Intelligent material selection method for fitness equipment preparation

Abstract

The present invention provides a method for intelligent material optimization for the preparation of fitness equipment, which relates to the field of data processing technology. By performing usage fitting according to user posture information and customized equipment type, a force data set of equipment and a contact data set of equipment are obtained, and the use environment information and the force data set of equipment are synchronized to the hard material analysis network analysis to obtain preferred coating material information and preferred force material information, and soft material optimization analysis is performed according to the equipment contact data set to obtain preferred soft material information, and customized fitness equipment production is performed according to the preferred material information. The technical problem that the customized fitness equipment actually produced by the prior art has a low degree of fit with users and a long fitness equipment customization cycle is solved. The technical effect of shortening the material selection cycle of customized fitness equipment and improving the technical effect of selecting materials for customized fitness equipment and the degree of fit of customized users is achieved.

Description

Technical Field

[0001] The invention relates to the technical field of data processing, and in particular to an intelligent material optimization method for preparing fitness equipment. Background Art

[0002] With the continuous development of the fitness industry, users' demand for fitness equipment has become more and more diversified. Therefore, the market for customized fitness equipment that can better meet users' different fitness needs and preferences has developed rapidly.

[0003] However, in the current preparation process of customized fitness equipment, the selection of raw materials for fitness equipment production is based on manual experience. The selection of raw materials based on manual experience has the defects of long material selection time, high subjectivity and prolonged customization cycle. At the same time, there is also the risk of lower compatibility between the customized fitness equipment actually produced and the users.

[0004] In summary, in the prior art, the selection of materials for customizing fitness equipment is relatively subjective and the material selection process is time-consuming, resulting in technical problems such as low adaptability of the customized fitness equipment actually produced to the user and a long fitness equipment customization cycle. Summary of the invention

[0005] The present application provides an intelligent material optimization method for the preparation of fitness equipment, which is used to solve the technical problems in the prior art that the selection of preparation materials when customizing fitness equipment is relatively subjective and the material selection process is time-consuming, resulting in a low degree of compatibility between the customized fitness equipment actually produced and the user and a long fitness equipment customization cycle.

[0006] In view of the above problems, the present application provides a material intelligent optimization method for preparing fitness equipment.

[0007] A material intelligent optimization method for preparing fitness equipment, the method comprising:

[0008] Interactively obtain customized input information of the customized user, wherein the customized input information includes user body information and customized equipment type; perform size specification fitting according to the user body information and the customized equipment type to obtain customized equipment specification information; perform modeling and restoration according to the customized equipment specification information to obtain an equipment dynamics model, and synchronize the user body information to the equipment dynamics model for use fitting to obtain an equipment force data set and an equipment contact data set; pre-construct a hard material analysis network, wherein the hard material analysis network includes a coating material analysis branch and a force material analysis branch; interactively obtain the use environment information of the customized user, and synchronize the use environment information and the equipment force data set to the coating material analysis branch and the force material analysis branch of the hard material analysis network respectively to obtain preferred coating material information and preferred force material information; perform soft material optimization analysis according to the equipment contact data set to obtain preferred soft material information; use the preferred coating material information, preferred force material information and preferred soft material information as material constraints, and use the customized equipment specification information as structural constraints to execute non-standardized production of customized fitness equipment.

[0009] The technical solution provided in this application has at least the following technical effects or advantages:

[0010] The method provided in the embodiment of the present application obtains the customized input information of the customized user through interaction, wherein the customized input information includes the user's body shape information and the customized equipment type; performs size specification fitting according to the user's body shape information and the customized equipment type to obtain the customized equipment specification information; performs modeling and restoration according to the customized equipment specification information to obtain the equipment dynamics model, and synchronizes the user's body shape information to the equipment dynamics model for use fitting to obtain the equipment force data set and the equipment contact data set; pre-constructs a hard material analysis network, wherein the hard material analysis network includes a coating material analysis branch and a force material analysis branch; interactively obtains the use environment information of the customized user, and synchronizes the use environment information and the equipment force data set to the coating material analysis branch and the force material analysis branch of the hard material analysis network respectively, to obtain preferred coating material information and preferred force material information; performs soft material optimization analysis according to the equipment contact data set to obtain preferred soft material information; uses the preferred coating material information, preferred force material information and preferred soft material information as material constraints, and uses the customized equipment specification information as structural constraints to perform non-standardized production of customized fitness equipment. The technical effect of shortening the material selection cycle of customized fitness equipment, improving the selected materials of customized fitness equipment and the adaptability of customized users is achieved, and indirectly achieving the technical effect of improving the experience of customized users when using customized fitness equipment for exercise. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A schematic diagram of the process flow of the intelligent material selection method for preparing fitness equipment provided in this application;

[0012] Figure 2 A schematic diagram of a process for obtaining preferred soft material information in the intelligent material selection method for preparing fitness equipment provided in this application. DETAILED DESCRIPTION

[0013] The present application provides a material intelligent optimization method for preparing fitness equipment, which is used to solve the technical problems in the prior art that the selection of materials for preparing customized fitness equipment is relatively subjective and the material selection process is time-consuming, resulting in low adaptability of the customized fitness equipment actually produced to the user and a long fitness equipment customization cycle. The technical effect of shortening the material selection cycle of customized fitness equipment, improving the adaptability of the selected materials of customized fitness equipment and customized users, and indirectly achieving the technical effect of improving the experience of customized users when exercising with customized fitness equipment.

[0014] Below, the technical solutions in the present invention will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments described herein. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. It should also be noted that, for the convenience of description, only the parts related to the present invention are shown in the accompanying drawings, rather than all of them.

[0015] like Figure 1 As shown, the present application provides a material intelligent optimization method for preparing fitness equipment, the method comprising:

[0016] A100: interactively obtaining customized input information of a customized user, wherein the customized input information includes user body information and customized equipment type;

[0017] Specifically, in this embodiment, the customized user is a customer who plans to customize fitness equipment for private use, and the customized equipment type is the type of fitness equipment planned to be customized, including but not limited to arm pullers, rowing machines, and core training equipment.

[0018] The user's body information includes a user's height parameter, a user's weight parameter, and a user's body proportion. In this embodiment, the customized input information is obtained for reference in analyzing and determining equipment specification parameters of the customized fitness equipment type.

[0019] A200: Performing size fitting according to the user's body shape information and the type of customized equipment to obtain customized equipment specification information;

[0020] In one embodiment, the size specification is fitted according to the user's body shape information and the customized equipment type to obtain customized equipment specification information. The method step A200 provided in the present application further includes:

[0021] A210: The user's body shape information includes a user's height parameter, a user's weight parameter, and a user's body proportion;

[0022] A220: interactively obtaining multiple sample body shape information and multiple sample fitness equipment specifications, wherein each sample body shape information includes a sample height parameter, a sample weight parameter, and a sample body proportion;

[0023] A230: reorganizing the plurality of sample body shape information and the plurality of sample fitness equipment specifications to obtain first branch training data, second branch training data and third branch training data, wherein the first branch training data includes a plurality of sample height parameters, a plurality of sample weight parameters and the plurality of sample fitness equipment specifications, the second branch training data includes a plurality of sample weight parameters, a plurality of sample body shape proportions and the plurality of sample fitness equipment specifications, and the third branch training data includes a plurality of sample height parameters, a plurality of sample body shape proportions and the plurality of sample fitness equipment specifications;

[0024] A240: constructing a standard specification fitting branch based on a convolutional neural network, and using the first branch training data, the second branch training data, and the third branch training data to train the standard specification fitting branch to obtain a first specification fitting branch, a second specification fitting branch, and a third specification fitting branch;

[0025] A250: Obtaining standard equipment according to the customized equipment type interaction, and constructing a size specification fusion layer based on standard specification information of the standard equipment;

[0026] A260: The first specification fitting branch, the second specification fitting branch and the third specification fitting branch are arranged in parallel, and the output ends of the first specification fitting branch, the second specification fitting branch and the third specification fitting branch are connected to the input end of the size specification fusion layer to complete the construction of the size specification fitting network;

[0027] A270: Synchronize the user's body shape information to the size specification fitting network to perform size specification fitting to obtain the customized equipment specification information.

[0028] In one embodiment, a standard specification fitting branch is constructed based on a convolutional neural network, and the first branch training data, the second branch training data, and the third branch training data are used to train the standard specification fitting branch to obtain a first specification fitting branch, a second specification fitting branch, and a third specification fitting branch. Step A240 of the method provided in the present application also includes:

[0029] A241: pre-constructing a first parameter convolution training layer and a second parameter convolution training layer, wherein the first parameter convolution training layer includes multiple layers of first parameter kernel functions, and the second parameter convolution training layer includes a second parameter kernel function;

[0030] A242: in the construction of the first specification fitting branch, the multi-layer first parameter kernel function of the first parameter convolution training layer is used to extract features of the multiple sample height parameters and the multiple sample weight parameters to obtain multi-layer body parameter convolution output samples;

[0031] A243: the second parameter kernel function of the second parameter convolution training layer is used to extract features of the multiple sample fitness equipment specifications to obtain multiple specification parameter convolution output samples;

[0032] A244: constructing and training a specification fitting function using the multi-layer body parameter convolution output samples and the plurality of specification parameter convolution output samples as training data until convergence and outputting a first specification fitting function;

[0033] A245: Synchronize the first specification fitting function to the first specification fitting branch.

[0034] Specifically, in this embodiment, the user body information includes a user height parameter, a user weight parameter and a user body proportion, and the user body proportion is specifically arm length: leg length: torso length.

[0035] Interactively obtain multiple sample body shape information and multiple sample fitness equipment specifications, wherein each sample body shape information includes a sample height parameter, a sample weight parameter and a sample body shape ratio, and reorganize the multiple sample body shape information and multiple sample fitness equipment specifications to obtain first branch training data, second branch training data and third branch training data, wherein the first branch training data includes multiple sample height parameters, multiple sample weight parameters and the multiple sample fitness equipment specifications, the second branch training data includes multiple sample weight parameters, multiple sample body shape ratios and the multiple sample fitness equipment specifications, and the third branch training data includes multiple sample height parameters, multiple sample body shape ratios and the multiple sample fitness equipment specifications.

[0036] The specific method of constructing a standard specification fitting branch based on a convolutional neural network and using the first branch training data as training data to generate the first specification fitting branch that can output an adjustment ratio for adjusting the specifications of standard fitness equipment according to a user height parameter and a user weight parameter is as follows:

[0037] A first parameter convolution training layer and a second parameter convolution training layer are pre-constructed, wherein the first parameter convolution training layer includes multiple layers of first parameter kernel functions, the second parameter convolution training layer includes a second parameter kernel function, and the first parameter convolution training layer and the second parameter convolution training layer are used to analyze the functional relationship between user height-user weight and the adjustment ratio of fitness equipment specifications.

[0038] In constructing the first specification fitting branch, the multi-layer first parameter kernel function included in the first parameter convolution training layer is used to extract features of the multiple sample height parameters and the multiple sample weight parameters to obtain multi-layer body parameter convolution output samples.

[0039] The second parameter kernel function of the second parameter convolution training layer is used to extract features of the multiple sample fitness equipment specifications to obtain multiple specification parameter convolution output samples.

[0040] The multi-layer body parameter convolution output samples and the multiple specification parameter convolution output samples are used to construct a fitting function, and the purpose of the fitting function is to quantify the numerical change relationship between the user's height-user weight and the fitness equipment specification adjustment ratio.

[0041] The multi-layer posture parameter convolution output samples and the multiple specification parameter convolution output samples are used as training data, and based on the existing data rule learning method (such as machine learning), functional relationship fitting training of the numerical change relationship between user height-user weight and the specification adjustment ratio of fitness equipment is performed until it is adjusted to the optimal state, and a first specification fitting function is output, and the first specification fitting function is synchronized to the first specification fitting branch.

[0042] By analogy, the second branch training data and the third branch training data are used to train the standard specification fitting branch to obtain the second specification fitting branch and the third specification fitting branch.

[0043] This embodiment extracts features from body parameters and fitness equipment specifications, and establishes a numerical change relationship model between the two through a machine learning method, thereby quickly giving a scientific fitness equipment specification adjustment amount when the user's body parameters are obtained.

[0044] Furthermore, this embodiment obtains universal standard equipment according to the interaction of the customized equipment type, and constructs a size specification fusion layer based on the standard specification information of the standard equipment. The standard specification information includes height information (seat height, handle height), width information (seat width, handle spacing), and angle information (seat tilt angle, pedal rotation angle) of the standard equipment.

[0045] The first specification fitting branch, the second specification fitting branch and the third specification fitting branch are arranged in parallel, and the output ends of the first specification fitting branch, the second specification fitting branch and the third specification fitting branch are connected to the input end of the size specification fusion layer to complete the construction of the size specification fitting network.

[0046] The same method as that for reorganizing the data of the multiple sample body shape information and the multiple sample fitness equipment specifications is adopted to reorganize the user height parameters, user weight parameters and user body proportions constituting the user body shape information to obtain the first branch input data, the second branch input data and the third branch input data.

[0047] The first branch input data, the second branch input data and the third branch input data are mapped and synchronized to the first specification fitting branch, the second specification fitting branch and the third specification fitting branch of the size specification fitting network to calculate the specification adjustment ratio of the fitness equipment, and obtain the first specification adjustment ratio, the second specification adjustment ratio and the third specification adjustment ratio. It should be understood that the adjustment objects of the first specification adjustment ratio, the second specification adjustment ratio and the third specification adjustment ratio are different, including but not limited to the seat inclination angle, the pedal inclination angle, and the handle spacing.

[0048] In the size specification fusion layer, the standard specification information is taken as the adjustment object, and the first specification adjustment ratio, the second specification adjustment ratio and the third specification adjustment ratio are used to perform mapping adjustment on the size specification ratio of the standard specification information to obtain the customized equipment specification information adapted to the user's body shape.

[0049] This embodiment achieves the technical effect of scientifically analyzing and adjusting various parameters (such as height, angle, and width ratio) of fitness equipment based on the user's body shape data by constructing the size specification fitting network to make the fitness equipment more suitable for the user's body shape characteristics.

[0050] A300: Modeling and restoring the customized equipment specification information to obtain an equipment dynamics model, and synchronizing the user's body information to the equipment dynamics model for use fitting to obtain an equipment force data set and an equipment contact data set;

[0051] In one embodiment, modeling and restoration are performed according to the customized equipment specification information to obtain an equipment dynamics model, and the user body information is synchronized to the equipment dynamics model for use fitting to obtain an equipment force data set and an equipment contact data set. Step A300 of the method provided in the present application also includes:

[0052] A310: Extracting features from the customized equipment specification information to obtain structural design features and connection design features;

[0053] A320: generating a rigid body model of the equipment according to the structural design features, and simulating the motion of the rigid body model of the equipment with the connection design features as constraints to generate a dynamic model of the equipment;

[0054] A330: interactively obtain the motion force characteristics of the standard equipment, and synchronize the user's body information and the motion force characteristics to the equipment dynamics model for use fitting to obtain a simulation dynamics model;

[0055] A340: performing force separation on the simulation dynamics model according to the structural design features to obtain force data of multiple components, wherein the force data of the multiple components constitute the equipment force data set;

[0056] A350: Traverse the simulation dynamics model to locate the motion action nodes and obtain multiple contact nodes, which constitute the equipment contact data set.

[0057] Specifically, in this embodiment, the three-dimensional model of the standard equipment is interactively obtained from the fitness equipment manufacturer, and local parameters of the obtained three-dimensional modeling are adjusted based on the customized equipment specification information to obtain the customized equipment three-dimensional model. For example, if the customized equipment specification information is the specification information for adjusting the seat height and angle ratio, the seat-related part of the three-dimensional model is correspondingly modified to meet the customization requirements.

[0058] Furthermore, feature extraction is performed on the three-dimensional model of the customized equipment to obtain structural design features and connection design features. The structural design features are the overall shape of the customized equipment and the multiple supporting structural components that constitute the three-dimensional model of the customized equipment. The connection design features are the connection relationship between the multiple supporting structural components to form the overall shape of the customized equipment.

[0059] The existing rigid body modeling technology is used to perform modeling with the structural design features as modeling metadata to generate a rigid body model of the equipment, and the motion effect of the rigid body model of the equipment is simulated with the connection design features as constraints to generate the equipment dynamics model. The equipment motion model can simulate the force characteristics of the user of the customized equipment during exercise.

[0060] This embodiment interactively obtains the motion force characteristics of a user of a specific weight when correctly using the standard specification equipment for exercise, and the motion force characteristics are the force action position of the standard specification equipment and the pressure value at the force action position.

[0061] The weight information of the user's body information and the weight information of a user with a specific weight are used to calculate the weight deviation, and based on the weight deviation calculation result mapping, the pressure value mapping calculation at the force acting position in the motion force feature is performed to obtain an optimized motion force feature.

[0062] According to the specification ratio deviation between the standard specification information and the customized equipment specification information, the force action position of the motion force feature is mapped proportionally and adjusted, and according to the optimized force action position, multiple optimized pressure values ​​in the optimized motion force feature are mapped and synchronized to the equipment dynamics model for use fitting to obtain a simulation dynamics model. The simulation dynamics model reflects the force conditions of each component of the customized equipment when the customized user uses the customized equipment.

[0063] The simulation dynamics model is subjected to force decomposition according to the structural design features to obtain force data of multiple components of multiple supporting structure components, and the force data of the multiple components constitute the equipment force data set.

[0064] According to the specification ratio deviation between the standard specification information and the customized equipment specification information, the force action position of the motion force feature is mapped and proportionally adjusted to obtain a plurality of conceptual contact sites, and the plurality of conceptual contact sites are used to traverse the simulation dynamics model to locate the motion action nodes, to obtain a plurality of contact nodes that the customized user may touch during the use of the customized fitness equipment, and the plurality of contact nodes constitute the equipment contact data set.

[0065] This embodiment achieves the technical effect of providing a reference for subsequent selection of hard materials and soft materials for fitness equipment by constructing a system that can fit and customize the force data of fitness equipment during use by users and the possible touch positions of fitness equipment during use by users.

[0066] A400: pre-constructing a hard material analysis network, wherein the hard material analysis network includes a coating material analysis branch and a stress-bearing material analysis branch;

[0067] In one embodiment, the method further comprises:

[0068] A410: interactively obtaining a plurality of sample environment information and a plurality of sample coating materials, and performing associated storage of the plurality of sample environment information and the plurality of sample coating materials based on a knowledge graph to generate a sample coating information storage layer;

[0069] A420: Pre-construct an environment similarity calculation function, and synchronize the environment similarity calculation function to the pre-constructed environment matching analysis layer. The environment similarity calculation function is as follows:

[0070] ;

[0071] Wherein, S is the environment similarity, and the usage environment information includes i usage environment parameters, is the kth usage environment parameter in the usage environment information, the sample environment information includes i sample environment parameters, is the kth sample environment parameter in the sample environment information;

[0072] A430: pre-constructing an environmental parameter discriminator and a similarity sorting engine, synchronizing the environmental parameter discriminator to the sample coating information storage layer, synchronizing the similarity sorting engine to the environmental matching analysis layer, connecting the output end of the sample coating information storage layer and the input end of the environmental matching analysis layer, and generating the coating material analysis branch;

[0073] A440: interactively obtain a performance deviation threshold and a performance deviation vector, and construct a performance deviation constraint layer based on the performance deviation threshold and the performance deviation vector;

[0074] A450: interactively obtaining multiple sample force performance parameters of multiple sample hard materials, and performing associated storage of the multiple sample hard materials and the multiple sample force performance parameters based on the knowledge graph to generate a sample performance storage layer;

[0075] A460: connecting the output end of the performance deviation constraint layer and the input end of the sample performance storage layer to complete the construction of the stress material analysis branch;

[0076] A470: The coating material analysis branch and the stress-bearing material analysis branch are arranged in parallel and the output end and the input end are adapted to generate the hard material analysis network.

[0077] Specifically, it should be understood that the multiple supporting structure components need to have load-bearing performance and corrosion resistance. The load-bearing performance depends on the hard material used to process the multiple supporting structure components, and the corrosion resistance depends on the anti-corrosion material coated on the surface of the hard material.

[0078] Based on this, this embodiment pre-constructs a hard material analysis network for analyzing and determining the preparation materials (hard materials and coating materials) of multiple supporting structure components for bearing loads of customized fitness equipment, wherein the hard material analysis network includes a coating material analysis branch for analyzing the surface coating materials of hard materials that resist corrosion from the external environment, and a hard material analysis branch for analyzing and determining multiple supporting structure components that are adapted to customized user forces.

[0079] The construction process of the coating material analysis branch is as follows:

[0080] Interactively obtain multiple sample environment information and multiple sample coating materials, and associate and store the multiple sample environment information and multiple sample coating materials based on the knowledge graph to generate a sample coating information storage layer.

[0081] A pre-built environment similarity calculation function is synchronized to the pre-built environment matching analysis layer. The environment similarity calculation function is as follows:

[0082] ;

[0083] Wherein, S is the environment similarity, and the usage environment information includes i usage environment parameters, is the kth usage environment parameter in the usage environment information, the sample environment information includes i sample environment parameters, is the kth sample environment parameter in the sample environment information;

[0084] The environment parameter discriminator and the similarity ranking engine are pre-built. The use method of the environment parameter discriminator and the similarity ranking engine is described in detail in the subsequent description of this embodiment.

[0085] The environmental parameter discriminator is synchronized to the sample coating information storage layer, the similarity sorting engine is synchronized to the environmental matching analysis layer, the output end of the sample coating information storage layer and the input end of the environmental matching analysis layer are connected to generate the coating material analysis branch.

[0086] The construction method of the stress material analysis branch is as follows:

[0087] The performance deviation threshold characterizing the allowable deviation range of the force performance and the performance deviation vector characterizing the allowable deviation direction of the force performance deviation are interactively obtained, and a performance deviation constraint layer for force performance deviation analysis is constructed based on the performance deviation threshold and the performance deviation vector.

[0088] Exemplarily, the performance deviation threshold is 200N, and the performance deviation vector is greater than a preset force parameter, that is, the actually obtained force performance is within 200N greater than the preset force parameter.

[0089] Interactively obtain multiple sample force performance parameters of multiple sample hard materials, and associate and store the multiple sample hard materials and multiple sample force performance parameters based on the knowledge graph to generate a sample performance storage layer; connect the output end of the performance deviation constraint layer and the input end of the sample performance storage layer to complete the construction of the force material analysis branch.

[0090] The coating material analysis branch and the stress-bearing material analysis branch are arranged in parallel and the output end and the input end are adapted to generate the hard material analysis network.

[0091] This embodiment constructs a dual-channel hard material analysis network to achieve simultaneous analysis and acquisition of hard materials and hard material surface coating materials, thereby shortening the material optimization cycle.

[0092] A500: interactively obtaining the usage environment information of the customized user, and synchronizing the usage environment information and the equipment force data set to the coating material analysis branch and the force material analysis branch of the hard material analysis network respectively, to obtain preferred coating material information and preferred force material information;

[0093] In one embodiment, the use environment information of the customized user is interactively obtained, and the use environment information and the equipment force data set are synchronized to the coating material analysis branch and the force material analysis branch of the hard material analysis network respectively, to obtain the preferred coating material information and the preferred force material information. The method step A500 provided in the present application also includes:

[0094] A510: In the environment matching analysis layer of the coating material analysis branch, the environment parameter discriminator constructs an environment parameter discrimination constraint based on the use environment information, and traverses the multiple sample environment information based on the environment parameter discrimination constraint to obtain multiple screening environment information and multiple screening coating materials corresponding to the multiple screening environment information;

[0095] A520: In the environment matching analysis layer of the coating material analysis branch, multiple environment similarities of the multiple screening environment information and the use environment information are calculated one by one based on the environment similarity calculation function, the multiple environment similarities are serialized based on the similarity sorting engine, and the preferred coating material information is obtained in the multiple screening coating material mapping calls based on the extreme value of the serialization result;

[0096] A530: generating a set of equipment stress threshold values ​​according to the performance deviation threshold and the equipment stress data set in the performance deviation constraint layer of the stress-bearing material analysis branch;

[0097] A540: In the sample performance storage layer of the stress-bearing material analysis branch, the performance deviation vector is used as a comparison constraint, and the equipment stress threshold set is used to traverse the multiple sample stress performance parameters to obtain multiple target hard materials. The multiple target hard materials constitute the preferred stress-bearing material information.

[0098] Specifically, in this embodiment, the usage environment information of the customized equipment used by the customized user is interactively obtained, and the usage environment information and the equipment force data set are synchronized to the coating material analysis branch and the force material analysis branch of the hard material analysis network respectively.

[0099] In the environmental matching analysis layer of the coating material analysis branch, the environmental parameter discriminator constructs an environmental parameter discrimination constraint based on the usage environment information, and traverses the multiple sample environmental information based on the environmental parameter discrimination constraint to obtain multiple screening environmental information, and each environmental information indicator of the multiple screening environmental information is inferior to the environmental information indicator corresponding to the usage environment information.

[0100] And according to the plurality of screening environment information obtained through screening, a plurality of screening coating materials are mapped and called in the plurality of sample coating materials.

[0101] In the environmental matching analysis layer of the coating material analysis branch, multiple environmental similarities of the multiple screening environment information and the usage environment information are calculated one by one based on the environmental similarity calculation function, the multiple environmental similarities are serialized based on the similarity sorting engine, and based on the serialization result, the screening environment information corresponding to the sequence extreme value is selected and called from the multiple screening coating materials as the preferred coating material information that can resist the environmental corrosion of the usage environment information.

[0102] In the performance deviation constraint layer of the stress material analysis branch, actual force deviation thresholds are calculated for multiple component stress data of the equipment stress data set according to the performance deviation threshold, so as to generate the equipment stress threshold set including multiple component stress thresholds.

[0103] In the sample performance storage layer of the stress-bearing material analysis branch, the performance deviation vector is used as a comparison constraint and the equipment stress threshold set is used to traverse the multiple sample stress performance parameters to obtain multiple target hard materials. The multiple target stress performance parameters corresponding to the multiple target hard materials all fall within the stress thresholds of the multiple components and are greater than the stress data of the multiple components. The multiple target hard materials constitute the preferred stress-bearing material information.

[0104] This embodiment is based on the dual-channel hard material analysis network to simultaneously analyze and obtain hard materials and hard material surface coating materials, thereby achieving the technical effect of quickly obtaining hard materials that can withstand the force applied by customized users when using fitness equipment and coating materials that can resist the corrosion of the customized user's use environment, thereby shortening the optimization cycle of fitness equipment preparation materials and improving the scientific nature of the selection of fitness equipment preparation materials.

[0105] A600: performing a soft material optimization analysis according to the equipment contact data set to obtain information on the preferred soft material;

[0106] In one embodiment, Figure 2 As shown, the soft material optimization analysis is performed according to the equipment contact data set to obtain the preferred soft material information. The method step A600 provided in the present application also includes:

[0107] A610: interactively obtaining the soft material comfort pre-entered by the customization user, and using the soft material comfort as a constraint to obtain multiple sample material attribute information of multiple sample soft materials, wherein each sample material attribute information includes a sample comfort index, a sample service life index, and a sample replacement cost index;

[0108] A620: Preset a comfort weight, a life weight and a replacement weight, and perform fitness calculation on the plurality of sample material attribute information based on the comfort weight, the life weight and the replacement weight to obtain fitness of the plurality of sample materials;

[0109] A630: Serialize the fitness of the plurality of sample materials and call the sample soft material corresponding to the extreme value of the fitness of the sample material as the preferred soft material information.

[0110] Specifically, the soft material is a layer of soft material covering specific parts of the fitness equipment, such as the grip or the surface that may come into contact with the user's body. The soft material is used to improve the safety of the fitness equipment, prevent the user from being injured due to accidental collision during use, and provide the fitness user with a softer and more comfortable gripping feel when the fitness user grips the grip of the fitness equipment.

[0111] Since soft materials are prone to aging and are subject to wear and tear during use by fitness users, resulting in performance degradation of the soft materials, the present embodiment uses the following method to optimize the soft materials.

[0112] Specifically, this embodiment interactively obtains the soft material comfort pre-entered by the customization user, and uses the soft material comfort as a constraint to obtain multiple sample material attribute information of multiple sample soft materials, wherein each sample material attribute information includes a sample comfort index, a sample service life index and a sample replacement cost index, and each sample comfort index is higher than the soft material comfort pre-entered by the customization user.

[0113] It should be understood that this embodiment quantifies the softness and hardness parameters of soft materials under the same force based on the expert evaluation method, obtains the sample comfort index, and quantifies the softness and hardness parameters pre-entered by the customization user as the soft material comfort.

[0114] This implementation presets comfort weights, life weights and replacement weights, and performs fitness calculations on the multiple sample material attribute information based on the comfort weights, life weights and replacement weights to obtain multiple sample material fitnesses, and then serializes the multiple sample material fitnesses and calls the sample soft materials corresponding to the extreme values ​​of the sample material fitnesses as the preferred soft material information.

[0115] This embodiment achieves the technical effect of quickly matching and obtaining the preferred soft material that meets the user's requirements for the comfort of the soft material and balances the replacement cost and service life of the soft material.

[0116] A700: Using the preferred coating material information, preferred stress-bearing material information and preferred soft material information as material constraints and the customized equipment specification information as structural constraints, perform non-standard production of customized fitness equipment.

[0117] Specifically, in this embodiment, the preferred coating material information, preferred force-bearing material information and preferred soft material information are used as material constraints, and the customized equipment specification information is used as structural constraints to perform non-standard production of customized fitness equipment to obtain customized fitness equipment suitable for the customized user.

[0118] This embodiment achieves the technical effect of shortening the material selection cycle of customized fitness equipment, improving the selected materials of customized fitness equipment and the adaptability of customized users, and indirectly achieves the technical effect of improving the experience of customized users when using customized fitness equipment for exercise.

[0119] Any of the methods or steps described above may be stored as computer instructions or programs in various types of computer memories, and the computer instructions or programs may be recognized by various types of computer processors to implement any of the methods or steps described above.

[0120] Based on the above specific embodiments of the present invention, any improvements and modifications made to the present invention by those skilled in the art without departing from the principles of the present invention shall fall within the patent protection scope of the present invention.

Claims

1. A material intelligent optimization method for preparing fitness equipment, characterized in that: The method comprises: Interactively obtain customized input information of the customized user, wherein the customized input information includes user body information and customized equipment type; Performing size fitting according to the user's body shape information and the type of customized equipment to obtain customized equipment specification information; Modeling and restoring the customized equipment specification information to obtain an equipment dynamics model, and synchronizing the user's body posture information to the equipment dynamics model for use fitting to obtain an equipment force data set and an equipment contact data set; Pre-constructing a hard material analysis network, wherein the hard material analysis network includes a coating material analysis branch and a stress-bearing material analysis branch; Interactively obtain the usage environment information of the customized user, and synchronize the usage environment information and the equipment force data set to the coating material analysis branch and the force material analysis branch of the hard material analysis network respectively, to obtain preferred coating material information and preferred force material information; Performing a soft material optimization analysis based on the equipment contact data set to obtain information on the preferred soft material; The preferred coating material information, preferred stress-bearing material information and preferred soft material information are used as material constraints, and the customized equipment specification information is used as structural constraints to execute non-standardized production of customized fitness equipment.

2. The method according to claim 1, characterized in that Performing size fitting according to the user's body shape information and the type of customized equipment to obtain customized equipment specification information, the method further includes: The user's body shape information includes a user's height parameter, a user's weight parameter, and a user's body proportion; Interactively obtain multiple sample body shape information and multiple sample fitness equipment specifications, wherein each sample body shape information includes a sample height parameter, a sample weight parameter, and a sample body proportion; Reorganize the plurality of sample body shape information and the plurality of sample fitness equipment specifications to obtain first branch training data, second branch training data and third branch training data, wherein the first branch training data includes a plurality of sample height parameters, a plurality of sample weight parameters and the plurality of sample fitness equipment specifications, the second branch training data includes a plurality of sample weight parameters, a plurality of sample body shape proportions and the plurality of sample fitness equipment specifications, and the third branch training data includes a plurality of sample height parameters, a plurality of sample body shape proportions and the plurality of sample fitness equipment specifications; Constructing a standard specification fitting branch based on a convolutional neural network, and using the first branch training data, the second branch training data, and the third branch training data to train the standard specification fitting branch, to obtain a first specification fitting branch, a second specification fitting branch, and a third specification fitting branch; Obtaining standard specification equipment interactively according to the customized equipment type, and constructing a size specification fusion layer based on standard specification information of the standard specification equipment; The first specification fitting branch, the second specification fitting branch and the third specification fitting branch are arranged in parallel, and the output ends of the first specification fitting branch, the second specification fitting branch and the third specification fitting branch are connected to the input end of the size specification fusion layer to complete the construction of the size specification fitting network; The user's body shape information is synchronized to the size specification fitting network to perform size specification fitting to obtain the customized equipment specification information.

3. The method according to claim 2, characterized in that A standard specification fitting branch is constructed based on a convolutional neural network, and the first branch training data, the second branch training data, and the third branch training data are used to train the standard specification fitting branch to obtain a first specification fitting branch, a second specification fitting branch, and a third specification fitting branch. The method further includes: Pre-constructing a first parameter convolution training layer and a second parameter convolution training layer, wherein the first parameter convolution training layer includes multiple layers of first parameter kernel functions, and the second parameter convolution training layer includes a second parameter kernel function; In the construction of the first specification fitting branch, the multi-layer first parameter kernel function of the first parameter convolution training layer is used to extract features of the multiple sample height parameters and the multiple sample weight parameters to obtain multi-layer body parameter convolution output samples; The second parameter kernel function of the second parameter convolution training layer is used to extract features of the multiple sample fitness equipment specifications to obtain multiple specification parameter convolution output samples; Using the multi-layer body parameter convolution output samples and the plurality of specification parameter convolution output samples as training data, constructing and training a specification fitting function until convergence and outputting a first specification fitting function; The first specification fitting function is synchronized to the first specification fitting branch.

4. The method according to claim 2, characterized in that Modeling and restoring are performed according to the customized equipment specification information to obtain an equipment dynamics model, and the user body information is synchronized to the equipment dynamics model for use fitting to obtain an equipment force data set and an equipment contact data set. The method further includes: Extracting features from the customized equipment specification information to obtain structural design features and connection design features; Generate a rigid body model of the equipment according to the structural design features, and simulate the motion of the rigid body model of the equipment with the connection design features as constraints to generate a dynamic model of the equipment; Interactively obtain the motion force characteristics of the standard equipment, and synchronize the user's body information and the motion force characteristics to the equipment dynamics model for use fitting to obtain a simulation dynamics model; According to the structural design features, the simulation dynamics model is subjected to force decomposition to obtain force data of multiple components, wherein the force data of the multiple components constitute the equipment force data set; The simulation dynamics model is traversed to locate the motion action nodes, and a plurality of contact nodes are obtained. The plurality of contact nodes constitute the equipment contact data set.

5. The method according to claim 1, characterized in that The method further comprises: Interactively obtain multiple sample environment information and multiple sample coating materials, and associate and store the multiple sample environment information and multiple sample coating materials based on the knowledge graph to generate a sample coating information storage layer; A pre-built environment similarity calculation function is synchronized to the pre-built environment matching analysis layer. The environment similarity calculation function is as follows: ； in, is the environmental similarity, and the usage environment information includes Use environment parameters, The usage environment information The sample environment information includes: Sample environmental parameters, The sample environment information Sample environmental parameters; Pre-build an environmental parameter discriminator and a similarity sorting engine, synchronize the environmental parameter discriminator to the sample coating information storage layer, synchronize the similarity sorting engine to the environmental matching analysis layer, connect the output end of the sample coating information storage layer and the input end of the environmental matching analysis layer, and generate the coating material analysis branch.

6. The method according to claim 5, characterized in that The method further comprises: interactively obtaining a performance deviation threshold and a performance deviation vector, and constructing a performance deviation constraint layer based on the performance deviation threshold and the performance deviation vector; Interactively obtain multiple sample force performance parameters of multiple sample hard materials, and associate and store the multiple sample hard materials and the multiple sample force performance parameters based on the knowledge graph to generate a sample performance storage layer; Connecting the output end of the performance deviation constraint layer and the input end of the sample performance storage layer to complete the construction of the stress material analysis branch; The coating material analysis branch and the stress-bearing material analysis branch are arranged in parallel and the output end and the input end are adapted to generate the hard material analysis network.

7. The method according to claim 6, characterized in that Interactively obtaining the usage environment information of the customized user, and synchronizing the usage environment information and the equipment force data set to the coating material analysis branch and the force material analysis branch of the hard material analysis network respectively, to obtain preferred coating material information and preferred force material information, the method further comprising: In the environment matching analysis layer of the coating material analysis branch, the environment parameter discriminator constructs an environment parameter discrimination constraint based on the use environment information, and traverses the multiple sample environment information based on the environment parameter discrimination constraint to obtain multiple screening environment information and multiple screening coating materials corresponding to the multiple screening environment information; In the environment matching analysis layer of the coating material analysis branch, multiple environment similarities of the multiple screening environment information and the use environment information are calculated one by one based on the environment similarity calculation function, the multiple environment similarities are serialized based on the similarity sorting engine, and the preferred coating material information is obtained in the multiple screening coating material mapping calls based on the extreme value of the serialization result; In the performance deviation constraint layer of the stress-bearing material analysis branch, a set of equipment stress threshold values ​​is generated according to the performance deviation threshold and the equipment stress data set; In the sample performance storage layer of the stress-bearing material analysis branch, the performance deviation vector is used as a comparison constraint and the equipment stress threshold set is used to traverse the multiple sample stress performance parameters to obtain multiple target hard materials, which constitute the preferred stress-bearing material information.

8. The method according to claim 1, characterized in that Performing a soft material optimization analysis according to the equipment contact data set to obtain preferred soft material information, the method further comprising: Interactively obtain the soft material comfort pre-entered by the customization user, and use the soft material comfort as a constraint to obtain multiple sample material attribute information of multiple sample soft materials, wherein each sample material attribute information includes a sample comfort index, a sample service life index, and a sample replacement cost index; Presetting a comfort weight, a life weight, and a replacement weight, and performing fitness calculation on the plurality of sample material attribute information based on the comfort weight, the life weight, and the replacement weight to obtain fitness of the plurality of sample materials; The fitness of the plurality of sample materials is serialized, and the sample soft material corresponding to the extreme value of the fitness of the sample material is called as the preferred soft material information.

Images