Browsing by Author "Rai, A."

Now showing 1 - 5 of 5

An integrated frequency domain decomposition and deep neural network approach for short-term PV power forecast
(Springer Science and Business Media Deutschland GmbH, 2025) Kumar, A.; Kashyap, Y.; Rai, A.
Weather disturbances and atmospheric parameters significantly influence the fluctuations in PV power output, which in turn affect the stability of grid operations. The current study proposed short-term PV power forecasting based on appropriate cutoff frequency in frequency domain and artificial intelligence method. Initially, the actual PV power data are decomposed into the frequency domain, and optimal cutoff frequency is determined by minimizing the squared difference of correlation between the decomposed components. Subsequently, the PV power is separated into low-frequency components (LFC) and high-frequency components (HFC). Then, long short-term memory (LSTM) and light gradient boosting machine (LGBM) models are then employed to forecast the LFC and HFC PV power. The final forecast output is generated using various recombination methods. The proposed combined forecast model, LFC-LGBM + HFC-LGBM, based on frequency domain decomposition (FDD) and LGBM approach, demonstrates superior performance compared to models (LFC-LSTM + HFC-LSTM), (LFC-LGBM + HFC-LSTM), and (LFC-LSTM + HFC-LGBM). The best-performing model (LFC-LGBM + HFC-LGBM) achieves a MAE of 4.9420%, a RMSE of 7.1047%, and a correlation index (R) of 0.9734 for 15-min ahead timesteps. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024.
Network optimization using femtocell deployment at macrocell edge in cognitive environment
(2016) Ghosh, J.; Bachhar, S.; Nandi, U.K.; Rai, A.; Roy, S.D.
This research focuses on the problem of cell edge user�s coverage in the context of femtocell networks operating within the locality of macrocell border where pathloss, shadowing, Rayleigh fading have been included into the environment. As macro cell edge users are located far-away from the macro base station (MBS), so that, the underprivileged users (cell edge users) get assisted by the cognitive-femto base station (FBS) to provide consistent quality of service (QoS). Considering various environment factors such as wall structure, number of walls, distance between MBS and users, interference effect (i.e., co-tier and crosstier), we compute downlink (DL) throughput of femto user (FU) for single input single output (SISO) system over a particular sub-channel, but also based on spectrum allocation and power adaptation, performance of two tier network is analyzed considering network coverage as the performance metric. Finally, the effectiveness of the scheme is verified by extensive matlab simulation. � Springer International Publishing Switzerland 2016.
Network optimization using femtocell deployment at macrocell edge in cognitive environment
(Springer Verlag service@springer.de, 2016) Ghosh, J.; Bachhar, S.; Nandi, U.K.; Rai, A.; Dhar Roy, S.D.
This research focuses on the problem of cell edge userâ€™s coverage in the context of femtocell networks operating within the locality of macrocell border where pathloss, shadowing, Rayleigh fading have been included into the environment. As macro cell edge users are located far-away from the macro base station (MBS), so that, the underprivileged users (cell edge users) get assisted by the cognitive-femto base station (FBS) to provide consistent quality of service (QoS). Considering various environment factors such as wall structure, number of walls, distance between MBS and users, interference effect (i.e., co-tier and crosstier), we compute downlink (DL) throughput of femto user (FU) for single input single output (SISO) system over a particular sub-channel, but also based on spectrum allocation and power adaptation, performance of two tier network is analyzed considering network coverage as the performance metric. Finally, the effectiveness of the scheme is verified by extensive matlab simulation. Â© Springer International Publishing Switzerland 2016.
On the kernelization complexity of string problems
(Elsevier B.V., 2018) Basavaraju, M.; Panolan, F.; Rai, A.; Ramanujan, M.S.; Saurabh, S.
In the CLOSEST STRING problem we are given an alphabet ?, a set of strings S={s1,s2,…,sk} over ? such that |si|=n and an integer d. The objective is to check whether there exists a string s over ? such that dH(s,si)?d, i?{1,…,k}, where dH(x,y) denotes the number of places strings x and y differ at. CLOSEST STRING is a prototype string problem. This problem together with several of its variants such as DISTINGUISHING STRING SELECTION and CLOSEST SUBSTRING have been extensively studied from parameterized complexity perspective. These problems have been studied with respect to parameters that are combinations of k, d, |?| and n. However, surprisingly the kernelization question for these problems (for the versions when they admit fixed-parameter tractable algorithms) is not studied at all. In this paper we fill this gap in the literature and do a comprehensive study of these problems from kernelization complexity perspective. We settle almost all the problems by either obtaining a polynomial kernel or showing that the problem does not admit a polynomial kernel under a standard assumption in complexity theory. © 2018 Elsevier B.V.
Optimized radial basis function neural network model for wind power prediction
(Institute of Electrical and Electronics Engineers Inc., 2016) Shetty, R.P.; Sathyabhama, A.; Srinivasa Pai, P.P.; Rai, A.
In this paper an effort has been done in developing a fast and efficient Radial Basis Function (RBF) neural network model to predict the power output of a wind turbine. The performance of the RBF neural network has been improved by making use of a hybrid Particle Swarm Optimization based Fuzzy C Means (PSO-FCM) clustering algorithm. Extreme Learning Machine (ELM) algorithm has been used to improve the speed of learning. Particle Swarm Optimization (PSO) has also been used to optimize the number of centers and width of the RBF units of the developed neural network model. The simulation results show that the model developed has a compact network structure and good generalization ability with 100% accuracies on training, test and validation data sets. The novelty of the present work is the use of PSO in optimizing the RBF neural network model and use of ELM in training the same. Â© 2016 IEEE.