/**
* Returns the state response matrix calculated from the front face of elemFrom to the back face
* of elemTo. This is a convenience wrapper to the real method in the trajectory class
*
* <p>This method was moved here from EnvelopeTrajectory/EnvelopeProbe where it was eliminated
* since <code>Trajectory</code> was genericized.
*
* @param strIdElemFrom String identifying starting lattice element
* @param strIdElemTo String identifying ending lattice element
* @return response matrix from elemFrom to elemTo
* @see EnvelopeTrajectory#computeTransferMatrix(String, String)
*/
public PhaseMatrix computeTransferMatrix(String strIdElemFrom, String strIdElemTo) {
Trajectory<TransferMapState> trajectory = this.getTrajectory();
// find starting index
int[] arrIndFrom = trajectory.indicesForElement(strIdElemFrom);
int[] arrIndTo = trajectory.indicesForElement(strIdElemTo);
if (arrIndFrom.length == 0 || arrIndTo.length == 0)
throw new IllegalArgumentException("unknown element id");
int indFrom, indTo;
indTo = arrIndTo[arrIndTo.length - 1]; // use last state before start element
TransferMapState stateTo = trajectory.stateWithIndex(indTo);
PhaseMatrix matTo = stateTo.getTransferMap().getFirstOrder();
indFrom = arrIndFrom[0] - 1;
if (indFrom < 0) return matTo; // response from beginning of machine
TransferMapState stateFrom = trajectory.stateWithIndex(indFrom);
PhaseMatrix matFrom = stateFrom.getTransferMap().getFirstOrder();
return matTo.times(matFrom.inverse());
}
/**
* Convenience method for computing the transfer map between two state locations, say
* <i>S</i><sub>1</sub> and <i>S</i><sub>2</sub>. Let <i>s</i><sub>0</sub> be the axis location of
* the beamline entrance, <i>s</i><sub>1</sub> the location of state <i>S</i><sub>1</sub>, and
* <i>s</i><sub>2</sub> the location of state <i>S</i><sub>2</sub>. Each state object
* <i>S<sub>n</sub></i> contains the transfer map
* <b>T</b>(<i>s<sub>n</sub></i>,<i>s</i><sub>0</sub>) which takes phases coordinates at the
* beamline entrance to the position of state <i>S<sub>n</sub></i>. The transfer map
* <b>T</b>(<i>s</i><sub>2</sub>,<i>s</i><sub>1</sub>) taking phase coordinates
* <b>z</b><sub>1</sub> (and covariance matrix <b>σ</b><sub>1</sub>) from position
* <i>s</i><sub>1</sub> to position <i>s</i><sub>2</sub> is then given by <br>
* <br>
* <b>T</b>(<i>s</i><sub>2</sub>,<i>s</i><sub>1</sub>) =
* <b>T</b>(<i>s</i><sub>2</sub>,<i>s</i><sub>0</sub>) ∘
* <b>T</b>(<i>s</i><sub>1</sub>,<i>s</i><sub>0</sub>)<sup>-1</sup> , <br>
* <br>
* where <b>T</b>(<i>s</i><sub>2</sub>,<i>s</i><sub>0</sub>) is the transfer map between the
* beamline entrance <i>s</i><sub>0</sub> and the position <i>s</i><sub>2</sub> of state
* <i>S</i><sub>2</sub>, and <b>T</b>(<i>s</i><sub>1</sub>,<i>s</i><sub>0</sub>) is the transfer
* map between the beamline entrance <i>s</i><sub>0</sub> and the position <i>s</i><sub>1</sub> of
* state <i>S</i><sub>1</sub>.
*
* @param state1 trajectory state <i>S</i><sub>1</sub> of starting location <i>s</i><sub>1</sub>
* @param state2 trajectory state <i>S</i><sub>2</sub> of final location <i>s</i><sub>2</sub>
* @return transfer map <b>T</b>(<i>s</i><sub>2</sub>,<i>s</i><sub>1</sub>) between locations
* <i>s</i><sub>1</sub> and <i>s</i><sub>2</sub>
* @author Christopher K. Allen
* @since Nov 4, 2014
*/
public static PhaseMap computeTransferMap(TransferMapState state1, TransferMapState state2) {
PhaseMap mapPhi1 = state1.getTransferMap();
PhaseMap mapPhi2 = state2.getTransferMap();
PhaseMap mapPhi1inv = mapPhi1.inverse();
PhaseMap mapPhi21 = mapPhi2.compose(mapPhi1inv);
return mapPhi21;
}
/**
* Convenience method for computing the transfer matrix between two state locations, say
* <i>S</i><sub>1</sub> and <i>S</i><sub>2</sub>. Let <i>s</i><sub>0</sub> be the axis location of
* the beamline entrance, <i>s</i><sub>1</sub> the location of state <i>S</i><sub>1</sub>, and
* <i>s</i><sub>2</sub> the location of state <i>S</i><sub>2</sub>. Each state object
* <i>S<sub>n</sub></i> contains the transfer matrix
* <b>Φ</b>(<i>s<sub>n</sub></i>,<i>s</i><sub>0</sub>) which takes phases coordinates at the
* beamline entrance to the position of state <i>S<sub>n</sub></i>. The transfer matrix
* <b>Φ</b>(<i>s</i><sub>2</sub>,<i>s</i><sub>1</sub>) taking phase coordinates
* <b>z</b><sub>1</sub> (and covariance matrix <b>σ</b><sub>1</sub>) from position
* <i>s</i><sub>1</sub> to position <i>s</i><sub>2</sub> is then given by <br>
* <br>
* <b>Φ</b>(<i>s</i><sub>2</sub>,<i>s</i><sub>1</sub>) =
* <b>Φ</b>(<i>s</i><sub>2</sub>,<i>s</i><sub>0</sub>)
* <b>Φ</b>(<i>s</i><sub>1</sub>,<i>s</i><sub>0</sub>)<sup>-1</sup> , <br>
* <br>
* where <b>Φ</b>(<i>s</i><sub>2</sub>,<i>s</i><sub>0</sub>) is the transfer matrix between
* the beamline entrance <i>s</i><sub>0</sub> and the position <i>s</i><sub>2</sub> of state
* <i>S</i><sub>2</sub>, and <b>Φ</b>(<i>s</i><sub>1</sub>,<i>s</i><sub>0</sub>) is the
* transfer matrix between the beamline entrance <i>s</i><sub>0</sub> and the position
* <i>s</i><sub>1</sub> of state <i>S</i><sub>1</sub>.
*
* @param state1 trajectory state <i>S</i><sub>1</sub> of starting location <i>s</i><sub>1</sub>
* @param state2 trajectory state <i>S</i><sub>2</sub> of final location <i>s</i><sub>2</sub>
* @return transfer matrix <b>Φ</b>(<i>s</i><sub>2</sub>,<i>s</i><sub>1</sub>) between
* locations <i>s</i><sub>1</sub> and <i>s</i><sub>2</sub>
* @author Christopher K. Allen
* @since Jun 23, 2014
*/
public static PhaseMatrix computeTransferMatrix(
TransferMapState state1, TransferMapState state2) {
PhaseMatrix matPhi1 = state1.getTransferMap().getFirstOrder();
PhaseMatrix matPhi2 = state2.getTransferMap().getFirstOrder();
PhaseMatrix matPhi1inv = matPhi1.inverse();
PhaseMatrix matPhi21 = matPhi2.times(matPhi1inv);
return matPhi21;
}
/**
* This is the phase advance for the given state location.
*
* <p>Compute and return the particle phase advance from the trajectory beginning to the given
* state location.
*
* <p>Internally the method calculates the phase advances using the initial and final
* Courant-Snyder α and β values for the matched beam at the trajectory beginning and
* the location of the given state, respectively. These Courant-Snyder parameters are computed as
* the matched beam envelope at the trajectory beginning and the given state location.
*
* <p>Let <b>Φ</b> represent the transfer matrix from the initial ring location to the given
* state location. The computed quantity is the general phase advance of the particle through the
* transfer matrix <b>Φ</b>, and no special requirements are placed upon <b>Φ</b> (e.g.,
* periodicity). One phase advance is provided for each phase plane, i.e.,
* (σ<sub><i>x</i></sub>, σ<sub><i>z</i></sub>, σ<sub><i>z</i></sub>).
*
* <p>The definition of phase advance σ is given by <br>
* <br>
* σ(<i>s</i>) ≡ ∫<sup><i>s</i></sup> [1/β(<i>t</i>)]<i>dt</i>
* , <br>
* <br>
* where β(<i>s</i>) is the Courant-Snyder, envelope function, and the integral is taken
* along the interval between the initial and final Courant-Snyder parameters.
*
* <p>The basic relation used to compute σ is the following: <br>
* <br>
* σ = sin<sup>-1</sup>
* φ<sub>12</sub>/(β<sub>1</sub>β<sub>2</sub>)<sup>½</sup> , <br>
* <br>
* where φ<sub>12</sub> is the element of <b>Φ</b> in the upper right corner of each
* 2×2 diagonal block, β<sub>1</sub> is the initial beta function value (provided) and
* β<sub>2</sub> is the final beta function value (provided).
*
* @param state state containing transfer map and location used in these calculations
* @see calculatePhaseAdvance(PhaseMatrix, Twiss[], Twiss[])
* @author Christopher K. Allen
* @since Aug 14, 2013
*/
@Override
public R3 computeBetatronPhase(TransferMapState state) {
PhaseMatrix matFullTrn = this.calculateFullLatticeMatrixAt(state);
Twiss[] arrTwsLoc = super.calculateMatchedTwiss(matFullTrn);
PhaseMatrix matPhiLoc = state.getTransferMap().getFirstOrder();
R3 vecPhsAdv = super.calculatePhaseAdvance(matPhiLoc, this.arrTwsMch, arrTwsLoc);
return vecPhsAdv;
}
/**
* Calculates the fixed point (closed orbit) in transverse phase space at the given state location
* in the presence of dispersion.
*
* <p>Let the full-turn map a the state location be denoted <b>Φ</b>. The transverse plane
* dispersion vector <b>Δ</b> is defined <br>
* <br>
* <b>Δ</b><sub><i>t</i></sub> ≡
* -(1/γ<sup>2</sup>)[d<i>x</i>/d<i>z'</i>, d<i>x'</i>/d<i>z'</i>, d<i>y</i>/d<i>z'</i>,
* d<i>y'</i>/d<i>z'</i>]<sup><i>T</i></sup> . <br>
* <br>
* It can be identified as the first 4 entries of the 6<sup><i>th</i></sup> column in the transfer
* matrix <b>Φ</b>. The above vector quantifies the change in the transverse particle phase
* coordinate position versus the change in particle momentum. The factor -(1/γ<sup>2</sup>)
* is needed to convert from longitudinal divergence angle <i>z'</i> used by XAL to momentum
* δ<i>p</i> ≡ Δ<i>p</i>/<i>p</i> used in the dispersion definition.
* Specifically, <br>
* <br>
* δ<i>p</i> ≡ Δ<i>p</i>/<i>p</i> = γ<sup>2</sup><i>z</i>'
* <br>
* <br>
* As such, the above vector can be better described <br>
* <br>
* <b>Δ</b><sub><i>t</i></sub> ≡ [Δ<i>x</i>/δ<i>p</i>,
* Δ<i>x'</i>/δ<i>p</i>, Δ<i>y</i>/δ<i>p</i>,
* Δ<i>y'</i>/δ<i>p</i>]<sup><i>T</i></sup> <br>
* <br>
* explicitly describing the change in transverse phase coordinate for fractional change in
* momentum δ<i>p</i>.
*
* <p>Since we are only concerned with transverse phase space coordinates, we restrict ourselves
* to the 4×4 upper diagonal block of <b>Φ</b>, which we denote take <b>T</b>. That is,
* <b>T</b> = π ⋅ <b>Φ</b> where π : <b>R</b><sup>6×6</sup> →
* <b>R</b><sup>4×4</sup> is the projection operator.
*
* <p>This method finds that point <b>z</b><sub><i>t</i></sub> ≡ (<i>x<sub>t</sub></i>,
* <i>x'<sub>t</sub></i>, <i>y<sub>t</sub></i>, <i>y'<sub>t</sub></i>) in transvse phase space
* that is invariant under the action of the ring for a given momentum spread δ<i>p</i>.
* That is, the particle ends up in the same location each revolution. With a finite momentum
* spread of δ<i>p</i> > 0 we require this require that <br>
* <br>
* <b>Tz</b><sub><i>t</i></sub> + δ<i>p</i><b>Δ</b><sub><i>t</i></sub> =
* <b>z</b><sub><i>t</i></sub> , <br>
* <br>
* which can be written <br>
* <br>
* <b>z</b><sub><i>t</i></sub> = δ<i>p</i>(<b>T</b> -
* <b>I</b>)<sup>-1</sup><b>Δ</b><sub><i>t</i></sub> , <br>
* <br>
* where <b>I</b> is the identity matrix. Dividing both sides by δ<i>p</i> yields the final
* result <br>
* <br>
* <b>z</b><sub>0</sub> ≡ <b>z</b><sub><i>t</i></sub>/δ<i>p</i> = (<b>T</b> -
* <b>I</b>)<sup>-1</sup><b>Δ</b><sub><i>t</i></sub> , <br>
* <br>
* which is the returned value of this method. It is normalized by δ<i>p</i> so that we can
* compute the closed orbit for any given momentum spread.
*
* @param state we are calculating the dispersion at this state location
* @return The closed orbit fixed point <b>z</b><sub>0</sub> for finite dispersion, normalized by
* momentum spread. Returned as an array
* [<i>x</i><sub>0</sub>,<i>x'</i><sub>0</sub>,<i>y</i><sub>0</sub>,<i>y'</i><sub>0</sub>]/δ<i>p</i>
* @see xal.tools.beam.calc.ISimEnvResults#computeChromDispersion(xal.model.probe.traj.ProbeState)
* @author Christopher K. Allen
* @since Nov 8, 2013
*/
@Override
public PhaseVector computeChromDispersion(TransferMapState state) {
PhaseMatrix matFullTn = this.calculateFullLatticeMatrixAt(state);
double dblGamma = state.getGamma();
// double[] arrDisp = super.calculateDispersion(matFullTn, dblGamma);
// R4 vecDispR4 = new R4(arrDisp);
R4 vecDispR4 = super.calculateDispersion(matFullTn, dblGamma);
PhaseVector vecDisp = PhaseVector.embed(vecDispR4);
return vecDisp;
}
/**
* Computes the chromatic aberration for one pass around the ring starting at the given state
* location, or from the entrance to state position for a linear machine. The returned vector is
* the displacement from the closed orbit caused by a unit momentum offset (δ<i>p</i> = 1).
* See the documentation in {@link ISimLocResults#computeChromAberration(ProbeState)} for a more
* detailed exposition.
*
* @see xal.tools.beam.calc.ISimLocResults#computeChromAberration(xal.model.probe.traj.ProbeState)
* @author Christopher K. Allen
* @since Nov 15, 2013
*/
@Override
public PhaseVector computeChromAberration(TransferMapState state) {
double dblGamma = state.getGamma();
// PhaseMap mapPhi = state.getTransferMap();
// PhaseMap mapPhi = state.getStateTransferMap();
// PhaseMatrix matPhi = mapPhi.getFirstOrder();
PhaseMatrix matPhi = this.calculateFullLatticeMatrixAt(state);
R6 vecDel = super.calculateAberration(matPhi, dblGamma);
return PhaseVector.embed(vecDel);
}
/**
* Calculates and returns the full lattice matrix for the machine at the given state location. Let
* <i>S<sub>n</sub></i> be the given state object at location <i>s<sub>n</sub></i>, and let
* <b>T</b><sub><i>n</i></sub> be the transfer matrix between locations <i>s</i><sub>0</sub> and
* <i>s<sub>n</sub></i> , where <i>s</i><sub>0</sub> is the location of the full transfer matrix
* <b>Φ</b><sub>0</sub> for this machine (end to end). Then the full turn matrix
* <b>Φ</b><sub><i>n</i></sub> for the machine at location <i>s<sub>n</sub></i> is given by
* <br>
* <br>
* <b>Φ</b><sub><i>n</i></sub> = <b>T</b><sub><i>n</i></sub> ⋅
* <b>Φ</b><sub>0</sub> ⋅ <b>T</b><sub><i>n</i></sub><sup>-1</sup> . <br>
* <br>
* That is, we conjugate the full transfer map for this machine by the transfer map for the given
* state.
*
* @param state state object <i>S<sub>n</sub></i> for location <i>s<sub>n</sub></i> containing
* transfer matrix <b>T</b><sub><i>n</i></sub>
* @return the full-turn map <b>Φ</b><sub><i>n</i></sub> at the location <i>s<sub>n</sub></i>
* of the given state
* @author Christopher K. Allen
* @since Oct 28, 2013
*/
protected PhaseMatrix calculateFullLatticeMatrixAt(TransferMapState state) {
PhaseMap mapPhiState = state.getTransferMap();
PhaseMatrix matPhiState = mapPhiState.getFirstOrder();
PhaseMatrix matPhiStInv = matPhiState.inverse();
PhaseMatrix matPhiFull = this.mapPhiFull.getFirstOrder();
PhaseMatrix matFullTnLoc;
matFullTnLoc = matPhiFull.times(matPhiStInv);
matFullTnLoc = matPhiState.times(matFullTnLoc);
return matFullTnLoc;
}
