Please use this identifier to cite or link to this item:
https://idr.nitk.ac.in/jspui/handle/123456789/14569
Title: | Investigations on Machinability Characteristics of EN47 Spring Steel using Optimization Techniques |
Authors: | Vasu, M. |
Supervisors: | Nayaka, H Shivananda. |
Keywords: | Department of Mechanical Engineering;EN47 Spring Steel;Coated tungsten carbide tool;Cutting forces;Surface roughness;Wet machining;Cryogenic machining;Tool Wear;Material removal rate;Chip morphology;Surface integrity;Artificial neural network;Particle Swarm Optimization;Response Surface Methodology;tool tip temperature;Desirability Approach |
Issue Date: | 2019 |
Publisher: | National Institute of Technology Karnataka, Surathkal |
Abstract: | Challenge of any manufacturing industry to give a better quality of products to society with minimum manufacturing cost, low manufacturing time and less consumption of raw material. Manufacturing involves various processes to convert raw material into finished products and hence meet demands with high-quality products. Selection of process parameters plays a significant role to satisfy all demands to ensure the quality of the product, increased production rate, and reduced operating cost. For such cases, optimization is essential to represent manufacturing process. Process parameters have been optimized by chosen best possible optimization techniques. Before conducting any experiments, selection of workpiece and tools it is necessary, to explore the literature to know, what has happened in earlier days. A literature survey has been done thoroughly existing statistical techniques are understood and implemented to optimize speed, feed, and depth of cut. EN47 spring steel has been chosen as work material which has a hardness of 45-48HRC. Hard turning process eliminates grinding process, and EN47 steel possesses low thermal conductivity and suitably oil hardened and tempered. Hardened spring steel offers excellent toughness and shock resistance, and are considered as suitable material for automobile applications. Other applications involve such as manufacturing of die, leaf spring for a heavy vehicle, crankshaft, spindles, pumps and steering knuckles and many general engineering applications. Experiments were performed using two different techniques, namely, one factor at a time (OFAT) approach and Design of Experiments (DOE). Cutting tool inserts are commercially available in the form of PVD coated TiAlN German make and are used during machining. Cutting forces, surface roughness, tool tip temperature, and material removal rate are estimated experimentally. From the experimental work, it is known that with an increase in nose radius, cutting forces, tool tip temperature, and material removal rate are increased, but surface roughness is decreased. Further, a tool with 0.8mm nose radius exhibits nominal performance in all output performances. 0.8mm nose radius tools are used to work in three different cutting environments, namely dry, wet and cryogenic. From the analysis, cryogenic machining showed better quality of the machined surface, tool wear also reduced and tool tip temperature decreased.viii Experiments were performed and analyzed using design of experiments (DOE) technique L27 full factorial design. A second order regression model was developed to know the interaction effect of output responses. Tool wear was analyzed by confocal microscope and SEM, with varying cutting time. ANOVA was used to identify the significant factor and percentage contribution for a particular output. Results from machining reveal that cutting force is mainly influenced by feed rate and depth of cut. Surface roughness was influenced by cutting speed and feed rate. Tool tip temperature was influenced by cutting speed and depth of cut. Material removal rate was influenced by speed, feed, and depth of cut. 3D response surface plots show interaction effect on each output response. Main effects plots show optimum condition for each output performance. Normal probability plots showed that the developed models are adequate by observing normal error distribution. Determination coefficient (R2) value should be in between 1 or 100% in the model. Multi-objective optimization was identified by Desirability Approach (DA) and Particle Swarm Optimization (PSO). Also, Artificial Neural Network (ANN) is used to predict experimental results and compared with RSM model, as well as, experimental value. Statistical analysis was done by Minitab and Design Expert Software. Validation was performed by ANN. MATLAB is used to develop artificial neural network model, as well as; codes are developed for PSO. From the experimental analysis, the developed model showed a significant and good agreement between the experimental value and predicted value. |
URI: | http://idr.nitk.ac.in/jspui/handle/123456789/14569 |
Appears in Collections: | 1. Ph.D Theses |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
155019ME15F15.pdf | 7.67 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.