Back propagation (BP) neural network based method for predicting service life of rubber absorber

Abstract

The invention discloses a back propagation (BP) neural network based method for predicting service life of a rubber absorber. According to the method, a BP neural network model of aging life of the rubber absorber is created firstly; according to the created BP neural network model of the aging life of the rubber absorber as well as fatigue life, a BP neural network model of the service life of the rubber absorber is created; and finally, the service life of the rubber absorber is predicted through the BP neural network model of the service life. In the method, aging factors and fatigue factors affecting the use of the rubber absorber are combined, a service life prediction model of the rubber absorber is reasonably created, the service life of the rubber absorber affected by both aging and fatigue is accurately predicted, and the defect of inaccurate prediction of service life of the rubber absorber due to separate research of aging life and fatigue life is overcome.

Description

technical field [0001] The invention relates to the technical field of life prediction of rubber shock absorbers, in particular to a method for predicting the service life of rubber shock absorbers based on BP neural network. Background technique [0002] Rubber shock absorbers have adjustable elastic parameters, can attenuate and absorb low-frequency, high-frequency vibration and noise, impact stiffness is greater than dynamic stiffness and static stiffness, and have the advantages of small size, light weight, and maintenance-free, so they are widely used in various fields. use. However, in the actual use of the rubber shock absorber, due to the fatigue damage of stress and strain and the aging of the rubber material, there is a problem of service life. The service life of the rubber shock absorber is also restricted by the aging of the shock absorber and the interaction of fatigue. . Once fatigue damage and aging of the rubber material occurs, the stiffness and damping p...

AI Technical Summary

Problems solved by technology

Whether it is based on "HG-T3087-2001 Rapid Determination Method for Storage Period of Static Seal Rubber Parts" and "GBT 20028-2005 Application of Arrhenius Diagram to Estimate Life and Maximum Service Temperature of Vulcanized Rubber or Thermoplastic Rubber", the Arrhenius equation is used to predict In the Arrhenius equation, the activation energy E is regarded as an invariant at each temperature, and the selection of a value in the empirical formula also greatly affects the prediction results, which cause the distortion of the rubber material life prediction life model question

At the same time, for rubber products such as rubber shock absorbers, the elongation at break of rubber samples is directly used to characterize the performance of rubber products, which will lead to inaccurate prediction results of the overall aging life of rubber products.

[0005] When studying the fatigue life of rubber, the crack growth energy criterion based on fracture mechanics is mostly used to characterize the fatigue life of rubber, but the research is based on the standard specimen of crack growth. The fatigue life of the product has certain limitations

Images