Electrical and Electronic Engineering - Journal Articles
Permanent URI for this collection
Browse
Recent Submissions
Item Near real-time machine learning framework in distribution networks with low-carbon technologies using smart meter data(Elsevier Ltd, 2025) Dokur, Emrah; Erdogan, Nuh; Sengor, Ibrahim; Yüzgeç, Uğur; Hayes, Barry P.The widespread adoption of low carbon technologies (LCTs) such as photovoltaics, electric vehicles, heat pumps, and energy storage units introduces challenges to distribution network congestion and power quality, particularly raising concerns about voltage stability. Enhanced voltage visibility in low-voltage (LV) networks is increasingly vital for an active grid management, making efficient voltage forecasting tools essential. This study introduces a novel data-driven approach for forecasting node voltages in LCT-enriched LV distribution networks. Using time series of power measurements from smart meter data, the study integrates an Extreme Learning Machine (ELM) with the Single Candidate Optimizer (SCO) to enhance both computational efficiency and forecasting accuracy. The model is validated using a realistic LCT-enriched LV network dataset and benchmarked against several established machine learning models. Results demonstrate that the SCO algorithm effectively optimizes ELM parameters, achieving up to a 17-fold reduction in computation time compared to the fastest metaheuristic methods implemented. The proposed model demonstrated superior accuracy, with an average voltage deviation of 0.56%. While the computation time per node achieved is not yet suitable for real-time applications, the study proves that SCO significantly enhances ELM performance.Item A digital twin of intelligent robotic grasping based on single-loop-optimized differentiable architecture search and sim-real collaborative learning(Springer Nature, 2024-10-14) Jiao, Qing; Hu, Weifei; Hao, Guangbo; Cheng, Jin; Peng, Xiang; Liu, Zhenyu; Tan, Jianrong; National Natural Science Foundation of China; Key Research and Development Program of Zhejiang Province; Natural Science Foundation of Zhejiang ProvinceThe effectiveness of deep learning models for vision-based intelligent robotic grasping (IRG) tasks typically hinges upon the deep neural network (DNN) architecture as well as the task-oriented annotated training samples. Nevertheless, current methods applied for designing DNN architectures depend on human expertise or discrete search by evolution and reinforcement learning algorithms, which leads to enormous computational cost. Moreover, DNNs trained solely on simulation-labeled data face challenges in direct real-world deployment. In response to these concerns, this paper proposes a new stable and fast differentiable architecture search method (DARTS) based on a single-loop optimization framework, named single-loop-optimized DARTS (SLO-DARTS). This method enables simultaneous updates to the weights and architecture parameters of neural networks by continuously relaxing the discrete search space. Additionally, a digital twin (DT) framework integrating the Grasp-CycleGAN method is developed to minimize the visual gap between simulated and real-world IRG scenarios, enhancing the transferability of DNNs trained in simulation. The DT framework can not only enhance the IRG accuracy but also save the costly expense of large-scale real labeled data collection. Experiments demonstrate that the proposed SLO-DARTS method achieves a time-efficient optimization process while delivering a DNN with improved prediction accuracy compared to the original dual-loop-optimized DARTS method. The developed DT framework produces IRG accuracies of 92.6%, 86.3%, and 83.7% for single household objects, single adversarial objects, and cluttered objects, respectively.Item Kineto-static analysis of a hybrid manipulator consisting of rigid and flexible limbs with locking function for planar shape morphing(Elsevier Ltd., 2024-10-05) Zhao, Yinjun; Xi, Fengfeng; Hao, Guangbo; Tian, Yingzhong; Li, Long; Wang, Jieyu; National Natural Science Foundation of China; Shanghai Pujiang Programme; China Scholarship CouncilThe incorporation of lockable passive backbones into active compliant morphing systems efficiently results in lightweight, high-load, and large deformation systems. However, there exist challenges in kineto-static analysis due to the interaction between rigid reconfigurable kinematic constraints and the nonlinear deformation of actuated flexible limbs. This paper addresses these issues by developing a kineto-static method to analyze the motion in a novel planar 3-DOF shape-morphing manipulator. The manipulator features two actuated flexible limbs with a lockable variable geometry truss (LVGT). In this study, two isostatic topologies are selected for reconfigurable motion control under external tip loads. A multi-step sequential control strategy is proposed to maneuver the manipulator's platform for desired poses. Then, a constrained-trajectory-based kinematic model is proposed for an inverse kinematic solution considering motion planning. Subsequently, a kineto-static model is introduced, considering constraints from rigid and flexible limbs, aiming to distribute distributing redundant actuation forces. Finally, nonlinear finite element analysis (FEA) and experiments are carried out to validate the effectiveness of the proposed method.Item Analysis and design optimization of a compliant robotic gripper mechanism with inverted flexure joints(Elsevier B.V., 2024-09-02) Kuresangsai, Pongsiri; Cole, Matthew O. T.; Hao, Guangbo; Chiang Mai UniversityFlexure-jointed grippers provide compliant grasping capability, have low-cost and flexible manufacturing, and are insusceptible to joint friction and wear. However, their grasp stiffness can be limited by flexure compliance such that loss-of-grasp is prone to occur for high object loads. This paper examines the application of inverted-flexure joints in a cable-driven gripper that can avoid flexure buckling and greatly enhance grasp stiffness and stability. To analyze behavior, an energy-based kinetostatic model is developed for a benchmark grasping problem and validated by hardware experiments. A multi-objective design optimization study is conducted, considering key metrics of peak flexure stress, grasp stiffness, and cable actuation force. Results show that the inverted-flexure design has significantly higher grasp stiffness (63% higher in a targeted design optimization) and requires lower actuation forces (¿20% lower in all optimization cases), compared with equivalent direct-flexure designs. An application study is conducted to validate the predicted operating performance under gravity loading of the grasped object. The results demonstrate that stable and high stiffness grasping can be achieved, even under overload conditions that lead to loss-of-grasp for conventional direct-flexure designs.Item The role of FPGAs in Modern Option Pricing techniques: A survey(MDPI, 2024-08-12) O'Mahony, Aidan; Hanzon, Bernard; Popovici, Emanuel; Science Foundation Ireland; Intel Corporation; Dell TechnologiesIn financial computation, Field Programmable Gate Arrays (FPGAs) have emerged as a transformative technology, particularly in the domain of option pricing. This study presents the impact of Field Programmable Gate Arrays (FPGAs) on computational methods in finance, with an emphasis on option pricing. Our review examined 99 selected studies from an initial pool of 131, revealing how FPGAs substantially enhance both the speed and energy efficiency of various financial models, particularly Black–Scholes and Monte Carlo simulations. Notably, the performance gains—ranging from 270- to 5400-times faster than conventional CPU implementations—are highly dependent on the specific option pricing model employed. These findings illustrate FPGAs’ capability to efficiently process complex financial computations while consuming less energy. Despite these benefits, this paper highlights persistent challenges in FPGA design optimization and programming complexity. This study not only emphasises the potential of FPGAs to further innovate financial computing but also outlines the critical areas for future research to overcome existing barriers and fully leverage FPGA technology in future financial applications.