Authors: Md Ashraful Islam, M M Naushad Ali, Moktadir Billah, Tajrian Mollick, Md. Ashaduzzaman Niloy, Ahmed Al Mansur

Abstract: This paper presents the techno-economic and environmental assessment of a hybrid renewable energy system integrating photovoltaic (PV), wind turbine, battery storage, and grid connection for sustainable power supply. Using HOMER Pro for optimal system sizing and the System Advisor Model (SAM) for performance validation, the study identifies the most cost-effective and reliable configuration for the selected infrastructure. The optimized system achieves an annual energy production of 1,234,709 kWh with zero unmet load, demonstrating high performance with a PV capacity factor of 16.9% and a performance ratio of 0.82. Economic analysis reveals a payback period of 10 years, an internal rate of return (IRR) of 20.88%, and a levelized cost of energy (LCOE) of 0.042 $/kWh. Furthermore, the system significantly reduces carbon emissions by approximately 282,754 kg CO2 annually. The results confirm that the proposed hybrid configuration is technically sound, economically viable, and environmentally sustainable, offering a practical solution for clean energy deployment.

DOI: 10.1109/QPAIN66474.2025.11172117 | ResearchGate Link | PDF

Authors: Shaharior Anik, Monjila Afrin Dorothi, Md. Ashaduzzaman Niloy

Abstract: We introduced and theoretically assessed a new hollow-core anti-resonant fiber (HC-ARF) with dual geometry cladding structure, suited for high power 2μm laser trans-mission. The fiber uses nested dual-layer circular and elliptical tubes to provide low-loss single-mode (LP01) transmission. We optimized the finite element method (FEM) to produce a loss of less than 1.5×10−2dB/m between 1900−2100nm with a remarkable low loss of 1.9×10−3 dB/m at 2025 nm. The concept achieves a high-order mode extinction ratio (HOMER) of over 1800 across a wide bandwidth. Furthermore, the fiber has exceptional bending tolerance, with a loss of 5.6×10−3 dB/m. at 1960 nm at bending radii above 17 cm. These findings demonstrate the HC-ARF's exceptional potential for efficient 2 µm. laser transmission.

DOI: 10.1109/QPAIN66474.2025.11171662 | ResearchGate Link | PDF

Authors: Md. Ashaduzzaman Niloy, Shahela Akter, Annafe Hossain Aronno, Adnan, Md Ashraful Islam, M M Naushad Ali

Abstract: This study evaluates and compares classical statistical and modern AI-based time-series forecasting models for predicting national power generation demand and natural gas reserve trends. Forecasting techniques including Holt's Linear Trend (HLT), ARIMA, SARIMAX, Polynomial Regression (PR), and the deep-learning model N-BEATS were trained on historical power and gas data. The models were assessed using MAE, MAPE, and RMSE. SARIMAX and N-BEATS achieved the best power generation forecast accuracies with MAPE values of 0.86% and 1.21%, respectively. In gas reserve forecasting, ARIMA (6, 3, 3) outperformed all others with a MAPE of 0.25%. Projections indicate a possible sharp decline in gas-based power generation share to 20.40% by 2032 under the N-BEATS model, compared to over 80% in some ARIMA scenarios. The study also quantifies the additional non-gas power generation needed up to 3903 MW by 2026 under N-BEATS to align with sustainable transition goals. These findings offer actionable insights for policymakers in energy planning and reinforce the utility of ML / DL techniques in modeling complex energy systems.

DOI: 10.1109/QPAIN66474.2025.11171820 | ResearchGate Link | PDF

Authors: Istiak Ahmed, Md. Ashaduzzaman Niloy, Marjan Al Haque, Showrov Rahman, Ferdous S. Azad, M.S. Hossain Lipu, Amanullah Maung Than Oo, Md. Abdur Razzak

Abstract: Power Factor Correction (PFC) ac–dc converters play a vital role in numerous power electronics applications as they can enhance power quality and achieve high levels of efficiency. The Single–Ended Primary–Inductor Converter (SEPIC) has emerged as a pivotal topology in the domain of PFC converters as it facilitates efficient power conversion in various industrial and domestic loads. However, despite its significance, a comprehensive review specifically focusing on the technical merit of its state–of–the–art topologies remains conspicuously absent in the literature. This paper endeavors to fill this gap by presenting a detailed synthesis of recent ac–dc SEPIC converters encompassing their classifications, performance characteristics, and applications. The working principles of the SEPIC converter with different control modes are provided in detail, and a classification section is introduced for the nominated converters according to the operating modes. The classification of ac–dc SEPIC converters based on the type of rectification, isolation, number of phases, and power levels are also demonstrated lucidly. A detailed chart of different topologies include information on topological structure, isolation, rated power, component counts, control and operating mode, conversion efficiency, input Power Factor (PF), and Total Harmonic Distortion (THD). The key design challenges and approximate cost estimation associated with semiconductor components are illustrated for various SEPIC converters. This study may serve as a valuable resource for researchers and manufacturers, providing insights, and fostering further advancements in the ever–evolving domain of power electronic converters.

DOI: 10.1016/j.rineng.2025.106814 | ResearchGate Link | PDF

Authors: Md. Ashaduzzaman Niloy, Sourav Chandra Das, M M Naushad Ali, Md Hasan Maruf, Ashik Ahmed, Razzaqul Ahshan

Abstract: Slime Mould Algorithm (SMA) is a highly promising meta-heuristic algorithm influenced by the food acquisition system of a Slime Mould. Chaos theory has proven its credibility in enhancing the performance of meta-heuristic optimization algorithms. In this work, the effect of incorporating three level chaos in improving the performance of SMA is studied. Ten well-known chaotic maps are integrated in initialization, exploration and exploitation stages of the original SMA, introducing ten new chaotic variants of SMA, named 3L-CSMA1 - 3L-CSMA10. These chaotic variants of SMA, are tested against twenty-two IEEE CEC benchmark functions from CEC'21 and CEC'22. To further validate the suitability of 3L-CSMAs against practical engineering problems, chaotic SMA adaptations are tested against multi-dimensional power system optimization problem Combined Heat and Power Economic Emission Dispatch (CHPEED). The simulation results revealed the significant improvement of 3L-CSMAs in finding the global optima, providing better fitness and improved solution to the complex functions compared to the original SMA and other advanced optimization algorithms.

DOI: 10.1109/ICCIT64611.2024.11022339 | ResearchGate Link | PDF

Authors: Md. Ashaduzzaman Niloy, MM Naushad Ali, Md Shahadat Hossen, Bibekananda Nath, Md Ashraful Islam, Shahela Akter, Ahmed Al Mansur

Abstract: Renewable energy systems (RES) present a sustainable and resilient alternative to traditional single-source energy systems, offering improved safety and cost-effectiveness in power generation. This study investigates the efficiency and economic viability of a photovoltaic (PV)-based energy system with integrated energy storage, tailored for a university campus setting. To optimize the utilization of available space, advanced optimization algorithms- Grey Wolf Optimizer (GWO), Whale Optimization Algorithm (WOA), and Artificial Hummingbird Al-gorithm (AHA) were employed to identify the optimal system size and associated costs. The findings highlight the strong potential of RES in urban environments. Specifically, the case study on the Green University of Bangladesh campus demonstrates the feasibility of the proposed system, achieving a cost of energy (COE) of $0.0185/kWh and a payback period of 8 years and 10 months. This research illustrates the capacity of PV-based systems to provide reliable and sustainable energy to urban campuses, delivering significant economic and environmental advantages that advocate for their wider adoption in similar settings.

DOI: 10.1109/STI64222.2024.10951103 | ResearchGate Link | PDF

Authors: Md. Ashaduzzaman Niloy, Kamran Ahmmed, M M Naushad Ali, Md Ashraful Islam, Md. Istiac Ahmed, Ridwan Abrar, Molla Shahadat Hossain Lipu

Abstract: In this research work, a renewable and clean energy system comprising of wind power, solar power and battery storage is designed with the primary objective of optimizing the total annual operating cost. Artificial Hummingbird Algorithm (AHA) is applied to size the system in such a way that it reduces the operating cost while satisfying the load demand and the regulatory constraints. A publicly available dataset mimicking the load demand, wind speed and solar insolation of an off-shore small island is used as simulation parameters and three test variants are defined to challenge AHA in solving the optimization problem. The simulation results revealed that the AHA has reduced the overall expenditure while fulfilling the system requirements and constraints to a level of $6691.88/annum or $0.39/kWh, which is proven to be superior when compared to other renowned algorithms like Particle Swarm Optimization, Grey Wolf Optimization and Artificial Bee Colony. The attained simulation results also highlight the efficacy and potential of AHA in solving complex power system design problems.

DOI: 10.1109/STI64222.2024.10951060 | ResearchGate Link | PDF

Abstract: This dataset contains high-resolution solar and wind measurement data collected from the Feni region, Bangladesh, spanning from 2017 to 2019. Logged at a 1-minute interval, the dataset provides a comprehensive record of atmospheric and meteorological conditions, essential for renewable energy analysis, climatological studies, and resource assessment.

The dataset includes wind data from multiple anemometers and wind vanes, along with solar radiation components such as Direct Normal Irradiance (DNI), Global Horizontal Irradiance (GHI), and Diffuse Horizontal Irradiance (DHI). Additionally, it records temperature, humidity, pressure, precipitation, and sun elevation parameters. Quality flags are included for each measured parameter to ensure data reliability.

DOI: https://dx.doi.org/10.21227/nx0k-h521 | ResearchGate Link

Abstract: This dataset contains high-resolution wind measurement data collected from 22 channels at varying heights, providing valuable insights for wind energy assessment, atmospheric research, and meteorological studies. The dataset includes wind speed, wind direction, and environmental parameters measured at multiple altitudes ranging from 10m to 120m. Each channel records parameters such as average wind speed, standard deviation, minimum and maximum values, gust speed, and wind vane direction. Additionally, atmospheric parameters such as temperature, relative humidity, and pressure are included. The data is stored in CSV format, enabling easy access and analysis for researchers and engineers. This dataset can support studies related to wind resource assessment, turbulence analysis, and renewable energy forecasting.

DOI: https://dx.doi.org/10.21227/enzm-qj50 | ResearchGate Link

Authors: Md. Ashaduzzaman Niloy, MM Naushad Ali, Md Ashraful Islam, Mohammad Asif Ul Haq, Molla Shahadat Hossain Lipu, Ahmed Al Mansur

Abstract: In Bangladesh, dairy industry plays a pivotal role in country's economic growth. Which is why integrating Biogas with Solar Photovoltaic (PV) systems offers a reliable and sustainable energy solution for dairy farms. By supplementing solar energy in a hybridized setup, biogas ensures a consistent power supply, while also contributing to waste management and environmental sustainability. In this paper, area, load demand and biogas-solar generation potential in a conventional Bangladeshi dairy farm is studied. Then four renowned nature-inspired algorithms, Grey Wolf Optimizer (GWO), Whale Optimization Algorithm (WOA), Artificial Hummingbird Algorithm (AHA), and Polar Fox Optimization Algorithm (PFOA) are applied to optimize the design parameters of a Hybrid Biogas-Solar System for the farm. The primary objectives of optimization are to minimize costs, maximize energy output, and promote self-sustainability in dairy farming. The simulation results revealed that the system provided by the Artificial Hummingbird Algorithm (AHA) achieved a levelized cost of energy (LCOE) of $0.045/kWh, a grid independent design with 0% loss of power supply probability (LPSP), and a payback period of 10 years and 4 months, making it the best-performing solution in terms of system reliability, self-sufficiency and energy efficiency compared to the GWO, WOA and PFOA.

DOI: 10.1109/ICREST63960.2025.10914450 | ResearchGate Link | PDF

Abstract: This dataset contains:

1. Hourly load demand of a typical single residence in Pacific Northwest, USA.

2. Hourly average incoming solar radiation and generated power from a single photovoltaic cell placed in Pacific Northwest, USA.

3. Hourly wind profile and generated power from a single wind turbine placed in Pacific Northwest, USA

This dataset was originally formulated by Kellogg et al. and first published in their research paper titled "Optimal unit sizing for a hybrid wind/photovoltaic generating system" in 1996. However, the given data were not numerical, which limited the use of that data in various applications like model training and optimization. Which is why in this dataset, the properly extracted numerical values of important parameters (ex: wind profile, wind power, insolation, solar power, demand) from the original paper is given for the researchers to use in their studies.

DOI: https://dx.doi.org/10.21227/55cq-x093 | ResearchGate Link

Authors: Md. Ashaduzzaman Niloy, Quazi D. M. Khosru

Abstract: In this paper, polynomial regression based machine learning model is applied to optimize the dielectric rod radius in a 60 degree bended Photonic Crystal (PhC) waveguide for achieving minimum transmission loss throughout the bended waveguide. A feedback based optimization process is proposed to study the electromagnetic transmission loss in a standard PhC model. The simulation results demonstrate a significant 67.79% reduction in transmission loss is achieved through the three stages of optimization compared to the initial unoptimized model, resulting in a transmission efficiency of 98.416%. Later, a comparative analysis of the proposed method and other similar research works revealed that the proposed feedback based method is superior in optimizing a 60 degree bended PhC for reducing electromagnetic transmission loss. A fabrication tolerance study revealed the ability of the proposed model to mitigate the increase in transmission loss due to fabrication inconsistencies.

DOI: 10.1109/ICECCE61019.2023.10442550 | ResearchGate Link | PDF

Authors: Marjan Al Haque, Md Ashaduzzaman Niloy, Alif Khan, Showrov Rahman, Istiak Ahmed, Ahmed Jobair

Abstract: This research presents a novel hybrid AC-DC Boost converter topology. The positive aspects of AC-DC conversion methods are combined in the proposed converter architecture to provide effective power transmission and enhanced grid compatibility. The circuit operations of the proposed converter are also described in detail. The performance of the converter has been evaluated with both the open-loop and closed-loop simulation models. The proposed circuit shows higher voltage gain performance by maintaining other factors including efficiency, total harmonic distortion (THD) and power factor (PF) within a good range. For open loop operation, the suggested converter offers conversion efficiencies of up to 99.77%. The performance of THD percentage shows remarkable result that is lower than the conventional circuit, while PF remains high for a wide range of operations. In closed-loop analysis, we achieved near unity (0.99) PF, more than 99% efficiency and lower THD than the conventional circuit.

DOI: 10.1109/ICPS60393.2023.10428708 | ResearchGate Link | PDF

Authors: Md Sayed Hasan Rifat, Md Ashaduzzaman Niloy, Mutasim Fuad Rizvi, Ashik Ahmed, Razzaqul Ahshan, Sarvar Hussain Nengroo, Sangkeum Lee

Abstract: A challenging engineering optimization problem in electrical power generation is the unit commitment problem (UCP). Determining the scheduling for the economic consumption of production assets over a specific period is the premier objective of UCP. This paper presents a take on solving UCP with Binary Slime Mould Algorithm (BSMA). SMA is a recently created optimization method that draws inspiration from nature and mimics the vegetative growth of slime mould. A binarized SMA with constraint handling is proposed and implemented to UCP to generate optimal scheduling for available power resources. To test BSMA as a UCP optimizer, IEEE standard generating systems ranging from 10 to 100 units along with IEEE 118-bus system are used, and the results are then compared with existing approaches. The comparison reveals the superiority of BSMA over all the classical and evolutionary approaches and most of the hybridized methods considered in this paper in terms of total cost and convergence characteristics.

DOI: 10.1109/ACCESS.2023.3273928 | ResearchGate Link | PDF

Authors: Md Ashaduzzaman Niloy, Faisal Hossain Reevu, Abrar Yeaser, Rubaiyat Islam Shupty, Abrar Shahriar Pramanik

Abstract: In this paper, a Chaotic Singer Grasshopper Optimization Algorithm (CS-GOA) is proposed to solve the Combined Economic and Emission Dispatch optimization problem. The premier objectives of CEED problem are to find the minimum fuel cost of power generation for the lowest possible pollutant emission, which is achieved simultaneously by a proper distribution of power generation duty among the generators. Price Penalty Factor (PPF) methodology is used to transform CEED into a single-objective optimization problem. A Chaotic Singer version of Grasshopper Optimization Algorithm is proposed and tested for solving CEED. The performance assessment tests consist of four different benchmark test systems with complexities such as different load demands and Valve-point loading of turbine. The attained simulation results highlight the superiority of CS-GOA in solving CEED problem in aspects of reduced fuel cost, emission and rapid convergence.

DOI: 10.1109/ICEEICT53905.2021.9667800 | ResearchGate Link | PDF

Authors: Md Ashaduzzaman Niloy, Md Sayed Hasan Rifat

Abstract: In this paper, the Slime Mould optimization Algorithm (SMOA) is proposed to find a cost effective solution to the Combined Economic and Emission Dispatch (CEED) problem. CEED is a multi-objective optimization problem where the main focus is on reducing power generation cost as well as the emission produced by the power generation while satisfying the regulatory constraints. With the term Price Penalty Factor (PPF), the multi-objective CEED is transformed into single-objective optimization problem. Then the SMOA is implemented on four benchmark test systems with complications such as valve-point loading effect and transmission losses. The test results were compared with other renowned optimization methods that were used in previous researches. The comparisons have shown superiority of SMOA over the other methods in terms of cost and emission reduction.

DOI: 10.1109/ICSCT53883.2021.9642577 | ResearchGate Link | PDF

1. Digital Logic Design: Calculator using Printed Circuit Boards and Integrated Circuits

Short Description: The project was to make a calculator using digital logic and combinational circuit which would have the capability to carry out addition, subtraction, multiplication and division operations.

Available files: Github | Project Report

2. Fault Analysis of Matter using Real Time Image Processing

Short Description: The project is mainly based on basic Image processing to find the accuracy of the objects of a production. The program will analyze the acquired images from live feed to detect the objects of unusual shapes amongst the desired output products.

Available files: Github | Project Report

3. CGPA Calculator using MATLAB GUI

Short Description: In this project, we had created a near proprietary CGPA calculator using Graphical User Interface (GUI) function of MATLAB. Using this, a user can generate result of a batch of students or a single student.

Available files: Github | Project Report | Video