Event-Triggered Fault Detection of Nonlinear Networked Systems

Area of Science:

• Control Systems Engineering
• Networked Systems Analysis
• Fuzzy Logic Applications

Background:

• Nonlinear discrete-time networked systems pose challenges for fault detection.
• Event-triggered schemes aim to optimize data transmission by reducing signal updates.
• Existing fault detection methods may not fully address the complexities of networked systems.

Purpose of the Study:

• To design a fault detection filter for nonlinear discrete-time networked systems using an event-triggered scheme.
• To ensure asymptotic stability and desired performance of the residual system.
• To develop a novel polynomial event-triggered mechanism for efficient signal transmission.

Main Methods:

• Design of a polynomial fuzzy fault detection filter.
• Development of a novel polynomial event-triggered scheme.
• Utilization of polynomial approximated membership functions via Taylor series expansion.
• Formulation of sufficient conditions using sum of squares (SOSs).

Main Results:

• A fault detection filter capable of generating a residual signal for fault identification.
• An effective polynomial event-triggered scheme for signal transmission.
• Guaranteed asymptotic stability and performance of the residual system.
• Conditions solvable using SOS tools in MATLAB.

Conclusions:

• The proposed polynomial fuzzy fault detection filter and event-triggered scheme are effective for nonlinear discrete-time networked systems.
• The methodology ensures system stability and performance while optimizing data transmission.
• The use of SOSs provides a computationally tractable approach for filter design and analysis.