Lifetime prediction for tantalum capacitors with multiple degradation measures and particle swarm optimization based grey model

Abstract

A lifetime prediction method for high-reliability tantalum (Ta) capacitors was proposed, based on multiple degradation measures and grey model (GM). For analyzing performance degradation data, a two-parameter model based on GM was developed. In order to improve the prediction accuracy of the two-parameter model, parameter selection based on particle swarm optimization (PSO) was used. Then, the new PSO-GM(1, 2, ω) optimization model was constructed, which was validated experimentally by conducting an accelerated testing on the Ta capacitors. The experiments were conducted at three different stress levels of 85, 120, and 145 °C. The results of two experiments were used in estimating the parameters. And the reliability of the Ta capacitors was estimated at the same stress conditions of the third experiment. The results indicate that the proposed method is valid and accurate.

Keywords

accelerated degradation test / capacitor / multiple degradation measure / particle swarm optimization / grey model

References

[1] FREEMAN Y, HAHN R, LESSNER P. Reliability and critical applications of tantalum capacitors [C]// CARTS Europe. Barcelona, Spain, 2007: 194–203.

[2] SotirisV. A., TseP. W., PechtM. G.. Anomaly detection through a bayesian support vector machine [J]. IEEE Transactions on Reliability, 2010, 59(2): 277-286

[3] ParkJ.-i., BaeS.-joo.. Direct prediction methods on lifetime distribution of organic light-emitting diodes from accelerated degradation tests [J]. IEEE Transactions on Reliability, 2010, 59(1): 74-90

[4] LuC. J., MeekerW. Q.. Using degradation measures to estimate a time-to-failure distribution [J]. Techno Metrics, 1993, 35(2): 161-174

[5] ListaV., GarbossaP., TomasiT.. Degradation based long-term reliability assessment for electronic components in submarine applications [J]. Micro-electronics Reliability, 2002, 42: 1389-1392

[6] DreesenR., CroesK., MancaJ.. A new degradation model and lifetime extrapolation technique for lightly doped drain nMOSFETs under hot-carrier degradation [J]. Microelectronics Reliability, 2001, 41: 437-443

[7] MarchandB., GhibaudoG., BalestraF.. A new hot carrier degradation law for MOSFET lifetime prediction [J]. Microelectronics Reliability, 1998, 38: 1103-1107

[8] GongH., WuY.-x., LiaoKai.. Prediction model of residual stress field in aluminum alloy plate [J]. Journal of Central South University of Technology, 2011, 18: 285-289

[9] REED E K. Tantalum chip capacitor reliability in high surge and ripple current applications [C]// Proceedings of the 44th CARTS. Washington D C, USA, 1994: 861–868.

[10] ZEDNICEK S, HORACEK I, PETRZILEK J, JACISKO P, GREGOROVA P, ZEDNICEK T. High CV tantalum capacitors-challenges and limitations [C]// CARTS Europe. Helsinki, Finland, 2008: 121–130.

[11] ZEDNICEK T, SIKULA J, LEIBOVITZ H. A study of field crystallization in tantalum capacitors and its effect on DCL and reliability [C]// Proceedings of the 29th CARTS. Jacksonville, FL, 2009. 5.3.1–11.

[12] LARSON D, MACDOUGALL F W, HARDY P. The impact of high energy density capacitors with metallized electrode in large capacitor banks for nuclear fusion application [C]// Proceedings of the 9th IEEE Pulsed Power Conference. Albuquerque, NM, 1993: 735–738.

[13] DengJ. L.The primary methods of grey system theory [M], 2005WuhanHuazhong University of Science and Technology Press

[14] HsuC.-c., ChenC.-yon.. Applications of improved grey prediction model for power demand forecasting [J]. Energy Conversion and Management, 2003, 44(14): 2241-2249

[15] WuW.-y., ChenS.-pei.. A prediction method using the grey model GMC (1, n) combined with the grey relational analysis: A case study on Internet access population forecast [J]. Applied Mathematics and Computation, 2005, 169(1): 198-217

[16] LiG.-d., YamaguchiD., MizutaniK.. New proposal and accuracy evaluation of grey prediction GM [J]. IEICE Transactions on Undamentals, 2007, E90-A(6): 1188-1197

[17] ZEDNICEK T, GILL J. Voltage derating results for solid tantalum and niobium capacitors [C]// CTI CARTS. Stuttgart, Germany, 2003: 44–50.

[18] ShiX. H., LiangY. C., LeebH. P.. An improved GA and a novel PSO-GA-based hybrid algorithm [J]. Information Processing Letters, 2005, 93(5): 255-261

[19] JHA G K, THULASIRAMAN P, THULASIRAM R K. PSO based neural network for time series forecasting [C]// International Joint Conference on Neural Networks, 2009: 1142–1427.

[20] FengH.-ming.. Self-generation RBFNs using evolutional PSO learning [J]. Neurocomputing, 2006, 70(1/2/3): 241-251

[21] PadgettW. J., TomlinsonM. A.. Inference from accelerated degradation and failure data based on Gaussian process models [J]. Lifetime Data Analysis, 2004, 10: 191-206

[22] ChenZ., ZhengS.. Lifetime distribution based degradation analysis [J]. IEEE Trans. Reliability, 2005, 54: 3-10