This study introduces innovative approach for predicting the State of Health (SOH) and Remaining Useful Life (RUL) of lithium-ion batteries, leveraging datasets from NASA and Oxford to establish a robust predictive framework. A key highlight is the development of a novel neural network architecture defined as AttentiveLSTM which integrates Long Short-Term Memory (LSTM) networks with Transformer mechanisms to enhance feature extraction and time-series forecasting. The research addresses critical challenges in the domain, including the non-linear behavior of battery degradation, diverse operating conditions, and the scarcity of historical data. The proposed AttentiveLSTM model surpasses existing approaches in predictive accuracy for both SOH and RUL. Additionally, it introduces an advanced objective function combining Denoising Autoencoder (DAE) loss functions with prediction loss to improve model performance. This work not only advances predictive modeling techniques but also contributes to the broader goal of enabling more efficient and sustainable use of electric vehicle (EV) batteries, thereby supporting the transition to eco-friendly industrial transportation systems.
In this paper, a photovoltaic (PV) energy extraction process with higher efficiency and wider scope of modular and portable programs using programmable logic controller (PLC) is proposed. Under dual layer maximum power point tracking (DLMPPT) concept, energy gains are simultaneously achieved as distributed MPPT (DMPPT) across PV modules and centralized MPPT (CMPPT) across PV arrays with newly modified “incremental conductance algorithm” using adaptive subroutine jump control at an appropriate ramp rate correction, to update pulse width modulation duty cycle for speedy MPPT action and resolving issue pertaining to multimaxima points under partial shading conditions as well. A dynamic energy balance between supply of PV energy to common dc bus and its quick transfer/consumption by load across inverter is a prime factor for enhanced DLMPPT benchmark in the process. This scheme is simulated on MATLAB, followed by hardware (5 kW) implementation for performance data generation and analysis on a yearly basis at a farm house located (lat = 26.397, lon = 74.728) in a village near Ajmer, India. It is found that DLMPPT is ahead in energy gains by 13.26% and 7.88% as compared to CMPPT and DMPPT, respectively.
1 September 2021 in Sustainable Energy, Grids and Networks
This work presents a step in micro-level Photovoltaic summation technology, for combining unsymmetrical PV modules with difference in specification, age, re-locating on nonfertile land or multi location mounting; by designing an Add on modular hardware (AOMH) interface to multi input single output (MISO) dc bus, using MISO controller exercising suitable MPPT under partial shading/long distance/mismatching PV module conditions along with sensor-less PV vector control inverter (PV-VCI) driving a 3 phase Asynchronous squirrel cage induction motor (ASCIM) dedicated to irrigation with auto configurable/tunable at start. PV source gets automatically added/removed in ramp profile in proportion to load in step thus keeping ON only when required; enabling good efficiency, minimum heating and prolonged life like that of PV module. The contribution includes: reduced PV energy cost per watt by 24.10%, improved ROI by 27.48%, water discharge yield by 11.06%, MPPT impact by 14.59% with 90.88% MISO system efficiency comparing a normal scheme.