Hybrid Computational Intelligence Framework With Enhanced Features Analysis to Predict the Fatigue Life of Critical Components in a PCB

The combination of data-driven insights, physics-based simulations, machine learning models, and analytical models provides an innovative solution for ensuring the reliability and lifetime of critical components in printed circuit board (PCB) assemblies under vibration loading at various boundary conditions. The present research introduces a hybrid computational intelligence framework (HCIF) that combines machine learning alongside conventional analytical models to improve the accuracy of fatigue life estimation predictions. By integrating the strengths of data-driven techniques with physics-based simulations, the framework addresses the limitations of traditional methods in capturing complex nonlinear interactions and fatigue failure under vibration loading. The innovative approach ensures a more comprehensive understanding of the factors influencing fatigue life while enhancing the precision and reliability of predictions. Employing data from finite element analysis (FEA) and experimental tests, the framework employs the random forest model to analyze complicated correlations between input parameters and predicted fatigue life. The HCIF exhibits excellent precision, with a coefficient of determination (<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${R} ^{2}$ </tex-math></inline-formula>) value of 0.89, an adjusted <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${R} ^{2}$ </tex-math></inline-formula> value of 0.88, and a low mean average error (MAE) score of 5.1, given that the target variable exceeds 100. Individual conditional expectation (ICE) plots demonstrate linear and nonlinear correlations, focusing on the threshold effect of stress on components. The framework minimizes the need for demanding physical testing, providing a powerful tool for optimizing component design and enhancing maintenance approaches. The HCIF offers the ability to significantly improve the durability and performance of critical electronic components in high-vibration conditions by accurately predicting fatigue life across various operating circumstances.