TY - GEN
T1 - Nonlinear functions activated noise-tolerant zeroing neural network for solving time-varying system of linear equations
AU - Lu, Huiyan
AU - Jin, Long
AU - Liu, Mei
AU - Hu, Bin
AU - Li, Kene
AU - Xiao, Lin
AU - Yi, Chenfu
N1 - Publisher Copyright:
© 2018 Technical Committee on Control Theory, Chinese Association of Automation.
PY - 2018/10/5
Y1 - 2018/10/5
N2 - Solving the problem of linear equations is extensively applied in the domains of science and technology, e.g., medicine, economy and so on. Usually, many practical problems in scientific and engineering areas can be converted into a system of linear equations depicted by the formula Mx = b and solved by the corresponding computing methods. In this paper, a nonlinear functions activated noise-tolerant zeroing neural network (NFNTZNN) is presented and exploited for dealing with the time-varying system of linear equations. Differing from the original gradient neural network (GNN) and existing noise-tolerant zeroing neural network (NTZNN), the proposed NFNTZNN model is activated by specially-constructed nonlinear activation functions, and therefore, has the better convergence speed. Simulative results are conducted to verify the efficiency and advantage of the NFNTZNN model for handling the time-varying system of linear equations.
AB - Solving the problem of linear equations is extensively applied in the domains of science and technology, e.g., medicine, economy and so on. Usually, many practical problems in scientific and engineering areas can be converted into a system of linear equations depicted by the formula Mx = b and solved by the corresponding computing methods. In this paper, a nonlinear functions activated noise-tolerant zeroing neural network (NFNTZNN) is presented and exploited for dealing with the time-varying system of linear equations. Differing from the original gradient neural network (GNN) and existing noise-tolerant zeroing neural network (NTZNN), the proposed NFNTZNN model is activated by specially-constructed nonlinear activation functions, and therefore, has the better convergence speed. Simulative results are conducted to verify the efficiency and advantage of the NFNTZNN model for handling the time-varying system of linear equations.
KW - Noise-tolerant zeroing neural network (NTZNN)
KW - Simulative results
KW - Time-varying system of linear equations
UR - https://www.scopus.com/pages/publications/85056094190
U2 - 10.23919/ChiCC.2018.8483009
DO - 10.23919/ChiCC.2018.8483009
M3 - Conference contribution
AN - SCOPUS:85056094190
T3 - Chinese Control Conference, CCC
SP - 1271
EP - 1276
BT - Proceedings of the 37th Chinese Control Conference, CCC 2018
A2 - Chen, Xin
A2 - Zhao, Qianchuan
PB - IEEE Computer Society
T2 - 37th Chinese Control Conference, CCC 2018
Y2 - 25 July 2018 through 27 July 2018
ER -