Life Prediction of Tantalum Capacitors Based on PSO-GM Model

Life prediction for Tantalum (Ta) capacitors is difficult by using conventional time-to-failure analysis method. Degradation analysis, which deals with parameters of performance degradation, is an efficient method to estimate reliability for highly reliable parts like Ta capacitors. Based on grey theory, the GM(1,2) model, a degradation analysis method for life prediction was proposed. Since the accuracy of GM(1,2) model was influenced deeply by parameters, weight parameter  was optimized by particle swarm optimization method(PSO), and PSO-GM model was established. The practical test showed that the prediction errors were reduced by using the proposed PSO-GM model. It indicates that the proposed method is valid and accurate.

[1]  Alexander Teverovsky,et al.  Degradation of leakage currents in solid tantalum capacitors under steady-state bias conditions , 2010, 2010 Proceedings 60th Electronic Components and Technology Conference (ECTC).

[2]  Peng Wang,et al.  Reliability prediction based on degradation modeling for systems with multiple degradation measures , 2004, Annual Symposium Reliability and Maintainability, 2004 - RAMS.

[3]  Peter W. Tse,et al.  Anomaly Detection Through a Bayesian Support Vector Machine , 2010, IEEE Transactions on Reliability.

[4]  Jiaoying Huang,et al.  Nonintrusive Pressure Measurement With Capacitance Method Based on FLANN , 2010, IEEE Transactions on Instrumentation and Measurement.

[5]  Wenzhan Dai,et al.  Improvement of background value and its application in non-equidistance GM(1,1) modeling , 2008, 2008 7th World Congress on Intelligent Control and Automation.

[6]  Gisela Hartler,et al.  Parameter Estimation for the Arrhenius Model , 1986, IEEE Transactions on Reliability.

[7]  Jiaoying Huang,et al.  Lifetime prediction for tantalum capacitors with multiple degradation measures and particle swarm optimization based grey model , 2012 .

[8]  Zhaoqin Peng,et al.  Capacitance Method and Data Fusion in Non-Intrusive Pressure Measurement , 2011 .

[9]  Wentian Ji,et al.  Particle Swarm Optimization Based GM(1,2) Method on Day-Ahead Electricity Price Forecasting with Predicted Error Improvement , 2010, 2010 2nd International Workshop on Database Technology and Applications.

[10]  Wann-Yih Wu,et al.  A prediction method using the grey model GMC(1, n) combined with the grey relational analysis: a case study on Internet access population forecast , 2005, Appl. Math. Comput..

[11]  Wayne Nelson,et al.  Analysis of Accelerated Life Test Data - Part I: The Arrhenius Model and Graphical Methods , 1971, IEEE Transactions on Electrical Insulation.

[12]  Jian Zhang,et al.  Experimental Parameter Investigations on Particle Swarm Optimization Acceleration Coefficients , 2012 .

[13]  Cheng Gao,et al.  Life prediction of Tantalum capacitor based on gray theory optimization model , 2011, 2011 IEEE International Conference on Quality and Reliability.

[14]  Cheng Gao,et al.  Particle swarm optimization based RVM classifier for non-linear circuit fault diagnosis , 2012 .