Non-Gaussian load damage equivalent test spectrum optimization method

Abstract

The invention provides a non-Gaussian load damage equivalent test spectrum optimization method. The method comprises the following operation steps of (1) calculating the FDS of an actually measured signal in a time domain, Q=10, b=4; (2) determining the equivalent test time, and calculating an initial PSD value by utilizing the FDS obtained in the step (1) and a formula (13); (3) calculating FDS in a frequency domain by using the PSD and a frequency domain method, wherein the frequency domain method assumes that a response stress peak value obeys Rayleigh distribution; (4) comparing the FDS calculated in the step (1) with the FDS calculated in the step (3), and if a large error exists, performing iterative optimization on the initial PSD; and (5) calculating the FDS in the frequency domainby using the PSD optimized in the step (4), comparing the FDS with the FDS in the step (1), and obtaining a final damage equivalent test spectrum when the error meets the requirement. According to the method, damage equivalent test spectrum precision of the non-Gaussian load can be effectively improved, over-test and under-test of the product are avoided, and evaluation precision of the reliability and fatigue life of the product is improved, so the economic benefit of a product production enterprise is improved.

Description

technical field [0001] The invention belongs to the technical field of mechanical vibration tests, in particular to a method for formulating and optimizing damage equivalent test spectra of non-Gaussian random loads in the field of mechanical vibrations. Background technique [0002] Random vibration testing is widely used to identify product design flaws and evaluate product reliability and service life. The vibration test spectrum is often described by Power Spectral Density (PSD), which implicitly assumes that the vibration load on the product obeys a Gaussian distribution. However, real random vibration loads, such as wind loads and road spectrum loads, often obey non-Gaussian distributions. Since parameters such as the kurtosis of non-Gaussian loads cannot be accelerated, there are problems of too long test time and high test cost when used for fatigue life evaluation of high-reliability and long-life products. One way to solve this problem is to evaluate the fatigue ...

AI Technical Summary

Problems solved by technology

[0030] However, the above method has the following defects: the existing technology assumes that the PSD of the measured non-Gaussian load is a smooth spectral shape within the half-power bandwidth of each single-degree-of-freedom system. When this condition is not satisfied, the equivalent The test spectrum will introduce obvious errors, resulting in over-tested or under-tested products

Images