Final Project: Wine Grape Variety Identification Machine#
Rough Sketch#
Overview#
This project aims to develop a system that identifies wine aroma compounds using machine learning techniques and displays the results on a screen. The goal is to create technology that automatically identifies wine grape varieties, supporting wine enthusiasts and professionals in their evaluation process.
System Configuration#
Gas Sensor Module#
The system uses an array of multiple high-sensitivity gas sensors to detect volatile organic compounds (VOCs) emitted from wine. Each sensor has different sensitivity to specific chemicals, functioning as an “electronic nose.”
Project Tasks#
1.Hardware Design and Implementation#
- Purchase of Xiao for machine learning
- Acquisition of gas sensors
- Gas sensor functionality testing
- Electronic circuit board preparation
- OLED display implementation
- Enclosure creation
2.Data Collection and Preprocessing#
- Collection of data from various wine samples
- Data labeling process
- Splitting data into test, validation, and training sets
3.Machine Learning Model Development#
- Finalization of machine learning clustering methods
- Training of the machine learning model
- Fine-tuning for optimal performance
4.System Integration and Testing#
- Design of OLED display information interface
- Integration of hardware and software components
- Final system testing and optimization
Project Timeline#
The project will proceed according to the following schedule:
Detailed Project Schedule#
| Domain | Task | Status | Start Date | Due Date |
|---|---|---|---|---|
| Hardware | Buy Xiao for machine learning | Done | 2025-04-13 | 2025-04-13 |
| Hardware | Buy gas sensor | Done | 2025-04-14 | 2025-04-15 |
| Hardware | Make the gas sensor work | Done | 2025-04-16 | 2025-04-19 |
| Hardware | Prepare electronic circuit board | Done | 2025-04-20 | 2025-04-22 |
| Software/Hardware | Display on OLED | Done | 2025-04-23 | 2025-04-26 |
| Data/Software | Collect data | Planned | 2025-04-27 | 2025-05-01 |
| Machine Learning | Finalize the machine learning clustering method | Planned | 2025-05-02 | 2025-05-04 |
| Data/Software | Label the data | Planned | 2025-05-05 | 2025-05-07 |
| Data/Software | Split the data into test, validation, and training sets | Planned | 2025-05-08 | 2025-05-09 |
| Machine Learning | Train the machine learning model | Planned | 2025-05-10 | 2025-05-14 |
| Machine Learning | Perform fine tuning | Planned | 2025-05-15 | 2025-05-18 |
| Design/Software | Design the OLED display information | Planned | 2025-05-19 | 2025-05-21 |
| Hardware | Create enclosures | Planned | 2025-05-22 | 2025-05-25 |
Expected Outcomes#
- A system capable of identifying multiple grape varieties with high accuracy
- Compact and user-friendly interface
- Design that considers future scalability
- Contribution to efficiency improvement in quality control and evaluation processes in the wine industry