Switched reluctance motors (SRM) are a type of electromagnetic machine that has piqued the interest of manufacturers, as opposed to induction, brushless, or permanent magnet machines. This is because the rotor is simple, robust, and lacks coils, windings, and permanent magnets. It can also operate in a wide range of power in the electric vehicle's drive, even in extreme conditions such as underground mines, ensuring a longer life of service. However, due to the toothed shape of the rotor, the SRM is characterized by vibration and acoustic noise. To solve this problem to better adapt the SRM to the electric vehicle, we propose to use intelligent techniques such as the controller (ANN) and the fractional order controller (PI α ). This article compares two intelligent speed controllers that use direct torque control (DTC) to reduce torque ripples. As a result, when associated with direct torque control, the Fractional Order Controller (PI α ) outperforms the Artificial Neural Network (ANN).

In this paper, a steady-state output power oscillation problem is overcome using the indirect control mode based-Perturb and Observe (P&O) implementation algorithm. This can be ensured through controlling the duty cycle input of the DC-DC boost converter using the proposed Linear Quadratic Regulator (LQR) controller. Their parameters are optimized using the Grasshopper Optimization Algorithm (GOA) where a good tracking behavior of a desired Maximum Power Point (MPP) can be guaranteed for various sudden changes in weather conditions such as absolute temperature and solar irradiance. The desired performances and robustness of the closed-loop system can be achieved by the two following stages. In the first stage, the standard P&O algorithm based-direct control mode generates a reference current perturbation using both existing electrical power and measured PV current. Accordingly, a current error perturbation is provided through the discrepancy between reference and measured currents. In the second stage, the previous current error provided in the inner control-loop is mitigated as much as possible using the stabilized LQR controller. The current control-loop problem is addressed with a detailed analysis technique of averaging and linearization, in which the linearization of actual PV-boost converter system around the desired MPP allows determining the corresponding linear plant-model. This leads to well optimize the LQR controller parameters. The performance and robustness provided by the P&O algorithm based-indirect duty cycle control are shown for sudden changes in solar irradiance and absolute temperature as well as in a wide variation of the resistive load.

Boiling heat transfer process is important because it is a way to increase the flux density transmitted at low temperature differences. To quantify the thermal exchanges, we performed an experimental study of the nitrogen pool boiling, in transient conditions, on a horizontal brass ribbon for a fixed flux density. The results show that there is no break between the monophasic convection zone and the nucleated boiling region. In the nucleated boiling zone, the temperature variations are very small. We also note that the overheating required to trigger boiling increases with the time delay after the activation of nucleation sites.

No abstract

No abstract