<p>Effective decoupling between the two integral parts, i.e., speed and torque of torque-producing-component of any electro-mechanical machine is of utmost importance for its selection and implementation for any sophisticated industrial application. In this paper, a model-predictive-controller has been developed for decoupling of a 5-phase vector-controlled permanent magnet synchronous machine drive system during transient-period. Meanwhile, the effect of moment-of-inertia is also included. The superiority of the proposed controller is proved after it is comparison with two well-established controllers based on proportional-integral and adaptive-neuro-fuzzy-inference-system approaches. The proposed controller enhances the dynamic performance &amp; life of the drive system and reduces the size and cost of the inverter. All the models have been developed using MATLAB/SIMULINK (off-line environment). However, the computation burden is 1000 folds (approx.) higher, thus the signals generated are not real-time clock synchronized, makes its practical implementation difficult. Thus, all these models have been converted to real-time environment using FPGA based OPAL-RT 4510 simulator with hardware-in-loop technology. Generation of real-time clock synchronized control signals in this environment makes its practical implementation easy in contrast to its off-line version.</p>