System modeling of micro‐grid with hybrid energy sources for optimal energy management—A hybrid elephant herding optimization algorithm‐adaptive neuro fuzzy inference system approach. (23rd June 2021)

• Record Type: Journal Article
• Authors: Durairasan, M.; Ramprakash, S.; Balasubramanian, Divya
• Other Names: Yan Su guestEditor.; Liu Yang guestEditor.
• Is Part Of: International journal of numerical modelling. Volume 34:Number 6(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-06-23
• Subjects: demand response -- energy management -- energy resources -- enhanced EHOA -- micro grid -- predicted load requirement -- system modeling; Electric networks -- Mathematical models -- Periodicals; Electronics -- Mathematical models -- Periodicals; 621.3011
• DOI: 10.1002/jnm.2915
• Languages: English
• ISSNs: 0894-3370

Abstract

Abstract: This manuscript proposes a hybrid method for system modeling and optimal allocation of low cost micro‐grid. The proposed hybrid method is the joint execution of the enhanced elephant herding optimization algorithm (EHOA) and adaptive neuro fuzzy inference system (ANFIS) named as EHO‐ANFIS. By using MG inputs, such as solar photovoltaic, wind turbine (WT), micro turbine (MT), fuel cell (FC), and battery energy storage system. EHO optimizes micro‐grid configuration in minimal fuel costs based on needed load requirement. Here, learning phase of ANFIS is utilized for predicting the load requirement. EHOA reduces operation and maintenance costs, emission cost on the basis of the predicted load requirement. The proposed method is executed in MATLAB/Simulink site and the robustness of the proposed method is compared with different existing methods. In the proposed method, the maximal generated power of photovoltaic represents 6 kW, wind turbine indicates 7.8 kW, micro turbine denotes 11.8 kW, FC implies 6.8 kW, and battery refers 3 kW. By utilizing genetic algorithm, the generated power of photovoltaic signifies 7 kW, WT implicates 6 kW, MT implicates 4 kW, FC refers 7 kW, and battery implies 14 kW. The proposed method has minimal cost effective depending on its load demand. The computational time of the proposed technique under 100, 250, 500, and 1000 trails is 5600, 14 000, 28 000, and 56 000 s.