Skip to content

Fault analysis API corrupts NumPy complex impedances #139

@marcostfermin

Description

@marcostfermin

Describe the bug

Several fault analysis functions attempt to coerce impedances to imaginary values using checks like if not isinstance(Z1, complex): Z1 *= 1j. NumPy complex scalars (e.g. np.complex128) are not instances of built-in complex, so already-complex impedances are multiplied by 1j again, producing incorrect results.

Reproduction Code

# Setup
import numpy as np
import electricpy as ep

# Code which causes failure (incorrect numeric result)
Vth = 1 + 0j
Z0 = np.complex128(0.1j)
Z1 = np.complex128(0.2j)
Z2 = np.complex128(0.3j)

If = ep.fault.phase_to_phase_fault(Vth, (Z0, Z1, Z2), sequence=True)
print(If)

Expected behavior

If impedances are already complex (including NumPy complex scalars), they should not be modified. Results should be identical to using built-in complex values.

Screenshots
If applicable, add screenshots to help explain your problem.

Desktop (please complete the following information):

  • OS: [e.g. Linux]
  • Python Version [e.g. 3.7]
  • Version [e.g. 0.2.1]

Additional context

Type checks should use np.iscomplexobj() or isinstance(x, (complex, np.complexfloating)).

Metadata

Metadata

Labels

bugSomething isn't working

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions