Abstract: A design scheme for a Duffing Chaotic circuit module is proposed based on a Multisim study. The operational amplifier is used to construct an integrating circuit, which forms the basis of the Duffing chaotic circuit. Results are visualized via phase trajectories on a virtual oscilloscope. The bifurcation results of the Duffing oscillator solved by the circuit module are consistent with those obtained by Matlab. The Duffing equation in a critical state is selected as a Chaos Inspection System (CIS) to identify weak signals buried in strong noise. Simulation examples demonstrate that the designed chaotic circuit successfully identifies various weak signals, including sine, triangle, and square waves, providing a theoretical foundation for hardware-based CIS in non-destructive testing.
Introduction
In fields such as Non-destructive Testing (NDT), detecting early-stage structural defects is critical for safety. However, the signals generated by tiny cracks or flaws are often incredibly weak and completely submerged in background noise. Traditional linear detection methods struggle when the Signal-to-Noise Ratio (SNR) is very low.
While the Duffing oscillator—a non-linear system known for its chaotic behavior—has proven to be an excellent tool for "sniffing out" these weak signals due to its extreme sensitivity to initial conditions and noise immunity, most research relies on software like Matlab. Software solutions are often slow and require a computer interface, limiting their use in portable or high-speed industrial hardware. This paper addresses the need for a dedicated circuit design that can perform this detection at the analog level.
Methodology: Circuit Design and Simulation
The core of the approach involves converting the second-order Duffing equation into a set of first-order differential equations suitable for circuit integration. The researchers designed a Chaos Inspection System (CIS) using a circuit-level simulation in Multisim.
The circuit construction utilizes high-precision components, including the 3554 AM operational amplifier for integration and the AD734BQ multiplier to handle the non-linear $x^3$ term of the Duffing equation. The system is tuned to a "critical state"—a tipping point where it is just about to transition from chaos to a periodic (ordered) state. Instead of standard time-domain waves, a virtual oscilloscope monitors "phase trajectories."
Key Results and Insights
The most significant finding is that the circuit accurately mimics the "phase transition" seen in mathematical models. Unlike linear filters that try to block noise, the Duffing oscillator uses noise immunity; the chaotic system ignores random noise but reacts violently to a periodic signal that matches its internal driving frequency.
Key findings include:
- Consistency: Bifurcation results in the Multisim circuit were consistent with numerical solutions obtained via Matlab.
- Critical Threshold: The researchers identified a critical driving amplitude ($F_c = 0.6747$ in the model) as the optimal threshold for detection.
- Successful Detection: The system successfully identified a 70 kHz weak signal even when completely submerged in Gaussian noise (SNR of 100).
- Waveform Agnosticism: While many systems are tuned only for sine waves, this circuit successfully detected triangle and square waves, proving the Duffing CIS is versatile for different industrial characteristic signals.
Conclusion
This work demonstrates that a hardware-equivalent circuit for a Duffing oscillator can reliably detect weak signals buried in heavy noise. By transitioning from software-based simulations to a Multisim circuit design, the team paved the way for high-efficiency, real-time hardware sensors. Future work will involve moving from simulation to physical PCB hardware and developing sweeping frequency architectures to detect signals of unknown frequencies.
Summary updated