🚀 Wire Bundle Optimizer

A Python/PyQt6 app to pack circular wires (with arbitrary diameters) into the smallest possible circular bundle.
It supports multi-start optimization, shielding layers, and live plotting.

---

Index

1. Features
2. Installation & Run
3. UI Walkthrough
4. Predefined Sizes (wire_types.yaml)
5. How it Works (Math)
6. Tips & Troubleshooting
7. Screenshots

---

Features

• Define any number of wire types (custom or from predefined sizes via YAML).
• Color-coded wires.
• Manufacturing margin (percentage) inflates radii to enforce spacing.
• Multi-start SLSQP with parallel runs to escape local minima.
• Inner exclusion constraint from prior shields (no wire can cross a shield).
• Live plot:
  • Wires placement
  • Dashed current outer boundary
  • Grey shield rings

---

Installation & Run

You can run the app in two ways:

Option 1: Using uv (recommended)

uv run main.py

This automatically manages a virtual environment and dependencies.

Option 2: Using pip

First install dependencies:

pip install PyQt6 numpy scipy joblib pyyaml

Then run:

python main.py

---

UI Walkthrough

1. 1. Define Wire Types

   • Set Count and Wire Diameter:

     • Custom: enter diameter in mm.
     • Predefined Sizes: pick a label defined in wire_types.yaml.

   • Pick a Color and click Add Wire. Identical (color + diameter) entries merge counts automatically.

2. 2. Optimization Parameters

   • Number of Solver Initializations: more restarts → better layout, slower.
   • Max Solver Iterations: SLSQP cap per run.
   • Manufacturing Tolerance Margin: extra spacing (percent) added to radii.

3. 3. Defined Wires

   • Shows the current working set.
   • Remove Selected Wire or Clear All (clears wires, layers, and results).

4. 4. Shielding

   • Set Shield thickness (mm).

   • Add Shielding is enabled only after a successful optimize. When clicked:

     • The current optimized bundle becomes a locked layer with a shield of the given thickness.
     • The inner exclusion radius is updated.
     • The working wire list is cleared for defining the next ring.

5. Optimize and Plot

   • Runs the multi-start optimizer and updates:

     • Plot (including historic layers)
     • Outer diameter in mm and inches.

6. Results

   • The plot is centered and scaled to fit.
   • The window is scrollable if content exceeds the viewport.

---

Predefined Sizes (wire_types.yaml)

Place this file next to main.py. It maps labels → diameter (mm):

# Example — adapt to your catalog
"16 AWG": 1.291
"18 AWG": 1.024
"20 AWG": 0.812
"1.00 mm": 1.000
"0.90 mm": 0.900
"Ethernet": 2.000

In the UI, choose Predefined Sizes to select one of these keys.

---

How it Works (Math)

Minimize the enclosing radius $R$ of a circle containing all wire disks.

Variables

• Centers $(x_i, y_i)$ for $i = 1..n$
• Outer radius $R$

Effective radii
$r_i^{\mathrm{eff}} = r_i (1 + m)$ where $m$ is the margin.

Constraints

1. Containment:
   $|c_i| + r_i^{\mathrm{eff}} \le R$

2. Non-overlap:
   $|c_i - c_j| \ge r_i^{\mathrm{eff}} + r_j^{\mathrm{eff}}$

3. Frozen core (from shields):
   $|c_i| \ge R_{\text{core}} + r_i^{\mathrm{eff}}$

Objective
Minimize $R$

---

Tips & Troubleshooting

• If optimization is slow, reduce initializations or max iterations.
• If wires overlap visually, increase margin a bit (e.g., 1–3%).
• Add Shielding stays disabled until a valid solution is available.
• Clear All resets everything (wires, layers, results, frozen core).

Screenshots

Main UI

This is the main window where you define wires, optimization parameters, and shielding layers:

---

Result

After optimization and adding shielding, the layout will look like this: