Optimization of the fatigue life of Epoxy Molding Compounds based on BP neural network prediction model

Abstract

Based on the data from the fatigue test of Epoxy Molding Compound (EMC), firstly with a focus on the application problem of the instability between fitting and prediction error of BP neural network (BPNN), the prediction model of fatigue life for EMC materials is established. In this approach, the network structure is improved with initiative way by reducing input from the perspective of nodes with principal component analysis (PCA). Secondly, in order to deal with the problem of the bottleneck in local flow minimum of BPNN, this study tries to find out the global minimum and improves the convergence performance of the BPNN combining genetic algorithm (GA). The stability and practicality of the GABPNN model is analyzed after training and verifying, and the effect of the input factors on the output factor is studied in turn. Finally, this paper makes use of well-trained GABPNN prediction model to analyze optimum design methods of parameters to predict the fatigue life. The prediction and optimization results show that the well-trained GABPNN model can be used in the forecasting and optimizing design of the fatigue and fracture reliability of the epoxy molding compounds, and is of much practical value.

Soft computing methods for fatigue life estimation: A review of the current state and future trends

“…Article/conference paper Publications using genetic algorithms Article Ma et al, 34 Zhang and Lin, 38 Vadood et al, 43 Liu et al, 46 Lotfi and Beiss, 49 Majidian and Saidi, 56 Gao et al, 76 Canyurt, 89 Karakas et al, 90 Chen et al, 91 Kamal et al, 92 Zhang et al, 93 Agius et al 94 and Deveci and Artem 95 Conference paper Susmikanti, 17 Rohman et al, 18 Cai et al, 30 Zhaohua, 35 Tian 79 and Xu and Cui 96 prediction method for different purposes, such as the creation of constant life diagrams. Mohanty et al 101 estimated the fatigue crack growth rate of aluminium alloys (2024-T3) using GP and ANN, whereas GP provided a higher precision than ANN.…”

“…Composite Al Assaf and El Kadi, 12 Bezazi et al, 13 Salmalian et al, 14 Salmalian et al, 16 Rohman et al, 18 Aymerich and Serra, 21 Junior et al, 26 Vassilopoulos et al, 28 Mathur et al, 29 Cai et al, 30 Liao et al, 31 Al-Assadi et al, 32 Kumar et al, 33 Uygur et al, 41 Xiang et al, 42 Vadood et al, 43 Al-Assaf, and El Kadi, 53 Vassilopoulos et al, 55 Xiao et al, 59 Al-Assadi et al, 61 El Kadi, 62 Tapkin, 68 Moghaddam et al, 69 Yan et al, 72 El Kadi and Al-Assaf, 74 El Kadi and Al-Assaf, 77 Lee et al, 78 Deveci and Artem, 95 Vassilopoulos et al, 100 Azarhoosh et al, 102 Ertas, 111 Ertas and Sonmez, 117 Ertas and Sonmez, 118 El Kadi et al, 125 Deveci and Artem 130 and Sai et al 131 Alloys Figueira Pujol and Andrade Pinto, 15 Susmikanti, 17 Venkatesh and Rack, 22 Pleune and Chopra, 23 Sohn and Bae, 24 Genel, 25 Marquardt and Zenner, 27 Zhaohua, 35 Xu et al,…”

“…Benchmarking Marquardt and Zenner, 27 Mathur et al, 29 Liao et al, 31 Artymiak et al 51 and Kim et al 133 Case study Yang et al, 1 Al Assaf and El Kadi, 12 Bezazi et al, 13 Salmalian et al, 14 Figueira Pujol and Andrade Pinto, 15 Salmalian et al, 16 Rohman et al, 18 Kong et al, 19 Han, 20 Aymerich and Serra, 21 Venkatesh and Rack, 22 Pleune and Chopra, 23 Sohn and Bae, 24 Genel, 25 Junior et al, 26 Vassilopoulos et al, 28 Cai et al, 30 Al-Assadi et al, 32 Kumar et al, 33 Ma et al, 34 Zhaohua, 35 Xu et al, 36 Barsoum et al, 37 Zhang and Lin, 38 Mohanty et al, 40 Uygur et al, 41 Xiang et al, 42 Vadood et al, 43 Mishra et al, 44 Mohanty, 45 Liu et al, 46 Martinez and Ponce, 47 Barbosa et al, 48 Lotfi and Beiss, 49 Razzaq et al, 50 Srinivasan et al, 52 Al-Assaf and El Kadi, 53 Park and Kang, 54 Vassilopoulos et al, 55 Majidian and Saidi,…”

Investigation on cracked inductors based on finite element analysis

“…To optimize the entire device, besides the above methods of modifying structural dimensions and changing filler plastic, the existing optimization design method of multivariate nonlinear [5,6] can be used to synthetically optimize parameters of various materials and size parameters of the device and find out the best materials and sizes that are suitable for the current process.…”

“…Many improvements can be drawn on, for example, there are optimization methods on dimensions, materials and process [1][2][3][4]. In addition, there is integrated optimization design method for multivariate nonlinear [5,6].…”