Artificial Deep Neural Network in Hybrid PV System for Controlling the Power Management

Abstract

The analysis of different components of a grid-linked hybrid energy system (HES) comprising a photovoltaic (PV) system is presented in this work. Due to the increase of the population and industries, power consumption is increasing every day. Due to environmental issues, traditional power plants alone are insufficient to supply customer demand. In this case, the most important thing is to discover another approach to meet customer demands. Most wealthy countries are now concentrating their efforts on developing sustainable materials and investing considerable amounts of money in product development. Wind, solar, fuel cells, and hydro/water resources are among the most environmentally benign renewable sources. To control the variability of PV generation, this sort of application necessitates the usage of energy storage systems (ESSs). Lithium-ion (Li-ion) batteries are the most often used ESSs; however, they have a short lifespan due to the applied stress. Hybrid energy storage systems (HESSs) started to evolve as a way to decrease the pressure on Li-ion batteries and increase their lifetime. This study represents a great power management technique for a PV system with Li-ion batteries and supercapacitor (SC) HESS based on an artificial neural network. The effectiveness of the suggested power management technique is demonstrated and validated using a conventional PV system. Computational models with short-term and long-term durations were used to illustrate their effectiveness. The findings reveal that Li-ion battery dynamical stress and peak value are reduced, resulting in longer battery life.

Review on smart grids and renewable integration: An artificial intelligence-powered perspective

“…Almost all forms of RES for the management, allocation, development of policies, calculation and optimization have benefited from the widespread use of AI strategies. The use of AI approaches in bioenergy, geothermal, solar, wind and hydropower is briefly covered in the study (Sahoo et al 2022). All of the power system operators have prioritized commercial energy trading from the outset.…”

Deep learning based buck-boost converter for PV modules

“…Further, the control of the output voltage, which is a challenging nonlinear problem, might be reduced by deep learning. Convolutional Neural Network (CNN) belongs to deep learning that learns to accomplish a task by creating maximum hidden layers and at the final step adding regression layer [27] , [28] , [29] , [30] . The main objective of deep learning is to teach computers to do what comes naturally to humans and automatically extract the features that are done manually in the case of machine learning [31] , [32] .…”

Robust Control of RSOC/Li-ion Battery Hybrid System Based on Modeling and Active Disturbance Rejection Technology

“…Two control problems in hybrid power generation systems are discussed: (i) regulation of the Li-ion battery charging and discharging temperatures along with voltage fluctuations; and (ii) control of thermoelectric performance and hydrogen production in a dual-mode (SOFC and SOEC modes) hybrid system. Although there is little previous research available in the literature on the control of RSOC/Li-ion battery hybrid systems, we hypothesize that this approach can be applied to the control of hybrid RSOC/Li-ion battery systems by learning from existing control methods for hybrid systems in related fields [26][27][28][29]. Therefore, the main contribution of this paper is the introduction of a simple and practical controller for RSOC/Li-ion battery hybrid systems.…”