鍩轰簬浜哄伐绁炵粡缃戠粶鐨勪簩闃剁郴缁熻鲸璇?/p>
鎽樿锛?/p>
BP
绁炵粡缃戠粶鏄宸弽鍚戜紶鎾缁忕綉缁滅殑绠�绉帮紝鎻愪緵浜嗕竴涓鐞嗛潪绾?/p>
鎬ч棶棰樼殑妯″瀷銆?/p>
鏈枃閽堝甯︽湁鍣０
(
)
v
k
鐨勪簩闃剁郴缁燂紝
鎻愬嚭浜嗘敼杩涚殑
BP
绁炵粡缃戠粶
瀵逛簩闃剁郴缁熺殑杈ㄨ瘑鏂规硶锛?/p>
浠ヨ揪鍒板绯荤粺鐨勭簿纭鲸璇嗭紱
閫氳繃浠跨湡瀹為獙鏁版嵁鍙緱锛?/p>
绁?/p>
缁忕綉缁滅殑杈撳嚭涓庤杈ㄨ瘑绯荤粺杈撳嚭涔嬮棿鐨勮宸緢灏忥紙褰?/p>
k>=8
鏃讹紝
error<0.1%
锛?/p>
锛涢
鍏堜粙缁嶄簡浜哄伐绁炵粡缃戠粶鐨勭郴缁熻鲸璇嗘柟闈㈢殑鍙戝睍涓庣爺绌剁幇鐘讹紝鐒跺悗浠嬬粛甯歌
BP
绠?/p>
娉曞拰鏀硅繘鐨?/p>
BP
绠楁硶锛屾渶鍚庨�氳繃涓�涓叿浣撶殑浜岄樁绯荤粺鐨勫疄渚嬪厖鍒嗚瘉鏄庝簡鏀硅繘
BP
绁炵粡缃戠粶鍏锋湁鐨勮壇濂借鲸璇嗘晥鏋滐紝瀹炵敤鎬у己銆?/p>
鍏抽敭瀛楋細
BP
绁炵粡缃戠粶锛涚郴缁熻鲸璇嗭紱浜岄樁闈炵嚎鎬х郴缁?/p>
Second-order system identification based on artificial neural
networks
WeiLu
(College of Electrical and Control Engineering, Xi鈥檃n University of Science and
Technology
锛?/p>
Xi鈥檃n 710054,China)
Abstract
锛?/p>
BP neural
network is
the abbreviation of erroneous reverse transmission
neural
network,
which
provides
a
model
of
dealing
with
nonlinear
problems.In
this
paper, the second-order system with noise, and puts forward the improved BP neural
network
to
second
order
system
modeling
method.
In
order
to
achieve
an
accurate
identification of the system.Through the simulation experiment the error between the
output
of
neural
network
and
the
output
of
identification
system
is
very
small(The
error<0.1%
when
k>=8).
First,
introduced
the
artificial
neural
network
system
identification aspects of development and research,Then, introduced the conventional
BP algorithm and improved BP algorithm,Finally, Through an example of a specific
second-order
system
fully
proved
that
the
improved
BP
neural
network
has
good
recognition results and practical.
Key words
锛?/p>
BP neural network
锛?/p>
System Identification
锛?/p>
Second-order nonlinear system
涓�
缁
鍦ㄨ嚜鐒剁瀛﹀拰绀句細绉戝鐨勫悇涓鍩熶腑锛岃秺鏉ヨ秺澶氶渶瑕佽鲸璇嗙郴缁熸ā鍨嬬殑闂
宸插箍娉涘紩璧蜂汉浠殑閲嶈锛屽緢澶氬鑰呭湪鐮旂┒鏈夊叧绾挎�у拰闈炵嚎鎬х殑杈ㄨ瘑闂銆?/p>