Pseudoprospective forecasting of failure time.

Area of Science:

• Engineering Mechanics
• Materials Science
• Geophysics

Background:

• Power-law precursory acceleration is a recognized method for predicting time to failure in materials and structures.
• A significant challenge in blind prediction is the unknown form of the power-law exponent.

Purpose of the Study:

• To develop a novel method for time to failure prediction that does not require prior knowledge of the power-law exponent.
• To establish a universally applicable forecasting technique for diverse failure phenomena.

Main Methods:

• A linear relationship with time (t) was identified for estimated failure times (t\[*]) calculated iteratively using monitored quantity updates.
• The monitored quantity was shown to be expressible as any power of the inverse rate.
• Projections of t\[*] for all exponents were found to intersect at a unique point on the t=t\[*] line, defining the failure time.

Main Results:

• A novel, exponent-independent method for time to failure prediction was demonstrated.
• The method's universality was confirmed across synthetic data, laboratory material failure experiments, and volcanic eruption data.
• The intersection point on the t=t\[*] line reliably predicts the exact failure time.

Conclusions:

• This work introduces a robust improvement in time to failure forecasting, particularly when the controlling power-law exponent is unknown.
• The findings provide a foundational basis for enhanced predictive modeling in engineering, materials science, and geophysics.