نویسندگان
1 دانشکده مهندسی مکانیک دانشگاه تربیت دبیر شهید رجایی تهران
2 دانشکده مهندسی مکانیک دانشگاه تبریز
چکیده
کلیدواژهها
عنوان مقاله [English]
نویسندگان [English]
The cutting tool wear degrades the quality, reliability and productivity of the product in the manufacturing process. Accordingly, an on-line monitoring of the cutting tool wear level is essential to preventany deterioration. Unfortunately, there is no direct method to measure the cutting toolwear on-line. Consequently, an indirect method can be adopted where wear will be estimatedfrom the measurement of one or more physical parameters appearing during themachining process such as vibrations, electrical current, cutting force, etc.In this paper, two techniques namely Adaptive Neuro - Fuzzy Inference System (ANFIS) and Multi-Layer Perceptron (MLP) have been used for prediction of tool wear in face milling. For this purpose, a series of experiment is carried out on a milling machine. It is observed that there was an increase in the current amplitude with increasing the tool wear. Besides, the effects of tool wear, feed, and depth of cut on the current are analyzed. Comparison of the tool wear detection techniques shows 92% of correct tool wear detection for ANFIS and 84% for MLP. As a result, ANFIS can be proposed as proper technique for intelligent fault detection of the tool wear and breakage due to its high efficiency in diagnosing wear and tool breakage.
کلیدواژهها [English]
)
