File Formats
PyPEEC Workflow
Description of the PyPEEC workflow (mesher, viewer, solver, and plotter). The input and output files of the different tools are shown. Additionally, a global configuration file is used for the logger.
Geometry File Format
The file_geometry file format is used by the mesher.
This file contains the definition of the voxel structure.
# mesher type
# - "voxel" for index arrays
# - "shape" for 2D vector shapes
# - "png" for PNG files
# - "stl" for STL files
"mesh_type": "'voxel' or 'shape' or 'png' or 'stl'"
# definition of the voxel structure
# - specific for the different mesher types
# - format definition is located in the next subsections
"data_voxelize": {}
# resampling of the voxel structure
# - use_reduce: reduce (or not) the voxel structure by removing unused voxels
# - use_resample: resample (or not) the voxel structure by subdividing voxels
# - resampling_factor: array with resampling factors (x, y, and z directions)
"resampling":
"use_reduce": true
"use_resample": true
"resampling_factor": [2, 2, 1]
# 2D array containing the points used for magnetic field evaluation
# - the number of points (n_pts) can be zero.
# - the array has the following size: (n_pts, 3)
# - the points should be located outside the active voxels
"pts_cloud":
- [-1.0, +1.0, +1.0]
- [+1.0, -1.0, +1.0]
- [+1.0, +1.0, -1.0]
# check (or not) the validity of the point cloud
"check_cloud": true
# perform (or not) the conflict resolution between the voxels
"check_conflict": true
# compute (or not) the connections between the domains
"check_connection": true
# definition of the conflict resolution between domains
# - during the voxelization, the same voxel can be assigned to several domains
# - the shared voxels are located at the boundaries between domain
# - the shared voxels are conflicts and should be assigned to a single domain
# - list of dicts with the conflict resolution rules
# - optional feature, the list can be empty if no conflict resolution is required
# - conflict definition
# - domain_resolve: list of domain names where the shared voxels should be removed
# - domain_keep: list of domain names where the shared voxels should be kept
"domain_conflict":
- {"domain_resolve": ["dom_cond"], "domain_keep": ["dom_src"]}
- {"domain_resolve": ["dom_cond"], "domain_keep": ["dom_sink"]}
# pledge the existence or absence of connections between domains
# - dict of dicts with the connection name and the connection definition
# - optional feature, the dict can be empty without having an impact on the results
# - connection definition
# - connected: boolean specified if the two domain lists should be connected
# - domain_group: list of list of domains where the connections are checks
# - the outer list domains are used for the connection checks
# - the inner list domains are merged together
"domain_connection":
"connected": {"connected": true, "domain_group": [["dom_cond"], ["dom_src"], ["dom_sink"]]}
"insulated": {"connected": false, "domain_group": [["dom_mag"], ["dom_src", "dom_cond", "dom_sink"]]}
Definition from Index Arrays
# voxel definition type ("voxel" for index arrays)
"mesh_type": "voxel"
# definition of the voxel structure
"data_voxelize":
# definition of the voxel structure
# - n: array with the number of voxels (x, y, and z directions)
# - d: array with the voxel dimensions (x, y, and z directions)
# - c: array with the coordinates of the voxel structure center (x, y, and z directions)
"param":
"n": [4, 4, 3]
"d": [10.0e-3, 10.0e-3, 10.0e-3]
"c": [0.0, 0.0, 0.0]
# definition of the voxel indices of the different domains
# - dict of arrays with the domain name and the voxel indices
# - required information, the dict cannot be empty
"domain_def":
"dom_src": [1, 2]
"dom_cond": [5, 6, 9, 10]
"dom_sink": [13, 14]
"dom_mag": [36, 37, 38, 39, 40, 41, 42, 43]
Definition from 2D Vector Shapes
# voxel definition type ("shape" for 2D vector shapes)
"mesh_type": "shape"
# definition of the voxel structure
"data_voxelize":
# definition of the voxel structure
# - dx: voxel dimension for the x direction
# - dy: voxel dimension for the y direction
# - dz: voxel dimension for the z direction
# - cz: coordinates of the voxel structure center for the z direction
# - tol: tolerance for simplifying the shapes (null for disabling simplification)
# - xy_min: array with the lower corner coordinates of the voxel structure (x and y directions)
# - xy_max: array with the upper corner coordinates of the voxel structure (x and y directions)
# - alternatively, the xy_min/xy_max can be set to null and the shape bounds are kept
"param":
"dx": 50.0e-6
"dy": 50.0e-6
"dz": 50.0e-6
"cz": 0.0e-6
"tol": 1.0e-6
"xy_min": [-20e-3, -20e-3]
"xy_max": [+20e-3, +20e-3]
# definition of the layer stack (voxels in the z-direction)
# - list of dicts with the definition of the layers
# - required information, the list cannot be empty
# - layer definition
# - n_layer: number of voxels in the z-direction for the layer
# - tag_layer: string with the name of the layer
"layer_stack":
- {"n_layer": 1, "tag_layer": "mag"}
- {"n_layer": 5, "tag_layer": "insulation_mag"}
- {"n_layer": 1, "tag_layer": "cond"}
- {"n_layer": 5, "tag_layer": "insulation_terminal"}
- {"n_layer": 1, "tag_layer": "terminal"}
# definition of the shapes composing the different domains
# - dict with the domain name and a list of shape to be assembled
# - required information, the dict cannot be empty
# - domain definition
# - shape_layer: list of layer where the shape is located
# - shape_operation: composition operation for assembling the shape ("add" or "sub")
# - shape_type: type of the shape ("pad" or "trace" or "polygon")
# - shape_data: data containing the shape geometry
# - "pad"
# - coord: array with the 2D coordinates composing the shape
# - diameter: diameter of the pads
# - "trace"
# - coord: array with the 2D coordinates composing the shape
# - width: thickness of the traces
# - "polygon"
# - coord_shell: array with the 2D coordinates composing the shape
# - coord_holes: list of arrays with the 2D coordinates of the holes
# - buffer: thickness of the buffer around the polygons
"geometry_shape":
"dom_cond":
-
"shape_layer": ["cond"]
"shape_operation": "add"
"shape_type": "trace"
"shape_data":
"width": 0.5e-3
"coord": [[0.0e-3, 0.0e-3], [5.0e-3, 5.0e-3]]
"dom_mag":
-
"shape_layer": ["mag"]
"shape_operation": "add"
"shape_type": "polygon"
"shape_data":
"buffer": 0.0
"coord_shell": [[0.0e-3, 0.0e-3], [5.0e-3, 0.0e-3], [5.0e-3, 5.0e-3], [0.0e-3, 5.0e-3]]
"coord_holes":
- [[1.0e-3, 1.0e-3], [2.0e-3, 1.0e-3], [2.0e-3, 2.0e-3], [1.0e-3, 2.0e-3]]
- [[3.0e-3, 3.0e-3], [4.0e-3, 3.0e-3], [4.0e-3, 4.0e-3], [3.0e-3, 4.0e-3]]
-
"shape_layer": ["mag"]
"shape_operation": "sub"
"shape_type": "polygon"
"shape_data":
"buffer": 0.0
"coord_shell": [[1.0e-3, 1.0e-3], [3.0e-3, 1.0e-3], [3.0e-3, 3.0e-3], [1.0e-3, 3.0e-3]]
"coord_holes": []
"dom_src":
-
"shape_layer": ["insulation_terminal", "terminal"]
"shape_operation": "add"
"shape_type": "pad"
"shape_data":
"diameter": 0.7e-3
"coord": [[0.0e-3, 0.0e-3]]
"dom_sink":
-
"shape_layer": ["insulation_terminal", "terminal"]
"shape_operation": "add"
"shape_type": "pad"
"shape_data":
"diameter": 0.7e-3
"coord": [[5.0e-3, 5.0e-3]]
Definition from PNG Files
# voxel definition type ("png" for PNG files)
"mesh_type": "png"
# definition of the voxel structure
"data_voxelize":
# definition of the voxel structure
# - d: array with the voxel dimensions (x, y, and z directions)
# - c: array with the coordinates of the voxel structure center (x, y, and z directions)
# - size: array with the size of the images (number of voxels in the x and y direction)
"param":
"d": [10.0e-3, 10.0e-3, 10.0e-3]
"c": [0.0, 0.0, 0.0]
"size": [49, 49]
# definition of the mapping between the image color and the different domains
# - dict with the domain name and the specified colors
# - required information, the dict cannot be empty
# - the colors are specified with an array of colors (RGBA format)
"domain_color":
"dom_src": [[255, 0, 0, 255]]
"dom_cond": [[0, 0, 0, 255]]
"dom_sink": [[0, 255, 0, 255]]
"dom_mag": [[0, 0, 255, 255]]
# definition of the layer stack (voxels in the z-direction)
# - list of dicts with the definition of the layers
# - required information, the list cannot be empty
# - layer definition
# - n_layer: number of voxels in the z-direction for the layer
# - filename_list: list of strings with the PNG files defining the layer
"layer_stack":
- {"n_layer": 1, "filename_list": ["png/layer_bottom.png"]}
- {"n_layer": 8, "filename_list": ["png/layer_mid.png"]}
- {"n_layer": 1, "filename_list": ["png/layer_top.png"]}
Definition from STL Files
# voxel definition type ("stl" for STL files)
"mesh_type": "stl"
# definition of the voxel structure
"data_voxelize":
# definition of the voxel structure
# - d: array with the voxel dimensions (x, y, and z directions)
# - xyz_min: array with the lower corner coordinates of the voxel structure (x, y, and z directions)
# - xyz_max: array with the upper corner coordinates of the voxel structure (x, y, and z directions)
# - alternatively, the xyz_min/xyz_max can be set to null and the STL mesh bounds are kept
"param":
"d": [10.0e-3, 10.0e-3, 10.0e-3]
"xyz_min": [-20e-3, -20e-3, -20e-3]
"xyz_max": [+20e-3, +20e-3, +20e-3]
# definition of the STL files of the different domains
# - dict with the domain name and the STL files
# - required information, the dict cannot be empty
# - domain definition
# - list of STL meshes composing the domain
# - scale: scaling factor for specifying the unit of the STL meshes (before offsetting)
# - offset: array with offsets for translating the STL meshes (after scaling)
# - filename: string with the STL filename
"domain_stl":
"dom_src": [{"scale": 1.0, "offset": [0.0, 0.0, 0.0], "filename": "stl/dom_src.stl"}]
"dom_cond": [{"scale": 1.0, "offset": [0.0, 0.0, 0.0], "filename": "stl/dom_cond.stl"}]
"dom_sink": [{"scale": 1.0, "offset": [0.0, 0.0, 0.0], "filename": "stl/dom_sink.stl"}]
"dom_mag": [{"scale": 1.0, "offset": [0.0, 0.0, 0.0], "filename": "stl/dom_mag.stl"}]
Problem File Format
The file_problem file format is used by the solver.
This file contains the definition of the magnetic problem to be solved.
# material definition
# - dict of dicts with the material name and the material definition
# - required information, the dict cannot be empty
# - electric and magnetic materials can be defined
# - material definition
# - domain_list: list of domains with the specified material
# - material_type: material type ("electric" or "magnetic" or "electromagnetic")
# - "electric": material with a finite conductivity (with a zero susceptibility)
# - "magnetic": material with a finite susceptibility (with a zero conductivity)
# - "electromagnetic": material with a finite susceptibility and permeability
# - orientation_type: material type ("isotropic" or "anisotropic")
# - "isotropic": the material parameters are anisotropic
# - "anisotropic": the material parameters anisotropic
# - var_type: type of material parameters ("lumped" or "distributed")
# - "lumped": the material parameters are homogeneous
# - "distributed": the material parameters are space-dependent
"material_def":
"mat_electric":
"domain_list": ["dom_src", "dom_cond", "dom_sink"]
"material_type": "electric"
"orientation_type": "isotropic"
"var_type": "lumped"
"mat_magnetic":
"domain_list": ["dom_mag"]
"material_type": "magnetic"
"orientation_type": "isotropic"
"var_type": "lumped"
# source definition
# - dict of dicts with the source name and the source definition
# - required information, the dict cannot be empty
# - sources can only be defined on electric material domains
# - source definition
# - domain_list: list of domains with the specified source
# - source_type: source type ("current" or "voltage")
# - var_type: type of source parameters ("lumped" or "distributed")
# - "lumped": the source parameters are homogeneous
# - "distributed": the source parameters are space-dependent
"source_def":
"src":
"domain_list": ["dom_src"]
"source_type": "current"
"var_type": "lumped"
"sink":
"domain_list": ["dom_sink"]
"source_type": "voltage"
"var_type": "lumped"
# definition of the sweep configurations to be run
# - dict with the sweep names and the sweep dependencies
# - required information, the dict cannot be empty
# - the dict keys represent the name of the sweeps
# - the dict values represent the name of the dependencies
# - the dependency is used as an initial guess for the solver
# - if null, no initial guess is used for the solver
"sweep_config":
"sim_dc": null
"sim_ac": "sim_dc"
# material and source parameters
# - dict of dicts with the material and source parameters
# - required information, the dict cannot be empty
# - each dict represents a specific sweep configuration
# - material and source parameters
# - freq: operating frequency for the problem (DC or AC)
# - material_val: material parameters
# - dict of dicts with the material name and the material definition
# - electric material definition
# - rho_re: material real resistivity (should be positive)
# - rho_im: material imaginary resistivity (should be positive)
# - definition: rho = rho_re + 1j*rho_im
# - magnetic material definition
# - chi_re: material real susceptibility (should be positive)
# - chi_im: material imaginary susceptibility (should be positive)
# - definition: chi = chi_re - 1j*chi_im
# - electromagnetic material definition
# - rho_re: material real resistivity (should be positive)
# - rho_im: material imaginary resistivity (should be positive)
# - chi_re: material real susceptibility (should be positive)
# - chi_im: material imaginary susceptibility (should be positive)
# - definition: rho = rho_re + 1j*rho_im
# - definition: chi = chi_re - 1j*chi_im
# - material values
# - "lumped" and "isotropic": scalar
# - the material parameters are isotropic
# - the same parameters are used for all the voxels
# - "lumped" and "anisotropic": array
# - the array contains the anisotropic parameters
# - the same parameters are used for all the voxels
# - the array has the following size: (3)
# - "distributed" and "isotropic": array
# - the material parameters are isotropic
# - different parameters are used for each voxel
# - the array has the following size: (n_voxel)
# - "distributed" and "anisotropic": 2D array
# - the material parameters are anisotropic
# - different parameters are used for each voxel
# - the array has the following size: (n_voxel, 3)
# - source_val: source parameters
# - dict of dicts with the source name and the source definition
# - current source definition
# - I_re: current source value (real part)
# - I_im: current source value (imaginary part)
# - Y_re: current internal admittance (real part)
# - Y_im: current internal admittance (imaginary part)
# - definition: I = I_re + 1j*I_im
# - definition: Y = Y_re + 1j*Y_im
# - voltage source definition
# - V_re: voltage source value (real part)
# - V_im: voltage source value (imaginary part)
# - Z_re: voltage internal impedance (real part)
# - Z_im: voltage internal impedance (imaginary part)
# - definition: V = V_re + 1j*V_im
# - definition: Z = Z_re + 1j*Z_im
# - source values
# - "lumped": scalar
# - the source is using lumped parameters
# - the same parameters are used for all the voxels
# - "distributed": array
# - the source is using distributed parameters
# - different parameters are used for each voxel
# - the array has the following size: (n_voxel)
"sweep_param":
"sim_dc":
"freq": 0.0
"material_val":
"mat_electric": {"rho_re": 1.0e-8, "rho_im": 0.0"}
"mat_magnetic": {"chi_re": 100.0, "chi_im": 10.0}
"source_val":
"src": {"I_re": 1.0, "I_im": 0.0, "Y_re": 2.0, "Y_im": 0.0}
"sink": {"V_re": 0.0, "V_im": 0.0, "Z_re": 2.0, "Z_im": 0.0}
"sim_ac":
"freq": 1.0e+3
"material_val":
"mat_electric": {"rho_re": 1.0e-8, "rho_im": 0.0"}
"mat_magnetic": {"chi_re": 100.0, "chi_im": 10.0}
"source_val":
"src": {"I_re": 1.0, "I_im": 0.0, "Y_re": 2.0, "Y_im": 0.0}
"sink": {"V_re": 0.0, "V_im": 0.0, "Z_re": 2.0, "Z_im": 0.0}
Other File Formats
The logger configuration file format is documented in
pypeec/data/logger.yaml.The logger configuration can be changed with the
PYTHONLOGGERenvironment variable.The
file_toleranceformat is documented inexamples/config/tolerance.yaml.The
file_viewerformat is documented inexamples/config/viewer.yaml.The
file_plotterformat is documented inexamples/config/plotter.yaml.