Control Method for Electric Vehicles' Rapid Charging Based on Artificial Neural Networks Utilizing Renewable Energy Sources for Smart Battery Thermal Management

Abstract

Sustainable or environmentally friendly transportation has been a driving force behind the development of new automobiles in response to rising public concern about air pollution. The broad adoption of battery-powered vehicles is anticipated to be accelerated by the development of rapid battery charging technologies that aim to minimize the amount of time that batteries need to be recharged. A rapid electric car charging system that is also connected to a selfsufficient DC microgrid that is powered by renewable energy sources is proposed in this research. To address the issue of power demand management in dynamic situations, this study develops a novel DNN-based control strategy that combines model predictive control, also known as MPC, with artificial neural networks (ANNs) instead of relying just on MPC. Fast direct current (dc) charging for electric vehicles (EVs) is supervised by AI (NN) that uses energy available from renewable sources with battery backup devices. A battery temperature management method is proposed to guarantee the safety and performance of EV batteries under harsh operating circumstances. After doing simulations and experiments under different operating circumstances, the results reveal that the suggested ANN MPC scheme outperforms the MPC controller. In this piece, we look at a photovoltaic-grid hybrid system that uses a water pump and an induction motor (IM). When transferring power, a straightforward dc-link voltage management method is used. When the water pump is not in use, the system's power is sent to the utility. Normally, it feeds the induction motor-driven water pump. Two current sensors & two voltage sensors are needed for this system to measure and estimate. Using a modified version of the space vectors modulation (SVM) method, the phase currents of induction motors are approximated from the dc-link current. Using a third-order integrator for flux estimate and an artificial neural network–based model reference adaptive system, this system is able to adjust the power flow according to demand and accomplish speed estimation. The speed of an IM-pump may be controlled via field-oriented control. The power transfer between the utility and the IM drive plus water pump is controlled by managing the dc-link voltage, which is achieved using a unit voltage generation method based on third-order integrators. This article discusses a photovoltaic-grid integrated system that uses an induction motor (IM) connected to a water pump. The system's appropriateness is supported by simulation results on the MATLAB/Simulink platform and by test results obtained with the aid of a prototype under different solar irradiance conditions. When transferring power, a straightforward dc-link voltage management method is used. When the water pump is not in use, the system's power is sent to the utility. Normally, it feeds the induction motor-driven water pump. Two current sensors or two voltage sensors are needed for this system to measure and estimate. Based on the direct current (dc) link current, the induction motor currents of phase are computed using a modified space matrix modulation (SVM) approach. Using a third-order integrator for flux estimate and an artificial neural network–based model reference adaptive system, this system is able to adjust the power flow according to demand and accomplish speed estimation. The speed of an IM-pump may be controlled via field-oriented control. The power transfer between the utility and the IM drive and water pump is controlled by managing the dc-link voltage, which is achieved using a unit voltage generation method based on third-order integrators. This article discusses a photovoltaic-grid integrated system that uses an induction motor (IM) connected to a water pump. The system's appropriateness is supported by simulation results on the MATLAB/Simulink platform and by test results obtained with the aid of a prototype under different solar irradiance conditions. When transferring power, a straightforward dc-link voltage management method is used. When the water pump is not in use, the system's power is sent to the utility. Normally, it feeds the induction motor-driven water pump. Two current sensors or two voltage sensors are needed for this system to measure and estimate. Based on the direct current (dc) link current, the induction motor phases are computed using a modified space matrix modulation (SVM) approach. Using a third-order integrator for flux estimate and an artificial neural network–based model reference adaptive system, this system is able to adjust the power flow according to demand and accomplish speed estimation. The speed of an IM-pump may be controlled via field-oriented control. The power transfer between the utility and the IM drive and water pump is controlled by managing the dc-link voltage, which is achieved using a unit voltage generation method based on third-order integrators. The system's suitability is supported by results from simulations run on the MATLAB/Simulink platform and tests conducted on a prototype under different solar irradiance conditions. The following terms are used to describe various concepts related to DC microgrids: fast electric vehicle charging, integration of renewable energy sources, model predictive control, power demand control, thermal management of batteries, experimental validation and simulation, system stability.