A Model Prediction Method for the Electromagnetic Force of an Axial Flux Eddy Current Brake Based on the ISSA Algorithm

Solving an accurate electromagnetic force calculation model is crucial for establishing a dynamic digital simulation model of the entire system of an axial flux eddy current brake (ECB). However, the electromagnetic force accuracy of the 2-D analytical model is relatively low, while the 3-D analytical model is difficult to solve directly. To address this issue, this article proposes a method for predicting the electromagnetic force of the axial ECB using the improved sparrow search algorithm (ISSA). First, by deriving the 2-D analytical model, the fundamental function characteristics of the electromagnetic force model are extracted, and a prediction model for the electromagnetic force with unknown parameters is established. Next, discrete data points from the 3-D finite element model (FEM) simulation are used as observation values. The proposed ISSA, which integrates genetic algorithms (GAs), cosine-sine search strategies, and t-distribution-based perturbation strategies, is applied to optimize the calculation with the sum of squared errors (SSEs) as the objective function, solving for the unknown parameters in the prediction model, and thereby establishing prediction models for electromagnetic torque and axial force. Experimental validation shows that the mean relative error (MRE) between the predicted and experimental values is within 3%, demonstrating the effectiveness of this method.