Appeal No. 1997-1111 Page 18
Application No. 08/105,899
(brief, page 17) that Hägglund does not suggest using a signal
generator with a fuzzy logic controller, and that (reply
brief, page 5) the fact that both Ueda and Hägglund mention
the Ziegler-Nichols method does not suggest the claimed steps
of appellant’s method.
We find that Hägglund discloses the use of a non-linear
circuit to provide a controlled induced oscillation for tuning
a PID. However, in Hägglund (col. 3, lines 5-11) the system
is brought into self oscillation by disconnecting the
integrating and derivative units I and D of the regulator, and
increasing the amplification of the proportional control unit
P, by manually moving adjusting means 9. Hägglund further
discloses (col. 4, lines 39-47) completely disconnecting the
PID controller. In addition, Hägglund discloses (col. 4,
lines 64 and 65) that “[a]lso the I- and D-units can be
connected individually or in combination-also with the P-
unit.” We are in general agreement with the examiner that PID
controllers and fuzzy controllers are both well known in the
art and that the Ziegler-Nichols method of tuning is used in
both PID and fuzzy controllers. However, we find no
suggestion to combine the teachings of Hägglund with Ueda for
Page: Previous 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Next
Last modified: November 3, 2007