ABSTRACT
This paper proposes a proficient hybrid approach based energy management system (EMS) of microgrid (MG) in grid-tied mode. The microgrid connected system like photovoltaic (PV), wind Turbine (WT), and microturbine (MT) battery. The major intention of the proposed approach is “to reduce the cost of electricity and enhancing the power flow (PF) among the source and load side.” The proposed method is the combination of Artificial Neural Network (ANN) and Hybrid Whale Optimization (HWO) approach, hence it is known as ANN-HWO. The necessary load requirement of grid connected microgrid scheme is constantly monitored by ANN. The HWOA develops the optimal combination of the microgrid based on the predicted load demand. Moreover, the two methods for microgrid energy management have been employed to minimize the impact of renewable energy predicting faults. The first method is focused on the various renewable energy sources (RESs) programming to reduction of electricity cost while the MG function. The second method is to balance the PF and minimize the forecasting errors impacts depending upon rule outlined from the scheduled power reference. The proposed technique is activated in MATLAB/Simulink site, and then the efficiency is classified with and without grid of MG structure. The effectiveness can be compared based on cost analysis and power generation of PV, MT, WT, and battery of existing methods. The power generation from the various sources based upon the ANN-HWO and existing methods is evaluated. Also, the total cost of the system also analyzed with different existing methods. The statistical evaluation, like mean, median, and standard deviation is also analyzed. The accuracy, specificity, recall and precision, RMSE, MAPE, MBE and consumption time of ANN-HWO, and existing techniques are also conducted. The mean, median, and standard deviation for 50 numbers of trails of the proposed technique is 1.3116, 1.2931, and 0.0354. The mean, median, and standard deviation for 100 numbers of trails of the proposed technique is 1.3232, 1.2786, and 0.0946. The RMSE, MAPE, MBE, and execution time for 50 numbers of trails of the proposed technique is 7.840, 0.748, 0.9971, and 1.011s. The RMSE, MAPE, MBE,and execution time for 100 numbers of trails of the proposed technique is 5.21, 0.93, 1.93, and 2.005 s. The proposed technique in 100, 200, 500, and 100 numbers of trails, the efficiency is 99.9673%, 99.7890%, 99.89402%, and 99.77879%.
Abbreviations
Table
Disclosure statement
No potential conflict of interest was reported by the author(s).
Additional information
Notes on contributors
Kallol Roy
Mr. Kallol Roy received B.E degree in Electrical Engineering from Burdwan University, Burdwan, West Bengal in 2002. He received M.E degree in Power Engineering from Jadavpur University, Kolkata, West Bengal in 2006. He is currently Assistant Professor in Electrical Engineering Department, University Institute of Technology, Burdwan University, West Bengal. He is currently working toward the Ph.D. degree at the Department of Power Engineering, Jadavpur University, Kolkata, West Bengal, India. His research interests include Micro Grid.
Kamal Krishna Mandal
Dr. Kamal Krishna Mandal received B.E from Jadavpur University, Kolkata, West Bengal in 1986. He received M.E from Allahabad University, Allahabad, Uttar Pradesh in 1998. He has received Ph.D. degree also. He is currently Professor in Power Engineering Department, Jadavpur University, Kolkata, West Bengal, India. His research interests include soft computing techniques, Power system optimization.
Atis Chandra Mandal
Dr. Atis Chandra Mandal received M.Sc in Physics from Burdwan University, Burdwan, West Bengal . He received Ph. D. degree from Jadavpur University, Kolkata, West Bengal, research done at Saha Institute of Nuclear Physics, Kolkata. He is currently Professor in Physics Department, Burdwan University, Burdwan, West Bengal, India. His research interests include soft computing techniques.