MATLAB does not do that.
What is happening is that when you use plot() on a finite set of data, then where there are places that the data suddenly shifts between being very large and very small (or very small and very large) then plot joins the two points. plot() is not adding-on extra lines: plot() is obeying your instructions to join the data that you passed in.
If you have discontinuities in the data you passed in, then plot() is always going to try to join the edges of the discontinuities.
The cure, to not have the edges joined, is to not pass plot() data that has discontinuities in it.
You have a couple of options:
- detect the jumps yourself and plot() in segments
- detect the jumps yourself and insert infinite or nan values at the discontinuity. plot() stops drawing at values that are not finite.
It is not possible for plot() to determine where discontinuities are and automatically handle them itself. plot() just draws what it is told to draw -- it can't tell a valid sudden steep area from discontinuities.
If you use fplot() then fplot() attempts to determine where discontinuities are, and optionally draws a dashed discontinuity line (defaults to drawing such lines.) This has performance implications.
, possibly accompanied by a sketch, would help us guide you on how to use specific MATLAB functions for plotting the graph.
