Thermal-mechanical coupling optimization method for shank structure of hydraulic robot

Abstract

The invention provides a thermal coupling optimization method for a shank structure of a hydraulic robot, and belongs to the technical field of hydraulic robots and topological optimizing.The thermal coupling optimization method comprises the steps that a node-based thermal coupling topological optimization algorithm is adopted, a complex structure is discretized through an unstructured tetrahedral mesh, and the discretized complex structure is obtained; and a Helmholtz filtering technology and a Heaviside projection technology are introduced, so that the fairness and manufacturability of an optimization result are ensured. And with the purpose of minimizing the flexibility of the structure, intelligent material distribution of the robot shank structure is achieved under the constraint of the material volume. And the hydraulic humanoid robot with the hydraulic flow channel integrated in the structure is constructed according to the optimization result. The quality of the shank structure of the hydraulic humanoid robot is reduced, and the heat dissipation efficiency is remarkably improved. The traditional defect that light weight and heat dissipation design are carried out independently is effectively overcome, performance collaborative optimization is achieved, and the dynamic response and continuous working capacity of the robot is improved.

Description

Technical Field

[0001] This invention belongs to the field of hydraulic robot and topology optimization technology, specifically relating to a method for optimizing the thermal coupling of the lower leg structure of a hydraulic robot. Background Technology

[0002] Humanoid robots are rapidly developing in the fields of artificial intelligence and robotics. Among them, hydraulic humanoid robots are robots driven by hydraulic actuators that move in a manner similar to humans. They possess advantages such as high output, strong bursts of power, and high load capacity, and have unique application potential in various fields such as heavy industrial production, military, special applications, and disaster relief. One of the core requirements in the structural design and optimization of hydraulic humanoid robots is to reduce the weight of the robot structure, especially the end limbs such as the lower legs, to improve the robot's dynamic response performance, reduce energy consumption, and increase endurance. Simultaneously, because the hydraulic pump and valve systems used in hydraulic humanoid robots are prone to heat generation, resulting in high hydraulic oil temperatures during operation, heat dissipation design is necessary to ensure stable operation over extended periods.

[0003] Currently, the structural design and optimization of robots mainly focus on lightweighting, with the primary goal of reducing robot weight. Research on heat dissipation optimization is limited, primarily concentrating on the heat dissipation of hydraulic systems such as hydraulic pumps and tanks, while less consideration is given to the passive heat dissipation potential arising from the robot's large surface area. A key reason for this is that most hydraulic robots use hydraulic hoses to transfer hydraulic oil rather than introducing it into the robot's structure, making it difficult to directly utilize the robot's heat dissipation capacity. By implementing hose-free design and incorporating hydraulic channels within the robot structure, the robot's simplicity and reliability can be effectively improved, and the robot's structure can be endowed with heat dissipation capabilities, thereby effectively improving the structure's heat dissipation efficiency through heat dissipation design. Meanwhile, in current robot design optimization, lightweight design and heat dissipation optimization are generally isolated, without simultaneous optimization of heat and force. Furthermore, due to the thermal expansion and contraction properties of materials, after the structure is heated by hydraulic oil, it will undergo slight deformation due to thermal stress, which will affect the displacement field formed by the robot structure under external load. Therefore, there is a weak coupling relationship between the thermal field and the force field of the robot structure. The thermo-mechanical coupling topology optimization method can effectively solve the thermo-mechanical coupling problem of robot structure and achieve simultaneous optimization of lightweighting and heat dissipation.

[0004] As a crucial end-effector of the lower limbs of a hydraulic humanoid robot, the robot's primary movement mode is rotation around the knee joint. Therefore, its mass is extremely sensitive, especially the rotational inertia around the knee joint, which directly affects the robot's dynamic motion performance. Simultaneously, the lower leg requires the installation of hydraulic cylinder actuators for the ankle joint to drive the robot's foot movement, necessitating the efficient flow of hydraulic fluid within the leg. To improve the dynamic motion performance and integration of the hydraulic humanoid robot, a thermo-coupling optimization design for the robot's lower leg is urgently needed.

[0005] Therefore, the main drawback of the existing technology is that the hydraulic robot body structure does not have the ability to circulate oil and dissipate heat, and that lightweighting and heat dissipation optimization are isolated in the process of structural optimization design. Summary of the Invention

[0006] To address the aforementioned deficiencies and improvement needs of existing technologies, this invention provides a method for optimizing the thermal coupling of the lower leg structure of a hydraulic robot.

[0007] To achieve the above objectives, the present invention provides the following technical solution: A method for optimizing the thermo-mechanical coupling of the lower leg structure of a hydraulic robot includes the following steps: An initial model of the robot's lower leg is established based on the size requirements, driving method, and external load. The initial model is then meshed using an unstructured four-node tetrahedral element mesh to obtain a mesh file containing model information. A parameter file including optimization parameters is defined. Using node density as a design variable, the design variable is initialized based on the mesh file and parameter file, and the continuous design variable is converted into element density through weighted calculation. Penalty functions for temperature field, stress field and thermal stress field are constructed. The unit density is used as the input of the penalty function to obtain the temperature field of the lower leg structure. The thermal stress field is calculated based on the temperature field of the lower leg structure. The total displacement field is obtained based on the combined action of the thermal stress field and the stress field formed by the external load force. The objective function for thermo-mechanical coupling optimization is constructed by minimizing structural flexibility. The total displacement field is input into the objective function, and the sensitivity of the objective function to the design variables is calculated. Based on the sensitivity, the design variables are iteratively adjusted to seek optimization. Iteration conditions are set, and after the conditions are met, the optimization results of the robot's lower leg topology model in STL format are output.

[0008] Preferably, the design variables are initialized based on the mesh file and parameter file, specifically by filtering the design variables in the mesh file and parameter file using a Helmholtz filter; the filtered design variables are then subjected to Heaviside projection, and the projected node design variables are converted into element densities.

[0009] Preferably, the penalty functions for the temperature field, stress field, and thermal stress field are specifically: ; ; ; in, The thermal conductivity of the unit cell. The Young's modulus of the unit cell. The thermal stress coefficient of the unit. For unit density, The coefficient of thermal expansion is These represent the thermal conductivity when the unit cell is not empty and the thermal conductivity when the unit cell is empty, respectively. These are the Young's moduli of the element under different states. These are the penalty coefficients for the corresponding coefficients.

[0010] Preferably, the objective function is: ; in, For design variables, Let be the objective function. For force load vector, For the total displacement field, This is the global thermal stiffness matrix. For temperature field, For thermal load vector, The global stiffness matrix, These are the mechanical force vector and the thermal stress vector, respectively. To optimize the size of the results, The volume of the structure at the target volume fraction. This is the unit density vector.

[0011] Preferably, the sensitivity is calculated using the following formula: ; in, This is the adjoint vector.

[0012] Preferably, the iteration condition specifically involves calculating the difference between the current objective function value and the objective function value of the previous round. If the difference is less than a set threshold, the iteration stops.

[0013] Preferably, when optimizing the design variables based on the sensitivity iterative adjustment, a pair of dynamically adjusted lower and upper bound asymptotes are set for each design variable using the moving asymptote method, and the asymptote positions are updated to tighten the search range, thereby updating the design variables.

[0014] Preferably, when performing finite element mesh generation, a physical domain is defined to divide the initial model into regions, distinguishing between optimizable regions, non-optimizable regions, and external domains; the optimizable regions are regions where material distribution can be adjusted through topology optimization; the non-optimizable regions are key functional structures of the lower leg that need to be fully preserved; and the external domains are virtual boundary regions that enclose the model.

[0015] The thermal coupling optimization method for the lower leg structure of a hydraulic robot provided by this invention has the following beneficial effects: This invention utilizes unstructured four-node tetrahedral element meshes to discretize complex-shaped robot structures in practical engineering applications. Node density is used as a design variable to represent the structural topology. Weighted calculations of node density convert continuous design variables into element densities, avoiding surface unevenness issues that may result from unstructured four-node tetrahedral elements. A thermo-coupling optimization objective function is set, and iterative topology optimization is performed on the design variables to achieve lightweighting and heat dissipation optimization, effectively improving robot structural performance. Structural simulation of the lower leg structure of a hydraulic humanoid robot based on the optimized data yields a lower leg structure with lighter mass and better heat dissipation, improving the robot's dynamic motion performance and ensuring stable operation. Attached Figure Description

[0016] To more clearly illustrate the embodiments and design schemes of the present invention, the accompanying drawings required for this embodiment will be briefly described below. The drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating a method for optimizing the thermo-mechanical coupling of the lower leg structure of a hydraulic robot, as provided by this invention.

[0018] Figure 2 This invention provides a design framework for optimizing the thermal coupling of robot structures.

[0019] Figure 3 This is the initial model of the lower leg structure of the hydraulic humanoid robot used for topology optimization in this embodiment of the invention.

[0020] Figure 4 This is the result of dividing the initial model into tetrahedral four-node finite element meshes in this embodiment of the invention.

[0021] Figure 5 This is the result of optimizing the lower leg of the hydraulic humanoid robot using the provided thermo-coupling topology optimization method in this embodiment of the invention. Figure 5 (a) is the oblique projection of the optimization result; Figure 5 (b) is the side view of the optimization results.

[0022] Figure 6 This is a model diagram of the topology optimization results after surface smoothing post-processing in an embodiment of the present invention. Figure 6 (a) is the oblique projection of the smoothed post-processing result; Figure 6 (b) is a side view of the smoothed post-processing result.

[0023] Figure 7 This is the final post-processing model obtained by further optimizing the smoothing post-processing results in this embodiment of the invention, taking into account practical applications.

[0024] Figure 8 This is a schematic diagram of the oil circuit structure inside the post-processing model in an embodiment of the present invention. Figure 8 (a) is a side sectional view; Figure 8 (b) is a front sectional view.

[0025] Figure 9 This is a comparison diagram of the initial model, the topology optimization model, and the final post-processing model in an embodiment of the present invention. Figure 9 (a) is the initial model; Figure 9 (b) is the topology optimization model; Figure 9 (c) is the final post-processed model, where the pink axis point in the figure represents the centroid position of the model.

[0026] Figure 10 This is a heat dissipation temperature distribution diagram of the initial model, topology optimization model, and final post-processing model in this embodiment of the invention, using internal hydraulic oil as the heat source. Figure 10 (a) is the heat dissipation temperature distribution diagram of the initial model; Figure 10 (b) is the heat dissipation temperature distribution diagram of the topology optimization model; Figure 10 (c) is the heat dissipation temperature distribution diagram of the final post-processing model.

[0027] Figure 11 This is a schematic diagram illustrating the application of the final post-processed lower leg model in the leg structure of a hydraulic humanoid robot in an embodiment of the present invention. Figure 11 (a) is a side view; Figure 11 (b) is an oblique view.

[0028] In all the accompanying drawings, the same reference numerals are used to denote the same elements or structures, wherein: 1-knee joint shaft hole, 2-ankle joint parallel mechanism hydraulic cylinder fixing platform, 3-ankle joint shaft hole, 4-knee joint linkage mechanism fixing hole, 5-initial model built-in hydraulic oil pipe, 6-hydraulic servo valve mounting platform, 7-ankle joint parallel mechanism hydraulic cylinder upper mounting hole, 8-joint IO communication controller mounting platform, 9-ankle joint parallel mechanism hydraulic cylinder lower mounting hole, 10-high pressure end oil supply pipe P, 11-low pressure end oil return pipe T, 12-hydraulic cylinder rodless chamber oil supply pipe A, 13-hydraulic cylinder rod chamber oil supply pipe B, 14-hydraulic humanoid robot thigh, 15-knee joint linkage mechanism, 16-joint IO communication controller, 17-hydraulic humanoid robot foot, 18-hydraulic servo valve, 19-ankle joint parallel mechanism hydraulic cylinder, 20-hydraulic humanoid robot lower leg structure. Detailed Implementation

[0029] To enable those skilled in the art to better understand and implement the technical solutions of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be construed as limiting the scope of protection of the present invention.

[0030] Example This invention provides a method for optimizing the thermo-mechanical coupling of the lower leg structure of a hydraulic robot, such as... Figure 1 As shown, the specific steps include: Step 1: Establish an initial model based on the robot's lower leg size requirements, drive method, and external load for subsequent optimization. Then, define and partition the physical domain of the initial model, and use a four-node tetrahedral finite element mesh to mesh the model to be optimized. Record the coordinate information and numbers of the elements (nodes) in the generated mesh to obtain a mesh file containing model information.

[0031] Step Two: First, the mesh file containing model information is used as input. This mesh file includes the model's geometric information, physical domain information, and mesh information including node and element details. Subsequent optimization processes will iterate based on the optimization parameters defined in the input parameter file. Using the mesh file and the parameter file defining the optimization parameters as input, a loop of finite element analysis and topology optimization is performed. After the convergence condition is met, the model optimization result file in STL format is output.

[0032] S21: Model identification and parameter initialization are performed based on mesh files and parameter files. Specifically, the node density used as a design variable in the mesh file and parameter file is filtered using a Helmholtz filter. The filtered design variables are then subjected to Heaviside projection to ensure clear boundaries of the optimization results and to achieve global optimum results, preventing them from getting trapped in local optima. The projected node design variables are converted into element design variables and used as element density for analysis and optimization.

[0033] S22: In terms of unit density As input, a physical field model is performed, specifically establishing penalty functions for the temperature field, stress field, and thermal stress field, as shown in the following equation: ; ; ; in, The thermal conductivity of the unit cell. The Young's modulus of the unit cell. The thermal stress coefficient of the unit. For unit density, The coefficient of thermal expansion is These represent the thermal conductivity when the element is not empty and the thermal conductivity when the element is empty, respectively (generally, the minimum value greater than 0 is used for calculation). Similarly, These represent the Young's modulus of the element under different states. These are the penalty coefficients for the corresponding factors, typically taken as 3. The thermal conductivity of the unit is calculated from these values. Young's modulus With thermal stress coefficient This allows us to calculate the temperature field inside the structure, and further calculate the thermal stress field. The combined effect of this thermal stress field and the stress field generated by the external load forces yields the total displacement field. .

[0034] S23: Perform finite element analysis and calculate the objective function. The objective function of the thermo-coupling optimization problem is to minimize the structural compliance under a specified target volume fraction. This requires ensuring that the element density is within the range of 0 to 1 and is constrained by weakly coupled thermo-mechanical analysis under steady-state heat conduction. The formula is as follows:

[0035] ; in, For design variables, Let be the objective function. For force load vector, For the total displacement field, This is the global thermal stiffness matrix. For temperature field, For thermal load vector, The global stiffness matrix, These are the mechanical force vector and the thermal stress vector, respectively. To optimize the size of the results, The volume of the structure at the target volume fraction. This is the unit density vector.

[0036] After calculating the objective function, sensitivity is analyzed based on the objective function, and an optimization solver based on the Moving Asymptote Method (MMA) drives the update of design variables. The process continues iterating or outputs the final result based on the convergence condition. The convergence condition is determined by the threshold of the Heaviside projection control parameters input from the parameter file in step three. During the iterative update of the design variables, the control parameters change synchronously until the threshold is reached, indicating that the global optimum has been achieved. The Moving Asymptote Method (MMA) sets a pair of dynamically adjusted asymptotes (lower and upper bounds) for each design variable. By continuously updating the asymptote positions, the search range is tightened, ultimately achieving the update of design variables and the optimization of the results.

[0037] Since the analysis of thermo-coupling problems is quite complex, the adjoint method is introduced for analysis. The sensitivity analysis results of the objective function are as follows: ; in, The adjoint vector is expressed as follows: ; After completing the optimization loop, the final output will be an STL format result file, which can be directly used for 3D printing or further processed according to the actual application requirements.

[0038] This paper utilizes unstructured four-node tetrahedral element meshes to discretize complex robot structures in practical engineering applications. A series of node densities are defined using various smoothing and filtering schemes to represent the structural topology, thereby smoothing the optimization results and effectively avoiding surface unevenness issues that may be caused by unstructured four-node tetrahedral elements. Penalty functions for temperature, stress, and thermal stress fields are established when considering steady-state heat conduction and weak thermo-mechanical coupling. An optimization design formula with compliance minimization as the optimization objective is developed, and corresponding sensitivity analysis is performed using the adjoint method. An optimization solver is used to drive the update of design variables, and convergence conditions are determined. Finally, the optimized model is output.

[0039] Step 3: The STL file output by the optimization process can be directly used for 3D printing in practical applications.

[0040] Based on the aforementioned design framework for thermo-mechanical coupling optimization of the lower leg structure of a hydraulic robot, the lower leg structure of the hydraulic robot is fabricated, such as... Figure 2 As shown, the process includes: First, considering the actual application of the hydraulic humanoid robot, selecting the structure to be optimized and determining the optimization objective to locate the design problem, and establishing an initial model based on the structural size requirements and boundary constraints. The established model is then input into the meshing software in STP format to define the physical domain, such as the delineation of optimizable regions, non-optimizable regions, and external domains, as well as the definition of boundary conditions such as load force, heat source, and fixed points. After completing the physical domain division, the generation domain of the mesh is performed. The model is divided into elements with four nodes by four-node tetrahedral finite elements, and the meshing is performed within the aforementioned physical domain and has a physical domain label. An element can only exist in one physical domain and cannot be in multiple physical domains simultaneously; therefore, the mesh is generally generated starting from the physical domain boundary. After meshing is completed, an MSH mesh file will be output, containing the model's geometric information, physical domain information, and mesh information including node and element information. The MSH mesh file, along with a TXT parameter file containing optimization parameter definitions, is used as input to begin processing... Figure 1 The thermo-coupling topology optimization process is shown below and will not be elaborated further. After completing the thermo-coupling topology optimization, post-processing is required based on the actual application conditions. For example, in the post-processing optimization of hydraulic humanoid robot structural components, this includes surface smoothing, fine design of internal oil pipes, and design of component mounting holes. The final post-processed model needs to be verified through simulation and performance testing to ensure that the initial design optimization goals are achieved. After completing all the above steps, the processing method and technology, such as additive manufacturing or machining, are selected based on a comprehensive consideration of actual application conditions, material selection, time costs, and manufacturing costs. After processing, experimental testing can be conducted. Once the testing meets the requirements, practical application can be implemented to ultimately achieve optimized application and improved performance of the robot structure.

[0041] like Figure 3For the initial model of the lower leg structure of the hydraulic humanoid robot used for topology optimization, the external dimensions of the initial model of the lower leg structure are determined according to the overall size of the robot's lower leg. A knee joint pivot hole is set according to the joint connection requirements to fix the knee joint pivot 1 and connect it to the thigh 14 of the hydraulic humanoid robot; an ankle joint pivot hole 3 is set to fix the ankle joint dual-degree-of-freedom cross axis and connect it to the foot 17 of the hydraulic humanoid robot. A knee joint linkage mechanism fixing hole 4 is set according to the actuator installation method to connect and install the knee joint linkage mechanism 15, enabling the knee joint linkage mechanism 15 to drive the robot's lower leg to achieve circular motion around the knee joint pivot hole 1 under the drive of the hydraulic cylinder. A two-ankle joint parallel mechanism hydraulic cylinder fixing platform is set to fix the ankle joint parallel mechanism hydraulic cylinder 19, so that the ankle joint parallel mechanism hydraulic cylinder 19 can drive the hydraulic humanoid robot foot 17 to rotate around the ankle joint pivot hole 3 through the guide rail and guide rod. Considering the model's oil transmission and heat dissipation, the initial model has an internal hydraulic oil pipe 5, which is set as a heat source for subsequent optimization.

[0042] exist Figure 4 In the middle, for example Figure 3 The initial model of the lower leg structure was meshed using unstructured four-node tetrahedral elements, which enabled high-precision discretization of the model. An appropriately sized outer domain was used to enclose the model, ensuring effective identification and optimization of all regions. The number of nodes and elements after meshing remained within a certain range, with a ratio approximately 1:4, depending on the model complexity, the number of physical domains, mesh precision, and the total number of mesh elements.

[0043] like Figure 5 As shown, the overall result of the thermal coupling topology optimization of the lower leg structure of the hydraulic humanoid robot is symmetrical. According to the boundary conditions of different regions, the middle oil pipe heat dissipation area shows the characteristic of heat dissipation. The cylindrical heat dissipation fins are evenly distributed on the periphery of the oil pipe and extend in an oblique direction, so that the optimization result has good heat dissipation capacity in this area. The upper and lower stress areas show the characteristics of lightweighting and load-bearing. The optimization result enhances the structural stress performance in the stress area with oblique reinforcing ribs and removes materials to achieve the goal of lightweighting. The middle area is used to install the guide rail of the ankle joint parallel mechanism hydraulic cylinder 19 to drive the foot 17 of the hydraulic humanoid robot.

[0044] like Figure 6 As shown, the smoothing post-processing model preserves Figure 5 The main features of the thermal coupling topology optimization results of the lower leg structure of the hydraulic humanoid robot are shown. The corners at the structural connection are rounded to make the overall surface of the model smoother and easier to manufacture.

[0045] like Figure 7 As shown, the final post-processing model is in Figure 6 Based on the smoothed post-processing model, and further considering practical applications, the smoothed post-processing result model was further optimized. An internal oil circuit was designed to connect and traverse the external oil circuit, hydraulic servo valve 18, and ankle joint parallel mechanism hydraulic cylinder 19. Considering the installation of the hydraulic servo valve 18, a hydraulic servo valve mounting platform 6 was designed to achieve hydraulic oil flow and pressure control, thereby driving the extension and retraction of the ankle joint parallel mechanism hydraulic cylinder 19. Considering the installation of the joint I / O communication controller 16, a joint I / O communication controller mounting platform 8 was designed. Considering the installation of the ankle joint parallel mechanism hydraulic cylinder 19, an upper mounting hole 7 was designed to fix the upper rodless chamber fixing hole of the hydraulic cylinder, and a lower mounting hole 9 was designed to fix the lower rod chamber fixing hole of the hydraulic cylinder, thus completing the fixed installation and oil flow of the ankle joint parallel mechanism hydraulic cylinder 19.

[0046] like Figure 8 As shown, the oil circuit structure inside the post-processing model introduces external oil from the top. At the corresponding position on the hydraulic servo valve mounting platform 6, oil is supplied to the hydraulic servo valve 18 via the high-pressure end oil supply pipe P10 and the low-pressure end oil return pipe T11. The high-pressure end oil supply pipe P10 outputs 21MPa high-pressure hydraulic oil, while the low-pressure end oil return pipe T11 outputs low-pressure hydraulic oil that flows back to the oil tank. The hydraulic servo valve 18, based on the control signal transmitted by the joint I / O communication controller 16, regulates the flow of high-pressure hydraulic oil through the hydraulic cylinder rodless chamber oil pipe A12 to extend the hydraulic cylinder piston rod, or through the hydraulic cylinder rod chamber oil pipe B13 to retract the hydraulic cylinder piston rod, thereby achieving the drive control of the ankle joint parallel mechanism hydraulic cylinder 19.

[0047] like Figure 9 As shown, the quality of the initial model, topology optimization model, and final post-processing model of the lower leg structure were calculated respectively. The pink axis point represents the center of mass of the model, and the material chosen is 7075 aluminum alloy; based on the mass and the perpendicular distance between the center of mass and the knee joint axis. The moment of inertia was calculated. The results are shown in Table 1 below.

[0048] Table 1 Mechanical parameters of the lower leg structure model As can be seen, after topology optimization and final post-processing, the model's mass decreased from 2.51 kg to 0.81 kg, only 32.3% of the initial mass, and the center-of-gravity distance decreased from 153.21 mm to 131.14 mm, resulting in a significant reduction in the moment of inertia from 0.0589 kg. m 2 Reduced to 0.0139 kg m 2 The value is only 23.6% of the initial rotational inertia, which shows the effectiveness of the provided thermo-coupling topology optimization method in terms of lightweighting and the rationality of post-processing. The final post-processed model has excellent lightweight performance and dynamic motion performance, which can effectively improve the motion performance of the lower leg of the hydraulic humanoid robot.

[0049] like Figure 10 As shown, the heat dissipation effect of the initial model, the topology-optimized model, and the final post-processing model using internal hydraulic oil as the heat source was simulated. The heat source temperature was set to 60℃, and the ambient temperature was set to 20℃. The final post-processing model showed a reasonable temperature distribution, achieving a temperature reduction of 5~10℃ at the hydraulic cylinder inlet. The heat dissipation fins located outside the hydraulic pipes effectively dissipated heat. The heat dissipation power of different models under this temperature condition was calculated, and the results are shown in Table 2 below.

[0050] Table 2 Heat dissipation power of the lower leg structure model After topology optimization and final post-processing, the model quality decreased by 67.7% while the heat dissipation power increased by 3.52%, demonstrating the effectiveness of the provided thermo-coupling topology optimization method in heat dissipation optimization and the rationality of the post-processing. The final post-processed model has excellent heat dissipation performance, which can effectively improve the heat dissipation performance of the robot's hydraulic system, enabling the robot to maintain a suitable working temperature range for a long time, which is conducive to improving the overall performance of the robot.

[0051] The lower leg structure of the hydraulic humanoid robot described in this invention is based on the thermo-coupling topology optimization method, which performs thermo-coupling configuration design, resulting in a lightweight and heat dissipation-enhancing structure that effectively improves the dynamic and heat dissipation performance of the lower leg structure. At the same time, the use of hose-free technology enables oil flow inside the lower leg structure, avoiding the adverse effects of hydraulic hoses on the robot and improving the integration of the robot structure.

[0052] like Figure 11As shown, the lower leg structure 20 of the hydraulic humanoid robot is connected to the thighs 14 and 16 and the foot 17, together forming the leg structure of the hydraulic humanoid robot, including four degrees of freedom: hip joint pitch, knee joint pitch, ankle joint pitch, and roll. A hydraulic servo valve 18 is installed in the left-right direction of the lower leg structure 20, and a joint I / O communication controller is installed in the front-back direction of the lower leg structure 20. Both their external dimensions are within the envelope of the lower leg structure 20, enabling effective collision protection for the hydraulic servo valve 18 and the joint I / O communication controller 16, improving the collision avoidance and fall resistance of the hydraulic humanoid robot, and ensuring the stability of the robot's movement. The ankle joint parallel mechanism hydraulic cylinder 19 is installed at the front of the lower leg structure 20, effectively controlling two degrees of freedom of the foot 17, ensuring the robot's flexibility and the balance and stability of foot contact with the ground.

[0053] It should be noted that the specific embodiments described above enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way. Therefore, although the present invention has been described in detail in this specification and embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the present invention; and all technical solutions and improvements that do not depart from the spirit and scope of the present invention are covered within the protection scope of the present invention patent. No reference numerals in the claims should be construed as limiting the scope of the claims. Any simple variations or equivalent substitutions of technical solutions that can be readily obtained by those skilled in the art within the scope of the technology disclosed in the present invention are within the protection scope of the present invention.

Claims

1. A method for optimizing the thermo-mechanical coupling of the lower leg structure of a hydraulic robot, characterized in that, Includes the following steps: An initial model of the robot's lower leg is established based on the size requirements, driving method, and external load. The initial model is then meshed using an unstructured four-node tetrahedral element mesh to obtain a mesh file containing model information. A parameter file including optimization parameters is defined. Using node density as a design variable, the design variable is initialized based on the mesh file and parameter file, and the continuous design variable is converted into element density through weighted calculation. Penalty functions for temperature field, stress field and thermal stress field are constructed. The unit density is used as the input of the penalty function to obtain the temperature field of the lower leg structure. The thermal stress field is calculated based on the temperature field of the lower leg structure. The total displacement field is obtained based on the combined action of the thermal stress field and the stress field formed by the external load force. The objective function for thermo-mechanical coupling optimization is constructed by minimizing structural flexibility. The total displacement field is input into the objective function, and the sensitivity of the objective function to the design variables is calculated. Based on the sensitivity, the design variables are iteratively adjusted to seek optimization. Iteration conditions are set, and after the conditions are met, the optimization results of the robot's lower leg topology model in STL format are output.

2. The method for optimizing the thermo-mechanical coupling of the lower leg structure of a hydraulic robot according to claim 1, characterized in that, The design variables are initialized based on the mesh file and parameter file, specifically by filtering the design variables in the mesh file and parameter file using a Helmholtz filter; the filtered design variables are then subjected to Heaviside projection, and the projected node design variables are converted into element densities.

3. The method for optimizing the thermo-mechanical coupling of the lower leg structure of a hydraulic robot according to claim 1, characterized in that, The penalty functions for the temperature field, stress field, and thermal stress field are specifically as follows: ； ； ； in, The thermal conductivity of the unit cell. The Young's modulus of the unit cell. The thermal stress coefficient of the unit. For unit density, The coefficient of thermal expansion is These represent the thermal conductivity when the unit cell is not empty and the thermal conductivity when the unit cell is empty, respectively. These are the Young's moduli of the element under different states. These are the penalty coefficients for the corresponding coefficients.

4. The method for optimizing the thermo-mechanical coupling of the lower leg structure of a hydraulic robot according to claim 1, characterized in that, The objective function is specifically: ； in, For design variables, Let be the objective function. For force load vector, For the total displacement field, This is the global thermal stiffness matrix. For the temperature field, For thermal load vector, The global stiffness matrix. These are the mechanical force vector and the thermal stress vector, respectively. To optimize the size of the results, The volume of the structure at the target volume fraction. This is the unit density vector.

5. The method for optimizing the thermo-mechanical coupling of the lower leg structure of a hydraulic robot according to claim 4, characterized in that, The sensitivity is specifically calculated using the following formula: ； in, This is the adjoint vector.

6. The method for optimizing the thermo-mechanical coupling of the lower leg structure of a hydraulic robot according to claim 1, characterized in that, The specific iteration condition is to calculate the difference between the current objective function value and the objective function value of the previous round. If the difference is less than a set threshold, the iteration stops.

7. The method for optimizing the thermo-mechanical coupling of the lower leg structure of a hydraulic robot according to claim 1, characterized in that, When iteratively adjusting the design variables based on the aforementioned sensitivity for optimization, the moving asymptote method is used to set a pair of dynamically adjusted lower and upper bound asymptotes for each design variable, update the asymptote positions to tighten the search range, and update the design variables.

8. The method for optimizing the thermo-mechanical coupling of the lower leg structure of a hydraulic robot according to claim 1, characterized in that, When performing finite element mesh generation, a physical domain is defined to divide the initial model into regions, distinguishing between optimizable regions, non-optimizable regions, and external domains. The optimizable regions are those where the material distribution can be adjusted through topology optimization. The non-optimizable regions are the key functional structures of the lower leg that need to be fully preserved. The external domain is the virtual boundary region that encloses the model.

Images