Search

In many human cases neurons are experimentally poorly accessible, introducing the need for computer simulations to study their behavior. In order to model nerve impulses, the electrical activity of neurons is described by systems of differential equations, which in the case of this work are based on the principle of the Hodgkin and Huxley equations. These were used to create models that describe the individual parts of a neuron and, subsequently, the representation of an entire neuron. In addition to straight nerve fibers, a 2D pathway in the basal turn of the human cochlear was simulated to account for real anatomical conditions. This allowed us to study the effects of displacements of the externally stimulating electrode for real cochlear implant situations. The relative spread (RS), introduced by Verveen and representing the coefficient of variation of the cumulative Gaussian distribution, served as the central metric for stochasticity.We showed (i) that the placement of the electrode over dendritic regions leads to significantly larger RS than stimulation over soma, indicating that the site of origin of the AP is important for the RS. (ii) the inverse relationship of RS to diameter, (iii) long internodes lead to a reduction in spontaneous firing behavior, (iv) there is an inverse linear relationship between RS and myelin conductivity and (v) that if the noise transmission time (Dt) is changed, the default value to 0.0025ms of knoise has to be multiplied by the factor (0.0025/Dt)^(1/2)*knoise(0.0025).

This paper presents a hybrid model for the prediction of magnetostriction in power transformers by leveraging the strengths of a data-driven approach and a physics-based model. Specifically, a non-linear physics-based model for magnetostriction as a function of the magnetic field is employed, the parameters of which are estimated as linear combinations of electrical coil measurements and coil dimensions. The model is validated in a practical scenario with coil data from two different suppliers, showing that the proposed approach captures the different magnetostrictive properties of the two suppliers and provides an estimation of magnetostriction in agreement with the measurement system in place. It is argued that the combination of a non-linear physics-based model with few parameters and a linear data-driven model to estimate these parameters is attractive both in terms of model accuracy and because it allows training the data-driven part with comparably small datasets.