Implementation of Machine Learning in Structural Reliability Analysis

Reliability is a probabilistic measure of structural safety. In Structural Reliability Analysis (SRA), both loads and resistances are modelled as probabilistic variables, and the failure of structure occurs when the total applied load is larger than the total resistance of the structure. The probability distribution of the loads as well as the resistance can depend upon multiple variables. Considering all these factors, the probability of failure of a structure is calculated.SRA can be used for systematic adjustment of structural safety factors, and for the probabilistic design and operation of structures. For example, SRA can be used to design a structure to operate during the desired lifetime safely, or it can be used for maintenance scheduling of structural systems to prevent potential failures. Machine learning (ML) is the study of computer algorithms that can improve automatically through experience and by the use of data. Machine learning and statistics are closely related fields in terms of methods, but distinct in their principal goal: statistics draws inferences from a sample, while machine learning finds generalizable predictive patterns. ML methods can be applied to analytical and numerical SRA methods, such as First/Second-Order Reliability Methods (FORM/SORM) and First Order Second Moment (FOSM)

Implementation of Machine Learning in Structural Reliability Analysis Keywords:

probabilistic models , Structural reliability analysis , Machine Learning