Fatigue Enhancement Mechanism Of The Omni-Axes Random Vibration Environment

With the increasing complexity of technology, the reliability of weapon equipments and mechatronic products is highly concerned. The problem of how to enhance the reliability of products has become the bottleneck to achieve the battle effectiveness of weapon equipments and the market competence of products under development.Reliability Enhancement Testing (RET) technology has been developed in western developed countries since 1990's, and has been successfully applied in the aspects of aeronautics, astronautics, national defense and etc. RET is accomplished by extreme environmental stresses to stimulate design and manufacture deficiency of products, then to modify and improve the reliability of products. Besides its advanced concept, RET greatly depends on the related equipments that provide extreme stresses environment. At present RET adopts a new type of Ultra High Stresses (UHS) system which can provide temperature and humidity environment, especially the omni-axes random vibration environment generated by Repetitive Shock (RS) of pneumatic hammers, which has extra high efficiency to stimulate failure, especial fatigue failure of products in engineering.Due to rather few reports about mechanism study of RET in abroad, this technology has been blocked in the interior. With the aid of the related key project of armament, this dissertation has studied the fatigue failure enhancement mechanism of omni-axis environment of the RS machine which is the most advanced equipment of RET in the interior. This dissertation is about to resolve what the omni-axis environment is and how it can efficiently stimulate fatigue failure. The study achievement will provide the theoretical instruction for RET spread of in the interior, and will form the basement for the development and optimizing design of RET equipments.In order to resolve the enhancement mechanism to stimulate fatigue failure of the omni-axes random vibration environment, this dissertation firstly studies the omni-axes random vibration environment, then describes it by two distinct characteristics: (1)single axis vibration is non-Gaussian, wideband and cyclostationary ; (2) multi-axes vibration is non-proportion simultaneous, then deduces a new analysis method of frequency domain to estimate fatigue life and damage under non-Gaussioan random stresses. The main content and conclusions of the dissertation are as following:1. Based on the sampled vibration signal of omni-axes random vibration environment, this dissertation studies its characteristics by related vibration description methods:(1) Cyclostationary of the RS vibration is validated by statistic method. The analysis methods in this dissertation is useful to other similar engineering data processing.(2) The two distinct characteristics of RS machine are summarized: (1)single axis vibration is super-Gaussian, wideband and cyclostationary; (2)multi-axial vibration is non-proportion simultaneous.(3) RS machine has more advantages in spectral bandwidth, peak probability distribution, loading rate and multi-axial vibration than ED. But RS machine has shortcoming, such as spectral valley and energy shortage in low band, uncontrollable frequency spectrum and so on. which indicates the possibility to improve the RS machine.2. Based on spectral fatigue life estimate theory and Hermite polynomial extension of non-Gaussian possibility density function, a new method was theoretically deduced to estimate fatigue life under non-gaussian random stresses, and was demonstrated with numerical examples under Gaussian stress, sub-Gaussian stress and super-Gaussian random stresses. The method is applicable to fatigue estimate under stresses whether it is Gaussian or non-Gaussian , and without frequency bandwidth limitation.3. This dissertation illustrated the fatigue failure enhancement mechanism of omni-axes random vibration environment from its characteristic of single axis vibration.(1) Based on theoretic method above, an inequation were deduced for random stresses of the equivalent level: super-Gaussian > Gaussian > sub-Gaussian.