NED University Journal of Research
ISSN 2304-716X
E-ISSN 2706-5758




NOVEL APPROACHES FOR ADDRESSING EPISTEMIC UNCERTAINTY IN FRACTURE MECHANICS

Author(s): Abdolrasoul Ranjbaran1, Mohammad Ranjbaran2, Fatema Ranjbaran3, Masoud Falamaki4, Shamsedin Hashemi5, Ali Mohammad Rousta6
1 Associate Professor, Department of Civil Engineering, Shiraz University, Shiraz, Iran, Ph. +98 917 314 5501, Fax: +98 711 6473161, Email: ranjbarn@shirazu.ac.ir.
2 Assistant Professor, Department of Chemical Engineering, Yasouj University, Yasouj, Iran, Email: mranjbaran24@gmail.com.
3 Postgraduate student, Department of Mechanical Engineering, Amirkabir University, Tehran, Iran, Email: narvan.m31@gmail.com.
4 Direct, RMS Construction and Consultancy Services Pty Ltd, Australia, Email: drfalamaki@gmail.com.
5 Assistant Professor, Department of Civil Engineering, Yasouj University, Yasouj, Iran, Email: shamsodin@gmail.com.
6 Assistant Professor, Department of Civil Engineering, Yasouj University, Yasouj, Iran, Email: arousta@mail.yu.ac.ir.

https://doi.org/10.35453/NEDJR-STMECH-2024-0002.R2


Volume: XXI

No. 3

Pages: 57-70

Date: 2024

Abstract:
This paper presents a novel optimisation approach for analysing cracked structural systems, consisting of four key steps. The first step (called the Golden Derivative) derives an equation at the discontinuity specific to cracked systems. The cracked system is transformed into an equivalent intact system for analysis in the second step. The third step converts the unique cracked system equation into a finite element equation, revealing epistemic uncertainty that leads to the formulation of Persian curves (PCs). These curves are developed using the following three methods: numerical experimentation, equivalent springs and logical reasoning, with consistent outcomes validating their effectiveness. The uncertainty is linked to the crack effect equation from fracture mechanics in the final step. It is shown by applying mathematical differentiation that this crack effect and classical fracture mechanics inherently involve epistemic uncertainty. A proposed remedy addresses this issue. The derivation of PC is based on pure mathematics and logical reasoning (bridging probabilistic and deterministic methods) making the results applicable across various fields for analysing real-world data.