Ease, and those living at high altitudes, where the air is thin, tend
To develop such an increase. This regulation draws its information
From the harmful effect (the lack of oxygen) itself, not from the
Cause (D) of the heart disease, or from the decision to live at a
Higher altitude.
From the point of view of communication, the new phenomena
Are easily related to those of the old. The difference is simply that
Now the information from D to R (which must pass if the regulator
R is to play any useful part whatever) comes through T. Instead of
D → T → E
↑
And the information available for regulatory purposes is whatever
Survives the coding imposed by its passage through T (S.8/5).
Sometimes the information available to R is forced to take an
Even longer route, so that R is affected only by the actual effect at
E. The diagram of immediate effects is then
D → T → E
↑
R
And we have the basic form of the simple “error-controlled servo-
Mechanism” or “closed loop regulator”, with its well-known feed-
Back from E to R. The reader should appreciate that this form differs
From that of the basic formulation (S.11/4) only in that the informa-
Tion about D gets to R by the longer route
D → T → E → R →
Again, the information available to R is only such as survives the
Transmission through T and E:
This form is of the greatest importance and widest applicability.
The remainder of the book will be devoted to it. (The other cases
Are essentially simpler and do not need so much consideration.)
D → T → E
We have
↑↓
R
R
R is thus getting its information about D by way of T:
D → T → R →
222
A fundamental property of the error-controlled regulator is
That it cannot be perfect in the sense of S.11/3.
Suppose we attempt to formulate the error-controlled system by
The method used in S.11/3 and 4. We take a table of double entry,
With D and R determining an outcome in E. Each column has a
Variety equal to that of D. What is new is that the rules must be
Modified. Whereas previously D made a selection (a particular
Disturbance), then R, and thus E was determined, the play now is
|
|
That after D’s initial selection, R must take a value that is a deter-
Minate function of the outcome E (for R is error-controlled). It is
Easily shown that with these conditions E’s variety will be as
Large as D’s— i.e. R can achieve no regulation, no matter how R is
Constructed (i.e. no matter what transformation is used to turn E’s
Value to an R-value).
If the formal proof is not required, a simpler line of reasoning
Can show why this must be so. As we saw, R gets its information
223
A N I N T R O D UC T I O N T O C Y B E R NE T I C S
TH E ERR O R- CO N TR O LLED REG U LA TO R
Through T and E. Suppose R is somehow regulating successfully;
Then this would imply that the variety at E is reduced below that
Of D— perhaps even reduced to zero. This very reduction makes
The channel
D → T → E →
To have a lessened capacity; if E should be held quite constant then
The channel is quite blocked. So the more successful R is in keeping
E constant, the more does R block the channel by which it is receiv-
Ing its necessary information. Clearly, any success by R can at best
Be partial.
Fortunately, in many cases complete regulation is not neces-
Sary. So far, we have rather assumed that the states of the essential
variables E were sharply divided into “normal” ( η ) and “lethal”, so
Occurrence of the “undesirable” states was wholly incompatible
With regulation. It often happens, however, that the systems show
Continuity, so that the states of the essential variables lie along a
Scale of undesirability. Thus a land animal can pass through many
Degrees of dehydration before dying of thirst; and a suitable reversal
From half way along the scale may justly be called “regulatory” if it
|
|
Дата добавления: 2019-11-16; просмотров: 204; Мы поможем в написании вашей работы! |
Мы поможем в написании ваших работ!