Search for: All records

The process instabilities intrinsic to the localized laser-powder bed interaction cause the formation of various defects in laser powder bed fusion (LPBF) additive manufacturing process. Particularly, the stochastic formation of large spatters leads to unpredictable defects in the as-printed parts. Here we report the elimination of large spatters through controlling laser-powder bed interaction instabilities by using nanoparticles. The elimination of large spatters results in 3D printing of defect lean sample with good consistency and enhanced properties. We reveal that two mechanisms work synergistically to eliminate all types of large spatters: (1) nanoparticle-enabled control of molten pool fluctuation eliminates the liquid breakup induced large spatters; (2) nanoparticle-enabled control of the liquid droplet coalescence eliminates liquid droplet colliding induced large spatters. The nanoparticle-enabled simultaneous stabilization of molten pool fluctuation and prevention of liquid droplet coalescence discovered here provide a potential way to achieve defect lean metal additive manufacturing.

 

Multiagent coordination is highly desirable with many uses in a variety of tasks. In nature, the phenomenon of coordinated flocking is highly common with applications related to defending or escaping from predators. In this article, a hybrid multiagent system that integrates consensus, cooperative learning, and flocking control to determine the direction of attacking predators and learns to flock away from them in a coordinated manner is proposed. This system is entirely distributed requiring only communication between neighboring agents. The fusion of consensus and collaborative reinforcement learning allows agents to cooperatively learn in a variety of multiagent coordination tasks, but this article focuses on flocking away from attacking predators. The results of the flocking show that the agents are able to effectively flock to a target without collision with each other or obstacles. Multiple reinforcement learning methods are evaluated for the task with cooperative learning utilizing function approximation for state-space reduction performing the best. The results of the proposed consensus algorithm show that it provides quick and accurate transmission of information between agents in the flock. Simulations are conducted to show and validate the proposed hybrid system in both one and two predator environments, resulting in an efficient cooperative learning behavior. In the future, the system of using consensus to determine the state and reinforcement learning to learn the states can be applied to additional multiagent tasks. 

Great Salt Lake (GSL), Utah, lost 1.89 ± 0.04 m of water during the 2012–2016 drought. During this timeframe, data from the Gravity Recovery and Climate Experiment mission underestimate this mass loss, while nearby Global Positioning System (GPS) stations exhibit significant shifts in position. We find that crustal deformation, from unloading the Earth's crust consistent with the observed GSL water loss alone, does not explain the GPS displacements, suggesting contributions from additional water storage loss surrounding GSL. This study applies a damped least squares inversion to the three‐dimensional GPS displacements to test a range of distributions of groundwater loads to fit the observations. When considering the horizontal and vertical displacements simultaneously, we find a realistic distribution of water loss while also resolving the observed water loss of the lake. Our preferred model identifies mass loss up to 64 km from the lake via two radial rings. The contribution of exterior groundwater loss is substantial (10.9 ± 2.8 km³vs. 5.5 ± 1.0 km³on the lake), and greatly improves the fit to the observations. Nearby groundwater wells exhibit significant water loss during the drought, which substantiates the presence of significant water loss outside of the lake, but also highlights greater spatial variation than our model can resolve. We observe seismicity modulation within the inferred load region, while the region outside the (un)loading reveals no significant modulation. Drier periods exhibit higher quantities of events than wetter periods and changes in trend of the earthquake rate are correlated with regional mass trends.