A method for customizing catheter tube design

Abstract

The application discloses a customized conduit tube shape optimization design method, which is used for solving the problem that the conduit needs to be cold calibrated, filed and sampled during engine assembly due to poor size consistency of a liquid rocket engine. The method comprises the following steps: firstly, measuring an engine conduit interface; then, assembling a conduit sample and the interface as an initial state of an optimization model; taking the coordinate values of each node of the conduit as variables, taking the conduit similarity as a target function, taking the distance and position requirements of the conduit from the engine and other components as constraint conditions, and solving by using a genetic algorithm to obtain the coordinate values of each node of the conduit, so that a complete tube shape is obtained. The method can quickly design the engine conduit tube shape meeting the requirements, avoids the processes of shape calibration, filing, on-site sampling and the like during engine assembly, and improves the precision and efficiency of the conduit production.

Description

TECHNICAL FIELD

[0001] The present application relates to a customized conduit shape optimization design, more particularly to a customized conduit shape optimization design method for a liquid rocket engine. BACKGROUND

[0002] The pipeline system is an important part of the liquid rocket engine, but due to poor product size consistency and other reasons, a large number of batched conduits cannot be normally matched, and the conduit direction and conduit interface need to be shaped and repaired by the experience of the assembly personnel according to the actual situation of the engine, and some conduits need to be sampled on site for bending production. The process of manual shaping and repairing and on-site sampling is extremely dependent on personnel experience, with high labor intensity and low efficiency; secondly, manual repair or sampling will cause differences in the direction of each conduit, which is not conducive to improving the overall consistency of the engine, so the assembly of the pipeline system has become a bottleneck link for the improvement of the overall assembly capacity of the liquid rocket engine. SUMMARY

[0003] In view of the defects and deficiencies in the prior art described above, the present application aims to overcome the technical problems of manual repair or sampling of the liquid rocket engine conduit, and provides a conduit shape optimization design method. The method first measures the engine conduit interface and constraint features, then assembles the conduit sample and the interface in the CAD software as the initial state of the optimization model, takes the conduit node coordinate values as variables, takes the conduit similarity as the objective function, takes the distance, position requirements and other constraints of the conduit from the engine and other components as constraints, solves it using genetic algorithm, obtains the conduit node coordinate values, and obtains the optimal conduit shape.

[0004] The technical scheme adopted by the present application to achieve the above-mentioned purpose is:

[0005] A customized conduit shape optimization design method, comprising the following steps:

[0006] Measurement stage: using a measuring device to measure the interface of the customized conduit of the liquid rocket engine, the position constraint features of the conduit and the pre-installed components on the engine, and extracting the measured data of the first and last end interfaces of the conduit and their position constraint features;

[0007] Design stage: according to the measured data of the customized conduit interface and its position constraint features, assembling the conduit sample and optimizing the shape to avoid interference with the pre-installed components, and obtaining the optimal conduit shape of the customized conduit.

[0008] The measured data of the first and last end interfaces of the conduit and the position constraint features includes the following steps:

[0009] Fitting the center and normal data of the cross section circle according to the first and last end interfaces of the conduit;

[0010] The preformed part is abstracted as a cylinder or a sphere, and the axis data of the cylinder and / or the center data of the sphere are extracted according to the position constraint features.

[0011] The design stage comprises the following steps:

[0012] Step 1: respectively establishing model instances of the conduit sample, the actually measured interface and the position constraint features; assembling the conduit and the interface, so that the straight line segments at the first end and the last end of the conduit respectively pass through the two interfaces; the conduit sample is the initial theoretical form of the conduit design;

[0013] Step 2: initializing the conduit shape, driving the key points on the axes of the two ends of the conduit, adjusting the running posture of the straight line segments at the two ends of the conduit, so that the straight line segments at the two ends of the conduit are respectively collinear with the normal lines of the two interfaces;

[0014] Step 3: establishing a mathematical optimization model, adjusting the key point coordinates of the conduit to change the shape and posture of the conduit under the condition of satisfying the spatial constraints, finding the optimal solution for making the designed conduit model infinitely approach the conduit sample form, and obtaining the optimal pipe type;

[0015] Step 4: outputting the optimal pipe type of the conduit, converting the included angle of the straight line pipe segment containing the extension line of the pipe segment into a bend angle again, and establishing a conventional solid model of the conduit.

[0016] The mathematical optimization model for the customized conduit pipe shape design comprises the following steps:

[0017] 1) Design variables

[0018] The conduit pipe shape is abstracted as a curve segment connected by multiple straight line segments, and the conduit has n key points, so there are n-1 straight line segments and n-2 bend angles;

[0019] 2) Objective function

[0020] In order to avoid interference between the designed conduit and surrounding parts, and to make the similarity between the designed conduit and the conduit sample as high as possible, the conduit sample model and the optimized model are translated and rotated to make the two best aligned, and the mean value of the position deviation of each corresponding point is taken as the similarity index as the objective function;

[0021] 3) Constraint conditions

[0022] The change range of the length of each pipe segment of the conduit should be within the control range;

[0023] The spacing requirements of the conduit and other preformed part features on the engine; the spacing requirements include two types: the distance between the straight line segment of the conduit and the point constraint feature on the engine and the distance between the straight line segment of the conduit and the axis of the preformed part on the engine;

[0024] The spatial constraint range of each key point coordinate.

[0025] The customized catheter tube shape design model is a mathematical model as follows:

[0026] Find x=(x1,x2,...,x 3(n-4)+2 )

[0027]

[0028] In the formula, x=(x1,x2,...,x 3(n-4)+2 ) is 3(n-4)+2 design variables, x 3(n-4)+2 represents the length l n-1 of the last line segment; The coordinates of the key points i of the optimized catheter model, is the coordinates of the i-th point of the coordinate set P ORT , P ORT is the original point coordinate set P O of the catheter, and P G is the coordinate set after alignment transformation; wherein d j represents the distance of the catheter from the j-th feature, d jmin represents the minimum required value of the distance; d lk represents the distance of the catheter from the k-th component axis, d lkmin represents the minimum required value of the distance; x kmin and x kmax are the upper and lower limits of the coordinate value x k , respectively.

[0029] The transformation relationship between the point set P ORT after alignment transformation and the original point coordinate set P O of the catheter is as follows:

[0030] P ORT =R*P O +T

[0031] In the formula, R represents a rotation matrix, and T represents a translation matrix. The best alignment is realized by singular value decomposition (SVD) on the original point coordinate set P O of the catheter and the coordinate set P G after shape adjustment of the catheter, so as to obtain the best rigid transformation of the catheter sample and the optimized model.

[0032] The mathematical model is solved by a genetic algorithm to obtain an optimal catheter model.

[0033] The present application has the following beneficial effects and advantages:

[0034] 1. Achieve the collaborative production of liquid rocket engine ducts, improve the production efficiency of ducts. Traditional customized duct production needs cold correction, filing or on-site sampling, which consumes a lot of time. The invention can quickly obtain the relevant parameters of the engine, automatically generate the duct design scheme according to the preset algorithm and model, greatly shorten the time cycle from demand to design completion.

[0035] 2. Improve the production accuracy of the duct. The invention can conduct accurate duct design based on the accurate parameters of the engine, such as interface data, constraint component position, etc. Ensure that the duct and other components of the engine are accurately matched, and improve the overall performance and reliability of the engine. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 Schematic diagram for measuring point cloud and feature extraction;

[0037] Figure 2 Schematic diagram for modeling interface and constraint features;

[0038] Figure 3 Schematic diagram for the assembly model of the duct sample and the interface;

[0039] Figure 4 Schematic diagram for duct sample shape initialization;

[0040] Figure 5 Schematic diagram for the optimal duct model;

[0041] Figure 6 Schematic diagram for duct rounding;

[0042] Figure 7 Schematic diagram for the three-dimensional duct. DETAILED DESCRIPTION

[0043] The invention will be further described in detail below in combination with the drawings and implementation examples.

[0044] The customized duct shape design includes two stages of engine feature measurement stage and duct shape design stage. In the measurement stage, the actual measurement of the liquid rocket engine duct interface features and constraint features is carried out by the measuring equipment, and the feature data is extracted as the input data of the customized duct shape design system. The measuring equipment used in this method is a three-dimensional laser scanner, which obtains the point cloud of the feature part, and then extracts the feature data. In the design stage, the actual data measured on site is accepted, the twin model of the duct interface and the constraint is established through the secondary development of the CAD software, and then the optimization design of the customized duct shape is carried out to obtain the optimal shape of the customized duct.

[0045] A customized duct shape design method includes the following steps:

[0046] Measurement stage: as Figure 1As shown, the interface and constraint features of the customized pipe of the liquid rocket engine are scanned and point clouds are obtained using a laser scanning device, and the measured data of the customized pipe interface and constraint features are extracted from the point clouds.

[0047] Design stage: as shown Figures 2-7 , the measured data of the customized pipe interface and constraint features are used to assemble the pipe sample and perform shape optimization to obtain the optimal pipe shape of the customized pipe.

[0048] The measurement stage includes the following steps:

[0049] The pipe interface and constraint features on the engine are measured using a measuring device to obtain point clouds.

[0050] The center data and normal data of the pipe interface, the axis data and center data of the constraint features are extracted.

[0051] The first interface coordinates of the pipe are (x S ,y S ,z S ), and the direction vector is The second interface coordinates are (x E ,y E ,z E ), and the direction vector is

[0052] The coordinates of the point (or ball) features on the engine are:

[0053]

[0054] In the formula, P j represents the coordinates of the jth point feature, represents the coordinate components of the jth constraint feature.

[0055] The axis of the cylindrical feature on the engine is extracted as:

[0056]

[0057] In the formula, P lk is the vector representation of a point on the axis of the engine component, is the direction vector of the axis.

[0058] The design stage includes the following steps:

[0059] Step 1: Establish the assembly model of the pipe sample and the interface. The pipe sample is the initial shape of the pipe, which is generally the theoretical shape during engine design. As shown Figure 2 , Figure 3As shown, the model of the catheter sample and the model of the measured interface, constraint and the like are respectively established by secondary development of CAD software; the catheter and the interface and the like are assembled in the assembly to make the first and the last linear segments of the catheter pass through the two interfaces respectively.

[0060] Step two: initialization of the shape of the catheter sample. As shown, the key points at the two ends of the catheter sample are driven, and the shape of the linear segments at the two ends of the catheter sample is adjusted to make the linear segments at the two ends of the catheter sample respectively collinear with the directions of the two interfaces. Figure 4

[0061] Step three: adjustment of the shape of the catheter and search for the optimal solution. The coordinates of the key points of the catheter are adjusted in the computer-aided design system to change the shape and posture of the catheter, so that the similarity between the designed catheter model and the catheter sample is as high as possible under the condition of meeting the spatial constraints, and the optimal tube shape is obtained by establishing a mathematical optimization model.

[0062] The customized catheter tube shape design mathematical optimization model comprises the following four parts:

[0063] (I) Design variables

[0064] Suppose that the catheter has n key points, then there are n-1 linear segments and n-2 angles. 3(n-4)+2 design variables x=(x1, x2, …, xn) are set. 3(n-4)+2

[0065] Where x1, x2, … x 3(n-4) represent the component coordinates of the key points 3 to the key points n-2 in turn x 3(n-4)+1 represents the length of the first linear segment of the catheter l1, x 3(n-4)+2 represents the length of the last linear segment l n-1 .

[0066] Since the interface position and the direction vector are known, the first linear segment and the last linear segment of the catheter pass through the catheter interface points, and the direction is consistent with the interface direction. The points 2 and the points (n-1) are determined by the lengths of the linear segments l1 and l n-1 , and the coordinates of the key points 2 and the key points (n-1) can be represented by the coordinates of the key points 1 and the key points n respectively.

[0067]

[0068]

[0069] (II) Objective function

[0070] After the translation and rotation of the catheter sample model and the optimized model, the two models are best aligned, and then the average of the sum of the position deviations of the corresponding points is calculated as the similarity index. ​​

[0071] P is the initial coordinate of the key point i of the catheter phantom model. O is the matrix representation of the key point set of the catheter phantom model. P is the coordinate of the key point i of the optimized catheter model. G is the matrix representation of the key point set of the optimized catheter model. Wherein P O is the catheter phantom parameter, which is a known quantity.

[0072]

[0073] The best rigid transformation of the catheter phantom and the optimized model is obtained by singular value decomposition (SVD).

[0074]

[0075] [U, S, V] = SVD(H)

[0076] R = VU T

[0077] T = -R * centroid O + centroid G

[0078] Wherein centroid O and centroid G are the average centers of the point set P O and the point set P G ; H is the covariance matrix.

[0079] P ORT = R * P O + T

[0080] Wherein P ORT is the coordinate set P O of the catheter after shape adjustment, and P G is the coordinate set after alignment. The similarity is defined as the average distance of the corresponding points between the point set P ORT and the point set P G .

[0081]

[0082] Wherein is the coordinate value of the i-th point in the point set P ORT .

[0083] The objective function is set as:

[0084] f(x) = φ(x) ​

[0085] (iii) Constraints

[0086] Constraint 1: The variation of the length of each conduit segment should be within the control range.

[0087]

[0088] l i denotes the nominal length of the i-th conduit segment, denotes the float range of the i-th conduit segment.

[0089] Constraint 2: The clearance requirement of the conduit from other features on the engine.

[0090] The clearance requirement of the conduit from a point feature on the engine:

[0091] d jmin ≤ d j

[0092] where d j denotes the clearance of the conduit from the j-th feature, d jmin denotes the minimum required value of the clearance.

[0093] The vector distance formula d j from a point to a line can be expressed as:

[0094]

[0095] where P j is the vector representation of the point feature on the engine, A j is the vector representation of a point on the conduit line segment, is the direction vector of the j-th conduit segment. The cross product of vectors is denoted by x, and the magnitude of a vector is denoted by ||.

[0096] The distance constraint of a straight conduit segment from an axis of a component on the engine is:

[0097] d lkmin ≤ d lk

[0098] where d lk denotes the clearance of the conduit from the k-th component axis, d lkmin denotes the minimum required value of the clearance.

[0099] The distance between two straight lines can be calculated by the following formula:

[0100]

[0101] where P lk and A lkare the vector representations of the engine upper part axis and any point on the corresponding straight line segment of the catheter respectively, |·| represents the length of the vector.

[0102] wherein are the directional vectors of the two straight lines The cross product of

[0103]

[0104] Constraint condition three: the spatial constraint range of each key point coordinate:

[0105]

[0106] wherein x kmin and x kmax are the upper and lower limits of the coordinate value x k

[0107] (Four) Mathematical optimization model

[0108] The customized catheter tube shape design model can be simplified as the following mathematical model:

[0109] Find x=(x1,x2,...,x 3(n-4)+2 )

[0110]

[0111] The above mathematical model is solved by genetic algorithm to obtain the optimal catheter model.

[0112] Step four: output the catheter CAD model. For the designed catheter model, the straight line pipe segment angle containing the pipe segment extension line is converted into the bend angle again, and the conventional solid model of the catheter is established.

[0113] The above-described specific embodiments further illustrate the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above-described specific embodiments are merely examples of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.​

Claims

1. A customized catheter tube shape optimization design method, characterized in that: The following steps are involved: Measurement phase: Use measurement equipment to measure the interfaces of the customized liquid rocket engine duct, the positional constraints between the duct and pre-installed components on the engine, and extract measured data on the interfaces at both ends of the duct and their positional constraints; Design phase: Based on the measured data of the customized catheter interface and its position constraint characteristics, the catheter sample is assembled and its shape is optimized to avoid interference with prefabricated components and obtain the optimal tube shape of the customized catheter; The mathematical optimization model for customized catheter tube shape design includes the following steps: 1) Design variables The catheter shape is abstracted as a curve segment formed by connecting multiple straight line segments. If the catheter has n key points, there are n-1 straight line segments and n-2 bends. 2) Objective function To avoid interference between the designed catheter and surrounding parts and to maximize the similarity between the designed catheter and the sample catheter, the sample catheter model and the optimized model were translated and rotated using singular value decomposition (SVD) to achieve optimal alignment. The mean of the position deviations of the corresponding points was used as the similarity index and the objective function. 3) Constraints The variation range of the length of each section of the catheter should be within the control range; Spacing requirements between the duct and other prefabricated features on the engine; these spacing requirements include two types: the distance between the duct straight segment and the point constraint feature on the engine, and the distance between the duct straight segment and the axis of the prefabricated component on the engine; The spatial constraint range of each key point coordinate.

2. A customized catheter shape optimization design method according to claim 1, characterized in that: The method of extracting measured data of the interfaces and position constraint features at both ends of the catheter comprises the following steps: Fit the center and normal data of the cross-section circle of the interface at both ends of the catheter; The prefabricated component is abstracted as a cylinder or a sphere, and the axis data of the cylinder and / or the center data of the sphere are extracted according to its position constraint features.

3. A customized catheter shape optimization design method according to claim 1, characterized in that: The design phase includes the following steps: Step 1: Create model instances of the catheter sample, the measured interface, and the position constraint features; assemble the catheter and the interface so that the straight segments at the beginning and end of the catheter pass through the interfaces on both sides; the catheter sample is the initial theoretical form of the catheter design; Step 2: Initialize the shape of the catheter, drive the key points on the axis of the two ends of the catheter, and adjust the direction of the straight segments at both ends of the catheter so that the straight segments at both ends of the catheter are collinear with the normals of the two interfaces; Step 3: Establish a mathematical optimization model. Under the condition of satisfying spatial constraints, adjust the coordinates of the key points of the catheter to change the shape and posture of the catheter, and find the optimal solution that makes the designed catheter model infinitely close to the shape of the catheter sample, thus obtaining the optimal tube shape; Step 4: Output the optimal tube shape of the catheter, convert the straight tube segment angle including the tube segment extension line into a curved angle again, and establish a conventional solid model of the catheter.

4. A customized catheter shape optimization design method according to claim 1, characterized in that: The customized catheter tube shape design model is the following mathematical model: Find x=(x1,x2,...,x 3(n-4)+2 ) Where x=(x1,x2,…,x 3(n-4)+2 ) is 3(n-4)+2 design variables, x 3(n-4)+2 Represents the length of the last line segment l n-1 ; The coordinates of the key point i of the catheter model after optimization, is the coordinate set P ORT The coordinates of the i-th point, P ORT is the original point coordinate set P of the catheter O The coordinate set P after adjustment with the catheter shape G Align the transformed coordinate set; where d j represents the distance between the catheter and the jth feature, d jmin Indicates the minimum required value of the spacing value; d lk represents the distance between the conduit and the axis of the kth component, d lkmin Indicates the minimum required value of the spacing; x kmin and x kmax The coordinate values ​​are x k upper and lower limits.

5. A customized catheter shape optimization design method according to claim 4, characterized in that: The point set P after the alignment transformation ORT and the catheter original point coordinate set P O The transformation relationship is: P ORT =R*P O +T Where R represents the rotation matrix, T represents the translation matrix, and the original point coordinate set P of the catheter is O The coordinate set P after adjusting the catheter shape G Singular value decomposition (SVD) was performed to achieve optimal alignment, thereby obtaining the optimal rigid transformation of the catheter sample and the optimized model.

6. A customized catheter shape optimization design method according to claim 4, characterized in that: The mathematical model is solved by a genetic algorithm to obtain an optimal catheter model.

Images