Paper Title
MHT-LSTM-Based Power Transmission By Considering Energy Management and Cost Reduction
Abstract
The rising popularity of battery-limited Electric Vehicles (EVs) put forward a significant problem of how to
charge vehicles efficiently. It is termed as the Electric Vehicle Charging Scheduling (EVCS) issue. Thus, in this work, a
fresh Mexican Hat Type function-centric Long Short-Term Memory (MHT-LSTM)-centered power transmission from Micro
Grids (MGs) is proposed by considering Energy Management (EM) and cost reduction. For producing current, the three
Renewable Energy Resources (RERs) like solar, wind, and biomass are used. Contrarily, to directly charge the vehicle and
swap the battery, the EV user sends the request to the nearest Charging Station (CS) and Battery Swapping Station (BSS),
respectively. Thus, for producing the charge, the nearest CS and BSS are detected utilizing MHT-LSTM, whereas the CSs
with limited resources are identified utilizing Depth First Search-basedTasmanian Devil Optimization (DFS-TDO). Demand
and supply are measured using power pitch analysis, and the power is produced directly to the prioritized CS. The outcomes
display that the proposed system attained a maximum efficiency of 65% for EM compared to existing methods and reduced
the overall cost by 45.6 % using renewable sources.
Keywords - Mexican Hat Type function based Long Short Term Memory (MHT-LSTM) neural network, Depth First Search
based TDO (DFS-TDO), Micro Grids (MGs), Electric Vehicle (EV) Energy Management (EM), and Cost.