THERMO-NET

📖 Overview

Thermal Efficiency Index (η_T)

"Heat is not the enemy — unmanaged entropy is. THERMO-NET: Mastering the Dissipation." — Samir Baladi, April 2026

THERMO-NET introduces the first physics-informed AI framework for quantitative characterization and suppression of irreversible entropy production in high-entropy physical systems — the Thermal Efficiency Index (η_T). Built on three mathematically rigorous constructs spanning Neural Heat Transport Operator, Local Entropy Production Minimizer, and Thermo-Informational Coupling Tensor.

91.3%

Mean η_T

5-regime cross-validation

87.9%

Entropy Reduction

vs uncontrolled baseline

200 μs

Prediction Horizon

Look-ahead dissipation control

5

Thermal Regimes

15 mK – 1200 K

η_T

Thermal Efficiency Index

// η_T Composite Formula (Equation from paper) η_T = η_Carnot - σ_integrated / η_Carnot ∈ [0, 1] // Carnot efficiency η_Carnot = 1 - T_cold / T_hot // AI Correction with Thermal/Mechanical/Informational Bias η_T_adj = σ(η_T_raw + β_therm + β_mech + β_info) // Python implementation from thermo_net import ETAParameters, compute_eta params = ETAParameters( nhto=0.88, lepm=0.85, tict=0.82 ) result = compute_eta(params, regime='cmos_node')

3 Constructs

Three Physics-Informed Constructs

Construct	Description	Domain
NHTO	Neural Heat Transport Operator	Non-Fourier heat transport · Learned κ(r,t,θ)
LEPM	Local Entropy Production Minimizer	Entropy production control · σ(r,t) minimization
TICT	Thermo-Informational Coupling Tensor	Landauer erasure · Information-thermal bridge

AI Architecture

Physics-Informed Neural Network + Neural ODE

// PINN penalty layer constraints // • Second Law compliance: σ(r,t) ≥ 0 everywhere // • Energy conservation: integrated power input equals heat output // • Landauer limit: k_B·T·ln(2) minimum per bit erasure // Python implementation from thermo_net import ThermoNet model = ThermoNet.load_pretrained("thermo_net_v1.0.0") result = model.predict(temperature_field, power_map)

Validation Scope

Five Thermal Regimes

92.1%

Sub-2nm CMOS Node (R1)

1.8nm · 300K · 100 W/cm² · 12 platforms

91.7%

Photonic Crystal Reservoir (R2)

Q=1e6 · 300K · 10 platforms

93.4%

Cryogenic Qubit Array (R3)

64 qubits · 15mK · 8 platforms

89.6%

Heat Engine (R4)

900K/400K · 1 atm · 7 platforms

90.8%

Thermoelectric Harvester (R5)

Si · 600K/350K · ZT=1.5 · 6 platforms

📦 Installation

Quick setup

# Clone repository git clone https://github.com/gitdeeper11/THERMO-NET.git cd THERMO-NET # Install package pip install -e . # Run analysis python bin/compute_eta.py --system test # Verify installation python -c "from thermo_net import __version__; print(__version__)"

🔧 API Reference

Python interface

ETAParameters

Three physics-informed construct container

from thermo_net import ETAParameters params = ETAParameters( nhto=0.88, lepm=0.85, tict=0.82 )

compute_eta

η_T computation with regime-specific normalization

from thermo_net import compute_eta result = compute_eta(params, regime='cmos_node') print(result.value) # η_T value print(result.status) # EXCELLENT/GOOD/MODERATE/CRITICAL/COLLAPSE

ThermalStateTracker

Main framework entry point for thermal analysis

from thermo_net import ThermalStateTracker tracker = ThermalStateTracker.load_pretrained("thermo_net_v1.0.0") result = tracker.predict(temperature_field, power_map) print(result.eta) # η_T value print(result.entropy_rate) # Entropy production rate

🧩 Core Modules

THERMO-NET architecture

core/

3 Constructs

NHTO, LEPM, TICT, η_T

transport/

Transport

Cattaneo-Vernotte, phonon BTE

entropy/

Entropy

Entropy production, Landauer cost

models/

AI Models

SIREN, LSTM, MPC, PINN

sensors/

Sensors

Temperature, heat flux, power

monitoring/

Monitor

Real-time entropy tracking

👤 Author

Principal investigator

🌡️

Samir Baladi

Interdisciplinary AI Researcher — Thermodynamic Intelligence & Entropy Control Division

Ronin Institute / Rite of Renaissance

Samir Baladi is an independent researcher affiliated with the Ronin Institute, developing the Rite of Renaissance interdisciplinary research program. THERMO-NET is a physics-informed AI framework for entropy production control, integrating neural heat transport, entropy minimization, Landauer erasure, and PINN architecture.

No conflicts of interest declared. All code and data are open-source under MIT License.

📧 gitdeeper@gmail.com 🔗 ORCID: 0009-0003-8903-0029 🐙 GitHub 🦊 GitLab

📝 Citation

How to cite

@software{baladi2026thermonet, author = {Samir Baladi}, title = {THERMO-NET: Neural Thermodynamic Dissipation Management for High-Entropy Physical Systems}, year = {2026}, version = {1.0.0}, publisher = {Zenodo}, doi = {10.5281/zenodo.19760903}, url = {https://doi.org/10.5281/zenodo.19760903}, note = {Physics-Informed AI Framework for Entropy Production Control} }

"Heat is not the enemy — unmanaged entropy is. THERMO-NET: Mastering the Dissipation. Irreversible entropy production in high-entropy physical systems is not an unavoidable physical tax — it is a predictable, multi-channel dynamical process that can be measured, predicted, and controlled with 91.3% efficiency."