High-reliability equal-strength design method and system for intelligent cubic press

Abstract

The invention relates to the technical field of equal-strength design, and provides a high-reliability equal-strength design method and system for an intelligent cubic press, which considers structural characteristics of the cubic press, performs physical field analysis, modeling and simulation, determines deformation, temperature, stress distribution and stress concentration areas of key components, predicts structural crack initiation of the key components, and provides a high-reliability equal-strength design method and system for an intelligent cubic press. Crack propagation simulation is carried out; establishing a fatigue strength life prediction model based on deep learning and a simulation data set; constructing a fatigue strength reliability model, and establishing a mathematical association relationship among fatigue life, fatigue damage and reliability; a reliability calculation index is introduced into multi-objective optimization of the structure, a high-precision numerical solution of reliability robustness optimization design based on structure simulation is developed, and equal-strength design of the cubic press is achieved. According to the method, the optimal combination of design parameters can be obtained, the equal-strength design of parts and the whole machine structure is formed, and development of intelligent, high-precision and large-scale cubic press equipment is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of equal strength design, and in particular to a high-reliability equal strength design method and system for an intelligent six-sided top press. Background Art

[0002] As China's most important diamond synthesis equipment, the six-sided top press is a typical static, high-temperature, ultra-high-pressure system. With the development and application of technologies such as microelectronics and precision hydraulic control, the scale-up of presses has accelerated, significantly improving their performance. However, the overall production efficiency of existing six-sided top presses remains relatively low, far behind the high output of two-sided top presses. The equipment's poor precision retention makes it difficult to meet the demands of high-quality diamond synthesis. The industry's commonly used six-sided top presses with cast hinged beams face significant obstacles in meeting the requirements of larger equipment sizes. These include excessive size, potential casting defects that lead to high press failure rates, difficulty in precisely controlling the structural material composition, poor performance stability, and short fatigue life, making intelligent synthesis difficult. As the press's cylinder diameter increases, the increasing operating pressures and temperatures place higher demands on the fatigue life reliability, centering accuracy, and overall control precision of the hinged beam and top hammer structures. This necessitates accurate assessment of the service performance and life reliability of key components (the hinged beam and top hammer) during the scale-up process, optimization of the components' structure, and improvement of the equipment's design and manufacturing capabilities.

[0003] The finite element method has become the most widely used numerical method in structural design and analysis, providing an accurate and reliable method for the design of the hinge beam structure of the six-sided top press. Through accurate finite element analysis, the actual characteristics of the hinge beam can be reflected, the stress and deformation distribution in its key areas can be analyzed, and the reliability of the hinge beam structure can be improved to ensure that the designed hinge beam structure is truly feasible in actual production. Sun Xuan et al. (Sun Xuan, Wang Bin, Zhang Yu, Zhou Fuping, et al. Strength calculation of the hinge beam of the six-sided top hydraulic press [J]. Journal of Guilin University of Technology, 2013, 33(02): 322-324.) used ANSYS to construct a finite element analysis model of the hinge beam structure, analyzed and calculated the hinge beam structure of the six-sided top press, and analyzed the reasons why the hinge beam lug and bottom may break. Li Ruiting et al. (Li Ruiting, Zhang Fuchao, Wang Qinmeng. Finite Element Analysis of Diamond Six-sided Press Strength [C] / / The 18th Annual Academic Conference of Beijing Mechanics Society. 2012: 377-378.) conducted a finite element numerical simulation of the diamond six-sided press, obtained the overall stress distribution characteristics and stress concentration areas of the diamond six-sided press, and analyzed the reasons for the frequent damage at the lug position. Wang Liangwen et al. (Wang Liangwen, Liu Jianshe, Wang Shuguang, et al. Finite Element Analysis of UDS-III Diamond Press Structure [J]. Mechanical Design, 2005, 22(03): 38-41.) used the finite element analysis method to simulate the actual working conditions of the press and conducted a comprehensive theoretical analysis of its strength performance, providing a comprehensive theoretical reference for the design and development of large-tonnage presses.

[0004] However, due to the unique high-temperature and ultra-high-pressure operating conditions of the equipment, finite element analysis of hexapod presses is often limited to the press's components. This "focus on a single point" approach makes it difficult to capture the multi-field coupled physical characteristics of high-temperature and ultra-high pressure, resulting in a "simulation without authenticity." Interactions between components and between components and the overall equipment are difficult to account for, resulting in an "incomplete simulation." This directly hinders the accurate characterization of the equipment's internal mechanical parameters. Structural performance is limited to offline analysis, lacking interaction and integration with Industrial Internet of Things data, leading to "overgeneralization." The coupling of multiple physical fields introduces numerous nonlinear factors, and temperature and pressure fluctuations also introduce uncertainties. Simulating crack propagation behavior within key components is difficult, necessitating the development of high-precision numerical simulation algorithms. To address the challenges of low fatigue strength and poor operational stability of traditional cast hinged beam hexapod presses, the development of high-reliability and strength design technologies for intelligent hexapod presses is crucial to improving the design and manufacturing of hexapod presses and enhancing the competitiveness of my country's superhard materials industry. Summary of the Invention

[0005] The purpose of the present invention is to provide a high-reliability and equal-strength design method and system for an intelligent six-sided top press. Taking into account the structural characteristics of the six-sided top press, multi-physical field coupling analysis, modeling, and simulation are carried out, and a fatigue strength life prediction model based on deep learning and simulation data sets is established. Reliability calculation indicators are introduced into the multi-objective optimization of the equipment structure to achieve reliability and robustness optimization design of the equipment, and promote the development of intelligent, high-precision, and large-scale six-sided top press equipment.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] The present invention proposes a high-reliability and equal-strength design method for an intelligent six-sided top press, comprising the following steps:

[0008] (1) Structural characteristics analysis and multi-physics field modeling: Based on the structural characteristics of the key components of the six-sided top press, a multi-physics field coupling simulation model including deformation field, temperature field, stress field and rheological field is established;

[0009] (2) Determination of crack initiation risk locations: Based on the multi-physics field coupling simulation model, a component-whole machine multi-physics field coupling simulation analysis is performed. The stress concentration area of ​​the key components is determined as the crack initiation risk location through a hybrid solution of the finite element method and the boundary element method.

[0010] (3) Crack initiation and propagation simulation: The stress concentration area of ​​the key components is subjected to thermal alternating load cycles. Based on the continuous damage mechanics method, a material damage evolution model of the area is constructed to calculate its fatigue cumulative damage. When the fatigue cumulative damage reaches the critical value, the crack initiation is determined, and the crack propagation simulation of the crack initiation area is performed using the boundary element and finite element coupling method;

[0011] (4) Construction of fatigue strength life prediction model: obtain the fracture mechanics parameter simulation data set, and establish a proxy mathematical model of fracture mechanics based on the deep neural network algorithm; integrate the IoT monitoring data of the six-sided top press, the proxy mathematical model of fracture mechanics and the crack propagation simulation results to construct a structural crack propagation model, and combine the dynamic Bayesian network and particle filter algorithm to improve the crack propagation prediction model and obtain the fatigue strength life prediction model;

[0012] (5) Fatigue strength reliability model construction: Based on the fatigue strength life prediction model, the IoT monitoring data, historical operation data and multi-physics field coupling simulation model of the six-sided top press are integrated to construct a fatigue strength reliability model and determine the mathematical correlation between fatigue life, fatigue damage and reliability;

[0013] (6) Reliability-driven multi-objective optimization design: Based on the mathematical correlation, the reliability calculation index is introduced into the multi-objective optimization of the key components of the six-screw press, a high-precision numerical solution algorithm based on Kriging surrogate model is developed, the optimal combination of design parameters is obtained, and the equal strength design of the key components and the whole structure of the six-screw press is formed.

[0014] Further, the design method further comprises:

[0015] Structural fatigue crack test verification: using wire cutting technology to prepare test pieces of key components, respectively, to carry out tensile test and fatigue test, to obtain the constitutive relation, S-N curve and fatigue life of the material of the key components, and to compare and verify the simulation results of the multi-physical field coupling simulation model, the prediction results of the fatigue strength life prediction model, and to improve the multi-physical field coupling simulation model and the fatigue strength life prediction model.

[0016] Further, the key components include hinge beam structure, top hammer structure and synthetic block structure.

[0017] Further, the step (1), structure characteristic analysis and multi-physical field modeling, specifically comprises:

[0018] Based on the structure characteristics of the hinge beam structure, the top hammer structure and the synthetic block structure of the six-screw press, a multi-physical field coupling simulation model containing deformation field, temperature field, stress field and rheological field is established;

[0019] Wherein, for the hinge beam structure, a three-dimensional geometric model of the hinge beam structure is constructed, including lugs, oil holes and pin shaft matching areas, local grid self-adaptive encryption technology is used for lug chamfer and oil hole fillet to avoid calculation grid distortion; define the material constitutive model of the hinge beam, and set the boundary conditions, including contact nonlinear analysis of the contact surface of the pin shaft and the lug, dynamic loading path of the oil pressure load of the hydraulic cylinder and symmetry constraint of the structure; a mechanical model of the hinge beam structure is established, and a deformation field, stress field and temperature field coupling simulation model is obtained;

[0020] For the top hammer structure, a top hammer structure model is constructed, the top hammer structure includes a hard alloy top hammer, a steel ring and a pad, wherein the steel ring and the hard alloy top hammer are connected through interference fit to apply radial pre-tightening force; a thermal-mechanical coupling simulation model of the top hammer is established to simulate the stress state of the top hammer in the ultra-high pressure sealed cavity, including compressive stress, shear stress, thermal stress and additional stress caused by synthetic block bias;

[0021] A geometric model of the composite block assembly was established for the composite block structure, including pyrophyllite, conductive components, and diamond growth chamber. A time-dependent function of the heating temperature was defined, and the pressure load on the composite block was calculated based on the data from the hydraulic system pressure sensor. Combined with the material rheological constitutive model, the finite element method was used to solve the coupled temperature, pressure, and rheological fields of the composite block.

[0022] The deformation field, stress field and temperature field coupling simulation model of the integrated hinge beam, the thermal-mechanical coupling model of the top hammer structure, the temperature field, pressure field and rheological field coupling model of the synthetic block are obtained to obtain the multi-physics field coupling simulation model of the six-sided top press.

[0023] Furthermore, based on the multi-physics coupling simulation model, component-whole machine multi-physics coupling simulation analysis is performed, specifically including:

[0024] Obtain the mechanical model and physical properties of the hinged beam structure, perform static analysis of the hinged beam, apply alternating loads and mesh discretization, configure the boundary element method, and solve;

[0025] A heating mathematical model of the anvil structure was established, and the boundary conditions of the anvil structure were analyzed. Based on the characteristics of multi-layer dielectric materials, it was determined that the carbide anvil was the area where stress concentration would cause cracks. The thermal-mechanical coupled boundary element method was used to analyze the thermal temperature field of the carbide anvil. The finite element method was used to analyze the steel ring and spacer, and the final results were coupled.

[0026] Based on the establishment of the geometric model and physical model of the synthetic block, the carbon convection diffusion field is analyzed, and the electric-thermal-mechanical-fluid coupling analysis and calculation are carried out to study the influence of carbon convection in the synthetic block and reveal the formation mechanism of growth defects on the surface of diamond synthesized by the temperature gradient method.

[0027] In the IoT system of the six-sided top press, real operating data collected by various sensors is compared with the simulation analysis results to verify the accuracy of the simulation results.

[0028] Furthermore, in step (3), the crack propagation simulation of the crack initiation area is performed using a boundary element and finite element coupling method, specifically including:

[0029] First, the sub-regions are divided. The crack region belongs to the stress singularity concentration area and is solved by the boundary element method. The area far from the crack is solved by the finite element method.

[0030] Then, the crack region is meshed and discretized, and boundary integral equations are configured, matrices assembled and solved;

[0031] Finally, the stress intensity factor, crack growth angle and crack growth direction are calculated to determine whether the crack has grown to the critical length and the fatigue life.

[0032] Furthermore, in step (4), a fracture mechanics parameter simulation data set is obtained, and a proxy mathematical model of fracture mechanics is established based on a deep neural network algorithm, specifically including:

[0033] A priori distribution model of the initial crack size and crack propagation parameters of the six-sided top press structure was established. Using the finite element and boundary element coupling method, fracture mechanics performance parameters including stress intensity factor and T stress were obtained, and a fracture mechanics parameter simulation data set was generated.

[0034] Based on the deep neural network algorithm, a proxy mathematical model of fracture mechanics is established, which includes crack propagation size, crack propagation parameters, crack propagation life and temperature, displacement, stress and strain of measurement points on key components.

[0035] Furthermore, in step (4), the crack growth prediction model is improved by combining the dynamic Bayesian network and the particle filter algorithm to obtain a fatigue strength life prediction model, which specifically includes:

[0036] By integrating IoT monitoring data, historical operation data, and multi-physics field coupling simulation models, a dynamic Bayesian network prediction model for crack size is constructed by considering three different types of dynamic Bayesian model parameters: deterministic parameters, model-specific uncertain parameters, and monitoring data.

[0037] After completing the Bayesian network prediction model, due to the existence of multiple uncertain variables, the model prediction will gradually deviate from the actual value. The particle filter inference algorithm is used to update the random variable nodes in the Bayesian network prediction model, probabilistically predict the crack expansion size, and obtain online crack expansion data. By inputting it into the Bayesian network prediction model and comparing it with the IoT monitoring data and historical operation data, the initial digital twin model is diagnosed and updated to achieve real-time, efficient, and intelligent analysis and prediction of fatigue crack location and expansion trend, and obtain a fatigue strength life prediction model.

[0038] Furthermore, the step (5), constructing the fatigue strength reliability model, specifically includes:

[0039] In the fatigue strength life prediction model, the company integrates IoT monitoring data, historical operation data, and multi-physics field coupling simulation models. Based on the fatigue life prediction method of damage mechanics and fracture mechanics, it uses Bayesian theory to quantify key parameters of equipment, such as structural parameters, material parameters, and stress load amplitude parameters, to overcome the random characteristics of uncertainty that fluctuates over time and space.

[0040] Determine the statistical characteristics of random parameters, build a mathematical relationship model between key parameters and fatigue life, obtain the probability distribution of parameters and the changing trend of reliability, unify them into the distribution of fatigue life, and build the corresponding relationship between fatigue life and fatigue damage;

[0041] Analyze the limit state of the hinge beam structure and the limit state of the top hammer system, consider the load effect and structural resistance of the hinge beam and top hammer structure, and establish the corresponding limit state equations;

[0042] Based on the neural network deep learning technology, sample points in the structural crack propagation simulation are extracted as sample data, the functional function of the fatigue residual life model of the structure is reconstructed, and the optimal proxy functional function is obtained. The fatigue reliability calculation is performed using the first-order second-moment method based on the fitted proxy functional function.

[0043] On the other hand, the present invention further provides a high-reliability and equal-strength design system for an intelligent six-sided top press, which is used to execute the above-mentioned high-reliability and equal-strength design method for an intelligent six-sided top press, comprising:

[0044] Structural characteristics analysis and multi-physics modeling module, used to establish a multi-physics coupling simulation model including deformation field, temperature field, stress field and rheological field based on the structural characteristics of the key components of the six-sided top press;

[0045] The module for determining the location of crack initiation risks is used to conduct component-to-whole-machine multi-physics coupling simulation analysis based on a multi-physics coupling simulation model. By using a hybrid solution of the finite element method and the boundary element method, the stress concentration area of ​​the key components is determined as the location of crack initiation risks.

[0046] The crack initiation and growth simulation module is used to apply thermal alternating load cycles to stress concentration areas of key components. Based on the continuous damage mechanics method, a damage evolution model of the material in this area is constructed to calculate its fatigue cumulative damage. When the fatigue cumulative damage reaches a critical value, crack initiation is determined and crack growth simulation is performed in the crack initiation area using the boundary element and finite element coupling method;

[0047] The fatigue strength and life prediction model construction module is used to obtain a simulation data set of fracture mechanics parameters and establish a proxy mathematical model of fracture mechanics based on a deep neural network algorithm. The module also integrates IoT monitoring data from the six-sided top press, a proxy mathematical model of fracture mechanics, and crack propagation simulation results to construct a structural crack propagation model. This model is then improved by combining a dynamic Bayesian network with a particle filter algorithm to produce a fatigue strength and life prediction model.

[0048] The fatigue strength reliability model construction module is used to build a fatigue strength reliability model based on the fatigue strength life prediction model, integrating the six-sided top press's IoT monitoring data, historical operation data, and multi-physics field coupling simulation model to determine the mathematical correlation between fatigue life, fatigue damage, and reliability;

[0049] A reliability-driven multi-objective optimization design module is used to introduce reliability calculation indicators into the multi-objective optimization of key components of the six-sided top press based on the mathematical correlation relationship, develop a high-precision numerical solution algorithm based on the Kriging proxy model, obtain the optimal combination of design parameters, and form an equal strength design of the key components of the six-sided top press and the entire machine structure.

[0050] Compared with the prior art, the beneficial effects of the present invention can be summarized as follows:

[0051] The present invention provides a high-reliability equal-strength design method and system for an intelligent six-sided top press, which analyzes the multi-physical field coupling characteristics of the key components of the six-sided top press and the entire machine, and realizes high-precision modeling of the multi-physical field coupling of the six-sided top press; constructs a component-entire machine multi-physical field coupling simulation analysis method to determine the deformation, temperature, stress distribution and stress concentration area of ​​the components, and predicts the initiation of cracks in the key component structure; establishes a fatigue strength life prediction model based on deep learning and simulation data sets; constructs a fatigue strength reliability model to establish a mathematical correlation between fatigue life, fatigue damage and reliability; introduces reliability calculation indicators into the multi-objective optimization of the equipment structure, and develops a high-precision numerical solution for reliability robustness optimization design based on equipment structure simulation. Under the premise of meeting the equipment function, installation and connection requirements, and processing technology, the present invention forms equal strength of components and the entire machine structure to achieve lightweight structure; in addition, the press deformation under real-time working conditions can be monitored, and compared and corrected with the deformation of the multi-physical field simulation model to optimize and determine the key parameters of the press structure.

[0052] The present invention can significantly improve the centering performance and precision, equipment stability, high reliability and fatigue life of the press, better meet the requirements of larger six-sided top presses, provide important theoretical and technical support for the design and manufacture of large-scale six-sided top presses in my country, promote the large-scale, precision and intelligent development of six-sided top presses, promote the high-end development of my country's diamond industry, and have important application value for the superhard material synthesis industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 This is a flow chart of the high reliability and equal strength design method of the intelligent six-sided top press in the present invention;

[0055] Figure 2 This is a schematic diagram of the six-sided top press structure and its sensor configuration according to an embodiment of the present invention;

[0056] Figure 3 This is a schematic diagram of the process of simulating crack propagation using the boundary element method according to an embodiment of the present invention;

[0057] Figure 4 This is a schematic diagram of the fatigue fracture prediction process of a six-sided top press driven by deep learning according to an embodiment of the present invention;

[0058] Figure 5 This is a schematic diagram of the tensile test design process according to an embodiment of the present invention;

[0059] Figure 6 This is a schematic diagram of the fatigue test design process according to an embodiment of the present invention;

[0060] Figure 7 This is a schematic diagram of the process of constructing and solving the fatigue reliability model of the six-sided top press according to an embodiment of the present invention with high precision. DETAILED DESCRIPTION

[0061] The following will illustrate the technical solutions and results of the embodiments of the present invention in conjunction with the accompanying drawings. The described embodiments are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0062] The purpose of the present invention is to provide a high-reliability equal strength design method and system for an intelligent six-sided top press. In view of the multi-physical field coupling characteristics involved in the production process of the six-sided top press, the mathematical and geometric model construction and solution of the multi-physical field are analyzed to achieve high-precision modeling of the multi-physical field coupling of the six-sided top press, and to construct a multi-physical field coupling simulation analysis method for components and the whole machine. In view of the fatigue fracture problem involved in the six-sided top press equipment, a prior distribution model of the initial crack size and crack propagation parameters of the equipment based on perception data is constructed, a fracture mechanics simulation data set is generated, a high-precision proxy model of equipment fracture mechanics based on deep learning and simulation data sets is established, a fatigue strength life assessment model of the equipment is constructed, and a multidisciplinary optimization mathematical model based on reliability design indicators is constructed; a high-precision numerical solution for reliability robustness optimization design is developed to obtain the optimal combination of design parameters and form an equal strength design of components and the whole machine structure.

[0063] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0064] like Figure 1 As shown, the present invention proposes a high reliability and equal strength design method for an intelligent six-sided top press, which specifically includes the following steps:

[0065] S1, Structural Characteristics Analysis and Multi-Physical Field Modeling: Based on the structural characteristics of the key components of the six-sided top press, a multi-physical field coupling simulation model including deformation field, temperature field, stress field and rheological field is established;

[0066] S2, Determination of crack initiation risk locations: Based on a multi-physics coupling simulation model, a component-to-whole machine multi-physics coupling simulation analysis is performed. A hybrid solution of the finite element method and the boundary element method is used to determine the stress concentration areas of key components as crack initiation risk locations;

[0067] S3, Crack Initiation and Growth Simulation: The stress concentration area of ​​the key components is subjected to thermal alternating load cycles. Based on the continuous damage mechanics method, a damage evolution model of the material in this area is constructed to calculate its fatigue cumulative damage. When the fatigue cumulative damage reaches a critical value, crack initiation is determined, and crack growth simulation is performed in the crack initiation area using the boundary element and finite element coupling method;

[0068] S4, Fatigue Strength Life Prediction Model Construction: Acquire a fracture mechanics parameter simulation data set and establish a proxy mathematical model of fracture mechanics based on a deep neural network algorithm. Combine the IoT monitoring data of the six-sided top press, the proxy mathematical model of fracture mechanics, and the crack propagation simulation results to construct a structural crack propagation model. Combined with a dynamic Bayesian network and particle filter algorithm, the crack propagation prediction model is improved to obtain a fatigue strength life prediction model.

[0069] S5, Structural fatigue crack test verification: Use wire cutting technology to prepare test specimens of key components, conduct tensile tests and fatigue tests respectively, obtain the constitutive relationship, SN curve and fatigue life of the materials of key components, and compare and verify with the simulation results of the multi-physics field coupling simulation model and the prediction results of the fatigue strength life prediction model to improve the multi-physics field coupling simulation model and the fatigue strength life prediction model.

[0070] S6, Fatigue Strength Reliability Model Construction: Based on the fatigue strength life prediction model, the fatigue strength reliability model is constructed by integrating the IoT monitoring data, historical operation data and multi-physics field coupling simulation model of the six-sided top press to determine the mathematical correlation between fatigue life, fatigue damage and reliability;

[0071] S7, reliability-driven multi-objective optimization design: Based on the mathematical correlation, the reliability calculation index is introduced into the multi-objective optimization of the key components of the six-sided top press. A high-precision numerical solution algorithm based on the Kriging proxy model is developed to obtain the optimal combination of design parameters, forming an equal strength design of the key components and the overall structure of the six-sided top press.

[0072] The embodiment of the present invention takes the diamond six-sided top press as an example, and conducts a specific analysis on key components such as the hinge beam structure, the top hammer structure and the synthetic block structure. The hinge beam of the diamond six-sided top press is a key pressure-bearing component, which is subjected to cyclic high oil pressure. Once the pressure exceeds the allowable stress, it will cause fatal damage to the structure. The top hammer will be subjected to harsh compression and shear stress in the high-temperature and high-pressure service environment of the synthetic cavity. If there are defects caused by the material and manufacturing process or uneven temperature, it will cause local stress concentration during operation, resulting in hammer cracking. Equal strength design and reliability robustness optimization design are carried out for the hinge beam structure of the six-sided top press to reveal the stress distribution law of key components of the structure, determine the deformation, temperature, stress distribution and stress concentration area of ​​the components, and predict the initiation of cracks in the key component structure; construct a fatigue crack propagation life assessment model for the hinge beam structure, and conduct reliability robustness optimization design to form equal strength of the components and the whole machine structure, and achieve lightweight structure.

[0073] Specifically, the high reliability and equal strength design method of the intelligent six-sided top press provided by the embodiment of the present invention is analyzed in each step as follows:

[0074] 1. Analysis of the structural characteristics of the six-sided diamond press

[0075] like Figure 2 As shown, the six-sided diamond press consists of a six-hinged beam system. The hydraulic control system, consisting of electronically controlled balancing and throttle valves, temperature sensors, and pressure sensors, collects the oil pressure and temperature of the hydraulic cylinder. The electronic control system also includes a programmable controller, an electric heater, a current transformer, and a voltage isolator. A displacement sensor is installed in the piston cylinder to measure piston stroke. The hydraulic system controls the pressurization, pressure maintenance, and pressure relief processes. The temperature within the synthesis chamber can be indirectly measured based on the relationship between the thermocouple electromotive force and temperature of a designated thermocouple, providing operational data for the six-sided diamond press. Diamond synthesis in the press undergoes multiple cycles of pressurization, pressure maintenance, and pressure relief. The hinged beam is a critical component that withstands ultra-high pressures. Under cyclical alternating loads, it is prone to cylinder cracking and lug breakage. Carbide anvils are subjected to long-term alternating high-temperature and ultra-high-pressure loads. From the perspective of structural characteristics and stress, cemented carbide is a brittle material. Under the action of multiple cyclic stresses, fatigue damage is prone to occur in stress concentration areas or defective areas, cracks are generated, and the top hammer is broken, which destroys the sealing environment in the high-pressure chamber and causes safety accidents.

[0076] 2. Multi-physics field analysis and modeling of diamond six-sided top press

[0077] The key components of the multi-physics field analysis of the six-sided top press are: the hinge beam structure, the top hammer structure, and the synthetic block structure. The hinge beam structure is a key pressure-bearing component that is subjected to alternating loads for a long time, and the failure forms are cylinder cracks and lug fractures. When the equipment is in operation, the pins of the hinge beam lugs are in contact and transmit the load between the hinge beams. During the diamond synthesis process, the oil cylinder that applies load to the inner wall of the hinge beam through oil pressure has a length equal to the piston stroke. From a force perspective, the most concentrated stress is in the lugs and oil holes of the hinge beam structure. When using finite element simulation, the computational mesh at the chamfers and circular holes will be distorted, resulting in a serious decrease in calculation accuracy or even distortion, requiring the use of high-precision numerical simulation technology.

[0078] 2.1 Analysis of hinge beam structure

[0079] The material of the hinge beam structure is 35CrMo alloy steel, with elastic modulus (E) and Poisson's ratio (v) E = 2.1×10 11 Pa, v = 0.3, density ρ = 7850 kg / m 3 , ultimate tensile strength σ b =1080MPa, yield strength σ s =9.3×10 8 Pa, the Lame constant is λ = Eν / (1+ν)(1-2ν). Based on the principle of virtual work, the static model of the hinge beam structure is as follows:

[0080]

[0081] The boundary conditions are

[0082] Where, represents the imaginary strain component, σ ij is the true stress component, b i is the component of the body force acting on the object, n j Represents the normal vector classification, is the component of the surface force acting on the surface of the object, represents the displacement component, is the virtual displacement component. During the pressure-building, pressure-maintaining, and pressure-releasing processes, the hinge beam structure experiences high-pressure oil pressure loads ranging from 0 to 100 MPa. The loads primarily impact the inner wall, chamfers, and bottom of the hinge beam. Constraints are applied at the hinge beam pins. Based on structural symmetry, a single hinge beam can be considered.

[0083] 2.2 Multi-physics field analysis of top hammer structure

[0084] The hammer structure consists of a carbide hammer, a steel ring, and a spacer. The steel ring forms an interference fit with the hammer head, applying radial preload to the hammer. Heating the upper and lower hammers generates high temperatures through resistance heating within the cavity. A hydraulic cylinder pushes the hammer, applying force to the synthesis block, creating an ultrahigh-pressure sealed cavity. The hammer is subject to multiple stresses, including compressive stress, shear stress, and thermal stress caused by high temperatures. Factors such as process defects during the synthesis process, offset and inaccurate alignment of the synthesis block, and temperature imbalances can also cause cracks and lead to fracture of the hammer. From a force perspective, during diamond synthesis, the temperature of the conductive hammer and steel ring increases dramatically, placing positive pressure on the front of the hammer. Under the rheological influence of phyllite in the ultrahigh-pressure working environment, the inclined surface is subjected to positive pressure and friction. This is a typical thermal-mechanical coupling problem under drastic temperature fluctuations.

[0085] The anvil structure has two material properties. The carbide anvil is YG8, a composite layered structure. It is subjected to both pressure and temperature loads, making it the most likely location for cracking or crushing. Accurate force analysis is required to identify the most likely locations for cracking or crushing, analyze the causes, and provide feedback to improve the anvil structure. The steel ring, small spacer, and large spacer are composed of low-alloy ultra-high-strength steel 45CrNiMoVA. The elastic modulus (E) and Poisson's ratio (v) of 45CrNiMoVA are E = 2.06 × 10 11 Pa, v = 0.3, density (ρ) is ρ = 7900 kg / m 3 , ultimate tensile strength σ b =1800MPa, yield strength σ s =1.65×10 9 Pa, shear strength σ c =1.8×10 9 Pa, thermal conductivity λ is 20W / m·k, thermal expansion coefficient α is 1.45×10 -5 m / k; the elastic modulus and Poisson's ratio of cemented carbide YG8 are E=6×10 11 Pa, v = 0.23, density ρ = 15000 kg / m 3 , ultimate tensile strength σb=4500MPa, yield strength σ s =5.5×10 9 Pa, shear strength σ c =2.3×10 9 Pa, thermal conductivity λ is 75.4W / m·k, and thermal expansion coefficient is 4.5×10 -6 m / k.

[0086] The boundary conditions of the temperature field of the top hammer structure are: the top hammer is subjected to temperature load, and a water cooling device is installed at the bottom of the steel ring, and the temperature is T f In addition, the water flow exchange coefficient (h) at the bottom of the steel ring is 1000W / m2 ·℃, the convection coefficient (h) of other parts directly in contact with air is 22W / m 2 ℃. Working oil pressure is applied to the bottom of the top hammer structure. In addition, the four small inclined surfaces of the top hammer are subjected to the positive pressure P2 of the synthesis chamber, and the material in the synthesis chamber exerts a friction force f1 on the four small inclined surfaces of the top hammer. Its thermodynamic model is:

[0087]

[0088] Where ρ is the density of the material, c is the specific heat capacity, t is the time, and λ x ,λ y ,λ z is the thermal conductivity coefficient of the material in the x, y, and z directions (here, λ x =λ y =λ z ), Φ0 is the density of the heat source, which needs to be calculated by the current power of the top hammer, and T represents the temperature.

[0089] The heat conduction boundary conditions considered are at the boundary The above is as follows:

[0090] (1) T(x, y, z, t) = T0(x, y, z, t), where T0 is the constant temperature of the top hammer surface;

[0091] (2) k is the thermal conductivity, is the derivative of temperature T along the normal direction outside the boundary, and q(x, y, z, t) is the known heat flux density function;

[0092] (3) Where h is the surface heat transfer coefficient, T f is the temperature of the surrounding fluid.

[0093]

[0094] Where, represents the true thermal strain, ε kl represents the true elastic strain, represents the imaginary strain component, σ ij is the true stress component, b i is the component of the body force acting on the object, n j Represents the normal vector classification, is the component of the surface force acting on the surface of the object, represents the displacement component, is the imaginary displacement component.

[0095] 2.3. Multi-physics analysis of composite blocks

[0096] The core of diamond synthesis is the matching of temperature, pressure and time in the synthesis chamber. Understanding the temperature and pressure distribution laws in the synthesis chamber can provide a theoretical basis for temperature and pressure control, process route formulation, cavity structure design, and assembly material selection, thereby achieving process optimization. Diamond synthesis blocks are grown under high-temperature and ultra-high-pressure sealing conditions. Due to the time-varying characteristics of material properties, the temperature and pressure distribution during diamond growth is relatively complex, making it difficult to accurately describe the pressure field, temperature field, and convection field of the synthesis block and the assembled material structure. The initial temperature of the synthesis block, the heating and temperature rise law of the press, the load it bears, the growth of diamond crystals, the carbon flow rate, etc. together constitute the thermal-mechanical-rheological coupling analysis problem of the synthesis block. The present invention intends to establish a time-effect function of the heating temperature, calculate the pressure load of the synthesis block based on the hydraulic pressure load, consider the rheological characteristics of the material, and use the finite element method to analyze the thermal-mechanical-rheological coupling problem of the synthesis block, analyze the advantages and disadvantages of the diamond synthesis block assembly method, and improve the stability of the diamond growth environment.

[0097] The block in the synthesis cavity is in a closed state subjected to high temperature and high pressure. Considering the properties of physical parameters such as synthesis rod, pyrophyllite, conductive steel bowl, carbon source sheet, catalyst sheet, the initial temperature of the synthesis block, the temperature load time function of the synthesis block, the pressure load of the synthesis block, the heating method and electric power, etc., the multi-physics field model in the synthesis cavity includes the synthesis block thermal-structural coupling model, the synthesis block temperature field electro-thermal coupling model, and the convection field electro-thermal-fluid coupling model.

[0098] To ensure the accuracy of subsequent simulation results, the six-sided ram press utilizes pressure sensors, current transformers, voltage isolators, displacement sensors, and thermocouples to obtain real-world operating data, including hydraulic system pressure, test heating primary current, transformer secondary current, heating ram voltage, piston stroke, composite block temperature, single-cylinder thrust, three-cylinder forward speed, return speed, overpressure speed, and pressure relief speed. Based on this real-world operating data, theoretical analysis and derivation are used to directly or indirectly obtain realistic boundary conditions, improving model accuracy. For missing data, such as the temperature within the composite block, simulation results, theoretical derivation, and a critical area load inversion method are constructed. Thermocouples installed within the composite block indirectly determine the temperature of calibrated points within the block. By comparing and correlating simulation results with the stress, strain, and temperature of easily calibrated points in key locations, load data for these critical areas can be inverted.

[0099] 3. Component-whole machine multi-physics field coupling simulation analysis method

[0100] The failure mode of the hinge beam single cylinder system is failure of the top hammer and the hinge beam. The hinge beam is a key pressure-bearing component, and cyclic high-pressure stress will cause the hinge beam structure lug to break or the cylinder to crack. The temperature of the top hammer hammer surface reaches 100°C, and it is subjected to complex and harsh compression and shearing in the synthetic cavity. If there are defects or uneven temperature, it will cause local stress concentration during operation, resulting in a cracked hammer or collapsed hammer. The reason for the failure of the top hammer or hinge beam is that the stress concentration in the local area exceeds the allowable stress of the material, causing crack initiation. The boundary element method with high displacement and stress calculation accuracy is selected to analyze the stress and strain of key components in the structure. It is worth noting that the regional mesh is reconstructed in the stress concentration area, which is the key area of ​​focus for crack initiation, to improve the calculation accuracy of its stress and strain.

[0101] Based on the multi-physics field analysis of the six-sided top press, the mechanical model and physical properties of the hinged beam structure were obtained. Since the temperature field has little effect on the hinged beam system, only static analysis was required. Alternating loads and mesh discretization were applied, and the boundary element method was configured for solution. For the top hammer system, it was first necessary to establish a mathematical model for the heating of the top hammer system and analyze the boundary conditions of the top hammer system (air convection heat transfer analysis and cooling water convection heat transfer analysis). The top hammer system is a typical multilayer dielectric material, and the carbide top hammer is the area most likely to cause stress concentration and crack initiation. Using the thermomechanical coupled boundary element method for analysis can effectively improve the analysis accuracy of its thermal temperature field. The steel ring and spacer can be analyzed using the finite element method. The final results are coupled. The high-pressure chamber provides an environment for diamond growth. Based on the geometric and physical models of the synthesis block, the carbon convection and diffusion fields are analyzed, and electro-thermal-mechanical-fluid coupled analysis and calculations are performed. The influence of carbon convection in the synthesis block is studied to reveal the formation mechanism of surface growth defects in diamond synthesized by the temperature gradient method. The detailed analysis is as follows:

[0102] 3.1 Statics Analysis of Hinge Beam

[0103] Based on formula (1), the static boundary integral equation of the hinge beam can be derived as follows:

[0104]

[0105] In formula (4), c lk (y) = 0.5, when l = k; c lk (y)=0, when l≠k. The range of l and k is 1, 2, 3; u k (x) represents the displacement component, t k (x) represents the surface force component.

[0106] and The basic solutions for displacement and traction are:

[0107]

[0108] r represents the distance between the source point y(y1,y2,y3) and the field point x(x1,x2,x3). When l=k; δ lk =1, when l≠k, δ lk =0. The value range of l and k is 1, 2, 3. ,l and r ,k Respectively represent r with respect to x l 、x k The derivative of n k and n l represents the components of the normal vector n, represents the directional derivative of r, G is the shear elastic modulus, and v is the Poisson's ratio.

[0109] The boundary conditions are as follows: u superior, u i (x) represents the displacement component at the field point, represents the displacement boundary imposed at the field point. The boundary condition is t Up, t i (x) represents the surface force component at the field point, Represents the traction boundary applied at the field point.

[0110] The stress boundary integral equation of the points in the domain can be expressed as:

[0111]

[0112] Where, σ ij (y) is the stress component at the point in the domain, represents the high-order fundamental solution of elasticity, where

[0113]

[0114] 3.2 Thermal-mechanical coupling simulation analysis of top hammer structure

[0115] Based on formula (2), the boundary integral equation of heat conduction of the top hammer structure can be derived as follows:

[0116] c(y)T(y)+ G q * (y, x)T(x)dS= G T * (y, x)q(x)dS (6)

[0117] Where c(y) = 0.5, T(y) represents the temperature at the source point y, T(x) represents the temperature at the field point x, q(x) represents the flow rate at the field point x, and q *(y,x) and T * (y,x) represents the basic solution of heat conduction, where k is the thermal conductivity coefficient.

[0118] The thermal-mechanical coupling boundary integral equation can be derived as:

[0119]

[0120] In formula (7), c lk (y) = 0.5, when l = k; c lk (y)=0, when l≠k. The value range of l and k is 1, 2, 3; u k (x) represents the displacement component, t k (x) represents the surface force component, T(x) represents the temperature, and a is the thermal expansion coefficient. and They represent the basic solutions for displacement and traction respectively:

[0121]

[0122]

[0123] r represents the distance between the source point y(y1,y2,y3) and the field point x(x1,x2,x3). When l=k; δ lk =1, when l≠k, δ lk =0. The value range of l and k is 1, 2, 3. ,l and r ,k Respectively represent r with respect to x l 、x k The derivative of n k and n l represents the components of the normal vector n, represents the directional derivative of r, and v represents Poisson's ratio.

[0124] The boundary conditions are as follows: u superior, u i (x) represents the displacement component at the field point, represents the displacement boundary imposed at the field point. The boundary condition is t Up, t i (x) represents the surface force component at the field point, represents the applied surface force boundary at the field point. G is the shear elastic modulus, and m is the Lame constant, m = E / 2 / (1+ν).

[0125] 3.3. Multi-physics coupling simulation analysis of synthetic blocks

[0126] The block in the synthesis cavity is subjected to high temperature and high pressure in a closed state. Based on the properties of the physical parameters of the synthesis rod, pyrophyllite, conductive steel bowl, carbon source sheet, catalyst sheet, etc., the initial temperature of the synthesis block, the time-dependent function of the synthesis block temperature load, the synthesis block pressure load, the heating method and electrical power, etc., the multi-physics field model in the synthesis cavity includes a synthesis block thermal-structural coupling model, a synthesis block temperature field electro-thermal coupling model, and a convection field electro-thermal-fluid coupling model. The time-dependent function of the temperature of the six-sided top press is T(t) = T max j(t)+T0, j(t) is a single-valued function about time t, T0 is the initial temperature. Assume that the direct heating time effect function of the six-sided top press is The relationship between steady-state heating temperature and heating power is established, and multi-physics field simulation is performed using finite element method to analyze the temperature and material rheological properties in the synthesis cavity.

[0127] It is worth noting that to ensure the accuracy of the calculation results, the six-sided top press IoT system utilizes pressure sensors, displacement sensors, and thermocouples to obtain real-world operating data such as the temperature of the composite block, single-cylinder thrust, three-cylinder forward speed, return speed, overpressure speed, and pressure relief speed. This data can be compared with simulation results to verify their accuracy. The temperature of calibration points in the composite block can be indirectly obtained using thermocouples, and the data obtained through simulation can be directly compared with the data at the measurement points. Simulation and measurement results are benchmarked to ensure data accuracy.

[0128] 4. Simulation method of crack initiation and propagation in key junctions of six-sided top press

[0129] The fatigue fracture process of key structures in a six-sided top press includes crack initiation and crack propagation. Fatigue crack initiation is affected by the observed length scale. From a macromechanical perspective, fatigue crack initiation is the stage in which the original microcrack expands to a macroscopically visible length, while fatigue crack propagation is the process in which the crack steadily propagates until the structure becomes unstable and damaged. Structural fatigue fracture prediction methods often only consider a subset of influencing factors or perform single-stage predictions, resulting in high computational complexity and lack of universal applicability. This invention, based on a developed Internet of Things (IoT) system, obtains key data such as the structure's material properties, historical operating data, and geometric parameters. Using multi-physics coupled simulation, regions of significant stress concentration in key components are identified and subjected to cyclical thermal alternating loads. Using a continuous damage mechanics approach, a material damage evolution model is constructed for this region, and the accumulated fatigue damage is calculated. When the accumulated fatigue damage reaches a critical value, microcracks initiate. The number of loading cycles at this stage is the crack initiation lifespan of the structure. For crack propagation, a coupled boundary element method (BEM) and finite element method (FEM) approach can be used. First, subregional partitioning is performed. The crack region is a region of concentrated stress singularities, and the BEM has significant advantages in solving such problems. Finite element analysis can be used for areas farther from the crack, effectively reducing computational costs and improving accuracy. The crack region is then discretized using a mesh, and boundary integral equations are configured, along with matrix assembly and solution. Finally, the stress intensity factor, crack growth angle, and crack growth direction are calculated to determine whether the crack has reached a critical length and thus the fatigue life.

[0130] like Figure 3 As shown in Figure 2, the process of using boundary elements to simulate crack growth is as follows:

[0131] By configuring the material parameters and applying the boundary conditions, the crack area is discretized. The boundary integral equation is configured according to the following rules. When it falls on the non-crack boundary S, the boundary integral equation is configured as formula (8):

[0132]

[0133] Where, represents the basic solution of the boundary element, as shown in formula (4). + Indicates that the field point falls on the upper boundary of the crack, x - Indicates that the field point falls on the lower boundary of the crack.

[0134] On the spot When it falls on the crack below S, the boundary integral equation is configured as formula (9):

[0135]

[0136] On the spot When it falls on the crack surface S, the boundary integral equation is configured as formula (10):

[0137]

[0138] Where, Indicates the top of the crack, Indicates the bottom of the crack. and u j (x + ) represents the displacement component above the crack. and u j (x - ) represents the displacement component below the crack. j (x + )and represents the surface force component on the crack, t j (x - )and represents the component of the traction force below the crack. After configuring the boundary integral equation, a singular element is added at the crack tip, and the finite partial integral and principal value integral definitions are used to handle supersingular and singular integrals. In the local natural coordinate system, the forms of singular and near-singular integrals are shown in formulas (11)-(13).

[0139]

[0140] The integrands of formulas (11)-(13) are expanded in polar coordinates near the source point to obtain the following form:

[0141]

[0142] In formulas (11)-(16), A represents the expansion of r in the local polar coordinate system, A i 、A j and A k Represents r ,i 、r ,j and r ,k Expansion in local polar coordinates. S -3 (θ) represents The coefficient, S -2 (θ) represents The coefficient of G ijk , F ijk Represents the basic solution expansion about the coefficients of the polynomial, N m represents the shape function. For convenience, N0 and N1 are the coefficients of the Taylor expansion in local polar coordinates with respect to ρ. After the integrand is expanded in polar coordinates, it can be processed using the finite partial integral and principal value integral definitions.

[0143] The stress intensity factor is obtained using the M integral (Formula (17))

[0144]

[0145] because:

[0146]

[0147] It can be obtained, assuming get assumed get assumed get

[0148] In formulas (17) and (18), n j represents the local normal vector, represent the stress and strain in the real state, represents the stress and strain under the auxiliary field. represents the derivative of the displacement component with respect to the local coordinate x in the real state, represents the derivative of the displacement component under the auxiliary field with respect to the local coordinate x. The asymptotic displacement field and stress field of the auxiliary field can be obtained from the stress intensity factor manual.

[0149] After the stress intensity factor is obtained, the crack growth criterion is applied. K I , K II , K III , respectively represent the stress intensity factors of type I, II, and III, c represents the material parameters, ΔK eq represents the equivalent stress intensity factor, and m represents the degree of the Paris formula.

[0150] The crack extension angle is sign represents the sign function.

[0151] Each step can extend the lifespan by Obtain, and obtain the overall fatigue life by superposition.

[0152] 5. Deep Learning-Driven Fatigue Life Prediction Method for Six-Sided Press

[0153] The accuracy of the fatigue strength life prediction model of the six-sided top press is easily affected by fatigue parameters. The acquisition of fatigue parameters is affected by uncertain factors such as test samples, working conditions, material properties, and has a large dispersion. Based on the developed Internet of Things system, key data such as material properties, historical operating data, and geometric parameters of the equipment structure are obtained. The present invention introduces deep learning technology to solve the problem of fatigue fracture prediction. A priori distribution model based on the initial crack size and crack propagation parameters of the six-sided top press structure is established to overcome the problem of large dispersion of fatigue parameters. The finite element and boundary element coupling method is used to obtain fracture mechanics performance parameters such as stress intensity factor and T stress, and a fracture mechanics parameter simulation data set is generated. Based on deep neural network technology, a proxy mathematical model of fracture mechanics of crack propagation size, crack propagation parameters, crack propagation life and temperature, displacement, stress and strain of measurement points on key components is established. Combined with crack propagation models (such as Pairs formula and Forman formula), rapid prediction of crack propagation can be achieved. Using the Internet of Things monitoring data of equipment structure, the proxy mathematical model of fracture mechanics and the crack propagation model, a structural crack propagation model (such as Figure 4 ).

[0154] The sources of uncertainty were analyzed. For example, temperature and pressure fluctuations can cause load uncertainty (variation in load amplitude). The time-varying characteristics of the high-pressure chamber's internal conditions lead to uncertainty in the material's physical properties, which in turn affect the crack growth rate and introduce uncertainty into the growth model. Crack size, location, and shape also exhibit randomness. The large number of parameters involved in considering these uncertainties makes the model overly complex and difficult to solve. The causal relationships between these uncertainties were analyzed, and sensitivity analysis was performed on the model parameters influencing them to identify the primary factors influencing the model. By integrating IoT monitoring data, historical operational data, and a multi-physics coupled physical model, a dynamic Bayesian network prediction model for crack size was constructed, considering three different types of dynamic Bayesian model parameters: deterministic parameters, model-specific uncertain parameters (load fluctuations, time-varying material properties), and monitoring data (crack distribution, size, and dimensions within the equipment). After the Bayesian prediction model was completed, the original model's predictions gradually deviated from the actual values ​​due to the presence of multiple uncertain variables. A particle filter inference algorithm was used to update the random variable nodes in the model, probabilistically predicting the crack growth size and obtaining online data such as real-time operational data and dynamic parameters of crack growth. By inputting into the dynamic Bayesian model and comparing it with the IoT monitoring results and historical operation data, the initial digital twin model is diagnosed and updated to achieve real-time, efficient, and intelligent analysis and prediction of fatigue crack locations and expansion trends, promoting the development of fatigue life prediction technology for six-sided top press structures and improving the safety of diamond six-sided top press equipment structures.

[0155] 6. Structural fatigue crack test verification plan

[0156] Fatigue crack test verification scheme for typical key structural components Figure 5 and Figure 6 As shown, it includes the design of tensile test process and fatigue test process. It is worth noting that the test specimen manufacturing methods for hinged beam structure and top hammer structure are different. The hinged beam structure is sampled by cutting from the mother body, and the test specimen is manufactured according to national standards. The cemented carbide top hammer is different. The cemented carbide top hammer is brittle and cannot be cut directly. The test specimen can be manufactured using the same vacuum sintering production process as the cemented carbide top hammer, the same production process, and manufactured in accordance with national standards. After the test specimen is successfully manufactured, the constitutive relationship of the material can be obtained through tensile testing. The SN curve and fatigue life of the material can be obtained through fatigue testing. Compare and verify with the simulation results to improve the simulation method.

[0157] 7. Construction of a six-sided press structure fatigue strength reliability model and high-precision solution method

[0158] like Figure 7 As shown in the figure, in the fatigue strength life prediction model of the six-sided top press structure, IoT monitoring data, historical operation data and multi-physics field coupling digital model are integrated. Based on the fatigue life prediction method of damage mechanics and fracture mechanics, Bayesian theory is used to quantify the key parameters of the equipment, such as structural parameters, material parameters and stress load amplitude, to overcome the random characteristics of uncertainty that fluctuates with time and space. The Bayesian theory is expressed as follows:

[0159]

[0160] Where p(w|D) is the posterior probability, p(D|w) is the likelihood function, p(w) is the prior probability, and p(D) is the evidence factor. D is the observed data, and w is the uncertainty parameter to be quantified.

[0161] Determine the statistical characteristics of random parameters, construct a mathematical relationship model between key parameters and fatigue life, obtain the probability distribution of parameters and the changing trend of reliability, unify them into the distribution of fatigue life, and use subsequent functional functions to calculate reliability.

[0162] If the performance function Z=RS conforms to the normal distribution, μ R is the expectation of R, is the variance of R, μ S is S's expectation, is the variance of S. The reliability is calculated as:

[0163]

[0164] If R and S conform to the log-normal distribution, the reliability is calculated as:

[0165]

[0166] In the formula, the load action on the hinge beam and top hammer structure is S and the structural resistance is R.

[0167]

[0168] The relationship between stress amplitude r and fatigue life N is:

[0169] r m N=C (21)

[0170] The total fatigue damage D is the damage d at each load level i The accumulation of d i It represents the amount of damage caused by cyclic loading at a certain stress level.

[0171] d1+d2+...+d k =Σd k =D (22)

[0172] n i is the actual number of cycles under the i-th stress level, N i Indicates the fatigue life of the material at this stress level (number of cycles to failure);

[0173]

[0174] When the cumulative sum of the damage from each load level equals 1, fatigue failure occurs, meaning the total damage reaches a critical value. A functional function for the fatigue residual life model is constructed based on the equipment's expected lifespan, N0. Using neural network deep learning technology, sample points from the fatigue fracture process of the six-sided top press are extracted as sample data to reconstruct the functional function for the fatigue residual life model of the six-sided top press structure and obtain the optimal proxy functional function. The fitted proxy functional function is then used to calculate fatigue reliability using the first-order second-moment method.

[0175] The calculation of failure probability is as follows: f =Φ(-β s ), β s is the reliability index, Φ(-β s ) is the cumulative distribution function of the standard normal distribution. The linear approximation of the performance function is the reference design point mean, is the variable x i The mean of the function The mean of the performance function Z. The variance of the performance function is is the variance of Z.

[0176] The present invention establishes the corresponding limit state equation by analyzing the limit state of the hinge beam structure and the limit state of the top hammer system, taking into account the load effect and structural resistance of the hinge beam and the top hammer structure. In combination with the expected life of the equipment, based on the stress intensity interference theory, the functional function of its fatigue remaining life model is constructed. Due to the coupling effect of multiple physical fields and the influence of various uncertain factors, the fatigue failure mechanism of the equipment is relatively complex, and the functional function of the fatigue remaining life model presents a nonlinear implicit form, which is difficult to solve directly and will bring about large calculation errors. Based on the neural network deep learning technology, the present invention extracts sample points in the structural crack propagation simulation as sample data, reconstructs the functional function of the fatigue remaining life model of the structure, and obtains the best proxy functional function. For the fitted proxy functional function, the first-order second moment method is used to calculate the fatigue reliability.

[0177] 8. Structural reliability optimization design method of six-sided top press

[0178] Under the influence of uncertain factors, the fatigue life of the six-sided top press has a large dispersion, which causes the fatigue life of the key component structure to often deviate far from its design life. In response to this problem, the target reliability is introduced as a constraint or optimization target into the multi-objective optimization of the structure, and the key structural and process parameters are determined as design variables (such as the thickness of the carbide layer, temperature, pressure index, etc.), the range of the design variables, and a multidisciplinary optimization mathematical model is constructed. In the embodiment of the present invention, the data relationship model between the structural performance of the six-sided top press and the design variables obtained by multi-physical field coupled fatigue fracture simulation is sampled, a Kriging proxy model is constructed and an optimization solution is performed. Specifically, the Kriging model of the random sampling points of the design variable probability distribution production estimates the expected variance of the predicted value, samples are drawn according to the expectation and variance of the sampling points, the mean and standard deviation of the sample response are calculated, and the optimization calculation is realized. A test system is constructed to monitor the deformation of the press under real-time working conditions, and to compare and correct it with the multi-physical field simulation model to optimize and determine the key parameters of the press structure and production process.

[0179] The application further provides a high-reliability equivalent strength design system of an intelligent cubic press, configured to execute the high-reliability equivalent strength design method of the intelligent cubic press, and comprising: a structure characteristic analysis and multi-physical field modeling module, configured to establish a multi-physical field coupling simulation model comprising a deformation field, a temperature field, a stress field and a rheological field based on the structure characteristics of key components of the cubic press; a crack initiation risk position determination module, configured to perform multi-physical field coupling simulation analysis of the components and the whole machine based on the multi-physical field coupling simulation model, determine the stress concentration area of the key components as the crack initiation risk position by mixed solving of the finite element method and the boundary element method; a crack initiation and propagation simulation module, configured to perform thermal and mechanical alternating load cyclic loading on the stress concentration area of the key components, construct a material damage evolution model of the area based on the continuous damage mechanics method, calculate the fatigue cumulative damage, determine the crack initiation when the fatigue cumulative fatigue damage reaches a critical value, and simulate the crack propagation in the crack initiation area by the method of coupling of the boundary element and the finite element; a fatigue strength life prediction model construction module, configured to obtain a fracture mechanics parameter simulation data set, and establish a proxy mathematical model of fracture mechanics based on a deep neural network algorithm; construct a structure crack propagation model by integrating the Internet of Things monitoring data of the cubic press, the proxy mathematical model of fracture mechanics and the crack propagation simulation result, and improve the crack propagation prediction model by combining the dynamic Bayesian network and the particle filtering algorithm to obtain the fatigue strength life prediction model; a fatigue strength reliability model construction module, configured to construct a fatigue strength reliability model by integrating the Internet of Things monitoring data of the cubic press, the historical operation data and the multi-physical field coupling simulation model based on the fatigue strength life prediction model, and determine the mathematical correlation relationship among the fatigue life, the fatigue damage and the reliability; and a reliability-driven multi-objective optimization design module, configured to introduce the reliability calculation index into the multi-objective optimization of the key components of the cubic press, develop a high-precision numerical solution algorithm based on the Kriging proxy model to obtain the optimal combination of the design parameters, and form the equivalent strength design of the key components and the whole machine structure of the cubic press.

[0180] The principles and implementation manners of the application are described in specific examples in the present application, and the above examples are only used to help understand the method and core idea of the application; meanwhile, for those skilled in the art, the specific implementation manners and application ranges will be changed according to the idea of the application. In conclusion, the content of the present specification should not be understood as a limitation of the application.

Claims

1. A high reliability and equal strength design method for an intelligent six-sided top press, characterized in that: The following steps are involved: (1) Structural characteristics analysis and multi-physics field modeling: Based on the structural characteristics of the key components of the six-sided top press, a multi-physics field coupling simulation model including deformation field, temperature field, stress field and rheological field is established; (2) Determination of crack initiation risk locations: Based on the multi-physics field coupling simulation model, a component-whole machine multi-physics field coupling simulation analysis is performed. The stress concentration area of ​​the key components is determined as the crack initiation risk location through a hybrid solution of the finite element method and the boundary element method. (3) Crack initiation and propagation simulation: The stress concentration area of ​​the key components is subjected to thermal alternating load cycles. Based on the continuous damage mechanics method, a material damage evolution model of the area is constructed to calculate its fatigue cumulative damage. When the fatigue cumulative damage reaches the critical value, the crack initiation is determined, and the crack propagation simulation of the crack initiation area is performed using the boundary element and finite element coupling method; (4) Construction of fatigue strength life prediction model: obtain the fracture mechanics parameter simulation data set, and establish a proxy mathematical model of fracture mechanics based on the deep neural network algorithm; integrate the IoT monitoring data of the six-sided top press, the proxy mathematical model of fracture mechanics and the crack propagation simulation results to construct a structural crack propagation model, and combine the dynamic Bayesian network and particle filter algorithm to improve the crack propagation prediction model and obtain the fatigue strength life prediction model; (5) Fatigue strength reliability model construction: Based on the fatigue strength life prediction model, the IoT monitoring data, historical operation data and multi-physics field coupling simulation model of the six-sided top press are integrated to construct a fatigue strength reliability model and determine the mathematical correlation between fatigue life, fatigue damage and reliability; (6) Reliability-driven multi-objective optimization design: Based on the mathematical correlation, the reliability calculation index is introduced into the multi-objective optimization of the key components of the six-sided top press, and a high-precision numerical solution algorithm based on the Kriging proxy model is developed to obtain the optimal combination of design parameters, forming an equal strength design of the key components of the six-sided top press and the entire machine structure.

2. The high reliability and equal strength design method of the intelligent six-sided top press according to claim 1 is characterized in that: The design method further includes: Structural fatigue crack test verification: Use wire cutting technology to prepare test specimens of key components, conduct tensile tests and fatigue tests respectively, obtain the constitutive relationship, SN curve and fatigue life of the materials of key components, and compare and verify with the simulation results of the multi-physics field coupling simulation model and the prediction results of the fatigue strength and life prediction model to improve the multi-physics field coupling simulation model and the fatigue strength and life prediction model.

3. The high reliability and equal strength design method of the intelligent six-sided top press according to claim 1 is characterized in that: The key components include a hinge beam structure, a top hammer structure and a composite block structure.

4. The high reliability and equal strength design method of the intelligent six-sided top press according to claim 3 is characterized in that: The step (1), structural characteristics analysis and multi-physics field modeling, specifically includes: Based on the structural characteristics of the hinge beam structure, top hammer structure and composite block structure of the six-sided top press, a multi-physics field coupling simulation model including deformation field, temperature field, stress field and rheological field is established; Specifically, for the hinged beam structure, a three-dimensional geometric model of the hinged beam structure is constructed, including the lug, oil hole, and pin matching area. Local mesh adaptive encryption technology is used for the lug chamfer and oil hole fillet to avoid computational mesh distortion. The material constitutive model of the hinged beam is defined, and boundary conditions are set, including contact nonlinear analysis of the contact surface between the pin and the lug, the dynamic loading path of the hydraulic cylinder oil pressure load, and structural symmetry constraints. A mechanical model of the hinged beam structure is established, and a coupled simulation model of the deformation field, stress field, and temperature field is obtained. A model of the anvil structure was constructed. The anvil structure includes a carbide anvil, a steel ring, and a spacer. The steel ring and the carbide anvil exert radial preload through an interference fit. A thermal-mechanical coupling simulation model of the anvil was established to simulate the stress state of the anvil in the ultra-high-pressure sealing cavity, including compressive stress, shear stress, thermal stress, and additional stress caused by the offset of the composite block. A geometric model of the composite block assembly was established for the composite block structure, including pyrophyllite, conductive components, and diamond growth chamber. A time-dependent function of the heating temperature was defined, and the pressure load on the composite block was calculated based on the data from the hydraulic system pressure sensor. Combined with the material rheological constitutive model, the finite element method was used to solve the coupled temperature, pressure, and rheological fields of the composite block. The deformation field, stress field and temperature field coupling simulation model of the integrated hinge beam, the thermal-mechanical coupling model of the top hammer structure, the temperature field, pressure field and rheological field coupling model of the synthetic block are obtained to obtain the multi-physics field coupling simulation model of the six-sided top press.

5. The high reliability and equal strength design method of the intelligent six-sided top press according to claim 4 is characterized in that: The step (2) is to perform component-whole machine multi-physics coupling simulation analysis based on the multi-physics coupling simulation model, specifically including: Obtain the mechanical model and physical properties of the hinged beam structure, perform static analysis of the hinged beam, apply alternating loads and mesh discretization, configure the boundary element method, and solve; A heating mathematical model of the anvil structure was established, and the boundary conditions of the anvil structure were analyzed. Based on the characteristics of multi-layer dielectric materials, it was determined that the carbide anvil was the area where stress concentration would cause cracks. The thermal-mechanical coupled boundary element method was used to analyze the thermal temperature field of the carbide anvil. The finite element method was used to analyze the steel ring and spacer, and the final results were coupled. Based on the establishment of the geometric model and physical model of the synthetic block, the carbon convection diffusion field is analyzed, and the electric-thermal-mechanical-fluid coupling analysis and calculation are carried out to study the influence of carbon convection in the synthetic block and reveal the formation mechanism of growth defects on the surface of diamond synthesized by the temperature gradient method. In the IoT system of the six-sided top press, real operating data collected by various sensors is compared with the simulation analysis results to verify the accuracy of the simulation results.

6. The high reliability and equal strength design method of the intelligent six-sided top press according to claim 1 is characterized in that: In the step (3), the crack propagation simulation is performed on the crack initiation area using the boundary element and finite element coupling method, which specifically includes: First, the sub-regions are divided. The crack region belongs to the stress singularity concentration area and is solved by the boundary element method. The area far from the crack is solved by the finite element method. Then, the crack region is meshed and discretized, and boundary integral equations are configured, matrices assembled and solved; Finally, the stress intensity factor, crack growth angle and crack growth direction are calculated to determine whether the crack has grown to the critical length and the fatigue life.

7. The high reliability and equal strength design method of the intelligent six-sided top press according to claim 1 is characterized in that: In the step (4), a fracture mechanics parameter simulation data set is obtained, and a proxy mathematical model of fracture mechanics is established based on a deep neural network algorithm, specifically including: A priori distribution model of the initial crack size and crack propagation parameters of the six-sided top press structure was established. Using the finite element and boundary element coupling method, fracture mechanics performance parameters including stress intensity factor and T stress were obtained, and a fracture mechanics parameter simulation data set was generated. Based on the deep neural network algorithm, a proxy mathematical model of fracture mechanics is established, which includes crack propagation size, crack propagation parameters, crack propagation life and temperature, displacement, stress and strain of measurement points on key components.

8. The high reliability and equal strength design method of the intelligent six-sided top press according to claim 7 is characterized in that: In the step (4), the crack growth prediction model is improved by combining the dynamic Bayesian network and the particle filter algorithm to obtain the fatigue strength life prediction model, which specifically includes: By integrating IoT monitoring data, historical operation data, and multi-physics field coupling simulation models, a dynamic Bayesian network prediction model for crack size is constructed by considering three different types of dynamic Bayesian model parameters: deterministic parameters, model-specific uncertain parameters, and monitoring data. After completing the Bayesian network prediction model, due to the existence of multiple uncertain variables, the model prediction will gradually deviate from the actual value. The particle filter inference algorithm is used to update the random variable nodes in the Bayesian network prediction model, probabilistically predict the crack expansion size, and obtain online crack expansion data. By inputting it into the Bayesian network prediction model and comparing it with the IoT monitoring data and historical operation data, the initial digital twin model is diagnosed and updated to achieve real-time, efficient, and intelligent analysis and prediction of fatigue crack location and expansion trend, and obtain a fatigue strength life prediction model.

9. The high reliability and equal strength design method of the intelligent six-sided top press according to claim 8 is characterized in that: The step (5), constructing a fatigue strength reliability model, specifically includes: In the fatigue strength life prediction model, the company integrates IoT monitoring data, historical operation data, and multi-physics field coupling simulation models. Based on the fatigue life prediction method of damage mechanics and fracture mechanics, it uses Bayesian theory to quantify key parameters of equipment, such as structural parameters, material parameters, and stress load amplitude parameters, to overcome the random characteristics of uncertainty that fluctuates over time and space. Determine the statistical characteristics of random parameters, build a mathematical relationship model between key parameters and fatigue life, obtain the probability distribution of parameters and the changing trend of reliability, unify them into the distribution of fatigue life, and build the corresponding relationship between fatigue life and fatigue damage; Analyze the limit state of the hinge beam structure and the limit state of the top hammer system, consider the load effect and structural resistance of the hinge beam and top hammer structure, and establish the corresponding limit state equations; Based on the neural network deep learning technology, sample points in the structural crack propagation simulation are extracted as sample data, the functional function of the fatigue residual life model of the structure is reconstructed, and the optimal proxy functional function is obtained. The fatigue reliability calculation is performed using the first-order second-moment method based on the fitted proxy functional function.

10. An intelligent six-sided top press high reliability and equal strength design system, characterized by: The high reliability and equal strength design method for an intelligent six-sided top press according to any one of claims 1 to 9 comprises: Structural characteristics analysis and multi-physics modeling module, used to establish a multi-physics coupling simulation model including deformation field, temperature field, stress field and rheological field based on the structural characteristics of the key components of the six-sided top press; The module for determining the location of crack initiation risks is used to conduct component-to-whole-machine multi-physics coupling simulation analysis based on a multi-physics coupling simulation model. By using a hybrid solution of the finite element method and the boundary element method, the stress concentration area of ​​the key components is determined as the location of crack initiation risks. The crack initiation and growth simulation module is used to apply thermal alternating load cycles to stress concentration areas of key components. Based on the continuous damage mechanics method, a damage evolution model of the material in this area is constructed to calculate its fatigue cumulative damage. When the fatigue cumulative damage reaches a critical value, crack initiation is determined and crack growth simulation is performed in the crack initiation area using the boundary element and finite element coupling method; The fatigue strength and life prediction model construction module is used to obtain a simulation data set of fracture mechanics parameters and establish a proxy mathematical model of fracture mechanics based on a deep neural network algorithm. The module also integrates IoT monitoring data from the six-sided top press, a proxy mathematical model of fracture mechanics, and crack propagation simulation results to construct a structural crack propagation model. This model is then improved by combining a dynamic Bayesian network with a particle filter algorithm to produce a fatigue strength and life prediction model. The fatigue strength reliability model construction module is used to build a fatigue strength reliability model based on the fatigue strength life prediction model, integrating the six-sided top press's IoT monitoring data, historical operation data, and multi-physics field coupling simulation model to determine the mathematical correlation between fatigue life, fatigue damage, and reliability; A reliability-driven multi-objective optimization design module is used to introduce reliability calculation indicators into the multi-objective optimization of key components of the six-sided top press based on the mathematical correlation relationship, develop a high-precision numerical solution algorithm based on the Kriging proxy model, obtain the optimal combination of design parameters, and form an equal strength design of the key components of the six-sided top press and the entire machine structure.

Images