Nowadays the increasing demand for highly effective cooling devices involving radiative-convective porous fin heat sink with functionally graded material (FGM) has gained intense research attention due to their extensive use in industrial, commercial and strategic types of equipment. The present study aims to introduce a novel application of stochastic numerical computing by exploitation of Levenberg–Marquardt backpropagation (LMB) competency for performance analysis of heat sink of functionally graded material of porous fin. The dataset for LMB is generated through the shooting method for the system dynamics by Thermo-geometric variation of conduction-radiation, conduction-convection and radiation. The process of training, testing and validation are employed for network modeling with LMB procedure for different scenarios of porous fin model. The accuracy of the results is analyzed by absolute error, mean square error, error histogram and regression measures for exhaustive numerical simulation studies of proposed LMB to study the thermal performance efficiency of the porous fin heat sink involving non-homogeneous index B of FGM.

如今，对涉及具有功能梯度材料 (FGM) 的辐射对流多孔翅片散热器的高效冷却装置的需求不断增长，由于它们在工业、商业和战略类型的设备中的广泛使用而引起了广泛的研究关注。本研究旨在通过利用 Levenberg-Marquardt 反向传播 (LMB) 能力来介绍随机数值计算的新应用，用于多孔翅片功能梯度材料散热器的性能分析。LMB的数据集是通过传导-辐射、传导-对流和辐射的热几何变化通过系统动力学的拍摄方法生成的。针对多孔翅片模型的不同场景，采用 LMB 程序对网络建模进行了训练、测试和验证过程。