.png){kind=icon}
.png){kind=icon}
Return
Neural network based data-driven differential equation modeling and state-fault estimation for fuzzy systems
.png){kind=icon}
PRE
AI
Y
S
DOI：10.1016/j.fss.2026.110033.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
For a class of Takagi-Sugeno (T-S) fuzzy systems with input disturbances and faults, an unknown input differential equation model via discrete sampled output data embedded neural networks is constructed to realize the system state and fault estimation. In the entire implementation process, different from the existing linear interpolation and cubic spline interpolation methods in the literature, the neural network approach is first introduced to process the discrete data sampled from the system output, so as to derive a fitting function that can effectively approximate the continuous output signal of the original T-S fuzzy system. Then, based on the obtained fitting function and the original system matrices, an unknown input differential equation model is established to achieve the system state and fault estimation with a certain accuracy. LMI-based conditions are given to guarantee the stability of the obtained estimation error dynamics. Two numerical models test the proposed neural network based data-driven differential equation modeling and fault estimation method.
Journal
IF:
2.7
Papers:
7.6K
Citations:
1.5W
Organization
No organization information available
Cited Papers