Numerical Optimization Applying Trust Region Algorithm to Optimize Parameters Related to Charge Transport Model in LDPE

Khaled Hallak (1) , Fulbert Baudoin (1) , Virginie Griseri (1) , Florian Bugarin (2) , Stéphane Segonds (2)

1 LAPLACE-DSF - Diélectriques Solides et Fiabilité
2 ICA - Institut Clément Ader

Recently, a numerical model has been developed for describing bipolar charge transport with two levels of traps (shallow and deep traps) in low density polyethylene (LDPE). This paper presents an optimization method, which is based on trust region algorithm. However, the enterprise has to be done with laying out the determination of physical parameters pertaining to bipolar charge transport model. Experimental data that have been used are net density charge measured by the pulse electro acoustic method (PEA) along with what is known as external charging current measurements. Simulated data are computed by the bipolar charge transport model, which is developed for LDPE. In essence, this paper sheds the light on the advanced algorithm that can give rise to a unique set of parameters for the model. Parameters will be reckoned using two different cost functions depending on current density and charge density data's. Conclusions on the best-to-use cost function are presented.

Mots clés: Density measurement; Computational modeling; Cost function; Numerical models; Current density; Convergence; Charge measurement

Domaines: Optimisation et contrôle [math.OC]; Traitement du signal et de l'image [eess.SP]; Robotique [cs.RO]

Khaled Hallak, Fulbert Baudoin, Virginie Griseri, Florian Bugarin, Stéphane Segonds. Numerical Optimization Applying Trust Region Algorithm to Optimize Parameters Related to Charge Transport Model in LDPE. IEEE Transactions on Dielectrics and Electrical Insulation, 2020, 27 (6), pp.2048-2055. ⟨10.1109/TDEI.2020.008876⟩. ⟨hal-03118105⟩