cvsolve: cvsolve

In sundialr: An Interface to 'SUNDIALS' Ordinary Differential Equation (ODE) Solvers

Description

CVSOLVE solver to solve stiff ODEs with discontinuties

Usage

cvsolve(
  time_vector,
  IC,
  input_function,
  Parameters,
  Events = NULL,
  reltolerance = 1e-04,
  abstolerance = 1e-04
)

Arguments

time_vector	time vector
IC	Initial Conditions
input_function	Right Hand Side function of ODEs
Parameters	Parameters input to ODEs
Events	Discontinuities in the solution (a DataFrame, default value is NULL)
reltolerance	Relative Tolerance (a scalar, default value = 1e-04)
abstolerance	Absolute Tolerance (a scalar or vector with length equal to ydot, default = 1e-04)

Examples

# Example of solving a set of ODEs with multiple discontinuities using cvsolve
# A simple One dimensional equation, y = -0.1 * y
# ODEs described by an R function
ODE_R <- function(t, y, p){

  # vector containing the right hand side gradients
  ydot = vector(mode = "numeric", length = length(y))

  # R indices start from 1
  ydot[1] = -p[1]*y[1]

  ydot

}

# R code to generate time vector, IC and solve the equations
TSAMP <- seq(from = 0, to = 100, by = 0.1)      # sampling time points
IC <- c(1)
params <- c(0.1)

# A dataset describing the dosing at times at which additions to y[1] are to be done
# Names of the columns don't matter, but they MUST be in the order of state index,
# times and Values at discontinuity.
TDOSE <- data.frame(ID = 1, TIMES = c(0, 10, 20, 30, 40, 50), VAL = 100)
df1 <- cvsolve(TSAMP, c(1), ODE_R, params)         # solving without any discontinuity
df2 <- cvsolve(TSAMP, c(1), ODE_R, params, TDOSE)  # solving with discontinuity

sundialr documentation built on May 16, 2021, 5:06 p.m.