Three-dimensional parametric simulation design method for intelligent bed structure

Abstract

The present invention relates to the technical field of smart bed structures, and in particular to a three-dimensional parametric simulation design method for smart bed structures. The method comprises the following steps: constructing a three-dimensional structural model based on the smart bed design structure to obtain a three-dimensional structural model of the smart bed; performing a body dynamic structural support analysis on the three-dimensional structural model of the smart bed, and designing an adaptive adjustment mechanism to obtain an adaptive adjustment mechanism of the smart bed; performing error parameter balance optimization on the adaptive adjustment mechanism of the smart bed, and performing optimization adjustment to obtain an adaptive adjustment optimization mechanism of the smart bed; constructing an adjustment simulation mechanism model for the adaptive adjustment optimization mechanism of the smart bed to obtain an adjustment simulation mechanism model of the smart bed; and sending the adjustment simulation mechanism model of the smart bed to the smart bed control unit to perform parametric simulation design of the smart bed. The present invention makes the smart bed structure technology more perfect by optimizing the smart bed structure technology.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent bed structures, and in particular to a three-dimensional parameterized simulation design method for intelligent bed structures. Background Art

[0002] The demand for smart beds in the smart home field is increasing. As an innovative home product that combines ergonomics, sensor technology and intelligent control, smart beds have become an important tool for improving sleep quality, realizing personalized health monitoring and optimizing the quality of life of users. The emergence of three-dimensional parametric simulation design methods not only improves the efficiency and accuracy of smart bed design, but also provides a more intelligent product development path for the smart home industry. With the continuous advancement of artificial intelligence and Internet of Things technologies, this method can be further combined with intelligent sensing systems and big data analysis to achieve personalized design and dynamic adjustment based on user data. In the future, this method is expected to become the core technology in the field of smart homes, injecting new vitality into the design and optimization of innovative products such as smart beds, and promoting the healthy home industry to move towards a more intelligent, personalized and efficient direction. However, a traditional three-dimensional parametric simulation design method for smart bed structures has the problem of being unable to accurately perform adaptive adjustments according to users of different body shapes, and the accuracy of adaptive adjustments is not high. Summary of the invention

[0003] Based on this, it is necessary to provide a three-dimensional parametric simulation design method for the smart bed structure to solve at least one of the above technical problems.

[0004] To achieve the above object, a three-dimensional parametric simulation design method for a smart bed structure is provided, the method comprising the following steps:

[0005] Step S1: Acquire the design structure and bed surface material data of the smart bed; construct a three-dimensional structure model based on the design structure of the smart bed to obtain a three-dimensional structure model of the smart bed;

[0006] Step S2: acquiring body shape data of people of different body shapes; performing body shape dynamic structural support analysis on the three-dimensional structure model of the smart bed based on the body shape data of people of different body shapes to obtain body shape dynamic structural support data; performing adaptive adjustment mechanism design based on the body shape dynamic structural support data to obtain an adaptive adjustment mechanism of the smart bed;

[0007] Step S3: performing error parameter balance optimization on the smart bed adaptive adjustment mechanism to obtain error parameter balance optimization data; optimizing and adjusting the smart bed adaptive adjustment mechanism based on the error parameter balance optimization data to obtain the smart bed adaptive adjustment optimization mechanism;

[0008] Step S4: construct an adjustment simulation mechanism model for the smart bed adaptive adjustment optimization mechanism to obtain the smart bed adjustment simulation mechanism model; send the smart bed adjustment simulation mechanism model to the smart bed control unit to execute the smart bed parametric simulation design.

[0009] The present invention first needs to collect the design structure information of the smart bed, including the physical properties of the bed frame, support system, bed surface material, etc. These data provide the necessary basis for subsequent three-dimensional modeling. Through these data, a high-precision three-dimensional structure model of the smart bed can be constructed to intuitively display the design details and functional layout of the bed. The beneficial effect of this step is to ensure that the bed structure can accurately express and reflect the mechanical properties and adaptability during actual use, laying a solid foundation for subsequent dynamic analysis and optimization. First, it is necessary to obtain the physical data of people of different body types, including physiological characteristics such as weight, height, and body shape distribution. Based on these data, the three-dimensional structure model of the smart bed is subjected to body shape dynamic support analysis to simulate the requirements of different body types for bed surface support. This analysis can provide dynamic feedback for the support system of the smart bed to ensure that different users can get appropriate support when using it. According to the analysis results, an adaptive adjustment mechanism of the smart bed can be designed so that the bed can automatically adjust the support system according to the changes in body shape, thereby improving comfort and health. The beneficial effect of this step is to enable the smart bed to have a highly personalized adjustment capability, provide accurate support according to the needs of different body types, and enhance the user experience. Error optimization of the adaptive adjustment mechanism of the smart bed. Since there are certain errors or deviations in the adjustment mechanism of the bed in actual applications, this will affect the performance of the smart bed. Therefore, by collecting and analyzing the error parameters, the optimization algorithm can be used to balance and adjust to ensure the precise adjustment of the smart bed under various conditions of use. Through error balance optimization, the response speed and adjustment accuracy of the adaptive mechanism can be improved, the impact of errors can be reduced, and the adjustment can be smoother and more stable. The beneficial effect of this step is to improve the adjustment accuracy and reliability of the smart bed in actual use, and ensure that the automatic adjustment mechanism of the bed can always provide ideal support and comfort. It is necessary to build a simulation mechanism model for the adjustment of the smart bed, and verify the performance of the adaptive adjustment mechanism by simulating the performance of the smart bed in different usage scenarios through computer simulation. The simulation model can predict the behavior of the smart bed through virtual testing before actual manufacturing, so as to find potential problems and make adjustments and optimizations. After the simulation model is completed, it is input into the control unit of the smart bed, and the parametric simulation design is automatically performed to further improve the adjustment mechanism. The beneficial effect of this step is that problems can be found in advance through virtual testing and simulation, the development cost of physical prototypes can be reduced, and the optimization iteration process of the product can be accelerated. Therefore, the present invention optimizes a traditional three-dimensional parametric simulation design method for smart bed structures, solves the problem that the traditional three-dimensional parametric simulation design method for smart bed structures cannot accurately perform adaptive adjustments according to users of different body shapes, and the adaptive adjustment accuracy is not high, improves the accuracy of adaptive adjustments according to users of different body shapes, and improves the adaptive adjustment accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic diagram of the steps of a three-dimensional parametric simulation design method for an intelligent bed structure;

[0011] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0012] Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0013] Figure 4 for Figure 3 Detailed implementation steps of step S33 are shown in the flowchart.

[0014] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0015] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0016] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0017] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0018] To achieve this, please refer to Figures 1 to 4 , a three-dimensional parametric simulation design method for an intelligent bed structure, the method comprising the following steps:

[0019] Step S1: Acquire the design structure and bed surface material data of the smart bed; construct a three-dimensional structure model based on the design structure of the smart bed to obtain a three-dimensional structure model of the smart bed;

[0020] Step S2: obtaining body shape data of people of different body shapes; performing body shape dynamic structure support analysis on the three-dimensional structure model of the smart bed based on the body shape data of people of different body shapes to obtain body shape dynamic structure support data; performing adaptive adjustment mechanism design based on the body shape dynamic structure support data to obtain an adaptive adjustment mechanism of the smart bed; wherein step S2 includes:

[0021] Step S21: Acquire body shape data of people of different body shapes;

[0022] Step S22: performing a body shape dynamic structure support analysis on the three-dimensional structure model of the intelligent bed based on the body shape data of human bodies of different body shapes, and obtaining body shape dynamic structure support data;

[0023] Step S23: performing mechanical response mapping analysis on the bed surface material data according to the body shape dynamic structure support data to obtain the bed surface material mechanical response mapping data;

[0024] Step S24: designing an adaptive adjustment mechanism based on the body shape dynamic structure support data and the bed surface material mechanical response mapping data to obtain an intelligent bed adaptive adjustment mechanism;

[0025] Step S3: performing error parameter balance optimization on the smart bed adaptive adjustment mechanism to obtain error parameter balance optimization data; optimizing and adjusting the smart bed adaptive adjustment mechanism based on the error parameter balance optimization data to obtain the smart bed adaptive adjustment optimization mechanism;

[0026] Step S4: construct an adjustment simulation mechanism model for the smart bed adaptive adjustment optimization mechanism to obtain the smart bed adjustment simulation mechanism model; send the smart bed adjustment simulation mechanism model to the smart bed control unit to execute the smart bed parametric simulation design.

[0027] In the embodiment of the present invention, reference Figure 1 FIG. 1 is a schematic diagram of a three-dimensional parameterized simulation design method for a smart bed structure according to the present invention. In this example, the three-dimensional parameterized simulation design method for a smart bed structure includes the following steps:

[0028] Step S1: Acquire the design structure and bed surface material data of the smart bed; construct a three-dimensional structure model based on the design structure of the smart bed to obtain a three-dimensional structure model of the smart bed;

[0029] In an embodiment of the present invention, when obtaining the design structure of the smart bed, the geometric parameters (such as length, width, and thickness) of each component are recorded one by one by physically disassembling and measuring the structural elements of the smart bed. In order to obtain the material data of the bed surface, mechanical testing equipment is used to measure different materials, including compression strength test, rebound rate measurement, and surface friction coefficient calculation. These data are numerically fitted by the piecewise approximation method to form a material property function expression. When constructing a three-dimensional structural model of the smart bed design structure, an initial geometric model is first established based on its geometric parameters, and the key connection points of the smart bed are marked by the distributed node marking method to generate a three-dimensional point cloud. These point cloud data are connected through the Delaunay triangulation algorithm to form a gridded geometric structure. Finite element analysis is used to statically simulate the stress conditions of the connection points of the overall structure to ensure that the mesh density meets the mechanical accuracy requirements and complete the construction of the three-dimensional structural model of the smart bed.

[0030] Step S2: obtaining body shape data of people of different body shapes; performing body shape dynamic structure support analysis on the three-dimensional structure model of the smart bed based on the body shape data of people of different body shapes to obtain body shape dynamic structure support data; performing adaptive adjustment mechanism design based on the body shape dynamic structure support data to obtain an adaptive adjustment mechanism of the smart bed; wherein step S2 includes:

[0031] Step S21: Acquire body shape data of people of different body shapes;

[0032] Step S22: performing a body shape dynamic structure support analysis on the three-dimensional structure model of the intelligent bed based on the body shape data of human bodies of different body shapes, and obtaining body shape dynamic structure support data;

[0033] Step S23: performing mechanical response mapping analysis on the bed surface material data according to the body shape dynamic structure support data to obtain the bed surface material mechanical response mapping data;

[0034] Step S24: designing an adaptive adjustment mechanism based on the body shape dynamic structure support data and the bed surface material mechanical response mapping data to obtain an intelligent bed adaptive adjustment mechanism;

[0035] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0036] Step S21: Acquire body shape data of people of different body shapes;

[0037] In an embodiment of the present invention, a basic sample library is established by collecting human body shape data. First, a three-dimensional scanner is used to scan the sample population to capture the geometric features of the whole body, including data such as height, weight, shoulder width, waist circumference, hip circumference, and limb length. These scanned data are saved in the form of point clouds, and a three-dimensional surface model of each body shape is generated by a polygon reconstruction algorithm. In order to enrich the accuracy of the data, anthropometric standards (such as ISO 7250) are used to annotate and correct key points and dimensions to ensure the consistency of the data globally. In order to reflect the distribution characteristics of different body shapes, the data is classified according to dimensions such as age, gender, and body mass index (BMI), and principal component analysis (PCA) is used to extract the main change patterns of body shape data. Finally, all body shape data are organized into a parameterized data set, including the numerical range of key dimensions and the corresponding three-dimensional point cloud data.

[0038] Step S22: performing a body shape dynamic structure support analysis on the three-dimensional structure model of the intelligent bed based on the body shape data of human bodies of different body shapes, and obtaining body shape dynamic structure support data;

[0039] In an embodiment of the present invention, based on the body shape data in step S21, combined with the three-dimensional structure model of the smart bed, a mechanical analysis method is used to calculate the dynamic support performance. The human body shape data is registered with the three-dimensional model of the smart bed, and the human body model and the bed surface model are accurately aligned through a rigid transformation algorithm, so that the support area of ​​each body shape on the bed surface can be clearly defined. The contact mechanics analysis method is used to calculate the force distribution of the smart bed under different body shapes. Assuming that the contact between the human body and the bed surface is an elastic contact relationship, the pressure distribution in the contact area is obtained by the Hertz contact theory. Then, a finite element analysis tool is used to simulate the deformation of the bed under different support states, and dynamic support data is generated with body shape as a variable. Finally, the analysis results are summarized as a body shape dynamic structure support data set, including contact area, maximum pressure, support uniformity index and bed deformation characteristic value.

[0040] Step S23: performing mechanical response mapping analysis on the bed surface material data according to the body shape dynamic structure support data to obtain the bed surface material mechanical response mapping data;

[0041] In an embodiment of the present invention, the mechanical response characteristics of the bed material are mapped and analyzed with the actual human body contact state according to the dynamic structural support data of the body shape. First, based on the elastic modulus and Poisson's ratio in the bed material data, the deformation of the bed under a given pressure distribution is calculated by mechanical formulas. Then, a correlation model between material response and support pressure is established, and the stress-strain distribution of different materials under dynamic support state is analyzed by adopting the linear elastic assumption. In order to improve the analysis accuracy, the Poisson effect is introduced to simulate the deformation of the material in the three-dimensional direction, and the local area is refined by high-resolution meshing. Combined with the contact characteristics of each body shape, the deformation variables and stress distribution diagrams of different areas of the bed are generated. Finally, all mapping results are integrated into the mechanical response mapping data of the bed material, including the stress field, strain field and contact pressure change curve under a specific body shape.

[0042] Step S24: Design an adaptive adjustment mechanism based on the body shape dynamic structure support data and the bed surface material mechanical response mapping data to obtain an intelligent bed adaptive adjustment mechanism.

[0043] In an embodiment of the present invention, the dynamic structural support data of the body shape and the mechanical response mapping data of the bed material are used as input, and the optimization design method is used to construct an adaptive adjustment mechanism for the smart bed. First, the stress-strain curve characteristics when different body shapes are in contact with the bed are analyzed to determine the optimal support distribution parameters. Then, according to the actual degrees of freedom of movement of the bed adjustment component, an adaptive adjustment strategy is designed, including angle adjustment, lifting movement, and local hardness adjustment. A gradient descent-based optimization algorithm is used to solve the multi-objective function of the adjustment mechanism to simultaneously optimize comfort and support uniformity. The adjustment strategy is verified by numerical simulation to ensure that it can achieve fast response and precise adjustment under different body shape conditions. Finally, the output adaptive adjustment mechanism of the smart bed describes the adjustment rules, parameter ranges, and response priorities in the form of logical relationships, providing a standardized design basis for the next step of control unit implementation.

[0044] In an embodiment of the present invention, the acquisition of human body shape data adopts three-dimensional laser scanning technology to scan the sample group to obtain three-dimensional appearance data of various body shapes. The scanned data is reconstructed by B-spline curve to establish a body shape characteristic curve database under various sleeping postures. These data are classified and clustered to be divided into sleeping posture characteristic data of different body shape groups. Based on the three-dimensional structure model of the intelligent bed, when the human body shape data is subjected to dynamic structural support analysis, the inverse matrix method is used to solve the contact force distribution between the bed structure and the human body force surface. According to the contact force distribution and the characteristic function of the bed material, the dynamic support surface is optimized and adjusted by the Lagrange multiplier method to generate body shape dynamic structural support data. The support data is input into the structural adjustment system, the elastic adjustment formula is derived by geometric nonlinear theory, the adaptive adjustment mechanism is designed, and the preliminary adjustment parameters are generated.

[0045] Step S3: performing error parameter balance optimization on the smart bed adaptive adjustment mechanism to obtain error parameter balance optimization data; optimizing and adjusting the smart bed adaptive adjustment mechanism based on the error parameter balance optimization data to obtain the smart bed adaptive adjustment optimization mechanism;

[0046] In an embodiment of the present invention, the error optimization of the adaptive adjustment mechanism of the smart bed is based on a multivariate nonlinear regression method. First, a large amount of adjustment test data is collected, including pressure feedback values ​​and adjustment deviation values. The data dimension is degraded by principal component analysis to extract the key factors affecting the error. The extracted factors are parameter normalized to generate optimization weights. Using the error optimization function, the error of the adjustment system is converged and optimized according to the gradient descent method. Through continuous iterations, the adjustment deviation is gradually reduced so that the optimization result is close to the theoretical ideal value, and finally the error parameter balance optimization data is generated. Based on these optimization data, the adjustment mechanism parameters are readjusted, and the dynamic response differences before and after the optimization are compared to further verify the accuracy of the optimization effect.

[0047] Step S4: construct an adjustment simulation mechanism model for the smart bed adaptive adjustment optimization mechanism to obtain the smart bed adjustment simulation mechanism model; send the smart bed adjustment simulation mechanism model to the smart bed control unit to execute the smart bed parametric simulation design.

[0048] In an embodiment of the present invention, the construction process of the adjustment simulation mechanism model is based on the multi-dimensional fitting of the pressure distribution matrix and the time dynamic curve. First, the optimized adaptive adjustment mechanism is subjected to a multi-scenario loading test to record the force changes in different areas of the bed surface during the change time. The dynamic pressure change model is constructed using the finite difference method, and the adjustment simulation mechanism model of the smart bed is generated by the tensor decomposition method. The adjustment simulation mechanism model is transmitted to the control unit of the smart bed through the hardware interface. When executing the simulation design, the control unit executes the dynamic adjustment process one by one and monitors the feedback in real time according to the parameterized formula in the adjustment simulation mechanism model to ensure the stability of the adjustment mechanism. Finally, by measuring the degree of fit between the actual adjustment data and the model prediction data, the implementation effect of the parameterized simulation design and the effectiveness of the model are verified.

[0049] Preferably, step S1 comprises the following steps:

[0050] Step S11: Acquire the design structure and bed surface material data of the smart bed;

[0051] Step S12: marking the structure connection points of the smart bed design structure to obtain the structure connection point data of the smart bed;

[0052] Step S13: defining the adjustable motion trajectory range of the smart bed design structure according to the smart bed structure connection point data to obtain the adjustable motion trajectory range data;

[0053] Step S14: construct a three-dimensional structure model based on the adjustable motion trajectory range data and the smart bed structure connection point data to obtain a three-dimensional structure model of the smart bed.

[0054] In the embodiment of the present invention, the acquisition of the design structure data of the smart bed is based on the physical measurement and data archiving method. By disassembling the various structural components of the smart bed, high-precision calipers and laser rangefinders are used to measure the geometric parameters of each component, such as length, width, thickness, angle and hole position, and the measurement results are recorded in a standardized data table. For complex structural parts, industrial CT scanning is used to reconstruct the internal geometric features in three dimensions to ensure the integrity and high precision of the data. The acquisition of the bed surface material data adopts the laboratory mechanical testing method. The bed surface samples are selected, and the compression strength test (measuring the deformation of the sample under different loading forces), elastic recovery rate measurement (evaluating its rebound performance through loading-unloading cycle test) and friction coefficient determination (measuring the friction characteristics of the sample surface through the sliding friction test bench) are carried out in turn. All data are converted into numerical functions that characterize material properties, such as elastic modulus and Poisson's ratio expressions, to provide basic material data for subsequent structural analysis. After completing the acquisition of the design structure data of the smart bed, the key connection points of the structure are marked using the physical connection point marking method. First, by analyzing the functional requirements and design characteristics of the smart bed, the key connection points that need to be marked are determined, including support points, motion hinges, fixed points, etc. Through the digital coding tool, a unique identifier is assigned to each connection point and marked in the design structure diagram. In the actual structure, the spatial position coordinates of each connection point are determined by a laser pointer, and the three-dimensional coordinate recording tool is used to accurately record them. To ensure the accuracy of the marking, the total station is used for multiple measurements, and the final coordinate value is determined after eliminating the error. The connection point data is uniformly stored as a structural connection point database, which contains key information such as the spatial position, type, and material properties of the connection point, providing a basis for the subsequent definition of the adjustable motion trajectory range. The definition of the adjustable motion trajectory range is based on geometric analysis and kinematic simulation. Using the smart bed structure connection point data obtained in the previous step, a kinematic model based on joints and connecting rods is constructed. Through the analytical geometry method, the degrees of freedom of each motion component in three-dimensional space are determined, including translation, rotation, and compound motion. For each motion connection point, the rigid body kinematics theory is applied, and the motion range of each joint under different initial states is calculated through matrix transformation. The motion trajectory of the entire bed is further modeled using the generalized coordinate method, and the complete motion envelope diagram is drawn by generating the trajectory curve of the extreme position point. These data are organized into structured adjustable motion trajectory range data, and the maximum adjustment range of each connection point is clarified, laying the foundation for the subsequent three-dimensional model construction. The construction of the three-dimensional structural model of the smart bed is based on the multi-level grid modeling method. First, the smart bed structure connection point data and the adjustable motion trajectory range data are input into the geometric modeling framework. The spatial triangulation technology is used to construct a preliminary skeleton model by connecting key points. According to the degree of freedom of motion of each connection point, the corresponding surface model is generated using a smooth interpolation algorithm to ensure that the model maintains continuity and smoothness within the range of motion.The finite element method was used to analyze the mechanical properties of the initially constructed three-dimensional model to verify the deformation and stability of the model under different stress conditions. According to the analysis results, the local weak areas in the model were meshed and the stress distribution was recalculated. Finally, a three-dimensional structure model of the smart bed that met the accuracy requirements was generated. The model contains the location, material properties and range of motion information of all connection points, and is output in a standardized file format to provide a basis for subsequent simulation and adjustment.

[0055] Preferably, step S22 includes the following steps:

[0056] Step S221: performing body curve analysis of different sleeping postures on the body shape data of people of different body shapes to obtain body shape curve data of different sleeping postures;

[0057] Step S222: performing three-dimensional feature point extraction processing on the body curve data of different sleeping postures to obtain a set of coordinates of key nodes of the human body;

[0058] Step S223: dynamically adjusting the force-bearing curved surface of the body curve data of different sleeping postures according to the human body key node coordinate set to obtain dynamic support force-bearing curved surface adjustment data;

[0059] Wherein, step S223 includes:

[0060] Step S2231: performing sleep turning force distribution analysis on the key node coordinate set of the human body to obtain node sleep turning force distribution data;

[0061] Step S2232: performing local curvature force interpolation processing of different sleeping postures on the body curve data of different sleeping postures according to the node sleep flipping force distribution data, so as to obtain local curvature force interpolation data of different sleeping postures;

[0062] Step S2233: performing estimation of joint sideways force difference on the local curvature force interpolation data of different sleeping postures to obtain joint sideways force difference data;

[0063] Step S2234: dynamically adjusting the force-bearing curved surface of the body curve data of different sleeping postures based on the local curvature force interpolation data of different sleeping postures and the joint side force difference data to obtain dynamic support force curved surface adjustment data;

[0064] Among them, in the embodiment of the present invention, when the human body key node coordinate set is subjected to sleep flip force distribution analysis, the rigid kinematic model is first used to simulate different sleep flip processes, including multiple posture transformation paths from supine to side-lying and from side-lying to prone. By analyzing the trajectory of the joint point during the flipping process, the spatial position and displacement vector of the key node in each posture change are calibrated. The force change of each key node during the flipping process is calculated using the finite element method, and the pressure distribution of its contact with the support surface is analyzed in combination with the force history data of the node. In this process, the force calculation results are verified by the distributed pressure sensor data to ensure that the force change of the key node conforms to the mechanical law of the actual flipping motion. Finally, the node sleep flip force distribution data is formed, and the data content includes the flipping trajectory, the node pressure change trend and the force extreme position during the flipping process. Using the node sleep flip force distribution data, combined with the body shape curve data of different sleeping postures, the force change of the local area of ​​the curve is first analyzed. The piecewise polynomial interpolation method is used to fit the force distribution curvature function within each key node and its neighborhood in the body shape curve, and a local force-curvature relationship model is established. The node pressure change trend is combined with the curvature fitting function to calculate the force gradient distribution of each curvature change point in different postures. The local force density and curvature correction value of the curve in each posture change are obtained through interpolation calculation. Finally, the local curvature force interpolation data of different sleeping postures are obtained, including the location of the local curvature change point, the force interpolation function parameters and the force gradient distribution matrix. The local curvature force interpolation data is input into the joint force analysis module to estimate the difference of the side-lying force characteristics of the joint area (such as the shoulder joint, hip joint and knee joint). First, the local support force of the joint area is extracted based on the three-dimensional coordinates of the node and the curvature characteristics of the neighborhood points. Then, the force difference of each joint in the side-lying posture and the supine posture is calculated by the mechanical equilibrium equation. The contact stress distribution analysis method is used to quantify the force peak and force distribution deviation of the joint area in different postures. By comparing the force uniformity of the joint area in the side-lying posture, the local adaptability of the intelligent bed support system is evaluated. Finally, the side-lying force difference data of the joint is generated, including the joint force distribution diagram, the difference quantification index and the regional pressure mapping result. Combining the local curvature force interpolation data and the joint side force difference data, the three-dimensional support surface of the smart bed is adjusted through the dynamic support optimization algorithm. First, the support point set of the surface is defined, and the key points of the body curve are projected into the three-dimensional coordinate system of the bed surface to construct the initial support surface. Then, the curvature parameters of the local area of ​​the surface are corrected using the joint force difference data to ensure that the pressure distribution of the support area and the human body curve is balanced. During the adjustment process, the gradient descent method is used to iteratively optimize the force uniformity of the surface, and gradually correct the height and curvature of the surface nodes. The force simulation of the optimized surface is carried out to verify its support performance and pressure distribution consistency under different postures.Finally, the dynamic support force surface adjustment data is generated, and the data content includes the optimized surface node coordinates, regional curvature parameters and the overall support force matrix.

[0065] Step S224: performing regional difference analysis on the dynamic support force curved surface adjustment data to obtain regional difference data of the support force curved surface;

[0066] Step S225: Perform body shape dynamic structure support analysis on the three-dimensional structure model of the smart bed according to the dynamic support force surface adjustment data and the support force surface area difference data to obtain body shape dynamic structure support data.

[0067] In the embodiment of the present invention, after obtaining the three-dimensional data of human bodies of different body types, they are classified into typical sleeping postures such as side lying, supine lying, and prone lying. First, the posture of the three-dimensional model of the human body is adjusted by rotation matrix and rigid transformation so that it simulates the real sleeping posture in space. Then, curve segmentation is performed on the adjusted model surface to extract the contour of the human body contacting the bed surface. This process uses a surface cutting algorithm, taking the lowest height area of ​​the model as a reference, so as to accurately divide the body curve in contact with the bed surface. The curve data is fitted using a multi-segment spline interpolation method to calculate the curvature distribution and curve equation under each posture. By analyzing the curvature and contact area, the regional characteristics and pressure distribution trends of the human body in contact with the bed surface under different postures are obtained. Finally, the body curve data of different sleeping postures are sorted out, including the curve parameters (such as curvature, length, contact point density) corresponding to each posture and the geometric characteristics of the contact area. The body curve data in step S221 is used as input, and a feature extraction algorithm based on convex hull and curvature change is used to calibrate the key nodes on the curve. First, the arc length of the curve is calculated by a geometric scanning method, and the initial point set is sampled at a fixed interval. Then, the curvature change rate of the curve is calculated by the quadratic difference method, and the curvature extreme point is selected as the initial feature point. Next, the contour boundary points of the curve are extracted by the three-dimensional convex hull algorithm to define the main support parts of the human body, such as the head, shoulder, waist and ankle positions. For each feature point, its coordinates in three-dimensional space are calculated to form a set of key node coordinates of the human body. Finally, these key node data are saved in the form of a matrix, and each row corresponds to the spatial coordinates of a node and its position index on the curve. Based on the key node coordinate set generated in step S222, combined with the body curve data under each sleeping posture, the dynamic support force surface is generated by the interpolation algorithm. First, quadratic spline interpolation is performed between the key nodes of the curve to generate a smooth three-dimensional surface to simulate the contact area between the human body curve and the bed surface. Then, the mechanical equilibrium equation is used to calculate the force state of each node on the surface to optimize its support performance under different postures. A dynamic adjustment mechanism is introduced to redistribute the curvature and contact area of ​​each part of the surface according to the difference in pressure distribution within the support surface. Through iterative optimization, the maximum contact pressure of each part is controlled within a preset range, thereby improving the overall support uniformity. Finally, the output dynamic support force surface adjustment data includes three-dimensional surface coordinates, node force matrix and pressure distribution after surface optimization. The dynamic support force surface adjustment data is input into the regional difference analysis process, and the force uniformity of different areas of the surface is calculated by the block partitioning method. First, the surface is divided into several sub-areas using the spatial partitioning algorithm, and the node distribution and area in each area are quantified. Then, the average pressure and pressure fluctuation range of each area are calculated by the node force matrix. The standard deviation analysis method and discretization difference index are used to quantify the force uniformity differences between the sub-areas, and the regional difference index of the overall support surface is calculated.Combined with the distribution histogram and color map, the force characteristics of different areas are output in a visual form to form the difference data of the support force surface area, including the pressure value, area ratio and difference parameter of each sub-area. The output data of step S223 and step S224 are used to analyze the dynamic adaptability of the intelligent bed structure model to different body shapes. First, the dynamic support force surface data is superimposed and matched with the three-dimensional structural model of the intelligent bed to ensure that the support surface and the contact area of ​​the bed body are completely consistent. Then, the difference data of the support force surface area is input into the mechanical analysis module to evaluate the stress distribution and deformation characteristics of each adjustment component of the bed body in different areas. The motion trajectory of the bed structure during the adjustment process is calculated by rigid body dynamics analysis, and the linkage design of each adjustment component of the bed body is optimized in combination with the displacement field distribution data. The support parameters are continuously adjusted by the iterative method, and the dynamic structural support data of the body shape is finally output, including the adjustment motion path, the bed force distribution curve and the overall support adaptation performance index.

[0068] Preferably, step S2233 includes the following steps:

[0069] The force gradient change between nodes is calculated for the local curvature force interpolation data of different sleeping postures to obtain the force gradient change data between nodes;

[0070] According to the force gradient change data between nodes, the partition equilibrium state analysis between nodes is performed to obtain the force equilibrium state data of the node partition;

[0071] Based on the node partition force balance state data, the node joint force offset value is calculated for the force gradient change data between nodes to obtain the node joint force offset value;

[0072] The joint lateral force difference is estimated according to the node joint force offset value to obtain the joint lateral force difference data.

[0073] In an embodiment of the present invention, the local curvature force interpolation data of different sleeping postures are mathematically analyzed. For each set of force data, a force vector relationship is established in units of adjacent nodes along the direction of curvature change, and the continuous change gradient of the force is analyzed. The vector differential method is applied to calculate the force difference between each two adjacent nodes, and a gradient change matrix is ​​established in combination with the curvature equation. Each element of the matrix represents the amplitude of the force gradient change between adjacent nodes. In order to avoid data redundancy and noise interference, the least squares method is used to smooth the gradient change data, and finally the force gradient change data between nodes is formed. After completing the force gradient change calculation, the force gradient change data is divided into multiple regions according to the geometric distribution of the nodes on the surface. Based on the finite element analysis method, the force distribution characteristics of the nodes in each region are analyzed, and the difference between the total force in the region and the total force of the regional boundary nodes is calculated. Through the regional balance equation, the force balance state of each region is iteratively calculated to ensure that the force difference between the internal nodes and the boundary nodes of the region meets the mechanical balance condition. Finally, the node partition force balance state data is output, and the data content includes the balance coefficient of each region and the force distribution uniformity index between the nodes. Based on the force balance state data of node partitions, the force transfer path between the region and the adjacent region is extracted by analyzing the balance state between regions. Along the path direction, the interpolation and extrapolation method is used to calculate the force offset value of the node under specific load conditions. In the specific operation, the force offset vector of the key node is obtained by the quadratic interpolation formula, and further based on the force difference of each node on the path and the partition balance state, the complete joint force offset value matrix is ​​iteratively calculated. The force offset value data not only describes the force transfer direction and amplitude between nodes, but also provides support for the subsequent mechanical performance evaluation. Using the node joint force offset value, the force change characteristics of different nodes are evaluated by constructing a force difference evaluation function. The force difference evaluation function comprehensively considers the force gradient change rate between joint nodes, the partition balance deviation, and the amplitude change of the node force offset value. In the calculation process, by traversing each node and its adjacent nodes, the consistency of its force direction and the volatility of the force offset value are analyzed, so as to obtain the degree of force imbalance between nodes. Finally, statistics are taken according to the curvature range and node distribution position to generate the joint lateral force difference data, which includes the force deviation value of each key node, the force difference map under the curvature distribution, and the average force fluctuation in the overall surface area.

[0074] Preferably, step S24 comprises the following steps:

[0075] Step S241: dividing the pressure difference distribution area based on the body shape dynamic structure support data and the bed surface material mechanical response mapping data to obtain the pressure difference distribution area data;

[0076] Step S242: performing deformation compression recovery rate analysis on the bed surface material data according to the pressure difference distribution area data to obtain bed surface deformation compression recovery rate data;

[0077] Step S243: performing posture change deformation recovery time series analysis on the bed deformation compression recovery rate data according to the pressure difference distribution area data to obtain posture change deformation recovery time series data;

[0078] Step S244: performing support force feedback evaluation based on the posture change deformation recovery time series data and the body dynamic structure support data to obtain support force feedback data;

[0079] Step S245: Design an adaptive adjustment mechanism according to the support force feedback data to obtain an adaptive adjustment mechanism for the smart bed.

[0080] In the embodiment of the present invention, the dynamic structure support data of the body shape and the mechanical response mapping data of the bed surface material are used, and the two are first aligned to the same three-dimensional space coordinate system through coordinate transformation. Bilinear interpolation calculation is applied to each data point to fill the gap between the data and ensure that the spatial distribution of the force response of the support surface and the material is consistent. Subsequently, the pressure difference matrix between the actual force distribution of the bed surface material and the theoretical force of the body shape support is extracted through difference calculation. In order to divide the pressure difference distribution area, a watershed algorithm based on the gradient direction is executed based on the difference matrix to calibrate the boundary of the area where the difference gradient changes significantly. The division results are classified according to the difference threshold level to generate a pressure distribution area map. The pressure difference distribution area data finally output includes a set of regional boundary points, a difference threshold classification, and a pressure distribution gradient change trend matrix. For the pressure difference distribution area data, the pressure center point and surrounding nodes in the area are selected, and the mechanical deformation experiment of the bed surface material of these nodes is performed. A high-precision displacement sensor is used to measure the deformation amount of the material under different pressures, and the compression rate and recovery rate of the material deformation are recorded in combination with the time dimension. During the experiment, the deformation curve of the material over time was measured and recorded point by point, and its initial recovery velocity and final recovery velocity were calculated. By fitting the nonlinear time series curve of the recovery process, the rebound time constant of the material after pressure release was extracted, and the deformation recovery ability of different regions was compared and analyzed. The experimental data were organized into bed deformation compression recovery rate data, including the recovery rate vector, rebound time constant distribution and deformation curve fitting parameters of each pressure area. Using the pressure difference distribution area data and bed deformation compression recovery rate data, the dynamic correlation between the deformation recovery rate of different pressure areas and the change of human posture was analyzed. First, the human posture switching process was simulated, and the start response time of bed deformation and the peak time of the recovery process were determined by analyzing the time series curve of the force change of each area of ​​the bed during posture switching. Using the piecewise linear interpolation method, the time series response data of the pressure area was mapped to the time axis of the overall deformation recovery process to form a dynamic time series of the overall deformation recovery of the bed. Combined with the joint force data in the recovery time series, the deformation coordination of each area of ​​the bed when the human body turned over or flipped was further analyzed. Finally, the posture change deformation recovery time series data is generated, including the recovery dynamic time series, the recovery time series peak of each posture switching node, and the regional deformation synchronization evaluation index. Combining the posture change deformation recovery time series data and the body dynamic structure support data, a bed support force feedback system is established. In the evaluation process, the support force data is first normalized to eliminate the influence of human body shape differences on the support force feedback. Then, based on the recovery time series data, the uniformity of the mechanical response of the bed support points under different postures is analyzed, and the time synchronization between the support force and the posture switching node is calculated. The feedback results are quantified, and multidimensional statistical analysis methods, including variance analysis and covariance analysis, are used to evaluate the dynamic change characteristics of the support force over time and the mechanical coupling relationship between different regions.Finally, the support force feedback data is generated, including the support point mechanical response matrix, the regional support uniformity distribution map and the time-series mechanical coupling coefficient table. According to the support force feedback data, an adaptive adjustment mechanism is formulated. First, the support point mechanical response matrix in the feedback data is analyzed to identify the areas with insufficient or excessive support force. Then, the gradient optimization algorithm is used to calculate the displacement adjustment amount of the bed surface adjustment device in each area to ensure that the support surface can match the changing needs of human body posture in future dynamic adjustments. For the over-large or insufficient support area, a multi-point displacement adjustment mechanism based on a stepper motor is designed to dynamically adjust the height and elastic parameters of the support surface. Combined with the timing characteristics of the feedback data, the trigger conditions and response priorities are set for the adjustment mechanism to ensure the real-time and accuracy of the adjustment process. The final output of the intelligent bed adaptive adjustment mechanism includes the working parameters of the adjustment mechanism, the support point optimization displacement vector and the dynamic adjustment control logic table.

[0081] Preferably, step S244 includes the following steps:

[0082] Perform regionalized deformation stress-time curve reconstruction on the posture change deformation recovery time series data to obtain regionalized stress-time characteristic data;

[0083] According to the regionalized stress-time characteristic data, dynamic partition support stability analysis is performed on the dynamic structural support data of the body shape to obtain dynamic partition support stability data;

[0084] The nonlinear mechanical coupling response analysis and processing is performed on the dynamic partition support stability data to obtain the support coupling force field data;

[0085] According to the support coupling force field data, the local stress peak distribution is analyzed and processed on the regionalized stress-time characteristic data to obtain the local stress peak distribution data;

[0086] Perform support force feedback weight distribution processing on the local stress peak distribution data to obtain support feedback weight data;

[0087] According to the support feedback weight data, support force feedback evaluation is performed on the dynamic partition support stability data and the body dynamic structure support data to obtain support force feedback data.

[0088] In an embodiment of the present invention, based on the posture change deformation recovery time series data, the data is first divided into regions, and the watershed segmentation method based on the force change gradient is used to calibrate the deformation boundary points in each region and their corresponding stress time series curves. The stress-time data of each region is smoothed using the cubic spline interpolation method to ensure the continuity and differentiability of the data. The data points in each region are fitted separately to reconstruct the regionalized deformation stress-time curve. During the reconstruction process, the boundary stress anomaly caused by the mechanical interaction between regions is corrected by linear superposition analysis. The regionalized stress-time characteristic data finally output includes the stress change curve equation of each region, the time response characteristic points and the overall deformation trend of the region after reconstruction. Combined with the regionalized stress-time characteristic data and the body dynamic structure support data, a mechanical stability analysis is applied to the support surface of each region. The finite difference method is used to calculate the spatial gradient of the support force and stress fluctuation in the region, and the dynamic stability index of the support surface is quantified. Subsequently, the Fourier transform is applied to decompose the regional stress time series characteristics to extract the low-frequency stability component and the high-frequency disturbance component. Through the anisotropic elastic modulus matrix of the support surface, the rate of disturbance propagation between regions is calculated, and the influence of regional support mechanical coupling on the overall stability is evaluated. The output dynamic partition support stability data includes stability time series curve, inter-region coupling fluctuation characteristic matrix and regional support stability classification evaluation table. Based on the dynamic partition support stability data, the support coupling characteristics of each region are simulated by nonlinear finite element method. First, by establishing a set of mechanical coupling equations, the support force of each region is associated with the disturbance of the adjacent region. Subsequently, the Runge-Kutta numerical solution is introduced to iteratively solve the force field distribution characteristics of the support coupling system at different time steps. During the analysis process, the response data of the coupling region is sampled in partitions, and the time distribution of the extreme value points of force field stability is analyzed by Laplace transform. Finally, the support coupling force field data is obtained, including the spatiotemporal distribution matrix of the inter-region coupling force field, the regional force field response characteristic function and the coupling influencing factor. Based on the support coupling force field data, the stress peak points of each region in the regionalized stress-time characteristic data are analyzed by constructing the stress gradient enhancement function. The polynomial fitting method is used to fit the local stress gradient changes in the region, and the weighted Lagrangian interpolation is applied to the area with abnormally high stress gradient to correct the stress offset phenomenon caused by force field coupling. By constructing a set of stress peak points in the region and calculating the distribution density of each peak point, a local stress peak distribution map is generated. The output local stress peak distribution data includes the spatial coordinates of the peak point, the stress peak time point, and the statistical results of the peak point density distribution in the region. According to the local stress peak distribution data, the dynamic partition weighting method is used to assign support force feedback weights to each region. First, the peak point density and time response frequency in the region are normalized to eliminate the influence of distribution differences.By introducing a weight function, the importance of the stress peak point in each area is associated with its support stability, so as to calculate the allocation priority of the support force feedback. Subsequently, the weight function parameters are adjusted using a multi-objective optimization algorithm so that the allocation result meets the design goal of the overall bed support balance. The support feedback weight data finally generated includes the feedback weight distribution vector in the region, the weight allocation priority list, and the overall bed support force optimization plan. The support feedback weight data, the dynamic partition support stability data, and the body dynamic structure support data are combined to evaluate the support force feedback. The support feedback error distribution map is constructed by calculating the deviation between the actual support force of each area and the theoretical support force after weight adjustment. The weighted average method is used to calculate the support force feedback balance index of the overall bed surface, and the time series analysis method is used to evaluate the dynamic stability of the support force feedback. The feedback results are summarized and summarized, and the support force feedback data is finally output, including the support feedback error matrix, the dynamic support force stability time series curve, and the support feedback balance evaluation report.

[0089] In this embodiment, reference Figure 3 The above is a schematic flow chart of the detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include:

[0090] Step S31: collecting adjustment simulation data of the smart bed adaptive adjustment mechanism to obtain adaptive adjustment simulation data;

[0091] Step S32: using a preset adjustment error identification model to collect adjustment errors from the adaptive adjustment simulation data to obtain adaptive adjustment error data;

[0092] Step S33: performing error parameter balance optimization on the adaptively adjusted error data to obtain error parameter balance optimization data;

[0093] Step S34: Optimize and adjust the smart bed self-adaptive adjustment mechanism based on the error parameter balance optimization data to obtain the smart bed self-adaptive adjustment optimization mechanism.

[0094] In the embodiment of the present invention, based on the adaptive adjustment mechanism of the smart bed, by establishing a real-time correspondence between the adjustment parameters and the deformation of the bed surface, the adjustment behavior under different user body shapes, postures and pressure distributions is simulated. First, in a laboratory environment, a representative pressure sensor matrix is ​​selected, and the sensor is embedded in the surface of the smart bed to record the changes in the force distribution of the bed surface during the adjustment process. By setting samples for different body shapes, weights and sleeping posture combinations, the adaptive adjustment operation is implemented one by one. The mechanical response data during the adjustment process is synchronously recorded using a mechanical monitor, including indicators such as support force changes, deformation amplitude and recovery time. The dynamic characteristics of the bed surface adjustment are recorded by high-speed video, and the deformation profile change characteristics are extracted by combining optical image analysis. The adaptive adjustment simulation data finally output includes a bed surface stress distribution diagram, a deformation dynamic curve and a pressure-adjustment relationship table. Based on the adaptive adjustment simulation data generated in step S31, the error recognition algorithm is used to analyze the adjustment error. By constructing a deviation calculation formula, the actual bed surface adjustment parameters are compared point by point with the theoretical design target to extract the error value. The error time series data is decomposed into frequency components using Fourier analysis, and the adjustment events corresponding to high-frequency errors and low-frequency errors are screened. The sources of errors are classified, and the errors caused by mechanical hysteresis, material elasticity differences and control delays are marked separately. Combined with the error spatial distribution analysis, the error points are mapped to the bed model in the form of a three-dimensional grid to generate an error distribution map. The final output of the adaptive adjustment error data includes an error statistics table, a frequency characteristic report and a three-dimensional visualization of the error distribution. Based on the adaptive adjustment error data generated in step S32, the error parameters are balanced by a multi-objective optimization method. By constructing the objective function, the influence of various types of errors is quantified, and the optimization constraints are set. The adjustment parameters are globally optimized using a genetic algorithm to gradually approach the parameter combination that minimizes the error. During the optimization process, the priorities of different error types are dynamically adjusted to ensure the diversity and adaptability of the optimization results. Combined with the linear programming method, the adjustment strategy of the regional error is further refined to ensure the uniformity and stability of the overall support effect of the bed surface. The final output of the error parameter balance optimization data includes the error balance optimization parameter set, the optimization iteration record and the error distribution map after optimization. According to the error parameter balance optimization data generated in step S33, the adaptive adjustment mechanism of the smart bed is optimized and adjusted. First, the optimization parameters are loaded into the adjustment control unit, and the deformation response of the bed surface is calibrated in real time by adjusting the hardware actuators (such as electric support rods or pneumatic adjustment modules). Subsequently, the changes in the adjustment behavior before and after the optimization are verified by experimental comparison, and the optimized mechanical response data and deformation characteristics are recorded. The stability and adaptability of the adjustment mechanism in different user scenarios are evaluated through three-dimensional parametric simulation of the optimization results. Combining mechanical simulation with dynamic image analysis, the performance improvement effect of the optimized adjustment mechanism is verified.The final output of the smart bed adaptive adjustment optimization mechanism includes an adjustment logic parameter table, an optimization performance evaluation report, and a post-optimization adjustment flow chart.

[0095] In this embodiment, reference Figure 4 The above is a flowchart of the detailed implementation steps of step S33. In this embodiment, the detailed implementation steps of step S33 include:

[0096] Step S331: extracting error parameters from the adaptive error adjustment data to obtain adaptive error adjustment parameters;

[0097] Step S332: performing parameter factor correlation analysis on the adaptive error adjustment data according to the adaptive error adjustment parameter to obtain parameter factor correlation data;

[0098] Step S333: performing multi-objective function fitting processing on the parameter factor correlation data to obtain error objective function parameter data;

[0099] Step S334: performing error parameter balance optimization on the adaptively adjusted error data according to the error objective function parameter data to obtain error parameter balance optimization data.

[0100] In an embodiment of the present invention, by analyzing the adaptive adjustment error data, a mathematical statistical method is used to extract key error parameters. First, the error data is divided into three categories: displacement error, time delay error, and pressure distribution error, and the noise data is eliminated by a hierarchical screening method. Subsequently, a deviation matrix generation method is used to archive each error type in the form of a multidimensional matrix, and the error parameters are partitioned and clustered. The main influencing factors in the error parameters are extracted by principal component analysis (PCA), and their contribution rate and sensitivity are calculated. The error parameters with high sensitivity are listed separately, and a numerical distribution diagram and a time series diagram are generated for each parameter to further refine the error change law. The adaptive error adjustment parameters finally output include an error parameter weight table, a key error indicator statistical report, and a three-dimensional surface diagram of the error distribution. Using the adaptive error adjustment parameters extracted in step S331, the error data is associated and modeled by a parameter factor association analysis method. First, according to the properties of different error parameters, a mathematical method for correlation analysis is selected, such as the Pearson correlation coefficient method for linear relationships and the Spearman rank correlation coefficient method for nonlinear relationships. By analyzing the error data pair by pair, the correlation strength between the error parameters is calculated, and a correlation matrix is ​​constructed. Combined with the factor analysis method, the coupling influence mechanism of each parameter on the adjustment result is identified, and the factor combination with high correlation is screened out. The heat map is used to display the strength distribution of factor associations, and the key interactive relationship between errors is clarified. The parameter factor association data finally output includes the association matrix, the factor sensitivity report and the error factor interaction diagram. Based on the parameter factor association data generated in step S332, the error optimization objective function is established by the multi-objective optimization method. First, the three objectives of minimizing errors, maximizing adjustment efficiency and enhancing system stability are selected, and the objective functions are constructed respectively. The numerical range of the error factor is fitted by piecewise linear regression, and the weight of the objective function is optimized to a dynamically adjustable parameter by the Lagrange multiplier method. The genetic algorithm is used to perform a global search on the objective function to generate the optimal solution set that meets the multi-objective optimization conditions. The accuracy of the objective function is verified by the goodness of fit test, and the function parameter curve graph and the error fitting surface graph are generated to intuitively display the results. The error objective function parameter data finally output includes the optimization objective function formula, the parameter optimization result table and the fitting verification report. The error objective function parameter data generated in step S333 is used to optimize the adaptive adjustment error data as a whole. First, the parameters of the error objective function are loaded into the optimization algorithm, and the initial conditions and boundary limits of the iterative optimization are set. The error distribution is globally balanced by the constrained optimization method, and the error parameter weights are dynamically adjusted to achieve multi-objective optimization. Combined with the dynamic programming method, the local extreme points of the error distribution are optimized to reduce the sudden errors of the system. Finally, the optimization results are evaluated through mechanical simulation tests, and a comparison chart of the error distribution before and after optimization and an optimization dynamic adjustment curve chart are generated.The final output error parameter balance optimization data includes error optimization result table, error balance distribution diagram and optimization test report.

[0101] Preferably, step S4 comprises the following steps:

[0102] Step S41: normalizing the smart bed self-adaptive adjustment optimization mechanism to obtain a smart bed self-adaptive adjustment normalization mechanism;

[0103] Step S42: constructing an adjustment simulation mechanism model for the smart bed adaptive adjustment normalization mechanism based on a policy gradient algorithm to obtain a smart bed adjustment simulation mechanism model;

[0104] Step S43: Send the smart bed adjustment simulation mechanism model to the smart bed control unit to perform parameterized simulation design of the smart bed.

[0105] In an embodiment of the present invention, each adjustment parameter in the adaptive adjustment optimization mechanism of the smart bed is normalized to ensure that all parameters are within the same dimension range, which is convenient for subsequent processing and analysis. The specific steps include: standardizing all original adjustment parameters to a unified numerical interval, such as [0,1], to avoid different scales from causing deviations to the final results. Based on the policy gradient algorithm, an adjustment simulation mechanism model of the adaptive adjustment normalization mechanism of the smart bed is constructed. First, the state space, action space and reward function in the adjustment process of the smart bed are defined. In this process, the state space describes the current parameter configuration of the smart bed and its adjustment status, the action space represents the adjustment actions that can be selected at each moment (such as bed frame angle, softness and hardness adjustment, etc.), and the reward function scores according to the performance of the smart bed in the adjustment process to evaluate the effect of each action. Specifically, the policy gradient algorithm in reinforcement learning is adopted, and the model is trained through interaction with the environment, so that the model can select the best adjustment action according to the current state, thereby optimizing the adjustment effect of the smart bed. When the algorithm is implemented, a parameterized policy network is first initialized, which accepts the current state information as input and outputs a series of adjustment actions and their corresponding probability distributions. Then, by sampling the output of the policy network, an action is selected, and a new state and reward are obtained after executing the action. Next, the policy network is updated by the gradient ascent method to ensure that the reward value can be maximized so that the model can gradually approach the optimal solution in each adjustment. This process is optimized through multiple iterations until the policy network can stably output an efficient adjustment strategy. Finally, the model is built, the intelligent bed adjustment simulation mechanism model is obtained, and its optimal adjustment strategy under different states is output. After the construction of the intelligent bed adjustment simulation mechanism model is completed in step S42, the next step is to send the model to the intelligent bed control unit to perform the intelligent bed parametric simulation design. First, the output data of the model (including the current state, action decision and reward function) is transmitted to the control unit through the communication interface of the intelligent bed. After the control unit receives the simulation mechanism model, it performs the corresponding intelligent bed parameter adjustment according to the decision rules in the model. In specific implementation, the control unit adjusts the parameters of the intelligent bed (such as bed shape, pressure distribution, hardness adjustment, etc.) in real time according to the optimal strategy in the adjustment model. In the actual execution process, the control unit continuously monitors the response of the bed, collects feedback data, and sends these data back to the model for real-time updating and optimization. This closed-loop feedback mechanism ensures that the smart bed maintains efficient and accurate adjustments in the parametric simulation design. In order to ensure the stability and reliability of the design process, the smart bed control unit also needs to have the ability to identify system failures or abnormal conditions. When the system cannot execute according to the simulation model, it can automatically trigger the fail-safe mode to prevent abnormal operations. Through this process, the smart bed can make adaptive dynamic adjustments based on the results of the three-dimensional parametric simulation design to ensure comfort and stability.

[0106] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0107] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A three-dimensional parametric simulation design method for a smart bed structure, characterized in that: The following steps are involved: Step S1: Obtaining the design structure and bed surface material data of the smart bed; A three-dimensional structural model is constructed based on the design structure of the smart bed to obtain a three-dimensional structural model of the smart bed; Step S2: acquiring body shape data of people of different body shapes; performing body shape dynamic structure support analysis on the three-dimensional structure model of the intelligent bed based on the body shape data of people of different body shapes, and obtaining body shape dynamic structure support data; Based on the body shape dynamic structure support data, an adaptive adjustment mechanism is designed to obtain an intelligent bed adaptive adjustment mechanism; wherein step S2 includes: Step S21: Acquire body shape data of people of different body shapes; Step S22: Based on the body shape data of human bodies of different body shapes, the three-dimensional structure model of the intelligent bed is subjected to body shape dynamic structure support analysis to obtain body shape dynamic structure support data. Step S22 is specifically as follows: Step S221: performing body curve analysis of different sleeping postures on the body shape data of people of different body shapes to obtain body shape curve data of different sleeping postures; Step S222: performing three-dimensional feature point extraction processing on the body curve data of different sleeping postures to obtain a set of coordinates of key nodes of the human body; Step S223: dynamically adjusting the body curve data of different sleeping postures according to the human body key node coordinate set to obtain dynamic support force curved surface adjustment data; wherein step S223 includes: Step S2231: performing sleep turning force distribution analysis on the key node coordinate set of the human body to obtain node sleep turning force distribution data; Step S2232: performing local curvature force interpolation processing of different sleeping postures on the body curve data of different sleeping postures according to the node sleep flipping force distribution data, so as to obtain local curvature force interpolation data of different sleeping postures; Step S2233: performing estimation of joint sideways force difference on the local curvature force interpolation data of different sleeping postures to obtain joint sideways force difference data; Step S2234: dynamically adjusting the force-bearing curved surface of the body curve data of different sleeping postures based on the local curvature force interpolation data of different sleeping postures and the joint side force difference data to obtain dynamic support force curved surface adjustment data; Step S224: performing regional difference analysis on the dynamic support force curved surface adjustment data to obtain regional difference data of the support force curved surface; Step S225: performing body shape dynamic structure support analysis on the three-dimensional structure model of the intelligent bed according to the dynamic support force curved surface adjustment data and the support force curved surface area difference data to obtain body shape dynamic structure support data; Step S23: performing mechanical response mapping analysis on the bed surface material data according to the body shape dynamic structure support data to obtain the bed surface material mechanical response mapping data; Step S24: Based on the body shape dynamic structure support data and the bed surface material mechanical response mapping data, an adaptive adjustment mechanism is designed to obtain an intelligent bed adaptive adjustment mechanism. Step S24 is specifically as follows: Step S241: dividing the pressure difference distribution area based on the body shape dynamic structure support data and the bed surface material mechanical response mapping data to obtain the pressure difference distribution area data; Step S242: performing deformation compression recovery rate analysis on the bed surface material data according to the pressure difference distribution area data to obtain bed surface deformation compression recovery rate data; Step S243: performing posture change deformation recovery time series analysis on the bed deformation compression recovery rate data according to the pressure difference distribution area data to obtain posture change deformation recovery time series data; Step S244: performing support force feedback evaluation based on the posture change deformation recovery time series data and the body dynamic structure support data to obtain support force feedback data; Step S245: designing an adaptive adjustment mechanism according to the support force feedback data to obtain an adaptive adjustment mechanism for the smart bed; Step S3: performing error parameter balance optimization on the smart bed adaptive adjustment mechanism to obtain error parameter balance optimization data; optimizing and adjusting the smart bed adaptive adjustment mechanism based on the error parameter balance optimization data to obtain the smart bed adaptive adjustment optimization mechanism; Step S4: construct an adjustment simulation mechanism model for the smart bed adaptive adjustment optimization mechanism to obtain the smart bed adjustment simulation mechanism model; send the smart bed adjustment simulation mechanism model to the smart bed control unit to execute the smart bed parametric simulation design.

2. The three-dimensional parametric simulation design method for the intelligent bed structure according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: Acquire the design structure and bed surface material data of the smart bed; Step S12: marking the structure connection points of the smart bed design structure to obtain the structure connection point data of the smart bed; Step S13: defining the adjustable motion trajectory range of the smart bed design structure according to the smart bed structure connection point data to obtain the adjustable motion trajectory range data; Step S14: construct a three-dimensional structure model based on the adjustable motion trajectory range data and the smart bed structure connection point data to obtain a three-dimensional structure model of the smart bed.

3. The three-dimensional parametric simulation design method for the intelligent bed structure according to claim 1 is characterized in that: Step S2233 includes the following steps: The force gradient change between nodes is calculated for the local curvature force interpolation data of different sleeping postures to obtain the force gradient change data between nodes; According to the force gradient change data between nodes, the partition equilibrium state analysis between nodes is performed to obtain the force equilibrium state data of the node partition; Based on the node partition force balance state data, the node joint force offset value is calculated for the force gradient change data between nodes to obtain the node joint force offset value; The joint lateral force difference is estimated according to the node joint force offset value to obtain the joint lateral force difference data.

4. The three-dimensional parametric simulation design method for the intelligent bed structure according to claim 1, characterized in that: Step S244 includes the following steps: Perform regionalized deformation stress-time curve reconstruction on the posture change deformation recovery time series data to obtain regionalized stress-time characteristic data; According to the regionalized stress-time characteristic data, dynamic partition support stability analysis is performed on the dynamic structural support data of the body shape to obtain dynamic partition support stability data; The nonlinear mechanical coupling response analysis and processing is performed on the dynamic partition support stability data to obtain the support coupling force field data; According to the support coupling force field data, the local stress peak distribution is analyzed and processed on the regionalized stress-time characteristic data to obtain the local stress peak distribution data; Perform support force feedback weight distribution processing on the local stress peak distribution data to obtain support feedback weight data; According to the support feedback weight data, support force feedback evaluation is performed on the dynamic partition support stability data and the body dynamic structure support data to obtain support force feedback data.

5. The three-dimensional parametric simulation design method for the intelligent bed structure according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: collecting adjustment simulation data of the smart bed adaptive adjustment mechanism to obtain adaptive adjustment simulation data; Step S32: using a preset adjustment error identification model to collect adjustment errors from the adaptive adjustment simulation data to obtain adaptive adjustment error data; Step S33: performing error parameter balance optimization on the adaptively adjusted error data to obtain error parameter balance optimization data; Step S34: Optimize and adjust the smart bed self-adaptive adjustment mechanism based on the error parameter balance optimization data to obtain the smart bed self-adaptive adjustment optimization mechanism.

6. The three-dimensional parametric simulation design method for the intelligent bed structure according to claim 5, characterized in that: Step S33 includes the following steps: Step S331: extracting error parameters from the adaptive error adjustment data to obtain adaptive error adjustment parameters; Step S332: performing parameter factor correlation analysis on the adaptive error adjustment data according to the adaptive error adjustment parameter to obtain parameter factor correlation data; Step S333: performing multi-objective function fitting processing on the parameter factor correlation data to obtain error objective function parameter data; Step S334: performing error parameter balance optimization on the adaptively adjusted error data according to the error objective function parameter data to obtain error parameter balance optimization data.

7. The three-dimensional parametric simulation design method for the intelligent bed structure according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: normalizing the smart bed self-adaptive adjustment optimization mechanism to obtain a smart bed self-adaptive adjustment normalization mechanism; Step S42: constructing an adjustment simulation mechanism model for the smart bed adaptive adjustment normalization mechanism based on a policy gradient algorithm to obtain a smart bed adjustment simulation mechanism model; Step S43: Send the smart bed adjustment simulation mechanism model to the smart bed control unit to perform parameterized simulation design of the smart bed.

Images