AAAyokunle Awelewa

Ayokunle Awelewa

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Ayokunle Awelewa has a total of 56 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    EVE: A Task-Adaptive Optimization Algorithm for Balanced Multitask Learning

    by Racheal Akinboyo, Olufunke Oladipupo, Ayokunle Awelewa

    Optimization plays a central role in multitask learning (MTL), yet widely used optimizers such as Adam and AdamW assume uniform parameter updates across tasks, often leading to optimization bias where simpler objectives dominate training. This paper proposes Extensive Variational Estimation (EVE), a task-adaptive optimizer that extends AdamW by integrating exponential moving average–based task-specific learning rate modulation, scheduled decay of shared parameters, and task-indexed parameter management. Unlike existing optimizers that remain task-agnostic, EVE dynamically adjusts updates according to per-task progress, thereby preserving stability across heterogeneous objectives. The algorithm was empirically validated on three representative multitask domains: image-level classification with pixel-level segmentation on the BUSI breast ultrasound dataset, classification with survival analysis on the SEER prostate cancer cohort, and semantic segmentation with depth estimation and surface normal on the NYUv2 dataset. With a higher Dice coefficient (0.795 ± 0.013) and competitive classification accuracy, EVE performed better on BUSI than Adam and AdamW. On SEER, EVE demonstrated stable performance, improved calibration, and a lower integrated Brier score (0.471 ± 0.035). EVE achieved semantic segmentation performance comparable to Adam (mIoU ≈ 0.370) and higher than AdamW, while maintaining depth estimation performance close to the baselines and exhibiting reduced variance across seeds, on NYUv2. Ablation and Sensitivity experiments proved the importance of EVE's smoothing and minimum learning rate components, and statistical testing revealed consistent directional advantages over Adam in BUSI and SEER, and over AdamW in NYUv2. The proposed optimizer is applicable to multitask learning settings where task heterogeneity and imbalance challenges exist.

  • Open Access

    Cuckoo search algorithm approach for optimal placement and sizing of distribution generation in radial distribution networks

    by Kayode Ojo, Seyi Fanifosi, Awelewa Ayokunle, Isaac Samuel

    Radial distribution networks (RDNs) often experience power loss due to improper distribution generation (DG) allocation. Strategic DG placement can reduce power loss, minimize costs, and improve voltage profiles and stability. This research optimizes DG placement and sizing in RDNs using the cuckoo search algorithm (CSA). The objective function considers losses across all network branches, and CSA identifies optimal DG locations and sizes. Tested on IEEE 33-bus, IEEE 69-bus, and Nigeria's Imalefalafia 32-bus RDN, the Cuckoo Search technique results in optimal DG locations at buses 6, 50, and 18 with corresponding sizes of 2.4576, 1.852, and 2.718 MW, respectively. Voltage improvements are 0.9509, 0.9817, and 0.9821 p.u, while total active and reactive power losses for IEEE 33-bus are reduced by 49.03% and 45.00%, and for IEEE 69-bus by 63.67% and 61.14%. The CSA approach significantly enhances voltage profiles and reduces power losses in these networks.

  • Open Access

    New droop-based control of parallel voltage source inverters in isolated microgrid

    by Timilehin F. Sanni, Ayokunle A. Awelewa, Anthony U. Adoghe, Adeola Balogun, Tobi Somefun

    Microgrids, featuring distributed generators like solar energy and hybrid energy storage systems, represent a significant step in addressing challenges related to the greenhouse effect and outdated transmission infrastructures. The operation and control of islanded microgrids, particularly in terms of grid voltage and frequency, rely on the synchronization of multiple parallel inverters connected to the distributed generators. However, to determine the necessary grid parameters for effective control, the presence of circulating currents from unbalanced grid voltages arises as a challenge. This situation necessitates the development of a new approach to achieve phase angle locking for grid synchronization, with the aim of maintaining the voltage within acceptable limits in islanded microgrids. This objective is realized through the creation of a microgrid network model, design of an adaptive filter, utilizing the double second-order generalized integrator–phase-locked loop (DSOGI-PLL), for dynamic voltage transformation. The design is evaluated by simulation using MATLAB/Simulink. The primary goal is to investigate the DSOGI-PLL-based droop control and compare its performance with the conventional synchronous reference frame–phase-locked loop (SRF-PLL) control approach. Notably, the DSOGI-PLL successfully eliminates the ripples in phase angle estimation, consequently enhancing the quality of voltage output in the microgrid.

  • Open Access

    Detection of Fruit Ripeness and Defectiveness Using Convolutional Neural Networks

    by Joshua S. Mommoh, James L. Obetta, Samuel N. John, Kennedy Okokpujie, Osemwegie N. Omoruyi, Ayokunle A. Awelewa

    The classification of fruit ripeness and detection of defects are critical processes in the agricultural industry to minimize losses during commercialization. This study evaluated the performance of three Convolutional Neural Network (CNN) architectures—Extreme Inception Network (XceptionNet), Wide Residual Network (Wide ResNet), and Inception Version 4 (Inception V4)—in predicting the ripeness and quality of tomatoes. A dataset comprising 2,589 images of beef tomatoes was assembled from Golden Fingers Farms and Ranches Limited, Abuja, Nigeria. The samples were categorized into six classes representing five progressive ripening stages and a defect class, based on the United States Department of Agriculture (USDA) colour chart. To enhance the dataset's size and diversity, image augmentation through geometric transformations was employed, increasing the dataset to 3,000 images. Fivefold cross-validation was conducted to ensure a robust evaluation of the models' performance. The Wide ResNet model demonstrated superior performance, achieving an average accuracy of 97.87%, surpassing the 96.85% and 96.23% achieved by XceptionNet and Inception V4, respectively. These findings underscore the potential of Wide ResNet as an effective tool for accurately detecting ripeness levels and defects in tomatoes. The comparative analysis highlights the effectiveness of deep learning (DL) techniques in addressing challenges in agricultural automation and quality assessment. The proposed methodology offers a scalable solution for implementing automated ripeness and defect detection systems, with significant implications for reducing waste and improving supply chain efficiency.


  • Open Access

    Progressive energy management technique for smart load control

    by Timilehin Sanni, Mobolade Olawale, Amarachi Attah, Anthony Adoghe, Ayokunle Awelewa

    Progressive energy management technique for smart load control

  • Open Access

    Control strategies in enhanced stand-alone mini-grid operations for the NESI–an overview

    by Adeola Balogun, Ayobami Olajube, Ayokunle Awelewa, Sodiq Agoro, Frank Okafor, Timilehin Sanni, Isaac Samuel, Adejumoke Ajilore

    Diverse control strategies for enhancing operations of isolated distribution grids are reviewed. Such distribution grids are called mini-grids or micro-grids, depending on their power flow capabilities. Robust control schemes identified in other climates for mini-grid and micro-grid operations are yet to be fully explored in the Nigerian electricity supply industry (NESI). Sustainable control strategies suitable for isolated distribution grids in the NESI predicate on capabilities for diverse objectives, such as energy conservation, affordability, efficient power throughput, and utilization, for enhanced resiliency and reliability. Consequently, the distributed control system in hierarchical layers is identified as a suitable choice for mini-grid operations in Nigeria because of its robustness in scalability and in energy conservation. However, the model predictive control (MPC) scheme is observed to be uniquely applicable in all of the hierarchical control layers. Therefore, a cascade-free MPC with improved robustness against sensitivity to system parameter variation is presented at the primary control layer for an H8 voltage source inverter (VSI) used for grid integration of the solar photovoltaic (PV) system. The H8 inverter gives more promising mitigation strategies against common-mode voltage and leakage current. Moreover, the control of DC link voltage for maximum power point tracking (MPPT) is achieved by the H8 inverter, thereby eliminating the need for a separate converter for MPPT. Thus, sustainability is achieved.


  • Open Access

    State of charge estimation based on a modified extended Kalman filter

    by Koto Omiloli, Ayokunle Awelewa, Isaac Samuel, Oghorchukwuyem Obiazi, James Katende

    The global transition from fossil-based automobile systems to their electric-driven counterparts has made the use of a storage device inevitable. Owing to its high energy density, lower self-discharge, and higher cycle lifetime the lithium-ion battery is of significant consideration and usage in electric vehicles. Nevertheless, the state of charge (SOC) of the battery, which cannot be measured directly, must be calculated using an estimator. This paper proposes, by means of a modified priori estimate and a compensating proportional gain, an improved extended Kalman filter (IEKF) for the estimation task due to its nonlinear application and adaptiveness to noise. The improvement was achieved by incorporating the residuals of the previous state matrices to the current state predictor and introducing an attenuating factor in the Kalman gain, which was chosen to counteract the effect of the measurement and process noise resulting in better accuracy performance than the conventional SOC curve fitting-based estimation and ampere hour methods. Simulation results show that the standard EKF estimator results in performance with an error bound of 12.9% due to an unstable start, while the modified EKF reduces the maximum error to within 2.05% demonstrating the quality of the estimator.


  • Open Access

    Robust hybrid estimator for the state of charge of a lithium-ion battery

    by Ayokunle Awelewa, Koto Omiloli, Isaac Samuel, Ayobami Olajube, Olawale Popoola

    The use of batteries for diverse energy storage applications is increasing, primarily because of their high energy density, and lithium-ion batteries (LiBs) are of particular significance in this regard. However, designing estimators that are robust to compute the state of charge (SOC) of these batteries in the presence of disturbance signals arising from different battery types remains a challenge. Hence, this paper presents a hybrid estimator that combines the extended Kalman filter (EKF) and sliding mode observer (SMO) via a switching function and tracking closed loop to achieve the qualities of noise cancellation and disturbance rejection. Hybridization was carried out in such a way that the inactive observer tracks the output of the used observer, simultaneously feeding back a zero-sum signal to the input gain of the used observer. The results obtained show that noise filtering is preserved at a convergence time of .01 s. Also, the state of charge estimation interval improves greatly from a range of [1, .93] and [.94, .84] obtained from the extended Kalman filter and sliding mode observer, respectively, to a range of [1, 0], in spite of the added disturbance signals from a lithium–nickel (INR 18650) battery type.


  • Open Access

    Experimental data on power quality assessment at point of common coupling of a steel mill to an electric power grid

    by Emenike Ugwuagbo, Adeola Balogun, Ayobami Olajube, Osita Omeje, Ayokunle Awelewa, Shehu Abba-Aliyu


  • Open Access

    ARTIFICIAL NEURAL NETWORK BASE SHORT-TERM ELECTRICITY LOAD FORECASTING: A CASE STUDY OF A 132/33KV TRANSMISSION SUB-STATION

    by Isaac Adekunle Samuel, Segun Ekundayo, Ayokunle Awelewa, Tobiloba Emmanuel Somefun, Adeyinka Adewale

    Artificial Neural Network Base Short-Term Electricity Load Forecasting: A Case Study of a 132/33kv Transmission Sub-Station