# Research on the Accelerated Fatigue Experiment Method of the Crankshaft Based on a Modified Particle Filtering Algorithm and the Fatigue Crack Growth Property

## TL;DR

This paper introduces a new method to shorten crankshaft fatigue experiments using a modified particle filtering algorithm for accurate life prediction.

## Contribution

A modified particle filtering algorithm is proposed to predict crankshaft fatigue life more accurately with less data.

## Key findings

- The modified algorithm predicts crankshaft fatigue life more accurately than traditional methods.
- Less experimental data is needed to achieve reliable predictions with the new approach.
- The method can reduce the need for full bending fatigue experiments in crankshaft design.

## Abstract

Crankshafts are among the most important parts of modern internal combustion engines. Owing to the power transmission demand, sufficiently high strength is usually necessary for the application of the component. In this paper, a new crankshaft bending experimental method was proposed to shorten the corresponding test. A modified particle filtering algorithm approach was proposed for predicting the remaining fatigue life of a crankshaft during bending fatigue experiments. The predicted fatigue life was used to replace the actual experimental results for further analysis if the accuracy requirements were fulfilled; in this way, the experimental duration was obviously shortened. The main conclusion drawn from the research is that, compared with the traditional particle filtering algorithm approach, the modified particle algorithm approach proposed in this paper can more accurately predict the remaining fatigue life of a crankshaft using less experimental data, which makes it possible to circumvent actual bending fatigue experiments of crankshafts in providing theoretical guidance for the design process.