Optimization method for deep sea diverter valve body topology
Patent Information
- Application Number
- CN202610875715.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-06-17
- Publication Date
- 2026-09-04
AI Technical Summary
传统阀体多沿用陆用或浅海场景的固定结构,难以满足深海装备小型化需求,还易因未适配深海流固耦合载荷特性产生应力集中,极端工况下形变量超标破坏配合间隙,叠加粘滞阻尼力、稳态液动力等随海深增大的影响
1.现有深海耐压结构拓扑优化方法,以结构整体柔度最小化为目标,约束条件仅设置体积分数上限,不涉及局部关键区域的变形控制,优化后可能会在配合圆柱面等局部位置出现超标变形,无法预测高压下因配合间隙破坏导致的阀芯卡滞。本发明通过构建全局柔度约束与局部抗变形的协同优化模型,首次在深海阀体拓扑优化中实现了整体轻量化与局部抗变形的同步控制,优化后阀体在11000m海深工况下,配合圆柱面变形量可严格控制在阈值内,有效避免了因形变破坏配合间隙导致的阀芯卡滞情况出现。
Smart Images

Figure CN122693349A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of lightweight design technology for hydraulic components, and specifically to an optimization method for the valve body topology of a deep-sea directional valve. Background Technology
[0002] The three-position four-way solenoid directional valve is a core control component of deep-sea hydraulic systems. Its valve body must withstand the increased resistance caused by the viscosity and density of the oil as the sea depth increases under extreme high pressure environments at all ocean depths, and maintain precise matching with the valve core to achieve jam-free directional switching. Traditional valve bodies mostly use fixed structures from land-based or shallow-sea scenarios, which are difficult to meet the miniaturization requirements of deep-sea equipment. They are also prone to stress concentration due to not being adapted to the fluid-structure interaction load characteristics of deep seas. Under extreme working conditions, excessive deformation can damage the fitting clearance, compounded by the effects of viscous damping forces and steady-state hydrodynamic forces that increase with sea depth.
[0003] Existing methods for designing deep-sea directional valve bodies suffer from three main problems: First, existing methods use minimizing structural flexibility as the sole objective and the upper limit of volume fraction as the sole constraint, without incorporating a local deformation control mechanism. The failure mode of deep-sea directional valves lies in the fact that under the hydrostatic pressure at a full ocean depth of 11,000 m, the valve body undergoes overall shrinkage deformation. If the radial displacement at the mating cylindrical surface of the valve core exceeds the design value of the mating clearance, it is typically on the order of 10-20 μm. The essence of this failure mode is the disruption of the mating relationship caused by local deformation. Therefore, given that ocean depth factors push the deformation of the mating cylindrical surface to the same order of magnitude as the clearance, optimizing solely with global flexibility as the objective may result in a pseudo-optimal solution where global stiffness is satisfied but local deformation of the mating cylindrical surface exceeds the limit. Second, the mating cylindrical surface of the valve core is a cylindrical surface, and its radial displacement response is the result of the coordination of displacements at all nodes on the surface domain. Existing methods can only constrain the displacement of a single node or point, failing to reflect the rigid body motion coordination relationship of multiple displacements on the mating cylindrical surface. Therefore, even if the single-point displacement is acceptable, the ellipticization of the mating cylindrical surface can still cause valve core jamming. Third, after simultaneously introducing global compliance constraints and mating cylindrical surface deformation constraints, the volume fraction threshold and constraint feasibility no longer have a simple monotonic relationship. Since this relationship cannot be analytically predicted, a single preset f value is very likely to fall into the infeasible region or suboptimal region, increasing the uncertainty of the optimization results.
[0004] Furthermore, the optimized structures produced by existing methods often contain non-manufacturable features such as tiny narrow slits and closed cavities, which are insufficient for synergy with advanced processes such as additive manufacturing, potentially weakening the optimization performance. Therefore, a topology optimization method for deep-sea directional valve bodies is needed, capable of simultaneously controlling global stiffness and local mating cylindrical surface deformation at the topology optimization level, and systematically searching for the optimal volume fraction, taking into account full-ocean-depth adaptability, lightweight design, resistance to jamming deformation, and manufacturability. Summary of the Invention
[0005] To address the aforementioned problems, the present invention aims to provide an optimization method for the topology of a deep-sea directional valve body. This method incorporates the deformation of the mating cylindrical surfaces of the valve body and valve core, along with the global flexibility constraints of the valve body, into the optimization formula. This ensures that material distribution in each iteration is simultaneously driven by the overall stiffness gradient of the valve body and the radial displacement gradient of the mating cylindrical surfaces. RBE2 rigid elements are used to establish the displacement coordination relationship between the master and slave nodes of the mating cylindrical surfaces, compressing the multi-point coupled displacement field on the mating cylindrical surfaces into a weighted average radial displacement. This allows the criterion for the deformation of the mating cylindrical surfaces of the valve body and valve core to participate in sensitivity analysis and iterative solutions, achieving a balance between valve body lightweighting and resistance to jamming deformation at the topology optimization level. Furthermore, since the volume fraction threshold under dual constraints cannot be analytically predicted, this method determines the optimal configuration of the valve body by performing parallel calculations of multiple f-values and selecting the best option, improving the global optimality and engineering reliability of the optimization results.
[0006] Specifically, this invention provides a method for optimizing the topology of a deep-sea directional valve body, the specific implementation steps of which include: S1. Obtain deep-sea operating condition data of the directional valve body, establish a discretized mesh model of the directional valve body, and assign a relative density variable ρ. i The interpolation relationship between the relative density variable and the elastic modulus in the discretized mesh model of the valve body is established by using the solid isotropic material penalty method, and the displacement coordination relationship between the valve body and the valve core mating cylindrical surface is established.
[0007] S2. Minimize the structural flexibility of the valve body as the optimization objective and construct a structural flexibility constraint model for the valve body. Based on the working environment of the directional valve body in the deep sea, the volume constraint of the valve body, the static equilibrium equation of the valve body, and the density limit of the valve body are used as constraint conditions.
[0008] S3. For the local deformation failure mode of the valve core mating cylindrical surface, the radial displacement u of the Ns slave nodes on the mating cylindrical surface in the discretized mesh model of the valve body obtained in step S1 is calculated. r,j Let j = 1, 2, ..., Ns. Using the displacement of the principal node M, the deformation constraint model of the mating cylindrical surface of the valve body and valve core in the directional valve is constructed as follows: ; Among them, w j U is the weighting coefficient. r,j Let J be the radial displacement at the j-th node; Based on the deformation constraint model of the mating cylindrical surface of the valve body and valve core in the reversing valve, the deformation constraint of the mating cylindrical surface is defined as follows: ; in, This is the upper limit of the allowable radial displacement of the mating cylindrical surface on the valve body.
[0009] S4. Based on the valve body structure flexibility constraint model obtained in step S2 and the valve body and valve core mating cylindrical surface deformation constraint model obtained in step S3, construct the valve body optimization model and optimization criteria.
[0010] S5. Set K different volume fraction thresholds. Utilizing the sensitivity of the valve body structure flexibility and the sensitivity to deformation of the cylindrical surface The optimization criteria for composition apply to the volume fraction threshold of each group. k By performing iterative double-constraint topology optimization, the candidate configuration sequence {Ω1, Ω2, ..., Ω} for the corresponding valve body is obtained. K}, and select the configuration with the smallest volume fraction f from the candidate configuration sequence as the optimal configuration Ω of the valve body.
[0011] Preferably, in step S1, the deep-sea operating condition data includes the hydrostatic pressure P at full ocean depth. ext The internal pressure P of the valve body int The parameters include: dynamic viscosity of the oil (μ), oil density (ρ), flow rate (Q), valve core movement speed (v), valve body and valve core clearance (Δr), sealing length (L), and related parameters of the undercut groove (λ). K L1, k s And the pressure difference ΔP.
[0012] Preferably, in step S1, based on the deep-sea operating conditions data of the valve body, the multiple resistances experienced by the valve core in deep sea are obtained, including the viscous damping force F. v Wedge gap damping force F w and the damping force F of the sinking groove K Viscous damping force F v Wedge gap damping force F w and the damping force F of the sinking groove K The expressions are as follows: ; ; ; in, For viscous damping force, The damping force is the wedge-shaped gap. For the damping force of the cutting groove, The dynamic viscosity of the oil. For the density of the oil, The valve core movement speed, This refers to the clearance between the valve body and the valve core. For the oil sealing length, The pressure difference across the valve core. Where is the valve core diameter. This refers to the oil flow rate; The height of the wedge gap; For flow coefficient, The friction coefficient of the cutting groove. For the geometric coefficient of the recessed groove.
[0013] Preferably, in step S1, the displacement coordination relationship between the valve body and the valve core mating cylindrical surface is as follows: at the mating cylindrical surface of the valve body and the valve core, a point on the central axis of the mating hole on the valve body is selected as the master node M, and the mesh node on the mating hole cylindrical surface is selected as the slave node S. The master node M and the slave node S are coupled by rigid element to establish the displacement coordination relationship.
[0014] Preferably, in step S1, the interpolation relationship between the density of each element and the elastic modulus in the discretized mesh model of the valve body is as follows: ; Where E0 is the elastic modulus of the valve body, E min =10 -9 E0, where p is the penalty factor. This represents the relative density variable of the mesh in the discretized mesh model of the valve body.
[0015] Preferably, in step S2, the structural flexibility constraint model of the valve body is: ; Where C represents the structural flexibility of the valve body. Let F be the global displacement vector, K be the global stiffness matrix, and F be the external load. The static equilibrium equation of the valve body is: K(ρ)·U = F; The valve body volume constraint is: ; in, V0 is the volume fraction threshold, which ranges from 50% to 70% of the original valve body structure volume, and V0 is the initial design domain volume. The density limit is: .
[0016] Preferably, in step S4, the optimized model of the valve body is: ; The optimization criteria are: ; =0; Wherein, λ is determined by the ratio of the sensitivity of the valve body structure's flexibility to the sensitivity of the deformation of the mating cylindrical surface. This refers to the total work done by the hydrostatic pressure at full ocean depth acting on the outer surface of the valve body. This refers to the radial displacement at the cylindrical surface that mates with the valve core. Let be the material retention rate of the i-th mesh element in the discretized mesh model of the valve body.
[0017] Preferably, in step S5, the design variable ρ is optimized using the optimization criterion method. i Perform iterative updates; the specific expression for iterative updates is: ; in, Let i be the material retention rate of the i-th mesh element of the valve body. For the flexibility of the valve body structure, This is the weighted average of the radial displacements of Ns slave nodes on the valve core mating cylindrical surface. Let be the sensitivity of the valve body's flexibility to the density of the i-th unit. The sensitivity of the valve core to the density of the i-th element is determined by the weighted average radial displacement of the cylindrical surface. Let be the initial volume of the i-th grid cell of the valve body. To satisfy the Lagrange multipliers that accommodate the deformation constraints of the cylindrical surface, For Lagrange multipliers constrained by the full volume, is the damping coefficient.
[0018] Preferably, in step S5, the iteration terminates after the optimization iteration for each volume fraction f simultaneously satisfies the following two criteria; Criterion 1: Stable mesh cell density distribution in the valve body: In two adjacent iterations, the maximum absolute change in the density of all network cells within the valve body design domain is less than the tolerance. : ; Criterion 2, convergence of valve body structural flexibility: In two adjacent iterations, the valve body flexibility... The relative change is lower than the tolerance. : ; Each group iterates independently until both of the above criteria are simultaneously satisfied, at which point the iteration terminates. After termination, it is checked whether the configuration of each group satisfies the deformation constraint of the mating cylindrical surface. Among the candidate configurations that satisfy the constraints, the valve body configuration with the smallest volume fraction f is selected as the optimal configuration.
[0019] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. Existing topology optimization methods for deep-sea pressure-resistant structures aim to minimize the overall structural flexibility. These methods only set upper limits for volume fraction constraints and do not address deformation control in critical local areas. After optimization, excessive deformation may occur at local locations such as the mating cylindrical surface, making it impossible to predict valve core jamming due to clearance failure under high pressure. This invention, by constructing a collaborative optimization model of global flexibility constraints and local deformation resistance, achieves, for the first time, simultaneous control of overall lightweighting and local deformation resistance in deep-sea valve body topology optimization. After optimization, the deformation of the mating cylindrical surface can be strictly controlled within a threshold at a sea depth of 11,000 m, effectively preventing valve core jamming caused by deformation damaging the mating clearance.
[0020] 2. Existing topology optimization methods only constrain the displacement of a single node or point, failing to model and constrain the deformation of local surface regions involving multi-point coupling, such as the mating cylindrical surface. This results in optimization results that cannot directly reflect the actual deformation coordination state of the valve body-valve core mating cylindrical surface. This invention employs rigid element coupling to model the nodal displacements of the mating cylindrical surface, establishing a weighted average radial displacement constraint. This transforms the multi-point coupled displacement response of the valve core mating cylindrical surface into a scalar constraint form that topology optimization can handle, overcoming the limitation of existing topology optimization methods in effectively modeling and constraining the local deformation of mating cylindrical surfaces.
[0021] 3. Existing topology optimization methods typically preset a single volume fraction threshold. When only volume constraints are set, a single threshold can generally guarantee optimization feasibility. However, when global compliance constraints and accommodating cylindrical surface deformation constraints are introduced simultaneously, a non-linear relationship emerges between the volume fraction threshold and constraint feasibility. If the threshold is too small, the deformation constraint is not satisfied; if the threshold is too large, the lightweighting effect is insufficient. A single empirical threshold is prone to falling into the infeasible or suboptimal region. This invention employs a parallel iterative screening mechanism with multiple sets of volume fractions to systematically search for the optimal volume fraction value within the common feasible region of global compliance constraints and accommodating cylindrical surface deformation constraints. This avoids empirically preset volume fraction thresholds and improves the global optimality and engineering reliability of the optimization results.
[0022] 4. Existing topology optimization methods do not adequately consider manufacturing process constraints. The optimized structure often contains non-manufacturing features such as closed cavities and tiny narrow gaps, requiring significant secondary modifications or even causing optimization performance failure, leading to a disconnect between design and manufacturing. This invention, after topology optimization, uses post-processing smoothing, removal of non-manufacturing features, and reverse geometric reconstruction to ensure that the optimized structure is compatible with additive manufacturing processes. The optimized valve body weight is reduced by more than 35% compared to the original structure, achieving integrated design and manufacturing collaboration. Attached Figure Description
[0023] Figure 1 This is a flowchart of the method for optimizing the valve body topology of a deep-sea directional valve according to the present invention; Figure 2This is a structural diagram of the valve body before optimization in the method for optimizing the valve body topology of a deep-sea directional valve according to the present invention. Figure 3 This is a structural diagram of the optimized valve body in the method for optimizing the valve body topology of the deep-sea directional valve of the present invention; Figure 4 This is a schematic diagram of the cylindrical mesh used in the optimization method of the valve body topology of the deep-sea directional valve of the present invention. Figure 5 This is a schematic diagram of the fixed support boundary in the method for optimizing the topology of the valve body of a deep-sea directional valve according to the present invention; Figure 6 This is a schematic diagram of internal pressure loading in the method for optimizing the valve body topology of a deep-sea directional valve according to the present invention. Figure 7 This is a schematic diagram of external pressure loading in the method for optimizing the valve body topology of a deep-sea directional valve according to the present invention. Figure 8 This is a schematic diagram of the 90% mass constraint result in the optimization method for the valve body topology of the deep-sea directional valve of the present invention; Figure 9 This is a schematic diagram of the 50% mass constraint result in the optimization method for the valve body topology of the deep-sea directional valve of the present invention; Figure 10 This is a schematic diagram of the 30% mass constraint result in the optimization method for the valve body topology of the deep-sea directional valve of the present invention; Figure 11 This is a schematic diagram of the optimal constraint results in the optimization method for the valve body topology of the deep-sea directional valve of the present invention; Figure 12 This is a schematic diagram of the overall deformation in the method for optimizing the topology of the valve body of a deep-sea directional valve according to the present invention. Figure 13 This is a schematic diagram of the deformation of the cylindrical surface in the optimization method of the valve body topology of the deep-sea directional valve of the present invention.
[0024] Key reference numerals: 1. Optimize the valve body before optimization; 2. Optimize the valve body after optimization; 3. Match the cylindrical surface; 4. Fix the support boundary; 5. Constrain the valve body during optimization; 6. Optimal configuration. Detailed Implementation
[0025] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.
[0026] like Figures 1-13As shown, the main logic of the optimization method for the valve body topology of a deep-sea directional valve is as follows: Based on the deep-sea working conditions where the hydrostatic pressure and oil properties vary with depth across the entire ocean depth, the valve body of the directional valve will simultaneously bear the overall contraction load and the radial extrusion load of the mating cylindrical surface. If only the minimization of global compliance is taken as the optimization objective, the local deformation of the mating cylindrical surface 3 may exceed the limit, leading to valve core jamming. Therefore, the mating cylindrical surface 3 must be introduced as a hard constraint in each iteration of the topology optimization. To this end, it is necessary to first transform the multi-point displacement coupling field of the mating cylindrical surface into a scalar constraint with computable sensitivity, and then find the solution with the smallest volume in the dual-constraint feasible region through parallel screening of multiple sets of volume fractions, such as... Figure 1 As shown, the specific implementation steps include: S1. Obtain deep-sea working condition data of the directional valve body, establish a discretized mesh model of the directional valve body, and establish the displacement coordination relationship between the valve body and the valve core mating cylindrical surface 3, so as to obtain a mechanical analysis basis that includes the design domain and non-design domain and has the ability to extract multi-point coupled response.
[0027] Specifically, deep-sea operating condition data, including the hydrostatic pressure P at full ocean depth. ext The internal pressure P of the valve body int The parameters include: dynamic viscosity of the oil (μ), oil density (ρ), flow rate (Q), valve core movement speed (v), valve body and valve core clearance (Δr), sealing length (L), and related parameters of the undercut groove (λ). K L1, k s And the pressure difference ΔP.
[0028] Based on the deep-sea operating conditions data of the valve body, the key resistance experienced by the valve core in deep-sea conditions is obtained, which serves as the basis for the load boundary conditions required for subsequent topology optimization. This includes the viscous damping force F. v Wedge gap damping force F w and the damping force F of the sinking groove K All resistances increase with increasing sea depth, and the viscous damping force F v Wedge gap damping force F w and the damping force F of the sinking groove K The expression is: ; ; ; in, For viscous damping force, The damping force is the wedge-shaped gap. For the damping force of the cutting groove, The dynamic viscosity of the oil. For the density of the oil, The speed of valve core movement; This refers to the clearance between the valve body and the valve core. For the oil sealing length, The pressure difference across the valve core. Where is the valve core diameter. For oil flow rate, The height of the wedge gap. For flow coefficient, The friction coefficient of the cutting groove. For the geometric coefficient of the recessed groove.
[0029] Furthermore, the hydrostatic pressure at full ocean depth is the primary load borne by the outer surface of the valve body. The oil viscosity μ and oil density ρ increase with depth, directly leading to an increase in viscous damping force during valve core movement. These parameters collectively constitute the core components of the load column vector F in subsequent steps, and their values directly determine the setting of the deformation threshold in the deformation constraint of the mating cylindrical surface 3—the mating clearance Δr and deformation must maintain a safety margin. This is the physical basis that distinguishes this method from existing optimization methods.
[0030] A 3D model is constructed based on the predetermined basic dimensions of the valve body. Solid elements are used to mesh the area where the valve core and valve body mate, and the mesh is refined to maintain a distortion rate within 5%. Design and non-design domains are defined. The non-design domain includes the valve core mating cylindrical surface 3, the oil passage port, and the mounting surface. At the mating cylindrical surface 3 of the valve body and valve core, a point on the central axis of the mating hole on the valve body is selected as the master node M, and the mesh nodes on the mating hole cylindrical surface are selected as slave nodes S. Using RBE2, rigid solid element type 2, the master node M and slave node S are coupled using rigid elements to establish a displacement coordination relationship. This allows the radial displacement response of each node on the mating cylindrical surface 3 to be uniformly extracted and constrained. The physical meaning of this coupling relationship is that when the mating cylindrical surface 3 is subjected to external loads, the radial displacements of each point are coordinated, and the whole conforms to a rigid body motion mode with the master node as the reference. This truly reflects the physical characteristics of the deformation of the valve core mating cylindrical surface—the mating cylindrical surface 3 must maintain cylindricity; otherwise, the valve core and valve body will jam due to "elliptical" surface contact.
[0031] like Figure 7 As shown, a full-ocean-depth hydrostatic pressure P is applied to the outer surface of the valve body 1 before optimization. ext full ocean depth hydrostatic pressure P ext For 110 MPa, such as Figure 6 As shown, a working pressure P is applied to the oil passage wall and port surface of the internal oil circuit. int The fixed support boundary 4, which applies three translational degrees of freedom to the mounting surface, is as follows: Figure 5 As shown, the specific values depend on the specific design pressure applied. Based on the general mesh discretization, the master-slave node displacement coordination relationship of the valve core mating cylindrical surface is established using RBE2 rigid elements, resulting in a discretized mesh model of the valve body containing both the design domain and the non-design domain, as shown below. Figure 3To accommodate the mesh node distribution on cylindrical surface 3, this model simultaneously possesses the ability to extract both global load response and local radial displacement of the cylindrical surface, providing a unified mechanical analysis basis for subsequent dual-constraint optimization.
[0032] Assign a relative density variable ρ to the mesh elements of the region to be optimized in the discretized mesh model of valve body 1 before optimization. i As a design variable for topology optimization, ρ i ∈[ρ min ,1],ρ min Take 10 -3 To prevent numerical singularities. ρ i =1 indicates that the element is a solid material, ρ i =ρ min This indicates that the element is empty. The optimization problem is transformed into: finding the ρ of each element within the design domain. i The optimal distribution ensures that the structure performs optimally while satisfying the constraints.
[0033] The interpolation relationship between the density and elastic modulus of each element in the discretized mesh model of the valve body is established using the solid isotropic material penalty method. The specific expression is as follows: ; Where E0 is the elastic modulus of the valve body, E min =10 -9 E0, where p is the penalty factor. This represents the relative density variable of the mesh in the discretized mesh model of the valve body.
[0034] The actual meaning of this interpolation relationship for the valve body is: the elastic modulus E0 of the valve body material is a constant, determined by the metal material selected for the valve body, for the target ρ i The unit with the intermediate value has its elastic modulus determined by ρ. i The nonlinear decay to values far below those calculated proportionally to volume makes the optimization of these uneconomical elements in terms of mechanical contribution shift towards ρ. i The convergence tends towards 0 or 1, generating a manufacturable configuration with clear boundaries. Here, p is a penalty factor, preferably p=3.
[0035] S2, when the hydrostatic pressure P at full ocean depth ext When uniform force is applied to the outer surface of valve body 1 before optimization, valve body 1 will undergo stiffness deformation. Therefore, minimizing the structural flexibility of the valve body is taken as the optimization objective, and a structural flexibility constraint model of the valve body is constructed. The specific expression is as follows: ; Where C represents the structural flexibility of the valve body. Let be the global displacement vector, representing the displacements of all nodes; K is the global stiffness matrix. F represents the external load.
[0036] Furthermore, the hydrostatic pressure P at full ocean depth ext, The total work done by the external load on the structural displacement of the valve body when it is uniformly applied to the outer surface of the valve body. For the valve body in this invention, the smaller C is, the smaller the overall contraction deformation of the valve body under uniformly distributed external pressure in the deep sea, that is, the greater the overall compressive stiffness of the valve body.
[0037] Based on the working environment of the valve body in the deep sea, the volume constraint of the valve body, the static equilibrium equation of the valve body, and the density limit of the valve body are used as constraint conditions.
[0038] The specific expression for the static equilibrium equation of the valve body is as follows: K(ρ)·U = F.
[0039] For the valve body, F includes the hydrostatic pressure P at full ocean depth. ext The uniformly distributed load acting on the outer surface of the valve body and the internal oil circuit working pressure P int The pressure load acting on the oil passage wall, and the constraint reaction force on the mounting surface. K is determined by the stiffness matrix of each element according to ρ. i It is assembled by weighted aggregation, and K must be reorganized and U solved in each iteration.
[0040] The specific expression for the valve body volume constraint is as follows: ; in, The volume fraction threshold is defined as 50%-70% of the structural volume of valve body 1 before optimization, and V0 is the initial design domain volume. For this valve body, V0 represents the space volume occupied by the design domain of valve body 1 before optimization, and f represents the designer's desired degree of weight reduction. For example, f=0.5 indicates a target weight reduction of 50%. This constraint requires that the equivalent material volume of valve body 2 after optimization does not exceed f times V0.
[0041] The specific expression for the density limit is: .
[0042] S3, due to the hydrostatic pressure P at full ocean depth ext When applied to the outer surface of the valve body, the valve body contracts as a whole, causing a radial inward displacement u on the mating cylindrical surface 3 of the valve core. r When u at each point on the mating cylindrical surface 3 r When the design safety margin of the fit clearance Δr is exceeded, the valve body and valve core change from a clearance fit to an interference fit. At this time, the resistance to valve core movement increases sharply, and reversing becomes stuck. Therefore, in the optimization, it is necessary to explicitly constrain the radial displacement of the fit cylindrical surface 3 to not exceed the upper limit determined by Δr. The radial displacement u of the Ns slave nodes on the fit cylindrical surface 3 in the discretized mesh model of the valve body obtained in step S1 is then used. r,j(j=1,2,...,Ns) is represented by the displacement of the principal node M, reflecting the deformation characteristics of the mating cylindrical surface 3 as a whole undergoing rigid body contraction—that is, the mating cylindrical surface 3 must maintain its cylindricity after compression; otherwise, even if the single-point radial displacement is qualified, the elliptic deformation of the mating cylindrical surface 3 will still lead to local interference between the valve core and the valve body. This multi-point coupled displacement field is compressed into a single scalar metric to construct a deformation constraint model of the mating cylindrical surface of the valve body and valve core in the directional valve. The specific expression is as follows: ; Among them, w j As a weighting coefficient, for the mating cylindrical surface 3 of this valve body, the hydrostatic pressure components borne by each node in the circumferential direction are the same, and the default weight is taken as equal weight w. j =1 / Ns; If there are weakened areas such as oil passage openings in the local area of cylindrical surface 3, a higher weight can be assigned to this area to reflect the deformation sensitivity at that location; U r,j The radial displacement at the j-th slave node can be obtained directly from the radial component of the displacement vector of that node.
[0043] Based on the deformation constraint model of the mating cylindrical surface of the valve body and valve core in the reversing valve, the specific expression for the deformation constraint of the mating cylindrical surface 3 is defined as follows: ; Among them, û rmax The allowable upper limit of radial displacement of the mating cylindrical surface 3 on the valve body is determined based on the mating clearance Δr between the valve body and the valve core. For this valve body, the design value of Δr is typically in the range of 10-20 μm. rmax Take 50%-70% of Δr, i.e. 5-14μm, to ensure that under full ocean depth load, the deformation of the mating cylindrical surface 3 still has a safety margin of not less than 30% from the gap closure.
[0044] S4. Based on the structural flexibility constraint model of the valve body and the deformation constraint model of the mating cylindrical surface of the valve body and valve core in the directional valve, construct the optimization model and optimization criteria for the valve body. The specific expression is as follows: ; The optimization criteria are determined by the optimization model. Request stationary values The physical meaning of this derivation is: for each network element within the valve body design domain, the element density... The direction of increase or decrease is determined by both sensitivities. The degree to which removing the material at that location causes a loss of overall valve body stiffness. The reaction removes the deterioration of the material's effect on the radial displacement of the valve core's mating cylindrical surface; the two are then driven by a weighted synthesis using Lagrange multipliers λ. The iterative update is as follows. Therefore, sensitivity is the input to the optimization criterion, and the optimization criterion is the user of sensitivity; the two are upstream and downstream in relationship, rather than parallel concepts. The specific expression is: ; =0; Here, λ is determined by the ratio of the sensitivity of the valve body structure's flexibility to the sensitivity of the deformation of the mating cylindrical surface. λ measures the "stiffness cost incurred to control the deformation of the mating cylindrical surface"—the larger λ is, the higher the flexibility cost is required to further reduce the deformation of the mating cylindrical surface 3 under the current density distribution, and the more material allocation tends to concentrate in the mating cylindrical surface region during optimization to enhance local stiffness; the smaller λ is, the more sufficient the deformation margin of the mating cylindrical surface 3 is, and more material can be allocated to the outer region of the valve body to improve the overall compressive stiffness. Thus, λ dynamically balances the requirements of global lightweighting and local anti-jamming in each iteration. This refers to the total work done when the hydrostatic pressure at full ocean depth acts on the outer surface of the valve body. The smaller the value, the stronger the overall resistance to pressure and contraction of the valve body. This refers to the radial displacement at point 3 on the mating cylindrical surface of the valve core. The material retention rate of the i-th mesh element in the valve body design domain is expressed as follows: .
[0045] Specifically, the above optimization model expression provides the mathematical basis for sensitivity analysis and iterative updates, where the optimization objective... Together with four constraints, this defines the criterion for determining the "feasibility and optimality" of the valve body density distribution under deep-sea conditions. Taking the partial derivative of this formulation yields the gradient directions in which elements the material should be added to and removed from in each iteration. This is further supported by the deformation constraint of cylindrical surface 3. After embedding with Lagrange multipliers λ and In the optimization criterion update, the update of the jointly driven element density is performed—this differs from general topology optimization which is solely driven by... The single drive creates a fundamental difference, allowing the material in the three regions of the cylindrical surface to receive additional protection during iteration, thereby achieving both lightweight valve body and resistance to jamming deformation at the topology optimization level.
[0046] In optimizing the model Based on this, the mesh units of the valve body are... Taking the partial derivative yields the objective function. right Sensitivity and deformation constraints of cylindrical surface 3 right Sensitivity The adjoint method is used to solve the deformation sensitivity field of the mating cylindrical surface 3, and the field is then filtered by sensitivity filtering. The filtering radius is taken as 1.2-1.8 times the characteristic size of the valve body mesh element to suppress the checkerboard effect. The two sensitivities constitute the driving terms for updating the design variables using the optimization criterion method. and The density of each unit in each iteration is jointly determined by the two through joint weighting using the Grand multiplier λ. The update direction. For each set of preset volume fraction threshold f, the optimization criterion iteration must be driven by the above sensitivity calculation. Sensitivity is the iterative driving force for the dual-constraint optimization under each set of f values.
[0047] S5. Based on the optimized model of the valve body obtained in step S4, set K different volume fraction thresholds. k Utilizing the sensitivity of the valve body structure flexibility and the sensitivity of deformation of cylindrical surface 3 The optimization criteria are used to perform double-constraint topology optimization iterations on each group, resulting in a candidate configuration sequence {Ω1, Ω2, ..., Ω} for the corresponding valve body. K}, and select the configuration with the smallest volume fraction f from the candidate configuration sequence as the optimal configuration of the valve body 6.
[0048] In step S5, each group iterates independently until the convergence criterion is met, and then terminates. Each candidate configuration is checked to see if it satisfies the deformation constraint of the mating cylindrical surface 3 defined in step S3. Candidate configurations that do not meet the constraints are marked as infeasible and eliminated. Among the remaining candidate configurations that meet the constraints, the configuration with the smallest volume fraction f is selected as the optimal configuration 6 of the valve body. That is, among all configurations that will not cause valve core jamming, the one with the lightest weight is selected. Thus, a unique output is determined from the candidate sequence, and the optimal configuration 6 is obtained.
[0049] Figure 8 The constraint optimization of valve body 5 under a volume fraction threshold of 90% mass constraint is given. Figure 9 The constraint optimization of valve body 5 is given under a volume fraction threshold of 50% mass constraint. Figure 10 Valve body 5 is presented in the constraint optimization under a volume fraction threshold of 30% mass constraint. As the mass constraint value decreases, the valve body configuration gradually transitions from a conservative form with relatively full material distribution to a lightweight form with a higher degree of perforation, directly reflecting the effect of volume constraint tightness on material retention. By comparing the radial displacement response of each configuration on the mating cylindrical surface under full ocean depth load, the configuration that meets the requirements is selected. The configuration with the smallest volume fraction is taken as the optimal solution.
[0050] In step S5, the design variable ρ is optimized using the optimization criterion method.i Perform iterative updates; the specific expression for iterative updates is: ; in, Let i be the material retention rate of the i-th mesh element of the valve body. For the structural flexibility of the valve body, The weighted average of the radial displacements of Ns slave nodes on the mating cylindrical surface 3 in the valve core measures the overall shrinkage of the mating cylindrical surface 3 under deep-sea pressure. This represents the sensitivity of the valve body's flexibility to the density of the i-th element. A larger value indicates a greater loss of overall stiffness after removing material from that location, meaning the material should be retained. This value represents the sensitivity of the 3-weighted average radial displacement of the mating cylindrical surface in the valve core to the density of the i-th element. A larger value indicates a greater increase in the radial displacement of the mating cylindrical surface after removing the valve body material at that location; removal may cause the valve core to jam. Let be the initial volume of the i-th grid cell of the valve body. To satisfy the Lagrange multipliers for the deformation constraint of the cylindrical surface 3, For Lagrange multipliers constrained by the full volume, is the damping coefficient, with a value ranging from 0.4 to 0.9.
[0051] In step S5, the optimization iteration for each volume fraction f terminates when both of the following two criteria are met.
[0052] Criterion 1: Stable mesh cell density distribution in the valve body: In two adjacent iterations, the maximum absolute change in the density of all network cells within the valve body design domain is less than the tolerance. : .
[0053] Criterion 2, Valve body structural flexibility convergence: In two adjacent iterations, the valve body flexibility... The relative change is lower than the tolerance. ,Right now: .
[0054] Each group iterates independently until both of the above criteria are simultaneously satisfied, at which point the iteration terminates. After termination, it is checked whether each group of candidate configurations satisfies the deformation constraint of the mating cylindrical surface 3. Eliminate infeasible configurations that do not meet the constraints, and select the valve body configuration with the smallest volume fraction f from the candidate configurations that meet the constraints as the optimal configuration.
[0055] Since the feasible region of the volume fraction threshold f under the dual constraint conditions cannot be analytically predicted before optimization—if f is set too small, such as f=0.3, then the material in the region of the mating cylindrical surface 3 in the valve body is insufficient to control the radial displacement. This will inevitably exceed ûr_max; if f is set too large, such as f=0.8, the valve body wall will be too thick, and the lightweighting effect will be insufficient. -- This method adopts a strategy of parallel calculation of multiple sets of f values: set K different f_k, covering 50%-70%, with a step size of 5%, for a total of 5 sets. Each set independently performs double-constraint topology optimization iteration, outputting the corresponding candidate configuration sequence of the valve body {Ω1, Ω2, ..., Ω K After convergence and feasibility tests, each candidate configuration is selected based on deterministic criteria, such as... Figure 11 As shown, this is the optimal configuration 6 selected from all candidate configurations. Figure 6 and Figure 7 As shown, a full-ocean-depth gravity load P is applied to the candidate configurations respectively. ext =110 MPa and internal working pressure P int Calculate the overall deformation cloud diagram and the radial displacement cloud diagram of the mating cylindrical surface 3 for each candidate configuration, as shown below. Figure 12 and Figure 13 As shown, it can be seen that while significantly reducing the volume and mass of the valve body, the reasonable rigidity of the valve body is still maintained, ensuring the deformation range under deep-sea static pressure. Simultaneously, the radial displacement of the cylindrical surface 3 is also within the maximum radial displacement û. r_max Within this range, it meets the usage requirements.
[0056] The optimal configuration 6 that satisfies the constraints is smoothed, and non-manufacturing features such as closed cavities or tiny narrow gaps are removed. It is then reverse-engineered geometrically reconstructed into a machinable solid model, namely the optimized valve body 2. A full-ocean-depth load is then reapplied to the optimized valve body 2, and finite element analysis is performed to verify the deformation of the optimized valve body 2 and the mating cylindrical surface 3, confirming that the response is controllable. Figure 10 and Figure 3 As shown, after binarizing and smoothing the density field of the optimized valve body 2, closed cavities and tiny narrow slits with feature sizes smaller than 0.1 mm were removed. The optimized valve body 2 was then obtained through inverse geometric reconstruction, resulting in a machinable design. The reconstructed optimized valve body 2 has smooth structural boundaries and no suspended closed cavities, meeting the process requirements for precision machining and additive manufacturing. Figure 2 and Figure 3 As shown, after reverse reconstruction of the optimized valve body 2, it is compared with the original valve body 1. Under the premise of maintaining the integrity of the valve core mating hole surface, oil circuit port and mounting surface, the optimized valve body 2 has formed a reasonable material hollow distribution in the non-critical load-bearing area, and the valve body weight is reduced by more than 35% compared with the original structure.
[0057] The essential difference between the method of this invention and existing optimization methods lies in the fact that it transforms the top-level engineering criterion of "non-jamming" of the deep-sea directional valve into a calculable hard constraint within the optimization iteration through a quantitative chain of "radial displacement of the mating cylindrical surface 3 → safety margin of the mating clearance → upper limit of scalar constraint". This differs from the approach of treating deformation as another objective function in multi-objective optimization—multi-objective optimization requires a trade-off between flexibility and deformation, and generating a set of Pareto solutions still requires manual selection; while this method treats deformation as a hard boundary that cannot be infringed. The reason is that in engineering, as long as the deformation of the mating cylindrical surface does not exceed the allowable clearance value, there is no need to pursue "smaller deformation". Excessively pursuing smaller deformation will lead to excessive thickening of the valve body wall and waste of material, creating unnecessary conflict with the lightweight objective.
[0058] This embodiment verifies the feasibility of the method of the present invention. This method achieves lightweight optimization of the valve body under full ocean depth load by inputting deep-sea operating condition data, constructing a parallel overall flexibility constraint model of the valve body and a deformation constraint model of the mating cylindrical surface 3, and through multiple sets of volume iteration screening. Simultaneously, it strictly controls the deformation of the mating cylindrical surface 3, thereby avoiding valve core jamming caused by deformation damaging the mating clearance. With non-jamming directional control as the primary goal, and ensuring the miniaturization of the valve body, this method yields a reliable deep-sea three-position four-way solenoid directional valve. The optimization results are affected by mesh quality, penalty factor, filtering parameters, and threshold selection. Practical application requires further verification through manufacturing processes and experiments.
[0059] After optimization, the optimal configuration 6 is binarized and the boundary is smoothed and reconstructed to remove tiny features smaller than 0.1mm. The optimized valve body 2 structure is adapted to the 3D printing integrated molding process, and the weight of the optimized valve body 2 is reduced by more than 35% compared with the three-dimensional weight of the valve body 1 before optimization.
[0060] The optimized valve body 2 has a maximum deformation that is more than 60% lower than the original structure under operating conditions at a sea depth of 11,000m. Furthermore, the maximum pressure difference in the internal flow field of the optimized valve body 2 does not exceed 13 MPa, meeting the usage requirements of a rated flow rate of 2.5 L / min.
[0061] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.
Claims
1. A method for optimizing the topology of a deep-sea directional valve body, characterized in that: The steps include: S1. Obtain deep-sea operating condition data of the directional valve body, establish a discretized mesh model of the directional valve body, and assign a relative density variable ρ. i The interpolation relationship between relative density variable and elastic modulus in the discretized mesh model of valve body is established by using the solid isotropic material penalty method, and the displacement coordination relationship between valve body and valve core mating cylindrical surface is established. S2. Minimize the structural flexibility of the valve body as the optimization objective and construct a structural flexibility constraint model for the valve body. Based on the working environment of the directional valve body in the deep sea, the volume constraint of the valve body, the static equilibrium equation of the valve body, and the density limit of the valve body are used as constraint conditions. S3. For the local deformation failure mode of the valve core mating cylindrical surface, the radial displacement u of the Ns slave nodes on the mating cylindrical surface in the discretized mesh model of the valve body obtained in step S1 is calculated. r,j Let j = 1, 2, ..., Ns. Using the displacement of the principal node M, the deformation constraint model of the mating cylindrical surface of the valve body and valve core in the directional valve is constructed as follows: ; Among them, w j U is the weighting coefficient. r,j Let J be the radial displacement at the j-th node; Based on the deformation constraint model of the mating cylindrical surface of the valve body and valve core in the reversing valve, the deformation constraint of the mating cylindrical surface is defined as follows: ; in, This is the upper limit of the allowable radial displacement of the mating cylindrical surface on the valve body; S4. Based on the valve body structure flexibility constraint model obtained in step S2 and the valve body and valve core mating cylindrical surface deformation constraint model obtained in step S3, construct the valve body optimization model and optimization criteria. S5. Set K different volume fraction thresholds. Utilizing the sensitivity of the valve body structure flexibility and the sensitivity to deformation of the cylindrical surface The optimization criteria for composition apply to the volume fraction threshold of each group. k By performing iterative double-constraint topology optimization, the candidate configuration sequence {Ω1, Ω2, ..., Ω} for the corresponding valve body is obtained. K }, and select the configuration with the smallest volume fraction f from the candidate configuration sequence as the optimal configuration Ω of the valve body.
2. The method for optimizing the valve body topology of a deep-sea directional valve according to claim 1, characterized in that: In step S1, the deep-sea operating condition data includes the hydrostatic pressure P at full ocean depth. ext The internal pressure P of the valve body int The parameters include: dynamic viscosity of the oil (μ), oil density (ρ), flow rate (Q), valve core movement speed (v), valve body and valve core clearance (Δr), sealing length (L), and related parameters of the undercut groove (λ). K L1, k s And the pressure difference ΔP.
3. The method for optimizing the topology of a deep-sea directional valve body according to claim 1, characterized in that: In step S1, based on the deep-sea operating conditions data of the valve body, the various resistances experienced by the valve core in deep-sea conditions are obtained, including the viscous damping force F. v Wedge gap damping force F w and the damping force F of the sinking groove K Viscous damping force F v Wedge gap damping force F w and the damping force F of the sinking groove K The expressions are as follows: ; ; ; in, For viscous damping force, The damping force is the wedge-shaped gap. For the damping force of the cutting groove, The dynamic viscosity of the oil. For the density of the oil, The valve core movement speed, This refers to the clearance between the valve body and the valve core. For the oil sealing length, The pressure difference across the valve core. Where is the valve core diameter. This refers to the oil flow rate; The height of the wedge gap; For flow coefficient, The friction coefficient of the cutting groove. For the geometric coefficient of the recessed groove.
4. The method for optimizing the valve body topology of a deep-sea directional valve according to claim 1, characterized in that: In step S1, the displacement coordination relationship between the valve body and the valve core mating cylindrical surface is as follows: at the mating cylindrical surface of the valve body and the valve core, a point on the central axis of the mating hole on the valve body is selected as the master node M, and the mesh node on the mating hole cylindrical surface is selected as the slave node S. The master node M and the slave node S are coupled by rigid element to establish the displacement coordination relationship.
5. The method for optimizing the valve body topology of a deep-sea directional valve according to claim 1, characterized in that: In step S1, the interpolation relationship between the element density and elastic modulus in the discretized mesh model of the valve body is as follows: ; Where E0 is the elastic modulus of the valve body, E min =10 -9 E0, where p is the penalty factor. This represents the relative density variable of the mesh in the discretized mesh model of the valve body.
6. The method for optimizing the valve body topology of a deep-sea directional valve according to claim 1, characterized in that: In step S2, the structural compliance constraint model of the valve body is as follows: ; Where C represents the structural flexibility of the valve body. Let F be the global displacement vector, K be the global stiffness matrix, and F be the external load. The static equilibrium equation of the valve body is: K(ρ)·U = F; The valve body volume constraint is: ; in, V0 is the volume fraction threshold, which ranges from 50% to 70% of the original valve body structure volume, and V0 is the initial design domain volume. The density limit is: 。 7. The method for optimizing the topology of a deep-sea directional valve body according to claim 1, characterized in that: In step S4, the optimized model of the valve body is: ; The optimization criteria are: ; =0; Wherein, λ is determined by the ratio of the sensitivity of the valve body structure's flexibility to the sensitivity of the deformation of the mating cylindrical surface. This refers to the total work done by the hydrostatic pressure at full ocean depth acting on the outer surface of the valve body. This refers to the radial displacement at the cylindrical surface that mates with the valve core. Let be the material retention rate of the i-th mesh element in the discretized mesh model of the valve body.
8. The method for optimizing the topology of a deep-sea directional valve body according to claim 1, characterized in that: In step S5, the design variable ρ is optimized using the optimization criterion method. i Perform iterative updates; the specific expression for iterative updates is: ; in, Let i be the material retention rate of the i-th mesh element of the valve body. For the flexibility of the valve body structure, This is the weighted average of the radial displacements of Ns slave nodes on the valve core mating cylindrical surface. Let be the sensitivity of the valve body's flexibility to the density of the i-th unit. The sensitivity of the valve core to the density of the i-th element is determined by the weighted average radial displacement of the cylindrical surface. Let be the initial volume of the i-th grid cell of the valve body. To satisfy the Lagrange multipliers that accommodate the deformation constraints of the cylindrical surface, For Lagrange multipliers constrained by the full volume, is the damping coefficient.
9. The method for optimizing the valve body topology of a deep-sea directional valve according to claim 1, characterized in that: In step S5, the optimization iteration for each volume fraction f terminates after satisfying the following two criteria simultaneously; Criterion 1: Stable mesh cell density distribution in the valve body: In two adjacent iterations, the maximum absolute change in the density of all network cells within the valve body design domain is less than the tolerance. ; ; Criterion 2, convergence of valve body structural flexibility: In two adjacent iterations, the valve body flexibility... The relative change is lower than the tolerance. : ; Each group iterates independently until both of the above criteria are simultaneously satisfied, at which point the iteration terminates. After termination, it is checked whether the configuration of each group satisfies the deformation constraint of the mating cylindrical surface. Among the candidate configurations that satisfy the constraints, the valve body configuration with the smallest volume fraction f is selected as the optimal configuration.
