CN107451378A - A kind of three-dimensional coordinates measurement blade profile samples point extracting method - Google Patents
A kind of three-dimensional coordinates measurement blade profile samples point extracting method Download PDFInfo
- Publication number
- CN107451378A CN107451378A CN201710789615.0A CN201710789615A CN107451378A CN 107451378 A CN107451378 A CN 107451378A CN 201710789615 A CN201710789615 A CN 201710789615A CN 107451378 A CN107451378 A CN 107451378A
- Authority
- CN
- China
- Prior art keywords
- point
- points
- sampled
- sampling
- blade profile
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/002—Measuring arrangements characterised by the use of optical techniques for measuring two or more coordinates
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Numerical Control (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of three-dimensional coordinates measurement blade profile to sample point extracting method, and this method includes：Section is intercepted to blade to be measured；Etc. arc length discrete cross section curve, and front and rear edge point fitting full accuracy circle is extracted, ask friendship to obtain waypoint with the layering section；Blade profile curve is divided into by four leading edge, blade back, leaf basin, trailing edge regions according to the waypoint spacing；The arc length such as use sampling in the region, the initial samples point position data of front and rear edge is extracted respectively, characteristic distance threshold value is set, using 3 points as a small-sized firefly group, when front and rear distance between two points on curve is more than or equal to the threshold value, realize that sampled point refreshes automatically according to curvature variation sampling location in the front and rear edge region, sampled point spacing sum and original arc length are contrasted in iterative process, it is determined whether retain the sampled result after refreshing.
Description
Technical field
The invention belongs to three coordinate machine field of measuring technique, and in particular to a kind of three-dimensional coordinates measurement blade profile sampled point carries
Take method.
Background technology
Blade is aero-engine, gas turbine, the important component of steam turbine, and its crudy directly affects aviation hair
The overall performance of motivation and life-span.It is a kind of accurate Blade measuring method using coordinate measuring machine (CMM) detection, and conventional survey
Amount method is compared, and possesses the features such as high measurement accuracy, error visualization, automaticity is high.Due to the complexity of blade in itself
It is required that the big measurement sampled point of extraction, possesses higher measurement difficulty.Usual coordinate measuring machine measurement efficiency is low, there is an urgent need to
The method that sampled point quantity keeps sampled point precision again can be reduced.The follow-up measurement data evaluation of blade establishes cross section information
On, handled for ease of three coordinate measuring engine measurement Track Pick-up and follow-up data, evaluating is calculated, it is therefore desirable to which blade is entered
The contour cross section-wise of row is sampled.The present invention proposes the adaptively sampled method of blade profile based on improvement glowworm swarm algorithm,
On the one hand can customize cloth to count out, another aspect sampled point is distributed in strict accordance with Curvature varying trend, to improving Blade measuring
Efficiency achieves noticeable achievement.
Common blade profile configuration sampling point extracting method mainly has：Equal arc method, Isoparametric method, curvature arc length are adaptive
Should.Equal arc method distribution is most uniform, easily causes blade front and rear edge Curvature varying large area sampling site and is distributed excessively sparse, curvature
It is excessively intensive to change the distribution of smaller area sampling site.Isoparametric method by parameter curve according to preset parameter range distribution sampled point, if
Front and rear edge region parameter section is too small equally to easily cause that sampling site is sparse, influences the precision and data assessment of measurement data reconstruct
Effect.Curvature arc in patent " CN201510379312.2_ Lanzhou University of Science ＆ Technologys, one kind become the adaptively sampled method of arc length "
Long adaptive approach, which exists, points to the problem of single, generally according to same directional curvature change profile, while with iterations
Increase, easily causing end points punishment cloth distortion and data redundancy reduces computational efficiency.The subject matter of the above-mentioned method of sampling is：
(1) uniform sampled point can not reflect actual physics model surface feature.(2) generally there is sampling site efficiency and sampling precision phase
The problem of mutually restricting, it is to improve sampling precision most efficient method to improve sampling site quantity, but can also reduce systematic sampling effect simultaneously
Rate, increase redundant data.
Glowworm swarm algorithm proposed by Cambridge scholar Yang in 2008, and its is right for the brightness decision of every firefly in algorithm
The Attraction Degree size of other fireflies of surrounding, brightness is more high then higher to the Attraction Degree of other fireflies, conversely then easily by it
The guiding of his high brightness firefly, moved to its direction.When the displacement and direction for judging firefly, first
Target firefly, i.e. other fireflies of brightness highest are found within its search radius, according to suction of other fireflies to it
Degree of drawing size judges direction and the distance of firefly movement.It is used for solving such as traveling salesman problem (Traveling
Salesman Problem, TSP) optimum path problems.Such as document " Li Mingfu, horse Jian Hua, Zhang Yuyan etc..Based on discrete firefly
Free-float space robot Sequence Planning [J] computer integrated manufacturing systems of fireworm algorithm, 2014,20 (11):2719-2727.”
The middle thinking for proposing to solve Blade measuring anchor point shortest path known to process using glowworm swarm algorithm, solves scattered data being
Point sequencing problem, this method have the shortcomings that amount of computing repeatedly is big, are suitable only for known point measurement Sequence Planning problem.Patent
" CN201110257951.3_ Harbin Engineering Universitys, a kind of naval vessel paths planning method based on glowworm swarm algorithm " proposes base
In the naval vessel trajectory planning algorithm of multi-target glowworm swarm algorithm；Patent " CN201210251782.7_ Harbin Engineering Universitys, one
Paths planning method of the kind based on multi-target glowworm swarm algorithm " proposes the robot obstacle-avoiding track based on multi-target glowworm swarm algorithm
Planning algorithm, both of which between known Origin And Destination to unknown path by calculate it is optimal in a manner of generate, too chase after
Seek globally optimal solution, local path has to be optimized, and it is applied to the prediction in unknown path, is not particularly suited for known curve and surface
Path point location generation." CN201610815269.4_ Harbin Institute of Technology, the big auspicious new and high technology of speeding of Harbin work have patent
Limit company, the PF space non-cooperative target orbital predictions method based on firefly group optimization " is proposed according to glowworm swarm algorithm pair
The certain point of known spatial locations calculates with other institute's Attraction Degrees a little, renewal and the point position of its Attraction Degree maximum, realizes rail
Road position prediction, relative to space certain point other positions point calculating not only algorithm complex it is higher and data update
During easily cause data corruption, inventive algorithm remains the order of sequence, and optimized algorithm greatly reduces computing
Amount, for the section feature of blade, the generation of primary study local sampling point and position optimization, and global conditions are set,
The distribution of global optimum can be expeditiously obtained on the basis of this.
Measured for known CAD model, need to obtain the enough model surface information in advance before CMM measurements, acquisition
Model information significant degree will directly affect follow-up measurement process.Sampled point is more high to model reduction precision, after being more beneficial for
Continuous actual measurement.However, the problem of CMM measurement efficiencies and precision are conflicts, the present invention proposes a kind of new blade sampling
Algorithm, it is intended to solve the problems, such as sampled point number it is less on the premise of sampled point to blade reduce precision it is not high enough, sampling number
Mesh can improve measurement flexibility through user configuration, reach the purpose of control sampled result, and final realize controls measurement result
Purpose.On the basis of point position does not determine, propose first by the way of block sampling region, the arc length mode such as recycling slightly carries
Initial samples point is taken, the improvement glowworm swarm algorithm of blade profile sampling is met with reference to each sample area characteristic Design.Improve the light of firefly
Worm algorithm by initial samples point in region it is discrete be small-sized firefly group, calculate sampling point position and curvature information in each group
Algorithm is substituted into, realizes that key point position automatically updates according to Curvature varying in group.The core of algorithm is the base determined in points
On plinth, optimum sampling position is determined in the range of region of search with the interaction relationship put by traveling through point, traversal generates each
Locally optimal solution, the final locally optimal solution are the optimal curvatures characteristic point in the region.The purpose of algorithm is desirable to lead to
Cross and obtain sampled point effective enough with strengthening follow-up corresponding measurement point validity in sample phase and reduce follow-up measurement data
The difficulty of processing.
The content of the invention
Present invention aim to address above mentioned problem, there is provided a kind of e measurement technology is simple, and measurement is counted out less, restoring data
A kind of high three-dimensional coordinates measurement blade profile sampling point extracting method of model accuracy.
In order to solve the above technical problems, the technical scheme is that：A kind of three-dimensional coordinates measurement blade profile sampled point carries
Take method, it is characterised in that comprise the following steps：
S1, leaf model data import, and import leaf model and store the data of leaf model；
S2, leaf model data hierarchy simultaneously store, by the data storage of the contour curve of blade profile；
S3, extraction cross-section data are simultaneously segmented, and mark off blade profile front and rear edge and leaf basin, the waypoint of blade back；
S4, the arc length such as the leaf basin blade back in blade profile are sampled；
S5, the leading edge trailing edge in blade profile is equidistantly sampled and optimizes sampled result；
S6, planning survey path, calculate normal vector of the sampled point on curved surface, and anchor point and rollback are asked for by biasing
Point, obtain the path planned；
S7, judge to check whether path with curve has intersection point, return to step S4 re-starts sampling if it intersection point be present；
S8, sampling terminate.
Preferably, the extraction cross-section data in the step S3 and it is segmented, centroid point is asked for by minimum circumscribed circle method,
Ask two points maximum with centroid point distance to be designated as leading edge estimated position points, trailing edge estimated position points respectively, estimate position in leading edge
Put a little and trailing edge estimated position points nearby search for the point composition front and rear edge point set of multiple adjacent positions respectively, before and after extraction
The parameter of the data digital simulation circle of edge point set, the number of change extraction point asks for the fitting circle under different points, in fitting essence
Highest point set is spent, maximum 2 points of extraction wherein consecutive points spacing is used as waypoint, and blade profile includes leading edge, leaf basin, blade back with after
Edge region.
Preferably, in the step S3, sampled point number can freely be set in each region, according to arc length such as set fixed-point numbers
Sampling, simplify calculating process.
Preferably, the step S5 is further comprising the steps of：
S51, setting characteristic distance threshold value, continuous 3 points of selection is one group of small-sized firefly group, calculates respectively at 3 points
The arc length of curvature and consecutive points；
S52, calculate before and after 2 points fluorescent value with and relative front and rear 2 fluorescent values of intermediate point；
S53, by contrasting front and rear 2 points of fluorescent value and the fluorescent value of front and rear 2 points relatively of intermediate point respectively, obtain among
Point moving direction simultaneously asks for displacement, renewal current point position by calculating Attraction Degree；
S54, to move front position point as the center of circle, with to be moved to rear location point intermediate position points make circle with section molded line
Friendship is asked, obtains the position of corresponding transfer point on curve, calculates the point after movement and source location spacing；
S55, updated the data if the distance between the point and source location after mobile is more than characteristic distance threshold value, after renewal
Location point is directly applied in next group of data calculating, until blade profile front and rear edge sampling is finished, single iteration mistake
The consecutive points distance that added up in journey calculates, and accumulation result relatively and determines sampling precision with actual arc length, judges whether to retain accordingly
Iteration sampled result.
Preferably, normal vector of the sampled point on curved surface bucket, normal bias sampling point position are calculated in the step S6
One chaining pin radius distance generates centre of sphere measurement point, continues outwards to generate measurement rollback point and avoidance point along normal vector direction,
Measurement track is generated according to described position data, the Intersection of checking track and curved surface is added in algorithm, if in the presence of intersecting
Step S4 is retracted, threshold value is changed, recalculates and extract sampled point, generation measurement track, until measurement track and blade surface
Without interference.
The beneficial effects of the invention are as follows：
1st, the present invention meets sampled point number in each sectional area for the extraction of blade sampled point and freely set, and reaches control
The purpose of sampled result processed, and on the premise of same sampling site number, sampled result is better than other sampling algorithms.
2nd, sampled point of the invention is set fewer, and sampling dot profile reduction precision and traditional algorithm contrast are higher.
3rd, the present invention is attracted each other rule using sampled point by amount of curvature, and curvature is relatively gradually leaned on a little louder compared with dot to curvature
Hold together, adjust automatically renewal sampling location, extraction possesses the point generation sampling point set of optimal adaptation degree, solves and becomes arc according to curvature
The long adaptive sampling site order easily occurred points to the problem of single, and enormously simplify calculating process, improves sampling efficiency.
4th, the application of improvement glowworm swarm algorithm realizes the cluster in regional space around each curvature characteristic point, basic herein
On generation point possess blade profile curve feature, be beneficial to follow-up collision detection and Surface Parameters and evaluate.
5th, it is different from glowworm swarm algorithm, the curvature Variation of blade profile molded line is combined, improves glowworm swarm algorithm,
Applied to sampling point location, this bionic localization method possesses certain learning and adapting capability, makes sampled result reduced form
Line precision is higher, is advantageous to the processing of follow-up measurement data.
Brief description of the drawings
Fig. 1 is the main flow chart of the present invention；
Fig. 2 is the design specific steps flow chart of the present invention；
Fig. 3 is the flow chart of step S5 in Fig. 1 of the present invention；
Fig. 4 is the blade profile sampled point extraction schematic flow sheet of the present invention；
Fig. 5 is importing model layers schematic diagram in the embodiment of the present invention；
Fig. 6 is that schematic cross-section is extracted in the embodiment of the present invention；
Fig. 7 is the section waypoint schematic diagram generated in the embodiment of the present invention；
Fig. 8 is the section sampled point schematic diagram generated in the embodiment of the present invention；
Fig. 9 is the measurement track schematic diagram of sampled point in the embodiment of the present invention；
Figure 10 is the single section gauge trajectory diagram generated in the embodiment of the present invention；
Figure 11 is the comparative result figure that the present invention uses algorithms of different.
Embodiment
The present invention is described further with specific embodiment below in conjunction with the accompanying drawings：
As shown in Fig. 1 to Figure 11, a kind of three-dimensional coordinates measurement blade profile sampling point extracting method provided by the invention, including
Following steps：
As depicted in figs. 1 and 2, S1, leaf model data import, and import leaf model and store the data of leaf model；
The surface parameter equation of modeling blade is expressed in the form of nurbs surface：
P in formula (1)i,jFor control point, wi,jFor weight, N shared by each control pointi,p(u) it is p B-spline function of u direction,
Ni,q(v) it is q B-spline function of u direction.The result that step S1 is finally run, as shown in Figure 5.
S2, leaf model data hierarchy simultaneously store, by the data storage of the contour curve of blade profile；
Establish space plane equation：
F (x, y, z)=Ax+By+Cz+D=0 (2)
The intersection point of Calculation Plane and spoon of blade, nurbs curve is taken to be fitted intersection point in the present embodiment：
P in formula (3)iFor control point, wiFor weight, N shared by each control pointi,k(u) it is k B-spline basic function of u direction.
By leaf model data hierarchy and store, as a result as shown in Figure 6.
S3, extraction individual-layer data are simultaneously segmented, and mark off blade front and rear edge and leaf basin, the waypoint of blade back；
Sampled point number can freely be set in each region, be sampled according to arc length such as set fixed-point numbers, simplify calculating process.Carry
Take individual-layer data and be segmented, centroid point is asked for by minimum circumscribed circle method, seek two points maximum with centroid point distance respectively
Leading edge estimated position points P1 and P4, trailing edge estimated position points P2 and P3 are designated as, in leading edge estimated position points and trailing edge estimated location
Point nearby searches for the point composition front and rear edge point set of multiple adjacent positions respectively, is calculated and intended by the data of the front and rear edge point set of extraction
The parameter of circle is closed, the number of change extraction point asks for the fitting circle under different points, in fitting precision highest point set, extracts it
Maximum 2 points of middle consecutive points spacing is used as waypoint, and blade back is formed between P1 and P2, and P2 and P3 form trailing region, P3 and P4 structures
Into leaf basin, P4 and P1 form front edge area, and blade profile includes front edge area, leaf basin, blade back and trailing region.
It the arc length mode such as can use to calculate curve C (u) discrete samplings after step S2 processing bent first here
Line total length：
A discrete point number N is defined, unit arc length d can be calculated
Unit arc length inverse iteration such as length of curve is calculated into formula and asks for the subsequent point parameter value t based on current location ui：
According to the method described above by cross section curve according to etc. arc length mode be separated into original point set.Utilize the institute in this point set
It is a little calculating centroid point (xo,yo), two points maximum with centroid point distance as x ＜ 0, x ＞ 0 are sought respectively, around 2 points
N points, generation point set (x are taken respectivelyi,yi), if disk equation is：
xi 2+yi 2+αxi+βyi+ γ=0 (8)
The coefficient of best-fit-circle is calculated when sampling site number is n：
Order
Simultaneously iteration asks for the optimal circle of wherein fitting result to the step of changing n value repetition above-mentioned fitting circle：
P (α, β, γ)=Min [Qn(α,β,γ)] (10)
As shown in fig. 7, making the point set of formula (10) establishment by extraction, then distance between two consecutive points is calculated, will wherein
Result of calculation two points of maximum are designated as waypoint, can obtain waypoint on cross section curve and be designated as P1, P2, P3, P4 respectively.
S4, the arc length such as the leaf basin blade back in blade are sampled；
Using leaf basin blade back waypoint [P1, P2] [P3, P4] spacing be far longer than front and rear edge waypoint [P1, P4] [P2,
P3] spacing the characteristics of, leaf basin blade back curve is obtained indirectly by way of calculating the distance of adjacent sectional point, using in step S2
Etc. arc length sample mode can obtain the sampled point of leaf basin blade back.
As shown in figure 3, the leading edge trailing edge in blade is equidistantly sampled by step S5 and optimizes sampled result；
Step S5 is further comprising the steps of：
S51, setting characteristic distance threshold value, continuous 3 points of selection is one group of small-sized firefly group, calculates respectively at 3 points
The arc length of curvature and consecutive points；
S52, calculate before and after 2 points fluorescent value with and relative front and rear 2 fluorescent values of intermediate point；
S53, by contrasting front and rear 2 points of fluorescent value and the fluorescent value of front and rear 2 points relatively of intermediate point respectively, obtain among
Point moving direction simultaneously asks for displacement, renewal current point position by calculating Attraction Degree；
S54, to move front position point as the center of circle, with to be moved to rear location point intermediate position points make circle with section molded line
Friendship is asked, obtains the position of corresponding transfer point on curve, calculates the point after movement and source location spacing；
S55, updated the data if the distance between the point and source location after mobile is more than characteristic distance threshold value, after renewal
Location point is directly applied in next group of data calculating, until blade profile front and rear edge sampling is finished, single iteration mistake
The consecutive points distance that added up in journey calculates, and accumulation result relatively and determines sampling precision with actual arc length, judges whether to retain accordingly
Iteration sampled result.
Search radius r and characteristic distance threshold epsilon are set, similarly according to the method in S2 by the arc length such as front and rear edge curve from
Dissipate, calculate the curvature of each discrete point：
Algorithm for design thinking of the present invention is that the curvature of point is bigger, and other small points of curvature are then drawn close to it, realize point coordinates
Automatically update.Fluorescein of the glowworm swarm algorithm using κ as each point is introduced, head and the tail waypoint remains stationary as.
Define firefly brightness：
I in formula (12)0For ri,jThe original intensity of firefly when=0, I is defined as in inventive algorithm0=κ, β table
Show the absorption coefficient of light, meet β ∈ [0 ,+∞), typically take β ∈ [0.1,10] in engineer applied, β=1, r taken in this algorithmi,jTable
Show the distance of two adjacent fireflies.It is continuous choose before etc. arc length it is discrete go out 3 points be one group of small-sized firefly group, distinguish
3 points of curvature κ 1, κ 2, κ 3 and consecutive points arc length L1, L3 are calculated, 2 points of fluorescent value before and after further calculatingWithWith relatively front and rear 2 fluorescent values of intermediate pointWithBy contrasting I respectively1With I2, 1I3With I,2I2With1I, obtain intermediate point moving direction.
Define Attraction Degree γi,j(r)：
γ in formula (13)0For maximum Attraction Degree, usual γ is typically made in engineer applied0=1, ri,jBetween expression firefly
Distance.
ri,j=| | Xj(t)-Xi(t)|| (14)
T represents iterations, X in formula (14)i(t) current location of current i-th of firefly is represented, in iterative process
Location updating：
Xi(t+1)=Xi(t)+γi,j(ri,j)[Xj(t)-Xi(t)]/2+αηi/2 (15)
α is arbitrary width in formula (15), typically takes α ∈ [0,1], ηiObey standardized normal distribution.
Glowworm swarm algorithm is moved with rectilinear direction, with reference to the cross section profile feature of blade, realize current point curve from
Dynamic location updating.With Xi(t) asked for for the center of circle with γi,j(ri,j)[Xj(t)-Xi(t)]/2+αηi/ 2 for radius circle in movement side
Upwards with the intersection point X of region inner section linet'(t+1)。
Calculate characteristic distance li：
If li>=ε, then update and store current point position；According to above-mentioned derivation, the inventive method principle specifically may be used
It is described as：Since second point compared with front and rear 2 curvature, if found in the search radius of second point brighter than its
Spend high point then to move to it, moving step length is according to current location with adaptively adjusting subsequent point positional distance on moving direction
It is whole, be sequentially completed location updating a little, treated that a location updating finishes, accumulation calculating is per adjacent 2 points distance, contrast
Original arc length, ask for the sampling point position of Optimal Distribution in an iterative process by means of which.
S6, planning survey path, calculate normal vector of the sampled point on curved surface, and anchor point and rollback are asked for by biasing
Point, obtain the path planned；
As shown in Figure 9 and Figure 10, by above step, constructed complete cross-section sample point data.Each point can be asked for
The normal vector on curved surface, normal bias obtain avoidance point and rollback point position generation measurement track.
The solution of the present invention is concluded can as follows：It is first directed to leaf model；Then section is obtained with equal altitude method；
Then realize that blade profile is segmented, cross section profile is divided into four leading edge, trailing edge, leaf basin, blade back parts；According still further to leaf basin blade back with before
The difference of edge trailing edge Curvature varying, the leaf basin blade back directly arc length mode such as use is sampled, arc length side is being waited to leading edge and trailing edge
Realize that the less point of curvature is drawn close automatically to the big point of curvature according to Curvature varying on the premise of formula sampling, it is lasting to refresh sampling knot
Fruit, until output optimum sampling point position；Finally utilize curved surface normal bias calculating and plotting measurement track.This method uses the light of firefly
Worm algorithm the point of arc length distribution such as drives according to curvature adjust automatically, makes a position integrated distribution in the larger region of curvature, significantly
Reduce redundant points quantity, each point is possessed validity, to point contact type measurement pattern and scanning measurement pattern all
With applicability.
The layering curve map of blade as shown in Figure 5 and Figure 6, reasonably by molded line curve segmentation be leading edge, trailing edge, leaf basin and
Blade back region, demonstrate the validity of this method.This method as shown in Figure 4 specifically generates the flow chart of sampled point, is described in detail
The process of the point self-adapted distribution of sampling.The sampling point distributions figure of generation as shown in figure 8, leaf basin blade back region according to etc. arc length side
Formula is uniformly distributed, and front and rear edge region adapts to be distributed in strict accordance with amount of curvature.If Fig. 9 and Figure 10 are the rails according to sampled point generation
Mark figure, with blade surface without interference.The present invention has made algorithm contrast respectively to 40,60,120,160 sampled points respectively, according to
Sampled point is generated each other away from accumulation calculating and original arc length difference evaluation algorithm superiority, is divided and is listed each algorithm profile calculating knot
Fruit is as shown in table 1.Shown in Figure 11, according to the result of calculation of table 1, profile errors pair have been carried out with equal arc method and Isoparametric method respectively
Than point quantity is fewer, and this algorithm action effect is more obvious.
The Experimental comparison results of table 1
The present invention's extracts blade profile sampled point based on improved glowworm swarm algorithm, can be not only used for three-dimensional coordinates measurement
Sampled point extracts, and after improving sample density, change PATH GENERATION is equally applicable to the sampling of optical scanner formula three-dimensional coordinates measurement
Point extraction.Calculate normal vector of the sampled point on curved surface bucket, the one chaining pin radius distance generation of normal bias sampling point position
Centre of sphere measurement point, continue outwards to generate measurement rollback point and avoidance point along normal vector direction, given birth to according to described position data
Into measurement track, the Intersection of checking track and curved surface is added in algorithm, step S4 is retracted if in the presence of intersecting, changes threshold value,
Recalculate and extract sampled point, generation measurement track, until measurement track with blade surface without interference.
S7, judge to check whether path with curve has intersection point, return to step S4 re-starts sampling if it intersection point be present；
S8, sampling terminate.
As shown in figure 11, it is compared by the result obtained by algorithms of different, used by the present embodiment obtained by method
As a result precision highest.
One of ordinary skill in the art will be appreciated that embodiment described here is to aid in reader and understands this hair
Bright principle, it should be understood that protection scope of the present invention is not limited to such especially statement and embodiment.This area
Those of ordinary skill can make according to these technical inspirations disclosed by the invention various does not depart from the other each of essence of the invention
The specific deformation of kind and combination, these deform and combined still within the scope of the present invention.
Claims (5)
1. a kind of three-dimensional coordinates measurement blade profile samples point extracting method, it is characterised in that comprises the following steps：
S1, leaf model data import, and import leaf model and store the data of leaf model；
S2, leaf model data hierarchy simultaneously store, by the data storage of the contour curve of blade profile；
S3, extraction cross-section data are simultaneously segmented, and mark off blade profile front and rear edge and leaf basin, the waypoint of blade back；
S4, the arc length such as the leaf basin blade back in blade profile are sampled；
S5, the leading edge trailing edge in blade profile is equidistantly sampled and optimizes sampled result；
S6, planning survey path, calculate normal vector of the sampled point on curved surface, ask for anchor point and rollback point by biasing, obtain
To the path planned；
S7, judge to check whether path with curve has intersection point, return to step S4 re-starts sampling if it intersection point be present；
S8, sampling terminate.
2. a kind of three-dimensional coordinates measurement blade profile sampling point extracting method according to claim 1, it is characterised in that described
Extraction layering in step S3 and is segmented cross-section data, and centroid point is asked for by minimum circumscribed circle method, is asked respectively and centroid point
Two maximum points of distance are designated as leading edge estimated position points, trailing edge estimated position points, estimate in leading edge estimated position points and trailing edge
Location point nearby searches for the point composition front and rear edge point set of multiple adjacent positions respectively, passes through the data meter of the front and rear edge point set of extraction
The parameter of fitting circle is calculated, the number of change extraction point is asked for the fitting circle under different points, in fitting precision highest point set, carried
Wherein maximum 2 points of consecutive points spacing is taken to be used as waypoint, blade profile includes leading edge, leaf basin, blade back and trailing region.
3. a kind of three-dimensional coordinates measurement blade profile sampling point extracting method according to claim 1, it is characterised in that described
In step S3, sampled point number can freely be set in each region, be sampled according to arc length such as set fixed-point numbers, simplify calculating process.
4. a kind of three-dimensional coordinates measurement blade profile sampling point extracting method according to claim 1, it is characterised in that described
Step S5 is further comprising the steps of：
S51, setting characteristic distance threshold value, continuous 3 points of selection is one group of small-sized firefly group, calculates 3 points of curvature respectively
With the arc length of consecutive points；
S52, calculate before and after 2 points fluorescent value with and relative front and rear 2 fluorescent values of intermediate point；
S53, by contrasting front and rear 2 points of fluorescent value and the fluorescent value of front and rear 2 points relatively of intermediate point respectively, obtain intermediate point and move
Simultaneously displacement, renewal current point position are asked for by calculating Attraction Degree in dynamic direction；
S54, to move front position point as the center of circle, make circle with the intermediate position points of the location point to after after movement and asked with section molded line
Hand over, obtain the position of corresponding transfer point on curve, calculate the point after movement and source location spacing；
S55, updated the data if the distance between the point and source location after mobile is more than characteristic distance threshold value, position after renewal
Point is directly applied in next group of data calculating, until blade profile front and rear edge sampling is finished, during single iteration
Cumulative consecutive points distance calculates, and accumulation result relatively and determines sampling precision with actual arc length, judges whether to retain iteration accordingly
Sampled result.
5. a kind of three-dimensional coordinates measurement blade profile sampling point extracting method according to claim 1, it is characterised in that described
Normal vector of the sampled point on curved surface bucket, the one chaining pin radius distance generation of normal bias sampling point position are calculated in step S6
Centre of sphere measurement point, continue outwards to generate measurement rollback point and avoidance point along normal vector direction, given birth to according to described position data
Into measurement track, the Intersection of checking track and curved surface is added in algorithm, step S4 is retracted if in the presence of intersecting, changes threshold value,
Recalculate and extract sampled point, generation measurement track, until measurement track with blade surface without interference.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710789615.0A CN107451378B (en) | 2017-09-05 | 2017-09-05 | Three-coordinate measuring blade section sampling point extraction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710789615.0A CN107451378B (en) | 2017-09-05 | 2017-09-05 | Three-coordinate measuring blade section sampling point extraction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107451378A true CN107451378A (en) | 2017-12-08 |
| CN107451378B CN107451378B (en) | 2021-01-05 |
Family
ID=60495183
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710789615.0A Active CN107451378B (en) | 2017-09-05 | 2017-09-05 | Three-coordinate measuring blade section sampling point extraction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107451378B (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108917687A (en) * | 2018-04-26 | 2018-11-30 | 太原理工大学 | A kind of blade of aviation engine front and rear edge slight camber detection method |
| CN109458963A (en) * | 2018-12-07 | 2019-03-12 | 中国航空工业集团公司济南特种结构研究所 | A kind of method of determining radome shunt bar characteristic point spatial position |
| CN109542106A (en) * | 2019-01-04 | 2019-03-29 | 电子科技大学 | A kind of paths planning method under mobile robot multi-constraint condition |
| CN109635322A (en) * | 2018-11-05 | 2019-04-16 | 武汉华锋惠众科技有限公司 | A kind of method that automobile panel process complementary surface section line is arranged automatically |
| CN109711376A (en) * | 2018-12-29 | 2019-05-03 | 重庆邮电大学 | A kind of multiple dimensioned sparse blue noise method of sampling based on optimal transmission theory |
| CN111060057A (en) * | 2019-12-25 | 2020-04-24 | 贵阳航发精密铸造有限公司 | Turbine blade profile measuring method based on three-coordinate measuring machine |
| CN111400667A (en) * | 2020-03-31 | 2020-07-10 | 华中科技大学 | Aviation blade profile detection method and system based on variable tolerance zone constraint |
| CN112015138A (en) * | 2020-09-04 | 2020-12-01 | 中国工程物理研究院机械制造工艺研究所 | Blade contour error evaluation method based on K nearest neighbor iterative nearest grid algorithm |
| CN112446123A (en) * | 2019-08-28 | 2021-03-05 | 电子科技大学 | Measuring head pose planning method for blisk three-coordinate measuring machine |
| CN113111405A (en) * | 2021-04-22 | 2021-07-13 | 大连大学 | NURBS curve fitting method based on improved second-order oscillation PSO algorithm |
| CN113970311A (en) * | 2021-10-12 | 2022-01-25 | 中国航空工业集团公司北京长城计量测试技术研究所 | Aero-engine blade vector approximation iterative measurement method |
| CN113989221A (en) * | 2021-10-26 | 2022-01-28 | 电子科技大学 | Abrasive belt grinding blisk collision detection method based on improved octree partitioning method |
| CN114048558A (en) * | 2021-10-26 | 2022-02-15 | 西北工业大学 | Method for modeling blade profile of gas compressor with non-uniform profile error |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1800777A (en) * | 2004-12-31 | 2006-07-12 | 上海萨克斯动力总成部件系统有限公司 | Three-coordinate measuring method and apparatus for measuring complicated curved surface |
| CN101149253A (en) * | 2007-10-16 | 2008-03-26 | 浙江大学 | Unknown free-form surface self-adaptive measuring method based on exploration method and measuring head device |
| CN103411574A (en) * | 2013-08-14 | 2013-11-27 | 西北工业大学 | Aviation engine blade profile three-coordinate measuring method |
| CN104392488A (en) * | 2014-12-11 | 2015-03-04 | 福州大学 | Automatic point cloud data rectification method aiming at laser scanner and three-coordinate measuring arm |
| CN105045973A (en) * | 2015-07-02 | 2015-11-11 | 兰州理工大学 | Variable-arc-length adaptive sampling method |
-
2017
- 2017-09-05 CN CN201710789615.0A patent/CN107451378B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1800777A (en) * | 2004-12-31 | 2006-07-12 | 上海萨克斯动力总成部件系统有限公司 | Three-coordinate measuring method and apparatus for measuring complicated curved surface |
| CN101149253A (en) * | 2007-10-16 | 2008-03-26 | 浙江大学 | Unknown free-form surface self-adaptive measuring method based on exploration method and measuring head device |
| CN103411574A (en) * | 2013-08-14 | 2013-11-27 | 西北工业大学 | Aviation engine blade profile three-coordinate measuring method |
| CN104392488A (en) * | 2014-12-11 | 2015-03-04 | 福州大学 | Automatic point cloud data rectification method aiming at laser scanner and three-coordinate measuring arm |
| CN105045973A (en) * | 2015-07-02 | 2015-11-11 | 兰州理工大学 | Variable-arc-length adaptive sampling method |
Non-Patent Citations (3)
| Title |
|---|
| LI CHAO等: "Calculation Method of Twist Error of Aero-Engine Blade Based on", 《INTERNATIONAL CONFERENCE ON MECHANICS AND MECHANICAL ENGINEERING.》 * |
| 彭志光: "基于改进凸包算法的叶片型面检测", 《中国优秀硕士学位论文全文数据库工程科技Ⅰ辑》 * |
| 李明富等: "基于离散萤火虫算法的自由曲面测量序列规划", 《计算机集成制造系》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108917687A (en) * | 2018-04-26 | 2018-11-30 | 太原理工大学 | A kind of blade of aviation engine front and rear edge slight camber detection method |
| CN108917687B (en) * | 2018-04-26 | 2020-08-04 | 太原理工大学 | Method for detecting tiny cambered surfaces of front edge and rear edge of blade of aero-engine |
| CN109635322A (en) * | 2018-11-05 | 2019-04-16 | 武汉华锋惠众科技有限公司 | A kind of method that automobile panel process complementary surface section line is arranged automatically |
| CN109458963A (en) * | 2018-12-07 | 2019-03-12 | 中国航空工业集团公司济南特种结构研究所 | A kind of method of determining radome shunt bar characteristic point spatial position |
| CN109711376A (en) * | 2018-12-29 | 2019-05-03 | 重庆邮电大学 | A kind of multiple dimensioned sparse blue noise method of sampling based on optimal transmission theory |
| CN109542106A (en) * | 2019-01-04 | 2019-03-29 | 电子科技大学 | A kind of paths planning method under mobile robot multi-constraint condition |
| CN112446123B (en) * | 2019-08-28 | 2022-12-30 | 电子科技大学 | Measuring head pose planning method for blisk three-coordinate measuring machine |
| CN112446123A (en) * | 2019-08-28 | 2021-03-05 | 电子科技大学 | Measuring head pose planning method for blisk three-coordinate measuring machine |
| CN111060057A (en) * | 2019-12-25 | 2020-04-24 | 贵阳航发精密铸造有限公司 | Turbine blade profile measuring method based on three-coordinate measuring machine |
| CN111060057B (en) * | 2019-12-25 | 2022-01-28 | 贵阳航发精密铸造有限公司 | Turbine blade profile measuring method based on three-coordinate measuring machine |
| CN111400667B (en) * | 2020-03-31 | 2021-11-02 | 华中科技大学 | Aviation blade profile detection method and system based on variable tolerance zone constraint |
| CN111400667A (en) * | 2020-03-31 | 2020-07-10 | 华中科技大学 | Aviation blade profile detection method and system based on variable tolerance zone constraint |
| CN112015138B (en) * | 2020-09-04 | 2021-08-10 | 中国工程物理研究院机械制造工艺研究所 | Blade contour error evaluation method based on K nearest neighbor iterative nearest grid algorithm |
| CN112015138A (en) * | 2020-09-04 | 2020-12-01 | 中国工程物理研究院机械制造工艺研究所 | Blade contour error evaluation method based on K nearest neighbor iterative nearest grid algorithm |
| CN113111405A (en) * | 2021-04-22 | 2021-07-13 | 大连大学 | NURBS curve fitting method based on improved second-order oscillation PSO algorithm |
| CN113111405B (en) * | 2021-04-22 | 2023-08-29 | 大连大学 | NURBS curve fitting method based on improved second-order oscillation PSO algorithm |
| CN113970311A (en) * | 2021-10-12 | 2022-01-25 | 中国航空工业集团公司北京长城计量测试技术研究所 | Aero-engine blade vector approximation iterative measurement method |
| CN113989221A (en) * | 2021-10-26 | 2022-01-28 | 电子科技大学 | Abrasive belt grinding blisk collision detection method based on improved octree partitioning method |
| CN114048558A (en) * | 2021-10-26 | 2022-02-15 | 西北工业大学 | Method for modeling blade profile of gas compressor with non-uniform profile error |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107451378B (en) | 2021-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107451378A (en) | A kind of three-dimensional coordinates measurement blade profile samples point extracting method | |
| GB2547816B (en) | Actually-measured marine environment data assimilation method based on sequence recursive filtering three-dimensional variation | |
| Praveen et al. | Low cost PSO using metamodels and inexact pre-evaluation: Application to aerodynamic shape design | |
| CN103679807B (en) | A kind of dispersion point cloud reconstructing method of band boundary constraint | |
| CN109683552B (en) | Numerical control machining path generation method on complex point cloud model guided by base curve | |
| CN109360272B (en) | Grid curved surface curve design method based on distance constraint | |
| CN102682172B (en) | Numerous-parameter optimization design method based on parameter classification for supercritical aerofoil | |
| CN103714577A (en) | Three-dimensional model simplification method suitable for model with textures | |
| CN113393577B (en) | Oblique photography terrain reconstruction method | |
| CN102800114B (en) | Data point cloud downsizing method based on Poisson-disk sampling | |
| CN109343468A (en) | A kind of blade multiaxis orbit generation method based on projection biasing | |
| CN106960102A (en) | A kind of space linearity assessment method based on secondary glowworm swarm algorithm of climbing the mountain | |
| Júnior et al. | Intelligent data-driven aerodynamic analysis and optimization of morphing configurations | |
| CN116306044A (en) | Uncertainty analysis method of full turbulence configuration and gradient optimization design method thereof | |
| KR102582665B1 (en) | System and method for evaluating patterns of integrated circuit | |
| CN105869210A (en) | Interpolation data processing method in three-dimensional geological surface model | |
| CN109671152B (en) | Line point cloud geometric characteristic estimation method based on Gaussian weighted discrete derivative | |
| CN108875141B (en) | Method for determining chip full-mask focusing parameters based on neural network model | |
| CN105869209A (en) | Deformed triangular data processing method in three-dimensional geological surface model | |
| CN113111553A (en) | Interpolation deformation grid-based large deformation motion numerical simulation method | |
| CN113281903B (en) | Simple and efficient iterative solution method for intersection point position of light and optical free-form surface | |
| CN105426626A (en) | Similar data style cluster based multiple-point geostatistics modeling method | |
| CN110163973A (en) | A kind of zone boundary ring method for fairing | |
| CN113449448B (en) | Layout optimization method for optical distributed measurement system | |
| CN105894580A (en) | Method for processing curved surface extension data in three-dimensional geological surface model |
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 |