Scholarlink

<p>This study focuses on the modeling and optimization of wire electric discharge machining (WEDM) parameters for Nimonic-C263 superalloy using artificial neural networks and response surface methodology (RSM). Due to its exceptional mechanical properties and corrosion resistance, Nimonic-C263, a superalloy based on nickel with high strength, finds widespread application in the aerospace and power generation industries. However, its machinability remains challenging due to its high hardness and toughness. This study considers the input parameters of spark energy (SE), spark frequency (SF), and peak current (PC). The evaluation of machining performance is based on cutting rate (CR) and surface roughness (SR). Experimental trials utilize the Box–Behnken design, with the collected data employed to construct predictive models through RSM-based regression analysis and ANN modeling. The ANN model is trained to utilize a feed-forward backpropagation algorithm and is validated through statistical performance measures, including mean squared error and coefficient of determination (<i>R</i><sup>2</sup>), to ensure high prediction accuracy. Comparative analysis reveals that the ANN model demonstrates superior predictive capability over RSM due to its ability to capture complex nonlinear regression equations between input parameters and machining responses. The optimized parameters significantly enhance machining performance, making WEDM a viable method for precision machining of Nimonic-C263. The spark energy directly influences the thermal energy necessary for the WEDM process. The maximum CR 8.985&#xa0;mm/min was attained SE: 9&#xa0;J, SF: 24&#xa0;Hz, and PC: 4&#xa0;A with a moderate SR of 2.045&#xa0;µm. Conversely, SE: 8&#xa0;J, SF: 26&#xa0;Hz, and PC: 3&#xa0;A yielded a cutting rate of 4.521&#xa0;mm/min with a surface roughness of 1.867&#xa0;µm. The findings of this research provide a robust framework for optimizing WEDM of nickel-based superalloys using AI-driven modeling techniques, contributing to improved efficiency and surface integrity in advanced manufacturing applications.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Hybrid Response Surface Methodology and Artificial Neural Network Approach for Performance Optimization in Wire Electrical Discharge Machining of Nimonic-C263 Superalloy

  • 期刊论文

  • 发表日期：

    2025年8月25日

查看全文
  • Indra Jeet,
  • A. K. Srivastava,
  • Himanshu Bisaria

摘要

This study focuses on the modeling and optimization of wire electric discharge machining (WEDM) parameters for Nimonic-C263 superalloy using artificial neural networks and response surface methodology (RSM). Due to its exceptional mechanical properties and corrosion resistance, Nimonic-C263, a superalloy based on nickel with high strength, finds widespread application in the aerospace and power generation industries. However, its machinability remains challenging due to its high hardness and toughness. This study considers the input parameters of spark energy (SE), spark frequency (SF), and peak current (PC). The evaluation of machining performance is based on cutting rate (CR) and surface roughness (SR). Experimental trials utilize the Box–Behnken design, with the collected data employed to construct predictive models through RSM-based regression analysis and ANN modeling. The ANN model is trained to utilize a feed-forward backpropagation algorithm and is validated through statistical performance measures, including mean squared error and coefficient of determination (R2), to ensure high prediction accuracy. Comparative analysis reveals that the ANN model demonstrates superior predictive capability over RSM due to its ability to capture complex nonlinear regression equations between input parameters and machining responses. The optimized parameters significantly enhance machining performance, making WEDM a viable method for precision machining of Nimonic-C263. The spark energy directly influences the thermal energy necessary for the WEDM process. The maximum CR 8.985 mm/min was attained SE: 9 J, SF: 24 Hz, and PC: 4 A with a moderate SR of 2.045 µm. Conversely, SE: 8 J, SF: 26 Hz, and PC: 3 A yielded a cutting rate of 4.521 mm/min with a surface roughness of 1.867 µm. The findings of this research provide a robust framework for optimizing WEDM of nickel-based superalloys using AI-driven modeling techniques, contributing to improved efficiency and surface integrity in advanced manufacturing applications.