteBook variables

variale can only be simple math expression, math.pi math.sin() are not supported.

partition, group and save multi-zone mesh

submesh is used

=============================================================

mesh conversion from salome to Fenics

meshio-convert (from meshio package) can translate the salome Med format into dolfin format, however, it seems boundary and subdomains are not converted.

Here is another route: Salome>Gmsh>Fenics, inspired by [I was finally able to import an assembly mesh into OpenFOAM format with the following Salome>Gmsh>OpenFOAM process]https://www.cfd-online.com/Forums/openfoam-meshing/82210-importing-multiple-meshes.html


-Using Salome:

1. Within the Geometry module explode surfaces needed for Boundary Conditions and renamed them

2. Fused all parts together into a single assembly

3. Partitioned this assembly using each of the parts included

4. Created 'Volume Groups' for each partitioned part

5. Move to the Meshing module and mesh the whole partitioned assembly (optional: make sub-meshes of the parts to better control mesh sizing)

6. Make 'Volume Groups' of each part within the mesh

7. Make 'Surface Groups' for all the boundary conditions

8. Export mesh as MED file (need to check 'Automatically Create Groups' box)



-Using Gmsh

9. Open mesh file in Gmsh and 'Save As' Gmsh mesh file (.msh file extension, make sure file is saved as Version 2.0 ACSII and un-check all check boxes)

10. Move Gmsh mesh file into an empty case file (make sure to have constant-&gt;polyMesh (empty) and system folder with necessary files)



-OpenFOAM mesh conversion/manipulation tools

11. Run [&gt;&gt; gmshToFoam] in OpenFOAM terminal within the case directory

12. Run [&gt;&gt; splitMeshRegions -cellZonesOnly -overwrite] (this is only used if there are multiple parts (volumes) to be split up)

-Using Salome:

1. Within the Geometry module explode surfaces needed for Boundary Conditions and renamed them

2. Fused all parts together into a single assembly

3. Partitioned this assembly using each of the parts included

4. Created 'Volume Groups' for each partitioned part

5. Move to the Meshing module and mesh the whole partitioned assembly (optional: make sub-meshes of the parts to better control mesh sizing)

6. Make 'Volume Groups' of each part within the mesh

7. Make 'Surface Groups' for all the boundary conditions

8. Export mesh as MED file (need to check 'Automatically Create Groups' box)

-Using Gmsh

9. Open mesh file in Gmsh and 'Save As' Gmsh mesh file (.msh file extension, make sure file is saved as Version 2.0 ACSII and un-check all check boxes)

10. Move Gmsh mesh file into an empty case file (make sure to have constant->polyMesh (empty) and system folder with necessary files)

-OpenFOAM mesh conversion/manipulation tools

11. Run [>> gmshToFoam] in OpenFOAM terminal within the case directory

12. Run [>> splitMeshRegions -cellZonesOnly -overwrite] (this is only used if there are multiple parts (volumes) to be split up)

To export dolfin meshing, using dolfin-convert your_gmsh.msh your_dolfin.xml

Extra notes:

In Salome-mesh export, Med version should select the lowest, however, gmsh may still complain the med lib version is too high. I use the gmsh 4 to read the med mesh sucessfully.

2. Make sure the gmsh “expert mode” is enabled in preference, so “gmsh v2 ASCII” option can be selected in gmsh export dialog. It seems quad mesh cells are not supported in dolfin-convert

Posted in Programming | Leave a comment

Coupling OpenFOAM with FeniCS for multiphysis simulation (OpenFOAM workshop 13 presentation)

Posted on June 25, 2018 by news

This is my presentation for OpenFOAM workshop 13 presentation, unfortunately, I can not attend the workshop in person.

Link to my presentation PDF:
OFW13-19-119__Coupling_OpenFOAM_with_FeniCS_for_multiphysis_simulation__QingfengXIA

Outline of this presentation:
1. physics simulation automation
2. CFD module for FreeCAD (GUI for OpenFOAM)
3. multiphysics coupling by OpenFOAM and FenicsSolver
4. Future plan

Posted in Uncategorized | Comments Off

preCICE build and test on Ubuntu 16.04

Posted on June 22, 2018 by news

please read PDF version for better code highlighting

preCICE build and test on Ubuntu 16.04

Posted in Uncategorized | Comments Off

FreeCAD 0.17 + OpenFOAM CFD module now works on windows 10 WSL

Posted on May 6, 2018 by news

after windows 10 2018 april update, it is possible to run my CFD module on FreeCAD 0.17

Limited by my time I can spend on this open source project, I will will make new feature in about half year, but maintain CFD module working with Stable FreeCAD release.

Posted in Uncategorized | Comments Off

I have developed several multi-physics solver based on Fenics and GUI within FreeCAD CFD workbench

Posted on November 27, 2017 by news

This is also announced in Fenics support forum

Screenshot of FenicsSolver in FreeCAD
Dear all,

I have developed several multi-physics solver based on Fenics, it is hosted on github:
https://github.com/qingfengxia/FenicsSolver
This solver is data driven (json), instead of programming for every case. It is still in early stage, I would pick up advice from community, especially standardization of json input data format, naming.

It is primarily used as a submodule for my CFD workbench for FreeCAD, giving GUI for boundary selection. I think a GUI for boundary selection is crucial for complicate geometry. I also use it for other projects with maximum code re-usage.
https://github.com/qingfengxia/Cfd

My future plan will be integrate Fenics with OpenFOAM which is quite advanced in multiphase flow and turbulent flow. Recently, my time is very limited without a permanent position in the academia, thus my update and response will be slow.

I should attend Fenics 2018 at Oxford to demonstrate some advanced usage.

Thanks

Qingfeng Xia @ Engineering Science Dept of Oxford University

Posted in Programming, Research | Comments Off

Add visual studio command prompt to windows explorer context menu

Posted on November 5, 2017 by news

Recently relised visual studio build tool (with cmake support), enables building windows c++ project without GUI. It is essential to accelerate the testing of cross-platform program.

It is convenient and decent to add Bash (WSL) and visual studio command line tool

1. git clone
2. cd source & mkdir build & cd build & cmake .. # cmake to generate msbuild/makefile
3. nmake # to compile

Continue reading →

Posted in Uncategorized | Comments Off

using cloud space as a private git repo

Posted on September 24, 2017 by news

[answer from stackoverflow]

It is absolutely possible – what you probably want to do is create a “bare” git repository in folder Y (git init --bare)

and then add that file location as a remote:

git remote add Y file:///path/to/Y
#example path: file:///media/sf_OneDrive/gitr<wbr />epo/FenicsSolver/

1 2	git remote add Y file:///path/to/Y #example path: file:///media/sf_OneDrive/gitr<wbr />epo/FenicsSolver/

======= push and pull has different repo ==========

git remote set-url origin --push --add &lt;a remote&gt;
git remote set-url origin --push --add &lt;another remote&gt;</code><code>

1 2	git remote set-url origin --push --add <a remote> git remote set-url origin --push --add <another remote></code><code>

========push to multiple repo======

The latest version of git (as of October 2012) allows you to do this from the command line:

git remote set-url origin --push --add user1@repo1
git remote set-url origin --push --add user2@repo2
git remote -v

git remote set-url origin --push --add user1@repo1

git remote set-url origin --push --add user2@repo2

git remote -v

It is also possible to push encrypted code to public repo, not tested.

Posted in Uncategorized | Comments Off

BoundaryLayer setup for Gmsh 3D meshing

Posted on June 16, 2017 by news

Precise control of the thickness of the first cell normal to wall, is crucial for turbulence models like k-epsilon, k-omega. Before moving to turbulence flow simulation for CFD workbench of FreeCAD

2D meshing with boundary layer, tutorial

<http://www.iesensor.com/blog/2017/05/24/gmsh_fenics_meshing/>

3D boundary meshing for simple cube is also done, based on FreeCAD FEM workbench example of 3D cantilever.

CAD and Gmsh setup files are embeded.

photo and progress will be updated on the FreeCAD Fem post

https://forum.freecadweb.org/viewtopic.php?f=18&t=22968

Continue reading →

Posted in Uncategorized | Comments Off

from gmsh geo file to fenics mesh with subdomain and boundary facet exported

Posted on May 24, 2017 by news

from dolfin import *
import os.path
from subprocess import Popen, PIPE, check_output

# https://fenicsproject.org/qa/9983/setting-boundary-conditions-for-meshes-generated-by-gmsh
# generete mesh by: gmsh test.geo -3
# dolfin-convert -i gmsh test.msh test.xml

geo_file_content = """
size = 1; //m
meshThickness = size / 3; 
gridsize = size / 5;

// All numbering counterclockwise from bottom-left corner
Point(1) = {0, 0, 0, gridsize};
Point(2) = {0, size, 0, gridsize};
Point(3) = {size, size, 0, gridsize};
Point(4) = {size, 0, 0, gridsize};
Line(1) = {1, 2};               
Line(2) = {2, 3};           
Line(3) = {3, 4};       
Line(4) = {4, 1};       
Line Loop(4) = {1, 2, 3, 4};
Plane Surface(8) = {4};
Physical Line(12) = {1};
Physical Line(23) = {2};
Physical Line(34) = {3};
Physical Line(41) = {4};

Physical Surface(1) = {8};  // do not comment out this line

//Transfinite Surface{8};  // fenics does not support recomine
//Recombine Surface{8};


// boundary layer
Field[4] = BoundaryLayer;
Field[4].EdgesList = {1,3};  // FacesList  for 3D
Field[4].hfar = 10;
Field[4].hwall_n = 0.015;
Field[4].hwall_t = 0.01;
Field[4].thickness = 0.02;
Field[4].ratio = 1.0;

BoundaryLayer Field = 4;

Mesh.CharacteristicLengthFromCurvature = 1;
Mesh.Algorithm = 8; // delannay=1
"""

fname="test"
with open(fname+".geo", 'w') as f:
    f.write(geo_file_content)

cmd = 'gmsh {}.geo -3'.format(fname)
print check_output([cmd], shell=True)  # run in shell mode in case you are not run in terminal

cmd = 'dolfin-convert -i gmsh {}.msh {}.xml'.format(fname, fname)
Popen([cmd], stdout=PIPE, shell=True).communicate()

mesh = Mesh(fname+".xml")
if os.path.exists( fname+"_physical_region.xml"):
    subdomains = MeshFunction("size_t", mesh, fname+"_physical_region.xml")
    plot(subdomains)
if os.path.exists( fname+"_facet_region.xml"):
    boundaries = MeshFunction("size_t", mesh, fname+"_facet_region.xml")
    plot(boundaries)

plot(mesh)
interactive()

from dolfin import *

import os.path

from subprocess import Popen, PIPE, check_output

# https://fenicsproject.org/qa/9983/setting-boundary-conditions-for-meshes-generated-by-gmsh

# generete mesh by: gmsh test.geo -3

# dolfin-convert -i gmsh test.msh test.xml

geo_file_content = """

size = 1; //m

meshThickness = size / 3;

gridsize = size / 5;

// All numbering counterclockwise from bottom-left corner

Point(1) = {0, 0, 0, gridsize};

Point(2) = {0, size, 0, gridsize};

Point(3) = {size, size, 0, gridsize};

Point(4) = {size, 0, 0, gridsize};

Line(1) = {1, 2};

Line(2) = {2, 3};

Line(3) = {3, 4};

Line(4) = {4, 1};

Line Loop(4) = {1, 2, 3, 4};

Plane Surface(8) = {4};

Physical Line(12) = {1};

Physical Line(23) = {2};

Physical Line(34) = {3};

Physical Line(41) = {4};

Physical Surface(1) = {8}; // do not comment out this line

//Transfinite Surface{8}; // fenics does not support recomine

//Recombine Surface{8};

// boundary layer

Field[4] = BoundaryLayer;

Field[4].EdgesList = {1,3}; // FacesList for 3D

Field[4].hfar = 10;

Field[4].hwall_n = 0.015;

Field[4].hwall_t = 0.01;

Field[4].thickness = 0.02;

Field[4].ratio = 1.0;

BoundaryLayer Field = 4;

Mesh.CharacteristicLengthFromCurvature = 1;

Mesh.Algorithm = 8; // delannay=1

"""

fname="test"

with open(fname+".geo", 'w') as f:

f.write(geo_file_content)

cmd = 'gmsh {}.geo -3'.format(fname)

print check_output([cmd], shell=True) # run in shell mode in case you are not run in terminal

cmd = 'dolfin-convert -i gmsh {}.msh {}.xml'.format(fname, fname)

Popen([cmd], stdout=PIPE, shell=True).communicate()

mesh = Mesh(fname+".xml")

if os.path.exists( fname+"_physical_region.xml"):

subdomains = MeshFunction("size_t", mesh, fname+"_physical_region.xml")

plot(subdomains)

if os.path.exists( fname+"_facet_region.xml"):

boundaries = MeshFunction("size_t", mesh, fname+"_facet_region.xml")

plot(boundaries)

plot(mesh)

interactive()

Posted in Programming, Research | Comments Off

FreeCAD compatible Fenics mshr geometry building

Posted on February 27, 2017 by news

Fenics geometry building API is based on CGAL, it can import CAD, but can not export CAD file like step. Here is a solution, not fully tested. This file may be further developed in github.

Continue reading →

Posted in Uncategorized | Comments Off

solved Fenics (2016.2) cpp code compiling error on ubuntu 16.04

Posted on February 27, 2017 by news

for any expression in Expression for Fenics 2016.2( installed from PPA )
expr = """ x[2]> {} ? {}:{}""".format(disc_zmax, lambda_p, lambda_d) print(expr) K = Expression(expr, element=V.ufl_element()) # error in compiling
instant error log:

”
— The imported target “vtkRenderingPythonTkWidgets” references the file
“/usr/lib/x86_64-linux-gnu/libvtkRenderingPythonTkWidgets.so”
but this file does not exist. Possible reasons include:
* The file was deleted, renamed, or moved to another location.
* An install or uninstall procedure did not complete successfully.
* The installation package was faulty and contained
“/usr/lib/cmake/vtk-6.2/VTKTargets.cmake”
but not all the files it references.
”

check if this file is missing or not
ls -l /usr/lib/python2.7/dist-packages/vtk/libvtkRendering*
python-vtk6 (VTK6.2) is installed, and installed file list:

/usr/lib/python2.7/dist-packages/vtk/libvtkRenderingPythonTkWidgets.x86_64-linux-gnu.so

sudo ln -s /usr/lib/python2.7/dist-packages/vtk/libvtkRenderingPythonTkWidgets.x86_64-linux-gnu-6.2.so /usr/lib/x86_64-linux-gnu/libvtkRenderingPythonTkWidgets.so

#this is the only file needs symbolic link to /usr/lib/x86_64-linux-gnu/

then, error is solved

Posted in Programming | Comments Off

Cmake cheatsheet

Posted on December 24, 2016 by news

#Cmake cheat sheet in markdown

Continue reading →

Posted in Uncategorized | Comments Off

c and cpp pitfalls, mixing signed and unsigned integer, implicit conversion

Posted on December 21, 2016 by news

1. mixing unsigned and sign int is a bad idea

first of all, remember: signed int types will be implicitly converted to unsigned if sizeof(signed int types) == sizeof(unsigned int types)
Google C++ coding standard, say “do not use unsigned integer”
C++ STL using size_t which is unsigned int to represent size of containters like std::vector, std::string
do not assume sizeof(int types), always #include <stdint.h> and use int64_t
compiler can warn warning: comparison between signed and unsigned integer expressions [-Wsign-compare] . However, mixed unsigned and signed int algorithm is not warned
JAVA before Java 8 SE does not support unsigned int, even in Java 8 SE you can not declare unsigned int types, but only using static methods, unsigned comparison.
There are good reason not to supported it.
python built-in integer/long is unsigned, but ctypes module defined c++ unsigned types; python 3 integer will not overflow
see also

http://blog.robertelder.org/signed-or-unsigned-part-2/

example and explanation from stackoverflow

2. Relevant quote from the Standard: Expressions

10 Many binary operators that expect operands of arithmetic or enumeration type cause conversions and yield result types in a similar way. The purpose is to yield a common type, which is also the type of the result. This pattern is called the usual arithmetic conversions, which are de?ned as follows:

2 clauses about equal types or types of equal sign omitted

Otherwise, if the operand that has unsigned integer type has rank greater than or equal to the rank of the type of the other operand, the operand with signed integer type shall be converted to the type of the operand with unsigned integer type.
Otherwise, if the type of the operand with signed integer type can represent all of the values of the type of the operand with unsigned integer type, the operand with unsigned integer type shall be converted to the type of the operand with signed integer type.
Otherwise, both operands shall be converted to the unsigned integer type corresponding to the type of the operand with signed integer type.

3. pitfall of overflow and underflow

overflow detection can not been done by compiler but user
float overflow could be captured and hook to abort routine
avoiding infinite looping for underflow by compiler warning
comparison of unsigned expression >= 0 is always true [-Wtype-limits]
assigned minus int to unsigned int should be prohibited, try hex instead or UINT_MAX – 1, since compiler will not warn you for implicit conversion into unsigned int types

4. pitfall of implicit conversion

see more c and c++ pitfalls

implicit conversion from bool to int: -0.5 < x < 0.5
implicit conversion int bool to pointer: must decleare ctor(ptr) as explicit in c++11, and using null_ptr
c++11 smart ptr like shared_ptr<> will not convert to raw ptr!
http://en.cppreference.com/w/cpp/language/implicit_conversion

Guru of the Week (GotW)

Guru of the Week (GotW) is a series of C++ programming problems created and written by Herb Sutter.

#include <iostream>
#include <limits>
#include <vector>


void std_container_test(){
    std::vector<int> v;
    int i = 10;
    if (i < v.size()-1) // there is a compiler warning here
        std::cout <<"underflow occurs for empty container size()-1\n";
}

void underflow_test(int input_s, unsigned int input_u){
	unsigned int size_u = 10u;
	unsigned int i_u;
	int size_s = 10;
	int i_s;
    // infinite looping for underflow
	for(i_u = size_u -1; i_u >=0; i_u--){
		std::cout <<"unsigned: " << ((input_u + i_u) / 2u) << "\n";
	}
}

void float_overflow_test() {

    float d = std::numeric_limits<float>::max();
    std::cout << "std::numeric_limits<float>::max() = " << d << std::endl;
    float q = d + 100;  //behavior is undefined,  could be inf, or remained as std::numeric_limits<float>::max() 
    std::cout << "q: " << q << std::endl;

}

void bool_implicit_coversion_pitfall_test() {
    int x = 0;
    if (!(-0.5 <= x <= 0.5))
        std::cout << "x=0, -0.5 <= x <= 0.5 is false"<<"\n";
}

void signed_int_test() {

    signed int s1 = -4;
    unsigned int u1 = 2;

    signed long long s2 = -4;
    unsigned int u2 = 2;

    signed long long int s3 = -4;
    unsigned long int u3 = 2;

    unsigned long int u4 = -2; // no worning at all

    if (s1>u1)
        std::cout << s1 << " > " << u1 << " is true? \n";
    std::cout << (s1 + u1) << "\n"; // 4294967294
    std::cout << (s2 + u2) << "\n"; // -2 
    std::cout << (s3 + u3) << "\n"; // 18446744073709551614 
    
    std::cout << " What are you expecting from `unsigned long int u4 = -2;`, actually stored integer is " << u4<< "\n"; 
}

int main()
{
    signed_int_test();
    bool_implicit_coversion_pitfall_test() ;
    std_container_test();
    float_overflow_test();
}

#include <iostream>

#include <limits>

#include <vector>

void std_container_test(){

std::vector<int> v;

int i = 10;

if (i < v.size()-1) // there is a compiler warning here

std::cout <<"underflow occurs for empty container size()-1\n";

}

void underflow_test(int input_s, unsigned int input_u){

unsigned int size_u = 10u;

unsigned int i_u;

int size_s = 10;

int i_s;

// infinite looping for underflow

for(i_u = size_u -1; i_u >=0; i_u--){

std::cout <<"unsigned: " << ((input_u + i_u) / 2u) << "\n";

}

void float_overflow_test() {

float d = std::numeric_limits<float>::max();

std::cout << "std::numeric_limits<float>::max() = " << d << std::endl;

float q = d + 100; //behavior is undefined, could be inf, or remained as std::numeric_limits<float>::max()

std::cout << "q: " << q << std::endl;

}

void bool_implicit_coversion_pitfall_test() {

int x = 0;

if (!(-0.5 <= x <= 0.5))

std::cout << "x=0, -0.5 <= x <= 0.5 is false"<<"\n";

}

void signed_int_test() {

signed int s1 = -4;

unsigned int u1 = 2;

signed long long s2 = -4;

unsigned int u2 = 2;

signed long long int s3 = -4;

unsigned long int u3 = 2;

unsigned long int u4 = -2; // no worning at all

if (s1>u1)

std::cout << s1 << " > " << u1 << " is true? \n";

std::cout << (s1 + u1) << "\n"; // 4294967294

std::cout << (s2 + u2) << "\n"; // -2

std::cout << (s3 + u3) << "\n"; // 18446744073709551614

std::cout << " What are you expecting from `unsigned long int u4 = -2;`, actually stored integer is " << u4<< "\n";

}

int main()

{

signed_int_test();

bool_implicit_coversion_pitfall_test() ;

std_container_test();

float_overflow_test();

}

Posted in Programming | Comments Off

compile c++ codes using Visual C++ Build Tools from the command prompt

Posted on December 6, 2016 by news

Introducing the Visual Studio Build Tools

“If you have installed Microsoft Visual C++ Build Tools 2015 on Windows 10, open the Start menu and choose All apps. Scroll down and open the Visual C++ Build Tools folder. Choose Visual C++ 2015 x86 Native Tools Command Prompt to open the command prompt window.”

1) for single file source file

cl -> gcc , but the command options is different,

cl /EHsc hello.cpp

https://msdn.microsoft.com/en-us/library/f35ctcxw.aspx

2) for a small project, CMake is recommended,

it is handy to write one from example CMakeLists.txt online.

In your source code folder: md build && cd build && cmake ..

then using nmake instead of make

3) for exsitent vs solution using msbuild:

compile visual studio solution/project designed by VS IDE, no corresponding command on Linux.

search “MSBuild Command-Line Reference”

In the near future, other light weight IDE might support detecting “Visual Studio Build Tools”, e.g QtCreator

Posted in Programming | Comments Off

Mibox 2 install youtube and play to kodi (just as chromecast)

Posted on November 8, 2016 by news

1. install shafa TV manager and/or ES file explorer to be enable install apk from USB/SD card
http://www.shafa.com/articles/yWNpNK6erhB6Q4dO.html

2. install Kodi/XMBC apk from shafa’s file manager (in toolbox) on USB storage

3. install apk of youtube for Google TV, and also enable USB keyboard in MiBox system setting -> Input.

3. enable DNLA, web http control, UPnP in kodi settings
http://kodi.wiki/view/web_interface
write down the IP (check on your router), port, user name and password on this setting page

For web http control, install the new “Chorus” theme addon

4. install “play to Kodi” Chrome extension, Firefox should be possible
then, right click to video link, kodi->play/queue
first of all, setup kodi extension to login to kodi web interface

5. disable screen saver, for Kodi

Posted in Uncategorized | Comments Off

Evaluation of OpenFOAM on bash on Ubuntu on windows 10

Posted on September 4, 2016 by news

## installation

After Anniversary Build (2016 Aug) for x65 windows 10, it is possible to evaluate this beta feature in developer mode without upgrade to claimed unstable insider version.

see detailed tutorial:

To install ubuntu on windows 10 from power shell prompt:

Enable-WindowsOptionalFeature -Online -FeatureName Microsoft-Windows-Subsystem-Linux

## overview
In addition to apt-get, some other tool are there like python2, and python3, gcc4.9, perl5. dbus, systemd
Xwindows has been explored by geeks, but it is not stable/usable to run GUI app.

After fireup bash, the current path is /mnt/c/Windwos/System32 . It is better to “cd /mnt/c//Documents/” at the end of .bashrc, not mess up system folder and for better locate file generated by bash commands.

## look around

– dpkg -l to see installed package, while not sure snap pkg will be supported.
– sudo apt-get update To update the repo, it will show trusty repo info
– ls /dev/
– ls /usr/include/, almost empty
– lsb_release -a shows it is 14.04.4
– uname -a
> Linux DESKTOP-DELL15 3.4.0+ #1 PREEMPT Thu Aug 1 17:06:05 CST 2013 x86_64 x86_64 x86_64 GNU/Linux

## test POSIX API

It may be used to easily run software previously POSIX only program on windows.
apt-get install gcc will install gcc 4.8 and *linux-libc-dev*

ls /usr/include/ will give unistd.h

## install OpenFOAM

see instruction:

two versions are available for ubuntu 14.04
openfoam30
openfoam4

OpenMPI may not work; ParaView is definetly not working.

apt-get install openfoam30, then apt-get remove openfoam30 (leave dependent pkg remained.)
install openfoam4 by force, will not install all paraview related tons of new packages.
apt-get download openfoam4 dpkg --force-all -i openfoam4
force remove this pkg can fix error message each time

## Run OpenFOAM solver

mkdir -p $FOAM_RUN cp -r $FOAM_TUTORIALS $FOAM_RUN cd $FOAM_RUN/tutorials

bash /OpenFOAM/qingfeng-4.0/incompressible/simpleFoam/windAroundBuildings$ ./Allrun Running surfaceFeatureExtract on /home/qingfeng/OpenFOAM/qingfeng-4.0/incompressible/simpleFoam/windAroundBuildings Running blockMesh on /home/qingfeng/OpenFOAM/qingfeng-4.0/incompressible/simpleFoam/windAroundBuildings Running snappyHexMesh on /home/qingfeng/OpenFOAM/qingfeng-4.0/incompressible/simpleFoam/windAroundBuildings Running simpleFoam on /home/qingfeng/OpenFOAM/qingfeng-4.0/incompressible/simpleFoam/windAroundBuildings

## Feature wanted

– right click to open bash, just as cmd prompt, or git-bash

## upgrade to ubuntu 16.04
upgrade to ubuntu 16.04 is planned by official support

apt-get update apt-get dist-upgrade do-release-upgrade -f DistUpgradeViewNonInteractive -d

lxrun.exe is used to manage WSL. This command can be used to install or uninstall the Ubuntu image.

Posted in Uncategorized | Comments Off

disable windows input method hotkey

Posted on August 21, 2016 by news

UPDATE: it seems the following methods still fails, as restart pc or windows update will invalidated the setting

===========

Avoid the annoying hotkey to trigger the IME shape/punctation in Chinese for half/full
see tutorial: http://superuser.com/questions/327479/ctrl-space-always-toggles-chinese-ime-windows-7
In addition to disapble Ctrl-Space, IME shape/punctation hotkeys are also disabled, which are shift-space and ctrl-period respectively. Once your edit the hotkey in GUI, which you will fail, you may need to rerun this reg file again. Logout to make it take effect

copy the following into notepad and save as “diasableIMEHotkey.reg” to anywhere in your PC. right click “merge” will import them into registry.

[HKEY_CURRENT_USER\Control Panel\Input Method\Hot Keys\00000010] for simplified Chinese, for traditional Chiese, use : [HKEY_CURRENT_USER\Control Panel\Input Method\Hot Keys\00000070]

Once your edit the hotkey in GUI, which you will fail, you may need to rerun this reg file again. Logout to make it take effect

Continue reading →

Posted in Uncategorized | Comments Off

Toshiba L650 laptop disassembling

Posted on May 29, 2016 by news

I got a 5 year old laptop, it does not boot, no screen show, no beeping. So I need to disassemble and check.

Generally, this laptop is easy to repalce harddisk or memory on the back. And the dissabling is straight forwards.

Continue reading →

Posted in Uncategorized | Comments Off

Introduction to Industrial IO (sensor framework for Linux)

Posted on May 29, 2016 by news

introduciton to Industrial IO

overview of industrial IO. In short, SPI and I2C sensor framework for Linux
sensor categories: “ADC, DAC, light, accel(erometer), gyro, magnetometer, humidity, temperature, amplifier, frequency, orientation, health”, etc
merged to mainline kernel 3.14, used by beagleBone, sysfs and char device based, see the IIO architecture

DAQmx base for linux: commercial lib from national instrument
comedi: PCI/PCIe, USB devices for data acquisition
hwmon: very low speed,traditionally for CPU temperature monitoring
HID: like mouse, keyboard, generally low speed input device
bluetooth smart:

user space IIO API

user space library libiio (https://wiki.analog.com/resources/tools-software/linux-software/libiio)

#include <iio.h> four basic classes: iio_contex, iio_device, iio_channel and iio_buffer
iio_trigger is kind of iio_device without channel but with name inc “trigger”
iio_context has internal operation functions structure, kernel driver
windows user can use libiio to access network iiod
python and cshaprt wrapping for user space API: python-iio libiio-cil-dev
XML context: attributes calibration data, scale, etc.

detailed

(http://events.linuxfoundation.org/sites/events/files/slides/lceu15_baluta.pdf) + scan_element(), iio_channel_is_scan_element(): elements in buffer is called scan /sys/bus/iio/devices/iio:devices/scan elements

read_raw() from register (possible 12bit) to integer in memory. NO integer to physical quantity (with unit) like comedilib?

trigger readings based on an external interrupt source hardware interrupt (IRQ pins) software interrupts (periodic timers, sysfs triggers)

design of IIO

modes: DIRECT,BUFFER SOFTWARE,BUFFER HARDWARE,BUFFER TRIGGERED
iio_triggered_buffer iio_events only found in kernel mode iio_core, see(linux/iio/triggered_buffer.h);trigger are software trigger, not hardware TTL signal in DAQ terminology?

documentation for IIO

user space library libiio
kernel document text does not provide enough information for user
analogdevice wiki on IIO
blog on IIO
mail list
high speed IIO

source code

supported hardware list

About 200 device drivers by the time of kernel 4.4; many drivers supports a varierty of models.

see/search kernel source tree linux/iio, there are a few subfolders corresponds to sensor categories: “ADC, DAC, light, accel(erometer), gyro, magnetometer, humidity, temperature, amplifier, frequency, orientation, health”, etc

test of IIO dummy device on ubuntu 16.04

“dummy” a virtual sensor for testing

sudo apt-get install libiio libiio-dev libiio-python libiio-utils
<span class="hljs-comment">#all trigger related function need kernel module to support</span>
sudo modprobe iio_dummy
iio_info
ls /sys/bus/iio/devices/iio\<span class="hljs-symbol">:device0</span>
cat /sys/bus/iio/devices/iio\<span class="hljs-symbol">:device0/name</span>
<span class="hljs-comment">#no need to enable?</span>
cat /sys/bus/iio/devices/iio\<span class="hljs-symbol">:device0/in_accel_x_raw</span>

sudo apt-get install libiio libiio-dev libiio-python libiio-utils

#all trigger related function need kernel module to support

sudo modprobe iio_dummy

iio_info

ls /sys/bus/iio/devices/iio\:device0

cat /sys/bus/iio/devices/iio\:device0/name

#no need to enable?

cat /sys/bus/iio/devices/iio\:device0/in_accel_x_raw

python-iio example

desired feature of IIO

user space tool to expose sensors to webUI?
devicetree: platform device support is supported such as beaglebone.
standardisation calibration data in XML attribute file to simulate: IEEE 1451.4 Plug and Play Sensors (TEDS)
possiblity of user space driver development based on user space SPI, i2c and libusb?
physical value conversion “The functions comedi_to_physical, comedi_to_phys, comedi_from_physical and comedi_from_phys are used to convert between Comedi’s integer data and floating point numbers corresponding to physical values (voltages, etc.). ”

Posted in Linux, Sensor | Comments Off

A recommendation for free of charge 2D CAD and office system, also cross-platform

Posted on May 17, 2016 by news

for office system, I recommend kingsoft WPS office, which is based on Qt, much faster than java based solution. It has Android, Linux version in alpha, in addition to mature windows ‘s version

http://wps-community.org/download.html

For 2D CAD drawing, I recommend Dassault system’ Draftsight, it is a professional level product. just need a email activation. Linux version is still in beta.

http://www.3ds.com/products-services/draftsight-cad-software/free-download/

Both are not open source, but free of charge for production usage.

Posted in Linux | Comments Off

Salome installation

mesh conversion from salome to Fenics

1. mixing unsigned and sign int is a bad idea

2. Relevant quote from the Standard: Expressions

3. pitfall of overflow and underflow

4. pitfall of implicit conversion

introduciton to Industrial IO

related projects/libs

user space IIO API

detailed

design of IIO

documentation for IIO

source code

supported hardware list

test of IIO dummy device on ubuntu 16.04

desired feature of IIO

Recent Posts

Categories

Recent Comments

Archives

Meta