Class NordsieckStepInterpolator

  • All Implemented Interfaces:
    java.io.Externalizable, java.io.Serializable, StepInterpolator

    public class NordsieckStepInterpolator
    extends AbstractStepInterpolator
    This class implements an interpolator for integrators using Nordsieck representation.

    This interpolator computes dense output around the current point. The interpolation equation is based on Taylor series formulas.

    Since:
    2.0
    Version:
    $Revision: 1073158 $ $Date: 2011-02-21 22:46:52 +0100 (lun. 21 févr. 2011) $
    See Also:
    AdamsBashforthIntegrator, AdamsMoultonIntegrator, Serialized Form
    • Field Detail

      • stateVariation

        protected double[] stateVariation
        State variation.
    • Constructor Detail

      • NordsieckStepInterpolator

        public NordsieckStepInterpolator()
        Simple constructor. This constructor builds an instance that is not usable yet, the AbstractStepInterpolator.reinitialize(double[], boolean) method should be called before using the instance in order to initialize the internal arrays. This constructor is used only in order to delay the initialization in some cases.
      • NordsieckStepInterpolator

        public NordsieckStepInterpolator​(NordsieckStepInterpolator interpolator)
        Copy constructor.
        Parameters:
        interpolator - interpolator to copy from. The copy is a deep copy: its arrays are separated from the original arrays of the instance
    • Method Detail

      • reinitialize

        public void reinitialize​(double[] y,
                                 boolean forward)
        Reinitialize the instance.

        Beware that all arrays must be references to integrator arrays, in order to ensure proper update without copy.

        Overrides:
        reinitialize in class AbstractStepInterpolator
        Parameters:
        y - reference to the integrator array holding the state at the end of the step
        forward - integration direction indicator
      • reinitialize

        public void reinitialize​(double time,
                                 double stepSize,
                                 double[] scaledDerivative,
                                 Array2DRowRealMatrix nordsieckVector)
        Reinitialize the instance.

        Beware that all arrays must be references to integrator arrays, in order to ensure proper update without copy.

        Parameters:
        time - time at which all arrays are defined
        stepSize - step size used in the scaled and nordsieck arrays
        scaledDerivative - reference to the integrator array holding the first scaled derivative
        nordsieckVector - reference to the integrator matrix holding the nordsieck vector
      • rescale

        public void rescale​(double stepSize)
        Rescale the instance.

        Since the scaled and Nordiseck arrays are shared with the caller, this method has the side effect of rescaling this arrays in the caller too.

        Parameters:
        stepSize - new step size to use in the scaled and nordsieck arrays
      • getInterpolatedStateVariation

        public double[] getInterpolatedStateVariation()
                                               throws DerivativeException
        Get the state vector variation from current to interpolated state.

        This method is aimed at computing y(tinterpolation) -y(tcurrent) accurately by avoiding the cancellation errors that would occur if the subtraction were performed explicitly.

        The returned vector is a reference to a reused array, so it should not be modified and it should be copied if it needs to be preserved across several calls.

        Returns:
        state vector at time AbstractStepInterpolator.getInterpolatedTime()
        Throws:
        DerivativeException - if this call induces an automatic step finalization that throws one
        See Also:
        AbstractStepInterpolator.getInterpolatedDerivatives()
      • computeInterpolatedStateAndDerivatives

        protected void computeInterpolatedStateAndDerivatives​(double theta,
                                                              double oneMinusThetaH)
        Compute the state and derivatives at the interpolated time. This is the main processing method that should be implemented by the derived classes to perform the interpolation.
        Specified by:
        computeInterpolatedStateAndDerivatives in class AbstractStepInterpolator
        Parameters:
        theta - normalized interpolation abscissa within the step (theta is zero at the previous time step and one at the current time step)
        oneMinusThetaH - time gap between the interpolated time and the current time
      • writeExternal

        public void writeExternal​(java.io.ObjectOutput out)
                           throws java.io.IOException
        Specified by:
        writeExternal in interface java.io.Externalizable
        Specified by:
        writeExternal in class AbstractStepInterpolator
        Throws:
        java.io.IOException
      • readExternal

        public void readExternal​(java.io.ObjectInput in)
                          throws java.io.IOException,
                                 java.lang.ClassNotFoundException
        Specified by:
        readExternal in interface java.io.Externalizable
        Specified by:
        readExternal in class AbstractStepInterpolator
        Throws:
        java.io.IOException
        java.lang.ClassNotFoundException