Elman neural network optimized by improved golden search optimization algorithm for optimal designing of the hybrid renewable energy systems
Suxia Chen,Jiachen Zhang,Guijie Zhang,Homayoun Ebrahimian
DOI: https://doi.org/10.1080/15567036.2024.2375740
2024-07-14
Energy Sources Part A Recovery Utilization and Environmental Effects
Abstract:Hybrid renewable energy systems (HRESs) could offer a promising solution to the challenges of providing a dependable, economical and sustainable energy's source. The integration of these technologies can help to overcome the limitations of intermittency, variability, and storage. Further, the economic and environmental benefits of HRESs make them an attractive option for widespread adoption in the future. However, designing an optimal configuration that can guarantee stable production at the minimum LCOE (Levelized Cost of Electricity) and LPSP (Loss of Power Supply Probability) values is the main concern of researchers. In this regard, the present study aims to optimize the configuration of a hybrid plant and determine the plant's economic viability. The proposed plant, containing of a storing unit (VRFB, Vanadium Redox Flow Batteries-based battery energy storing), a solar photovoltaic (PV) plant, a wind farm, and a hydrogen storage system (for running an FC (Fuel Cell)), is designed to provide a stable production to the grid at a minimum LCOE value under least LPSP value and extra energy output. The development of the proposed optimal model for the HRES is based on an Elman Neural Network (ENN) model established by a new optimization algorithm (i.e. Improved Golden Search Optimization Algorithm (IGSOA)). The model is applied to a case study, San Luis Ayucan in Mexico, to verify the method efficacy. From the results, the storing system, including in cooperation battery and hydrogen indicate the main cost driver for each model arrangement with total cost percentage of 63.4%, 70.2%, 73.2%, and 73.7% for the proposed ENN/IGSOA, ENN, SVR, and ANN, respectively. Further, an ANN model had zero's both an excess energy and an LPSP value; so, the system needs to rely heavily on energy storing to ensure that the stable load can be fulfilled. As a result, this increases the system's total rate and subsequently results in a higher LCOE.
energy & fuels,engineering, chemical