Département de Génie Electrique, Ecole Nationale Supérieure d’Ingénieurs (ENSI), Centre d’Excellence Régional pour la Maîtrise de l’Electricité (CERME), Université de Lomé (UL), BP: 1515, Lomé, Togo
This paper studies the control of a DC machine in a Matlab/Simulink environment, more specifically the shunt machine. We first highlight the modeling of a shunt machine and then control it by acting on each parameter, first in open loop, then in closed loop, while studying the system’s performance. Finally, introduce the appropriate correction to improve system performance. The second part consisted in simulating the operation of the shunt-excited DC machine in a Matlab/Simulink environment. The more the electric motors are loaded, the lower the rotational speed. In order to bring the motor speed back to its nominal value, two types of control were proposed in this work: control by variation of the armature voltage U_a and control by variation of the excitation current I_e. Simulation of these two types of control, in our case using Matlab/Simulink software, showed the strengths and weaknesses of each type of control, depending on whether a PI corrector is integrated or not.
This work has enabled us to distinguish two categories of incidents which lead to interruptions in the supply of electric energy on the Lomé electricity network. These are both external and internal incidents. Internal faults mainly consist of faults on the underground and overhead cables. They represent 40.02% of incidents. These have an impact on undistributed energy and on the quality of distributed energy. Statistical processing and the Monte Carlo method are used. It has been shown that there is a correlation between the number of incidents and undistributed energy. To reduce undistributed energy, we suggest that the distributor favor double derivation and artery cutting like topology.