A Neural Network Method for Approximating Preisach’s Model of Magnetic Hysteresis Using Particle Swarm Optimization

This paper describes approximating the discrete Preisach model of hysteresis using an artificial neural network with two different training algorithms. The particle swarm optimization algorithm is the main point of focus and the Adam gradient descent algorithm is used as a basis for comparison. The two optimization algorithms were used to train a feedforward neural network and the training data consists of a major hysteresis loop constructed with a pair of hyperbolic tangent functions.