.Modelica_LinearSystems2.UsersGuide.ReleaseNotes.Version_2_1

Information

This release is backward compatible to the previous version 2.0. It contains the following improvements:

New functions/blocks:

• Function sylvester to solve Sylvester equations "A*X + X*B = C" using a Schur method.
• Function nullspace to calculate the orthogonal nullspace and the dimension (nullity) of the nullspace of a matrix.
• Functions conditionNumber and rcond to calculate the condition number and the reciprocal condition number of a matrix respectively.
• Block Controller.Interpolator to increase the sampling frequency with linear interpolation and optional mean-value filtering.
• Blocks have been added to Controller.Examples to demonstrate the discretization methods (Discretization1, Discretization2, DiscretizationSeries) and the new Interpolator block (Interpolator).

Other improvements:

• The instance name is now displayed above the icon of every Controller block.
• The sample factor is now displayed in the icon of every Controller block.
• Access to resources (e.g. image files) has been changed from file names to URIs.
• The sequence of pole shifting in poleAssignmentMI has been changed.
• Bode diagrams of state space systems are now based on numerical more reliable zeros and poles representation instead of on transfer functions.

Bug fixes:

• Redundant results for initial responses of multi input state space systems have been fixed.
• Several functions have been modified to work (or to terminate correctly) for state space system without in- and/or outputs:
  • Analysis.analysis
  • Analysis.invariantZeros.
  • Check for controllability and observability.
  • Staircase algorithm to reduce a system to controller Hessenberg form.
  • Computation of controllable/uncontrollable poles.
• The design of FIR filters in block Controller.FilterFIR was disabled. It is now enabled and the previously wrong design of highpass filters with odd order was corrected, by increasing the order by one in this case.