Preparation method of pressure self-adaptive nonlinear deformation rubber sealing element

Abstract

The invention provides a preparation method of a pressure self-adaptive non-linear deformation rubber sealing element, and belongs to the technical field of physical and chemical methods.According to the method, nitrile rubber and EPDM are mixed according to a certain proportion, nano-silica filler and a pressure-sensitive deformation triggering agent are added, and a multi-stage mixing and vulcanization technology is adopted to construct a differentiated cross-linked network. By adopting the technologies of reducing treatment, circumferential prestress application, nonlinear deformation area distribution treatment and the like and combining double-layer mold compression molding and nitrogen protection vulcanization, the sealing element forms the gradient strain response characteristic. An elastic modulus distribution optimization function, a Pareto optimization method and a pre-trained sealing strain response prediction network model are innovatively introduced, calculation-driven sealing element design and preparation are achieved, and the problems that in the prior art, a rubber sealing element cannot achieve self-adaptive deformation under the variable pressure working condition, and the sealing performance is poor are solved. And the sealing performance is not stable.

Description

Technical Field

[0001] The invention belongs to the technical field of physical and chemical methods, and in particular relates to a method for preparing a pressure-adaptive nonlinear deformation rubber seal. Background Art

[0002] Rubber seals are widely used in sealing systems in the fields of aerospace, petrochemicals, and automobile manufacturing, and are key components to ensure the safe operation of equipment. Traditional rubber seals are mainly made of rubber materials with a single formula, and are prepared through simple mold forming and vulcanization processes. Their sealing performance depends on a fixed elastic modulus and geometric structure design.

[0003] However, in actual application environments, seals are usually faced with complex and changing pressure conditions, and traditional rubber seals with fixed elastic properties are difficult to adapt to such changes. When the working pressure is lower than the design pressure, the seal contact pressure is insufficient and leakage is likely to occur; when the working pressure exceeds the design value, the seal is excessively deformed, which not only accelerates material aging, but may also cause seal failure. This performance limitation is particularly obvious in harsh environments such as high temperature, high pressure, and strong corrosion.

[0004] At present, in engineering applications, the main way to cope with changing pressure conditions is to increase sealing redundancy or adopt multiple sealing structures. However, these methods often increase the complexity of the structure, the difficulty of installation and the cost of use, and it is still difficult to fundamentally solve the technical problem of rubber seals maintaining the best sealing state under different pressure conditions. How to make rubber seals have the ability to adaptively deform under pressure has become a technical problem that the industry urgently needs to solve. In other words, there is a technical problem in the prior art that rubber seals cannot achieve adaptive deformation under changing pressure conditions, resulting in unstable sealing performance. Summary of the invention

[0005] In view of this, the present invention provides a method for preparing a pressure-adaptive nonlinear deformation rubber seal, which can solve the technical problem in the prior art that the rubber seal cannot achieve adaptive deformation under changing pressure conditions, resulting in unstable sealing performance.

[0006] The present invention is implemented as follows: the present invention provides a method for preparing a pressure-adaptive nonlinear deformation rubber seal, including the following steps: mixing nitrile rubber and EPDM, adding nano-silica filler and pressure-sensitive deformation triggering agent to knead to form a basic rubber mixture; adding vulcanizing agent and accelerator to the basic rubber mixture and continuing to knead to obtain a pre-vulcanized rubber; placing the pre-vulcanized rubber in a pressure-strain adjustment cavity, calculating the optimal circumferential prestress distribution through the elastic modulus distribution optimization function to perform a diameter change treatment, so that the rubber seal exhibits gradient deformation characteristics under different pressure conditions; using a double-layer mold for compression molding; introducing nitrogen during the compression molding process for atmosphere protection; performing a primary vulcanization treatment on the rubber seal to form a primary cross-linking network; performing a nonlinear deformation region distribution treatment on the rubber seal after the primary vulcanization, constructing a multi-objective optimization model of the relationship between the seal stress distribution and the deformation, and solving and determining the stress application parameters of each region; performing a secondary vulcanization treatment to complete the final cross-linking network construction; and performing a pressure adaptability test on the prepared rubber seal. The EPDM is ethylene propylene diene monomer rubber.

[0007] Among them, nitrile rubber and EPDM are mixed in a mass ratio of 4:6 to 6:4, and 2% to 5% by mass of nano-silica filler and 1% to 3% by mass of pressure-sensitive deformation trigger are added. The mixture is kneaded at 50 to 70°C for 30 to 45 minutes to form a basic rubber mixture.

[0008] Among them, 1% to 2% by mass of a vulcanizer and 0.5% to 1.5% by mass of an accelerator are added to the basic rubber mixture, and the mixing is continued for 15 to 25 minutes to obtain a uniformly dispersed pre-vulcanized rubber compound.

[0009] Specifically, the optimal circumferential prestress distribution is calculated through the elastic modulus distribution optimization function, the circumferential prestress range is adjusted to 5 to 10 MPa, and the diameter change treatment is performed to make the rubber seal show gradient deformation characteristics under different pressure conditions.

[0010] Specifically, double-layer mold pressing is used, in which the surface of the inner mold is coated with a self-lubricating nano-coating, and the outer mold is micro-textured, the pressing temperature is controlled at 140 to 160°C, and the pressure is 15 to 25 MPa.

[0011] Among them, nitrogen is introduced into the atmosphere for protection during the pressing process. Specifically, the air pressure is controlled at 0.2 to 0.5 MPa and maintained for 2 to 4 hours, so that the rubber seal forms an internal microscopic stress network structure.

[0012] The rubber seal is subjected to a vulcanization treatment to form a primary cross-linking network. Specifically, the molded rubber seal is subjected to a vulcanization treatment at a temperature of 80 to 100° C. for 60 to 90 minutes to form a primary cross-linking network.

[0013] Among them, the secondary vulcanization treatment is carried out to complete the final cross-linking network construction. Specifically, the treated rubber seal is placed under 120 to 140°C for secondary vulcanization treatment, and the pre-trained seal strain response prediction network model is used to determine the optimal vulcanization time and temperature curve. The vulcanization time range is controlled within 120 to 180 minutes to complete the final cross-linking network construction.

[0014] Among them, the pressure adaptability performance test of the prepared rubber seals includes using the ideal pressure deformation curve generated by the pre-trained sealing strain response prediction network model as a reference standard, cyclically loading 20 times in the pressure range of 1 to 30 MPa, and checking that its nonlinear deformation recovery rate should be greater than 95%, and the sealing performance retention rate should be greater than 98%.

[0015] Among them, the elastic modulus distribution optimization function is used to calculate the optimal elastic modulus distribution gradient according to the working environment pressure range and material properties of the seal. The input includes the initial elastic modulus of the pre-vulcanized rubber, the expected working pressure range, the geometric size parameters of the rubber seal, the contact surface friction coefficient and the allowable deformation range. The output is the optimal elastic modulus distribution curve along the radial distribution of the rubber seal and the corresponding circumferential prestress application scheme.

[0016] Compared with the prior art, the present invention provides a method for preparing a pressure-adaptive nonlinear deformation rubber seal. The present invention proposes a method for preparing a pressure-adaptive nonlinear deformation rubber seal. Through the precise ratio of nitrile rubber and EPDM, the composite reinforcement of nanofillers, the introduction of pressure-sensitive deformation triggers and a multi-stage vulcanization process, a new sealing material system with gradient strain response characteristics is constructed.

[0017] The seal breaks through the limitation of the single elastic response of traditional rubber seals and realizes adaptive deformation behavior under different pressure conditions. When the pressure is low, the pressure-sensitive deformation trigger in the seal causes the rubber molecular chain to rearrange in a directional manner and enhance the contact area; when the pressure increases, the nonlinear deformation area distribution design makes the various parts of the seal deform differently and automatically optimizes the contact pressure distribution. Through the application of elastic modulus distribution optimization function and Pareto optimization method, the seal forms an ideal gradient strain response characteristic and can maintain excellent sealing performance in a wide pressure range of 1 to 30 MPa.

[0018] The present invention solves the problem of adaptive sealing of rubber seals under changing pressure conditions, and significantly improves the reliability and life of the sealing system under complex working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a flow chart of the method of the present invention.

[0020] Figure 2 Graph showing radial elastic modulus distribution and circumferential prestress distribution of the rubber seal in Example 2.

[0021] Figure 3 The temperature-time curve and crosslinking degree development curve during the secondary vulcanization process in Example 2 are shown.

[0022] Figure 4 This is a pressure deformation characteristic curve diagram of the adaptive nonlinear deformation rubber seal in Example 2. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0024] like Figure 1 FIG. 1 is a flow chart of a method for preparing a pressure-adaptive nonlinear deformation rubber seal provided by the present invention. The method comprises the following steps: S01. Select nitrile rubber and EPDM in a mass ratio of 4:6 to 6:4, add 2% to 5% by mass of nano-silicon dioxide filler and 1% to 3% by mass of pressure-sensitive deformation trigger, and mix at 50 to 70° C. for 30 to 45 minutes to form a basic rubber mixture; S02, adding 1% to 2% by mass of a vulcanizing agent and 0.5% to 1.5% by mass of an accelerator to the base rubber mixture, and continuing to mix for 15 to 25 minutes to obtain a uniformly dispersed pre-vulcanized rubber material; S03, placing the pre-vulcanized rubber material in a pressure-strain adjustment cavity, calculating the optimal circumferential prestress distribution by using an elastic modulus distribution optimization function, adjusting the circumferential prestress range to 5 to 10 MPa, and performing a diameter reduction process so that the rubber seal exhibits gradient deformation characteristics under different pressure conditions; S04, using double-layer molds for pressing, wherein the inner mold surface is coated with a self-lubricating nano-coating, and the outer mold is micro-textured, the pressing temperature is controlled at 140 to 160°C, and the pressure is 15 to 25 MPa; S05, introducing nitrogen into the compression molding process for atmosphere protection, controlling the air pressure at 0.2 to 0.5 MPa, and maintaining it for 2 to 4 hours, so that the rubber seal forms an internal microscopic stress network structure; S06, subjecting the molded rubber seal to a vulcanization treatment at a temperature of 80 to 100° C. for 60 to 90 minutes to form a primary cross-linking network; S07, performing nonlinear deformation regional distribution processing on the rubber seal after primary vulcanization, using Pareto optimization method to construct a multi-objective optimization model of the relationship between stress distribution and deformation of the seal, and determining stress application parameters in each region by solving the Pareto optimal solution set, so that the seal forms an optimal gradient strain response characteristic; S08, placing the treated rubber seal at 120 to 140° C. for secondary vulcanization treatment, using a pre-trained seal strain response prediction network model to determine an optimal vulcanization time and temperature curve, wherein the vulcanization time range is controlled within a range of 120 to 180 minutes, and the final cross-linking network construction is completed; S09. Conduct a pressure adaptability test on the prepared rubber seal, use the ideal pressure deformation curve generated by the pre-trained seal strain response prediction network model as a reference standard, and cyclically load 20 times within the pressure range of 1 to 30 MPa to test its nonlinear deformation recovery rate should be greater than 95%, and the sealing performance retention rate should be greater than 98%.

[0025] Among them, the pressure-sensitive deformation trigger specifically refers to an organosilicon-modified polyamide compound that can change the molecular arrangement under the action of external pressure and promote the directional deformation of the rubber material, and its molecular structure contains reversible deformation groups.

[0026] Among them, the diameter-changing treatment specifically refers to the process of precisely controlling the radial dimension change process of the rubber seal to form a predetermined stress distribution inside the material, thereby giving the seal the ability to adaptively deform under different pressure conditions.

[0027] Among them, micro-texturing treatment specifically refers to the surface processing technology of making regular or irregular micro-texture structures with a depth of 10 to 50 microns on the mold surface, which is used to regulate the surface morphology characteristics and friction coefficient of rubber seals.

[0028] Among them, the nonlinear deformation area distribution processing specifically refers to a process method of controlling the cross-linking density and filler distribution of different parts of the rubber seal to make the seal produce non-uniform deformation when under pressure, thereby achieving adaptive sealing under different pressure conditions.

[0029] Among them, the gradient strain response characteristics specifically refer to the performance characteristics that the rubber seal exhibits different degrees of deformation and deformation rate in different areas in response to external pressure stimulation, so that the overall structure can automatically adjust the sealing contact area and sealing pressure according to the working pressure.

[0030] Among them, the Pareto optimization method specifically refers to a mathematical optimization method that constructs a multi-objective optimization problem and seeks a set of solutions so that the improvement of any objective function will inevitably lead to the deterioration of other objective functions, thereby finding the best balance point between multiple conflicting optimization objectives.

[0031] Among them, the elastic modulus distribution optimization function is used to calculate the optimal elastic modulus distribution gradient according to the working environment pressure range and material properties of the seal. The input includes the initial elastic modulus of the pre-vulcanized rubber, the expected working pressure range, the geometric size parameters of the rubber seal, the contact surface friction coefficient and the allowable deformation range. The output is the optimal elastic modulus distribution curve along the radial distribution of the rubber seal and the corresponding circumferential prestress application scheme.

[0032] Among them, the specific structure of the pre-trained sealing strain response prediction network model is a hybrid architecture based on the combination of graph convolutional network and recurrent neural network, wherein the graph convolutional network is used to capture the spatial correlation and stress transfer characteristics between the various parts of the rubber seal, and the recurrent neural network is used to model the dynamic deformation behavior of the rubber seal over time under different pressure conditions. The graph attention mechanism parameters in the pre-trained sealing strain response prediction network model are dynamically adjusted through three key parameters: the elastic modulus of the pre-vulcanized rubber, the distribution density of the nano-silica filler, and the expected working pressure range, thereby achieving accurate prediction of the deformation behavior of the rubber seal with different material formulations and structural designs.

[0033] Among them, the steps of establishing a training data set in the pre-training process of the pre-trained sealing strain response prediction network model specifically include collecting a large number of real deformation data of rubber seals with different formulations and structures under various pressure conditions, using finite element analysis software to generate theoretical deformation simulation data as data enhancement, and combining actual sealing performance test data to establish a three-dimensional correlation database of pressure deformation sealing performance, using data cleaning and normalization processing to ensure data quality, and finally dividing the data set into training set, verification set and test set in a ratio of 8:1:1.

[0034] Among them, the pre-training steps of the pre-trained sealing strain response prediction network model specifically include first pre-training on simulated data to learn basic material mechanics laws and strain response characteristics, then fine-tuning using real test data to adapt to the non-ideal behavior of actual materials, using adaptive learning rate optimizer and early stopping strategy to prevent overfitting, introducing pressure gradient as an auxiliary supervision signal during training to improve the model's sensitivity to pressure changes, and finally ensuring the generalization ability of the model through cross-validation on test data of rubber seals with different formulations, so that the pre-trained sealing strain response prediction network model can accurately predict the deformation behavior and sealing performance of the rubber seals with unseen rubber formulations and structural designs under various pressure conditions.

[0035] Among them, the ideal pressure-deformation curve specifically refers to a mathematical relationship curve that describes the ideal deformation state of the rubber seal under different pressure conditions, which is used to guide the design optimization and performance evaluation of the rubber seal, and is generated by the pre-trained seal strain response prediction network model based on material properties and structural parameters.

[0036] The specific implementation of the above steps is described in detail below. The specific implementation of step S01 is to first accurately weigh the nitrile rubber and EPDM, and the ratio can be selected to be 5:5 by mass. This ratio can simultaneously utilize the excellent oil resistance of nitrile rubber and the high temperature resistance of EPDM. Weigh a nano-silica filler with a mass fraction of 3.5%. The particle size of the filler is controlled within the range of 30~50 nanometers, has a high specific surface area characteristic, and can provide a good reinforcement effect. Select a pressure-sensitive deformation trigger with a mass fraction of 2%. The trigger is an organosilicon-modified polyamide compound, and the molecular structure contains aromatic biphenyl groups and siloxane segments. This structure will produce reversible conformational changes under pressure. The above materials are placed in a closed mixer, the temperature is set to 60°C, the speed is controlled at 45~60 rpm, and the mixing time is 40 minutes. The mixing process adopts segmented temperature control technology, and the first 20 minutes are kept at 55°C to promote material softening; the temperature is raised to 60°C in the last 20 minutes to ensure that the components are fully compatible. The purpose of this step is to obtain a basic rubber mixture with a uniform dispersion structure through precise proportioning and sufficient mixing, so as to form a stable matrix material for subsequent seals.

[0037] The specific implementation method of step S02 is to accurately weigh and add a vulcanizing agent with a mass fraction of 1.5% to the basic rubber mixture, preferably tetramethylthiuram disulfide, which has a high cross-linking efficiency and good thermal stability. At the same time, an accelerator with a mass fraction of 1% is added, and N-cyclohexyl-2-benzothiazole sulfonamide is selected to provide an appropriate vulcanization rate and good cross-linking density control. The mixture is placed in an open rubber mixer, the temperature is controlled at 40-45°C, the speed is 20-25 rpm, and the mixing time is 20 minutes. The mixing process adopts an intermittent operation method, and the machine is stopped once every 5 minutes to perform scraping and turning operations to ensure that each component is evenly dispersed. The pre-vulcanized rubber compound after mixing is subjected to aging test, and the Martin aging resistance coefficient is not less than 0.85, indicating that the material has good cross-linking potential and thermal stability. Through this step, the uniform dispersion of the vulcanization system is achieved, which provides the basic conditions for the formation of a stable cross-linking network for rubber seals.

[0038] The specific implementation method of step S03 is to place the pre-vulcanized rubber in a special pressure-strain adjustment chamber, which consists of two layers of coaxial cylinders, the inner layer can expand radially, and the outer layer has high rigidity. Apply radial pressure to the adjustment chamber, the initial pressure is 3 MPa, slowly increase to 8 MPa, and maintain for 30 minutes. The optimal circumferential prestress distribution is calculated by the elastic modulus distribution optimization function, which is based on finite element analysis and strain energy density theory, considering the nonlinear elastic behavior of the material. The input parameters include the initial elastic modulus of the pre-vulcanized rubber (usually 4~6 MPa), the expected working pressure range (1~30 MPa), the geometric size parameters of the seal (inner diameter, outer diameter, cross-sectional shape, etc.), the contact surface friction coefficient (usually 0.2~0.4) and the deformation tolerance range (the maximum deformation does not exceed 25% of the original size). The optimization calculation uses adaptive mesh refinement technology to improve the calculation accuracy in the key area of ​​the seal. The calculation results guided the adjustment of the circumferential prestress, which was set to 7.5 MPa, and then the diameter change treatment was performed to make the rubber material form different pre-deformation states at different radial positions. The diameter change treatment adopted a segmented control strategy, with the pre-strain rate in the inner layer area controlled at 8-12%, the middle layer area controlled at 58%, and the outer layer area controlled at 35%. Through this step, the rubber seal is given a gradient deformation characteristic, so that it can show adaptive deformation ability under different pressure conditions.

[0039] The specific implementation method of step S04 is to design and manufacture a double-layer mold system, wherein the inner mold is made of high-precision stainless steel material with a surface roughness controlled below Ra0.2μm, and the outer mold is made of high-strength aluminum alloy material. The inner mold surface is coated with a self-lubricating nano coating, which is a composite of polytetrafluoroethylene and molybdenum disulfide, with a thickness controlled at 15~20 microns, and is prepared by plasma-assisted chemical vapor deposition. The outer mold surface is micro-textured, and laser micro-machining technology is used to produce a cross-grid micro-texture structure with a depth of 30 microns, a width of 50 microns, and a spacing of 200 microns. The pre-vulcanized rubber is placed in the mold, the mold is closed, and placed in a flat vulcanizer, and the temperature is raised to 150°C, and the heating rate is controlled at 3~5°C / minute. The pressure is increased to 20 MPa, the pressure remains stable, and the fluctuation range is controlled at ±0.5 MPa. The purpose of this step is to form a precise geometric shape and surface characteristics of the seal through a precisely controlled compression molding process combined with special mold surface treatment technology, laying the foundation for achieving excellent sealing performance.

[0040] The specific implementation method of step S05 is to introduce high-purity nitrogen (purity ≥ 99.99%) into the mold cavity through the preset gas channel on the mold during the compression molding process to replace the air in the mold cavity. The nitrogen pressure is controlled at 0.35 MPa and the flow rate is 0.5~1 liter / minute. The nitrogen environment in the mold cavity is maintained for 3 hours while maintaining a constant mold temperature and pressure. The nitrogen environment effectively inhibits the interference of oxygen on the rubber cross-linking reaction and reduces the oxidative degradation phenomenon. Through proton nuclear magnetic resonance spectrum analysis, it is found that the free end density of the molecular chain of the rubber seal vulcanized under nitrogen protection is reduced by 18~25%, indicating that a more complete three-dimensional network structure has been formed. In addition, the nitrogen environment also promotes the arrangement of the rubber molecular chain in a specific orientation, forming a highly ordered microscopic stress network structure. This microscopic stress network structure gives the seal a stress memory effect and can return to a preset deformation state after being compressed. The role of this step is to control the cross-linking reaction environment of the rubber molecular chain through atmosphere protection, construct the internal microscopic stress network structure of the rubber seal, and improve its deformation recovery ability.

[0041] The specific implementation method of step S06 is to take out the molded rubber seal, remove the burrs generated by the mold parting surface, and place it in a hot air circulation oven with precise temperature control. The set temperature is 90°C, and the temperature uniformity is controlled within the range of ±2°C. The primary vulcanization time is 75 minutes, and the vulcanization process adopts a programmed temperature rise mode, preheating at 80°C for the first 15 minutes, and then maintaining at 90°C for 60 minutes. During the vulcanization process, the Mooney viscosity of the rubber material is detected every 15 minutes, and its change trend is recorded. When the Mooney viscosity reaches 2.5 to 3 times the initial value, it indicates that an appropriate primary cross-linking network has been formed. The primary vulcanization process mainly forms a cross-linked structure of sulfur-sulfur bonds and sulfur-carbon bonds. This structure has certain strength and elasticity, but retains some active sites to prepare for subsequent secondary vulcanization. After the primary vulcanization is completed, the seal is naturally cooled to room temperature, and any form of external force distortion and stretching is avoided during the period. The cross-sectional structure of the rubber seal is observed by scanning electron microscopy, and uniformly distributed cross-linked networks and filler particles can be seen. The purpose of this step is to form a primary cross-linking network for the rubber seal through a controlled primary vulcanization process, laying the foundation for subsequent nonlinear deformation area distribution processing.

[0042] The specific implementation method of step S07 is to perform nonlinear deformation area distribution processing on the rubber seal after primary vulcanization. First, three-dimensional laser scanning modeling of the seal is performed to obtain its precise geometric data. Based on the geometric model, a finite element analysis model of the seal is constructed, and the grid size is controlled at 0.2~0.5 mm. The key areas of the seal are defined, including the sealing contact surface, the stress concentration area and the deformation transition area. The multi-objective optimization model is constructed using the Pareto optimization method, and the optimization objectives include three mutually constrained objectives: uniformity of sealing contact pressure, deformation recovery rate and rationality of stress distribution. The objective function adopts the weighted summation method, and the weight coefficient is determined by the hierarchical analysis method. The weight of contact pressure uniformity is 0.4, the weight of deformation recovery rate is 0.35, and the weight of stress distribution rationality is 0.25. The optimization variables include the cross-linking density, filler distribution density and prestress distribution of each region. The constraints include that the maximum stress does not exceed 70% of the ultimate strength of the material, and the maximum deformation variable does not exceed 35% of the original size. The multi-objective optimization problem was solved by genetic algorithm, with the population size set to 100, the number of iterations to 500, the crossover probability to 0.8, and the mutation probability to 0.05. After obtaining the Pareto optimal solution set, the TOPSIS method (topic ideal solution sorting technique) was used to select the solution with the best comprehensive performance and determine the stress application parameters for each region. According to the optimization results, heat treatment and mechanical deformation treatment were applied to different regions of the seal to form a gradient distribution of crosslinking density and filler distribution. The purpose of this step is to achieve different deformation characteristics of the seal in different regions through multi-objective optimization and regionalization, so as to obtain the optimal gradient strain response characteristics.

[0043] The specific implementation method of step S08 is to place the rubber seal after nonlinear deformation area distribution treatment in a precision temperature-controlled pressure vulcanization tank for secondary vulcanization treatment. Nitrogen is filled into the pressure vulcanization tank, and the pressure is controlled at 0.3 MPa. The optimal vulcanization time and temperature curve are determined by a pre-trained seal strain response prediction network model. The network model is based on a hybrid architecture of a graph convolutional network and a recurrent neural network, and includes 3 layers of graph convolutional layers and 2 layers of bidirectional long short-term memory network layers. The graph convolutional network adopts an edge attention mechanism to capture the spatial correlation between different parts of the seal. The recurrent neural network adopts a gating mechanism to effectively model the dynamic deformation behavior of rubber seals under pressure cycles. The model input includes material formula parameters, geometric dimension data, primary vulcanization state parameters and expected working conditions of the seal. The model outputs a vulcanization temperature-time curve and a predicted deformation-pressure relationship curve. According to the model prediction, the secondary vulcanization temperature is set to 130°C and the vulcanization time is set to 150 minutes. The vulcanization process adopts a segmented temperature control strategy, preheating at 120°C for the first 30 minutes, then maintaining at 130°C for 90 minutes, and finally completing deep crosslinking at 140°C for 30 minutes. The resistance change of the seal is monitored in real time during the vulcanization process. When the resistance value stabilizes at a constant value, it indicates that the crosslinking reaction is basically completed. After the secondary vulcanization is completed, the seal is slowly cooled to room temperature, and the cooling rate is controlled at 1~2°C / minute to avoid internal stress concentration caused by rapid cooling. The purpose of this step is to complete the construction of the final crosslinking network of the rubber seal through a controlled secondary vulcanization process, so that it has stable nonlinear deformation characteristics and adaptive sealing capabilities.

[0044] The specific implementation method of step S09 is to conduct a comprehensive pressure adaptability performance test on the prepared rubber seal. First, the seal is installed on a sealing performance test device, which can provide an adjustable pressure in the range of 1 to 30 MPa and a temperature environment simulating actual working conditions. The pre-trained seal strain response prediction network model is used to generate an ideal pressure deformation curve as a reference standard. The ideal curve describes the optimal deformation state that the seal should have at each pressure point, forming a benchmark for performance evaluation. The test process adopts a step-by-step pressurization scheme, starting from 1 MPa, increasing by 2 MPa each time until 30 MPa, and then decreasing to 1 MPa with the same step length, forming a complete loading-unloading cycle. Stay at each pressure point for 3 minutes, and record the deformation amount, contact pressure distribution and leakage rate of the seal. Perform 20 consecutive cyclic loading tests to evaluate the durability and stability of the seal. Calculate the nonlinear deformation recovery rate, that is, the ratio of the deformation recovery amount after unloading to the maximum deformation amount, which is required to be greater than 95%. At the same time, measure the sealing performance retention rate, that is, the ratio of the sealing efficiency of the 20th cycle test to the first time, which is required to be greater than 98%. The temperature distribution of the seal during the pressure cycle is monitored by an infrared thermal imager to evaluate its energy dissipation characteristics. The local deformation distribution of the seal is analyzed using digital image correlation technology to verify its gradient strain response characteristics. The purpose of this step is to verify the pressure adaptability and long-term stability of the rubber seal through rigorous performance testing to ensure that it meets the requirements of engineering applications.

[0045] The mathematical model or calculation process involved in the present invention is described in detail below.

[0046] Step S03 involves the calculation process of the elastic modulus distribution optimization function, which is based on finite element analysis and strain energy density theory and takes into account the nonlinear elastic behavior of the material. The specific expression is as follows: ; In the formula, is the radial position The elastic modulus at ; is the basic elastic modulus, with an initial value of 4 to 6 MPa; is the amplitude of elastic modulus change, ranging from 1 to 3 MPa; is the elastic modulus gradient coefficient, ranging from 1.5 to 3.0; is the radial coordinate; is the inner diameter of the seal; is the outer diameter of the seal.

[0047] This equation describes the distribution of the elastic modulus of the rubber seal from the inner diameter to the outer diameter, in the form of an exponential decay function. The exponential decay function can simulate the nonlinear performance change of the rubber material in the radial direction. The elastic modulus of the inner layer changes quickly, while the outer layer changes slowly, which is consistent with the actual working state of the rubber seal. The exponential form is chosen instead of the nonlinear or polynomial form because the stress-strain relationship of the rubber material itself has exponential characteristics, and the exponential function has good convergence in numerical calculations.

[0048] Based on the elastic modulus distribution, calculate the circumferential prestress distribution: ; In the formula, is the radial position Circumferential prestress at is the reference circumferential prestress, with a value range of 5~10 MPa; is the stress-elastic modulus correlation index, ranging from 0.6 to 0.9; is the stress fluctuation amplitude coefficient, and its value range is 0.10.3; is the number of stress fluctuation cycles, usually taken as 24.

[0049] This equation relates the circumferential prestress to the elastic modulus and introduces a periodic fluctuation term to simulate the non-uniform contact in the actual sealing environment. It reflects the nonlinear mechanical properties of rubber materials, that is, the stress and elastic modulus are not a simple linear relationship. The sine term introduces a controllable periodic change, so that the seal forms alternating high and low stress areas in the circumferential direction, enhancing its adaptability to irregular surfaces.

[0050] Calculate the optimal pre-strain distribution: ; In the formula, is the radial position Prestrain at is the Poisson’s ratio of rubber material, ranging from 0.48 to 0.5.

[0051] This equation is based on the strain calculation formula of annular components in elastic mechanics, taking into account the Poisson effect and geometric nonlinearity of the material. This reflects the radial-circumferential coupling effect in the ring structure, that is, the circumferential strain will cause radial deformation. The term takes into account the effect of the ring geometry on the strain distribution. The strain gradient is larger near the inner diameter and smaller near the outer diameter.

[0052] In step S07, the Pareto optimization method is used to construct a multi-objective optimization model of the relationship between the seal stress distribution and deformation. The specific expression is as follows: Multi-objective optimization function: ; In the formula, Optimize function vectors for multiple objectives; is the decision variable vector, which includes the cross-linking density, filler distribution density and prestress distribution of each region; is the seal contact pressure uniformity objective function; is the deformation recovery rate objective function; is the objective function of stress distribution rationality.

[0053] Sealing contact pressure uniformity objective function: ; In the formula, For the Sealing pressure at each contact point; is the average sealing pressure of all contact points; is the total number of contact points; is the trade-off coefficient, and its value range is 0.2~0.5.

[0054] The function consists of two parts: the first part calculates the variance of the contact pressure to evaluate the overall uniformity of the pressure distribution; the second part considers the maximum pressure deviation to prevent local extreme pressure points. The square term makes large deviation points more severely penalized, prompting the optimization process to focus more on uniform distribution.

[0055] Deformation recovery rate objective function: ; In the formula, For the Residual deformation of a region; For the The maximum deformation of a region; For the The area of ​​the region; is the total number of regions.

[0056] This function calculates the overall deformation recovery rate, taking into account the area weighting of different areas. Area weighting is used instead of simple averaging because different areas of the seal have different importance, and areas with large contact areas have a greater impact on the sealing performance. The fractional form directly reflects the physical definition of the recovery rate, which is the ratio of the recovered deformation to the maximum deformation.

[0057] Stress distribution rationality objective function: ; In the formula, For the Region stress components; is the corresponding stress limit; is the number of stress components considered; For the The weight coefficient of the stress component; For the The average paradigm equivalent stress of the area; is the optimal stress ratio of adjacent regions, ranging from 1.2 to 1.5; is the stress gradient weight coefficient, and its value range is 0.3~0.6.

[0058] The function is divided into two parts: the first part evaluates whether the stress in each area exceeds the material limit, using a penalty function, and only when it exceeds the limit is it included in the objective function; the second part evaluates whether the stress gradient between adjacent areas is reasonable to ensure the formation of an appropriate stress gradient distribution. The use of square terms and absolute value terms makes the optimization process more focused on eliminating outliers.

[0059] Comprehensive objective function: ; In the formula, is the comprehensive objective function; After normalization, The objective function value; are weight coefficients, which are 0.4, 0.35, and 0.25 respectively.

[0060] Normalization of objective function: ; In the formula, and Respectively The minimum and maximum values ​​of the objective function in the feasible region.

[0061] The normalization process converts objective functions of different dimensions and magnitudes to the same scale, so that the weight coefficient can directly reflect the relative importance of each objective.

[0062] In step S08, the pre-trained seal strain response prediction network model is used to determine the optimal vulcanization time and temperature curve. The model is based on a hybrid architecture of a graph convolutional network and a recurrent neural network. The optimization equation of the vulcanization time and temperature curve is specifically expressed as follows: ; In the formula, For time The vulcanization temperature at Is the basic vulcanization temperature, the value range is 110~120℃; is the initial temperature increase, ranging from 5 to 15°C; It is the temperature increase in the later stage, and the value range is 10~20℃; is the initial heating rate coefficient, ranging from 0.05 to 0.1 / min; is the late heating rate coefficient, ranging from 0.02 to 0.04 / min; It is the turning point time of the temperature curve, and its value range is 30~45 minutes.

[0063] The equation describes a two-stage temperature curve: in the first stage, the temperature gradually decreases from a high point to promote uniform heat distribution of the rubber material; in the second stage, the temperature gradually increases to accelerate the deep cross-linking reaction. The exponential function is used to describe the temperature change because the heat conduction process conforms to the exponential decay law and the exponential function can provide a smooth temperature transition. The combination of the two exponential terms makes the temperature curve flexible enough to adapt to the vulcanization requirements of different material formulations.

[0064] Based on the temperature curve, predict the development of crosslinking degree: ; In the formula, For time The degree of cross-linking at is the maximum possible cross-linking degree, ranging from 0.9 to 0.98; is the temperature-dependent reaction rate constant; is the reaction order, ranging from 1.2 to 1.8.

[0065] ; In the formula, is the pre-exponential factor, the value range is 10 5 ~10 7 / minute; is the activation energy, ranging from 80 to 120 kJ / mol; is the gas constant, 8.314 J / (mol·K); is the Kelvin temperature.

[0066] The integral equation is based on the kinetics of the crosslinking reaction and describes the development of the degree of crosslinking over time. The integral within the exponential term takes into account the cumulative effect of the temperature history on the crosslinking process. The term reflects the effect of reactant concentration on the reaction rate. As the degree of cross-linking increases, the reaction rate gradually decreases. The Arrhenius equation is used to describe the effect of temperature on the reaction rate, which reflects the essence of high temperature accelerating the cross-linking reaction.

[0067] Based on the predicted degree of crosslinking, calculate the optimal vulcanization time: ; In the formula, is the optimal vulcanization time; is the target cross-linking degree, ranging from 0.85 to 0.95; is the cross-linking rate threshold, ranging from 0.001 to 0.003 / min.

[0068] The equation determines two conditions for the optimal vulcanization time: one is to achieve the target crosslinking degree, and the other is that the crosslinking rate drops below the threshold, indicating that the reaction is basically completed. Minimization operation is used to find the shortest time that meets the conditions to avoid unnecessary energy consumption and the risk of over-vulcanization.

[0069] In step S09, the ideal pressure-strain curve generated by the pre-trained sealing strain response prediction network model is used as a reference standard. The mathematical expression of the ideal pressure-strain curve is as follows: ; In the formula, For pressure The ideal deformation under ; is the initial deformation, with a value range of 0.050.2 mm; is the power function coefficient, ranging from 0.01 to 0.05; is the power exponent, ranging from 0.6 to 0.8; is the saturation deformation coefficient, and its value range is 0.51.5 mm; is the deformation saturation rate coefficient, and its value range is 0.1~0.3 / MPa.

[0070] The equation consists of three parts: initial shape Represents the deformation during assembly preload; power function term Describes the elastic deformation in the low-pressure area. The power function relationship reflects the nonlinear elastic properties of the rubber material. The deformation behavior of the high-pressure region tending to saturation is described, simulating the compression limit of the rubber material under high pressure. The combination of power function and exponential function can accurately describe the deformation characteristics of the rubber material in the full pressure range, which is better than simple linear or polynomial models.

[0071] Based on the ideal deformation curve, calculate the nonlinear deformation recovery rate: ; In the formula, is the nonlinear deformation recovery rate; is the deformation under maximum pressure; is the residual deformation after unloading.

[0072] This equation directly reflects the physical definition of deformation recovery, which is the ratio of the recovered deformation to the maximum deformation. The percentage form is used to more intuitively express the degree of recovery, which is convenient for comparison with the 95% threshold requirement.

[0073] Calculate the sealing performance retention rate: ; In the formula, is the sealing performance retention rate; is the sealing efficiency of the first cycle test; The sealing efficiency is the 20th cycle test.

[0074] Sealing efficiency calculation: ; In the formula, For sealing efficiency; is the measured leakage rate; This is the standard allowable leakage rate.

[0075] This set of equations evaluates the performance stability of the seal after multiple pressure cycles. The ratio directly reflects the degree of performance degradation, which is convenient for comparison with the 98% threshold requirement. The sealing efficiency is quantified by comparing the leakage rate with the standard allowable value, which is consistent with the evaluation method in engineering practice.

[0076] Optionally, step S07 also involves a genetic algorithm and a TOPSIS method used in the optimization solution process, which are described in detail as follows.

[0077] Mathematical expression of genetic algorithm in Pareto optimization: ; In the formula, For the Generation population; For the Generation population; is the selection operator; is the crossover operator; is the mutation operator.

[0078] The selection operator calculates the fitness of the individual based on the objective function value: ; In the formula, For individuals The fitness value of For individuals The comprehensive objective function value.

[0079] Using the roulette wheel selection method, individuals The probability of being selected is: ; In the formula, For individuals The probability of being selected; is the population size, which is set to 100.

[0080] The TOPSIS method (top-optimal solution ranking technique) is used to select the best solution from the Pareto optimal solution set: First calculate the positive and negative ideal solutions: ; ; In the formula, For the A positive ideal solution to the objective function; For the Negative ideal solution of the objective function; is a solution in the Pareto optimal solution set.

[0081] Compute the distance of each solution to the positive and negative ideal solutions: ; ; In the formula, For solution The distance to the positive ideal solution; For solution The distance to the negative ideal solution; For the The weight of the objective function.

[0082] Compute the closeness coefficient for each solution: ; In the formula, For solution The proximity coefficient ranges from 0 to 1. The larger the value, the better the solution.

[0083] Selection of the optimal solution: ; In the formula, is the optimal solution finally selected.

[0084] Specifically, the principle of the present invention is: the technical principle of the present invention is based on the cross-integration of materials science, computer science and mechanics, and realizes pressure-adaptive deformation characteristics by regulating the microstructure and macroscopic properties of rubber seals at multiple levels. At the material formulation level, nitrile rubber provides good oil resistance and mechanical strength, while EPDM enhances heat resistance and aging resistance. The two are mixed in proportion to form a basic composite system. Nano-silica fillers are evenly dispersed in the rubber matrix to form a "physical cross-linking point" reinforcement network, which improves the stiffness and strength of the material; and the pressure-sensitive deformation trigger is a key functional additive. The reversible deformation group in its molecular structure can change conformation under the action of external pressure, induce the directional arrangement of rubber molecular chains, and produce controllable micro-deformation.

[0085] At the process design level, the variable diameter treatment and prestressing technology enable the seal to form a preset internal stress distribution during the preparation process, laying the foundation for the subsequent nonlinear deformation response. In the double-layer mold design, the inner self-lubricating nano-coating reduces the demolding resistance, and the outer micro-texturing treatment accurately controls the surface morphology and friction characteristics. The multi-stage vulcanization process forms a cross-linked network of different densities through precise control of temperature and time, so that different areas of the seal present differentiated deformation characteristics.

[0086] The most innovative part of this invention is the introduction of advanced computational models to assist in optimizing the design. The elastic modulus distribution optimization function is based on the principle of material mechanics to calculate the optimal elastic modulus gradient distribution; the Pareto optimization method balances the conflicting indicators such as deformability, sealing and life under multi-objective constraints; the pre-trained seal strain response prediction network model combines the graph convolutional network with the recurrent neural network to accurately simulate and predict the dynamic deformation behavior of the seal under different pressures. This computationally driven design method enables the seal to form a synergistic effect at three levels: microscopic molecular structure, mesoscopic cross-linking network and macroscopic geometric shape, achieving adaptive deformation capability under changing pressure conditions.

[0087] A specific embodiment 1 of the present invention is provided below, and the specific implementation method of each step in this embodiment 1 is described in detail as follows.

[0088] The specific implementation of steps S01-S02 is the same as above and will not be repeated here.

[0089] The specific implementation method of step S03 is to place the pre-vulcanized rubber in a special pressure-strain adjustment chamber, which consists of two layers of coaxial cylinders, the inner layer can expand radially, and the outer layer has high rigidity. Apply radial pressure to the adjustment chamber, the initial pressure is 3 MPa, slowly increase to 8 MPa, and maintain for 30 minutes. The optimal circumferential prestress distribution is calculated by the elastic modulus distribution optimization function, which is based on finite element analysis and strain energy density theory, taking into account the nonlinear elastic behavior of the material. The elastic modulus distribution optimization function is specifically expressed as follows: ; In the formula, is the radial position The elastic modulus at ; is the basic elastic modulus, with an initial value of 4 to 6 MPa; is the amplitude of elastic modulus change, ranging from 1 to 3 MPa; is the elastic modulus gradient coefficient, ranging from 1.5 to 3.0; is the radial coordinate; is the inner diameter of the seal; is the outer diameter of the seal.

[0090] Based on the elastic modulus distribution, calculate the circumferential prestress distribution: ; In the formula, is the radial position Circumferential prestress at is the reference circumferential prestress, with a value range of 5~10 MPa; is the stress-elastic modulus correlation index, ranging from 0.6 to 0.9; is the stress fluctuation amplitude coefficient, ranging from 0.1 to 0.3; is the number of stress fluctuation cycles, usually 2~4.

[0091] Calculate the optimal pre-strain distribution: ;

[0092] In the formula, is the radial position Prestrain at is the Poisson’s ratio of rubber material, ranging from 0.48 to 0.5.

[0093] The input parameters include the initial elastic modulus of the pre-vulcanized rubber (usually 4~6 MPa), the expected working pressure range (1~30 MPa), the geometric size parameters of the seal (inner diameter, outer diameter, cross-sectional shape, etc.), the contact surface friction coefficient (usually 0.2~0.4) and the allowable deformation range (the maximum deformation does not exceed 25% of the original size). The optimization calculation uses adaptive mesh refinement technology to improve the calculation accuracy in the key areas of the seal. The calculation results guide the adjustment of the circumferential prestress, which is set to 7.5 MPa, and then the diameter is changed to make the rubber form different pre-deformation states at different radial positions. The diameter change treatment adopts a segmented control strategy, and the pre-strain rate of the inner layer area is controlled at 81~2%, the middle layer area is controlled at 58%, and the outer layer area is controlled at 35%. Through this step, the rubber seal is given a gradient deformation characteristic, so that it can show adaptive deformation ability under different pressure conditions.

[0094] The specific implementation of steps S04-S06 is the same as above and will not be repeated here.

[0095] The specific implementation method of step S07 is to perform nonlinear deformation area distribution processing on the rubber seal after primary vulcanization. First, three-dimensional laser scanning modeling of the seal is performed to obtain its precise geometric data. Based on the geometric model, a finite element analysis model of the seal is constructed, and the grid size is controlled at 0.2~0.5 mm. The key areas of the seal are defined, including the sealing contact surface, the stress concentration area and the deformation transition area. The Pareto optimization method is used to construct a multi-objective optimization model, and the optimization objectives include three mutually constrained objectives: uniformity of sealing contact pressure, deformation recovery rate and rationality of stress distribution. The multi-objective optimization function is expressed as follows: ; In the formula, Optimize function vectors for multiple objectives; is the decision variable vector, which includes the cross-linking density, filler distribution density and prestress distribution of each region; is the seal contact pressure uniformity objective function; is the deformation recovery rate objective function; is the objective function of stress distribution rationality.

[0096] Sealing contact pressure uniformity objective function: ; In the formula, For the Sealing pressure at each contact point; is the average sealing pressure of all contact points; is the total number of contact points; is the trade-off coefficient, and its value range is 0.2~0.5.

[0097] Deformation recovery rate objective function: ; In the formula, For the Residual deformation of a region; For the The maximum deformation of a region; For the The area of ​​the region; is the total number of regions.

[0098] Stress distribution rationality objective function: ; In the formula, For the Region stress components; is the corresponding stress limit; is the number of stress components considered; For the The weight coefficient of the stress component; For the The average paradigm equivalent stress of the area; is the optimal stress ratio of adjacent regions, ranging from 1.2 to 1.5; is the stress gradient weight coefficient, and its value range is 0.3~0.6.

[0099] Comprehensive objective function: ; In the formula, is the comprehensive objective function; After normalization, The objective function value; are weight coefficients, which are 0.4, 0.35, and 0.25 respectively.

[0100] Normalization of objective function: ; In the formula, and Respectively The minimum and maximum values ​​of the objective function in the feasible region.

[0101] Mathematical expression of genetic algorithm in Pareto optimization: ; In the formula, For the Generation population; For the Generation population; is the selection operator; is the crossover operator; is the mutation operator.

[0102] The selection operator calculates the fitness of the individual based on the objective function value: ; In the formula, For individuals The fitness value of For individuals The comprehensive objective function value.

[0103] Using the roulette wheel selection method, individuals The probability of being selected is: ; In the formula, For individuals The probability of being selected; is the population size, which is set to 100.

[0104] The TOPSIS method (top-optimal solution ranking technique) is used to select the optimal solution from the Pareto optimal solution set: First calculate the positive and negative ideal solutions: ; ; In the formula, For the A positive ideal solution to the objective function; For the Negative ideal solution of the objective function; is a solution in the Pareto optimal solution set.

[0105] Compute the distance of each solution to the positive and negative ideal solutions: ; ; In the formula, For solution The distance to the positive ideal solution; For solution The distance to the negative ideal solution; For the The weight of the objective function.

[0106] Compute the closeness coefficient for each solution: ; In the formula, For solution The proximity coefficient ranges from 0 to 1. The larger the value, the better the solution.

[0107] Selection of the optimal solution: ; In the formula, is the optimal solution finally selected.

[0108] The multi-objective optimization problem was solved by genetic algorithm, with the population size set to 100, the number of iterations to 500, the crossover probability to 0.8, and the mutation probability to 0.05. After obtaining the Pareto optimal solution set, the TOPSIS method was used to select the solution with the best comprehensive performance and determine the stress application parameters for each region. According to the optimization results, heat treatment and mechanical deformation treatment were applied to different regions of the seal to form a gradient distribution of cross-linking density and filler distribution. The purpose of this step is to achieve different deformation characteristics of the seal in different regions through multi-objective optimization and regionalization, so as to obtain the optimal gradient strain response characteristics.

[0109] The specific implementation method of step S08 is to place the rubber seal after the nonlinear deformation area distribution treatment in a precision temperature-controlled pressure vulcanization tank for secondary vulcanization treatment. Nitrogen is filled into the pressure vulcanization tank, and the pressure is controlled at 0.3 MPa. The optimal vulcanization time and temperature curve is determined by a pre-trained seal strain response prediction network model. The network model is based on a hybrid architecture of a graph convolutional network and a recurrent neural network, and includes 3 layers of graph convolutional layers and 2 layers of bidirectional long short-term memory network layers. The optimization equation of the vulcanization time and temperature curve is specifically expressed as follows: ; In the formula, For time The vulcanization temperature at Is the basic vulcanization temperature, the value range is 110~120℃; is the initial temperature increase, ranging from 5 to 15°C; It is the temperature increase in the later stage, and the value range is 10~20℃; is the initial heating rate coefficient, ranging from 0.05 to 0.1 / min; is the late heating rate coefficient, ranging from 0.02 to 0.04 / min; It is the turning point time of the temperature curve, and its value range is 30~45 minutes.

[0110] Based on the temperature curve, predict the development of crosslinking degree: ; In the formula, For time The degree of cross-linking at is the maximum possible cross-linking degree, ranging from 0.9 to 0.98; is the temperature-dependent reaction rate constant; is the reaction order, ranging from 1.2 to 1.8.

[0111] ; In the formula, is the pre-exponential factor, the value range is 10 5 ~10 7 / minute; is the activation energy, ranging from 80 to 120 kJ / mol; is the gas constant, 8.314 J / (mol·K); is the Kelvin temperature.

[0112] Based on the predicted degree of crosslinking, calculate the optimal vulcanization time: ; In the formula, is the optimal vulcanization time; is the target cross-linking degree, ranging from 0.85 to 0.95; is the cross-linking rate threshold, ranging from 0.001 to 0.003 / min.

[0113] The graph convolutional network uses an edge attention mechanism to capture the spatial correlation between different parts of the seal. The recurrent neural network uses a gating mechanism to effectively model the dynamic deformation behavior of rubber seals under pressure cycles. The model input includes the material formula parameters, geometric dimension data, primary vulcanization state parameters and expected working conditions of the seal. The model outputs the vulcanization temperature-time curve and the predicted deformation-pressure relationship curve. According to the model prediction, the secondary vulcanization temperature is set to 130℃ and the vulcanization time is 150 minutes. The vulcanization process adopts a segmented temperature control strategy, preheating at 120℃ for the first 30 minutes, then maintaining at 130℃ for 90 minutes, and finally completing deep crosslinking at 140℃ for 30 minutes. The resistance change of the seal is monitored in real time during the vulcanization process. When the resistance value stabilizes at a constant value, it indicates that the crosslinking reaction is basically completed. After the secondary vulcanization is completed, the seal is slowly cooled to room temperature, and the cooling rate is controlled at 1~2℃ / minute to avoid internal stress concentration caused by rapid cooling. The purpose of this step is to complete the construction of the final cross-linked network of the rubber seal through a controlled secondary vulcanization process, so that it has stable nonlinear deformation characteristics and adaptive sealing capabilities.

[0114] The specific implementation method of step S09 is to conduct a comprehensive pressure adaptability performance test on the prepared rubber seal. First, the seal is installed on a sealing performance test device, which can provide an adjustable pressure in the range of 1 to 30 MPa and a temperature environment simulating actual working conditions. The pre-trained seal strain response prediction network model is used to generate an ideal pressure deformation curve as a reference standard. The mathematical expression of the ideal pressure deformation curve is as follows: ; In the formula, For pressure The ideal deformation under ; is the initial deformation, ranging from 0.05 to 0.2 mm; is the power function coefficient, ranging from 0.01 to 0.05; is the power exponent, ranging from 0.6 to 0.8; is the saturation deformation coefficient, and its value range is 0.5~1.5 mm; is the deformation saturation rate coefficient, and its value range is 0.1~0.3 / MPa.

[0115] This ideal curve describes the optimal deformation state of the seal at each pressure point and forms a benchmark for performance evaluation. The test process adopts a step-by-step pressurization scheme, starting from 1 MPa, increasing by 2 MPa each time until 30 MPa, and then decreasing to 1 MPa in the same step length, forming a complete loading-unloading cycle. Stay at each pressure point for 3 minutes, and record the deformation amount, contact pressure distribution and leakage rate of the seal. Perform 20 consecutive cyclic loading tests to evaluate the durability and stability of the seal. Based on the ideal deformation curve, calculate the nonlinear deformation recovery rate: ; In the formula, is the nonlinear deformation recovery rate; is the deformation under maximum pressure; is the residual deformation after unloading.

[0116] Calculate the sealing performance retention rate: ; In the formula, is the sealing performance retention rate; is the sealing efficiency of the first cycle test; It is the sealing efficiency of the 20th cycle test.

[0117] Sealing efficiency calculation: ; In the formula, For sealing efficiency; is the measured leakage rate; This is the standard allowable leakage rate.

[0118] The nonlinear deformation recovery rate is required to be greater than 95%, and the sealing performance retention rate is required to be greater than 98%. The temperature distribution of the seal during the pressure cycle is monitored by an infrared thermal imager to evaluate its energy dissipation characteristics. Digital image correlation technology is used to analyze the local deformation distribution of the seal to verify its gradient strain response characteristics. The purpose of this step is to verify the pressure adaptability and long-term stability of the rubber seal through rigorous performance testing to ensure that it meets the requirements of engineering applications.

[0119] In order to better understand and implement the present invention, Example 2 of a specific application scenario of the present invention is provided below: In a certain shaft sealing device, axial sealing is its key subsystem, which needs to maintain long-term stable sealing performance under high temperature, high pressure and irradiation environment. The operating pressure range of the shaft sealing system is 2~28 MPa, the temperature range is 60~120℃, and it needs to withstand certain radial vibration and axial movement. Traditional sealing structures have problems such as unstable sealing, excessive deformation and premature failure under pressure fluctuation conditions. Using the pressure-adaptive nonlinear deformation rubber seal preparation method of the present invention, researchers have successfully developed a new type of sealing assembly. The specific implementation process is as follows.

[0120] First, the researchers selected nitrile rubber and EPDM in a mass ratio of 5.5:4.5 according to step S01, added 3.2% of nano-silica filler (particle size 45 nanometers) and 2.2% of pressure-sensitive deformation trigger. In a closed mixer, the temperature was set to 58°C, the speed was 50 rpm, and the mixing time was 42 minutes (maintained at 54°C for the first 22 minutes and heated to 58°C for the last 20 minutes) to form a basic rubber mixture.

[0121] According to step S02, 1.4% of tetramethylthiuram disulfide as a vulcanizing agent and 0.9% of N-cyclohexyl-2-benzothiazole sulfenamide as an accelerator are added to the base rubber mixture. In the open rubber mixer, the temperature is controlled at 43°C, the speed is 22 rpm, the mixing time is 18 minutes, and the machine is stopped once every 4.5 minutes to perform scraping and turning operations to obtain a uniformly dispersed pre-vulcanized rubber compound. The material formula composition is shown in Table 1:

[0122] Table 1 Rubber seal material formula composition table

[0123] According to step S03, the pre-vulcanized rubber is placed in a special pressure-strain adjustment chamber, and the elastic modulus distribution optimization function is applied to calculate the optimal circumferential prestress distribution. According to the geometric parameters of the seal with an inner diameter of 32 mm and an outer diameter of 52 mm, the radial elastic modulus distribution is calculated. The circumferential prestress is finally set to 7.2 MPa. Based on this, the seal is subjected to diameter reduction treatment, and the prestrain rate of the inner layer area is 10.3%, the middle layer area is 6.8%, and the outer layer area is 4.2%. Figure 2 The radial elastic modulus distribution and circumferential prestress distribution curve of the rubber seal are shown. The blue curve shows the change of elastic modulus E(r) with radial position r. From the inner diameter of 32 mm to the outer diameter of 52 mm, the elastic modulus shows a nonlinear growth trend. The red curve shows the circumferential prestress The distribution fluctuates with the radial position, reflecting the differentiated design of the prestress distribution in different areas of the seal. This gradient distribution of elastic modulus and periodic fluctuation of prestress distribution enable the seal to exhibit non-uniform deformation characteristics when under pressure, thereby achieving adaptive sealing under different pressure conditions.

[0124] According to step S04, the researchers designed and produced a double-layer mold system. The inner mold surface was coated with a 18-micron thick polytetrafluoroethylene and molybdenum disulfide composite self-lubricating nano-coating. The outer mold surface was micro-textured to produce a cross-grid micro-texture structure with a depth of 28 microns, a width of 48 microns, and a spacing of 210 microns. The pre-vulcanized rubber was placed in the mold, heated to 152°C at a heating rate of 4°C / min in a flat vulcanizer, and pressurized to 18.5 MPa, with a fluctuation range of ±0.4 MPa.

[0125] According to step S05, high-purity nitrogen with a purity of 99.995% was introduced during the pressing process, the pressure was controlled at 0.32 MPa, the flow rate was 0.8 L / min, and it was maintained for 3.2 hours while maintaining a constant mold temperature and pressure.

[0126] According to step S06, the molded rubber seal is placed in a hot air circulation oven for a vulcanization treatment. The temperature is set to 88°C, the temperature uniformity is controlled within the range of ±1.8°C, and the first 15 minutes are preheated at 80°C, and then maintained at 88°C for 65 minutes. During the vulcanization process, the Mooney viscosity of the rubber is tested every 15 minutes, and the vulcanization is terminated when the Mooney viscosity reaches 2.8 times the initial value. The final Mooney viscosity test results are shown in Table 2: Table 2 Changes in Mooney viscosity during primary vulcanization

[0127] According to step S07, the nonlinear deformation area distribution processing is performed on the rubber seal after primary vulcanization. First, three-dimensional laser scanning modeling is performed to construct a finite element analysis model, and the grid size is controlled at 0.35 mm. The Pareto optimization method is used to construct a multi-objective optimization model, including three optimization goals: uniformity of sealing contact pressure, deformation recovery rate, and rationality of stress distribution. The optimization variables include the cross-linking density, filler distribution density, and prestress distribution in each region. The constraints include that the maximum stress does not exceed 68% of the ultimate strength of the material, and the maximum deformation does not exceed 32% of the original size. The multi-objective optimization problem is solved by a genetic algorithm with a population size of 100, a number of iterations of 480, a crossover probability of 0.82, and a mutation probability of 0.045. The comparison of sealing performance parameters before and after optimization is shown in Table 3: Table 3 Comparison of sealing performance parameters before and after optimization

[0128] According to step S08, the rubber seal after the nonlinear deformation area distribution treatment is placed in a precision temperature-controlled pressure vulcanization tank for secondary vulcanization treatment. Nitrogen is filled into the pressure vulcanization tank, and the pressure is controlled at 0.32 MPa. The optimal vulcanization time and temperature curve are determined by the pre-trained seal strain response prediction network model. The vulcanization temperature-time curve is calculated. The secondary vulcanization process adopts a three-stage temperature control strategy, preheating at 122°C for the first 28 minutes, then maintaining at 128°C for 92 minutes, and finally completing deep crosslinking at 138°C for 32 minutes, with a total vulcanization time of 152 minutes. The development of the degree of crosslinking is predicted. When the degree of crosslinking reaches 0.92 and the crosslinking rate drops below 0.0018 / minute, the optimal vulcanization time is determined to be 152 minutes. Figure 3 The temperature-time curve and cross-linking degree development curve during the secondary vulcanization process are shown. The red curve shows the change of vulcanization temperature T(t) with time t. The initial temperature is high, then slowly decreases, then gradually increases, and finally reaches a stable state. The blue curve shows the development process of the cross-linking degree α(t), which is a typical S-shaped curve, slow in the early stage, accelerated in the middle stage, and tends to saturation in the later stage. Three vulcanization stages are marked in the figure: preheating stage (0-28 minutes), holding stage (28-120 minutes) and deep cross-linking stage (120-152 minutes). The optimal vulcanization time point was determined to be 152 minutes, when the cross-linking degree reached 0.92 and the cross-linking rate dropped to a sufficiently low level, indicating that the cross-linking reaction was basically completed.

[0129] According to step S09, a comprehensive pressure adaptability test is performed on the prepared rubber seal. The pre-trained seal strain response prediction network model is used to generate an ideal pressure deformation curve: The test process starts from 2 MPa, increases by 2 MPa each time until 28 MPa, and then decreases to 2 MPa with the same step length, forming a complete loading-unloading cycle. Stay at each pressure point for 3 minutes, and record the deformation, contact pressure distribution and leakage rate of the seal. After 20 consecutive cyclic loading tests, the nonlinear deformation recovery rate is 97.3% and the sealing performance retention rate is 99.1%. The sealing performance test results are shown in Table 4: Table 4 Sealing performance test results

[0130] The rubber seal has undergone 5,000 hours of long-term stability testing and, under harsh conditions of working pressure of 25 MPa and temperature of 110°C, the sealing performance retention rate still reached 96.8%, demonstrating excellent durability. Figure 4 The pressure deformation characteristic curve of the adaptive nonlinear deformation rubber seal is shown. The blue solid line represents the ideal deformation curve of the seal of the present invention; the red dotted line represents the deformation curve of the traditional linear seal; and the green scattered points represent the actual test data. It can be seen from the figure that in the low-pressure area (0-15 MPa), the deformation amount of the adaptive seal is similar to that of the traditional seal, maintaining good initial sealing performance; while in the high-pressure area (15-30 MPa), the deformation increment of the adaptive seal is significantly reduced, showing strain hardening characteristics, effectively preventing excessive deformation. The deformation recovery rate is also marked in the figure as 97.3%, which is higher than that of the traditional seal, indicating that the seal of the present invention has excellent deformation recovery ability. Compared with the traditional rubber seal, the adaptive nonlinear deformation rubber seal of the present invention has a significantly reduced increase in the deformation amount in the high-pressure area, showing obvious strain hardening characteristics; in the low-pressure area, it maintains a high compliance to ensure good initial sealing performance. During the actual operation of 8 months, no leakage occurred, and the maintenance cycle was extended from the original 3 months to 8 months, which greatly improved the operating reliability and economy of the system.

[0131] There are three main technical problems in the traditional preparation method of rubber seals: first, the sealing material exhibits linear or nearly linear deformation characteristics under different pressure conditions, which leads to excessive compression deformation under high pressure conditions, causing damage to the seal; second, the traditional seal has a uniform structure and lacks regional design, and cannot be optimized for the stress conditions of different parts; third, the conventional vulcanization process cannot accurately control the formation process of the cross-linked network, resulting in large fluctuations in the performance of the seal. The preparation method of the pressure-adaptive nonlinear deformation rubber seal of the present invention successfully solves the above problems by introducing a pressure-sensitive deformation trigger, using an elastic modulus distribution optimization function to design a gradient structure, using a multi-objective optimization model to determine the nonlinear deformation regional distribution, and using a pre-trained seal strain response prediction network model to guide the vulcanization process. Compared with the traditional technology, the present invention has the following obvious improvements: the seal exhibits good flexibility in the low-pressure area to ensure the initial sealing effect; it exhibits nonlinear hardening characteristics in the high-pressure area to prevent excessive deformation; it has a stress memory effect and can recover to a preset deformation state after being compressed; it forms a gradient distribution of cross-linking density and filler distribution, so that the seal has different deformation characteristics in different areas; the sealing performance is stable and reliable, and the service life is significantly extended. These improvements make the rubber seal of the present invention particularly suitable for harsh environments with large pressure fluctuations and complex working conditions.

[0132] It should be noted that the variables involved in the present invention are explained in detail as shown in Tables 5 and 6 below.

[0133] Table 5 Variable explanation table (Part I)

[0134] Table 6 Variable Explanation Table (Part 2)

[0135] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for preparing a pressure-adaptive nonlinear deformation rubber seal, characterized in that: The method comprises the following steps: mixing nitrile rubber and EPDM, adding nano-silica filler and pressure-sensitive deformation triggering agent for kneading to form a basic rubber mixture; adding vulcanizing agent and accelerator to the basic rubber mixture for further kneading to obtain a pre-vulcanized rubber compound; placing the pre-vulcanized rubber compound in a pressure-strain regulating cavity, calculating the optimal circumferential prestress distribution through an elastic modulus distribution optimization function for diameter reduction treatment, so that the rubber seal exhibits gradient deformation characteristics under different pressure conditions; and adopting a double-layer mold for compression molding; During the pressing process, nitrogen is introduced for atmosphere protection; the rubber seal is subjected to a primary vulcanization treatment to form a primary cross-linking network; the rubber seal after the primary vulcanization is subjected to nonlinear deformation regional distribution treatment, a multi-objective optimization model of the relationship between seal stress distribution and deformation is constructed, and the stress application parameters of each region are determined by solving; Perform secondary vulcanization treatment to complete the final cross-linking network construction; The prepared rubber seals are subjected to pressure adaptability tests.

2. The method for preparing the pressure-adaptive nonlinear deformation rubber seal according to claim 1, characterized in that: Nitrile rubber and EPDM are mixed in a mass ratio of 4:6 to 6:4, and 2% to 5% by mass of nano-silica filler and 1% to 3% by mass of pressure-sensitive deformation trigger are added. The mixture is kneaded at 50 to 70°C for 30 to 45 minutes to form a basic rubber mixture.

3. The method for preparing the pressure-adaptive nonlinear deformation rubber seal according to claim 2, characterized in that: Add 1% to 2% by mass of a vulcanizer and 0.5% to 1.5% by mass of an accelerator to the basic rubber mixture, and continue mixing for 15 to 25 minutes to obtain a uniformly dispersed pre-vulcanized rubber compound.

4. The method for preparing the pressure-adaptive nonlinear deformation rubber seal according to claim 3, characterized in that: The optimal circumferential prestress distribution is calculated through the elastic modulus distribution optimization function to perform diameter variation treatment. Specifically, the optimal circumferential prestress distribution is calculated through the elastic modulus distribution optimization function, the circumferential prestress range is adjusted to 5 to 10 MPa, and the diameter variation treatment is performed to make the rubber seal exhibit gradient deformation characteristics under different pressure conditions.

5. The method for preparing the pressure-adaptive nonlinear deformation rubber seal according to claim 4, characterized in that: The double-layer mold pressing molding is specifically, wherein the surface of the inner mold is coated with a self-lubricating nano-coating, and the outer mold is micro-textured, the pressing temperature is controlled at 140 to 160°C, and the pressure is 15 to 25 MPa.

6. The method for preparing the pressure-adaptive nonlinear deformation rubber seal according to claim 5, characterized in that: During the pressing process, nitrogen is introduced for atmosphere protection. Specifically, the air pressure is controlled at 0.2 to 0.5 MPa and maintained for 2 to 4 hours to form an internal microscopic stress network structure for the rubber seal.

7. The method for preparing the pressure-adaptive nonlinear deformation rubber seal according to claim 6, characterized in that: The rubber seal is subjected to a vulcanization treatment to form a primary cross-linking network. Specifically, the molded rubber seal is subjected to a vulcanization treatment at a temperature of 80 to 100° C. for 60 to 90 minutes to form a primary cross-linking network.

8. The method for preparing the pressure-adaptive nonlinear deformation rubber seal according to claim 7, characterized in that: The secondary vulcanization treatment is carried out to complete the final cross-linking network construction. Specifically, the treated rubber seal is placed under 120 to 140°C conditions for secondary vulcanization treatment, and the pre-trained seal strain response prediction network model is used to determine the optimal vulcanization time and temperature curve. The vulcanization time range is controlled within 120 to 180 minutes to complete the final cross-linking network construction.

9. The method for preparing the pressure-adaptive nonlinear deformation rubber seal according to claim 8, characterized in that: The pressure adaptability performance test of the prepared rubber seals includes using the ideal pressure deformation curve generated by the pre-trained seal strain response prediction network model as a reference standard, cyclically loading 20 times in the pressure range of 1 to 30 MPa, and checking that its nonlinear deformation recovery rate should be greater than 95%, and the sealing performance retention rate should be greater than 98%.

10. The method for preparing the pressure-adaptive nonlinear deformation rubber seal according to claim 9, characterized in that: The elastic modulus distribution optimization function is used to calculate the optimal elastic modulus distribution gradient according to the working environment pressure range and material properties of the seal. The input includes the initial elastic modulus of the pre-vulcanized rubber, the expected working pressure range, the geometric size parameters of the rubber seal, the contact surface friction coefficient and the allowable deformation range. The output is the optimal elastic modulus distribution curve along the radial distribution of the rubber seal and the corresponding circumferential prestressing scheme.

Images