Department of Electrical and Electronics Engineering

Permanent URI for this collectionhttp://localhost:4000/handle/123456789/1925

Browse

Filters

2021 - 20222

Settings

Author: search.filters.author.Bansal, Hari OmSubject: search.filters.subject.Neural networks

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Fuzzy logic and Elman neural network tuned energy management strategies for a power-split HEVs
    (Elsevier, 2021-06) Bansal, Hari Om; Singh, Dheerendra
    This paper focuses on optimal energy sharing between the two sources i.e., the internal combustion engine and the battery-powered electric motor in a hybrid electric vehicle (HEV). It is necessary that these sources operate in their efficient operating region while fulfilling the energy demanded by the vehicle to obtain the maximum fuel economy. As both of these sources have different operating characteristic and vehicle running conditions, the situation requires a smart controller to address this problem appropriately. In this work, fuzzy logic and Elman neural network-based adaptive energy management strategies (EMS) in an HEV are designed and implemented. The input parameters to these EMS are torque demand, battery state of charge, and regenerative braking. The proposed strategy aims to maximise the fuel economy while maintaining the battery health. A power-split HEV along with EMS is designed, modelled and simulated in MATLAB/Simulink first and then the whole system is validated in real-time using controller hardware in the loop testing platform (CHIL). The FPGA based MicroLabBox CHIL has been employed to test the system behaviour in real-time. The proposed EMS have been compared with conventional strategies and the comparison reveals that the Elman neural network-based method results in higher fuel economy, faster response, and minimal mismatch between desired and attained vehicle speeds.
  • No Thumbnail Available
    Item
    Transformer-based time series prediction of the maximum power point for solar photovoltaic cells
    (Wiley, 2022-06) Bansal, Hari Om
    This paper proposes an improved deep learning-based maximum power point tracking (MPPT) in solar photovoltaic cells considering various time series-based environmental inputs. Generally, artificial neural network-based MPPT algorithms use basic neural network architectures and inputs which do not represent the ambient conditions in a comprehensive manner. In this article, the ambient conditions of a location are represented through a comprehensive set of environmental features. Furthermore, the inclusion of time-based features in the input data is considered to model cyclic patterns temporally within the atmospheric conditions leading to robust modeling of the MPPT algorithm. A transformer-based deep learning architecture is trained as a time series prediction model using multidimensional time series input features. The model is trained on a dataset containing typical meteorological-year data points of ambient weather conditions from 50 locations. The attention mechanism in the transformer modules allows the model to learn temporal patterns in the data efficiently. The proposed model achieves a 0.47% mean average percentage error of prediction on non-zero operating voltage points in a test dataset consisting of data collected over a period of 200 consecutive hours; resulting in the average power efficiency of 99.54% and peak power efficiency of 99.98%. The proposed model is validated through real-time simulations. The proposed model performs power point tracking in a robust, dynamic, and nonlatent manner, over a wide range of atmospheric conditions.