A python library computing carson's equations.
Project description
carsons
This is an implementation of Carson's Equations, a mathematical model for deriving the equivalent impedance of an AC transmission or distribution line.
Implementation
carsons
is developed using python 3.6 support for unicode characters
like π, ƒ, ρ, μ, ω etc. This feature allows us to avoid translating the
problem into a more typical programming syntax, so the code is dense and
can easily be compared to published formulations of the problem.
For example, we implement the kron reduction, a matrix decomposition step, using unicode notation to indicate the slightly different meaning of impedance values before and after a kron reduction:
def perform_kron_reduction(z_primitive):
Ẑpp, Ẑpn = z_primitive[0:3, 0:3], z_primitive[0:3, 3:]
Ẑnp, Ẑnn = z_primitive[3:, 0:3], z_primitive[3:, 3:]
Z_abc = Ẑpp - Ẑpn @ inv(Ẑnn) @ Ẑnp
return Z_abc
Take a look at the source code to see more cool unicode tricks!
Installation
~/$ pip install carsons
Usage
Carsons model requires a line model object that maps each phase to properties of the conductor for that phase.
from carsons import CarsonsEquations, calculate_impedance
class Line:
geometric_mean_radius: {
'A': geometric_mean_radius_A in meters
...
}
resistance: {
'A': per-length resistance of conductor A in ohms/meters
...
}
wire_positions: {
'A': (x, y) cross-sectional position of the conductor in meters
...
}
phases: {'A', ... }
# map of phases 'A', 'B', 'C' and 'N<>' which are described in the
# gmr, r and phase_positions attributes
line_impedance = calculate_impedance(CarsonsEquations(Line()))
The model supports any combination of ABC phasings (for example BC, BCN etc...) including systems with multiple neutral cables; any phases that are not present in the model will have zeros in the columns and rows corresponding to that phase.
Multiple neutrals are supported, as long as they have unique labels
starting with N
(e.g. Neutral1
, Neutral2
).
Intermediate results such as primitive impedance matrix are also available.
z_primitive = CarsonsEquations(Line()).build_z_primitive()
For examples of how to use the model, see the overhead wire tests.
carsons
is tested against several cable configurations from the IEEE
test feeders, as well as
examples from EPRI's OpenDSS documentation.
Concentric Neutral Cable
carsons
also supports modelling of concentric neutral cables of any
phasings. Its usage is very similar to the example above, only requiring
a few more parameters about the neutral conductors in the line model
object.
from carsons import (ConcentricNeutralCarsonsEquations,
calculate_impedance)
class Cable:
resistance: {
'A': per-length resistance of conductor A in ohm/meters
...
}
geometric_mean_radius: {
'A': geometric mean radius of conductor A in meters
...
}
wire_positions: {
'A': (x, y) cross-sectional position of conductor A in meters
...
}
phases: {'A', 'NA', ... }
neutral_strand_gmr: {
'NA': neutral strand gmr of phase A in meters
...
}
neutral_strand_resistance: {
'NA': neutral strand resistance of phase A in ohm/meters
...
}
neutral_strand_diameter: {
'NA': neutral strand diameter of phase A in meters
...
}
diameter_over_neutral: {
'NA': diameter over neutral of phase A in meters
...
}
neutral_strand_count: {
'NA': neutral strand count of phase A
...
}
cable_impedance = calculate_impedance(ConcentricNeutralCarsonsEquations(Cable()))
For examples of how to use the model, see the concentric cable tests.
Multi-Conductor Cable
carsons
also supports modelling of phased duplex, triplex, quadruplex cables and triplex secondary.
It only requires a few more parameters to describe cable's geometry.
from carsons import (MultiConductorCarsonsEquations,
calculate_impedance)
class Cable:
resistance: {
'A': per-length resistance of conductor A in ohm/meters
...
}
geometric_mean_radius: {
'A': geometric mean radius of conductor A in meters
...
}
wire_positions: {
'A': (x, y) cross-sectional position of conductor A in meters
...
}
outside_radius: {
'A': outside radius of conductor A, including insulation and jacket thickness
...
}
insulation_thickness: {
'A': insulation thickness of conductor A
...
}
phases: {'A', ... }
cable_impedance = calculate_impedance(MultiConductorCarsonsEquations(Cable()))
To model a triplex secondary cable, the inputs should be keyed on secondary conductors S1
and S2
. The impedance result
is a 2 x 2 matrix.
class Cable:
resistance: {
'S1': per-length resistance of conductor S1 in ohm/meters
...
}
geometric_mean_radius: {
'S1': geometric mean radius of conductor S1 in meters
...
}
wire_positions: {
'S1': (x, y) cross-sectional position of conductor S1 in meters
...
}
outside_radius: {
'S1': outside radius of conductor S1, including insulation and jacket thickness
...
}
insulation_thickness: {
'S1': insulation thickness of conductor S1
...
}
phases: {'S1', ... }
For examples of how to use the model, see the multi-conductor cable tests.
Problem Description
Carsons equations model an AC transmission or distribution line into an equivalent set of phase-phase impedances, which can be used to model the line in a power flow analysis.
For example, say we have a 4-wire system on a utility pole, with A
,
B
, C
phase conductors as well as a neutral cable N. We know that
when conductors carry electrical current, they exhibit a magnetic field
--- so its pretty easy to imagine that, e.g., the magnetic field
produced by A
would interact with the B
, C
, and N
conductors.
B
O
|
|
A N | C
O O| O
----------|-----------
|
|
|
|
|
|
|
|
|
|
|
|
|
==============[Ground]============================
/ / / / / / / / /
/ / / / / / /
/ / / / /
A* N* C*
0 0 0
B*
0
Figure: Cross-section of a 4-wire distribution line, with
ground return.
However, each conductor also has a ground return path (or 'image') ---
shown as A*
, B*
, C*
, and N*
in the figure above --- which is a
magnetically induced current path in the ground. When A produces a
magnetic field, that field also interacts with B*
, C*
, N*
, and
A*
. Carsons equations model all these interactions and reduce them to
an equivalent impedance matrix that makes it much easier to model this
system.
In addition carsons
implements the kron reduction, a conversion that
approximates the impedances caused by neutral cables by incorporating
them into the impedances for phase A
, B
, and C
. Since most AC and
DC powerflow formulations don't model the neutral cable, this is a
valuable simplification.
References
The following works were used to produce this formulation:
- Leonard L. Grigsby -- Electrical Power Generation, Transmission and Distribution
- William H. Kersting -- Distribution System Modelling and Analysis 2e
- William H. Kersting, Distribution System Analysis Subcommittee -- Radial Distribution Test Feeders
- Timothy Vismore -- The Vismor Milieu
- Daniel Van Dommelen, Albert Van Ranst, Robert Poncelet -- GIC Influence on Power Systems calculated by Carson's method
- Andrea Ballanti, Roger Dugan -- Cable Modelling in OpenDSS
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for carsons-0.5.1a1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 972447f3ccd68202015b57b900ab11a1ac2b6ca9159eb57a20f0478a6395233e |
|
MD5 | 9c1d191119870425aa40c1f6dbd2f8b9 |
|
BLAKE2b-256 | 6f833a72235b6cc35fed796c75283969f54b3b49f76c1a67a34d2c181774a1df |