Dynamic Characteristics Analysis of Helical Gear System Under Stochastic Disturbance and Friction

This study investigates the nonlinear dynamic behavior of an 8-DOF single-stage helical gear transmission system considering stochastic disturbances and tooth surface friction. A comprehensive mathematical model is established to analyze bifurcation characteristics and chaotic responses under varying operational conditions.

1. Mathematical Modeling

The dynamic model considers bending-torsion-axial coupling effects with the following governing equations:

$$m_e\ddot{x}_n + c_m\dot{x}_n + k_hf(x_n) = F_0 + F_{ah}(t) – m_e\ddot{e}(t) + \eta\mu F_f(s_1+s_2)$$
$$f(x_n) =
\begin{cases}
x_n – b & x_n > b \\
0 & |x_n| \leq b \\
x_n + b & x_n < -b
\end{cases}$$

The dimensionless equations are derived as:

$$
\begin{cases}
\dot{X}_1 = X_2 \\
\dot{X}_2 = -2\xi_{11}X_2 – k_{11}X_1 – \mu\eta(2\xi_{12}X_{14} + k_{12}X_{13}) \\
\vdots \\
\dot{X}_{14} = \ddot{X}_4 – \ddot{X}_{10} + \text{Stochastic terms} + \text{Friction terms}
\end{cases}
$$

2. Friction Analysis

The time-varying friction force is calculated using:

$$F_f = \mu F_m \frac{L_{right} – L_{left}}{L}$$

Key parameters for helical gear analysis:

Parameter Pinion Gear
Number of teeth 28 70
Module (mm) 4 4
Helix angle (°) 18 18
Contact ratio 2.15

3. Numerical Simulation

The bifurcation characteristics under different conditions are analyzed using Runge-Kutta method:

$$X_{n+1} = X_n + \frac{h}{6}(k_1 + 2k_2 + 2k_3 + k_4)$$

Key findings from numerical simulations:

Condition Bifurcation Pattern Critical Frequency
No friction Period-2 → Chaos ω = 1.069
μ = 0.05 Period-1 → Period-4 ω = 0.879
Stochastic disturbance Early chaos ω = 0.943

4. Stochastic Effects Analysis

The system response under random excitation follows:

$$\ddot{x} + 2\xi\omega_n\dot{x} + \omega_n^2x = \eta(t)$$

Where the stochastic disturbance η(t) satisfies:

$$E[\eta(t)] = 0, \quad E[\eta(t)\eta(t+\tau)] = D\delta(\tau)$$

5. Dynamic Response Characteristics

The Poincaré maps reveal three distinct motion states:

$$
\begin{aligned}
&\text{Periodic: } \Sigma_{points} = 1 \\
&\text{Quasi-periodic: } \Sigma_{points} = \infty \\
&\text{Chaotic: } \Sigma_{points} = \text{Fractal structure}
\end{aligned}
$$

The time-domain response demonstrates:

$$RMS = \sqrt{\frac{1}{T}\int_0^T x^2(t)dt} \propto \mu^{0.31}\omega^{1.07}$$

6. Conclusion

Key conclusions for helical gear system dynamics:

  • Chaotic thresholds decrease by 12.7% with stochastic excitation
  • Friction coefficient increases vibration amplitude by 18-23%
  • Critical mesh frequency ranges: 850-1250 Hz

The proposed model provides theoretical guidance for optimizing helical gear transmission systems under complex operational conditions.

Scroll to Top