Journal of Engineering Research and Reports

Federated Learning and Graph Neural Networks for Transmission System Stability: A Privacy-Preserving Artificial Intelligence Framework for Smart Grids

Imoh Isaiah Udofot — 2025-10-18

Integrating renewables and grid-edge technologies has increased modern power systems’ complexity and data intensity, necessitating advanced Artificial Intelligence (AI) for stability management. However, centralized AI faced fundamental limitations due to data privacy, sovereignty, and communication constraints. This review explored the fusion of Federated Learning (FL) and Graph Neural Networks (GNNs) as a transformative privacy-preserving paradigm. FL enabled collaborative training across decentralized utilities without raw data sharing, while GNNs natively modelled grid topology. The study systematically analyzed FL–GNN applications in state estimation, stability assessment, anomaly detection, and resilient control. Key challenges—including data heterogeneity, communication efficiency, and privacy–accuracy trade-offs—were critically examined. The review concluded by outlining future pathways, such as physics-informed models and digital twin integration, highlighting the potential of FL and GNNs for secure and intelligent grid management. Reported implementations show up to 8% higher predictive accuracy and over 10% improvement in real-time control efficiency compared to centralized models.

Development of an Automatic Assembly System for RV Reducers Using SolidWorks and VB.NET

Li Junqi — 2025-10-18

Objective: RV reducer is one of the core components of industrial robots, but its complex structure makes the design process cumbersome. Therefore, parametric design was applied to certain design aspects of the RV reducer to shorten its design cycle.

Methods: Based on the structural analysis of the RV reducer, this paper takes the RV-110E reducer produced by a certain factory as an example, achieving parametric modeling of the entire RV reducer. Through VB.NET programming language in Visual Studio 2025 software, the program design for the RV reducer assembly process was completed to achieve automated assembly of the RV reducer.

Results: Through the assembly method based on spatial coordinate transformation, a set of automatic assembly programs was designed to assemble the reducer automatically, reducing the assembly time from 6 hours to 5 minutes.

Conclusion: The results demonstrate the rationality and necessity of this program design, which can significantly reduce the design cycle of RV reducers, lower labor costs, and greatly improve production efficiency and yield rate.

Torrefaction of Groundnut Shell and Straw for Improving the Fuel Quality and Bioenergy Application

Victor Achirgbenda — 2025-10-18

Groundnut residues, specifically shells and straw, are abundant agricultural by-products in Nigeria and other developing nations. These residues are often underutilised or openly burned, contributing to environmental pollution. This study investigates torrefaction as a thermochemical pretreatment process to improve the fuel quality of groundnut residues. Groundnut shell and straw were mixed in different ratios (100:0, 75:25, 50:50, 25:75, and 0:100) and torrefied under varying temperatures (250, 270 and 290⁰C) and residence times (15, 30 and 60 minutes). Proximate and ultimate analyses, higher heating value (HHV), mass yield, energy yield, and energy density ratio performance were evaluated. Results showed that torrefaction reduced moisture and volatile matter content while increasing fixed carbon and HHV (up to 24 MJ/kg). Improvements were also observed in hydrophobicity, grindability, bulk density, and durability of the pellets. The findings demonstrate that torrefaction significantly upgrades groundnut residues into viable solid biofuels for domestic and industrial applications.

Optimization of Hybrid Renewable Energy Systems for Sustainable Rural Electrification in Kwara State, Nigeria: A Techno-Economic and Environmental Assessment

Muniru Olajide Okelola — 2025-10-18

Access to reliable and affordable electricity remains a major challenge for rural communities in Nigeria, particularly in Kwara State, where weak grid infrastructure and reliance on costly fossil fuels constrain socio-economic development. Despite abundant renewable resources such as solar, wind, hydro, and biomass, these remain underutilized. This study designed and optimized Hybrid Renewable Energy Systems (HRES) for three off-grid communities—Owode, Abe-Eya, and Ila-Oja Aboto—using a mixed-method approach that combined household energy surveys, field-based resource assessments, and techno-economic simulations. HOMER Pro (v3.14.2) was used to develop optimal hybrid configurations integrating solar photovoltaic (PV), wind, and biomass systems, while RETScreen Expert evaluated financial and environmental performance. Results showed strong solar irradiance (~5.85 kWh/m²/day), moderate wind speeds (3.8–4.2 m/s) in Abe-Eya, and abundant biomass availability (1.5–2.2 tons per household per year) in Owode and Ila-Oja Aboto. The optimal configuration included 54 × 215 W PV modules, 2 × 1 kW wind turbines, 4 × 1 kW batteries, and 34 kVA converters, with installation costs of ₦22.32 million (Owode), ₦24.48 million (Abe-Eya), and ₦24.96 million (Ila-Oja Aboto), generating 25,754–31,643 kWh annually. Economic analysis indicated strong viability with NPV > ₦550 million, IRR of 142–158%, payback periods of 3.2–3.5 years, and benefit–cost ratios of 1.15–1.17. Greenhouse gas emissions decreased by 63–70%, from 6,000–7,500 kg CO₂/year to 613–959 kg CO₂/year. Sensitivity analysis confirmed robustness under ±25% cost variation. The study demonstrates that community-specific solar-biomass and solar-wind hybrid systems provide reliable, cost-effective, and sustainable electrification pathways for rural Nigeria, advancing the national energy transition and Sustainable Development Goal 7.

Analysis of Rain Attenuation over the Geo-climatic Regions of Nigeria Using Attenuation Prediction Models

Ozoemelam Onyebuchi — 2025-10-24

At frequencies above 10GHz, rain drops absorb part of the energy (signal) and part are scattered in different directions, causing unavailability of signals and making the system link unreliable. This study presents analysis of rain-induced attenuation on satellite communications at Ku band and the effects on telecommunications, broadcasting and navigations over the geo-climatic regions in Nigerian. Rainfall parameters were recorded using the ISS automatic weather station, which consists of various weather instruments integrated into one. Rain-induced attenuation data were measured using a satellite beacon setup and the measured values were compared with existing prediction models. Results shows significant attenuation levels ranging from 8.4 to 11.5 dB, demonstrating substantial impact of intense rainfall on radio wave propagation above 10 GHz. The ITU-R model showed mixed performance across locations. Its annual cumulative rain rate at 0.1% (35.6mm/hr.) and 0.01% (95.2mm/hr.) of time exceedance means that satellite link is expected to be available for 99.9% and 99.99% of time with rain attenuation exceeding the threshold of approximately 526min and 52.6min equivalent number of minutes annually. However, in Akure, the ITU-R model underestimated the measured values at 0.1% (38.83mm/hr.) and 0.01% (106.8mm/hr.) respectively with about 11% deviation and it overestimated ENUGU site values within the same range of values. This implies that any rain prediction done with ITU-R values in Akure location, the satellite link availability will experience a down time for more than 526min and 52.6min annually. While the overestimated will experience shorter equivalent number of times like in the case of Enugu within the same range of values. At GUSAU and ABUJA sites, the ITU model consistently overestimated measured rain attenuation. Among the attenuation prediction models, Moupfouma model demonstrated superior accuracy, closely following measured rain attenuation values across all locations.

Design and Fabrication of Vacuum Dehydration Machine for Cassava Processing

Raji, N.A. — 2025-10-24

The importance of removing cyanide content in cassava mash cannot be overemphasized. The removal of cyanide by dehydration not only makes the cassava safe for consumption but also could improve it flavour. The design of a vacuum dehydration machine used for cassava processing is presented. The analysis of the various components including the slurry drum, belt drive system, transmission shaft, and the selection of a suitable prime mover was done based on the power requirement for the operation of the machine. The design analysis revealed that the machine may be driven by a 1 horsepower engine. The total length of flat-belt to drive the pulley was at an angle of lap on the smaller pulley of 2.53 rad. The resultant load act on a shaft of diameter 21 mm with maximum bending moment of 16.51 Nm. Total weights of drum filter and pipes acting on the reservoir is 124N. The result of the performance evaluation shows that the machine performance improves with increasing vacuum pressure up to a threshold where the moisture content remain constant. The result exposed the effectiveness and efficiency of the machine at 95%.

Self-Healing Health Records: Autonomous Data Integrity Models for Corruption-resistant Electronic Medical Records (EMR)

Seun Michael Oyekunle — 2025-10-21

This study investigates self-healing health records through the development of autonomous data integrity models for corruption-resistant Electronic Medical Records (EMRs). Using a convergent parallel mixed-methods design, a systematic review of 204 studies (2010–2025) identified key EMR integrity challenges, notably data completeness and accuracy issues. Comparative analysis of 72 self-healing technologies revealed database-level systems as most effective, demonstrating high reliability and rapid recovery performance. A four-phase framework: Detection, Diagnosis, Recovery, and Learning was developed, reducing error recurrence by over 60%. Case studies, including implementation at the Cleveland Clinic, confirmed significant improvements in patient safety and cost efficiency. While integration complexity and moderate user acceptance remain challenges, the framework provides a scalable, evidence-based foundation for autonomous EMR systems. This research advances healthcare informatics by enhancing data integrity, promoting patient safety, and paving the way for intelligent, self-correcting health record infrastructures.