Omdurman Islamic University — Faculty of Engineering Sciences

Confronting
Chaos

Improving Photovoltaic Energy Systems using Physics-Guided Hybrid Forecasting Model

SCROLL TO BEGIN

The Challenge We Face

In control engineering, we search for stability and accuracy. But we chose to confront the most complex non-linear system: the Earth's atmosphere.

The atmosphere is a chaotic dynamical system — a suspended speck of dust in Sudan's skies can overturn the energy balance in seconds.

We prove that merging atmospheric physics with deep AI can decode this chaos and transform reactive systems into proactive ones.

KEY MISSION ℹ️
🎯

Design a system that anticipates atmospheric volatility instead of merely reacting to it — using physics-guided AI and hierarchical control.

72 hours × 3 scenarios × 15 features × 5hr prediction horizon

Paradigm Shift

Toggle between the old reactive approach and our proactive solution.

🌪️

Atmospheric Volatility ℹ️

Dust storms (Haboob) cause GHI drops exceeding 60% in minutes, destroying grid stability.

⏱️

Phase Lag Failure ℹ️

Standard AI predicts yesterday's weather instead of tomorrow's — useless for control.

🔋

Reactive Control ℹ️

Systems respond after changes occur. Always one step behind — energy waste and degradation.

🧮

Computational Burden ℹ️

Transformer models (>2M params) are too heavy for real-time edge deployment.

🧠

Physics-Guided AI ℹ️

CNN-BiLSTM with 15 physics features. 7 are zero-cost from solar geometry.

Dual-Level Architecture ℹ️

Level 1: Reference model. Level 2: Lightweight surrogate.

🎯

Proactive MPC Control ℹ️

MPC uses future predictions to act before weather events impact the system.

🔧

FOPID Precision ℹ️

Fractional-Order PID with (λ, μ) provides superior flexibility over integer PID.

Chaos ↔ Order

Left: Chaotic atmospheric data — unpredictable motion.

Right: Physics-guided flow — our hybrid model imposing constraints.

Neural Data Flow

CNN: Extracts spatial patterns from atmospheric data with high precision.

BiLSTM: Comprehends temporal sequencing over time and uncovers complex dependencies.

Objectives & Scope

01 — Hybrid Prediction Architecture ℹ️

CNN for spatial feature extraction + BiLSTM for temporal sequence — guided by 15 physics-informed features, optimized via Bayesian hyperparameter tuning.

02 — Advanced Hierarchical Control ℹ️

FOPID + MPC strategy for intelligent proactive energy and water management — shifting from reactive to anticipatory decision-making.

03 — Integrated Simulation Platform ℹ️

MATLAB/Simulink simulation combining forecasting + PV pumping under hyper-realistic scenarios — benchmarking against LSTM, GRU, Transformer baselines.

04 — Economic & Practical Impact ℹ️

Eliminating expensive sensors through zero-cost solar geometry features and extending mechanical equipment lifespan via proactive, wear-reducing control.
0
FEATURES
0
ZERO-COST
0
PARAMETERS
0
CHAPTERS

Begin The Journey

The introduction is complete. Explore the control system or dive into the chapters.

CONTROL SYSTEM → EXPLORE CHAPTERS CONCLUSION 🏁