Create Your First Project
Start adding your projects to your portfolio. Click on "Manage Projects" to get started
Fuzzy Logic System for Optimizing Drilling Parameters and Tool Condition Monitoring
Project type
Dynamics & Control
Date
Jan 2022 - Feb 2022
Skills
MATLAB
This project focuses on the design and implementation of a Fuzzy Logic System aimed at selecting optimal machining parameters and monitoring the condition of cutting tools during material removal processes. The system was developed based on the study "Fuzzy Logic Based Data Selection for the Drilling Process" (Hashmi, K., Graham, I.D., & Mills, B., Liverpool John Moores University, 2000) and was tailored to adjust the drilling speed according to the hardness of the workpiece material.
Key Features:
- The project involved defining fuzzy input and output variables, with hardness as the input and drilling speed as the output. The fuzzy sets were constructed using triangular membership functions with a 50% overlap between adjacent sets to ensure smooth transitions.
- Mamdani-type fuzzy rules were implemented to correlate the input (material hardness) with the output (drilling speed). Each rule was designed based on expert knowledge of the drilling process, ensuring that the system would select appropriate machining parameters under varying conditions.
- The "centroid" method was used for defuzzification to calculate the optimal drilling speed based on the weighted average of all possible outputs. This approach ensures that the selected drilling speed is the most appropriate for the given hardness, leading to improved machining performance.
- The developed fuzzy logic system was tested against the original system from the referenced study. Comparisons were made between the proposed system's outputs and the expected values, with analyses focusing on absolute and relative errors. The results showed that the system effectively matched or exceeded the performance of the original, particularly in mid-range hardness values.
- The project also explored potential improvements, including adjusting the number of fuzzy sets, modifying overlap percentages, experimenting with different types of membership functions (e.g., Gaussian), and varying defuzzification methods.
