Power Plant Energy Prediction with Artificial Neural Networks (ANN)

📝 Table of Contents

• Project Overview
• Key Features
• Dataset Description
• Model Architecture
• Training Parameters
• Performance Metrics
• Installation Guide
• Usage Instructions
• Results Visualization
• License
• Contact Information

🌟 Project Overview

This project implements an Artificial Neural Network (ANN) to predict the net hourly electrical energy output (PE) of a combined cycle power plant. The model achieves a Mean Squared Error (MSE) of 26.86 on the training data, demonstrating strong predictive capability for energy generation forecasting.

🚀 Key Features

• Complete data preprocessing pipeline
• ANN implementation with TensorFlow/Keras
• Comprehensive model evaluation
• Visual comparison of actual vs predicted values
• Batch training optimization

📊 Dataset Description

File: Folds5x2_pp.xlsx (9,568 samples, 5 features)

Features:

• AT: Ambient Temperature (°C)
• V: Exhaust Vacuum (cm Hg)
• AP: Ambient Pressure (millibar)
• RH: Relative Humidity (%)

Target Variable:

• PE: Net Hourly Electrical Energy Output (MW)

🧠 Model Architecture

ann = tf.keras.models.Sequential()
ann.add(tf.keras.layers.Dense(units=6, activation='relu', input_shape=(X_train.shape[1],)))
ann.add(tf.keras.layers.Dense(units=6, activation='relu'))
ann.add(tf.keras.layers.Dense(units=1))

⚙️ Training Parameters

Neural Network Configuration:

Parameter	Value	Description
Optimizer	Adam	Adaptive Moment Estimation
Loss Function	MSE	Regression loss metric
Batch Size	32	Samples per weight update
Epochs	100	Full training cycles
Test Size	20%	Holdout for evaluation

📈 Performance Metrics

Training Progress

• Initial Loss: 125,187.69 (Epoch 1)
• Final Loss: 26.86 (Epoch 100)
• 99.79% loss reduction during training

📥 Installation Guide

1. Clone repository:

git clone https://github.com/barisgudul/ANN-Energy-Prediction.git
cd ANN-Energy-Prediction

2. Create virtual environment:

# Unix/macOS
python -m venv churn_env
source churn_env/bin/activate

# Windows
python -m venv churn_env
churn_env\Scripts\activate

🖥️ Usage Instructions

1. Place dataset in ANN_for_Power_Plant_Energy\Dataset\Folds5x2_pp Run Jupyter notebook:

jupyter notebook main.ipynb

2. Execute cells sequentially
3. Check Results/ directory for generated outputs

📊 Results Visualization

Actual vs Predicted Values

Actual vs Predicted

Key Observations:

• Tight clustering around diagonal line indicates strong predictive accuracy
• Model maintains performance across different energy output ranges
• Minor variance in higher energy output ranges

📄 License

MIT License

Permissions:
✅ Free academic/research use
✅ Modification and redistribution
❌ Commercial use requires written consent

Full license terms available in LICENSE file.

---

📧 Contact Information

Project Maintainer

Contribution Guidelines:
We welcome collaborations! Please reach out via email before submitting PRs.

Technical Support:
For implementation assistance, include "[Power Plant]" in your email subject line.