CN103471552A - Carbon fiber reinforced polymer (CFRP) machined surface appearance representation method - Google Patents

Abstract

The invention provides a carbon fiber reinforced polymer (CFRP) machined surface appearance representation method. The method includes the steps that a to-be-represented workpiece is a continuous curve block on a bounded domain, and a surface appearance characteristic function F(x,y) is measured; wavelet component division is carried out on the appearance characteristic function F(x,y) of the three-dimensional surface of the bounded domain, and representation of the surface appearance is further carried out. The method is based on a wavelet assessment reference, machined surface appearance of the CFRP is truly reflected, and meanwhile the machined surface appearance which is precisely represented is obtained by calculating representation parameters. According to a wavelet theory, a multi-fractal spectrum of the machined surface of the CFRP is calculated by means of a wavelet transform method, and multi-fractal parameters are identified to represent characteristics of the machined surface appearance of the CFRP. An appearance representation model is built on the basis of multi-fractal characteristic analysis of the machined surface of the CFRP, the characteristics of the whole machined surface can be described, and local appearance characteristics can be described.

Description

Carbon fibre composite cut surface topography characterizing method

Technical field

The present invention relates to a kind of carbon fibre composite cut surface topography characterizing method.

Background technology

Carbon fibre composite (CFRP, Carbon Fiber Reinforced Polymer) be a kind of high-tech new material of mechanical property excellence, it with light-high-strength, high ratio modulus, low thermal coefficient of expansion and outstanding thermotolerance and the advantage such as wear-resistant in Aero-Space, communications and transportation, sports equipment, civil construction and other industrial field widespread use such as material for characteristic.With the processing mechanism of ordinary metallic material, its material property is anisotropy.And CFRP machining surface defect is more, these defects are not directly cut out by blade entirely, and its cutting surface can produce many special appearances.

Carbon fibre composite finished surface morphology characterization technology still adopts or uses signature criteria and the Evaluation model of the metal material workpiece surface quality of setting up based on the assumption of isotropy at present, even so that adopt FA surface roughness measuring instrument, and utilize the surface quality characterization parameter of setting up for metal material surface of the work characteristics to remove to set up CFRP surface texture access model, drawn conclusion and actual conditions still have larger error.Even with respect to metal material workpiece cutting surface, regular, the difference of CFRP and metal material maximum is that it has anisotropic characteristics, the cutting surface microscopic appearance geometric properties of its workpiece is different from the metal material cutting surface, the cut theoretical roughness of surface of ordinary metallic material can calculate with formula, the machining surface defect of CFRP is a lot, comprise that fiber is lax, fibre breakage, fiber exposes along its length or comes off, fiber and matrix unsticking, layering etc., on finished surface, existing outstanding fiber also has groove and the hole that loses fiber and stay, these paddy peak profiles are not directly cut out by blade, therefore CFRP is after cut, should not adopt the formula of existing common material to calculate its theoretical roughness of surface.Machine direction, processing conditions are the principal elements that determines CFRP finished surface pattern, and even in processing, different pressure and shearing forces also causes different surface topographies, and the machining surface of CFRP is more than plain metal machining surface complexity.

The CFRP workpiece is with its non-easy clamping and be difficult to ensure card cutting high-aspect-ratio, complex-shaped, has determined that its processing has minimum yardstick, in the minute manufacturing group of very high degree of precision, has specific (special) requirements.At present technology carbon fibre composite is used basically or the range parameter of take two-dimentional roughness as main, be aided with hybrid parameter.In view of the CFRP finished surface belongs to the anisotropy type, the two dimension roughness is very low to the evaluating precision on this class surface, and the sign of two-dimensional surface profile roughness is limited for the use of surface information with analysis, be difficult to workpiece surface appearance is estimated comprehensively and effectively, the spatial characteristics of CFRP surface of the work structure, determined only ability accurate Characterization surfaceness in three-dimensional scope, this is also the essence of reflection material surface feature.

Summary of the invention

In order to overcome the deficiencies in the prior art, the invention provides a kind of 3D evaluation and characterization technique that can truly reflect CFRP workpiece cutting surface microscopic appearance, exploitation forms the 3 d surface topography representation system.

A kind of carbon fibre composite cut surface topography characterizing method, comprise the following steps,

Steps A, be made as the continuous curve surface piece on bounded domain by workpiece to be characterized, and effects on surface shape characteristic function F (x, y) is measured;

Step B, carry out the decomposition of small echo composition to the shape characteristic function F (x, y) of bounded domain three-dimensional surface, note F (x, y)=F _i+1, by F _i+1be decomposed into the surface topography function G _iand sign reference field function F _i, and by F _icarry out lower one deck decomposition, obtain F _i+1=G _i+ G _i-1+ G _i-2+ ...+G _m+ F _m, wherein, F _mbe the sign reference field of wavelet decomposition.

Further, described step B meets following formula:

h ₁(x,y)=F(x,y)-h ₂(x,y)

Wherein, h ₁(x, y) is the dimensional discrepancy with respect to the ideal surfaced shape, h ₂(x, y) means the evaluation reference field.

Further, described step B meets following formula:

F _i+1=F _i+G _i

In formula $G_i=\underset{j,k\∈Z}{\mathrm{\Σ}}{r^1}_{i,j,k}{{\mathrm{\θ}}^1}_{i,j,k}(x,y)+{r^2}_{i,j,k}{{\mathrm{\θ}}^2}_{i,j,k}(x,y)+{r^3}_{i,j,k}{{\mathrm{\θ}}^3}_{i,j,k}(x,y)$

Wherein $s_{i,j,k}(x,y)=\underset{p,q\∈Z,}{\mathrm{\Σ}}{\stackrel{\‾}{h}}_{p-2j}{\stackrel{\‾}{h}}_{q-2k}{s}_{i+1,p,q};$

${r^1}_{i,j,k}(x,y)=\underset{p,q\∈Z,}{\mathrm{\Σ}}{\stackrel{\‾}{h}}_{p-2j}{{\stackrel{\‾}{G}}_{q-2k}s}_{i+1,p,q};$

${r^2}_{i,j,k}(x,y)=\underset{p,q\∈Z,}{\mathrm{\Σ}}{\stackrel{\‾}{G}}_{p-2j}{{\stackrel{\‾}{h}}_{q-2k}s}_{i+1,p,q};$

${r^3}_{i,j,k}(x,y)=\underset{p,q\∈Z,}{\mathrm{\Σ}}{\stackrel{\‾}{G}}_{p-2j}{{\stackrel{\‾}{G}}_{q-2k}s}_{i+1,p,q}.$

Further, further comprising the steps in described step B,

The best fit approximation space of F (x, y) is found out in initialization;

Interative computation, decompose F (x, y);

Calculate iteration according to the Chi-square method of the goodness of fit and finally decompose end condition.

Further, the characterization result of checking multifractal spectra, used 3D amplitude roughness parameter S _qwith space pattern parameter S _althe effects on surface pattern is evaluated.

Further, described step C meets following formula:

$S_q=\sqrt{\frac{1}{\mathrm{MN}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{\left[h(x_i,y_i)\right]}^2}$

M in formula, N is sampling number, h (x _i, y _i) be the height of sampled point with respect to the evaluation reference field;

S _albe defined as:

$S_{a1}=\min \left(\sqrt{{{\mathrm{\τ}}_x}^2+{{\mathrm{\τ}}_y}^2}\right),R({\mathrm{\τ}}_x,{\mathrm{\τ}}_y)\≤0.2$

R (τ _x, τ _y) be surperficial two-dimensional autocorrelation function (AACF), τ _x, τ _ybe respectively x, the auto-correlation length on the y direction, x direction and y direction are surface pairwise orthogonal direction arbitrarily.

Further, described measuring surface form adopts the Talysurf5P-120 surface topographic apparatus fo of Rank Taylor Hobson company.

Further, in described step C, adopt two-dimentional multifractal to fall Fluctuation Method and analyze surface topography.

Further, the width △ a that defines fractal spectrum is:

△a=a _max-a _min

The spectral difference △ f of definition multifractal spectra is:

△f=f(a _min)-f(a _max)。

Compared to prior art, the present invention, from wavelet theory, adopts the multifractal spectra of small wave converting method calculating CFRP cutting surface, the feature of definition multifractal parameter characterization CFR cutting surface pattern.Set up the morphology characterization model based on CFRP cutting surface Multifractal Analysis, not only can describe the feature of finished surface integral body, can also describe its local shape characteristic

The accompanying drawing explanation

The Decomposition iteration process schematic diagram that accompanying drawing 1 is carbon fibre composite cut surface topography characterizing method of the present invention.

The algorithm flow chart of accompanying drawing 2 carbon fibre composite cut of the present invention surface topography characterizing method.

Accompanying drawing 3 is CFRP pattern two-dimensional wavelet transformation schematic diagram.

Accompanying drawing 4 is CFRP pattern 3 D wavelet transformation schematic diagram.

Accompanying drawing 5 is CFRP finished surface shape appearance figures.

Accompanying drawing 6 is different a _pthe lower CFRP finished surface pattern multifractal spectrogram of value.

Embodiment

Below in conjunction with accompanying drawing explanation and embodiment, the present invention is further described.

Refer to Fig. 1 to Fig. 6, the invention provides a kind of carbon fibre composite cut surface topography characterizing method, the carbon fibre composite finished surface carries out morphology characterization.The present invention is on the basis of parameter characterization workpiece surface roughness, the research wavelet theory is at the CFRP(carbon fibre composite, Carbon Fiber Reinforced Polymer) application that in the material cutting surface, pattern extracts, set up the mathematical model that wavelet analysis method extracts 3 d surface topography, a kind of CFRP finished surface morphology characterization method based on small echo evaluation benchmark is provided, design can truly reflect the evaluating system of CFRP cutting surface pattern, simultaneously by the computational representation parameter, draw the method for accurate sign cutting surface pattern, carbon fibre composite cut surface topography characterizing method of the present invention, comprise the following steps:

Steps A, be made as the continuous curve surface piece on bounded domain by workpiece to be characterized, and effects on surface shape characteristic function F (x, y) is measured.The continuous curve surface piece that single CFRP workpiece surface appearance can be regarded as by some bounded domains forms, so only need to study the compositions such as roughness how to extract surface topography on every bounded domain, texture degree, track geometry error, just can reach the target of extracting the processing work surface topography.

If F (x, y) is certain the piece zone on surface topography, known, F (x, y) is the closed region Upper bounded function.Prove according to the mathematics relevant knowledge, establish surface topography function F (x, y) bounded and continuous, F on closed region Ω ²(x, y) can amass on Ω.

Simultaneously, establish R ²(M ²) expression M ²the space collection of all quadractically integrable functions, known, the space that the upper all quadractically integrable functions of closed region Ω form is R ²(M ²) subset, and certain the piece zone F (x, y) on CFRP workpiece processing work surface topography is R ²(M ²) in function.

The three-dimensional model of small echo evaluation benchmark, need 2-d wavelet, by the tensor product acquisition of one dimension small echo.

If the quadrature multiresolution analysis { U that monobasic scaling function φ generates _i} _{i ∈ Z}, definition U _iwith it self Tensor Products, be:

${U_i}^2=U_i\⊗U_i=\{s\left(x\right)t\left(y\right){\}}_{s\left(x\right)\∈U_i,t\left(y\right)\∈U_i}$

Due to U _iorthonormal basis be { 2 ^i/2φ (2 ⁱx-j) } _{j ∈ Z}so, U _i ²orthonormal basis is: { 2 ⁱφ (2 ⁱx-j) φ (2 ⁱy-k) } _{j, k ∈ Z}

For binary function f (x, y), note f _ijk(x, y)=2 ⁱh (2 ⁱx-j, 2 ⁱy-k)

Note

for U _i ²orthonormal basis.Therefore

generate a R ²(M ²) quadrature multiresolution analysis { U _i ²} _{i ∈ Z}, establish V _iu _iand U _i+1orthogonal space mend, i.e. U _i+1=U _i⊕ V _i, order again

generate V ₀,

at this moment ${U_{i+1}}^2=U_{i+1}\⊗U_{i+1}={U_i}^2\⊕{V_i}^2.$ If { θ _ijk ¹(x, y), θ _ijk ²(x, y), θ _ijk ²(x, y) } _{i, j, k ∈ Z}form R ²(M ²) orthonormal basis.Thereby known, it is feasible that CFRP workpiece processing wavelet method extracts 3 d surface topography.

Step B, carry out the decomposition of small echo composition to the shape characteristic function F (x, y) of bounded domain three-dimensional surface.

Suppose h ₁(x, y) is the dimensional discrepancy with respect to the ideal surfaced shape, and the ideal surfaced shape is evaluated benchmark exactly, h ₂(x, y) means the evaluation reference field, and it is the summation of other compositions such as morphology, shape error, and the computing formula that surfaceness is extracted is:

h ₁(x,y)=F(x,y)-h ₂(x,y) (1)

When CFRP workpiece machining 3 D surface F (x, y) is carried out to the separation and extraction of small echo composition on different scale, need the 2-d wavelet method, to this, at first need on the i layer, define a two-dimentional scaling function and three small echo translation basis functions.

$\left\{\begin{array}{c}{\mathrm{\φ}}_{i,j,k}(x,y)=2^{i/2}\mathrm{\φ}(2^{i}x-j,2^{i}y-k)\\ {{\mathrm{\θ}}^l}_{i,j,k}(x,y)=2^{i/2}\mathrm{\θ}(2^{i}x-j,2^{i}y-k),l=\mathrm{1,2,3}\end{array}\right.---\left(2\right)$

Here φ (x, y) is two-dimentional scaling function, θ ¹(x, y), θ ²(x, y), θ ³(x, y) is three wavelet functions.

If { U _i ²} _{i ∈ Z}for square integrable space R ²(M ²) a quadrature multiresolution analysis generating, therefore make F (x, y) ∈ U _i+1 ², have:

$F(x,y)=\underset{j,k\∈Z}{\mathrm{\Σ}}{s}_{i+1,j,k}{\mathrm{\φ}}_{i+1,j,k}(x,y)---\left(3\right)$

Note F (x, y)=F _i+1, can draw, at space R ²(M ²) in, the small echo extraction algorithm of CFRP workpiece machining surface three-dimensional roughness reference field is:

F _i+1=F _i+G _i （4）

In formula $G_i=\underset{j,k\∈Z}{\mathrm{\Σ}}{r^1}_{i,j,k}{{\mathrm{\θ}}^1}_{i,j,k}(x,y)+{r^2}_{i,j,k}{{\mathrm{\θ}}^2}_{i,j,k}(x,y)+{r^3}_{i,j,k}{{\mathrm{\θ}}^3}_{i,j,k}(x,y)$

Wherein $s_{i,j,k}(x,y)=\underset{p,q\∈Z,}{\mathrm{\Σ}}{\stackrel{\‾}{h}}_{p-2j}{\stackrel{\‾}{h}}_{q-2k}{s}_{i+1,p,q},$ ${r^1}_{i,j,k}(x,y)=\underset{p,q\∈Z,}{\mathrm{\Σ}}{\stackrel{\‾}{h}}_{p-2j}{{\stackrel{\‾}{G}}_{q-2k}s}_{i+1,p,q},$ ${r^2}_{i,j,k}(x,y)=\underset{p,q\∈Z,}{\mathrm{\Σ}}{\stackrel{\‾}{G}}_{p-2j}{{\stackrel{\‾}{h}}_{q-2k}s}_{i+1,p,q},$ ${r^3}_{i,j,k}(x,y)=\underset{p,q\∈Z,}{\mathrm{\Σ}}{\stackrel{\‾}{G}}_{p-2j}{{\stackrel{\‾}{G}}_{q-2k}s}_{i+1,p,q}$

Here G _ipart belongs to U _i+1 ², but

it represents the detail signal of HFS, and F _irepresent the low frequency approximation signal.

Continuation is to F _icarrying out lower one deck decomposition, is F _i=F _i-1+ G _i-1, with method, can obtain F _i+1=G _i+ G _i-1+ G _i-2+ ...+G _m+ F _m(5)

Surface topography belongs to radio-frequency component, corresponding, characterizes reference field and belongs to low-frequency component.So F _mbe the sign reference field of wavelet decomposition.

Be illustrated in figure 2 algorithm flow chart.The algorithm implementation procedure that small echo extracts CFRP finished surface pattern comprises the following steps:

(1) the best fit approximation space is looked in initialization.Find the best fit approximation space U of F _i+1 ², approximate satisfied (3) formula,

be also F (x, y) ≈ F _i+1, make its maximal degree ground reflect the various information of F (x, y).

(2) interative computation.After looking for the best fit approximation initial space, by F _i+1start, it is decomposed into to low other approximate part F of one-level _iwith pattern part G _i, be formula (4).Then continue to decompose, refer to Fig. 1.In Fig. 1, t is the wavelet decomposition number of times, F _i-t+1for F than the approximate part of low level, F in its correspondence (5) formula _mpart, same, G _i-t+1g in corresponding (5) formula _mpart.

(3) end condition.Calculate t according to the Chi-square method of the goodness of fit, this is also that iteration is finally decomposed end condition.

The characterization result of checking multifractal spectra, used 3D amplitude roughness parameter S _qwith space pattern parameter S _althe effects on surface pattern is evaluated.

Step C: in order to contrast the characterization result of multifractal spectra, use 3D range parameter S _q(root-mean-square-deviation) and space pattern parameter S _al(auto-correlation that decays the soonest length) is characterized pattern.

$S_q=\sqrt{\frac{1}{\mathrm{MN}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{\left[h(x_i,y_i)\right]}^2}---\left(6\right)$

M in formula, N is sampling number, h (x _i, y _i) be the height of sampled point with respect to the evaluation reference field.

S _skfor the symmetric tolerance of surface deviation with respect to reference surface, defined formula is as follows:

S _albe defined as:

$S_{a1}=\min \left(\sqrt{{{\mathrm{\τ}}_x}^2+{{\mathrm{\τ}}_y}^2}\right),R({\mathrm{\τ}}_x,{\mathrm{\τ}}_y)\≤0.2---\left(7\right)$

R (τ _x, τ _y) be surperficial two-dimensional autocorrelation function (AACF), τ _x, τ _ybe respectively x, the auto-correlation length on the y direction, x direction and y direction are surface pairwise orthogonal direction arbitrarily.S in above formula _althat AACF decays at 0.2 o'clock, the shortest auto-correlation length on any possible direction.S _althe three-dimensional characterization parameter of characterization of surfaces constituent, S _alillustrate that greatly surperficial low-frequency component is many, surface is mainly to consist of the long wave composition; S _althe radio-frequency component on the bright surface of novel is many, is mainly to consist of the shortwave composition.

The computing method of multifractal spectra are many, as box counting dimension method (DB), difference box counting dimension method (OBC), two blanket (Peleg_blanket) cladding process, the variational method with based on discrete fraction Blang increment random field (DFBIR) etc., the result that different computing method are calculated differs greatly sometimes, in view of wavelet analysis is the modern fractal instrument of calculating commonly used, the wavelet transformation that this project is taked to introduce above carrys out the scheme of Calculating material surface Multifractal Dimension, and Fig. 3 and Fig. 4 are respectively CFRP pattern two-dimensional wavelet transformation and 3 D wavelet design sketch.

The present invention adopts is that two-dimentional multifractal falls Fluctuation Method (ZD-MF-DFA) and analyzes surface topography, and ZD-MF-DFA is the popularization that the Pulsation analysis method falls in multifractal.

Pulsatile function F always falls in the q rank of calculating under different curve yardstick s _q(s), between them, pass is:

F _q(s)～s ^h(q) (8)

Wherein scaling exponent h (q) is called as generalized Hurst index, if research object be one fractal, h (q) is exactly a constant; If research object is a multifractal, h (q) is the function of a decline.For the q value, be positive situation, h (q) has described the Scaling behavior of the great fluctuation process in curved surface of studying; Otherwise, mean the Scaling behavior of the minor swing in curved surface of studying.

After obtaining h (q), can obtain performance figure τ (q) according to following relation:

τ(q)=qh(q)-D _f (9)

D wherein _fthe fractal dimension of support set, for workpiece three-dimensional surface in the present invention, D _f=3, according to Legendre transformation, can obtain the expression formula of singularity intensity a and multifractal spectra f (a):

$\left\{\begin{array}{c}a\left(q\right)={\mathrm{\τ}}^{\′}\left(q\right)=h\left(q\right)+q{h}^{\′}\left(q\right)\\ f\left(a\right)=\mathrm{qa}\left(q\right)-\mathrm{\τ}\left(q\right)\end{array}\right.---\left(10\right)$

F (a) and a are a set of parameters of describing multifractal, the also number of corresponding probability subset of a at this moment, and the explanation τ (q) proposed by Bacry can be expressed as formula again with the relation between the Legendre of multifractal spectra f (a) variation:

$f\left(a\right)=\underset{q\∈R}{\min }\left(q(a+1/2-\mathrm{\τ}\left(q\right))\right)---\left(11\right)$

Multifractal spectra is to distribute and calculate each regional probability according to actual material surface height, adds up all probability difference, by weighting, processes surface is divided into to the research of many a subsets, and each subset represents a kind of probability distribution, i.e. P (ε) ∝ ε ^a, the width △ a that defines fractal spectrum is:

△a=a _max-a _min

(12)

The width △ a of multifractal spectra has reflected the size of probability distribution scope, and probability distribution is more inhomogeneous, and corresponding f (a) curve is wider, and △ a is larger, otherwise less.Corresponding to material surface, △ a correspondence the fluctuating quantity of finished surface pattern height, reflected machined surface roughness.

The spectral difference △ f of definition multifractal spectra is:

△f=f(a _min)-f(a _max)

(13)

The size of △ f can count there is maximum, the ratio between the number of minimum probability unit, what it reflected is the probability distribution of crest and the trough on materials processing surface.When △ f<0, multifractal spectra is bottom right bending shape, and its corresponding finished surface pattern height is higher, and the shared probability in indication workpiece machining surface peak is large; As △ f > 0 the time, multifractal spectra is lower-left bending shape, and its corresponding finished surface pattern height is lower, and the shared probability of paddy that the indication workpiece machining surface occurs is large; When △ f=0, the peak that the indication workpiece machining surface occurs, the probability of paddy equate.

CFRP finished surface morphology characterization based on multifractal spectra:

The material of laboratory sample is the C/CCFRP compound substance, adopts PVD diamond cutter cut, machined parameters configuration: speed of mainshaft n=2500r/min, cutting depth a _p=0.4mm, speed of feed v _f=120mm/mi, the n measuring surface form adopts the Talysurf5P-120 surface topographic apparatus fo of Rank Taylor Hobson company, adopts wavelet analysis to calculate its multifractal spectra, and the C/CCFRP composite material surface pattern processed is as shown in Figure 5.

According to above-mentioned Multifractal Analysis, change the cutting depth value, prepare a computer program the multifractal spectrogram that obtains as shown in Figure 6.The multifractal spectra parameter calculated is in Table 1.

CFRP cutting surface multifractal spectra characteristic parameter under the different cutting depth of table 1 and amplitude and spatial parameter calculated value

From Fig. 6 and Biao, multifractal spectra is all the curved state in bottom right, multifractal spectra parameter △ f is negative value, also can find out, C line △ a minimum in Fig. 6, illustrate that finished surface is relatively the most smooth, minimum in A line △ f absolute value three, illustrate that the crest of its finished surface and the probability distribution of trough are tending towards equal.B line △ a maximum, illustrate that its finished surface is the most coarse, and the △ f of absolute value maximum is the C line, illustrates that other finished surface of crest likelihood ratio that finished surface occurs wants many, but that its finished surface is still the crest ratio trough is more.

The curved shape in three bottom right all shows that the crest of finished surface regional area appearance is sharper, and this is that fibre breakage often appears in the CFRP finished surface, the phenomenon of the outstanding matrix of the fibers such as fiber tearing.This also matches with actual conditions.

In order to verify the characterization result of multifractal spectra, use 3D amplitude roughness parameter S _qwith space pattern parameter S _alshape appearance figure shown in Fig. 5 is evaluated.S _qbe the amplitude roughness evaluation parameter, by top multifractal analysis of spectrum, what in the three, finished surface was the most smooth is the C line, the S that it is corresponding _qvalue should be also minimum; The B line is the most coarse, the S that it is corresponding _qvalue should be also maximum, this and table 2S _qresults of calculation is consistent.S _albe the rate of decay of surperficial autocorrelation, the parameter of surperficial constituent frequency is described, its value is larger, illustrates that surface mainly consists of the long wave composition, otherwise, be that the shortwave composition occupies the majority, as shown in table 2 calculates, above the S of three's maximum _alin the C line, show that its surperficial crest is also maximum in the three surface; Maximum S _alin the A line, show that its surperficial crest is also minimum in the three surface, this is also consistent with top multifractal spectra discussion result.

The present invention, from wavelet theory, adopts the multifractal spectra of small wave converting method calculating CFRP cutting surface, the feature of definition multifractal parameter characterization CFR cutting surface pattern.Wavelet theory and multifractal parameter are that the useful of current three-dimensional appearance parameter evaluation supplemented, set up the morphology characterization model based on CFRP cutting surface Multifractal Analysis, not only the feature of finished surface integral body can be described, its local shape characteristic can also be described, wherein △ a can characterize the roughness of finished surface, the crest that the positive negativity explanation finished surface of △ f occurs and the probability of trough, the vertical drop of its absolute value reflection finished surface.

Surface topography is the stochastic process of a non-stationary, and finished surface has geometrical self-similarity and from the feature of affinity.Traditional is the evaluation to the Spatial Statistical Character of material surface with the parameter characterization method, development along with the present information treatment technology, especially the proposition of Concept of Fractal, for new thinking has appearred in the sign of surperficial new look, therefore, in surface topography characterization model process of establishing based on Multifractal Analysis, take wavelet analysis technology as means, find out the fractal parameter of energy characterization of surfaces pattern inherent feature, useful supplementing carried out in spatial parameter evaluation to material surface, and this is one of bright spot of the present invention.

For common metal material, because its cutting surface is directly to be cut out by cutter, so surface uniform and certain rule is arranged, adopt one or some profiles to reach well approximate to cutting surface.And, for CFRP, due to its structure and performance, its cutting surface is not directly cut out by cutter, surface presents various feature, adopts one or some the profiles can not these surface expressions are clear.Therefore, to these surfaces, should consider more suitably characterizing method and more suitably characterization parameter, to reach the pattern information of more truly extracting the CFRP cutting surface, more accurately to reflect the shape characteristic that it is surperficial, this is two of bright spot of the present invention.

CFRP is a kind of high-tech new material of mechanical property excellence, it with light-high-strength, high ratio modulus, low thermal coefficient of expansion and outstanding thermotolerance and the advantage such as wear-resistant in the field widespread uses such as material for Aero-Space, communications and transportation, sports equipment, civil construction and other industrial characteristic, the finished surface microscopic appearance of CFRP material workpiece is used function that important impact is arranged on it, the surface topography of the C/C compound substance that for example, aeroplane brake discs, automobile braking device are used can exert an influence to its friction, polishing machine; The surface topography that is used for the C/C compound substance of rocket afterbody affects its flame resistance energy.On the basis of Measurement accuracy, pass through parameter characterization, research surface quality model is used the impact of function on workpiece, and determine using the larger characterization parameter of function effect, thereby determine the descriptive model of surface quality, can be correct Choice and process technique, the performance of giving full play to compound substance provides theoretical foundation and technical support.

Because the characteristic that CFRP is good with it is increasingly extensive with the application in the fields such as material in above industrial characteristic, also more and more higher to the requirement of CFRP machined surface quality, because surface quality will affect the performance of CFRP performance.How to improve the CFRP machined surface quality, reduce its surfaceness and seem extremely important, and will control the CFRP workpiece surface roughness, must model machining composite material surfaceness model, therefrom find to affect its factor.For metal, the influence factor funtcional relationship of cutting surface roughness has ripe theory and experimental formula, so just can be according to by desired surface quality, controlling these factors in process of manufacture, thus increase work efficiency, reduce production cost.Yet, due to the design feature of CFRP self, in cutting process, can not continue to use some conclusions and the experimental formula of metal material.The present invention is directed to the factor that affects CFRP cut surfaceness is studied, sum up the rule that affects of each factor, and then determine the machining scheme adopted, and the prerequisite of these researchs is exactly the CFRP workpiece cutting surface roughness that can accurately characterize under different processing conditionss.Therefore, the present invention has great importance to research CFRP surface roughness affect factor.

Above content is in conjunction with concrete preferred implementation further description made for the present invention, can not assert that specific embodiment of the invention is confined to these explanations.For the general technical staff of the technical field of the invention, without departing from the inventive concept of the premise, can also make some simple deduction or replace, all should be considered as belonging to protection scope of the present invention.

Claims (9)

1. a carbon fibre composite cut surface topography characterizing method is characterized in that: comprises the following steps,

Steps A, be made as the continuous curve surface piece on bounded domain by workpiece to be characterized, and effects on surface shape characteristic function F (x, y) is measured;

Step B, carry out the decomposition of small echo composition to the shape characteristic function F (x, y) of bounded domain three-dimensional surface, note F (x, y)=F _i+1, by F _i+1be decomposed into the surface topography function G _iand sign reference field function F _i, and by F _icarry out lower one deck decomposition,

Obtain F _i+1=G _i+ G _i-1+ G _i-2+ ...+G _m+ F _m, wherein, F _mbe the sign reference field of wavelet decomposition.

2. carbon fibre composite cut surface topography characterizing method according to claim 1, it is characterized in that: described step B meets following formula:

h ₁(x,y)=F(x,y)-h ₂(x,y)

Wherein, h ₁(x, y) is the dimensional discrepancy with respect to the ideal surfaced shape, h ₂(x, y) means the evaluation reference field.

3. carbon fibre composite cut surface topography characterizing method according to claim 1, it is characterized in that: described step B meets following formula:

F _i+1=F _i+G _i

In formula

Wherein

4. carbon fibre composite cut surface topography characterizing method according to claim 1 is characterized in that: further comprising the steps in described step B,

The best fit approximation space of F (x, y) is found out in initialization;

Interative computation, decompose F (x, y);

Calculate iteration according to the Chi-square method of the goodness of fit and finally decompose end condition.

5. carbon fibre composite cut surface topography characterizing method according to claim 1, it is characterized in that: described carbon fibre composite cut surface topography characterizing method also comprises step C,

The characterization result of checking multifractal spectra, used 3D amplitude roughness parameter S _qwith space pattern parameter S _althe effects on surface pattern is evaluated.

6. carbon fibre composite cut surface topography characterizing method according to claim 5, it is characterized in that: described step C meets following formula:

M in formula, N is sampling number, h (x _i, y _i) be the height of sampled point with respect to the evaluation reference field;

S _albe defined as:

R (τ _x, τ _y) be surperficial two-dimensional autocorrelation function (AACF), τ _x, τ _ybe respectively x, the auto-correlation length on the y direction, x direction and y direction are surface pairwise orthogonal direction arbitrarily.

7. carbon fibre composite cut surface topography characterizing method according to claim 5, it is characterized in that: described measuring surface form adopts the Talysurf5P-120 surface topographic apparatus fo of Rank Taylor Hobson company.

8. carbon fibre composite cut surface topography characterizing method according to claim 5, is characterized in that: in described step C, adopt two-dimentional multifractal to fall Fluctuation Method and analyze surface topography.

9. carbon fibre composite cut surface topography characterizing method according to claim 6 is characterized in that:

The width Delta a that defines fractal spectrum is:

Δa=a _max-a _min

The spectral difference Δ f of definition multifractal spectra is:

Δf=f(a _min)-f(a _max)，

The size of the width Delta a reflection probability distribution scope of described fractal spectrum, the crest on described Δ f reflection materials processing surface and the probability distribution of trough.

Images