A simplified three-dimensional space vector algorithm for three-phase four-leg converters based on classifier neural networks

Four-leg voltage source converters have successfully been used to nullify the zero-sequence current generated by unbalanced or nonlinear loads. This paper introduces an intelligent and computationally efficient neuro-computing classification algorithm for the implementation of three-dimensional Space Vector Modulation (SVM) on four-leg voltage-source inverters. The proposed technique avoids non-linear function approximations, and removes any dependency on complicated look-up tables. This advantage makes the use of the proposed technique of major gains in modern power systems and applications such as packaging technology and Application-Specific Integrated Circuit (ASIC) design. The proposed scheme is validated by analytical analysis, and simulations on a four-leg voltage-source converter.