moorepants

GUID,Title,Abstract,status,"file url","Download count",URL,"Preprint DOI","Publisher DOI","Date created","Date modified"
mkzp5,"A stochastic multiscale peridynamic model for corrosion-induced fracture","Concrete fracture caused by corrosion of the reinforcing bars plays a key role in accelerating rebar corrosion and causing subsequent structure failure. To better predict this process, we introduce a three-phase stochastic peridynamic model, with the simplest constitutive relation (linear elastic with brittle failure), that avoids the need for explicit concrete microstructure geometry representations. The model links information from the composition at the microscale (phase volume fractions) to the macroscale fracture behavior, while costing the same as a fully homogenized model. We show that a similar peridynamic homogenized model fails in capturing the correct fracture modes/patterns in these problems where the microstructure controls failure behavior. The multiscale model is used to study fracture in reinfo

moorepants

name: filter
channels:
  - conda-forge
  - defaults
dependencies:
  - numpy
  - scipy
  - matplotlib
  - sympy
  - dynamicisttoolkit

moorepants

# channel priority strict (pull from defaults, then look to conda forge, then
# look to twh)
# defaults
# conda-forge
# twh
# these are all the requested packages (conda names)
anaconda
alpaca_kernel_2
anaconda
asciitree

moorepants

function [x, P] = kalman_filter(A, B, C, Q, R, x, u, y, P)
% kalman_filter - Returns the next state estimate and its covariance matrix
% using the Kalman Filter predict & update. Note that this implementation is
% not numerically stable, use production code for non-textbook problems.
%
% Syntax: [x, P] = kalman_filter(A, B, C, Q, R, x, u, y, P)
%
% Inputs:
%   A - State matrix, size 5 x 5
%   B - Input matrix, size 5 x 1

moorepants

import numpy as np
import matplotlib.pyplot as plt
import matplotlib.patches as pat
import sklearn.cluster as clu

# set the seed if you want the same random numbers on eac execution
#np.random.seed(5)

# generate data for random overlapping rectangles
num_rectangles = 50

moorepants

# This is a simple point mass "jumping" simulation. Given a point mass of a
# realistic human mass that starts some height above the ground with a force
# that acts vertically between the ground and mass that represents the force
# that legs can generate, we simulate from standstill to maximum height in the
# air and calculate the work done by all forces on the mass as well as the
# potential and kinetic energy at all times in the simulation. There is also a
# Coulomb friction term that can represent some dissapative resistance to the
# motion.
#
#   O   |g

moorepants

Testing results.

Timing the functions.

Timing: cython
cython time: 0.00288254904747 s

Timing: numpy_broadcast
numpy_broadcast time: 0.00597401690483 s

moorepants

moorepants

moorepants

Data Management Plan

This assignment will utilize two types of data:

1. Existing static and motion data collected from bicycle and bicyclists during prior research. For example, time series from rate gyros or motion capture systems, geometry of vehicles, and anthropomorphic data of the riders.
2. Bicycle and rider motion data generated from simulations.