Water turbine top cover bolt fatigue prediction method

Abstract

The invention discloses a water turbine top cover bolt fatigue prediction method, which comprises the following steps: in a time domain range, solving the gradient of a stress load to a time history for acquired stress load-time history data to obtain the power density of a stress amplitude; performing power density conversion on the S-N curve of the top cover bolt material to obtain an S-N powerdensity curve of the bolt material; in the frequency domain range, performing short-time Fourier transform on the power density of the stress amplitude of the acquired data to obtain a curve that thestress amplitude changes along with the frequency at a certain moment; in combination with a bolt material S-N power density curve, performing fatigue life prediction on the top cover bolt by applyinga linear fatigue accumulation method. According to the method, fatigue life methods of a time domain and a frequency domain are fused, the process that the loading frequency of the stress load generates tiny cracks on the bolt is accurately represented, and meanwhile, the magnitude of the loading frequency represents the limit degree of high-cycle circulation. According to the prediction method,the damage degree of the bolt can be accurately judged, and the fatigue life of the top cover bolt is predicted.

Description

technical field [0001] The invention relates to the technical field of bolt fatigue prediction, in particular to a method for predicting the fatigue of bolts on the roof of a water turbine. Background technique [0002] The top cover of the water turbine is the main part supporting the water guide mechanism and the guide bearing parts, and is also one of the main flow-passing parts of the unit. It is a welded structure, and the main flange adopts a double-plate upper flange structure. The seat ring is connected. [0003] The roof bolts need to withstand loads under various working conditions during actual operation, including low-cycle high-level loads such as start-up, shutdown, and load shedding, high-cycle loads during normal operation of the unit, and changes in working conditions. Sudden loads borne at the time. These random alternating loads will cause cracks and defects to form inside the bolts, which will cause fatigue damage to the bolts as the working conditions ...

AI Technical Summary

Problems solved by technology

[0005] The existing bolt fatigue life prediction method based on the rainflow method, because the traditional rainflow method cannot take all the full cycles in the stress-strain, and even cannot take out some full cycle meters with obvious closed loops; the rainflow counting method The cumulative damage effect of variable amplitude load is only considered in the time domain, and the effect of loading frequency is not considered. At the same time, the average value obtained by cycle counting ignores the influence of some low-cycle and large-scale loads, so there are many shortcomings and deficiencies, resulting in:

[0006] 1. The traditional rainflow method is not accurate enough for bolt fatigue prediction;

[0007] 2. Predicting the life of bolts requires a large number of time series samples, and the amount of calculation is large

Images