Cheatsheet


Spring analogy for flexible structures:

  • Bar, fixed in one end, deformation in the other end: k=3EIl3k=\frac{3EI}{l^3}
  • Bar, supported by two simple supports, deformation in the center: k=48EIl3k=\frac{48EI}{l^3}
  • Solid axle, deformation by torsion: k=Gπr42lk=G\frac{\pi r^4}{2l}

Equivalent rigidities:

  • Spring in series: k=k1k2k1+k2k=\frac{k_1k_2}{k_1+k_2}
  • Spring in parallel: k=k1+k2k=k_1+k_2

Lagrange’s Equation:

 ⁣d ⁣dt( ⁣dT ⁣du˙) ⁣dT ⁣du+ ⁣dD ⁣du˙+ ⁣dU ⁣du=F\frac{\dd}{\dd t} \left( \frac{\dd T}{\dd \dot u} \right) - \frac{\dd T}{\dd u} + \frac{\dd D}{\dd \dot u} + \frac{\dd U}{\dd u}=F

Energies:

T=12mv2;D=12cv2;U=12mu2T=\frac{1}{2}mv^2\quad;\quad D=\frac{1}{2}cv^2\quad;\quad U=\frac{1}{2}mu^2

Frequencies:

ωn=km;fn=12πkm\omega_n=\sqrt{\frac{k}{m}}\quad;\quad f_n=\frac{1}{2\pi}\sqrt{\frac{k}{m}}

Damping coefficient:

ζ=cccr;ccr=2mωn=2km\zeta=\frac{c}{c_{cr}}\quad;\quad c_{cr}=2m\omega_n=2\sqrt{km}
  • Can be measured by comparing amplitudes
    • Two consecutive amplitudes: ζ12πlnuiui+1\zeta\approx\frac{1}{2\pi}\ln\frac{u_i}{u_{i+1}}
    • NN cycles: ζ12πNln(ukuk+N)\zeta\approx\frac{1}{2\pi N}\ln\left( \frac{u_k}{u_{k+N}} \right)

General solutions:

mu¨+ku=0u(t)=C1cosωnt+C2sinωntmu¨+cu˙+ku=0u(t)=eζωnt(C1cosωdt+C2sinωdt)\begin{aligned} m\ddot u+ku=0 \quad & \rightarrow \quad u(t)=C_1\cos\omega_nt+C_2\sin\omega_nt\\ m\ddot u+c\dot u+ku=0 \quad& \rightarrow \quad u(t)=e^{\zeta\omega_n t}(C_1\cos\omega_d t+C_2\sin\omega_d t) \end{aligned}
  • ωd=ωn1ζ2ωn\omega_d=\omega_n\sqrt{1-\zeta^2}\approx\omega_n: Damped natural frequency

Fourier’s series:

f(t)=a0+n=1(ancosnω0t+bnsinnω0t)f(t)=c0+n=1cncos(nω0tαn)\begin{aligned} f(t)&=a_0+\sum_{n=1}^\infty \left( a_n\cos n\omega_0 t + b_n\sin n\omega_0t \right)\\ f(t)&=c_0+\sum_{n=1}^\infty c_n\cos(n\omega_0 t-\alpha_n) \end{aligned}

Coefficients:

a0=1T00T0x(t) ⁣dt;an=2T00T0x(t)cosnωt ⁣dt;bn=2T00T0x(t)sinnωt ⁣dtcn=an2+bn2;tanαn=bnan\begin{gathered} a_0=\frac{1}{T_0}\int_0^{T_0}x(t)\dd t\quad;\quad a_n=\frac{2}{T_0}\int_0^{T_0}x(t)\cos n\omega t\dd t\quad;\quad b_n=\frac{2}{T_0}\int_0^{T_0} x(t)\sin n\omega t\dd t \\ c_n=\sqrt{a_n^2 + b_n^2} \quad;\quad \tan\alpha_n=\frac{b_n}{a_n} \end{gathered}

Power spectral density:

x2(t)=1T0Tx2(t) ⁣dt=a02+121(an2+bn2)=c02+121cn2\begin{aligned} \langle x^2(t) \rangle&=\frac{1}{T}\int_0^Tx^2(t)\dd t\\ &=a_0^2+\frac{1}{2}\sum_1^\infty\left( a_n^2+b_n^2 \right)\\ &=c_0^2+\frac{1}{2}\sum_1^\infty c_n^2 \end{aligned}

Data collecting parameters:

Δf=fsN=1T;Δt=TN=1fs\Delta f=\frac{f_s}{N}=\frac{1}{T}\quad;\quad \Delta t=\frac{T}{N}=\frac{1}{f_s}
  • Anti-aliasing condition: fsfny=2ff_s\ge f_{ny}=2f
  • Anti-leakage condition: T0/TZT_0/T\in\mathbb{Z}

Modal analysis:

Mu¨+Cu˙+Ku=0\mathbf{M\ddot u} + \mathbf{C\dot u} + \mathbf{Ku} = \mathbf 0
  • To obtain natural frequencies: det(Kωn2M)=0\det\left( \mathbf K -\omega_n^2\mathbf M \right)=0
  • Modal vectors are the eigenvectors of the natural frequencies.
  • Modal matrix is: Φ=[ϕ1,ϕ2,ϕ3]\mathbf\Phi=[\mathbf\phi_1, \mathbf\phi_2,\mathbf\phi_3]
    • Ordered from the lowest resonance frequency to the greatest.
    • If a resonance frequency is repeated, additional modal vectors can be obtained by: ϕiTMϕj=0\mathbf\phi_i^T\mathbf M\mathbf\phi_j=0

Mapping to modal space:

M=ΦTMΦ;η=ΦTu\mathbf{\mathcal M}=\mathbf\Phi^T\mathbf M\mathbf\Phi \quad;\quad \mathbf\eta= \mathbf\Phi^T\mathbf u