Skip to main content
Log in

In-process surface quality monitoring of the slender workpiece machining with digital twin approach

  • Published:
Journal of Intelligent Manufacturing Aims and scope Submit manuscript

Abstract

In-process monitoring of production quality plays a significant role in intelligent manufacturing. Both part deformation and vibration happen simultaneously in machining processes. They are two prominent issues that can affect the surface quality of machined parts, especially those with low rigidity. The purpose of this study is to explore a hybrid modeling solution to simultaneously monitor the diametrical errors and early chatter vibrations when turning a slender workpiece. A generic analytical model of the slender workpiece turning is formulated based on dynamics of machining. It is proved that the system complies with the principle of superposition. Accordingly, the explicit expressions free from cutting force modeling that is widely used in literature are derived for characterizing the finished surface quality with analytical and finite element methods, respectively. A data-driven model is also developed using the wavelet packet transform to the displacement signals. The independent decompositions of the displacement signals are then correlated with both the dimensional error model and the turning chatter model. Interconnecting the dynamics-based model and the data-driven model contributes to a digital-twin prototype, which allows for in-process detection of the geometrical distortion and the onset of chatter on the part surface. Finally, two different machining cases were performed to verify the proposed methodology. The results show that the developed model consisting of the formulated deflection correlation and chatter indicator is capable of simultaneously evaluating and detecting the dimensional error prediction and the early-onset chatter. By comparison with the analytical modeling, the high fidelity digital twin using the finite element modeling could exhibit higher prediction accuracy. The proposed monitoring strategy could provide a pragmatic approach to online quality control for intelligent machining of flexible workpieces on the shop floor.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15
Fig. 16

Similar content being viewed by others

Data availability

Data sets generated during the current study are available from the corresponding author upon reasonable request.

Abbreviations

\(A_{j} (q),{\kern 1pt} {\kern 1pt} {\kern 1pt} D_{j} (q)\) :

Scaling and wavelet coefficients at the jth decomposition level

\(d\) :

Depth of cut (mm)

\(D,D_{f}\) :

Initial and finished diameters of workpiece (mm)

\(E\) :

Young’s modulus (MPa)

\(E_{j,i}\) :

Wavelet packet energy in the specified frequency range

\(E_{rms}\) :

Root mean square of energy in the specified frequency range

\(\overline{E}\) :

Expected value of energy in the specified frequency range

\(f_{n}\) :

Nyquist frequency (Hz)

\(f_{0}\) :

Feed rate (mm/rev)

\(f(t)\) :

Uncut chip thickness (mm)

\(F(t)\) :

Excitation force (N)

\(F_{c} (t)\) :

Radial cutting force (N)

\(F_{0} (t)\) :

Centrifugal force of the rotating workpiece (N)

\(g\) :

High-pass coefficient of the wavelet function

\(h\) :

Low-pass coefficient of the scaled function

\(k_{c} (z_{c} )\) :

Workpiece stiffness at the cutting point (N/m)

\(k_{m} (z_{m} )\) :

Workpiece cross stiffness at the measuring point (N/m)

\(k_{\lim }\) :

Limiting structural stiffness of workpiece (N/m)

\(K_{E}\) :

Wavelet packet energy kurtosis

\(L\) :

Length of workpiece (mm)

\(m,c,k\) :

Equivalent mass, damping, and stiffness of the SDOF system

\(M\) :

Total number of samples of the signal for chatter detection

\(n\) :

Filter length

\(N\) :

Spindle rotation speed (rpm)

\(N_{0}\) :

Total number of the original signal

\(q\) :

Number of wavelet coefficients with \(q = 1,2,...,N_{0} \cdot 2^{ - j}\)

\(X,Y,Z\) :

Coordinate axes

\(z_{c}\) :

Tool and workpiece contact location (mm)

\(z_{m}\) :

Displacement sensor installation location (mm)

\(\delta (z_{c} )\) :

Workpiece deflection at the cutting point (μm)

\(\delta (z_{m} )\) :

Workpiece deflection at the measuring point (μm)

\(\zeta\) :

Damping ratio of the SDOF system

\(\rho\) :

Material density of the workpiece (kg/m3)

\(\omega\) :

Angular velocity of the spindle (rad/s)

\(\omega_{n}\) :

Natural frequency of the SDOF system (rad/s)

References

Download references

Acknowledgements

The financial support of National Natural Science Foundation of China (Grant Nos. 52175108, 51805352) is gratefully acknowledged. Dr. A. Longstaff would like to acknowledge the UK’s Engineering and Physical Sciences Research Council (EPSRC) funding of the Future Metrology Hub (Grant Ref: EP/P006930/1).

Funding

Supported by National Natural Science Foundation of China (Grant Nos. 52175108, 51805352), and UK’s Engineering and Physical Sciences Research Council (Grant Ref: EP/P006930/1).

Author information

Authors and Affiliations

Authors

Contributions

KL was in charge of funding acquisition, project administration, methodology, and original draft preparation; ZL conducted modeling, experiments and signal processing; AL carried out funding acquisition, review and editing. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Kaibo Lu.

Ethics declarations

Competing interests

The authors declare no conflicts of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix A

Appendix A

For turning slender workpieces, the governing equation of motion of the SDOF system is expressed as:

$$ \begin{aligned} &m\ddot{x}(t) + c\dot{x}(t) + kx(t)\\ &\quad = K_{r} df_{0} + K_{r} d\left[ {x(t - \tau ) - x(t)} \right] + F_{0} \sin \omega t \end{aligned}$$
(A.1)

Assume that \(x_{1} (t)\) is a particular solution to the equation

$$ m\ddot{x}_{1} (t) + c\dot{x}_{1} (t) + kx_{1} (t) = K_{r} df_{0} $$
(A.2)

and \(x_{2} (t)\) is a particular solution to the equation

$$ m\ddot{x}_{2} (t) + c\dot{x}_{2} (t) + kx_{2} (t) = K_{r} d\left[ {x_{2} (t - \tau ) - x_{2} (t)} \right] $$
(A.3)

and \(x_{3} (t)\) is a particular solution to the equation

$$ m\ddot{x}_{3} (t) + c\dot{x}_{3} (t) + kx_{3} (t) = F_{0} \sin \omega t $$
(A.4)

Let \(x(t)\) be equal to the sum of the above responses \(x_{i} (t){\kern 1pt}\)(i = 1,2,3), so that

$$ x(t){\kern 1pt} = x_{1} (t){\kern 1pt} + x_{2} (t){\kern 1pt} + x_{3} (t){\kern 1pt} $$
(A.5)

Suppose \(x(t){\kern 1pt}\) is a solution to the differential equation (A.1), we substitute it into the left-hand side of Eq. (A.1) and obtain,

$$ \begin{aligned} & m[\ddot{x}_{1} (t) + \ddot{x}_{2} (t) + \ddot{x}_{3} (t)] + c[\dot{x}_{1} (t) + \dot{x}_{2} (t) + \dot{x}_{3} (t)]\\ &\qquad + k[x_{1} (t) + x_{2} (t) + x_{3} (t)] \hfill \\ &\quad = [m\ddot{x}_{1} (t) + c\dot{x}_{1} (t) + kx_{1} (t)]\\ &\qquad + [m\ddot{x}_{2} (t) + c\dot{x}_{2} (t) + kx_{2} (t)] \\ &\qquad+ [m\ddot{x}_{3} (t) + c\dot{x}_{3} (t) + kx_{3} (t)] \hfill \\ \end{aligned} $$
(A.6)

Applying Eqs. (A.2), (A.3), and (A.4) to Eq. (A.6) yields

$$\begin{aligned} & m\ddot{x}(t) + c\dot{x}(t) + kx(t)\\ &\qquad = K_{r} df_{0} + K_{r} d\left[ {x_{2} (t - \tau ) - x_{2} (t)} \right] + F_{0} \sin \omega t \end{aligned}$$
(A.7)

On the other hand, substituting Eq. (A.5) into the right-hand side of Eq. (A.1), we have

$$ \begin{aligned} & K_{r} df_{0} + K_{r} d\left[ {x(t - \tau ) - x(t)} \right] + F_{0} \sin \omega t\\ &\quad = K_{r} df_{0} + K_{r} d\{ \left[ {x_{1} (t - \tau ) - x_{1} (t)} \right] \\ &\qquad+ \left[ {x_{2} (t - \tau ) - x_{2} (t)} \right] \\ &\qquad + \left[ {x_{3} (t - \tau ) - x_{3} (t)} \right]\} + F_{0} \sin \omega t \hfill \\ \end{aligned} $$
(A.8)

Generally speaking, the differential equation does not satisfy the superposition principle due to the existence of the delayed term in it. But in this given case, according to fundamentals of mechanical vibrations the responses of \(x_{1} (t){\kern 1pt}\) and \(x_{3} (t){\kern 1pt}\) have the forms with the initial conditions equal to zero, respectively

$$ x_{1} (t) = {{K_{r} df_{0} } \mathord{\left/ {\vphantom {{K_{r} df_{0} } k}} \right. \kern-0pt} k} $$
(A.9)
$$ x_{3} (t) = A\sin (\omega t - \varphi ) $$
(A.10)

where \(A = {{F_{0} } \mathord{\left/ {\vphantom {{F_{0} } {[(k - m\omega^{2} )^{2} + c^{2} \omega^{2} ]}}} \right. \kern-0pt} {[(k - m\omega^{2} )^{2} + c^{2} \omega^{2} ]}}^{0.5}\) and \(\varphi = \tan^{ - 1} [{{c\omega } \mathord{\left/ {\vphantom {{c\omega } {(k - }}} \right. \kern-0pt} {(k - }}m\omega^{2} )]\) denote the amplitude and phase angle of the response, respectively.

Clearly, the response \(x_{1} (t){\kern 1pt}\) is a constant. Thus, we can obtain

$$ x_{1} (t - \tau ) - x_{1} (t) = 0 $$
(A.11)

Meanwhile, since the time delay is equal to the period of the spindle rotation, that is, \(\tau = \frac{60}{N} = \frac{2\pi }{\omega }\), we apply Eq.(A.10) and also derive

$$ x_{3} (t - \tau ) - x_{3} (t) = 0 $$
(A.12)

Finally, inserting Eqs. (A.11) and (A.12) into Eq. (A.8) produces the same expression as Eq. (A.7). As a result, it is proved that the combination of the response \(x_{i} (t){\kern 1pt}\) under each individual excitation is a solution to Eq. (A.1) and thereby the given system complies with the principle of superposition.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Lu, K., Li, Z. & Longstaff, A. In-process surface quality monitoring of the slender workpiece machining with digital twin approach. J Intell Manuf (2024). https://doi.org/10.1007/s10845-024-02353-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s10845-024-02353-y

Keywords

Navigation