Power management strategy based on Elman neural network for grid-connected photovoltaic-wind-battery hybrid system. (April 2022)

• Record Type: Journal Article
• Authors: Boualem, S.; Kraa, O.; Benmeddour, M.; Kermadi, M.; Maamir, M.; Cherif, Hakima
• Is Part Of: Computers & electrical engineering. Volume 99(2022)
• Publication Date: 2022-04
• Subjects: Energy management strategy -- Elman neural network -- Multilayer perceptron neural network -- Stateflow -- Grid-connected -- Fuzzy logic -- Model predictive control; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2022.107823
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: The energy management strategy based on Stateflow strategy is presented. Three control strategies are discussed and demonstrated. A comparison between the proposed ENN controller and MLP controller was conducted. The Elman neural network (ENN) controller has higher accuracy than MLP controller. Abstract: In this study, an improved energy management controller (EMC) is proposed for a grid-connected hybrid system (HS), composed of wind–photovoltaic generation and an energy storage system (ESS). The main contributions of this study are as follows: first, we developed an intelligent supervisory controller based on a recurrent neural network, namely Elman neural network (ENN), to alleviate the complexity and requirement of a rule-based structure or prior mathematical modeling. Second, we designed an energy management strategy (EMS) using Stateflow (SF) approach to extract the training and testing datasets for construction of the ENN controller. The proposed framework was compared with a multilayer perceptron neural network (MLPNN) strategy to evaluate their performances. The simulation results indicated that the strategy proposed is a suitable option to efficient energy management control based on a predictive model, as it is not very complex and does not require a high processing machine. Graphical abstract: Image, graphical abstract