<p>In order to optimize the manufacture of an advanced transformation-induced plasticity (TRIP)-aided martensitic steel, a novel kernel-based gradient evolution approach is integrated into the multi-objective adaptative memory procedure and combined with a support vector regression model. To achieve this, a number of heat treatments are carried out at a temperature that is appropriate for manufacturing galvanized steel. Thus, the most critical process variables (cooling rates and a galvanizing temperature isothermal holding period) were tuned to provide the required mechanical property values. Generally speaking, the support vector regression model is taken as the goal function since it represents the extremely nonlinear relationship between the experimental parameters and the desired mechanical properties in a reasonable way. Additionally, the proposed method exhibits an exceptional performance with respect to the well-known multi-objective genetic algorithm, since it discovered a robust, wide Pareto front. Additionally, the ranges of the manufacturing variables recommended to produce transformation-induced plasticity-assisted martensitic steels are 57–63 <InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="170_2025_16115_Article_IEq1.gif" Format="GIF" Height="7" Rendition="HTML" Resolution="72" Type="Linedraw" Width="9" /> </InlineMediaObject> <EquationSource Format="TEX">$^{\circ }$</EquationSource> <EquationSource Format="MATHML"><math> <mmultiscripts> <mrow /> <mrow /> <mo>∘</mo> </mmultiscripts> </math></EquationSource> </InlineEquation>C/<InlineEquation ID="IEq2"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="170_2025_16115_Article_IEq2.gif" Format="GIF" Height="10" Rendition="HTML" Resolution="72" Type="Linedraw" Width="13" /> </InlineMediaObject> <EquationSource Format="TEX">$\varvec{s}$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi mathvariant="bold-italic">s</mi> </mrow> </math></EquationSource> </InlineEquation> for the first cooling rate (<InlineEquation ID="IEq3"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="170_2025_16115_Article_IEq3.gif" Format="GIF" Height="16" Rendition="HTML" Resolution="72" Type="Linedraw" Width="34" /> </InlineMediaObject> <EquationSource Format="TEX">$CR_{1}$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>C</mi> <msub> <mi>R</mi> <mn>1</mn> </msub> </mrow> </math></EquationSource> </InlineEquation>), 33–37 <i>s</i> for the holding period (<InlineEquation ID="IEq4"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="170_2025_16115_Article_IEq4.gif" Format="GIF" Height="16" Rendition="HTML" Resolution="72" Type="Linedraw" Width="14" /> </InlineMediaObject> <EquationSource Format="TEX">$t_{2}$</EquationSource> <EquationSource Format="MATHML"><math> <msub> <mi>t</mi> <mn>2</mn> </msub> </math></EquationSource> </InlineEquation>), and 1–2 <InlineEquation ID="IEq5"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="170_2025_16115_Article_IEq1.gif" Format="GIF" Height="7" Rendition="HTML" Resolution="72" Type="Linedraw" Width="9" /> </InlineMediaObject> <EquationSource Format="TEX">$^{\circ }$</EquationSource> <EquationSource Format="MATHML"><math> <mmultiscripts> <mrow /> <mrow /> <mo>∘</mo> </mmultiscripts> </math></EquationSource> </InlineEquation>C/<InlineEquation ID="IEq6"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="170_2025_16115_Article_IEq2.gif" Format="GIF" Height="10" Rendition="HTML" Resolution="72" Type="Linedraw" Width="13" /> </InlineMediaObject> <EquationSource Format="TEX">$\varvec{s}$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi mathvariant="bold-italic">s</mi> </mrow> </math></EquationSource> </InlineEquation> for the second cooling rate (<InlineEquation ID="IEq7"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="170_2025_16115_Article_IEq7.gif" Format="GIF" Height="16" Rendition="HTML" Resolution="72" Type="Linedraw" Width="34" /> </InlineMediaObject> <EquationSource Format="TEX">$CR_{2}$</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi>C</mi> <msub> <mi>R</mi> <mn>2</mn> </msub> </mrow> </math></EquationSource> </InlineEquation>).</p>

A novel multi-objective optimization intelligent technique applied to the design and manufacturing process of advanced transformation-induced plasticity aided martensitic steel

期刊论文
发表日期：

2025年7月18日

查看全文

Carlos O. Flor-Sánchez,
Edgar O. Reséndiz-Flores,
Gerardo Altamirano-Guerrero,
Rogelio Deaquino-Lara

摘要

In order to optimize the manufacture of an advanced transformation-induced plasticity (TRIP)-aided martensitic steel, a novel kernel-based gradient evolution approach is integrated into the multi-objective adaptative memory procedure and combined with a support vector regression model. To achieve this, a number of heat treatments are carried out at a temperature that is appropriate for manufacturing galvanized steel. Thus, the most critical process variables (cooling rates and a galvanizing temperature isothermal holding period) were tuned to provide the required mechanical property values. Generally speaking, the support vector regression model is taken as the goal function since it represents the extremely nonlinear relationship between the experimental parameters and the desired mechanical properties in a reasonable way. Additionally, the proposed method exhibits an exceptional performance with respect to the well-known multi-objective genetic algorithm, since it discovered a robust, wide Pareto front. Additionally, the ranges of the manufacturing variables recommended to produce transformation-induced plasticity-assisted martensitic steels are 57–63 $^{\circ }$ $^{\circ}$ C/ $\varvec{s}$ $s$ for the first cooling rate ( $CR_{1}$ $C R_{1}$ ), 33–37 s for the holding period ( $t_{2}$ $t_{2}$ ), and 1–2 $^{\circ }$ $^{\circ}$ C/ $\varvec{s}$ $s$ for the second cooling rate ( $CR_{2}$ $C R_{2}$ ).