Version 0.2.5

• ModiaMath initialization improved:

  • KINSOL tolerance FTOL takes IDA tolerance into acount.
  • More information about the model is provided in case KINSOL fails.

• Interface to Modia improved:
  New keyword option hev (step size used for implicit Euler during initialization and at events) added to "simulate".

• ModiaMath result handling improved:

  • The result may contain single values (such as parameter J=0.1).
    When using plot(..) on such a result variable, a constant line is plotted
    between the first and the last point of the x-axis.
  • New function ModiaMath.resultTable(result) to transform the variable information
    in the result data structure in a DataFrames table
    (containing variable name + type + size ) that can then be printed.
  • New functions ModiaMath.closeFigure(figure) and
    ModiaMath.closeAllFigures() to close a specific figure or close all figures.
  • Changing some default options of PyPlot. In particular, (a) the labels on the x- and y-axis
    use exponential notation (e.g. 1e5), if the numbers are larger as 1e4 or smaller as 1e-3
    (PyPlot default is 1e7 and 1e-7), (b) smaller fonts and linewidth are used.
    Default options are changed via PyCall. If PyCall is not in the Julia environment, PyCall is added.
  • If a vector or matrix of subplots is defined, then the x-axis labels
    are only displayed for the last subplot row.
  • Result data is always converted to Float64 before passing it to PyPlot.
    Therefore, Julia numbers can be plotted, even if the types of the numbers are not supported by PyPlot
    (for example rational numbers can be plotted).