Control of switching angles for a CMLI using ANN

The field of neural networks has a history of some five decades but has found solid applications only in the past fifteen years, and the field is still developing rapidly. Thus, it is distinctly different from the fields of control systems or optimization where the terminology, basic mathematics, and design procedures have been firmly established and applied for many years. In this work, a neural network fitting tool has been used. Neural network fitting tool helps us to select data, create and train a network, and evaluate its performance using mean square error and regression analysis. Then trained neural network is used for supplying switching angles for cascaded multi-level inverter. Thus, different switching angles corresponding to a particular value of modulation index (m) are fed to CMLI by trained Neural Network. Simulated line to line output voltage and its harmonic spectrum has been calculated and compared with analytically obtained values of total harmonic distortion. Analytic results are obtained for the switching angles computed using optimization based method for elimination of lower order harmonics in MATLAB 7.10.0.