Parameter Identification Method of DC-DC Converter Based on Neural Network

Traditional electronic converter parameter identification methods usually need a lot of training data and high precision measuring equipment, and there are precision problems for complex converter. To solve these problems, a method of parameter identification of DC-DC converter based on neural network is proposed in this paper. Based on the discrete model of DC-DC converter and the structure characteristics of traditional neural network, a new neural network construction method is proposed, and the capacitance and inductance values are predicted online by gradient descent method. The method has the advantages of easy implementation, high precision and no additional sensor. At present, the simulation result has verified the feasibility of this method, and the corresponding experiment is in progress.