Article
Automation & Control Systems
Guangzhu Peng, C. L. Philip Chen, Wei He, Chenguang Yang
Summary: This article introduces a neural network based admittance control scheme for robotic manipulators to achieve compliant behavior and estimate external torque. An adaptive neural controller with dynamic learning framework is developed to deal with uncertainties in the robot system. Experiments on the Baxter robot have validated the effectiveness of the proposed method.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2021)
Article
Mathematics
Dwi Sudarno Putra, Seng-Chi Chen, Hoai-Hung Khong, Chin-Feng Chang
Summary: An observer is a crucial part of sensorless control of a PMSM, relying on mathematical equations and motor parameter values. This article presents an intelligent observer based on machine learning, which utilizes a modified Jordan neural network to process sensor inputs and generate position and speed feedback information. The observer is trained using simulation data from five PMSMs with different parameters, with successful control demonstrated in both simulation and experimental hardware. The average positioning error was 0.0078 p.u in simulation and 0.0100 p.u in experimentation.
Article
Automation & Control Systems
Chenguang Yang, Guangzhu Peng, Long Cheng, Jing Na, Zhijun Li
Summary: This paper proposes a force sensorless control scheme based on neural networks with an observer approach for impedance control, combined with radial basis function NN to compensate for uncertainties, and an error transformation algorithm to achieve prescribed tracking precision.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Yu Lu, Changyun Wen, Tielong Shen, Weidong Zhang
Summary: This article investigates the formation scaling control problem of ASVs with uncertainties and input saturation, developing a novel bearing-based adaptive neural formation scaling control scheme. The main idea is to use a small number of leader ASVs to program trajectories and steer remaining ASVs to follow leaders via adaptive neural techniques, achieving desired formation scaling maneuver with guaranteed formation errors.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Engineering, Multidisciplinary
George-Iulian Uleru, Mircea Hulea, Adrian Burlacu
Summary: This research introduces a structure that controls finger motions using spiking neural networks, demonstrating its feasibility and accuracy in simulating human finger movements. Additionally, compared to traditional microcontrollers, using SNN provides enhanced biological plausibility.
Article
Engineering, Electrical & Electronic
Luy Nguyen Tan, Thanh Pham Cong, Duy Pham Cong
Summary: This article proposes neural network based observer schemes and a sensorless robust optimal control scheme for partially unknown permanent magnet synchronous motors with disturbances and saturating voltages. The schemes include NN-observer schemes to estimate back-electromotive force and tracking errors of rotor position and speed, as well as a sensorless saturated robust optimal control scheme dealing with general disturbances and saturating voltages. The effectiveness of the proposed schemes is tested through simulations and comparative experiments.
IEEE TRANSACTIONS ON POWER ELECTRONICS
(2021)
Article
Chemistry, Analytical
Jokin Uralde, Eneko Artetxe, Oscar Barambones, Isidro Calvo, Pablo Fernandez-Bustamante, Imanol Martin
Summary: Piezoelectric actuators (PEA) are high-precision devices used in applications requiring micrometric displacements. However, PEAs present non-linearity phenomena that introduce drawbacks at high precision applications. This paper presents a high precision control scheme to be used at PEAs based on the model-based predictive control (MPC) scheme, which uses artificial neural networks (ANN) to simplify the obtaining of the model. The experimental results show that the MPC control strategy achieves higher accuracy at high precision PEA applications.
Article
Engineering, Electrical & Electronic
Adel Rahoui, Ali Bechouche, Hamid Seddiki, Djaffar Ould Abdeslam
Summary: A new neural network-based virtual flux estimator is proposed for sensorless control of a pulsewidth modulation rectifier under unbalanced and distorted grid conditions. Through optimal tuning, accurate and fast estimation is achieved, with feasibility and robustness verified through experiments. This approach demonstrates better current waveforms and reduced settling time compared to conventional methods under unbalanced grid conditions.
IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS
(2021)
Article
Optics
Carlo M. Valensise, Alessandro Giuseppi, Giulio Cerullo, Dario Polli
Summary: This study demonstrates the ability of a deep reinforcement learning agent to generate a long-term stable white-light continuum in a bulk medium without prior knowledge of the system dynamics. It shows that deep reinforcement learning can effectively be used to control complex nonlinear optical experiments.
Article
Green & Sustainable Science & Technology
Sven Myrdahl Opalic, Morten Goodwin, Lei Jiao, Henrik Kofoed Nielsen, Mohan Lal Kolhe
Summary: This paper focuses on the application of reinforcement learning techniques in controlling the energy optimization problem of battery energy storage systems in a smart warehouse. By using an intelligent energy management system and an optimization algorithm, the goal of minimizing energy costs is achieved.
JOURNAL OF CLEANER PRODUCTION
(2022)
Article
Chemistry, Analytical
Marcel Nicola, Claudiu-Ionel Nicola, Cosmin Ionete, Dorin Sendrescu, Monica Roman
Summary: This paper summarizes a robust controller for a permanent magnet synchronous motor (PMSM) that can maintain performance over a wide range of parameter variations. It also introduces the synthesis and implementation of the robust control, as well as an improved version using a reinforcement learning twin-delayed deep deterministic policy gradient (RL-TD3) agent. Numerical simulations validate the superior performance of the RL-TD3 agents in both sensored and sensorless control.
Article
Construction & Building Technology
Yue Li, Zheming Tong
Summary: The study developed a DL-based MPC framework for realtime control of building thermal environment, integrating encoder-decoder recurrent neural network and co-simulation platform. The DL-based MPC algorithm achieved approximately 4% and 7% energy savings compared to adaptive and traditional PID control methods, showing promising application prospects for building automation.
JOURNAL OF BUILDING ENGINEERING
(2021)
Article
Mathematics
Harshit Mohan, Gopal Agrawal, Vibhu Jately, Abhishek Sharma, Brian Azzopardi
Summary: To reduce pollution and energy consumption, electric vehicles (EVs) are gaining more attention in the automotive industry. High efficiency, compactness, lightweight, low cost, and easy recyclability are desired in the electric motors used in EVs. Various motor control strategies and sensorless speed control techniques are employed to achieve better dynamic performance and increased reliability.
Article
Energy & Fuels
Linfei Yin, Bin Zhang
Summary: This paper proposes a real-time control framework based on a relaxed deep generative adversarial network method to address the frequency stability and economic dispatch issues of integrated energy systems. By combining deep learning and relaxed operation, the proposed method can generate multiple convergent and high-performance generation commands, thereby improving energy efficiency and achieving multi-energy complementarity.
Article
Chemistry, Multidisciplinary
Huapeng Zhang, Saewoong Oh, Manmatha Mahato, Hyunjoon Yoo, Il-Kwon Oh
Summary: This study presents a novel knot-architectured fabric actuator (KAFA) with superior features such as self-locking crossing, mechanical robustness, low-cost fabrication, high force generation, and large actuation strain. KAFA operates through the shape recovery of constituent nitinol fibers heated by Joule heating, resulting in reliable actuation and exceptionally high force. The study also demonstrates the potential applications of KAFA in wearable actuation devices and adaptive surfaces.
ADVANCED FUNCTIONAL MATERIALS
(2022)
Article
Engineering, Electrical & Electronic
E. Asua, V. Etxebarria, J. Feutchwanger, J. Portilla
SENSORS AND ACTUATORS A-PHYSICAL
(2019)
Review
Physics, Applied
M. Goiriena-Goikoetxea, D. Munoz, I. Orue, M. L. Fernandez-Gubieda, J. Bokor, A. Muela, A. Garcia-Arribas
APPLIED PHYSICS REVIEWS
(2020)
Article
Materials Science, Multidisciplinary
D. de Cos, N. Lete, M. L. Fdez-Gubieda, A. Garcia-Arribas
JOURNAL OF MAGNETISM AND MAGNETIC MATERIALS
(2020)
Article
Chemistry, Analytical
Alfredo Garcia-Arribas
Article
Chemistry, Analytical
Beatriz Sisniega, Ariane Sagasti Sedano, Jon Gutierrez, Alfredo Garcia-Arribas
Article
Physics, Applied
M. L. Fdez-Gubieda, J. Alonso, A. Garcia-Prieto, A. Garcia-Arribas, L. Fernandez Barquin, A. Muela
JOURNAL OF APPLIED PHYSICS
(2020)
Article
Chemistry, Physical
Beatriz Sisniega, Jon Gutierrez, Virginia Muto, Alfredo Garcia-Arribas
Article
Engineering, Electrical & Electronic
Julio Lucas, Victor Etxebarria
Summary: This article completes the complex formalism of linear beam dynamics by obtaining a general differential equation and solving it, showing that the general transformation of a linear beam line is a complex Moebius transformation. This opens up the possibility of studying the effect of the beam line on complete regions of the complex plane, rather than just a single point. Additionally, a new description of betatron functions in case of a mismatched injection in a circular accelerator is provided.
IEEE TRANSACTIONS ON NUCLEAR SCIENCE
(2021)
Article
Engineering, Electrical & Electronic
Beatriz Sisniega, Jon Gutierrez, Alfredo Garcia-Arribas
Summary: This study uses a magnetoelastic resonance sensor to monitor the precipitation reaction of calcium oxalate. The sensor measures the changes in resonance frequency to improve the detection resolution and explores the impact of solution concentration on the precipitation reaction. The results show that the sensor can track the reaction in solutions as low as 1 mM and detect a mass of precipitate as small as 2 μg.
IEEE TRANSACTIONS ON MAGNETICS
(2022)
Article
Chemistry, Multidisciplinary
Lourdes Marcano, Inaki Orue, David Gandia, Lucia Gandarias, Markus Weigand, Radu Marius Abrudan, Ana Garcia-Prieto, Alfredo Garcia-Arribas, Alicia Muela, M. Luisa Fdez-Gubieda, Sergio Valencia
Summary: The use of nanomagnets in biomedical applications has increased in recent years. However, the magnetic anisotropy of individual magnetic nanostructures is still mostly unknown. Current methods for measuring magnetic signals are limited in spatial resolution or magnetic field strength, and cannot be applied to nanomagnets in biological systems. In this study, we present a hybrid experimental/theoretical method to determine the magnetic anisotropy constant and magnetic easy axis of individual magnetic nanostructures embedded in biological systems.
Article
Chemistry, Multidisciplinary
Estibaliz Asua, Jon Gutierrez-Zaballa, Oscar Mata-Carballeira, Jon Ander Ruiz, Ines del Campo
Summary: This study aims to investigate the variation of passenger's comfort evaluation parameters depending on driving style, car, and road. The results show a high dependence of comfort evaluation variables on road type, and demonstrate that driving style and vehicle dynamics amplify or attenuate these values. Additionally, it has been found that longitudinal and lateral accelerations have a greater effect on comfort than vertical accelerations.
APPLIED SCIENCES-BASEL
(2022)
Article
Chemistry, Multidisciplinary
Jorge Feuchtwanger, Victor Etxebarria, Joaquin Portilla, Josu Jugo, Inigo Arredondo, Inari Badillo, Estibaliz Asua, Nicolas Vallis, Mikel Elorza, Benat Alberdi, Rafael Enparantza, Iratxe Ariz, Inigo Munoz, Unai Etxebeste, Inaki Hernandez
Summary: This paper describes a new compact linear proton accelerator project (LINAC 7) that has been designed and built at the Beam Laboratory of the University of the Basque Country (UPV/EHU). The project is a collaboration between the University, a research technology center, and a private company, aiming to develop a compact, low-current proton accelerator capable of accelerating particles up to 7 MeV. The paper provides an overview of the accelerator design, summarizes the progress and testing of the built components, and discusses the ongoing design of components needed to achieve the desired energy of 7 MeV.
APPLIED SCIENCES-BASEL
(2022)
Article
Engineering, Multidisciplinary
Victor Etxebarria, Jorge Feuchtwanger, Joaquin Portilla, Josu Jugo, Inigo Arredondo, Inari Badillo, Estibalitz Asua, Rafael Enparantza, Iratxe Ariz, Unai Etxebeste, Inaki Hernandez
Summary: Linac 7 is a new generation linear proton accelerator designed and built by the Beam Laboratory of the University of the Basque Country UPV/ EHU. The Linac 7 project aims to produce pharmaceuticals locally near large clinical centers, which is considered one of its most important health applications. Currently, medical radioisotopes are manufactured outside hospitals, leading to issues such as long transports, rapid decay of radioisotopes during transport, and limited options for their use. In contrast, the compact Linac 7 can meet biomedical needs, including specific doses of pharmaceuticals on demand from medical staff in hospitals.
Article
Chemistry, Multidisciplinary
Eider Berganza, Miriam Jaafar, Jose A. Fernandez-Roldan, Maite Goiriena-Goikoetxea, Javier Pablo-Navarro, Alfredo Garcia-Arribas, Konstantin Guslienko, Cesar Magen, Jose M. De Teresa, Oksana Chubykalo-Fesenko, Agustina Asenjo
Article
Chemistry, Multidisciplinary
David Gandia, Lucia Gandarias, Lourdes Marcano, Inaki Orue, David Gil-Carton, Javier Alonso, Alfredo Garcia-Arribas, Alicia Muela, Ma Luisa Fdez-Gubieda