First I want to tell you that I am very grateful for your comments.
You comments about the forward referencing are appropriate, and
we have already done some changes to reduce them, and we will
look a bit more into it.
The idea has been to give the main ideas in chapter 3 such that
it can be read almost independently. I still think is a good
idea, but we must change the style of writing some
places so that the students do not feel they need to read
the later chapters in order to follow chapter 3.
The only thing I diagree on is the use of e. In other areas than control
an error is always the actual value minus the desired, e = y-r.
I have also found by checking control books that lately this
convention has become more common in control, at least in graduate
books (Morari and Zafiriou, Boyd and Barratt, Mu control toolbox
manual). Also, in the one-degree feedback control system
the input to the controller is ym-r = y + n - r, which is not
the error, but the measured one.
I think we have a comment about this in the book already.
I would be happy to receive additional comments from you regarding
inappropriate forward referencing.
Best regards,
Sigurd Skogestad
~