skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.


Title: Machine Learning-Enhanced Model Predictive Control for Incremental Bending of Skeletal Fixation Plates
Abstract Skeletal fixation plates are essential components in craniomaxillofacial (CMF) reconstructive surgery to connect skeletal disunions. To ensure that these plates achieve geometric conformity to the CMF skeleton of individual patients, a pre-operative procedure involving manual plate bending is traditionally required. However, manual adjustment of the fixation plate can be time-consuming and is prone to geometric error due to the springback effect and human inspection limitations. This work represents a first step towards autonomous incremental plate bending for CMF reconstructive surgery through machine learning-enabled springback prediction and feedback bending control. Specifically, a Gaussian process is first investigated to complement the physics-based Gardiner equation to improve the accuracy of springback effect estimation, which is then incorporated into nonlinear model predictive controller to determine the optimal sequence of bending inputs to achieve geometric conformity. Evaluation using a simulated environment for bending confirms the effectiveness of the developed approach.  more » « less
Award ID(s):
2133630
PAR ID:
10581267
Author(s) / Creator(s):
; ; ; ; ; ; ; ;
Publisher / Repository:
American Society of Mechanical Engineers
Date Published:
ISBN:
978-0-7918-8788-2
Format(s):
Medium: X
Location:
Seattle, Washington, USA
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract There are many sources of manufacturing variations in sheet metal assemblies, such as automotive bodies. These include non-isotropic material properties from cold rolling, springback in stamping, and distortion from residual stresses when components are clamped and spot welded. FE simulations have been used to predict these variations in order to better design tooling and processes. Such simulations require expertise in complex, multi-stage nonlinear analysis. We are investigating the feasibility of training machine learning algorithms in order to democratize these types of analyses. This requires the curation of large, validated, and balanced data sets. To this end, we have developed a multi-stage finite element simulation workflow encompassing component stamping and joining with a focus on examining deformations due to springback in two-part assemblies. Three connected simulations comprise the workflow: (1) component stamping with capture of springback, (2) assembly clamping, and (3) assembly joining, then release. The workflow utilizes explicit dynamic finite element analysis (FEA) and includes the transfer of intermediate solutions (geometries/stresses), as well as extraction of key geometric parameters of springback from both component- and assembly-level simulations. The NUMISHEET 1993 U-draw/bending benchmark was referenced for its tooling geometry and utilized for verification of the forming process simulation; variations of material and geometry were also simulated. In summary, this work provides a means of generating a design space of flexible two-part assemblies for applications such as dataset generation, design optimization, and machine learning. 
    more » « less
  2. Abstract BackgroundLongitudinal skeletal growth takes place in the cartilaginous growth plates. While growth plates are found at either end of conventional long bones, they occur at a variety of locations in the mammalian skeleton. For example, the metacarpals and metatarsals (MT) in the hands and feet form only a single growth plate at one end, and the pisiform in the wrist is the only carpal bone to contain a growth plate. We take advantage of this natural anatomical variation to test which components of the PTHrP/Ihh feedback loop, a fundamental regulator of chondrocyte differentiation, are specific to growth plate function. ResultsParathyroid hormone‐like hormone(Pthlh), the gene that transcribes parathyroid hormone‐related peptide (PTHrP), is expressed in the reserve zone of the growth plate‐forming end of the MT. At the opposite end, the absence of a PTHrP+ reserve zone results in premature chondrocyte differentiation andIndian hedgehog(Ihh) expression.Pthlhis expressed in the reserve zone of the developing pisiform, confirming the existence of a true growth plate. ConclusionA pool of PTHrP+ reserve zone chondrocytes is a defining characteristic of growth plates, and its patterning may be key to evolved differences in growth plate location in the mammalian skeleton. 
    more » « less
  3. Modeling springback in sheet materials is challenging in aluminum alloys, especially when a complex strain path is applied. This paper presents results from pure bending experiments on AA 6016-T4 sheet material, where various plastic pre-strains were first applied prior to bending. A crystal plasticity based elasto-plastic selfconsistent (EPSC) model that includes the effect of backstress in the hardening law was used to predict final part shape after unloading. The backstress term in the model was calibrated using geometrically necessary dislocation (GND) content, measured experimentally by high resolution electron backscattered diffraction (HREBSD). The EPSC model predicted springforward angles for unstrained 1 mm AA 6016-T4 sheet with an error of 0.4% (0.3◦) in the worst case, while the J2 plasticity isotropic model overpredicted springforward angles by as much as 2.4% (2◦). For cases where uniaxial, plane-strain, and biaxial pre-strains were first imparted to the sheets before bending, the EPSC model with backstress accurately predicted the transition from springforward to springback, while the EPSC model without backstress did not. Backstress influence on model accuracy, which increased with greater pre-strain levels, appears to be correlated to the statistically stored dislocation (SSD) density computed by the model at the end of each pre-strain step. 
    more » « less
  4. null (Ed.)
    In this work we investigate the effects of two distinct actuation methods on the hydrodynamics of elastic rectangular plates oscillating at resonance. Plates are driven by plunging motion at the root or actuated by a distributed internal bending moment at Reynolds numbers between 500 and 4000. The latter actuation method represents internally actuated smart materials and emulates the natural ability of swimming animals to continuously change their shapes with muscles. We conduct experiments with plunging elastic plates and piezoelectric plate actuators that are simulated using a fully coupled three-dimensional computational model based on the lattice Boltzmann method. After experimental validation the computational model is employed to probe plate hydrodynamics for a wide range of parameters, including large oscillation amplitudes which prompts nonlinear effects. The comparison between the two actuation methods reveals that, for the same level of tip deflection, externally actuated plates significantly outperform internally actuated plates in terms of thrust production and hydrodynamic efficiency. The reduced performance of internally actuated plates is associated with their suboptimal bending shapes which leads to a trailing edge geometry with enhanced vorticity generation and viscous dissipation. Furthermore, the difference in actuation methods impacts the inertia coefficient characterizing the plate oscillations, especially for large amplitudes. It is found that the inertia coefficient strongly depends on the tip deflection amplitude and the Reynolds number, and actuation method, especially for larger amplitudes. 
    more » « less
  5. Abstract Oceanic plates experience extensive normal faulting as they bend and subduct, enabling fracturing of the incoming lithosphere. Debate remains about the relative importance of pre‐existing faults, plate curvature and other factors controlling the extent and style of bending‐related faulting. The subduction zone off the Alaska Peninsula is an ideal place to investigate controls on bending faulting as the orientation of the abyssal‐hill fabric with respect to the trench and plate curvature vary along the margin. Here, we characterize faulting between longitudes 161°W and 155°W using newly collected multibeam bathymetry data. We also use a compilation of seismic reflection data to constrain patterns of sediment thickness on the incoming plate. Although sediment thickness increases over 1 km from 156°W to 160°W, most sediments were deposited prior to the onset of bending faulting and thus should have limited impact on the expression of bend‐related fault strikes and throws in bathymetry data. Where magnetic anomalies trend subparallel to the trench (<30°) west of ∼156°W, bending faults parallel magnetic anomalies, implying that bending faults reactivate pre‐existing structures. Where magnetic anomalies are highly oblique (>30°) to the trench east of 156°W, no bending faults are observed. Summed fault throws increase to the west, including where pre‐existing structure orientations are constant (between 157 and 161°W), suggesting that another factor such as the increase in slab curvature must influence bending faulting. However, the westward increase in summed fault throws is more abrupt than expected for gradual changes in slab bending alone, suggesting potential feedbacks between pre‐existing structures, slab dip, and faulting. 
    more » « less