An Integrated Optimization Design Method of Single-Phase PV Inverter Based on Machine Learning

In order to minimize the filter mass and loss, optimize controller parameters, this paper proposes an integrated optimization design method for filter and controller of single-phase grid-connected PV inverter based on machine learning (ML). Support vector machine (SVM) is used to judge the feasibility of the filter design scheme, and artificial neural network (ANN) establishes the mapping relationship from system parameters to optimization goals. In the model, the optional capacitance data is discretized, ensuring that the results are current commercial models. Similarly, the parameters of the current loop controller are optimized by ML to obtain the minimum current ripple. In this paper, the performance comparison of the system with different filter and controller design schemes is conducted in simulation, the results indicate the validity and superiority of the proposed method.