Neural Network Model for Machinability Investigations on CNC Turning of AA5052 for Marine Applications with MQL

2022-28-0515

12/23/2022

Features
Event
International Conference on Advances in Design, Materials, Manufacturing and Surface Engineering for Mobility
Authors Abstract
Content
Aluminium alloys are attracting importance in various engineering industries because of their exceptional characteristics such as strength, resistance to oxidation etc., AA5052 is an alloy that categorized under Al-Mg series, commonly adopted in anti-rust applications, especially for desalination applications because of its good corrosion resistance in seawater at temperatures up to 125°C, low cost, good thermal conductivity, and non-toxicity of its corrosion products. Minimum Quantity Lubrication (MQL) is one of the approaches that are economically affordable and also eco-friendly used in various machining operations. This present exploration details the investigation CNC turning of AA5052 alloy with conventional Tungsten Carbide (WC) tool inserts under MQL conditions. There are two different natural cutting fluids were engaged such as live oil and coconut oil. Taguchi’s experimental design concept is adopted for planning the experiments by considering speed, feed and depth of cut are considered as independent process variables. An L27orthogonal array has been opted for performing the experiments. Material Removal Rate and Surface Roughness are considered as performance measures. The optimum process parameters were determined by Taguchi’s single response analysis. The significance process variables are analysed by ANOVA analysis. Appropriate use of artificial intelligent decision-making tools assists the manufacturer to attain benefits in various engineering applications. By keeping this viewpoint, an artificial neural network model is developed for predicting the desired performance measures. The performance of the developed model is also analysed. The predicted values were further compared with the actual values obtained from the experimentation. The closeness among these values shows the effectiveness of the developed model and it proves that the model is capable of predicting the desired performance measures with precision.
Meta TagsDetails
DOI
https://doi.org/10.4271/2022-28-0515
Pages
6
Citation
Katta, L., Natarajan, M., Pasupuleti, T., Sivaiah, P. et al., "Neural Network Model for Machinability Investigations on CNC Turning of AA5052 for Marine Applications with MQL," SAE Technical Paper 2022-28-0515, 2022, https://doi.org/10.4271/2022-28-0515.
Additional Details
Publisher
Published
Dec 23, 2022
Product Code
2022-28-0515
Content Type
Technical Paper
Language
English