Wholesale Long Range Electric Motorcycle Racing Powerful Off Road Customized Electric Motorcycle Wuxi Jiangyin Xufeng Electric Bicycle Co., Ltd. , https://www.xufengrelxrides.com
1 Introduction Cold forging technology is an important part of advanced manufacturing technology. In the research of cold precision forging technology, the research of CAD/CAM and expert system is an important part of CIMS. Due to the complexity of cold forging parts and the market requirements for development cycle, part quality and cost, the research of KBE design system combining knowledge engineering and CAD has become an inevitable trend. Based on the comprehensive experience knowledge and finite element analysis results, the advanced knowledge engineering method is used to establish a gear cold cold forging process mold design system based on FEM and KBE.
The integrated frame structure of the gear cold forging KBE design system The gear cold forging KBE design system should integrate the key technologies of various knowledge engineering such as expert system and neural network to realize the effective integration of various intelligent elements. In the design of integrated systems, the framework of the system is a key issue.
Gear cold forging intelligent design system, due to the uniqueness of cold precision plastic processing, the complexity of mold design problems, the speciality of graphics processing and the ambiguity of design knowledge, artificial intelligence technology such as neural network, expert system and fuzzy logic must be adopted. Based on the above IIS ideas, a more realistic integrated intelligent design system model is established. Therefore, information on precision forgings, precision forging process design and mold design is proposed to be transferred between three complex systems, each system processing differently. Type information, division of labor and cooperation, to better complete the process of intelligent design of molds and molds.
Finite element analysis of 3 cold precision forging gears Because the gear cold forging technology is still not mature, it is still difficult to establish a gear cold forging KBE design system. Therefore, the finite element method is used to simulate the gear cold forging process. Can be used as a reliable basis.
The cold precision forging of gears is analyzed by the elastic-plastic finite element method for the typical cold-forging process of cylindrical gears.
3.1 The main theoretical hypothesis of finite element simulation (1) According to the UL (Lagrange) method, the virtual work equation can be expressed as: ΘVTjiδvi, jdV=ΘSPiδvidS ΘVWiδvidS(1)(2) The constitutive equation of Mises material based on flow theory It is: σij=Dijklepεkl(2)(3) The incompressible condition of plastic deformation is: Tji=σij-σikεki-σkjεki σikνj,k(3)(4) The interface friction model adopts Coulomb friction model: f=μfn(4) (5) The boundary constraint model is: set the mold to move in the Z direction, and the normal unit vector outside the surface of the mold that is in contact with a node N is n={n1, n2, n3}, and the relative velocity of the node N in the X coordinate For: dr={dN,dN 1,dN 2-ωP}(5)3.2 Numerical simulation and result analysis Considering the approximate symmetry of the gear parts, the 1/2 tooth of the deformed body is taken as the object of finite element simulation and will pass The vertical section of the tooth tip and the center of the root is used as the main section for studying the metal flow law. After the analysis of different cold forging processes, the closed-type extrusion-constrained hole split two-step forming method has the best formability.
It can be seen from the simulation results that the whole deformation process is roughly divided into three stages, namely, the initial stage of deformation, the filling period of the tooth cavity and the final filling period. The unit deformation force of the precision forging is about 4 to 6 times of the yield stress σs. Numerical simulation results and experimental simulations The results are consistent. Closed squeezing-constrained hole splitting two-step forming method for gear precision forging process is very beneficial for reducing load and improving corner filling capacity.
Fuzzy prediction subsystem gear cold precision forging rebound quantity prediction involves many factors such as forging material, deformation condition, forging part size, etc. Therefore, predicting springback is a relatively complicated task, and now using neural network technology to achieve springback Fuzzy prediction.
In this study, eight input single output fuzzy predictor structures are designed, among which 8 input nodes are plastic processing technology, basic contour size, mold material elastic modulus, workpiece material strength limit, workpiece material elongation, deformation degree, workpiece material. The composition (C 0.12Cr), the shape complexity coefficient, and the output node is the rebound amount.
(1) Fuzzifier design. The fuzzy quantization function of the precise quantity of the system is designed as follows. For the elastic modulus E, the upper type function is obtained: μ(E)=11 (E-1.5)E>1.51E<1.5(6) for the fuzzy quantization of the symbol quantity.
(2) Inverse fuzzer design. The system chooses the centroid method to perform inverse fuzzy quantization for the fuzzy decision method: u=∑ni=1μ(ui)ui∑ni=1μ(ui)(7)(3) training algorithm and prediction result analysis. When using the MBP algorithm, an impulse increment value is added to each weight correction process of the BP algorithm, namely: Wrs(t 1)=Wrs(t) ηδsyr Δrs(t)(8)Δrs(t)=α[Wrs (t)-Wrs(t-1)](9) where α is a momentum factor and 0<α<1.
Training results and relative errors obtained by MBP algorithm training.
5ES subsystem expert system technology utilizes expert experience to automate the design process. Process, mold and CAM instruction self-design can be realized in the design experience knowledge application framework and rule representation.
5.1 Forging design sub-knowledge forging knowledge design knowledge can be expressed by production rules, the following is part of the knowledge base: RuleNo.: 2 (REM "pit diameter size design") IF (dental circle radius is Rn) AND (reference The distance from the surface to the center of the circle is Hw) The diameter of the THEN pit is: Dya=2[RnRn-(Hw ΔD)2]5.2 The process and mold design sub-knowledge forgings have complex shapes and large deformations, and adopt two-step molding. Process. From the results of the above finite element analysis, it is known that the closed-type squeeze-constrained hole split two-step forming method is the best process plan. The main task of the process design is to design the shape and shape of the pre-forging, and the mold design is mainly to design the core outline size and The cavity size is two parts.
(1) Correction of the toothed die. This system takes the displacement correction method as the basic idea, and adopts the following gear cold-forging tooth-shaped concave die correction design method. The diameter of the top circle is selected according to the lower deviation of the molded part, and the determining equation of the displacement coefficient is established accordingly: m =(z 2ha 2x)=da δ(10) where da―——the nominal diameter of the tooth tip circle, mmm―——module δ―——the deviation of the top of the tooth, mmz―——the number of teeth ha————the top of the tooth The displacement coefficient of the high-coefficient concave mold tooth shape is: the value of the elastic deformation is obtained from the prediction result of the fuzzy neural network.
(2) Process mold and electrode design knowledge. The system transforms the process and mold design experience into a fixed framework, some of which are represented by rules, and the knowledge representation technology of the sub-knowledge uses a combination of rules and frameworks. Taking the NC machining of pre-forging die electrode as an example, the CAM instruction design framework is expressed as: 5.3 subsystem reasoning mechanism ES system mainly includes domain blackboard, control blackboard, environment blackboard, input information conversion rule base, framework rule base And the framework facts library 6 parts, using the blackboard reasoning algorithm, the reasoning step is divided into 3 steps: filling the field blackboard, controlling the blackboard matching, filling the environment blackboard, and displaying it by the post-processor graphic display.
6 gear cold precision forging design system development and experimental verification Gear cold precision forging design system with VisualBasic, AutoCAD, Matlab and other languages ​​as programming tools, graphics display program using AutoCAD-based ACTIVE technology. The system interface and design results are shown in Figures 6 and 7. Experiments on the mold designed with this system, through the comprehensive inspection of the accuracy of the obtained cold precision forging gears, the precision range of No. 15 steel cold precision forged spur gears is 6-9 grades in GB10095-88 standard. In addition to the error, the accuracy of other indicators can reach 6-8.
7 Conclusion Through the research of gear forging integrated intelligent design system, we can know:
(1) It is proved by FEM analysis that the straight spur gear is formed by the closed squeezing-constrained hole split two-step forming method, and the precision forging process is reliable.
(2) The knowledge of process and mold design can be combined with rules and frameworks and the application of blackboard inference mechanism.
(3) The gear die correction can be modified by the displacement coefficient method combined with the rebound amount fuzzy prediction.
(4) The intelligent design system under the GC/GD-ES-ANN integrated model can successfully complete the design task.