| The rotation frame is the main load-bearing and transmission component of the "SRFT" amusement equipment.The rationality of the rotation frame structure design directly determines the running capacity,ride safety and comfort of the amusement equipment.In this paper,the rotation frame is set as the research object.Based on it,has completed the relative static calculation,dynamic characteristic analysis,parameter sensitivity analysis,structure optimization and quality analysis.The main points are as follows:1.According to the design requirements and specifications of "SRFT" amusement equipment,the strength,stiffness and local stability of the main components for the rotation frame are verified by the classical mechanics theory and finite element analysis software.The finite element software ANSYS Workbench was used to establish the overall model of the rotation frame.Through the simulation of various working conditions of the rotation frame,the working condition 2 was clarified as the most dangerous working condition.At the same time,it is found that the structure has a large margin in terms of strength and stiffness,and has certain optimization possibilities.2.The modal analysis of the rotation frame was completed by the modal analysis module of ANSYS Workbench,obtain the first 8 modal frequencies of the structure,also the harmonic response analysis of the structure is completed based on the modal analysis.It is concluded that the deformation of structure at 2.3 Hz is the largest.By studying the effect of the excitation frequency generated by the rotating main shaft and the hydraulic motor on the rotation frame,it is shown that the main shaft and the hydraulic motor never causing resonance of the structure.In this paper,the linear buckling analysis of the structure is carried out,and the buckling characteristic value of the first-order buckling instability of the structure is 22.52,which indicates that the structure has strong resistance to instability.3.In order to find out the design variables that have the most significant influence on the strength,stiffness and stability of the rotation frame,the optimal Latin hypercube sampling method is used and obtain 140 samples for 12 input parameters and 5 output parameters.The BP neural network approximation model was established by using these samples,and the training of neural network was completed by GDX training algorithm.The BP neural network model was sampled 1400 times randomly by descriptive Monte Carlo numerical simulation method.The pareto influence degree distribution map of 12 design variables for 5 design indicators was obtained,and the design variables that had the greatest influence on the design indicators was analyzed.It provide a basis for the selection of variables in structural design optimization.4.Combined with the sensitivity analysis results,six design variables witch has significant influence on the weight of the rotation frame were selected for re-sampling.By comparing the fitting precision of the three proxy models to the samples,the response surface model with relatively high fitting accuracy is used to replace the finite element model of the rotation frame,and the multi-island genetic algorithm is used to optimize the design of the structure.After optimization,the weight of the structure decreased by 14.21%,and the effect was remarkable.In order to test the reliability of the structure after optimization,the parameters such as load and material are used as the interference factors,and the 6σ mass analysis of the rotation frame is carried out by descriptive Monte Carlo sampling method.The analysis results show that the σ level of optimized structure is higher at 8σ,and the reliability is close to 100%. |
