National Institute of Technology, Silchar

About: National Institute of Technology, Silchar is a education organization based out in Silchar, Assam, India. It is known for research contribution in the topics: Computer science & Control theory. The organization has 1934 authors who have published 4219 publications receiving 41149 citations. The organization is also known as: NIT Silchar.

Prediction of sea surface temperatures using deep learning neural networks

Abstract: Sea surface temperature (SST) prediction has widespread applications in the field of marine ecology, fisheries, sports and climate change studies. At present, the real-time SST forecasts are made by numerical models which are categorically based on physics-based assumptions subjected to boundary and initial conditions. They are more suited to a large spatial region than in a specific location. In this study, location-specific SST forecasts were made by combining deep learning neural networks with numerical estimators at five different locations around India for three different time horizons (daily, weekly and monthly). Firstly, forecasts were made with traditional neural networks (NNs) and then through deep learning networks. The NNs significantly improved on the results achieved by numerical forecasts which were further enhanced by the deep learning long short-term memory (LSTM) neural network over all timescales and at all the selected sites. The model was performed successfully in terms of various statistical parameters with correlation values nearing 1.0 while minimizing the errors. Additionally, a comparative study with a linear system, the autoregressive integrated moving average with exogenous input was made. The predictive skills of deep learning LSTMs are found to be more attractive than the other existing techniques (linear or other NNs) due to their ability of learning long time dependencies and extracting features from a sample space.

Analysis of mixing performances in microchannel with obstacles of different aspect ratios

Abstract: In this study, the mixing performance and pressure drop characteristics have been numerically analyzed for flow through rectangular microchannel with obstacles in the walls arranged in a staggered ...

Numerical study of thermo-hydraulic characteristics for forced convective flow through wavy channel at different Prandtl numbers

Abstract: The present work numerically investigates the thermo-fluidic and entropy generation characteristics for laminar forced convective flow through wavy channel at different Prandtl number (Pr). Results are presented for the following range of parameters: Reynolds number $$5 \le \text{Re} \le 200$$ , Prandtl number $$0.72 \le \Pr \le 100$$ , dimensionless amplitude $$0.3 \le \alpha \le 0.7$$ and dimensionless wavelength $$0.5 \le \lambda \le 1.5$$ . It is observed that with increase in Pr, the thickness of the thermal boundary layer at trough region decreases slowly for smaller Re, whereas at higher Re, the rate of decrement is higher. The average Nusselt number increases with Pr for all amplitude, wavelength and Reynolds number. The relative heat transfer enhancement compared to equivalent plane channel is presented in terms of enhancement ratio (ER), and it shows a non-monotonic variation of ER with Pr at lower Re and a monotonic one at higher Re. The combined alteration of rate of heat transfer and pressure drop as compared to plane channel is enumerated by performance factor (PF), and the variation of PF with Pr shows non-monotonic behaviour at lower Re and monotonic one at higher Re. The variation of PF shows non-monotonic variation with Re for higher Pr and for smaller wavelength, whereas it monotonically decreases for all Pr at higher wavelength. Thermal entropy generation contribution is higher over the viscous one for all the cases considered. The local thermal entropy generation distribution varies with Re, Pr and geometrical configuration of the channel. For smaller amplitude (α = 0.3), the total entropy generation is minimum in the considered range of Re and Pr.

Plug in hybrid vehicle-wind-diesel autonomous hybrid power system: frequency control using FA and CSA optimized controller

Abstract: Large integration of renewable energy in hybrid power system in isolated mode of operation make frequency control a challenging task. This paper investigates the performance of Cuckoo Search Algorithm (CSA) and Firefly Algorithm (FA) based frequency control strategy of such a hybrid power system, which is a unique work. The generating units of the system are plug in hybrid vehicle (PHEV), wind turbine generators, a diesel engine generator (DEG) and battery energy storage system (BESS). The proportional plus integral (PI)/proportional integral derivative (PID) controllers are employed with PHEV, DEG and BESS to adjust the total active power generation in accordance to the load demand. Addition of PHEV reduces the reliance on the DEG or BESS as a result of variability and uncertainty of wind power. Different disturbance conditions such as step perturbations, random variations of load as well as wind output power, have been considered in the case studies under Matlab simulation to assess the performance of CSA and FA based control strategy. Analysis indicates that CSA based PID controller provides better response compare to GA, PSO and FA based PI/PID controller and CSA based PI controller. Sensitivity analysis has been carried out to check the robustness of FA and CSA optimized PI/PID controller gains.