CN108763668A - The model of gear region parameter method replaced with boundary based on subdivision technology - Google Patents

The model of gear region parameter method replaced with boundary based on subdivision technology Download PDF

Info

Publication number
CN108763668A
CN108763668A CN201810463056.9A CN201810463056A CN108763668A CN 108763668 A CN108763668 A CN 108763668A CN 201810463056 A CN201810463056 A CN 201810463056A CN 108763668 A CN108763668 A CN 108763668A
Authority
CN
China
Prior art keywords
point
curve
model
segmentation
boundary
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201810463056.9A
Other languages
Chinese (zh)
Other versions
CN108763668B (en
Inventor
徐岗
李博剑
许金兰
肖周芳
吴卿
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hangzhou Dianzi University
Original Assignee
Hangzhou Dianzi University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hangzhou Dianzi University filed Critical Hangzhou Dianzi University
Priority to CN201810463056.9A priority Critical patent/CN108763668B/en
Publication of CN108763668A publication Critical patent/CN108763668A/en
Application granted granted Critical
Publication of CN108763668B publication Critical patent/CN108763668B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00Computer-aided design [CAD]
    • G06F30/10Geometric CAD
    • G06F30/17Mechanical parametric or variational design

Landscapes

  • Physics & Mathematics (AREA)
  • Geometry (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Mathematics (AREA)
  • Image Generation (AREA)

Abstract

The invention discloses the model of gear region parameter methods replaced with boundary based on subdivision technology.Graphical model with complex region is difficult the parametrization for accomplishing high quality.The present invention reduces the concave-convex degree of each section of curve by curve segmentation method, ensures the network of quadrilaterals lattice model that not will produce overturning, to optimize the quality of model of gear parametrization, and meets it in engineer application for waiting validity of geometrical analysis to require;Pass through the modification to model of gear Polygonal Boundary average distance value, the distance between uniform boundary vertex, to improve the production rate of quadrilateral mesh later;Ensure that the shape of the model of gear will not change by using the method that boundary is replaced, the use of method for fairing and C-C divided methods is then optimized the model of gear on smoothness and continuity, to meet smoothness and robustness requirement of the engineer application for equal geometrical analysis.

Description

The model of gear region parameter method replaced with boundary based on subdivision technology
Technical field
The invention belongs to computer aided design and engineering application fields, and in particular to be replaced based on subdivision technology and boundary Model of gear region parameter method.
Background technology
CAD (CAD) technology and computer-aided engineering (CAE) technology today computer application and In development, there is very far-reaching influence, they are all obtained in fields such as aerospace, automobile making, architectural design, mechano-electronics To being widely applied.In cad technique, body form is indicated by boundary model, and boundary model contains The information of geometrical model curved inner surface.Equal geometrical analysis are frequently necessary to carry out the parametrization inside geometrical model region, therefore are The equal geometrical analysis that allow that CAD model is more applicable for, it is necessary to the method that research builds region parameter by giving data boundary Technology.
Equal geometrical analysis are a kind of new techniques of the precise geometrical expression progress analog simulation based on CAD model, it It proposes to realize that the seamless fusion of CAD/CAE opened up a new way, while being also that the geometric modeling field for having become ripe is filled with New vigor.In equal geometrical analysis, the parametrization of computational domain is still current one of research hotspot, frequent in finite element analysis The parameterized model for needing high-precision and validity, to be applied in the research and design of high-level CAD or CAE models.And it is Reach these requirements, it is necessary to being optimized for the grid of parametrization and some special areas, specific to some The complexity of method, region largely affects the validity and quality of parametrization.
The parametrization in a region is needed in equal geometrical analysis to meet three requirements:1) it does not generate selfing, namely Mapping from parameter field to physical domain is interior penetrates;2) isoparametric element element should as much as possible uniformly;3) isoparametric element element should to the greatest extent can Energy ground is orthogonal.Currently, the boundary information indicated by B-spline curves can be handled there are many method, they include variation Harmonic wave method, technology of dividing and ruling, reflection method, the decomposition method based on profile, multi-panel piece parametric method and non-standard B-spline Parametric method.But the parametrization of some simple regions can only be handled in existing method mostly, and for complex region Some graphical models, be but difficult to accomplish the parametrization of high quality.Such as a kind of situation be handle some have concave-convex degree compared with When big net region, it is easy to form the grid of overturning, the serious quality for affecting region parameter at this.
Invention content
The purpose of the present invention is the deficiencies for existing method, propose a kind of gear replaced with boundary based on subdivision technology Model area parametric method, this method have mainly used the thought that a kind of boundary replaces and C-C are combined to segment, realized to by curve The parametrization that the model of gear of boundary composition works well, and obtain final parameterized model.
The present invention is specific as follows:
Step 1 samples the inside and outside boundary of the model of gear, on inside and outside boundary since first sampled point, 4 sampled points are taken to be fitted to a cubic uniform B-splines successively every time, in adjacent two cubic uniform B-splines, after When first sampled point used when one cubic uniform B-splines fitting is fitted for previous cubic uniform B-splines The last one sampled point used;After fitting, inner boundary is made of 4 cubic uniform B-splines, outer boundary by 24 three times Uniform B-Spline Curve forms;Cubic uniform B-splines expression formula is:
Wherein, PiFor the control point position for building cubic uniform B-splines, i takes 0,1,2 or 3, Ni,3(t) it is B Spline base function, P (t) are the match point position on the cubic uniform B-splines formed, parameter t ∈ node intervals [0,1].
Step 2 passes through the inside and outside boundary through step 1 treated the model of gear curvilinear characteristic point and carries out curve minute It cuts, then converts the inside and outside boundary of the model of gear to Polygonal Boundary geometrical model, it is specific as follows:
Step a, by the method for parabola interpolation, the inside and outside borderline curvilinear characteristic point of the model of gear, curve are detected Characteristic point includes the point with extreme curvature and inflection point of curve, and curve segmentation is carried out at curvilinear characteristic point.
The process that parabola interpolation detects point with extreme curvature is specific as follows:By the inside and outside every borderline of the model of gear The node interval [0,1] of cubic uniform B-splines is divided into 20 equal portions section [tn-1,tn], n=1,2 ..., 20, then according to Secondary traversal [tn-1,tn+1], computation interval [tn-1,tn+1] in three match point curvature k (tn-1)、k(tn) and k (tn+1), if full Sufficient condition k (tn)>k(tn-1) and k (tn)>k(tn+1), then current interval is section where bent curvature of a curve maximum point, and if Meet condition k (tn)<k(tn-1) and k (tn)<k(tn+1), then current interval is section where curvature minimum point.Curvature maximum Section where section where point or curvature minimum point passes through three point (t of interpolationn-1,k(tn-1)), (tn,k(tn)) and (tn+1, k(tn+1)) obtain parabola f (m)=a2m2+a1m+a0, wherein a0,a1,a2For coefficient, then by setting leading for f (m) Number f ' (m)=0 acquire the m values of the point in parabola on symmetry axis, are denoted as t ':
Using parabola peak or the limit curvature of minimum point point t ' as cubic uniform B-splines in section [tn-1, tn] in curvature maximum point or curvature minimum point.
The process of parabola interpolation detection curve inflection point is specific as follows:By inside and outside borderline every three of the model of gear The node interval [0,1] of secondary Uniform B-Spline Curve is divided into 20 equal portions section [tn-1,tn], [t is then traversed successivelyn-1,tn], meter Calculate section [tn-1,tn] two endpoints curvature value, meet k (tn-1)*k(tn)<When 0, there are points of inflexion on a curve in the section.So Midpoint (the t in the parameter section is calculated afterwardsn-0.5,k(tn-0.5)), wherein tn-0.5=1/2* (tn-1+tn).If | k (tn-0.5)-0 |<0.0001, just directly by tn-0.5As point of inflexion on a curve, otherwise, still through (tn-1,k(tn-1)), (tn-0.5,k(tn-0.5)) (tn,k(tn)) one parabola f (m)=b of three point interpolation2m2+b1m+b0, wherein b0,b1,b2For coefficient, finds out f (m)=0 and exist Section [tn-1,tn] in unique root, and using root as cubic uniform B-splines in section [tn-1,tn] in inflection point.
Step b, using the curvilinear characteristic point of cubic uniform B-splines in step a as cut-point, to the curve weight of segmentation New distribution control point;By node interval [ta, tb] two-end-point taAnd tbThe t substituted into formula (1) is found out at segmentation curve two-end-point Control vertex PaAnd Pb.The curve of segmentation and the match point position correspondence of cubic uniform B-splines are as follows:
In formula, Pab(t ") is segmentation curve matching point position, parameter t " ∈ node intervals [ta, tb];According to the song of segmentation The match point position correspondence of line and cubic uniform B-splines calculates segmentation curve residue intermediate control point position difference For:
Wherein, P0And P1For the control point position of former cubic uniform B-splines, Pa1And Pa2It is two among segmentation curve Control point position, to be that segmentation curve is assigned with new control point position according to former cubic uniform B-splines, respectively Pa, Pa1, Pa2, Pb
Step 3 first traverses the segmentation point sequence of acquisition, calculates the distance between adjacent segmentation point, and calculate The average value h of each adjacent segmentation point distance.When the distance between adjacent segmentation point is less than 0.5 λ h, wherein 0 < λ < 1, it will be adjacent The latter divides point deletion in cut-point;The distance between adjacent segmentation point be more than 1.5 λ h when, by between adjacent segmentation point away from From subtracting one after the result rounding plus 0.5 λ h resulting values divided by λ h as the quantity for being inserted into cut-point, thus to the adjacent segmentation Section [t belonging to pointn-1,tn] divided equally, to realize the segmentation of corresponding B-spline Curve, then according in step 2 The method that step b redistributes control point redistributes control point to the curve of segmentation.Finally, it is bent to be sequentially connected every segmentation The control point at line both ends, to convert curved boundary to obtain the Polygonal Boundary geometrical model of the model of gear;Use quadrangle Subdivision algorithm generates the network of quadrilaterals lattice model of the model of gear, and carries out SmoothNumerical TechniqueandIts to grid using iteration method for fairing.
Step 4 controls grid model to the quadrilateral mesh model construction of the model of gear and carries out C-C sub-dividing operations, together When by replacing boundary Control point and curve node be inserted into that treated keeping step 1 is fitted through cubic uniform B-splines Model of gear curved boundary shape.
Step a, the control point of network of quadrilaterals lattice model each quadrilateral surface sheet border and inside is calculated using linear interpolation, Four vertex including quad patch generate the control point sequence of a 4*4 in each quadrilateral surface on piece, then will pass through The control point on segmentation curved boundary in step 3 after curve segmentation be substituted into corresponding quadrilateral mesh model boundary pass through it is linear On the control point that interpolation obtains, to construct the control grid model for parametrization.
Step b, the internal vertex of control grid model is handled using C-C divided methods, generates C-C subdivided meshes models.
Step c, the method for being inserted into node and control point using B-spline curves increases every in step 3 after curve segmentation The number of control points of boundary curve, it is specific as follows to match the number on C-C subdivided meshes model boundaries vertex:
After generating C-C subdivided meshes models, the number of vertices corresponding to the boundary of each quad patch increases to 7 It is a, still ensureing that knot vector is uniform, three times uniformly by every in the boundary curve after curve segmentation in step 3 B-spline curves increase by 3 nodes, while changing every cubic uniform B-spline song in the boundary curve in step 3 after curve segmentation The control point position of centre two of line, and increase by 3 control points, the position at 7 new control points calculate separately for:
P′0=P0
P′6=P3
Wherein, Pr', r=0,1 ..., 6 be the position at new control point, Ps, s=0, after 1,2,3 is curve segmentation in step 3 Boundary curve in every cubic uniform B-splines original 4 control point positions.
It is inserted into after node, the new control point of each edge circle curve is substituted into corresponding quadrilateral mesh model boundary successively On control point after C-C is segmented.
Step 5, the C-C subdivided meshes model for obtaining the step c in step 4 are as the control net for drawing B-spline surface Lattice compare the network of quadrilaterals lattice model generated in step 3, and each quad patch, which corresponds to drafting one, 7 rows 7 row control point Cubic uniform B-spline curved surface completes the parametrization of plane domain.
Further, the calculation formula of curvature is:
In formula, det is determinant operation, and P ' (t) are the first derivative of P (t), and P " (t) are the second dervative of P (t);x (t), y (t) is the coordinate value at P (t) points, and x ' (t), x " (t) are respectively the first derivative and second dervative of x (t), y ' (t), y " (t) be respectively y (t) first derivative and second dervative, | | P ' (t) | | be P ' (t) modular arithmetic.
The device have the advantages that:
The present invention proposes a kind of complex region parametric method of high quality, passes through curve segmentation method, curve segmentation The concave-convex degree for reducing each section of curve afterwards ensures not will produce the network of quadrilaterals lattice model of overturning, to optimize with this The quality of model of gear parametrization, and it is met in engineer application for waiting validity of geometrical analysis to require.Later By the modification to model of gear Polygonal Boundary average distance value, the distance between uniform border vertices, to improve The production rate of quadrilateral mesh.In addition, ensure that the shape of the model of gear will not change by using the method that boundary is replaced, light Use along method and C-C divided methods is then optimized the model of gear on smoothness and continuity, to meet Smoothness and robustness requirement of the engineer application for equal geometrical analysis.
Description of the drawings
Fig. 1 is a model of gear used in the embodiment of the present invention;
Fig. 2 is the Polygonal Boundary geometrical model of the model of gear after curved boundary conversion;
Fig. 3 is the schematic diagram of the limit curvature point and inflection point of B-spline curves;
Fig. 4 is segmentation and the control point reallocation schematic diagram of B-spline curves;
Fig. 5 is that Fig. 2 carries out the quadrilateral mesh illustraton of model after quadrangle grids;
Fig. 6 is deletion and the insertion schematic diagram of curve segmentation point;
Fig. 7 is substituted into corresponding quadrilateral mesh model boundary for the control point on the segmentation curved boundary after curve segmentation and leads to Cross the schematic diagram on the control point that linear interpolation obtains;
Fig. 8 is the final parameterized model of Fig. 1 middle gear models.
Specific implementation mode
The curved boundary model that a gear shape is provided below in conjunction with practical application carries out region using the present invention to it Parametrization allows that the model of gear is more applicable for and waits geometrical analysis to obtain the parameterized model needed in engineer application.
It is specific as follows based on the model of gear region parameter method that subdivision technology is replaced with boundary:
Step 1, Fig. 1 are a model of gear, are sampled to the inside and outside boundary of the model of gear, on inside and outside boundary Since first sampled point, 4 sampled points is taken to be fitted to a cubic uniform B-splines successively every time, adjacent two three In secondary Uniform B-Spline Curve, first sampled point that when latter cubic uniform B-splines fitting uses for previous item three times The last one sampled point that Uniform B-Spline Curve uses when being fitted;Control point is adjusted, cubic uniform B-splines is made to be fitted tooth Take turns the curved boundary of model;After fitting, inner boundary is made of 4 cubic uniform B-splines, and outer boundary is uniform three times by 24 B-spline curves form, and every cubic uniform B-splines are containing there are four control points;Cubic uniform B-splines expression formula is:
Wherein, PiFor the control point position for building cubic uniform B-splines, i takes 0,1,2 or 3, Ni,3(t) it is B Spline base function, P (t) are the match point position on the cubic uniform B-splines formed, parameter t ∈ node intervals [0,1].
Step 2 passes through the inside and outside boundary through step 1 treated the model of gear curvilinear characteristic point and carries out curve minute It cuts, then converts the inside and outside boundary of the model of gear to Polygonal Boundary geometrical model, it is specific as follows:
Step a, by the method for parabola interpolation, the inside and outside borderline curvilinear characteristic point of the model of gear, curve are detected Characteristic point includes the point with extreme curvature and inflection point (such as Fig. 3) of curve, and curve segmentation is carried out at curvilinear characteristic point.The song of curve Rate extreme point is the more apparent place of the concave-convex degree of curve itself, and the concave-convex degree of each section of curve is reduced after curve segmentation, And it thereby guarantees that in the network of quadrilaterals lattice model being subsequently generated without there is the case where overturning grid (such as Fig. 5).
The process that parabola interpolation detects point with extreme curvature is specific as follows:By the inside and outside every borderline of the model of gear The node interval [0,1] of cubic uniform B-splines is divided into 20 equal portions section [tn-1,tn], n=1,2 ..., 20, then according to Secondary traversal [tn-1,tn+1], computation interval [tn-1,tn+1] in three match point curvature k (tn-1)、k(tn) and k (tn+1), if full Sufficient condition k (tn)>k(tn-1) and k (tn)>k(tn+1), then current interval is section where bent curvature of a curve maximum point, and if Meet condition k (tn)<k(tn-1) and k (tn)<k(tn+1), then current interval is section where curvature minimum point.Curvature maximum Section where section where point or curvature minimum point passes through three point (t of interpolationn-1,k(tn-1)), (tn,k(tn)) and (tn+1, k(tn+1)) obtain parabola f (m)=a2m2+a1m+a0, wherein a0,a1,a2For coefficient, then by setting leading for f (m) Number f ' (m)=0 acquire the m values of the point in parabola on symmetry axis, are denoted as t ':
Using parabola peak or the limit curvature of minimum point point t ' as cubic uniform B-splines in section [tn-1, tn] in curvature maximum point or curvature minimum point.
The process of parabola interpolation detection curve inflection point is specific as follows:By inside and outside borderline every three of the model of gear The node interval [0,1] of secondary Uniform B-Spline Curve is divided into 20 equal portions section [tn-1,tn], [t is then traversed successivelyn-1,tn], meter Calculate section [tn-1,tn] two endpoints curvature value, meet k (tn-1)*k(tn)<When 0, illustrate turning there are curve in the section Point.Then midpoint (the t in the parameter section is calculatedn-0.5,k(tn-0.5)), wherein tn-0.5=1/2* (tn-1+tn).If | k (tn-0.5)-0|<0.0001, just directly by tn-0.5As point of inflexion on a curve, otherwise, still through (tn-1,k(tn-1)), (tn-0.5, k(tn-0.5)) and (tn,k(tn)) one parabola f (m)=b of three point interpolation2m2+b1m+b0, wherein b0,b1,b2For coefficient, f is found out (m)=0 in section [tn-1,tn] in unique root, and using root as cubic uniform B-splines in section [tn-1,tn] in Inflection point.
The calculation formula of curvature is:
In formula, det is determinant operation, and P ' (t) are the first derivative of P (t), and P " (t) are the second dervative of P (t);x (t), y (t) is the coordinate value at P (t) points, and x ' (t), x " (t) are respectively the first derivative and second dervative of x (t), y ' (t), y " (t) be respectively y (t) first derivative and second dervative, | | P ' (t) | | be P ' (t) modular arithmetic.
Step b, as shown in figure 4, using the curvilinear characteristic point of cubic uniform B-splines in step a as cut-point, to dividing The curve cut redistributes control point;By node interval [ta, tb] two-end-point taAnd tbIt is bent that the t substituted into formula (1) finds out segmentation Control vertex P at line two-end-pointaAnd Pb.The curve of segmentation and the match point position correspondence of cubic uniform B-splines are such as Under:
In formula, Pab(t ") is segmentation curve matching point position, parameter t " ∈ node intervals [ta, tb];According to the song of segmentation The match point position correspondence of line and cubic uniform B-splines calculates segmentation curve residue intermediate control point position difference For:
Wherein, P0And P1For the control point position of former cubic uniform B-splines, Pa1And Pa2It is two among segmentation curve Control point position, so just according to former cubic uniform B-splines be segmentation curve be assigned with new control point position, respectively Pa, Pa1, Pa2, Pb
Step 3, as shown in fig. 6, traversed first to the segmentation point sequence of acquisition, calculate between adjacent segmentation point away from From, and calculate the average value h of each adjacent segmentation point distance.(λ is in 0 < λ < 1 when the distance between adjacent segmentation point is less than 0.5 λ h In take a value), by the latter cut-point in adjacent segmentation point (the P points in such as Fig. 6) delete;The distance between adjacent segmentation point When more than 1.5 λ h, subtract one as slotting after the distance between adjacent segmentation point to be added to the result rounding of 0.5 λ h resulting values divided by λ h Enter the quantity of cut-point, to the section [t belonging to the adjacent segmentation pointn-1,tn] divided equally, to realize corresponding B three times The segmentation of spline curve redistributes the curve of segmentation then according to the method that step b redistributes control point in step 2 Control point (the Q points in such as Fig. 6).Finally, it is sequentially connected the control point at every segmentation curve both ends, to convert curved boundary The Polygonal Boundary geometrical model of the model of gear being made of 110 vertex is obtained, as shown in Fig. 2, its vertex distance reaches The effect of relatively uniform distribution, while the quantity of former boundary curve has risen to 110 from 28.
The network of quadrilaterals lattice model of the model of gear is generated using quadrangle grids algorithm, and using iteration method for fairing to net Lattice carry out SmoothNumerical TechniqueandIts, to obtain network of quadrilaterals lattice model that is uniform and smooth and having preferable robustness.It is gear as shown in Figure 5 The quadrilateral mesh model result that model generates, wherein quadrilateral mesh quantity are 196.
Step 4 controls grid model to the quadrilateral mesh model construction of the model of gear and carries out C-C sub-dividing operations, together When by replacing boundary Control point and curve node be inserted into that treated keeping step 1 is fitted through cubic uniform B-splines Model of gear curved boundary shape.
Step a, as shown in fig. 7, calculating each quadrilateral surface sheet border of network of quadrilaterals lattice model and interior using linear interpolation The control point in portion includes four vertex (d of quad patch0,0、d0,3、d3,0And d3,3, wherein d0,0As P0, d0,3As P3) Inside, the control point sequence (i.e. 4 rows 4 arrange control point) that a 4*4 is generated in each quadrilateral surface on piece, then will pass through step The control point on segmentation curved boundary in 3 after curve segmentation is substituted into corresponding quadrilateral mesh model boundary and passes through linear interpolation Obtained control point d0,1、d0,2On, to construct the control grid model for parametrization.
Step b, the internal vertex of control grid model is handled using C-C divided methods, generates C-C subdivided meshes models, The effect of C-C subdivisions is tessellated mesh, improves the continuity and smoothness of grid.
Step c, the method for being inserted into node and control point using B-spline curves increases every in step 3 after curve segmentation The number of control points of boundary curve, to match the number on C-C subdivided meshes model boundaries vertex.
After generating C-C subdivided meshes models, the number of vertices corresponding to the boundary of each quad patch increases to 7 It is a, still ensureing that knot vector is uniform, three times uniformly by every in the boundary curve after curve segmentation in step 3 B-spline curves increase by 3 nodes, while changing every cubic uniform B-spline song in the boundary curve in step 3 after curve segmentation The control point position of centre two of line, and increase by 3 control points, the position at 7 new control points calculate separately for:
P′0=P0
P′6=P3
Wherein, Pr', r=0,1 ..., 6 be the position at new control point, Ps, s=0, after 1,2,3 is curve segmentation in step 3 Boundary curve in every cubic uniform B-splines original 4 control point positions.
It is inserted into after node, the new control point of each edge circle curve is substituted into corresponding quadrilateral mesh model boundary successively On control point after C-C is segmented.It can realize step 1 in this way treated and be fitted through cubic uniform B-splines The effect of model of gear curved boundary shape invariance can equally retain on the parameterized model after parametrization through uniform three times The model of gear curved boundary shape of B-spline curves fitting.
Step 5, as shown in figure 8, the C-C subdivided meshes model that the step c in step 4 is obtained as drawing B-spline song The control grid in face compares the network of quadrilaterals lattice model generated in step 3, and each quad patch, which corresponds to drafting one, 7 rows 7 The cubic uniform B-spline curved surface at control point is arranged to complete the parametrization of plane domain, thus obtains the parametrization of final high quality As a result.Fig. 8 is the final parameterized model of Fig. 1 middle gear models.
We have just obtained the high quality parameterized model of gear in this way, substantially met CAD or CAE application in for The demand of parameterized model slickness and robustness.
The present invention reduces the concave-convex degree of each section of curve by curve segmentation method after curve segmentation, ensured with this The network of quadrilaterals lattice model that not will produce overturning to optimize the quality of model of gear parametrization, and meets it in work For the validity requirement of equal geometrical analysis in Cheng Yingyong.Later by being repaiied to model of gear Polygonal Boundary average distance value Change, the distance between uniform border vertices, to improve the production rate of quadrilateral mesh.In addition, being replaced by using boundary Method ensure that the shape of the model of gear will not change, the use of method for fairing and C-C divided methods is then in smoothness and continuous Property on the model of gear is optimized, to meet smoothness and robustness requirement of the engineer application for equal geometrical analysis.
To sum up, the present invention is satisfied high-precision and is had by the combined use of subdivision technology and boundary replacement method Effect property require Gear Parameterization model, it is adapted to the parametrization of complex region, so as to be applied to high-level CAD or In the research and design of CAE models.

Claims (2)

1. the model of gear region parameter method replaced with boundary based on subdivision technology, it is characterised in that:This method is specifically such as Under:
Step 1 samples the inside and outside boundary of the model of gear, on inside and outside boundary since first sampled point, every time 4 sampled points are taken to be fitted to a cubic uniform B-splines successively, in adjacent two cubic uniform B-splines, latter item First sampled point that cubic uniform B-splines use when being fitted uses when being fitted for previous cubic uniform B-splines The last one sampled point;After fitting, inner boundary is made of 4 cubic uniform B-splines, and outer boundary is uniform three times by 24 B-spline curves form;Cubic uniform B-splines expression formula is:
Wherein, PiFor the control point position for building cubic uniform B-splines, i takes 0,1,2 or 3, Ni,3(t) it is B-spline base Function, P (t) are the match point position on the cubic uniform B-splines formed, parameter t ∈ node intervals [0,1];
Step 2 carries out curve segmentation to the inside and outside boundary through step 1 treated the model of gear by curvilinear characteristic point, so Convert the inside and outside boundary of the model of gear to Polygonal Boundary geometrical model afterwards, it is specific as follows:
Step a, by the method for parabola interpolation, the inside and outside borderline curvilinear characteristic point of the model of gear, curvilinear characteristic are detected Point includes the point with extreme curvature and inflection point of curve, and curve segmentation is carried out at curvilinear characteristic point;
The process that parabola interpolation detects point with extreme curvature is specific as follows:Three times by inside and outside borderline every of the model of gear The node interval [0,1] of Uniform B-Spline Curve is divided into 20 equal portions section [tn-1,tn], n=1,2 ..., 20, then successively time Go through [tn-1,tn+1], computation interval [tn-1,tn+1] in three match point curvature k (tn-1)、k(tn) and k (tn+1), if meeting item Part k (tn)>k(tn-1) and k (tn)>k(tn+1), then current interval is section where bent curvature of a curve maximum point, and if met Condition k (tn)<k(tn-1) and k (tn)<k(tn+1), then current interval is section where curvature minimum point;Curvature maximum point institute Pass through three point (t of interpolation in section where section or curvature minimum pointn-1,k(tn-1)), (tn,k(tn)) and (tn+1,k (tn+1)) obtain parabola f (m)=a2m2+a1m+a0, wherein a0,a1,a2For coefficient, then pass through the derivative of setting f (m) F ' (m)=0 acquires the m values of the point in parabola on symmetry axis, is denoted as t ':
Using parabola peak or the limit curvature of minimum point point t ' as cubic uniform B-splines in section [tn-1,tn] in Curvature maximum point or curvature minimum point;
The process of parabola interpolation detection curve inflection point is specific as follows:Three times by inside and outside borderline every of the model of gear The node interval [0,1] of even B-spline curves is divided into 20 equal portions section [tn-1,tn], [t is then traversed successivelyn-1,tn], calculate area Between [tn-1,tn] two endpoints curvature value, meet k (tn-1)*k(tn)<When 0, there are points of inflexion on a curve in the section;Then it counts Calculate the midpoint (t in the parameter sectionn-0.5,k(tn-0.5)), wherein tn-0.5=1/2* (tn-1+tn);If | k (tn-0.5)-0|< 0.0001, just directly by tn-0.5As point of inflexion on a curve, otherwise, still through (tn-1,k(tn-1)), (tn-0.5,k(tn-0.5)) and (tn,k(tn)) one parabola f (m)=b of three point interpolation2m2+b1m+b0, wherein b0,b1,b2For coefficient, f (m)=0 is found out in area Between [tn-1,tn] in unique root, and using root as cubic uniform B-splines in section [tn-1,tn] in inflection point;
Step b, using the curvilinear characteristic point of cubic uniform B-splines in step a as cut-point, the curve of segmentation is divided again With control point;By node interval [ta, tb] two-end-point taAnd tbThe t substituted into formula (1) finds out the control at segmentation curve two-end-point Vertex P processedaAnd Pb;The curve of segmentation and the match point position correspondence of cubic uniform B-splines are as follows:
In formula, Pab(t ") is segmentation curve matching point position, parameter t " ∈ node intervals [ta, tb];According to the curve of segmentation and three The match point position correspondence of secondary Uniform B-Spline Curve calculates segmentation curve residue intermediate control point position:
Wherein, P0And P1For the control point position of former cubic uniform B-splines, Pa1And Pa2For two controls among segmentation curve Point position, to be that segmentation curve is assigned with new control point position, respectively P according to former cubic uniform B-splinesa, Pa1, Pa2, Pb
Step 3 first traverses the segmentation point sequence of acquisition, calculates the distance between adjacent segmentation point, and calculate each phase The average value h of adjacent cut-point distance;When the distance between adjacent segmentation point is less than 0.5 λ h, wherein 0 < λ < 1, by adjacent segmentation The latter divides point deletion in point;When the distance between adjacent segmentation point is more than 1.5 λ h, the distance between adjacent segmentation point is added Subtract one after the result rounding of upper 0.5 λ h resulting values divided by λ h as the quantity for being inserted into cut-point, to the adjacent segmentation point institute Section [the t of categoryn-1,tn] divided equally, to realize the segmentation of corresponding B-spline Curve, then according to step b in step 2 The method for redistributing control point redistributes control point to the curve of segmentation;Finally, every segmentation curve both ends are sequentially connected Control point, so that curved boundary be converted to obtain the Polygonal Boundary geometrical model of the model of gear;It is calculated using quadrangle grids Method generates the network of quadrilaterals lattice model of the model of gear, and carries out SmoothNumerical TechniqueandIts to grid using iteration method for fairing;
Step 4 controls grid model to the quadrilateral mesh model construction of the model of gear and carries out C-C sub-dividing operations, leads to simultaneously It crosses and replaces boundary Control point and the insertion of curve node to keep step 1 tooth that treated is fitted through cubic uniform B-splines Take turns model curved boundary shape;
Step a, the control point of network of quadrilaterals lattice model each quadrilateral surface sheet border and inside is calculated using linear interpolation, including Four vertex of quad patch generate the control point sequence of a 4*4 in each quadrilateral surface on piece, then will pass through step The control point on segmentation curved boundary in 3 after curve segmentation is substituted into corresponding quadrilateral mesh model boundary and passes through linear interpolation On obtained control point, to construct the control grid model for parametrization;
Step b, the internal vertex of control grid model is handled using C-C divided methods, generates C-C subdivided meshes models;
Step c, the method for being inserted into node and control point using B-spline curves increases each edge circle after curve segmentation in step 3 The number of control points of curve, it is specific as follows to match the number on C-C subdivided meshes model boundaries vertex:
After generating C-C subdivided meshes models, the number of vertices corresponding to the boundary of each quad patch increases to 7, In the case of still ensureing that knot vector is uniform, by every cubic uniform B-spline in the boundary curve after curve segmentation in step 3 Curve increases by 3 nodes, while changing in the boundary curve in step 3 after curve segmentation every cubic uniform B-splines Intermediate two control point positions, and increase by 3 control points, the position at 7 new control points calculate separately for:
P′0=P0
P′6=P3
Wherein, Pr', r=0,1 ..., 6 be the position at new control point, Ps, s=0,1,2,3 is the side after curve segmentation in step 3 Original 4 control point positions of every cubic uniform B-splines in boundary's curve;
It is inserted into after node, the new control point of each edge circle curve is substituted into corresponding quadrilateral mesh model boundary successively to be passed through On control point after C-C subdivisions;
Step 5, the C-C subdivided meshes model for obtaining the step c in step 4 are right as the control grid for drawing B-spline surface According to the network of quadrilaterals lattice model generated in step 3, each quad patch, which corresponds to drafting one, has 7 rows 7 row control point three times Uniform B-spline surfaces complete the parametrization of plane domain.
2. the model of gear region parameter method according to claim 1 replaced with boundary based on subdivision technology, special Sign is:The calculation formula of curvature is:
In formula, det is determinant operation, and P ' (t) are the first derivative of P (t), and P " (t) are the second dervative of P (t);X (t), y (t) it is the coordinate value at P (t) points, x ' (t), x " (t) are respectively the first derivative and second dervative of x (t), and y ' (t), y " (t) divide Not Wei y (t) first derivative and second dervative, | | P ' (t) | | be P ' (t) modular arithmetic.
CN201810463056.9A 2018-05-15 2018-05-15 Gear model region parameterization method based on subdivision technology and boundary replacement Active CN108763668B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810463056.9A CN108763668B (en) 2018-05-15 2018-05-15 Gear model region parameterization method based on subdivision technology and boundary replacement

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810463056.9A CN108763668B (en) 2018-05-15 2018-05-15 Gear model region parameterization method based on subdivision technology and boundary replacement

Publications (2)

Publication Number Publication Date
CN108763668A true CN108763668A (en) 2018-11-06
CN108763668B CN108763668B (en) 2022-03-01

Family

ID=64007746

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810463056.9A Active CN108763668B (en) 2018-05-15 2018-05-15 Gear model region parameterization method based on subdivision technology and boundary replacement

Country Status (1)

Country Link
CN (1) CN108763668B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110890074A (en) * 2019-10-30 2020-03-17 华显光电技术(惠州)有限公司 Display screen public voltage adjusting method and display screen
CN111784796A (en) * 2020-06-22 2020-10-16 上海米哈游天命科技有限公司 Terrain grid generation method, device, equipment and medium
CN113658133A (en) * 2021-08-16 2021-11-16 江苏鑫丰源机电有限公司 Gear surface defect detection method and system based on image processing

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1499447A (en) * 2002-11-12 2004-05-26 松下电器产业株式会社 Curve image processor and its processing method
CN101714259A (en) * 2008-10-06 2010-05-26 Arm有限公司 Graphics processing systems
US20130135302A1 (en) * 2011-11-29 2013-05-30 Dassault Systemes Creating a Surface from a Plurality of 3D Curves
EP2660782A1 (en) * 2012-05-02 2013-11-06 Dassault Systèmes Designing a 3D modeled object
CN103903018A (en) * 2014-04-02 2014-07-02 浙江师范大学 Method and system for positioning license plate in complex scene
CN107085865A (en) * 2017-05-12 2017-08-22 杭州电子科技大学 Quadrangle dividing method applied to finite element analysis

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1499447A (en) * 2002-11-12 2004-05-26 松下电器产业株式会社 Curve image processor and its processing method
CN101714259A (en) * 2008-10-06 2010-05-26 Arm有限公司 Graphics processing systems
US20130135302A1 (en) * 2011-11-29 2013-05-30 Dassault Systemes Creating a Surface from a Plurality of 3D Curves
EP2660782A1 (en) * 2012-05-02 2013-11-06 Dassault Systèmes Designing a 3D modeled object
CN103903018A (en) * 2014-04-02 2014-07-02 浙江师范大学 Method and system for positioning license plate in complex scene
CN107085865A (en) * 2017-05-12 2017-08-22 杭州电子科技大学 Quadrangle dividing method applied to finite element analysis

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
D. BURKHART等: "Iso-geometric Finite Element Analysis Based on Catmull-Clark Subdivision Solids", 《EUROGRAPHICS SYMPOSIUM ON GEOMETRY PROCESSING 2010》 *
QING PAN等: "Isogeometric analysis based on extended Loop’s subdivision", 《JOURNAL OF COMPUTATIONAL PHYSICS 299》 *
杨雨: "面向数控加工的网格曲面规范化方法研究", 《中国优秀博硕士学位论文全文数据库(硕士) 工程科技Ⅰ辑》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110890074A (en) * 2019-10-30 2020-03-17 华显光电技术(惠州)有限公司 Display screen public voltage adjusting method and display screen
CN110890074B (en) * 2019-10-30 2022-10-28 华显光电技术(惠州)有限公司 Display screen public voltage adjusting method and display screen
CN111784796A (en) * 2020-06-22 2020-10-16 上海米哈游天命科技有限公司 Terrain grid generation method, device, equipment and medium
CN113658133A (en) * 2021-08-16 2021-11-16 江苏鑫丰源机电有限公司 Gear surface defect detection method and system based on image processing

Also Published As

Publication number Publication date
CN108763668B (en) 2022-03-01

Similar Documents

Publication Publication Date Title
CN110516388B (en) Harmonic mapping-based curved surface discrete point cloud model circular cutter path generation method
CN109711048A (en) It is a kind of to consider pneumatic and structure polymorphic type design variable Wing design method
Farouki The characterization of parametric surface sections
CN111797555B (en) Geometric reconstruction method based on finite element model
CN108763668A (en) The model of gear region parameter method replaced with boundary based on subdivision technology
CN109377561A (en) A kind of digital-to-analogue surface mesh generation method based on conformal geometry
Farin et al. Geometric Modeling
CN105160700B (en) A kind of cross section curve reconstructing method for reconstructing three-dimensional model
CN106981097A (en) A kind of T spline surface approximating methods based on subregion Local Fairing weight factor
CN107730587A (en) One kind is based on picture quick three-dimensional Interactive Modeling method
CN116522496A (en) Automatic three-dimensional airfoil quadrilateral mesh generation method based on symmetric direction field
CN108230452B (en) Model hole filling method based on texture synthesis
CN110689620B (en) Multi-level optimized grid surface discrete spline curve design method
CN110188395A (en) A kind of dimension increase formula calculating fluid grid generation method based on line face body
CN111079353B (en) Method and device for rapid and uniform grid division applied to complex fluid analysis
Wu et al. A global approach to multi-axis swept mesh generation
CN103942397A (en) Shape-correction gear digital modeling method based on power function
CN111274732A (en) Grid repairing method based on &#39;connection relation-position&#39; iterative optimization
Ye et al. Constructive modeling of G1 bifurcation
CN108876922A (en) A kind of mesh amending method based on the regularization of interior dihedral angle supplementary angle
CN112085832B (en) Advertisement word 3D model generation method
CN107798732A (en) A kind of free form surface configuration control method
CN117172399B (en) Automatic wire laying track planning method based on heuristic algorithm
CN112967333B (en) Complex point cloud skeleton extraction method and system based on grading
CN117593485B (en) Three-dimensional model simplifying method and system based on Haoskov distance perception

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant