Abstract
Adaptive neuro-fuzzy inference systems (ANFIS) were used for on-line classification and measurement of tool wear for the boring of titanium parts. The input vectors consist of extracted features from cutting force data. A total of fourteen features were extracted by processing cutting force signals using virtual instrumentation. Feature selection was carried out using a Sequential Forward Search (SFS) algorithm to select the best combination of features. For the on-line classification, the outputs are boring tool conditions, which are either usable or worn out. For the on-line measurement, the outputs are estimated values of the tool wear. Using ANFIS, three features were selected for the on-line classification of boring tools. They are the average longitudinal force, average of the ratio between the tangential and radial forces, and kurtosis of the longitudinal force. Only one feature, kurtosis of the longitudinal force, was needed for the on-line measurement of tool wear using ANFIS. A 3×5 ANFIS can achieve a 100% success rate for the on-line classification of boring tool conditions. Using a 1×5 ANFIS, the average flank wear estimation error is below 5% for on-line measurement of tool wear.