Research

In order to regulate the output voltage of power converters, there are a lot of modulation techniques such as pulse width modulation (PWM) and space vector modulation (SVM). Though they create advantages, they involve much more complexity in the calculation. So, most of them are complicated and rigid to put in. To control the voltage, Model Predictive Control (MPC) can be the best control scheme. The simple concept and fastv dynamic response are the main advantages and this scheme does not need internal current control loops and modulators. In this thesis, to regulate the output voltage of a three-phase four-leg inverter, MPC have been followed. Three-phase four-leg inverter is used instead of a three-phase three leg inverter, as it can supply power to three-phase loads under heavy, light, unbalanced, and balanced load and source conditions and it can also generate three output voltages independently with one additional leg. For the implementation of this model, we need to sense the filter current, capacitor voltage, and load current. This sensing increases the hardware implementation cost. For this reason, we are formulating a current-sensorless three-phase four-leg LC filtered voltage source inverter here. In this formulated system, we reconstructed the system model for the capacitor current and capacitor voltage. And only the capacitor voltage will be sensed that will estimate the capacitor current for the system which will eliminate the current sensor that was previously used in the typical four-leg LC-controlled VSI. Furthermore, as the output of this current sensor-less three phase four-leg inverter is better than the typical one for greater inductance, then we can make it grid-connected to supply the grid if it has an extra amount of power. This current sensor-less inverter system can also contribute to the micro-grid power system in the recent future to make the proper utilization of power.