CN115292845B - Multi-objective optimization-based arc-shaped flexible spherical hinge design method and device - Google Patents

Multi-objective optimization-based arc-shaped flexible spherical hinge design method and device Download PDF

Info

Publication number
CN115292845B
CN115292845B CN202210967972.2A CN202210967972A CN115292845B CN 115292845 B CN115292845 B CN 115292845B CN 202210967972 A CN202210967972 A CN 202210967972A CN 115292845 B CN115292845 B CN 115292845B
Authority
CN
China
Prior art keywords
parameters
model
design
flexible hinge
neural network
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202210967972.2A
Other languages
Chinese (zh)
Other versions
CN115292845A (en
Inventor
徐振邦
徐安鹏
王学问
韩思凡
于阳
杨金河
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Changchun Institute of Optics Fine Mechanics and Physics of CAS
Original Assignee
Changchun Institute of Optics Fine Mechanics and Physics of CAS
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Changchun Institute of Optics Fine Mechanics and Physics of CAS filed Critical Changchun Institute of Optics Fine Mechanics and Physics of CAS
Priority to CN202210967972.2A priority Critical patent/CN115292845B/en
Publication of CN115292845A publication Critical patent/CN115292845A/en
Application granted granted Critical
Publication of CN115292845B publication Critical patent/CN115292845B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/17—Mechanical parametric or variational design
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
    • G16C60/00—Computational materials science, i.e. ICT specially adapted for investigating the physical or chemical properties of materials or phenomena associated with their design, synthesis, processing, characterisation or utilisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/04—Constraint-based CAD
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Software Systems (AREA)
  • Computer Hardware Design (AREA)
  • Biomedical Technology (AREA)
  • Health & Medical Sciences (AREA)
  • Mathematical Physics (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Mathematical Analysis (AREA)
  • Computational Mathematics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Medical Informatics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention relates to the field of mechanical structure optimization, in particular to a method and a device for designing an arc-shaped flexible spherical hinge based on multi-objective optimization. The method and the device comprise the following steps: calculating structural parameters of the flexible hinge, and carrying out structural modeling according to initial values of the parameters to obtain a theoretical model; calculating finite element models under different structural parameters by changing design parameters and utilizing a finite element method, and obtaining a flexibility matrix according to the finite element models under different structural parameters; designing a neural network model, taking different design parameters as the input of the neural network model, and taking different flexibility matrixes as the output of the neural network model; correcting the theoretical model by using the designed neural network model to obtain a theoretical optimization model; based on the theoretical optimization model, constraint conditions are obtained according to actual design requirements. The invention solves the parameter optimization design problem of the arc-shaped flexible spherical hinge in the prior art.

Description

Multi-objective optimization-based arc-shaped flexible spherical hinge design method and device
Technical Field
The invention relates to the field of mechanical structure optimization, in particular to a method and a device for designing an arc-shaped flexible spherical hinge based on multi-objective optimization.
Background
With the development of high-precision systems and micro-nano operation technologies, there is an urgent need for elements capable of moving in a small space, which not only ensures that high movement accuracy and high resolution can be achieved, but also gives a very high index for miniaturization of its shape and maximization of fatigue strength. The majority of joints of the parallel pointing mechanism are conventional rigid kinematic pairs, such as a sliding pair, a revolute pair, a hook hinge, a ball pair and the like, and the problems of friction, return clearance, creeping and the like inevitably exist, so that the parallel pointing mechanism is one of main reasons that the mechanical mechanism is not easy to obtain high precision.
The flexible hinge can well solve the precision problem caused by the reasons of gap creeping and the like due to the special structure. The integrated structural design of the flexible joint enables the flexible joint to have ultrahigh motion sensitivity, and is very suitable for completing single-axis or multi-axis rotation around a specific axis. The flexible hinge connection precise pointing mechanism has wide demands in the fields of future precise pointing of optical systems, semiconductor production, space detection and other ultra-precise engineering. The integrated design of the flexible hinge is utilized to replace the traditional rigid hinge, so that the space occupation of the mechanism is greatly reduced, and the flexible hinge has the advantage of being unable to replace the requirements of small stroke and small deformation on the aspect of high-precision motion control.
Although there are many design methods for the flexible hinge in theory at present, in the actual design process, the methods still encounter many difficulties, such as that the special structure of the flexible hinge can only be optimally designed according to different parameters, and engineering experience is utilized, although the calculation design can be performed by using the finite element method, the calculation is needed to be continuously performed, and the calculation force is too much relied on, so that the design period of the flexible hinge is greatly increased, the design period of the whole device is increased, and the future high-precision pointing device has strong limitation on large-scale application.
In addition, the optimization design of the flexible hinge is mainly topology optimization, and because the topology optimization is mainly that the computer performs finite element analysis to search for the optimal solution, no special requirements are made on the specific shape of the design part, the flexible hinge design finished product is difficult to process because the practical constraint condition is not combined in the design process, and the standardized production and application of the flexible hinge are greatly limited.
The invention patent application with publication number of CN106096158A discloses a topology optimization design method of a flexible hinge, which comprises the following steps: step 1: establishing a flexible hinge topological optimization design model, setting a typical incision flexible hinge contour as the shape of a design domain, and defining a rigid region (a non-design domain); step 2: establishing a flexible hinge topological optimization finite element model; step 3: establishing a flexible hinge topology optimization problem mathematical model based on the finite element model; step 4: calculating the sensitivity of the flexible hinge topology optimization problem; step 5: solving a flexible hinge topology optimization problem by adopting an optimization algorithm, and updating design variables to obtain a final topology result diagram; step 6: and extracting the outline of the final topological result diagram obtained by topological optimization, and obtaining the novel flexible hinge through proper modification.
However, the optimization design for the flexible hinge is mainly focused in the design of the single-axis flexible hinge, but the modeling of the single-axis flexible hinge is relatively simple, is equivalent to a rigid plane kinematic pair, and can only bend and deform along one axis, so that the application requirements cannot be met under the condition that a large number of flexible spherical hinges and flexible hook hinges are needed for a high-precision pointing platform in the future.
Disclosure of Invention
The embodiment of the invention provides a method and a device for designing an arc-shaped flexible spherical hinge based on multi-objective optimization, which at least solve the problem of parameter optimization design of the arc-shaped flexible spherical hinge in the prior art.
According to an embodiment of the invention, there is provided a multi-objective optimization-based arc-shaped flexible spherical hinge design method, which includes the following steps:
according to specific design requirements, designing initial values of parameters of the flexible hinge;
calculating structural parameters of the flexible hinge, and carrying out structural modeling according to initial values of the parameters to obtain a theoretical model;
calculating finite element models under different structural parameters by changing design parameters and utilizing a finite element method, and obtaining a flexibility matrix according to the finite element models under different structural parameters;
designing a neural network model, taking different design parameters as the input of the neural network model, and taking different flexibility matrixes as the output of the neural network model;
Correcting the theoretical model by using the designed neural network model to obtain a theoretical optimization model;
Based on the theoretical optimization model, constraint conditions are obtained according to actual design requirements.
Further, the method further comprises:
and optimizing structural parameters by utilizing a multi-objective optimization algorithm based on deep reinforcement learning to obtain the optimal flexible hinge size parameters.
Further, the initial values of the design parameters of the flexible hinge according to the specific design requirements comprise:
The design parameters are the minimum section diameter t, the circular arc central angle theta m, the circular arc radius R and the deflection angle theta of the whole length of the flexible hinge, the input parameters are expressed as X= [ tθ m R theta ] by a matrix, the material properties of the flexible hinge are set, the material properties are respectively the material density ρ unit kg/m 3 of the flexible hinge, the elastic modulus E unit of the flexible hinge is MPa, the shearing model G unit of the material is MPa, and the actually required target conditions are set.
Further, calculating structural parameters of the flexible hinge, and performing structural modeling according to initial values of the parameters, wherein obtaining the theoretical model comprises:
Calculating length information of the flexible hinge, wherein the thickness of the flexible hinge is expressed as t (x) =t+2R (1-cos theta), theta epsilon [ -theta m,θm ], the total length of the flexible hinge is expressed as L=2Rsin theta m +2t (x), the length of a flexible part of the flexible hinge is expressed as l=2Rsin theta m, and finally, using UG software to carry out structural modeling on initial values of parameters.
Further, by changing design parameters and calculating finite element models under different structural parameters by using a finite element method, obtaining a flexibility matrix according to the finite element models under different structural parameters includes:
The compliance of the flexible hinge in each direction is deduced by using a second theorem of material mechanics, the deformation of the flexible hinge is written into a matrix form, delta represents the deformation in six directions, C represents the translational and torsional matrixes in six directions, F represents the force and moment matrixes in six directions, and the relation is expressed as follows:
Δ＝CF；
The method comprises the steps of calculating a finite element model under different structural parameters by changing design parameters X= [ tθ m Rθ ], obtaining a flexibility matrix C by using a finite element method, determining three coefficients by using a control variable method, changing parameters to be changed by using a progressive method, and obtaining the flexibility matrix according to the finite element model under different structural parameters.
Further, designing the neural network model, taking different design parameters as the input of the neural network model, and taking different compliance matrices as the output of the neural network model comprises:
designing a neural network model, taking different design parameters X= [ tθ m Rθ ] as the input of the neural network model, taking different flexibility matrixes obtained by finite element software as the output of the neural network model, and carrying out a system identification program to obtain the influence condition of different structural parameter values on the flexibility matrixes.
Further, correcting the theoretical model by using the designed neural network model, and obtaining the theoretical optimization model comprises the following steps:
And correcting the theoretical model by using the obtained neural network model, and regularizing the data to obtain a more accurate theoretical optimization model.
Further, optimizing the structural parameters using a multi-objective optimization algorithm based on deep reinforcement learning includes:
the design of the rotation capacity of the flexible hinge, wherein the rotation capacity is defined as the rotation angle thetat of the unit torque in the y direction around the y axis, and the larger the thetat is, the larger the rotation range of the hinge is; θ t＝C1, the design rule of the flexible hinge indicates that the flexible hinge needs stronger rotation capability, so that C1 should be kept as large as possible;
The relative flexibility design is that the flexible hinge deformation is realized by increasing the flexibility of the movement direction, and the increase of the flexibility of the expected direction can also lead to the increase of the flexibility of the vertical movement direction, so that the actual deformation deviates from the expected direction, the actual deformation of the structure tends to be complex, the concept of relative rigidity is introduced, and when rho is larger, the structure is more stable; wherein:
The general optimization objective is max (C 1)max(C2)min(C3);
In addition, the maximum value and the minimum value of the parameter X= [ tθ m Rθ ] are constrained according to the actual situation, and the numerical value of the structural size is constrained.
According to another embodiment of the present invention, there is provided a multi-objective optimization-based arc-shaped flexible spherical hinge design apparatus, including:
the parameter initial value design unit is used for designing a parameter initial value of the flexible hinge according to specific design requirements;
the structure modeling unit is used for calculating the structure parameters of the flexible hinge and carrying out structure modeling according to the initial values of the parameters to obtain a theoretical model;
The flexibility matrix acquisition unit is used for obtaining a flexibility matrix according to the finite element model under different structural parameters by changing design parameters and calculating the finite element model under different structural parameters by utilizing a finite element method;
The neural network model design unit is used for designing a neural network model, taking different design parameters as the input of the neural network model, and taking different flexibility matrixes as the output of the neural network model;
The theoretical optimization model acquisition unit is used for correcting the theoretical model by utilizing the designed neural network model to obtain the theoretical optimization model;
and the constraint condition acquisition unit is used for obtaining constraint conditions according to actual design requirements based on the theoretical optimization model.
Further, the apparatus further comprises:
And the structural parameter optimization unit is used for optimizing structural parameters by utilizing a multi-objective optimization algorithm based on deep reinforcement learning to obtain the optimal flexible hinge size parameters.
The processor is used for running a program, wherein the program runs to execute any one of the arc-shaped flexible spherical hinge design methods based on multi-objective optimization.
The processor is used for running a program, wherein the program runs to execute any one of the arc-shaped flexible spherical hinge design methods based on multi-objective optimization.
The method and the device for designing the circular arc-shaped flexible spherical hinge based on the multi-objective optimization are mainly used for designing the circular arc-shaped flexible spherical hinge by utilizing an intelligent multi-objective optimization algorithm in order to manually limit target parameters to be designed, so that the design period of the circular arc-shaped flexible spherical hinge is greatly shortened, the production cost is reduced, the whole design period of a product is shortened, a foundation is laid for the production, the processing and the popularization and the use of the flexible spherical hinge, and the contradiction between the high bearing capacity of the circular arc-shaped flexible spherical hinge and the reduced structure size of the hinge and the contradiction between the rigidity is reduced to obtain a large rotation angle and guarantee precision are solved. On the basis of meeting the rigidity and the motion precision simultaneously, a novel flexible hinge is designed and prepared so as to realize the motion output of gapless and large motion stroke.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
FIG. 1 is a flow chart of a multi-objective optimization-based arc-shaped flexible spherical hinge design method;
FIG. 2 is a preferred flow chart of the method for designing a circular arc type flexible spherical hinge based on multi-objective optimization of the present invention;
FIG. 3 is a structural parameter diagram of a flexible hinge in the design method of the circular arc flexible spherical hinge based on multi-objective optimization;
FIG. 4 is a diagram showing stress conditions of a flexible hinge in the design method of the circular arc flexible spherical hinge based on multi-objective optimization;
FIG. 5 is a finite element model diagram of a flexible hinge in the arc-shaped flexible spherical hinge design method based on multi-objective optimization;
FIG. 6 is a flow chart of a multi-objective optimization algorithm in the design method of the circular arc flexible spherical hinge based on multi-objective optimization;
FIG. 7 is a block diagram of a multi-objective optimization-based circular arc flexible spherical hinge design device according to the present invention;
Fig. 8 is a preferred block diagram of the arc-shaped flexible spherical hinge design device based on multi-objective optimization of the invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
According to an embodiment of the present invention, there is provided a method for designing a circular arc type flexible spherical hinge based on multi-objective optimization, referring to fig. 1, including the steps of:
s100, designing an initial value of a parameter of the flexible hinge according to specific design requirements;
s200, calculating structural parameters of the flexible hinge, and carrying out structural modeling according to initial values of the parameters to obtain a theoretical model;
S300, calculating finite element models under different structural parameters by changing design parameters and utilizing a finite element method, and obtaining a flexibility matrix according to the finite element models under different structural parameters;
s400, designing a neural network model, taking different design parameters as the input of the neural network model, and taking different flexibility matrixes as the output of the neural network model;
S500, correcting a theoretical model by using the designed neural network model to obtain a theoretical optimization model;
and S600, obtaining constraint conditions according to actual design requirements based on a theoretical optimization model.
The arc-shaped flexible spherical hinge design method based on multi-objective optimization in the embodiment of the invention is mainly used for designing the arc-shaped flexible spherical hinge by utilizing an intelligent multi-objective optimization algorithm in order to artificially limit target parameters to be designed, so that the design period of the arc-shaped flexible spherical hinge is greatly shortened, the production cost is reduced, the whole design period of a product is shortened, a foundation is laid for the production, the processing and the popularization and the use of the flexible spherical hinge, and the contradiction between the high bearing capacity of the arc-shaped flexible spherical hinge and the reduced structure size of the hinge and the contradiction between the rigidity is reduced to obtain a large rotation angle and guarantee precision are solved. On the basis of meeting the rigidity and the motion precision simultaneously, a novel flexible hinge is designed and prepared so as to realize the motion output of gapless and large motion stroke.
Wherein, referring to fig. 2, the method further comprises:
s700: and optimizing structural parameters by utilizing a multi-objective optimization algorithm based on deep reinforcement learning to obtain the optimal flexible hinge size parameters.
Wherein, according to the specific design requirement to flexible hinge design parameter initial value include:
The design parameters are the minimum section diameter t, the circular arc central angle theta m, the circular arc radius R and the deflection angle theta of the whole length of the flexible hinge, the input parameters are expressed as X= [ tθ m R theta ] by a matrix, the material properties of the flexible hinge are set, the material properties are respectively the material density ρ unit kg/m 3 of the flexible hinge, the elastic modulus E unit of the flexible hinge is MPa, the shearing model G unit of the material is MPa, and the actually required target conditions are set.
Calculating structural parameters of the flexible hinge, and performing structural modeling according to initial values of the parameters, wherein obtaining a theoretical model comprises:
Calculating length information of the flexible hinge, wherein the thickness of the flexible hinge is expressed as t (x) =t+2R (1-cos theta), theta epsilon [ -theta m,θm ], the total length of the flexible hinge is expressed as L=2Rsin theta m +2t (x), the length of a flexible part of the flexible hinge is expressed as l=2Rsin theta m, and finally, using UG software to carry out structural modeling on initial values of parameters.
The method for obtaining the flexibility matrix according to the finite element model under different structural parameters comprises the following steps of:
The compliance of the flexible hinge in each direction is deduced by using a second theorem of material mechanics, the deformation of the flexible hinge is written into a matrix form, delta represents the deformation in six directions, C represents the translational and torsional matrixes in six directions, F represents the force and moment matrixes in six directions, and the relation is expressed as follows:
Δ＝CF；
The method comprises the steps of calculating a finite element model under different structural parameters by changing design parameters X= [ tθ m Rθ ], obtaining a flexibility matrix C by using a finite element method, determining three coefficients by using a control variable method, changing parameters to be changed by using a progressive method, and obtaining the flexibility matrix according to the finite element model under different structural parameters.
The design of the neural network model, taking different design parameters as the input of the neural network model, and taking different flexibility matrixes as the output of the neural network model comprises the following steps:
designing a neural network model, taking different design parameters X= [ tθ m Rθ ] as the input of the neural network model, taking different flexibility matrixes obtained by finite element software as the output of the neural network model, and carrying out a system identification program to obtain the influence condition of different structural parameter values on the flexibility matrixes.
The method for obtaining the theoretical optimization model comprises the following steps of:
And correcting the theoretical model by using the obtained neural network model, and regularizing the data to obtain a more accurate theoretical optimization model.
Wherein, utilizing the multi-objective optimization algorithm based on the deep reinforcement learning to optimize the structural parameters comprises:
the design of the rotation capacity of the flexible hinge, wherein the rotation capacity is defined as the rotation angle thetat of the unit torque in the y direction around the y axis, and the larger the thetat is, the larger the rotation range of the hinge is; θ t＝C1, the design rule of the flexible hinge indicates that the flexible hinge needs stronger rotation capability, so that C1 should be kept as large as possible;
The relative flexibility design is that the flexible hinge deformation is realized by increasing the flexibility of the movement direction, and the increase of the flexibility of the expected direction can also lead to the increase of the flexibility of the vertical movement direction, so that the actual deformation deviates from the expected direction, the actual deformation of the structure tends to be complex, the concept of relative rigidity is introduced, and when rho is larger, the structure is more stable; wherein:
The general optimization objective is max (C 1)max(C2)min(C3);
In addition, the maximum value and the minimum value of the parameter X= [ tθ m Rθ ] are constrained according to the actual situation, and the numerical value of the structural size is constrained.
The following describes the design method of the circular arc type flexible spherical hinge based on multi-objective optimization in detail by using a specific embodiment:
Aiming at a flexible system with multiple degrees of freedom in space, a flexible spherical hinge is one of important solutions, and the flexible spherical hinge has important significance for the design of flexible spherical hinges in the application fields of future micro-electromechanical systems (MEMS), high-precision micro-displacement work tables, biological medicine robots and the like. The flexible hinge is designed on the conceptual level by adopting the topology optimization method, so that a novel flexible hinge with more complex structure and more excellent performance can be designed, and the flexible hinge has larger flexibility, higher precision and smaller maximum stress.
In order to solve the problem of parameter optimization design of the circular arc-shaped flexible spherical hinge, the invention provides a circular arc-shaped flexible spherical hinge design method based on multi-objective optimization, which is mainly used for artificially limiting target parameters to be designed, and the intelligent multi-objective optimization algorithm is utilized for designing the circular arc-shaped flexible spherical hinge, so that the design period of the circular arc-shaped flexible spherical hinge is greatly reduced, the production cost is reduced, the whole design period of a product is shortened, a foundation is laid for the production, the processing and the popularization of the flexible spherical hinge, and the contradiction between the high bearing capacity of the circular arc-shaped flexible spherical hinge and the reduced structure size of the hinge and the contradiction between the rigidity is reduced to obtain a large rotation angle and guarantee precision are solved. On the basis of meeting the rigidity and the motion precision simultaneously, a novel flexible hinge is designed and prepared so as to realize the motion output of gapless and large motion stroke.
In order to achieve the above purpose, the technical scheme adopted by the invention mainly comprises the following 7 parts:
(1) According to specific design requirements, the initial values of design parameters of the flexible hinge are respectively the minimum section diameter t, the circular arc central angle theta m, the circular arc radius R and the deflection angle theta affecting the whole length of the flexible hinge, then the input parameters can be expressed as X= [ tθ m R theta ] by using a matrix, the material properties of the flexible hinge are set, the material density rho unit kg/m 3 of the flexible hinge is respectively, the elastic modulus E unit of the flexible hinge is MPa, the shearing model G unit of the material is MPa, and the actually required target conditions are set;
(2) Calculating information such as the length of the flexible hinge, wherein the flexible hinge structure parameter is shown in fig. 3, the flexible hinge thickness can be expressed as t (x) =t+2r (1-cos θ), θ e [ - θ m,θm ], the flexible hinge total length can be expressed as l=2rsin θ m +2t (x), the flexible hinge flexible part length can be expressed as l=2rsin θ m, and finally, the initial parameter value is structurally modeled by using UG software;
(3) Referring to fig. 4, the compliance of the flexible hinge in each direction is deduced by using the second theorem of the material mechanics, the deformation of the flexible hinge is written in the form of a matrix, Δ represents the deformation in six directions, C represents the translational and torsional matrices in six directions, and F represents the force and moment matrices applied in six directions, the relationship of which is expressed as follows:
Δ＝CF；
(4) Calculating a finite element model under different structural parameters by changing a design parameter X= [ tθ m Rθ ], obtaining a flexibility matrix C by using a finite element method, determining three coefficients by using a control variable method, changing parameters to be changed by using a progressive method, and obtaining the flexibility matrix according to the finite element model under different structural parameters;
(5) Designing a neural network model, taking different design parameters X= [ tθ m Rθ ] as the input of the neural network model, taking different flexibility matrixes obtained by finite element software as the output of the model, and carrying out a system identification program to obtain the influence condition of different structural parameter values on the flexibility matrixes;
(6) Correcting the theoretical model obtained in the step (2) by utilizing the neural network model obtained in the step (5), and regularizing the data to obtain a more accurate theoretical model;
(7) Based on the theoretical optimization model obtained in the step (6), constraint conditions are obtained according to actual design requirements;
(8) And optimizing the input structural parameter X= [ tθ m Rθ ] by using a multi-objective optimization algorithm based on deep reinforcement learning to obtain the optimal flexible hinge size parameter.
The technical key points of the invention are at least as follows:
1. and determining target conditions required by practical application according to the initial values of the design parameters of the flexible hinge according to specific design requirements.
2. And calculating the required structural size of the flexible hinge through a formula, and carrying out structural modeling on the structure of the flexible hinge by utilizing UG modeling software.
3. And obtaining a calculation formula of the flexibility matrix model of the flexible hinge through theoretical derivation.
4. And changing design parameters of the flexible hinge by a finite element method to obtain a flexibility matrix of the flexible hinge under different parameters.
5 . Designing a neural network model, taking different design parameters X= [ tθ m Rθ ] as the input of the neural network model, taking different flexibility matrixes obtained by finite element software as the output of the model, and carrying out a system identification program to obtain the influence condition of different structural parameter values on the flexibility matrixes.
6. And (3) correcting the theoretical model obtained in the step (2) by using the neural network model obtained in the step (5), and regularizing the data to obtain a more accurate theoretical model.
7. Based on the theoretical optimization model obtained in the step (6), constraint conditions are obtained according to actual design requirements, and the input structural parameter X= [ tθ m Rθ ] is optimized by utilizing a multi-objective optimization algorithm based on deep reinforcement learning, so that the optimal flexible hinge size parameter is obtained.
The protection points of the invention are at least as follows:
1. And obtaining a flexibility relation matrix of the flexible hinge under different structural parameters by a finite element analysis method, and recording original simulation data.
2. And the relation between the flexible hinge structure parameter and the flexibility matrix is identified through a neural network algorithm, so that the theoretical model is optimized, and the accuracy of the theoretical model is improved.
3. And (3) analyzing to obtain an optimized theoretical model, and reducing modeling errors according to comparison with actual simulation data to obtain the influence of different structural parameters on the flexible hinge flexibility matrix.
4. The constraint conditions of the structural parameters are designed according to actual needs, and the flexibility effect is expected to be achieved.
5. And optimizing the four input structural parameters C= [ tθ m Rθ ] by using a multi-objective optimization algorithm based on deep reinforcement learning to obtain the optimal flexible hinge size parameter.
According to the technical scheme of the invention, the invention has at least the following advantages and effects:
1. according to the invention, the cylindrical flexible hinge is optimized by utilizing a multi-objective optimization algorithm, so that the design period of the flexible hinge is greatly shortened, the flexible hinge is ensured to meet the design requirement, the processing procedure is simplified, and the design parameters are simplified.
2. The linear displacement range of the spatial micro-motion pointing mechanism of the flexible hinge designed by the invention is in the micron level, and the angular displacement range is in the milliradian level, so that in the design process of the flexible hinge: the hinge has simple structure and small size; the rigidity in the rotation direction is small, and the rigidity in other directions is large; the deformation precision is high; the controllability is good; the bearing capacity is high.
3. The standardized design program of the flexible hinge is provided, the popularization and the use of the flexible hinge are facilitated, and the contradiction between the high bearing capacity of the circular arc-shaped flexible spherical hinge and the reduced structure size of the hinge and the contradiction between the rigidity reduction to obtain a large rotation angle and guarantee precision are solved. On the basis of meeting the rigidity and the motion precision simultaneously, a novel flexible hinge is designed and prepared so as to realize the motion output of gapless and large motion stroke.
4. The problem of inaccurate multi-objective optimization of a large amount of data in the flexible hinge parameter optimization process is solved, and compared with other optimization methods, the efficiency and the accuracy of simulation results are greatly improved.
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
(1) According to specific design requirements, the initial values of the design parameters of the flexible hinge are respectively the minimum section diameter t, the circular arc central angle theta m, the circular arc radius R and the deflection angle theta affecting the whole length of the flexible hinge, then the input parameters can be expressed as X= [ ttheta m R theta ] by using a matrix, the material properties of the flexible hinge are set, the material density rho unit kg/m 3 of the flexible hinge is respectively set, the elastic modulus E unit of the flexible hinge is MPa, the shearing model G unit of the material is MPa, and the actually required target conditions are set;
(2) Calculating information such as the length of the flexible hinge, wherein the flexible hinge structure parameter is shown in fig. 3, the flexible hinge thickness can be expressed as t (x) =t+2r (1-cos θ), the total length of the flexible hinge with the length of θ e [ - θ m,θm ] can be expressed as l=2rsin θ m +2t (x), the flexible hinge flexible part length can be expressed as l=2rsin θ m, and finally, the initial value of the parameter is structurally modeled by using UG software; wherein fig. 3 is a structural parameter diagram of the flexible hinge, the parameters are a minimum section diameter t, a circular arc central angle θ m, a circular arc radius R, and a deflection angle θ affecting the overall length of the flexible hinge, the thickness of the flexible hinge can be expressed as t (x) =t+2r (1-cos θ), θ e [ - θ m,θm ], the total length of the flexible hinge can be expressed as l=2rsin θ m +2t (x), and the flexible portion length of the flexible hinge can be expressed as l=2rsin θ m;
(3) Referring to fig. 4, the compliance of the flexible hinge in each direction is deduced by using the second theorem of the material mechanics, and is respectively simplified into a form that deformation of the flexible hinge can be written as a matrix, delta represents deformation in six directions, C represents translation and torsion matrices in six directions, F represents force and moment matrices applied in six directions, and the relationship is represented as follows:
Δ＝CF；
Wherein Δ＝[Δx Δy Δz Δαx Δαy Δαz]TF＝[Fx Fy Fz Mx My Mz]T; is a graph of the stress on a flexible hinge, assuming that the flexible hinge is stressed at one end and at the other end.
The axisymmetric properties of the flexible hinge are as follows:
and G and E are fixed values, and the flexibility in three directions is in a linear relationship.
And C αz-Fy＝Cαy-Fz＝Cy-Mz＝Cz-My is used for the preparation of the medicine,
The compliance matrix is mainly influenced by three variables of C ay-MyCαz-Fy and C x-Fx and is uniformly rewritten into C 1＝Cay-MyC2＝θαz- FyC3＝Cx-Fx;
The theoretical calculation formula is obtained:
wherein f1 f 2f 3 is a parameter matrix related to the minimum section diameter t, the central angle θm of the arc and the radius R of the arc, respectively
Wherein the method comprises the steps of
(4) Calculating a finite element model under different structural parameters by changing design parameters x= [ tθ m Rθ ], using a finite element method to obtain a flexibility matrix C as shown in figure 5, determining three coefficients by using a control variable method, changing parameters to be changed by using a progressive method, and obtaining the flexibility matrix according to the finite element model under different structural parameters; fig. 5 is a finite element model diagram of a flexible hinge, wherein one end of the flexible hinge is fixed for modal analysis, and the flexibility of each direction is calculated.
(5) Designing a neural network model, taking different design parameters x= [ tθ m R θ ] as the input of the neural network model, taking different flexibility matrixes obtained by finite element software as the output of the model, and carrying out a system identification program to obtain the influence condition of different structural parameter values on the flexibility matrixes:
(5.1) selecting RBF neural network structure as the neural network model structure model;
(5.2) regularizing the finite element analysis data obtained in the step (4) and dividing the data into a training set and a test set, wherein the training set generally occupies 70% of the total data, and the test set occupies 30%;
And (5.3) initializing the RBF neural network, and respectively giving initial values To the input layer weight Wih, the hidden layer weight Who, the output layer weight Who, the hidden layer threshold Th and the output layer threshold To.
(5.4) The input layer of the neural network model is a minimum section diameter t, a circular arc central angle θ m, a circular arc radius R, and a deflection angle θ affecting the overall length of the flexible hinge, and the output layer of the neural network is a compliance matrix Crbf, wherein C rbf can be expressed as:
(6) Correcting the theoretical model obtained in the step (2) by utilizing the neural network model obtained in the step (5) to obtain a more accurate theoretical model;
(7) Based on the theoretical optimization model obtained in the step (6), constraint conditions are obtained according to actual design requirements:
(7.1) design of rotation capability of the flexible hinge, wherein the movement of the flexible hinge is realized by means of elastic deformation of the unit, when the deformation exceeds the elastic deformation range of the material, plastic deformation occurs, the material is permanently deformed, and the movement condition is unpredictable, so that the flexible hinge is limited to a limited movement space. The turning capability may be defined as the rotational angle θt about the y-axis with a unit torque in the y-direction, with a greater θt indicating a greater range of hinge rotation. θ t＝C1 the design rules of the flexible hinge dictate that the flexible hinge requires a strong rotational capability so that C1 should be kept as large as possible.
(7.2) Relative compliance design, the compliant hinge deformation is achieved by increasing compliance in the direction of motion, but an increase in compliance in the desired direction will also result in an increase in compliance in the direction of perpendicular motion, such that the actual deformation deviates from the desired direction, the actual deformation of the structure tends to be complex, so introducing the concept of relative stiffness, it is apparent that the more stable the structure is when ρ is greater.
The general optimization objective is max (C 1)max(C2)min(C3);
In addition, the maximum value and the minimum value of the parameter X= [ tθ m Rθ ] are constrained according to the actual situation, and the numerical value of the structural size is constrained.
(8) And optimizing the input structural parameter X= [ tθ m Rθ ] by using a multi-objective optimization algorithm based on deep reinforcement learning to obtain the optimal flexible hinge size parameter.
Referring to fig. 6, a flowchart of a multi-objective optimization algorithm design is shown, and the steps of the multi-objective optimization algorithm design for deep reinforcement learning are as follows:
(8.1) decomposing the optimization target into a plurality of sub-problems based on a decomposition strategy;
(8.2) modeling the sub-problems obtained by decomposition by utilizing a neighborhood-based parameter migration strategy and sub-problems based on a deep neural Network, and modeling the sub-problems by adopting a Pointer Network model of a Pointer Network;
(8.3) training the model based on a reinforcement learning algorithm;
(8.4) after the optimization algorithm model is completed, optimizing the flexible hinge structure parameters according to constraint conditions;
(8.5) obtaining the optimal structural parameters under the constraint condition, and outputting the specific structural parameters of the flexible hinge;
(8.6) obtaining a structural parameter model of the flexible hinge.
Example 2
According to another embodiment of the present invention, there is provided a multi-objective optimization-based arc-shaped flexible spherical hinge design device, referring to fig. 7, including:
A parameter initial value design unit 201, configured to design a parameter initial value for the flexible hinge according to specific design requirements;
the structure modeling unit 202 is configured to calculate a structural parameter of the flexible hinge, and perform structure modeling according to an initial value of the parameter, so as to obtain a theoretical model;
The compliance matrix obtaining unit 203 is configured to obtain a compliance matrix according to a finite element model under different structural parameters by changing design parameters and calculating the finite element model under different structural parameters by using a finite element method;
The neural network model design unit 204 is configured to design a neural network model, input the neural network model with different design parameters, and output the neural network model with different flexibility matrices;
The theoretical optimization model obtaining unit 205 is configured to correct the theoretical model by using the designed neural network model to obtain a theoretical optimization model;
constraint acquiring unit 206 is configured to obtain constraint according to an actual design requirement based on the theoretical optimization model.
Wherein, referring to fig. 8, the apparatus further comprises:
The structural parameter optimizing unit 207 is configured to optimize structural parameters by using a multi-objective optimization algorithm based on deep reinforcement learning, so as to obtain an optimal flexible hinge size parameter.
The arc-shaped flexible spherical hinge design device based on multi-objective optimization in the embodiment of the invention is mainly used for designing the arc-shaped flexible spherical hinge by utilizing an intelligent multi-objective optimization algorithm in order to manually limit target parameters to be designed, so that the design period of the arc-shaped flexible spherical hinge is greatly shortened, the production cost is reduced, the whole design period of a product is shortened, a foundation is laid for the production, processing and popularization and use of the flexible spherical hinge, and the contradiction between the high bearing capacity of the arc-shaped flexible spherical hinge and the reduced structure size of the hinge and the contradiction between the reduction of rigidity so as to obtain a large rotation angle and guarantee precision are solved. On the basis of meeting the rigidity and the motion precision simultaneously, a novel flexible hinge is designed and prepared so as to realize the motion output of gapless and large motion stroke.
The following describes the arc-shaped flexible spherical hinge design device based on multi-objective optimization in detail by using a specific embodiment:
Aiming at a flexible system with multiple degrees of freedom in space, a flexible spherical hinge is one of important solutions, and the flexible spherical hinge has important significance for the design of flexible spherical hinges in the application fields of future micro-electromechanical systems (MEMS), high-precision micro-displacement work tables, biological medicine robots and the like. The flexible hinge is designed on the conceptual level by adopting the topology optimization method, so that a novel flexible hinge with more complex structure and more excellent performance can be designed, and the flexible hinge has larger flexibility, higher precision and smaller maximum stress.
In order to solve the problem of parameter optimization design of the circular arc-shaped flexible spherical hinge, the invention provides a circular arc-shaped flexible spherical hinge design device based on multi-objective optimization, which is mainly used for manually limiting target parameters to be designed, and the intelligent multi-objective optimization algorithm is utilized for designing the circular arc-shaped flexible spherical hinge, so that the design period of the circular arc-shaped flexible spherical hinge is greatly reduced, the production cost is reduced, the whole design period of a product is shortened, a foundation is laid for the production, the processing and the popularization of the flexible spherical hinge, the contradiction between the high bearing capacity of the circular arc-shaped flexible spherical hinge and the reduced structure size of the hinge is solved, and the rigidity is reduced so as to obtain the contradiction between the large rotation angle and the guarantee precision. On the basis of meeting the rigidity and the motion precision simultaneously, a novel flexible hinge is designed and prepared so as to realize the motion output of gapless and large motion stroke.
In order to achieve the above purpose, the technical scheme adopted by the invention mainly comprises the following 7 parts:
(1) According to specific design requirements, the initial values of design parameters of the flexible hinge are respectively the minimum section diameter t, the circular arc central angle theta m, the circular arc radius R and the deflection angle theta affecting the whole length of the flexible hinge, then the input parameters can be expressed as X= [ tθ m R theta ] by using a matrix, the material properties of the flexible hinge are set, the material density rho unit kg/m 3 of the flexible hinge is respectively, the elastic modulus E unit of the flexible hinge is MPa, the shearing model G unit of the material is MPa, and the actually required target conditions are set;
(2) Calculating information such as the length of the flexible hinge, wherein the flexible hinge structure parameter is shown in fig. 3, the flexible hinge thickness can be expressed as t (x) =t+2r (1-cos θ), θ e [ - θ m,θm ], the flexible hinge total length can be expressed as l=2rsin θ m +2t (x), the flexible hinge flexible part length can be expressed as l=2rsin θ m, and finally, the initial parameter value is structurally modeled by using UG software;
(3) Referring to fig. 4, the compliance of the flexible hinge in each direction is deduced by using the second theorem of the material mechanics, the deformation of the flexible hinge is written in the form of a matrix, Δ represents the deformation in six directions, C represents the translational and torsional matrices in six directions, and F represents the force and moment matrices applied in six directions, the relationship of which is expressed as follows:
Δ＝CF；
(4) Calculating a finite element model under different structural parameters by changing a design parameter X= [ tθ m Rθ ], obtaining a flexibility matrix C by using a finite element method, determining three coefficients by using a control variable method, changing parameters to be changed by using a progressive method, and obtaining the flexibility matrix according to the finite element model under different structural parameters;
(5) Designing a neural network model, taking different design parameters X= [ tθ m Rθ ] as the input of the neural network model, taking different flexibility matrixes obtained by finite element software as the output of the model, and carrying out a system identification program to obtain the influence condition of different structural parameter values on the flexibility matrixes;
(6) Correcting the theoretical model obtained in the step (2) by utilizing the neural network model obtained in the step (5), and regularizing the data to obtain a more accurate theoretical model;
(7) Based on the theoretical optimization model obtained in the step (6), constraint conditions are obtained according to actual design requirements;
(8) And optimizing the input structural parameter X= [ tθ m Rθ ] by using a multi-objective optimization algorithm based on deep reinforcement learning to obtain the optimal flexible hinge size parameter.
The technical key points of the invention are at least as follows:
1. and determining target conditions required by practical application according to the initial values of the design parameters of the flexible hinge according to specific design requirements.
2. And calculating the required structural size of the flexible hinge through a formula, and carrying out structural modeling on the structure of the flexible hinge by utilizing UG modeling software.
3. And obtaining a calculation formula of the flexibility matrix model of the flexible hinge through theoretical derivation.
4. And changing design parameters of the flexible hinge by a finite element method to obtain a flexibility matrix of the flexible hinge under different parameters.
5. Designing a neural network model, taking different design parameters X= [ tθ m Rθ ] as the input of the neural network model, taking different flexibility matrixes obtained by finite element software as the output of the model, and carrying out a system identification program to obtain the influence condition of different structural parameter values on the flexibility matrixes.
6. And (3) correcting the theoretical model obtained in the step (2) by using the neural network model obtained in the step (5), and regularizing the data to obtain a more accurate theoretical model.
7. Based on the theoretical optimization model obtained in the step (6), constraint conditions are obtained according to actual design requirements, and the input structural parameter X= [ tθ m Rθ ] is optimized by utilizing a multi-objective optimization algorithm based on deep reinforcement learning, so that the optimal flexible hinge size parameter is obtained.
The protection points of the invention are at least as follows:
1. And obtaining a flexibility relation matrix of the flexible hinge under different structural parameters by a finite element analysis method, and recording original simulation data.
2. And the relation between the flexible hinge structure parameter and the flexibility matrix is identified through a neural network algorithm, so that the theoretical model is optimized, and the accuracy of the theoretical model is improved.
3. And (3) analyzing to obtain an optimized theoretical model, and reducing modeling errors according to comparison with actual simulation data to obtain the influence of different structural parameters on the flexible hinge flexibility matrix.
4. The constraint conditions of the structural parameters are designed according to actual needs, and the flexibility effect is expected to be achieved.
5. And optimizing the four input structural parameters X= [ tθ m Rθ ] by using a multi-objective optimization algorithm based on deep reinforcement learning to obtain the optimal flexible hinge size parameter.
According to the technical scheme of the invention, the invention has at least the following advantages and effects:
1. according to the invention, the cylindrical flexible hinge is optimized by utilizing a multi-objective optimization algorithm, so that the design period of the flexible hinge is greatly shortened, the flexible hinge is ensured to meet the design requirement, the processing procedure is simplified, and the design parameters are simplified.
2. The linear displacement range of the spatial micro-motion pointing mechanism of the flexible hinge designed by the invention is in the micron level, and the angular displacement range is in the milliradian level, so that in the design process of the flexible hinge: the hinge has simple structure and small size; the rigidity in the rotation direction is small, and the rigidity in other directions is large; the deformation precision is high; the controllability is good; the bearing capacity is high.
3. The standardized design program of the flexible hinge is provided, the popularization and the use of the flexible hinge are facilitated, and the contradiction between the high bearing capacity of the circular arc-shaped flexible spherical hinge and the reduced structure size of the hinge and the contradiction between the rigidity reduction to obtain a large rotation angle and guarantee precision are solved. On the basis of meeting the rigidity and the motion precision simultaneously, a novel flexible hinge is designed and prepared so as to realize the motion output of gapless and large motion stroke.
4. The problem of inaccurate multi-objective optimization of a large amount of data in the flexible hinge parameter optimization process is solved, and compared with other optimization methods, the efficiency and the accuracy of simulation results are greatly improved.
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
(1) According to specific design requirements, the initial values of the design parameters of the flexible hinge are respectively the minimum section diameter t, the circular arc central angle theta m, the circular arc radius R and the deflection angle theta affecting the whole length of the flexible hinge, then the input parameters can be expressed as X= [ ttheta m R theta ] by using a matrix, the material properties of the flexible hinge are set, the material density rho unit kg/m 3 of the flexible hinge is respectively set, the elastic modulus E unit of the flexible hinge is MPa, the shearing model G unit of the material is MPa, and the actually required target conditions are set;
(2) Calculating information such as the length of the flexible hinge, wherein the flexible hinge structure parameter is shown in fig. 3, the flexible hinge thickness can be expressed as t (x) =t+2r (1-cos θ), the total length of the flexible hinge with the length of θ e [ - θ m,θm ] can be expressed as l=2rsin θ m +2t (x), the flexible hinge flexible part length can be expressed as l=2rsin θ m, and finally, the initial value of the parameter is structurally modeled by using UG software; wherein fig. 3 is a structural parameter diagram of the flexible hinge, the parameters are a minimum section diameter t, a circular arc central angle θ m, a circular arc radius R, and a deflection angle θ affecting the overall length of the flexible hinge, the thickness of the flexible hinge can be expressed as t (x) =t+2r (1-cos θ), θ e [ - θ m,θm ], the total length of the flexible hinge can be expressed as l=2rsin θ m +2t (x), and the flexible portion length of the flexible hinge can be expressed as l=2rsin θ m;
(3) Referring to fig. 4, the compliance of the flexible hinge in each direction is deduced by using the second theorem of the material mechanics, and is respectively simplified into a form that deformation of the flexible hinge can be written as a matrix, delta represents deformation in six directions, C represents translation and torsion matrices in six directions, F represents force and moment matrices applied in six directions, and the relationship is represented as follows:
Δ＝CF；
Wherein Δ＝[Δx Δy Δz Δαx Δαy Δαz]TF＝[Fx Fy Fz Mx My Mz]T; is a graph of the stress on a flexible hinge, assuming that the flexible hinge is stressed at one end and at the other end.
The axisymmetric properties of the flexible hinge are as follows:
and G and E are fixed values, and the flexibility in three directions is in a linear relationship.
And C αz-Fy＝Cay-Fz＝Cy-Mz＝Cz-My is used for the preparation of the medicine,
The compliance matrix is mainly influenced by three variables of C ay-MyCαz-Fy and C x-Fx and is uniformly rewritten into C 1＝Cay-MyC2＝Cαz- FyC3＝Cx-Fx;
The theoretical calculation formula is obtained:
wherein f1 f 2f 3 is a parameter matrix related to the minimum section diameter t, the central angle θm of the arc and the radius R of the arc, respectively
Wherein the method comprises the steps of
(4) Calculating a finite element model under different structural parameters by changing a design parameter X= [ tθ m Rθ ], using a finite element method to obtain a flexibility matrix C as shown in figure 5, determining three coefficients by using a control variable method, changing parameters to be changed by using a progressive method, and obtaining the flexibility matrix according to the finite element model under different structural parameters; fig. 5 is a finite element model diagram of a flexible hinge, wherein one end of the flexible hinge is fixed for modal analysis, and the flexibility of each direction is calculated.
(5) Designing a neural network model, taking different design parameters X= [ tθ m Rθ ] as the input of the neural network model, taking different flexibility matrixes obtained by finite element software as the output of the model, and carrying out a system identification program to obtain the influence condition of different structural parameter values on the flexibility matrixes:
(5.1) selecting RBF neural network structure as the neural network model structure model;
(5.2) regularizing the finite element analysis data obtained in the step (4) and dividing the data into a training set and a test set, wherein the training set generally occupies 70% of the total data, and the test set occupies 30%;
And (5.3) initializing the RBF neural network, and respectively giving initial values To the input layer weight Wih, the hidden layer weight Who, the output layer weight Who, the hidden layer threshold Th and the output layer threshold To.
(5.4) The input layer of the neural network model is a minimum section diameter t, a circular arc central angle θ m, a circular arc radius R, and a deflection angle θ affecting the overall length of the flexible hinge, and the output layer of the neural network is a compliance matrix Crbf, wherein C rbf can be expressed as:
(6) Correcting the theoretical model obtained in the step (2) by utilizing the neural network model obtained in the step (5) to obtain a more accurate theoretical model;
(7) Based on the theoretical optimization model obtained in the step (6), constraint conditions are obtained according to actual design requirements:
(7.1) design of rotation capability of the flexible hinge, wherein the movement of the flexible hinge is realized by means of elastic deformation of the unit, when the deformation exceeds the elastic deformation range of the material, plastic deformation occurs, the material is permanently deformed, and the movement condition is unpredictable, so that the flexible hinge is limited to a limited movement space. The turning capability may be defined as the rotational angle θt about the y-axis with a unit torque in the y-direction, with a greater θt indicating a greater range of hinge rotation. θ t＝C1 the design rules of the flexible hinge dictate that the flexible hinge requires a strong rotational capability so that C1 should be kept as large as possible.
(7.2) Relative compliance design, the compliant hinge deformation is achieved by increasing compliance in the direction of motion, but an increase in compliance in the desired direction will also result in an increase in compliance in the direction of perpendicular motion, such that the actual deformation deviates from the desired direction, the actual deformation of the structure tends to be complex, so introducing the concept of relative stiffness, it is apparent that the more stable the structure is when ρ is greater.
The general optimization objective is max (C 1)max(C2)min(C3);
In addition, the maximum value and the minimum value of the parameter X= [ tθ m Rθ ] are constrained according to the actual situation, and the numerical value of the structural size is constrained.
(8) And optimizing the input structural parameter X= [ tθ m Rθ ] by using a multi-objective optimization algorithm based on deep reinforcement learning to obtain the optimal flexible hinge size parameter.
Referring to fig. 6, a flowchart of a multi-objective optimization algorithm design is shown, and the steps of the multi-objective optimization algorithm design for deep reinforcement learning are as follows:
(8.1) decomposing the optimization target into a plurality of sub-problems based on a decomposition strategy;
(8.2) modeling the sub-problems obtained by decomposition by utilizing a neighborhood-based parameter migration strategy and sub-problems based on a deep neural Network, and modeling the sub-problems by adopting a Pointer Network model of a Pointer Network;
(8.3) training the model based on a reinforcement learning algorithm;
(8.4) after the optimization algorithm model is completed, optimizing the flexible hinge structure parameters according to constraint conditions;
(8.5) obtaining the optimal structural parameters under the constraint condition, and outputting the specific structural parameters of the flexible hinge;
(8.6) obtaining a structural parameter model of the flexible hinge.
Example 3
A storage medium storing a program file capable of implementing any one of the above-described multi-objective optimization-based arc-shaped flexible spherical hinge design methods.
Example 4
The processor is used for running a program, wherein the program runs to execute any one of the arc-shaped flexible spherical hinge design methods based on multi-objective optimization.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
In the foregoing embodiments of the present invention, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed technology may be implemented in other manners. The system embodiments described above are merely exemplary, and for example, the division of units may be a logic function division, and there may be another division manner in actual implementation, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method of the various embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which are intended to be comprehended within the scope of the present invention.

Claims (7)

1. The design method of the circular arc type flexible spherical hinge based on multi-objective optimization is characterized by comprising the following steps of:
according to specific design requirements, designing initial values of parameters of the flexible hinge;
calculating structural parameters of the flexible hinge, and carrying out structural modeling according to initial values of the parameters to obtain a theoretical model;
calculating finite element models under different structural parameters by changing design parameters and utilizing a finite element method, and obtaining a flexibility matrix according to the finite element models under different structural parameters;
designing a neural network model, taking different design parameters as the input of the neural network model, and taking different flexibility matrixes as the output of the neural network model;
Correcting the theoretical model by using the designed neural network model to obtain a theoretical optimization model;
Based on a theoretical optimization model, constraint conditions are obtained according to actual design requirements;
The initial values of the design parameters of the flexible hinge according to the specific design requirements comprise:
The design parameters are the minimum section diameter t, the circular arc central angle theta m, the circular arc radius R and the deflection angle theta of the whole length of the flexible hinge, the input parameters are expressed as X= [ tθ m R theta ] by a matrix, the material properties of the flexible hinge are set, the material properties are respectively the material density rho unit kg/m 3 of the flexible hinge, the elastic modulus E unit of the flexible hinge is MPa, the shearing model G unit of the material is MPa, and the actually required target conditions are set;
Calculating structural parameters of the flexible hinge, and performing structural modeling according to initial values of the parameters, wherein obtaining a theoretical model comprises:
Calculating length information of a flexible hinge, wherein the thickness of the flexible hinge is expressed as t (x) =t+2R (1-cos theta), theta epsilon [ -theta m,θm ], the total length of the flexible hinge is expressed as L= Rsin theta m +2t (x), the length of a flexible part of the flexible hinge is expressed as l= Rsin theta m, and finally, using UG software to carry out structural modeling on initial values of parameters;
The constraint condition obtaining method based on the theoretical optimization model according to the actual design requirement comprises the following steps:
The rotation capacity of the flexible hinge is designed, wherein the rotation capacity is defined as that the larger the rotation angle theta t,θt around the y axis under the action of the unit torque in the y direction is, the larger the rotation range of the hinge is; θ t＝C1, the design rule of the flexible hinge indicates that the flexible hinge needs a strong rotation capability so that C 1 should be kept as large as possible;
The relative flexibility design introduces the concept of relative stiffness, and it is obvious that the more stable the structure is when ρ is larger;
the optimization target is max (C 1)max(C2)min(C3);
And constraining the maximum value and the minimum value of the parameter X= [ tθ m Rθ ] according to the actual situation, and constraining the numerical value of the structural size.
2. The multi-objective optimization-based arc-shaped flexible spherical hinge design method according to claim 1, further comprising:
and optimizing structural parameters by utilizing a multi-objective optimization algorithm based on deep reinforcement learning to obtain the optimal flexible hinge size parameters.
3. The method for designing the circular arc type flexible spherical hinge based on the multi-objective optimization according to claim 1, wherein the calculating the finite element model under different structural parameters by changing the design parameters and using the finite element method, and obtaining the flexibility matrix according to the finite element model under different structural parameters comprises:
The compliance of the flexible hinge in all directions is deduced by using a second theorem of material mechanics, the deformation of the flexible hinge is written into a matrix form, delta represents the deformation in six directions, C represents the compliance matrix, F represents the force and moment matrix born in the six directions, and the relation is expressed as follows:
Δ＝CF；
The method comprises the steps of calculating a finite element model under different structural parameters by changing design parameters X= [ tθ m Rθ ], obtaining a flexibility matrix C by using a finite element method, determining three coefficients by using a control variable method, changing parameters to be changed by using a progressive method, and obtaining the flexibility matrix according to the finite element model under different structural parameters.
4. The method for designing a circular arc type flexible spherical hinge based on multi-objective optimization according to claim 3, wherein the designing the neural network model, taking different design parameters as the input of the neural network model, and taking different flexibility matrices as the output of the neural network model comprises:
designing a neural network model, taking different design parameters X= [ tθ m Rθ ] as the input of the neural network model, taking different flexibility matrixes obtained by finite element software as the output of the neural network model, and carrying out a system identification program to obtain the influence condition of different structural parameter values on the flexibility matrixes.
5. The method for designing a circular arc type flexible spherical hinge based on multi-objective optimization according to claim 4, wherein the correcting the theoretical model by using the neural network model obtained by design, to obtain the theoretical optimization model comprises:
And correcting the theoretical model by using the obtained neural network model, and regularizing the data to obtain a more accurate theoretical optimization model.
6. A multi-objective optimization-based arc-shaped flexible spherical hinge design device, which is characterized by adopting the multi-objective optimization-based arc-shaped flexible spherical hinge design method according to any one of claims 1-5; comprising the following steps:
the parameter initial value design unit is used for designing a parameter initial value of the flexible hinge according to specific design requirements;
the structure modeling unit is used for calculating the structure parameters of the flexible hinge and carrying out structure modeling according to the initial values of the parameters to obtain a theoretical model;
The flexibility matrix acquisition unit is used for obtaining a flexibility matrix according to the finite element model under different structural parameters by changing design parameters and calculating the finite element model under different structural parameters by utilizing a finite element method;
The neural network model design unit is used for designing a neural network model, taking different design parameters as the input of the neural network model, and taking different flexibility matrixes as the output of the neural network model;
The theoretical optimization model acquisition unit is used for correcting the theoretical model by utilizing the designed neural network model to obtain the theoretical optimization model;
and the constraint condition acquisition unit is used for obtaining constraint conditions according to actual design requirements based on the theoretical optimization model.
7. The multi-objective optimization-based circular arc type flexible spherical hinge design device according to claim 6, further comprising:
And the structural parameter optimization unit is used for optimizing structural parameters by utilizing a multi-objective optimization algorithm based on deep reinforcement learning to obtain the optimal flexible hinge size parameters.
CN202210967972.2A 2022-08-12 2022-08-12 Multi-objective optimization-based arc-shaped flexible spherical hinge design method and device Active CN115292845B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210967972.2A CN115292845B (en) 2022-08-12 2022-08-12 Multi-objective optimization-based arc-shaped flexible spherical hinge design method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210967972.2A CN115292845B (en) 2022-08-12 2022-08-12 Multi-objective optimization-based arc-shaped flexible spherical hinge design method and device

Publications (2)

Publication Number Publication Date
CN115292845A CN115292845A (en) 2022-11-04
CN115292845B true CN115292845B (en) 2024-04-23

Family

ID=83827267

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210967972.2A Active CN115292845B (en) 2022-08-12 2022-08-12 Multi-objective optimization-based arc-shaped flexible spherical hinge design method and device

Country Status (1)

Country Link
CN (1) CN115292845B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104268324A (en) * 2014-09-17 2015-01-07 华南理工大学 Multi-DOF (degree of freedom) topological optimization method for compliant parallel mechanism
CN106407582A (en) * 2016-09-28 2017-02-15 中国工程物理研究院总体工程研究所 High-efficiency calculating method used for static and dynamic analysis of flexible mechanism
WO2017215217A1 (en) * 2016-06-16 2017-12-21 华南理工大学 Topology optimization design method for flexible hinge
WO2021159569A1 (en) * 2020-02-10 2021-08-19 Dalian University Of Technology Method for engineering structural fiexibility identification considering the impact of noise

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104268324A (en) * 2014-09-17 2015-01-07 华南理工大学 Multi-DOF (degree of freedom) topological optimization method for compliant parallel mechanism
WO2017215217A1 (en) * 2016-06-16 2017-12-21 华南理工大学 Topology optimization design method for flexible hinge
CN106407582A (en) * 2016-09-28 2017-02-15 中国工程物理研究院总体工程研究所 High-efficiency calculating method used for static and dynamic analysis of flexible mechanism
WO2021159569A1 (en) * 2020-02-10 2021-08-19 Dalian University Of Technology Method for engineering structural fiexibility identification considering the impact of noise

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
直梁圆角形柔性铰链的柔度矩阵分析;赵磊;巩岩;华洋洋;;中国机械工程(第18期);全文 *

Also Published As

Publication number Publication date
CN115292845A (en) 2022-11-04

Similar Documents

Publication Publication Date Title
Armanini et al. Soft robots modeling: A structured overview
Zhang et al. Topology optimization of compliant mechanisms
Megaro et al. A computational design tool for compliant mechanisms.
Zhan et al. Error modelling and motion reliability analysis of a planar parallel manipulator with multiple uncertainties
Fang et al. Kinematics of soft robots by geometric computing
Lou et al. Optimization algorithms for kinematically optimal design of parallel manipulators
Wen et al. Conicity and cylindricity error evaluation using particle swarm optimization
CN111752151A (en) Adaptive tracking and compensating method and system for grinding and polishing industrial blade
Sun et al. Enhancing torsional stiffness of continuum robots using 3-D topology optimized flexure joints
Gilbert On the mathematical modeling of slender biomedical continuum robots
CN118013865B (en) Dynamic parameterization modeling and rigidity reinforcing method for aviation mobile industrial robot
CN114714348A (en) Industrial robot absolute positioning precision improving method
CN115292845B (en) Multi-objective optimization-based arc-shaped flexible spherical hinge design method and device
da Silva et al. Computer-aided integrated design for machines with varying dynamics
Shabana et al. Relative orientation constraints in the nonlinear large displacement analysis: application to soft materials
Karmakar et al. Forward kinematics solution for a general Stewart platform through iteration based simulation
Lee et al. Smooth boundary topology optimization using B-spline and hole generation
Feng et al. Calibration of a six-DOF parallel manipulator for chromosome dissection
Avilés et al. A procedure for the optimal synthesis of planar mechanisms based on non‐linear position problems
Yang et al. Optimal design and tracking control of a superelastic flexure hinge based 3-PRR compliant parallel manipulator
Dao Cuckoo search algorithm: statistical-based optimization approach and engineering applications
Yu et al. A GPS-based force rendering model for virtual assembly of mechanical parts
Tran et al. A new optimal design synthesis method for flexure-based mechanism: recent advance of metaheuristic-based artificial intelligence for precision micropositioning system
Zhang et al. Differentiable Surrogate Models for Design and Trajectory Optimization of Auxetic Soft Robots
Chen et al. Joint equivalence design and analysis of a tensegrity joint

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant