In the biomedical industry, biomaterials are vital for advancing medical treatments, particularly in developing implants used in orthopedics, dentistry, and cardiovascular care, which can be permanent or temporary. Biodegradable materials, like magnesium alloys, naturally break down in the body, removing the need for secondary surgeries. However, challenges remain in preventing premature degradation and improving their bio functionality. Thus, this work intends to optimize the powder-mixed electrical discharge machining (EDM) for processing biomedical alloy AZ31BMg, utilizing the soft computing method, which includes the mixed-variable Jaya algorithm. Further investigation of the effect of cryogenically treated tools on machine samples for their vitro cytotoxicity, biocompatibility, antibacterial properties, and surface properties. The EDM parameters (polarity, current, electrode materials, pulse-off time, pulse-on time, and powder mixed dielectric) were experimentally examined via Taguchi design of experiments to address the required outputs: material removal rate, tool wear ratio, and surface roughness. The experimental findings were processed to address the response correlations and trade-offs and transform the response data into an overall process index. They were fed into the Bayesian regularized neural network, which generated an outstanding mapping of the EDM parameters vs. the overall process index in terms of an excellent generalization with no overfitting. Since the observed machining parameters include continuous numerical and attribute variables, the mixed-variable Jaya (MixVarJaya) was designed and implemented to optimize the parameters while considering their heterogeneous nature. The obtained optimal machining conditions resulted in a high process index (<InlineEquation ID="IEq1"> <EquationSource Format="TEX">$\gamma = 0.93607$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>γ</mi> <mo>=</mo> <mn>0.93607</mn> </mrow> </math></EquationSource> </InlineEquation>), demonstrating a highly satisfactory model are P straight, Cp 4A, EM Tungsten, Ton 200 μs, Toff = 150 μs, DE EDM oil + Zn nanopowder. Further, experimentation is performed using cryogenically treated tungsten tool electrodes at the optimal parametric setting for comparative evaluation of biological responses biocompatibility, cytotoxicity, and antibacterial. The formation of apatite content on TTS is greater than that of UTS; hence, more apatite formation revealed more biocompatibility. The MTT assessment for cytotoxicity of TTS showed 97.41% cell vitality, higher than UTS’s 79.89% cell vitality. Additionally, using TTS and UTS, respectively, S. aureus and E. coli bacteria were inhibited on machined surfaces by 78.29% and 73.38%. AZ31BMg processing in the powder-mixed EDM using a cryogenically treated tool enhances biocompatibility.

Multi-Objective Optimization of Powder Mixed Electrical Discharge Machining Parameters in the Processing of AZ31BMg Using a Soft Computing Approach Based on the Mixed-Variable Jaya Algorithm to Improve the Bio-Functionality

期刊论文
发表日期：

2025年5月26日

查看全文

Sandeep Kumar,
Abhishek Singh

摘要

In the biomedical industry, biomaterials are vital for advancing medical treatments, particularly in developing implants used in orthopedics, dentistry, and cardiovascular care, which can be permanent or temporary. Biodegradable materials, like magnesium alloys, naturally break down in the body, removing the need for secondary surgeries. However, challenges remain in preventing premature degradation and improving their bio functionality. Thus, this work intends to optimize the powder-mixed electrical discharge machining (EDM) for processing biomedical alloy AZ31BMg, utilizing the soft computing method, which includes the mixed-variable Jaya algorithm. Further investigation of the effect of cryogenically treated tools on machine samples for their vitro cytotoxicity, biocompatibility, antibacterial properties, and surface properties. The EDM parameters (polarity, current, electrode materials, pulse-off time, pulse-on time, and powder mixed dielectric) were experimentally examined via Taguchi design of experiments to address the required outputs: material removal rate, tool wear ratio, and surface roughness. The experimental findings were processed to address the response correlations and trade-offs and transform the response data into an overall process index. They were fed into the Bayesian regularized neural network, which generated an outstanding mapping of the EDM parameters vs. the overall process index in terms of an excellent generalization with no overfitting. Since the observed machining parameters include continuous numerical and attribute variables, the mixed-variable Jaya (MixVarJaya) was designed and implemented to optimize the parameters while considering their heterogeneous nature. The obtained optimal machining conditions resulted in a high process index ( $\gamma = 0.93607$ $γ = 0.93607$ ), demonstrating a highly satisfactory model are P straight, C_p 4A, EM Tungsten, T_on 200 μs, T_off = 150 μs, DE EDM oil + Zn nanopowder. Further, experimentation is performed using cryogenically treated tungsten tool electrodes at the optimal parametric setting for comparative evaluation of biological responses biocompatibility, cytotoxicity, and antibacterial. The formation of apatite content on TTS is greater than that of UTS; hence, more apatite formation revealed more biocompatibility. The MTT assessment for cytotoxicity of TTS showed 97.41% cell vitality, higher than UTS’s 79.89% cell vitality. Additionally, using TTS and UTS, respectively, S. aureus and E. coli bacteria were inhibited on machined surfaces by 78.29% and 73.38%. AZ31BMg processing in the powder-mixed EDM using a cryogenically treated tool enhances biocompatibility.