Department of Mathematics

Permanent URI for this collectionhttp://localhost:4000/handle/123456789/1920

Browse

Subject: search.filters.subject.Neural networks

Search Results

Now showing 1 - 6 of 6
  • No Thumbnail Available
    Item
    Application of Spiking Neural Networks in Renewable Energy Forecasting
    (Springer, 2024-12) Pasari, Sumanta
    Considering the high consumption rates of the non-renewable energy sources as well as their adverse climatic impacts, renewable energy has become a widespread topic of discussion. Among the available renewable resources, solar and wind are the highest contributors. However, the high influence of atmospheric parameters and higher cost involved in energy production prevent the widespread use of renewable energy among common public. The location identification for optimum energy production is also a crucial step for setting up future energy plants. In this regard, here we propose a novel strategy to compare prediction results in terms of loss made by traditional convolutional neural network (CNN) with that of spiking neural network (SNN). Though the SNNs are popularly used for vision related tasks, here we evaluate their efficacy in analyzing time series data of solar irradiance and wind speed. In summary, we provide a comprehensive discussion on SNN and their significance on energy forecasting.
  • No Thumbnail Available
    Item
    Earthquake forecasting using artificial neural networks
    (ISPRS, 2018) Pasari, Sumanta
    Earthquake is one of the most devastating natural calamities that takes thousands of lives and leaves millions more homeless and deprives them of the basic necessities. Earthquake forecasting can minimize the death count and economic loss encountered by the affected region to a great extent. This study presents an earthquake forecasting system by using Artificial Neural Networks (ANN). Two different techniques are used with the first focusing on the accuracy evaluation of multilayer perceptron using different inputs and different set of hyper-parameters. The limitation of earthquake data in the first experiment led us to explore another technique, known as nowcasting of earthquakes. The nowcasting technique determines the current progression of earthquake cycle of higher magnitude earthquakes by taking into account the number of smaller earthquake events in the same region. To implement the nowcasting method, a Long Short Term Memory (LSTM) neural network architecture is considered because such networks are one of the most recent and promising developments in the time-series analysis. Results of different experiments are discussed along with their consequences.
  • No Thumbnail Available
    Item
    Wind Energy Prediction Using Artificial Neural Networks
    (Springer, 2020-07) Pasari, Sumanta
    Renewable energy sources are one of the most vital alternatives to the conventional non-replenishable energy generating systems. Among several renewable power sources, the installed wind power capacity contributes to almost half of the total capacity. However, the variability and seasonality in wind speed, wind direction, atmospheric pressure, relative humidity and precipitation cause wind power generation to be highly volatile. In this regard, the present study aims to develop a wind speed prediction scheme using artificial neural network (ANN) techniques. Single step and multistep recurrent neural networks (RNNs) are implemented. The long short term memory (LSTM), rectified linear unit (ReLU) activation function and Adam optimization algorithm are considered to carry out daily to monthly prediction using the RNN process. Results, based on the data from Charanka solar energy park in Gujarat, indicate that root means square (RMS) errors for univariate single layer, multivariate single layer and univariate two-layer models are 0.601, 0.782 and 1.120, respectively. Therefore, a univariate single layer RNN architecture is recommended for wind speed prediction. We envisage that a multilayer RNN model may improve the prediction accuracy over longer period.
  • No Thumbnail Available
    Item
    Earthquake Prediction Using Deep Neural Networks
    (IEEE, 2022) Pasari, Sumanta
    Reliable prediction of earthquakes has numerous societal and engineering benefits. In recent years, the exponentially rising volume of seismic data has led to the development of several automatic earthquake detection algorithms through machine learning approaches. In this study, we propose a fully functional and efficient earthquake detector cum forecaster based on deep neural networks of long-short-term memory (LSTM) units. The model captures inherent temporal characteristics of earthquake data. For illustration, we consider an earthquake catalog from the Himalaya and its neighboring regions. The proposed LSTM model shows satisfactory performance for small to medium-sized earthquakes. We also implement a baseline artificial neural network (ANN) model to perform a suitable comparison. It is observed that both ANN and LSTM models fail to produce desired result for large events.
  • No Thumbnail Available
    Item
    Earthquake Magnitude Prediction in Chile Using Neural Network
    (IEEE, 2022) Pasari, Sumanta
    In this study, we implement an earthquake magnitude prediction model using a neural network for a test region in Chile. For this, the epicenter of earthquake is located on a mesh with dimensions of 1°×1°. We adopt a zonation scheme originally proposed by Reyes and Cardenas [1]. The scheme uses increments in b−value and other input parameters to incorporate G-R linear relation and Bath’s law. The model enables the prediction of the maximum magnitude for a given cell within the next five days. Common seismological parameters are used for the performance evaluation of the model. Results show satisfactory performance of the proposed model in comparison to other existing models.
  • No Thumbnail Available
    Item
    Air Quality Prediction Using the Fractional Gradient-Based Recurrent Neural Network
    (Hindawi Publishing Corporation, 2022) Agarwal, Shivi; Mathur, Trilok
    In this study, the air quality index (AQI) of Indian cities of different tiers is predicted by using the vanilla recurrent neural network (RNN). AQI is used to measure the air quality of any region which is calculated on the basis of the concentration of ground-level ozone, particle pollution, carbon monoxide, and sulphur dioxide in air. Thus, the present air quality of an area is dependent on current weather conditions, vehicle traffic in that area, or anything that increases air pollution. Also, the current air quality is dependent on the climate conditions and industrialization in that area. Thus, the AQI is history-dependent. To capture this dependency, the memory property of fractional derivatives is exploited in this algorithm and the fractional gradient descent algorithm involving Caputo’s derivative has been used in the backpropagation algorithm for training of the RNN. Due to the availability of a large amount of data and high computation support, deep neural networks are capable of giving state-of-the-art results in the time series prediction. But, in this study, the basic vanilla RNN has been chosen to check the effectiveness of fractional derivatives. The AQI and gases affecting AQI prediction results for different cities show that the proposed algorithm leads to higher accuracy. It has been observed that the results of the vanilla RNN with fractional derivatives are comparable to long short-term memory (LSTM). 1. Introduction