Calibration_HBAN.Rd

% Generated by roxygen2 (4.0.1): do not edit by hand
\encoding{UTF-8}
\name{Calibration_HBAN}
\alias{Calibration_HBAN}
\title{Calibration algorithm which minimises the error criterion using the Irstea-HBAN procedure}
\usage{
Calibration_HBAN(InputsModel, RunOptions, InputsCrit, CalibOptions, FUN_MOD,
  FUN_CRIT, FUN_TRANSFO = NULL, quiet = FALSE)
}
\arguments{
\item{InputsModel}{[object of class \emph{InputsModel}] see \code{\link{CreateInputsModel}} for details}

\item{RunOptions}{[object of class \emph{RunOptions}] see \code{\link{CreateRunOptions}} for details}

\item{InputsCrit}{[object of class \emph{InputsCrit}] see \code{\link{CreateInputsCrit}} for details}

\item{CalibOptions}{[object of class \emph{CalibOptions}] see \code{\link{CreateCalibOptions}} for details}

\item{FUN_MOD}{[function] hydrological model function (e.g. RunModel_GR4J, RunModel_CemaNeigeGR4J)}

\item{FUN_CRIT}{[function] error criterion function (e.g. ErrorCrit_RMSE, ErrorCrit_NSE)}

\item{FUN_TRANSFO}{(optional) [function] model parameters transformation function, if the FUN_MOD used is native in the package FUN_TRANSFO is automatically defined}

\item{quiet}{(optional) [boolean] boolean indicating if the function is run in quiet mode or not, default=FALSE}
}
\value{
[list] list containing the function outputs organised as follows:
         \tabular{ll}{
         \emph{$ParamFinalR  }  \tab   [numeric] parameter set obtained at the end of the calibration \cr
         \emph{$CritFinal    }  \tab   [numeric] error criterion obtained at the end of the calibration \cr
         \emph{$NIter        }  \tab   [numeric] number of iterations during the calibration \cr
         \emph{$NRuns        }  \tab   [numeric] number of model runs done during the calibration \cr
         \emph{$HistParamR   }  \tab   [numeric] table showing the progression steps in the search for optimal set: parameter values \cr
         \emph{$HistCrit     }  \tab   [numeric] table showing the progression steps in the search for optimal set: criterion values \cr
         \emph{$MatBoolCrit  }  \tab   [boolean] table giving the requested and actual time steps when the model is calibrated \cr
         \emph{$CritName     }  \tab   [character] name of the calibration criterion \cr
         \emph{$CritBestValue}  \tab   [numeric] theoretical best criterion value \cr
         }
}
\description{
Calibration algorithm which minimises the error criterion. \cr
\cr
The algorithm is based on a local search procedure.
First, a screening is performed using either a rough predefined grid or a list of parameter sets
and then a simple steepest descent local search algorithm is performed.
}
\details{
A screening is first performed either from a rough predefined grid (considering various initial
values for each paramete) or from a list of initial parameter sets. \cr
The best set identified in this screening is then used as a starting point for the steepest
descent local search algorithm. \cr
For this search, the parameters are used in a transformed version, to obtain uniform
variation ranges (and thus a similar pace), while the true ranges might be quite different. \cr
At each iteration, we start from a parameter set of NParam values (NParam being the number of
free parameters of the chosen hydrological model) and we determine the 2*NParam-1 new candidates
by changing one by one the different parameters (+/- pace). \cr
All these candidates are tested and the best one kept to be the starting point for the next
iteration. At the end of each iteration, the pace is either increased or decreased to adapt
the progression speed. A diagonal progress can occasionally be done.                        \cr
The calibration algorithm stops when the pace becomes too small.                            \cr

To optimise the exploration of the parameter space, transformation functions are used to convert
the model parameters. This is done using the TransfoParam functions.
}
\examples{
## load of catchment data
require(airGR)
data(L0123001)