Search

CN-122020866-A - Method and system for optimizing air-structure coupling data transmission of aircraft

CN122020866ACN 122020866 ACN122020866 ACN 122020866ACN-122020866-A

Abstract

The invention belongs to the technical field of aerodynamic/structural design of aircrafts, and discloses an aerodynamic-structural coupling data transmission optimization method and system of an aircrafts, wherein the method comprises the steps of constructing a k-d tree according to initial structural grid node coordinates, calculating aerodynamic-structural gap dimensions and structural grid gap dimensions by using k neighbor queries, and determining self-adaptive neighborhood and self-adaptive nuclear width; the method comprises the steps of carrying out load-displacement transmission to obtain structural grid node load and pneumatic grid node displacement, deforming pneumatic grid node coordinates according to the structural grid node load obtained by the round, carrying out structural displacement and pneumatic load solving by the round if the load distribution uniformity reaches the standard, otherwise, determining self-adaptive neighborhood and self-adaptive kernel width again until the load distribution uniformity reaches the standard, and returning to start a new round of iteration until the iteration ending condition is reached. The invention obviously improves the spatial uniformity of the structural side load field and the smoothness of the displacement field, and improves the numerical robustness and engineering applicability of the load transmission process.

Inventors

  • LIU YAOLONG
  • ZHENG ZIJIE
  • ZHENG YAO

Assignees

  • 浙江大学

Dates

Publication Date
20260512
Application Date
20260414

Claims (10)

  1. 1. A method for optimizing aircraft aerodynamic-structural coupling data transfer, comprising: respectively constructing an initial pneumatic grid node coordinate and an initial structure grid node coordinate based on the geometric data of the aircraft, completing first-round pneumatic load solving according to the initial pneumatic grid node coordinate, and initializing the structure displacement to be zero; constructing a k-d tree according to the node coordinates of the initial structural grid, inquiring and calculating the aerodynamic-structural gap scale and the structural grid gap scale by using k neighbor, and determining the self-adaptive neighborhood and the self-adaptive kernel width; Load-displacement transmission is carried out based on the self-adaptive neighborhood and the self-adaptive kernel width, so that structural grid node load and pneumatic grid node displacement are obtained; Judging whether the load distribution uniformity meets the standard according to the structural grid node load obtained by the wheel, deforming the pneumatic grid node coordinates if the load distribution uniformity meets the standard, and carrying out structural displacement and pneumatic load solving of the wheel; And returning to carry out load-displacement transmission to start a new iteration until reaching the iteration ending condition, and outputting the pneumatic grid node coordinates, pneumatic load and structural displacement.
  2. 2. The method of optimizing aircraft aero-structural coupling data transfer of claim 1, wherein said calculating aero-structural gap dimensions and structural grid pitch dimensions using k-nearest neighbor queries comprises: For each pneumatic grid node, inquiring the distance between the k near structural grid nodes by using a k-d tree, and taking the median of the distances inquired by all the pneumatic grid nodes as the pneumatic-structural gap scale; and querying the distances of the k+1 near structural grid nodes before using the k-d tree for each structural grid node, and taking the median of the distances queried by all the structural grid nodes as the structural grid interval scale.
  3. 3. The method of aircraft aero-structural coupling data transfer optimization of claim 1, wherein said determining an adaptive neighborhood and an adaptive kernel width comprises: Constructing a dimensionless gap ratio according to the ratio of the aerodynamic-structural gap dimension to the structural grid spacing dimension; determining an adaptive neighborhood based on the reference neighborhood and the dimensionless gap ratio comprises the steps of when the dimensionless gap ratio tends to be 1, the adaptive neighborhood tends to be the reference neighborhood, and when the dimensionless gap ratio increases, the adaptive neighborhood is based on the reference neighborhood according to the following steps of And tends to the neighborhood maximum value, when the dimensionless gap ratio is reduced, the adaptive neighborhood is based on the reference neighborhood according to And tends to be neighborhood minimum, wherein, In order to obtain the non-dimensional gap ratio, Is a neighborhood sensitivity factor; The adaptive kernel width is determined based on the aerodynamic-structure gap dimension, including decreasing the adaptive kernel width when the aerodynamic-structure gap dimension increases and increasing the adaptive kernel width when the aerodynamic-structure gap dimension decreases.
  4. 4. The aircraft aero-structure coupled data transfer optimization method of claim 3, wherein the adaptive neighborhood is determined as Wherein In order to adapt the neighborhood to be used, As a reference neighborhood, As a neighborhood minimum value, As a neighborhood maximum value, As a function of Round, Is a clipping function.
  5. 5. The aircraft aero-structure coupled data transfer optimization method of claim 3, wherein the adaptive core width is determined as Wherein In order to adapt the core width to the size of the core, In order to be a kernel-wide sensitivity factor, Is the dimension of the gap between the pneumatic structure and the gap between the pneumatic structure, Is a numerical security term that prevents denominators from becoming zero.
  6. 6. The method for optimizing the transmission of aerodynamic-structural coupling data of an aircraft according to claim 1, wherein the load-displacement transmission is performed based on an adaptive neighborhood and an adaptive kernel width, and wherein the adopted weight function is a gaussian function, and the gaussian weight of the gaussian function is determined as follows: Constructing an initial Gaussian weight of each structural grid node in the self-adaptive neighborhood according to the self-adaptive kernel width; and carrying out normalization processing on the initial Gaussian weights according to the initial Gaussian weights of all the structural grid nodes in the self-adaptive neighborhood to obtain the Gaussian weights of the structural grid nodes.
  7. 7. The method for optimizing air-structure coupling data transmission of an aircraft according to claim 1, wherein the step of determining whether the load distribution uniformity meets the standard according to the structural grid node load obtained by the present wheel comprises the following steps: taking the modular length of the structural grid node load, and performing data cleaning, sequencing and mean value calculation; Based on the sequencing result, taking the ratio of the p50 load module length to the p95 load module length as a quantile ratio; calculating standard deviation and a coefficient of the base coefficient of the load module length according to the sequencing result; If the fractional number ratio, the standard deviation and the coefficient of the Kernel are respectively smaller than the corresponding threshold values, judging that the load distribution uniformity reaches the standard, otherwise, judging that the load distribution uniformity does not reach the standard.
  8. 8. The aircraft aero-structure coupled data transfer optimization method of claim 3, wherein the redefining the adaptive neighborhood and the adaptive kernel width comprises: The reference neighborhood and the neighborhood sensitivity factor are adjusted, and the adjustment logic is that the first priority continuously increases the reference neighborhood until reaching the neighborhood maximum value, and the second priority continuously increases the neighborhood sensitivity factor until reaching the neighborhood sensitivity factor maximum value.
  9. 9. The aircraft aero-structure coupled data transfer optimization method of claim 5, wherein the redefining the adaptive neighborhood and the adaptive kernel width comprises: The reference neighborhood, the neighborhood sensitivity factor and the kernel width sensitivity factor are adjusted, and the adjustment logic is that the first priority continuously increases the reference neighborhood until reaching the neighborhood maximum value, the second priority continuously increases the neighborhood sensitivity factor until reaching the neighborhood sensitivity factor maximum value, and the third priority continuously increases the kernel width sensitivity factor until reaching the kernel width sensitivity factor maximum value.
  10. 10. An aircraft aero-structure coupled data transfer optimization system, comprising: The multi-disciplinary coupling module is used for respectively constructing initial pneumatic grid node coordinates and initial structural grid node coordinates based on the geometric data of the aircraft, completing first-round pneumatic load solving according to the initial pneumatic grid node coordinates, initializing structural displacement to be zero, and carrying out structural displacement and pneumatic load solving in subsequent iteration; The geometric preprocessing module is used for constructing a k-d tree according to the node coordinates of the initial structural grid, and calculating the pneumatic-structural gap dimension and the structural grid spacing dimension by utilizing k neighbor query; the parameter self-adaption module is used for determining a self-adaption neighborhood and a self-adaption kernel width; The load-displacement transmission module is used for carrying out load-displacement transmission based on the self-adaptive neighborhood and the self-adaptive nuclear width to obtain the load of the structural grid node and the displacement of the pneumatic grid node; the load transfer effect evaluation module is used for judging whether the load distribution uniformity meets the standard according to the structural grid node load obtained by the wheel, deforming the pneumatic grid node coordinates if the load distribution uniformity meets the standard, and carrying out structural displacement and pneumatic load solving of the wheel; and the iteration output module is used for returning to the load-displacement transmission module to start a new iteration of load-displacement transmission until reaching the iteration ending condition and outputting the pneumatic grid node coordinates, the pneumatic load and the structural displacement.

Description

Method and system for optimizing air-structure coupling data transmission of aircraft Technical Field The invention belongs to the technical field of aircraft pneumatic/structural design, and particularly relates to an aircraft pneumatic-structural coupling data transmission optimization method and system. Background In high-fidelity pneumatic-structural coupling design and optimization, a numerical coupling mode is often adopted to transfer load and displacement between a pneumatic grid and a structural grid. Because the two types of grids are generally inconsistent in geometric position, grid scale and topological structure, the following two types of load transmission technical routes are generally adopted in engineering: 1. the global interpolation type method based on Radial Basis Function (RBF) takes pneumatic nodes and structural nodes as point sets, and field quantity mapping is realized by constructing a global interpolation matrix taking the radial basis function as a core. The basic structure is that RBF basis function matrixes are built on all selected nodes, interpolation coefficients are solved, so that the pressure or displacement field on the pneumatic grid can be represented by linear combination of the basis functions, and corresponding load or displacement values are calculated on the positions of the structural nodes by utilizing the interpolation representation. The working principle of the method is based on global smooth interpolation, higher interpolation precision can be provided theoretically, but the interpolation matrix is a dense matrix, the dimension is directly related to the number of nodes, and in order to reduce the calculation scale, the node sub-sampling or representative point selection strategy is often combined in engineering. 2. The load-displacement extrapolation (MELD) method based on matching belongs to the local weighted least squares fitting method. The basic structure is that for each pneumatic node, a plurality of structure nodes in the neighborhood are selected to form a local point set, and the number of the neighborhood nodes is as followsThe method comprises the steps of setting according to experience by a user in advance, adopting a weighted least square method to fit a rigid motion (a rotation matrix and a translation vector) in the local neighborhood, mapping structural side displacement to pneumatic nodes to construct a displacement transfer operator, and deducing a corresponding load transfer operator from a displacement transfer relation through a virtual work consistency principle to realize conservation of force and moment. In the MELD method, a Gaussian weight function is generally adopted, the weight is attenuated along with the distance between the nodes of the pneumatic-structure, and the weight attenuation coefficient is reducedIs empirically set in advance. The working principle of the method is that under the assumption of local approximate rigid motion, the consistency of force, moment conservation and virtual work in the transmission process is ensured through weighted fitting. Although the existing method is mature in the aspects of force conservation, moment conservation and virtual work consistency, the following defects and reasons still exist under the complex geometric and engineering application scenes: 1. The lack of quantitative description and control on load distribution uniformity is that the existing methods and documents generally take total force errors, total moment errors and virtual work errors as main evaluation indexes, and the spatial distribution quality of the load on the structural side among nodes lacks unified measurement and design targets. Even if the conservation and virtual work are consistent, it may still occur that a few nodes are subjected to extreme loads far above the overall level, resulting in non-physical stress hot spots and numerical oscillations. The root cause is that statistics such as a coefficient of Kernel, a fractional bit ratio (p 95/p 50), a standard deviation and the like are not introduced in the prior art to explicitly evaluate and restrict the load uniformity, so that the key performance of the load uniformity is in an unmodeled and uncontrolled state for a long time. 2. The neighborhood size and weight attenuation parameters of the MELD method are fixed empirical values, and the geometric adaptivity is lacking, namely the number of neighborhood nodes in the traditional MELD methodAnd weight decay coefficientTypically, the configuration is uniform over the entire wing or structure, and is not explicitly related to factors such as aerodynamic-structural gap size, structural grid local density, and curved geometry. Fixed and bigger in the area with smaller gap and denser gridEasily causes excessive smooth load, is fixed and smaller in a region with larger gaps or sparse gridsAnd the rigid body fitting matrix is possibly sick, so that the numerical value is unstable. Also, fixThe weight at