Reliability Analysis On Kinematics Accuracy Of Gear Mechanism

With the development of modern mechanical systems towards high speed,light type,and high precision,the elastic deformation and the vibration of components have attracted sufficient attention.The elastic deformation and vibration will increase the error of the real motion and the expected motion of the mechanical system which leads to the reduction of the motion accuracy and reliability of the mechanical system.In the usual mechanical system,the gear mechanism is an important device for motion and power transmission.And its dynamic performance directly affects the motion accuracy of the entire mechanical system.As an elastic system,in addition to being subject to dynamic excitation,its structural parameters,material parameters and other parameters are inevitably random,which adds the controlling difficulty of its motion accuracy.Therefore,the dynamic transmission error of gear mechanism is analyzed from the perspective of improving dynamic kinematic accuracy.And combined with neural network,uncertainty modeling,instantaneous reliability analysis,sensitivity analysis and time-dependent reliability analysis are carried out.It mainly includes the following aspects:1.The kinematic error analysis of gear mechanism.Error excitation and stiffness excitation are the internal incentive that can not be ignored in gear mechanism.Based on the deformation of gear engagement,the static transmission error and comprehensive stiffness of gear meshing are derived.And the error excitation is expressed in the form of the sum of sinusoidal signal and random error.For the torsional vibration model of the gear pair,the fourth-order RungeKutta algorithm is used to solve the dynamic transmission error of the gear mechanism.2.BP network model for dynamic transmission error of gear mechanism.In order to obtain the functional relationship between the dynamic transmission error and random variables such as structural parameters,random errors,and material properties of the gear mechanism,a neural network is constructed and trained so that it can express the unknown function.Then the agent model of the dynamic transmission error is obtained.3.The instantaneous reliability and sensitivity analysis of gear mechanism.Based on BP network model of dynamic transfer error and considering the randomness of variables in the model,the uncertainty modeling of dynamic motion accuracy of gear mechanism is carried out.The instantaneous reliability of the motion accuracy of the gear mechanism is solved by using the first-order reliability method(FORM).And based on the sensitivity analysis,the influence degree of the distributed parameter of random variables on the dynamic kinematic accuracy of gear mechanism is studied.4.Time-dependent reliability analysis of gear mechanism.In order to evaluate the reliability of the motion accuracy of the gear mechanism within a specified operating time,the time variable is introduced into the instantaneous reliability model of gear mechanism to establish the time-dependent reliability analysis model of gear mechanism.The envelope method is applied to solve the time-dependent reliability of dynamic transmission error of gear mechanism,and Monte Carlo simulation is applied to verify the model and method proposed in this paper.In conclusion,based on the dynamic modeling and analysis of gear mechanism,the agent model of dynamic motion error of this kind of mechanism is established by using BP neural network,and the instantaneous reliability and time-dependent reliability analysis models of gear mechanism are established based on the agent model.Then,first-order reliability method and envelope method are adopted to get instantaneous reliability and time-dependent reliability of the mechanism.The research provides an effective method and solution for improving the kinematic accuracy of gear mechanism.