https://github.com/r-mahesh45/gas-turbine-energy-yield-prediction-forest-fires-burned-area-forecast
This project focuses on predicting turbine energy yield (TEY) using ambient and turbine parameters from a gas turbine dataset. It involves building neural network models to analyze 36,733 instances of sensor data, evaluating model performance using metrics like MSE and R², and optimizing predictions for improved energy efficiency.
https://github.com/r-mahesh45/gas-turbine-energy-yield-prediction-forest-fires-burned-area-forecast
forcasting neural-network tensorflow
Last synced: 8 months ago
JSON representation
This project focuses on predicting turbine energy yield (TEY) using ambient and turbine parameters from a gas turbine dataset. It involves building neural network models to analyze 36,733 instances of sensor data, evaluating model performance using metrics like MSE and R², and optimizing predictions for improved energy efficiency.
- Host: GitHub
- URL: https://github.com/r-mahesh45/gas-turbine-energy-yield-prediction-forest-fires-burned-area-forecast
- Owner: R-Mahesh45
- Created: 2024-03-07T10:32:05.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-01-04T12:52:02.000Z (11 months ago)
- Last Synced: 2025-01-30T07:16:11.643Z (10 months ago)
- Topics: forcasting, neural-network, tensorflow
- Language: Jupyter Notebook
- Homepage:
- Size: 5.85 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Gas Turbine Energy Yield Prediction & Forest Fires Burned Area Forecast
This repository contains two predictive modeling projects leveraging Neural Networks.
1. **Gas Turbine Energy Yield Prediction:** Predicting turbine energy yield (TEY) using ambient variables as features.
2. **Forest Fires Burned Area Forecast:** Predicting burned areas of forest fires using neural networks.
## Table of Contents
- [Objective](#objective)
- [Datasets](#datasets)
- [Methodology](#methodology)
- [Models](#models)
- [Results](#results)
- [Technologies Used](#technologies-used)
- [Installation](#installation)
- [Usage](#usage)
- [License](#license)
---
## Objective
- **Gas Turbines:** Predict turbine energy yield (TEY) using ambient and gas turbine parameters.
- **Forest Fires:** Predict the burned area of forest fires based on environmental and fire-related parameters.
---
## Datasets
- **Gas Turbines:** Contains 36,733 instances of 11 sensor measurements aggregated hourly. Key variables include ambient temperature, pressure, humidity, and turbine parameters.
- **Forest Fires:** Environmental factors and fire-related data to forecast burned area.
---
## Methodology
1. Data Preprocessing: Standardized datasets and removed inconsistencies using `pandas` and `numpy`.
2. Neural Network Modeling: Constructed models with multiple layers using `keras` and evaluated with RMSE and accuracy metrics.
3. Performance Evaluation: Compared models using RMSE for Gas Turbines and Accuracy for Forest Fires.
---
## Models
### Gas Turbines:
- **Model Architecture:**
```python
model = Sequential()
model.add(Dense(10, input_dim=10, activation='relu'))
model.add(Dense(5, activation='relu'))
model.add(Dense(4, activation='relu'))
model.add(Dense(1, activation='sigmoid'))
```
- **Compilation & Evaluation:**
```python
model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['mse'])
```
- **RMSE:** 17989.7305
- **MSE (%):** 1798973.05%
### Forest Fires:
- **Model Architecture:**
```python
model = Sequential()
model.add(Dense(20, input_dim=28, activation='relu'))
model.add(Dense(10, activation='relu'))
model.add(Dense(10, activation='relu'))
model.add(Dense(1, activation='sigmoid'))
```
- **Compilation & Evaluation:**
```python
model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
```
- **Accuracy:** 99.03%
---
## Results
- **Gas Turbines:** Achieved RMSE of 17989.7305, demonstrating the model's ability to predict turbine energy yield.
- **Forest Fires:** Achieved 99.03% accuracy in predicting burned areas of forest fires.
---
## Technologies Used
- **Programming Languages:** Python
- **Libraries:**
- `keras` (Deep Learning)
- `numpy` (Numerical Computing)
- `pandas` (Data Manipulation)
- `scikit-learn` (Model Evaluation)
---
## Installation
1. Clone the repository:
```bash
git clone https://github.com/R-Mahesh45/project-name.git
```
2. Install required libraries:
```bash
pip install -r requirements.txt
```
---
## Usage
1. Open the respective scripts for Gas Turbines or Forest Fires.
2. Run the scripts to preprocess data, build the model, and evaluate performance.