The creep behavior of 316 L stainless steel at room temperature was evaluated as a function of time and applied stress using a new high-Throughput approach. Several common creep models were evaluated against the observations, leading to deeper analysis of a stress-dependent modified logarithmic creep model. Within this model, multiple sources of uncertainty were compared. Aleatoric stochastic variation between samples under nominally identical conditions was identified as the primary contributor to uncertainty in creep response. Under any particular set of conditions, the sample-To-sample variability in creep strain was as high as a factor of two, highlighting the engineering importance of characterizing large statistical datasets. The model's extrapolation capabilities were assessed by comparing predictions derived from calibration on partial, shorter-duration subsets of the data. These findings underscore the importance of accounting for stochastic effects in predictive modeling of aging phenomena.
Mechanical metamaterials are artificial materials with unique global properties due to the structural geometry and material composition of their unit cell. Typically, mechanical metamaterial unit cells are designed such that, when tessellated, they exhibit unique mechanical properties such as zero or negative Poisson's ratio and negative stiffness. Beyond these applications, mechanical metamaterials can be used to achieve tailorable nonlinear deformation responses. Computational methods such as gradient-based topology optimization (TO) and size/shape optimization (SSO) can be implemented to design these metamaterials. However, both methods can lead to suboptimal solutions or a lack of generalizability. Therefore, this research used deep reinforcement learning (DRL), a subset of deep machine learning that teaches an agent to complete tasks through interactive experiences, to design mechanical metamaterials with specific nonlinear deformation responses in compression or tension. The agent learned to design the unit cells by sequentially adding material to a discrete design domain and being rewarded for achieving the desired deformation response. After training, the agent successfully designed unit cells to exhibit desired deformation responses not experienced during training. This work shows the potential of DRL as a high-level design tool for a wide array of engineering applications.
Mechanical metamaterials are regularly implemented in engineering applications due to their unique properties derived from their structural geometry and material composition. This study incorporates deep reinforcement learning, a subset of machine learning that teaches an agent to complete a task through interactive experiences, into mechanical metamaterial design. The approach creates a design environment for the reinforcement learning agent to iteratively construct metamaterials with tailorable deformation and hysteretic characteristics. Validation involved producing metamaterials with a thermoplastic polyurethane (TPU) base material that exhibited the deformation response of expanded thermoplastic polyurethane (E-TPU) while maximizing or minimizing hysteresis in cyclic compression. This alignment confirmed the feasibility of tailoring deformation and energy manipulation using mechanical metamaterials. The agent's generalizability was tested by tasking it to create various metamaterials with distinct loading deformation responses and specific hysteresis goals in a simulated setting. The agent consistently delivered metamaterials that met loading curve criteria and demonstrated favorable energy return. This work demonstrates the potential of deep reinforcement learning as a rapid and effective tool for designing mechanical metamaterials with customizable traits. It ushers in the possibility of on-demand metamaterial design solutions, opening avenues across industries like footwear, wearables, and medical equipment.
