7.42 load_coupling_response_matrix

• type: setup command.
• function: read a previously-saved coupling-correction response matrix (produced by compute_coupling_response_matrix) and install it for use by a subsequent correct_coupling command, avoiding the cost of rebuilding the matrix.
• sequence: must follow run_setup.
• Command syntax, including use of equations and subcommands, is discussed in 7.2.

• filename — The (incomplete) name of the SDDS file written by a prior compute_coupling_response_matrix command. The file carries the parameter being adjusted (ElementParameter), the perturbation that was used to build it, and the full list of BPM and skew elements (by name and occurrence); these are matched against the current beamline to install the matrix. An error is reported if any expected element is not present in the current beamline. The file’s CorrectionElements and CorrectionItems parameters (the family-list strings used when the matrix was built) are also loaded into the active correction state, so a subsequent correct_coupling may omit correction_elements and items and inherit them from the matrix file. When those parameters are restated on correct_coupling they must agree with the loaded values or the run aborts with an explicit diagnostic.
• cross_filename — Optional. The (incomplete) name of the second SDDS file produced by a prior compute_coupling_response_matrix run with cross_filename set. When supplied, the cross-plane response matrix is also installed: the H and V probe-corrector inventories (with their items and occurrences), the cross-plane BPM inventories, and the response coefficients are restored, and the cross_h_steering / cross_v_steering / probe-kick magnitude from the file become the defaults inherited by a subsequent correct_coupling. The user still has to opt in by setting cross_response_weight > 0 on correct_coupling; the loaded data alone does not auto-enable the channel. Default NULL – no cross-plane channel is installed, preserving the original η_y-only behavior.
• verbosity — Higher values print more diagnostic information.

If either compute_coupling_response_matrix or load_coupling_response_matrix is given before correct_coupling, the precomputed matrix is used and the knob/BPM/item selection on correct_coupling is not required. When the user does restate correction_elements or items on correct_coupling, the values must match those carried by the matrix file or the run aborts. In all cases lower_limits and upper_limits may be supplied freely on correct_coupling to clamp the inherited knobs; their token count must agree with the inherited family count. If neither command provides a matrix, correct_coupling builds it itself according to its own use_perturbed_matrix control.