Several challenges are encountered in the development of maraging steels with desired combinations of mechanical properties. These include the need to explore a large process design space, the time- and effort-consuming standardized testing protocols for property evaluations, and the lack of a formal design of experiments strategy that guides the selection of the process conditions for the next set of experiments based on a thorough analyses of the previously accumulated data. In this work, we explored the effect of the different combinations of the aging temperature and the aging time on the yield strength of 350-grade maraging steels using high-throughput protocols. For this purpose, a total of 21 small volumes of differently aged samples were produced and studied using spherical nanoindentation stress–strain protocols. Furthermore, the results of these tests were modeled using Gaussian process regression (GPR) to establish data-driven linkages between the yield strengths and the aging parameters. The predicted yield strengths were found to be in reasonable agreement with the experimentally measured ones. It is also demonstrated that the GPR model can provide objective guidance in the selection of the next set of experiments.

在具有期望的机械性能组合的马氏体时效钢的开发中遇到了几个挑战。其中包括需要探索较大的过程设计空间，用于性能评估的费时且费力的标准化测试协议，以及缺乏正式的实验策略设计来指导下一组实验的工艺条件的选择基于对先前积累的数据的全面分析。在这项工作中，我们使用高通量协议探索了时效温度和时效时间的不同组合对350级马氏体时效钢屈服强度的影响。为此，使用球形纳米压痕应力-应变方案生产并研究了总共21小份不同老化的样品。此外，使用高斯过程回归（GPR）对这些测试的结果进行建模，以建立屈服强度与时效参数之间的数据驱动链接。发现预测的屈服强度与实验测量的屈服强度在合理的范围内。还证明了GPR模型可以为选择下一组实验提供客观指导。