A Data-Driven Method for Determining the Loading Spectrum in Fatigue Tests for Working Device of Engineering...

Area of Science:

• Mechanical Engineering
• Materials Science
• Computational Engineering

Background:

• The working device of wheel loaders experiences significant fatigue damage due to operational loads.
• Accurate load spectrum determination is crucial for fatigue life assessment but lacks standardized methods in construction machinery.
• Existing methods struggle with the complexity of real-world operational loads on heavy equipment.

Purpose of the Study:

• To develop a novel method for determining the load spectrum of construction machinery working devices.
• To enable reliable fatigue life assessment for guiding structural design.
• To address the challenge of non-standardized load spectrum determination in the field.

Main Methods:

• Application of random loads to a fatigue testing platform and measurement of output loads.
• Development of a numerical model using data identification techniques and the NARX (Nonlinear Autoregressive with Exogenous Inputs) model.
• Inverse determination of input loads using genetic algorithms and measured target output loads.

Main Results:

• A data-driven numerical model for the loader's working device was successfully established.
• The inverse virtual iteration technique effectively determined the input load spectrum.
• Validation confirmed the method's reliability in fatigue damage assessment by comparing output loads.

Conclusions:

• The proposed inverse virtual iteration technique, integrating NARX models and genetic algorithms, offers a reliable approach for load spectrum determination.
• This method provides a valuable reference for the fatigue life assessment and structural design of construction machinery.
• The findings contribute to improving the durability and safety of heavy construction equipment.