Prediction of micro abrasive intermittent jet machining process using adaptive neuro-fuzzy inference system

• Main Authors: Shashank, V., Varma, C. V. Mahendra, Chaudhari, Devendra, Sasank, V. Sai, Jagadesh, T.
• Format: Conference Proceeding
• Language: English
• Subjects: Abrasive machining; Abrasives; Adaptive systems; Artificial neural networks; Back propagation; Biomedical materials; Electronic devices; Experimentation; Fuzzy logic; Fuzzy systems; Inference; Mathematical models; Penetration depth; Predictions; Process parameters; Surgical implants; Syringes; Weighting functions
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/1.5120234

Micro abrasive jet machining products have been predominantly used in biomedical applications such as needle for syringes, neuro surgical implants and also for cutting of glass in electronic devices. However, this is one of the complex micro manufacturing processes as it involves several process parameters. The penetration depth is unknown so it is difficult to calculate the amount of abrasives required for micro abrasive jet machining process. And moreover nano abrasives particles are costly and experimentation is expensive. So the objective of this paper is to develop an adaptive neuro-fuzzy inference system (ANFIS) for understanding the physics of the process and predict the process parameters to get better feature geometry for micro manufacturing applications. ANFIS model has been developed by considering hybrid and back propagation algorithm, weight function, number of data in training and testing, different layers and shape of membership function. The proposed model has been calibrated and verified using neuro fuzzy model and experimental results. There is a good agreement between the results measured experimentally and predicted diameter and depth of the hole geometry. The average prediction error for the diameter and depth of machined holes are 4.19 % and 8.86 % respectively.