ZPE / ZPF Phenomenological Kernel

An instrumentation-first Python kernel for testable ZPF-style boundary-coupled spectral experiments.

This repository open-sources a single-file phenomenological stack inspired by Douglas Miller's public ZPF Technologies testable vacuum framework and built for observable-first lab workflows. It is not a claim of new physics. It is a tool for generating forward predictions, comparing them against measurements, and making the assumptions explicit enough to test.

What It Does

The kernel implements a forward spectral model:

spectrum(omega) = g(omega) x P_occ(omega; T) x N_b(omega; geometry, phi_q, loss, x)

where:

• g(omega) is the mode supply
• P_occ(omega; T) is thermal / ZPF participation
• N_b(omega) is a phenomenological boundary-coupling surrogate
• phi_q = (V_eff / V_conf)^(1/3) is the quantum packing fraction

Observable heads include:

• impedance shift
• phase sensitivity
• spectral change
• nonlinear response
• force gradient
• energy proxy

Capabilities

• forward spectral model with explicit g x P_occ x N_b factorization
• analytic box geometry or real mesh ingestion through optional trimesh
• quantum packing fraction helpers and phi_q sweeps
• temperature-vs-packing phase diagrams
• lossy force-gradient scans
• CSV/JSON lab-data loader
• robust least-squares fitting through optional scipy
• JSON summaries and plot outputs for reproducible experiment handoff

Quick Start

Requires Python 3.10+ with numpy and matplotlib. Optional fitting and mesh support use scipy and trimesh.

git clone https://github.com/soulhash-labs/zpe.git
cd zpe
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
python zpf_phenom_kernel.py demo

Outputs are written to phenomenology_out/ by default:

• summary.json
• spectrum.png

For fitting and mesh support:

pip install -r requirements-optional.txt

CLI Examples

Run a single forward model:

python zpf_phenom_kernel.py demo --temp-K 300 --phi-q 0.4 --outdir out/demo

Generate a temperature-vs-packing phase diagram:

python zpf_phenom_kernel.py phase-diagram \
  --observable energy_proxy \
  --temp-min 5 --temp-max 800 \
  --phi-min 0.05 --phi-max 1.2 \
  --outdir out/phase

Run a lossy force-gradient scan:

python zpf_phenom_kernel.py loss-scan \
  --loss-min 1e-6 --loss-max 1e-1 \
  --gap-min -1e-3 --gap-max 1e-3 \
  --outdir out/loss

Fit against a CSV or JSON lab dataset:

python zpf_phenom_kernel.py fit-data \
  --dataset measurements.csv \
  --data-targets impedance_shift,force_gradient \
  --outdir out/fit

Run a synthetic recovery demo, which requires scipy:

python zpf_phenom_kernel.py fit-demo --outdir out/fit-demo

Data Format

CSV input expects independent variables plus one or more target columns:

temperature_K,phi_q,gap_x_m,impedance_shift,force_gradient,sigma_impedance_shift,sigma_force_gradient
300,0.40,0.0,1.2e-6,-3.0e-24,1.0e-7,2.0e-25

Recognized independent-variable aliases:

• temperature_K, temp_K, temperature
• phi_q, phi
• gap_x_m, gap, displacement_m

JSON input can be either dict-of-arrays:

{
  "temperature_K": [300],
  "phi_q": [0.4],
  "gap_x_m": [0.0],
  "targets": {
    "impedance_shift": [1.2e-6],
    "force_gradient": [-3.0e-24]
  }
}

or a list of row objects with the same fields.

Geometry

Analytic box geometry is built in:

python zpf_phenom_kernel.py demo --extents 0.04,0.03,0.02

Real mesh support is available when trimesh is installed:

python zpf_phenom_kernel.py demo --mesh cavity.stl

The mesh path is reduced to a geometry descriptor with extents, area, volume, and metadata. This is a boundary surrogate, not a full FEM/BEM Maxwell solve.

Scientific Scope

This kernel is deliberately phenomenological:

• it is not a full electromagnetic solver
• it is not a claim of zero-point energy extraction
• it is not proof of a vacuum-energy mechanism
• it is a test harness for observables, parameter sweeps, and lab-data fitting

The intended workflow is simple:

1. define geometry and experimental conditions
2. generate model predictions
3. measure the corresponding observable heads
4. fit parameters against real CSV/JSON data
5. reject, revise, or refine the phenomenological assumptions

No hand-wavy theory. Just observables you can actually measure.

License Recommendation

This repository uses Apache-2.0.

That is the recommended license for this project because it is permissive for open science and commercial lab adoption while also providing an explicit patent grant. If the project later needs a stronger copyleft requirement, GPLv3 would be the obvious alternative, but Apache-2.0 is the best default for broad experimental reuse.

Citation / Attribution

Framework inspiration:

• Douglas Miller / ZPF Technologies on X: https://x.com/zpfTechnologies
• ZPF Technologies on Substack: https://substack.com/@zpftechnologies

If you use this kernel in an experiment, include:

soulhash-labs/zpe: ZPE / ZPF Phenomenological Kernel
https://github.com/soulhash-labs/zpe

Run It Today

python zpf_phenom_kernel.py demo

This is the bridge between a testable vacuum note and a real experiment.

Built instrumentation-first. No new physics claimed. Just tools to test it.