Skip to main content

CONTROL Y GESTION DE ENERGIA EN VEHICULOS ELECTRICOS HIBRIDOS CON PILAS DE COMBUSTIBLE

RTI2018-096001-B-C32

•
Funding agency name Agencia Estatal de Investigación
Funding agency acronym AEI
Program Programa Estatal de I+D+i Orientado a los Retos de la Sociedad
Call Proyectos I+D+i Retos de Investigación
Call year 2018
State Plan 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Beneficiary centre UNIVERSITAT POLITECNICA DE CATALUNYA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 31
Encontrada(s) 1 página(s)

DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Contribución de congreso (ConferenceOutput)

Implementación y validación experimental del control de un sistema híbrido basado en pilas de combustible para vehículos eléctricos

  • Anderson, Jorge L.
  • Moré, Jerónimo J.
  • Puleston, Pablo Federico
  • Roda, Vicente
  • Costa Castelló, Ramon
Trabajo presentado en el XXVII Congreso Argentino de Control Automático (AADECA), celebrado de forma virtual del 28 al 30 de octubre de 2020, En este trabajo se presenta el diseño, implementación y validación experimental de un controlador por modos ´ deslizantes de segundo orden aplicado a un modulo de generación´ hibrida basado en una pila de combustible y supercapacitores. El controlador desarrollado se encuentra basado en el algoritmo Super-Twisting. Para simplificar el diseño de las acciones de ˜ control del sistema, se emplea una transformación no lineal que ´ permite independizar el diseño de las variables de deslizamiento. ˜ El sistema de control propuesto es ensayado experimentalmente en una plataforma que incorpora una pila de combustible de 1.2kW y supercapacitores de uso comercial. El modulo ´ de generación hıbrida controlado es entonces evaluado para satisfacer la demanda de potencia de un vehículo eléctrico en ´ condiciones de conducción urbana, con la presencia de múltiples ´ etapas de aceleración y desaceleración. Los resultados obtenidos ´ muestran una buena regulación de la tensión del bus común y un ´ seguimiento robusto de las referencias impuestas, aun frente a la ´ presencia de fuertes variaciones y perturbaciones de la potencia de carga, Este trabajo fue realizado bajo el apoyo de la FI-UNLP, CONICET y Agencia I+D+i. El trabajo fue parcialmente financiado por el proyecto nacional de Espana DOVELAR ref. ˜ RTI2018-096001-B-C32 (MCIU/AEI/FEDER, UE), la Agencia Nacional de Investigacion de España a través del Sello de ´ Excelencia Mar´ıa de Maeztu al IRI (MDM-2016-0656) y la Asociación Universitaria Iberoamericana de Postgrado (AUIP).
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Contribución de congreso (ConferenceOutput)

Control no lineal adaptativo con identificación dispersa, Nonlinear adaptive control with sparse identification

  • Domingo-Enrich, Roger
  • Cecilia, Andreu
  • Costa Castelló, Ramon
Trabajo presentado en las XLII Jornadas de Automática, celebradas en Castellón (España), del 1 al 3 de septiembre de 2021, [EN] This work presents a controller with parameter estimation focused on non-linear, unknown and time-variant systems. The proposal consists of an on-line learning algorithm to estimate the system equations, and a controller that uses this information to stabilize the plant. Our approach is to combine a non-linear observer and a parameter estimator using sparse optimization techniques. The resulting algorithm operates in two stages: the learning stage, where unknown functions are learned on-line, and the execution stage, where the plant is controlled with pre-established objectives. The proposed algorithm can autonomously decide what is its current stage by using a learning metric. We have validated our algorithm using numerical simulation in a roto-magnet plant., [ES] En este trabajo se propone un controlador con estimación de parámetros enfocada a sistemas no lineales, desconocidos y cambiantes en el tiempo. Se presenta un algoritmo de aprendizaje en línea para estimar las ecuaciones del sistema y un controlador que usa esta información para estabilizar la planta. Para ello, se han combinado un observador no lineal y un estimador de parámetros con técnicas de optimización dispersa. En el algoritmo hay dos fases diferenciadas: la fase de aprendizaje, donde se aprenden en línea las funciones desconocidas, y la fase de ejecución, donde se controla la planta con unos objetivos preestablecidos. El algoritmo propuesto es capaz de decidir de forma autónoma en qué fase se encuentra, mediante una métrica de aprendizaje. Este algoritmo se ha validado por simulación numérica en una planta roto-magnet., Este trabajo se ha realizado con el apoyo de la Agencia Estatal de Investigación española a través del sello de excelencia de María de Maeztu, IRI (MDM-2016-0656) y a través del proyecto DOVELAR (ref. RTI2018-096001-B-C32).
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Contribución de congreso (ConferenceOutput)

Estimación del estado de carga y coeficientes de difusión en baterías de flujo redox

  • Clemente, Alejandro
  • Cecilia, Andreu
  • Costa Castelló, Ramon
Trabajo presentado en las XLII Jornadas de Automática, celebradas en Castellón (España), del 1 al 3 de septiembre de 2021, [ES] En este articulo se presenta el diseño de un observador de estado y parámetros para la estimación del estado de carga y algunos de los parámetros de difusión de una batería de flujo redox. El observador es de alta ganancia y se basa en un modelo dinámico no lineal, que utiliza un enfoque realista diferenciando entre las concentraciones totales de electrolito en las dos partes del sistema. Se muestra que el estado de carga se puede estimar mediante la medición del voltaje de celda y un observador de alta ganancia. No obstante, la precisión de la estimación se ve afectada por la incertidumbre en los coeficientes de difusión del sistema. Por esta razón, el observador se robustece mediante una estimación de parámetros adaptativos de inmersión e invariancia. Finalmente, los resultados se validan en una simulación numérica., [EN] This paper presents an observer for the estimation of the state of charge of redox flow batteries. The observer is high gain and is based on a non-linear dynamic model, which uses a realistic approach differentiating between the total electrolyte concentrations in the two parts of the system. It is shown that the state of charge can be estimated by measuring cell voltage and a high gain observer. However, the precision of the estimate is affected by the uncertainty in the diffusion coefficients of the system. For this reason, the observer is strengthened by estimating adaptive parameters of immersion and invariance. Finally, the results are validated in a numerical simulation., Este trabajo ha sido financiado por el CSIC bajo el proyecto PTI FLOWBAT 2021 (ref.642 201980E101), el Ministerio de Economía, Industria y Competitividad bajo el Proyecto DOVELAR (ref.RTI2018-096001-B-C32), y bajo la Agencia Estatatal de Investigación mediante la unidad de MARIA de Maeztu Seal of Excellence to IRI (MDM-2016-0656).
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Contribución de congreso (ConferenceOutput)

Flow controlling tuning for the voltage of a redox flow battery considering the effect of overpotentials

  • Clemente, Alejandro
  • Costa Castelló, Ramon
Trabajo presentado en el 26th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA ), celebrado en Västerås (Suecia), del 7 al 10 de septiembre de 2021, This conference paper presents a comparison between a H¿ control technique and a classical PID, applied on a redox flow battery system. The study presents a dynamic nonlinear electrochemical model that considers the effect of over-potentials losses in the computation of the voltage measurement. The controller is designed to regulate the output voltage of the battery. The H¿ controller is designed using the classical weighting function approach while the PID controller is tuned using a particle swarm optimizer. Finally, a comparison between the designed H¿ controller, and the classic PID is presented., This research was supported by the CSIC under the PTI FLOWBAT 2021 project (ref. 642 201980E101), the María de Maeztu Seal of Excellence to IRI (ref. MDM-2016-0656) and the Spanish Ministry of Economy and Competitiveness under Project DOVELAR (ref. RTI2018-096001-B-C32)
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Contribución de congreso (ConferenceOutput)

Combined heat and power using high-temperature proton exchange membrane fuel cells for housing facilities

  • Sanz i López, Víctor
  • Costa Castelló, Ramon
  • López Diest, Guillermo
  • Batlle, Carles
Trabajo presentado en el 26th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA ), celebrado en Västerås (Suecia), del 7 al 10 de septiembre de 2021, Recently, new alternatives to conventional energy sources such as fossil fuels are arising due to global problems related to climate change effect and energy shortage. In this context, fuel cells and combined heat and power technologies appear as a possible solution due to their ability to provide both electrical and thermal energy more efficiently compared to traditional methods. Related to this, high-temperature proton exchange membrane fuel cells offer the possibility of implementing combined heat and power systems, and they are also considered an efficient technology that emits less greenhouse gases. In this article a model predictive control based energy management system for a specific house is presented. Simulation and control models of the system are presented, together with dimensions and energy profiles used. Finally, control objectives and the proposed control algorithm are detailed, and the results when trying to match residential heat and power demands are discussed., This work has been partially funded by the Spanish national project DOVELAR (ref. RTI2018-096001-B-C32). This work is partially funded by AGAUR of Generalitat de Catalunya through the Advanced Control Systems (SAC) group grant (2017 SGR 482)
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Contribución de congreso (ConferenceOutput)

On state-estimation in weakly-observable scenarios and implicitly regularized observers

  • Cecilia, Andreu
  • Costa Castelló, Ramon
Trabajo presentado en la 60th Conference on Decision and Control (CDC), celebrada online del 14 al 17 de diciembre de 2021, This work proposes a framework to design observers for systems that present low observability. It is shown that, in these scenarios, the estimation problem becomes ill-posed, which drastically limits the performance of standard observers, specially in the presence of noise. Consequently, this paper presents a method to design an observer that optimizes some potential function to be defined by the designer. This allows to implicitly regularize the estimation and recover a well-posed problem. The proposed technique is validated in a set of weakly-observable systems and the performance is compared with a common Kalman filter-like observer., This work has been partially funded by the Spanish State Research Agency through the María de Maeztu Seal of Excellence to IRI (MDM-2016-0656) and by the project DOVELAR (ref. RTI2018-096001-B-C32).
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Contribución de congreso (ConferenceOutput)

Library-based adaptive observation through a sparsity-promoting adaptive observer

  • Cecilia, Andreu
  • Costa Castelló, Ramon
Trabajo presentado en la European Control Conference (ECC 2021), celebrada online del 29 de junio al 2 de julio de 2021, This paper proposes an adaptive observer for a class of nonlinear system with linear parametrization. The main novelty of the technique is that the regressor vector is considered to be unknown. Instead, a library of candidate non-linear functions is implemented, which transforms the original parameter vector into a new one that is characterized by being sparse. In such problem, it is shown that standard adaptive observers cannot recover the original vector due to a lack of persistence of excitation. Instead, a parameter-adaptation with an implicit l1 regularization is implemented. It is shown that this new observer can recover the parameter vector under standard assumptions of sparse signal recovery. The results are validated in a numerical simulation., This work has been partially funded by the Spanish State Research Agency through the María de Maeztu Seal of Excellence to IRI (MDM-2016-0656), by the project DOVELAR (ref. RTI2018-096001-B-C32) and by the PTI FLOWBAT 2021 project (ref. 642 201980E101)
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Contribución de congreso (ConferenceOutput)

Experimental validation of a vanadium redox flow battery model via Particle Swarm Optimization

  • Clemente, Alejandro
  • Costa Castelló, Ramon
  • Montiel, Manuel
  • Lozano Fantoba, Antonio
  • Barreras Toledo, Félix
Trabajo presentado en el 8th Symposium on Hydrogen, Fuel Cells and Advanced Batteries, celebrado en Buenos Aires (Argentina), del 10 al 13 de julio de 2022, Among all types of energy storage systems (ESS), vanadium redox flow batteries (VRFB) stand out for their high efficiency (70-90 %), long life cycle, safety and the possibility to decouple both energy and power sizes [1]. One of the main challenges in the study and analysis of VRFB consists on the correct caracterization of its performance through mathematical models. The importance of developing mathematical models is reflected in the literature, where numerous studies widely validated and referenced can be found, such as the electrochemical model developed by M. Skyllas-Kazacos [2], who pioneered the use of vanadium in redox flow batteries. Models can be classified as static or dynamic, depending on whether the effect of time is taken into acount, and as distributed or lumped parameter models, depending on the space dimension [3]. Tools as COMSOL are used to analyze and develop distributed parameter models, to study effects of flow design, components materials or degradation mechanisms [4]. However, for control purposes, the distributed models are rarely used due to its complexity and relevant computational cost, both in resources and time. For that reason, a vast majority of works in the literature concerning VRFB control, use lumped parameters models. Most of these lumped parameters models use different hypothesis to relax the problem, facilitating its understanding and analysis. This is the case of Skyllas-Kazacos electrochemical models mentioned, where the same flow rate is considered in both semicells,same species concentration inside the cell and tanks, as well as a constant temperature to estimate the open circuit voltage (OCV) by means of the Nernst equation [5]. Therefore, it is necessary to develop an analysis in order to define which variables and parameters will be modelled, and under which assumptions. Thus, in this work, a complete model that considers the most important effects of a VRFB is presented. It can be divided into four submodels, according to the previous parameters, which are: electrochemical, voltage, thermal and hydraulic ones. This model has been developed in a Matlab-Simulink environment as a tool to analyze the performance of a VRFB in a wide range of possibilities, examining the behaviour of the species and the voltage, temperature and pressure variables, incorporating some improvements with respect to other works. For all submodels, a clear distinction is made between both sides of the system (catholyte and anolyte semicells), not only in terms of dimensions and flow rates, but also on the initial conditions., This research is part of the CSIC program for the Spanish Recovery, Transformation and Resilience Plan funded by the Recovery and Resilience Facility of the European Union, established by the Regulation (EU) 2020/2094, CSIC Interdisciplinary Thematic Platform (PTI+) Transición Energética Sostenible+ (PTI-TRANSENER+) and the Spanish Ministry of Economy and Competitiveness under Project DOVELAR (ref. RTI2018-096001-B-C32).
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Contribución de congreso (ConferenceOutput)

State Machine-based architecture to control PEMFC system processes in a fuel cell electric vehicle

  • Molavi, Ali
  • Husar, Attila
  • Serra-Prat, María
  • Hjortberg, Hampus
  • Nilsson, Niclas
  • Kogler, Markus
  • Sanchez Monreal, Juan
  • Eldigair, Yousif
Trabajo presentado en la European Hydrogen Energy Conference, celebrada en Madrid (España), del 18 al 20 de mayo de 2022, In this paper, a state machine-based control architecture for a Fuel Cell Electric Vehicle (FCEV) is proposed. The considered Fuel Cell System (FCS) is a complex system with multiple different states of operation. This includes the start-up, run, shut-down and idle states, which have been identified as the states of a State Machine (SM). Furthermore, the FCS incorporates several subsystems; namely, as the cathode, anode and thermal subsystems and the DC/DC converter that all have different modes of operation. In each specific state of the FCS, the subsystems should operate in a specific predefined mode. A control structure is therefore necessary to coordinate these subsystems and the stack in the different FCS operation states and the operation modes of the different subsystems. This work describes in detail the SM designed for the fuel cell system of the FCEV of the INN-BALANCE project., This work was supported in part by the Spanish national project DOVELAR (RTI2018-096001-B-C32, MINECO/FEDER) also in part by the Fuel Cells and Hydrogen 2 Joint Undertaking under Grant INN-BALANCE 735969.This Joint Undertaking receives support from the European Union's Horizon 2020 research and innovation program and Hydrogen Europe and N.ERGHY.
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

H8 interactive controller design for teaching purposes

  • Díaz Martínez, José Manuel
  • Dormido Bencomo, Sebastián
  • Nicolau, Bernat
  • Costa Castelló, Ramon|||0000-0003-2553-5901
H8-based controller design is one of the most powerful methodologies for controller design in the frequency domain. Unfortunately, its use requires advanced knowledge of control theory. The use of interactive tools allows to a non-specialized user to use these techniques and to learn the fundamental concepts behind the H8 theory. In this work a graphical and completely interactive tool used to introduce students in H8 control is presented., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

PEMFC state and parameter estimation through a high-gain based adaptive observer

  • Cecilia Piñol, Andreu|||0000-0002-5579-4157
  • Serra, Maria|||0000-0002-9885-8093
  • Costa Castelló, Ramon|||0000-0003-2553-5901
This work presents the design of an adaptive observer to estimate the cathode catalytic layer's water content of a polymer electrolyte membrane fuel cell and some of the parameters related to its water dynamics. However, in general, existing adaptive observer algorithms require a certain relative degree condition which is not satisfied in the concerned fuel cell system. This conflict is solved by modifying the adaptive observer strategy with an auxiliary signal that does satisfy the relative degree condition. This signal is estimated through a high-gain observer. The viability of the presented observer is validated through numerical simulations., This work has been partially funded by the Spanish State Research Agency through the Maria de Maeztu Seal of Excellence to IRI (MDM-2016-0656), by the project DOVELAR (ref. RTI2018-096001-B-C32) and by AGAUR of Generalitat de Catalunya through theAdvanced Control Systems grant (SAC) (2017 SGR 482, Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Flow controlling tuning for the voltage of a redox flow battery considering the effect of overpotentials

  • Clemente Escuin, Antonio
  • Costa Castelló, Ramon|||0000-0003-2553-5901
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, This conference paper presents a comparison between a H8 control technique and a classical PID, applied on a redox flow battery system. The study presents a dynamic nonlinear electrochemical model that considers the effect of over-potentials losses in the computation of the voltage measurement. The controller is designed to regulate the output voltage of the battery. The H8 controller is designed using the classical weighting function approach while the PID controller is tuned using a particle swarm optimizer. Finally, a comparison between the designed H8 controller, and the classic PID is presented., This research was supported by the CSIC under the PTI FLOWBAT 2021 project (ref. 642 201980E101), the María de Maeztu Seal of Excellence to IRI (ref. MDM-2016-0656) and the Spanish Ministry of Economy and Competitiveness under Project DOVELAR (ref. RTI2018-096001-B-C32), Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Library-based adaptive observation through a sparsity-promoting adaptive observer

  • Cecilia Piñol, Andreu|||0000-0002-5579-4157
  • Costa Castelló, Ramon|||0000-0003-2553-5901
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, This paper proposes an adaptive observer for a class of nonlinear system with linear parametrization. The main novelty of the technique is that the regressor vector is considered to be unknown. Instead, a library of candidate non-linear functions is implemented, which transforms the original parameter vector into a new one that is characterized by being sparse. In such problem, it is shown that standard adaptive observers cannot recover the original vector due to a lack of persistence of excitation. Instead, a parameter-adaptation with an implicit l1 regularization is implemented. It is shown that this new observer can recover the parameter vector under standard assumptions of sparse signal recovery. The results are validated in a numerical simulation., This work has been partially funded by the Spanish State Research Agency through the María de Maeztu Seal of Excellence to IRI (MDM-2016-0656), by the project DOVELAR (ref. RTI2018-096001-B-C32) and by the PTI FLOWBAT 2021 project (ref. 642 201980E101)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Implementación y validación experimental del control de un sistema híbrido basado en pilas de combustible para vehículos eléctricos

  • Anderson, Jorge
  • Moré,, J.J.
  • Puleston, Pablo Federico
  • Roda Serrat, Vicente
  • Costa Castelló, Ramon|||0000-0003-2553-5901
En este trabajo se presenta el diseño, implementación y validación experimental de un controlador por modos deslizantes de segundo orden aplicado a un módulo de generación híbrida basado en una pila de combustible y supercapacitores. El controlador desarrollado se encuentra basado en el algoritmo Super-Twisting. Para simplificar el diseño de las acciones de control del sistema, se emplea una transformación no lineal que permite independizar el diseño de las variables de deslizamiento. El sistema de control propuesto es ensayado experimentalmente en una plataforma que incorpora una pila de combustible de 1.2kW y supercapacitores de uso comercial. El módulo de generación híbrida controlado es entonces evaluado para satisfacer la demanda de potencia de un vehículo eléctrico en condiciones de conducción urbana, con la presencia de múltiples etapas de aceleración y desaceleración. Los resultados obtenidos muestran una buena regulación de la tensión del bus común y un seguimiento robusto de las referencias impuestas, aún frente a la presencia de fuertes variaciones y perturbaciones de la potencia de carga., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Stochastic neural radiance fields: quantifying uncertainty in implicit 3D representations

  • Shen, Jianxiong
  • Ruiz Ovejero, Adrià
  • Agudo Martínez, Antonio|||0000-0001-6845-4998
  • Moreno-Noguer, Francesc
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, Neural Radiance Fields (NeRF) has become a popular framework for learning implicit 3D representations and addressing different tasks such as novel-view synthesis or depth-map estimation. However, in downstream applications where decisions need to be made based on automatic predictions, it is critical to leverage the confidence associated with the model estimations. Whereas uncertainty quantification is a long-standing problem in Machine Learning, it has been largely overlooked in the recent NeRF literature. In this context, we propose Stochastic Neural Radiance Fields (S-NeRF), a generalization of standard NeRF that learns a probability distribution over all the possible radiance fields modeling the scene. This distribution allows to quantify the uncertainty associated with the scene information provided by the model. S-NeRF optimization is posed as a Bayesian learning problem that is efficiently addressed using the Variational Inference framework. Exhaustive experiments over benchmark datasets demonstrate that S-NeRF is able to provide more reliable predictions and confidence values than generic approaches previously proposed for uncertainty estimation in other domains., This work is supported partly by the Chinese Scholarship Council (CSC) under grant (201906120031), by the Spanish government under project MoHuCo PID2020- 120049RB-I00, the ERA-Net Chistera project IPALM PCI2019-103386 and María de Maeztu Seal of Excellence MDM-2016-0656. Adria Ruiz acknowledges financial support from MICINN through the program Juan de la Cierva., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Parameter fitting of a reversible solid oxide stack model

  • Lleonsí Carrillo, Ignasi
  • Bernadet, Lucile
  • Torrell, Marc
  • Serra, Maria|||0000-0002-9885-8093
Ponència presentada a la European Hydrogen Energy Conference, celebrada a Madrid (España), del 18 al 20 de maig de 2022, In this work, the fitting of five parameters of a Reversible Solid Oxide (RSO) stack model is done based on data obtained at different conditions of operation. Two types of experimental curves are used: polarization curves and sweeps. The system suffered a degradation and the analysis has permitted to study which one of the parameters was affected, indicating which part of the stack was damaged., Project funded partially by Ajuntament de Barcelona, in the framework of Pla Barcelona Ciència and partially by the Spanish State Research Agency through the project DOVELAR (ref. RTI2018-096001-B-C32)
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

SOC and diffusion rate estimation in redox flow batteries: An I&I-based high-gain observer approach

  • Clemente Escuin, Antonio
  • Cecilia Piñol, Andreu|||0000-0002-5579-4157
  • Costa Castelló, Ramon|||0000-0003-2553-5901
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, This paper presents an adaptive non-linear observer for the state of charge estimation in vanadium redox flow batteries. The study is based directly on the use of non-linear equations that describe the evolution of the species concentration inside the system, and a nonlinear cell voltage expression that takes into account the effect of overpotentials. Moreover, a more realistic approach is used which does not consider that the electrolyte concentration is the same in the catholyte and anolyte sides of the system. It is shown that the state of charge can be estimated through the measurement of the output voltage and a high-gain observer. Nonetheless, the accuracy of the estimation is affected by uncertainty in the system diffusion rates. For this reason, the observer is robustified by means of an immersive and invariance adaptive parameter estimation. The results are validated in a numerical simulation., This research was funded by the CSIC under the PTI FLOWBAT 2021 project (ref. 642 201980E101), the Spanish Ministry of Economy and Competitiveness under Project DOVELAR (ref. RTI2018-096001-B-C32), and by the Spanish State Research Agency through the María de Maeztu Seal of Excellence to IRI (MDM-2016-0656)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Control no lineal adaptativo con identificación dispersa

  • Domingo-Enrich, Roger
  • Cecilia Piñol, Andreu|||0000-0002-5579-4157
  • Costa Castelló, Ramon|||0000-0003-2553-5901
This work presents a controller with parameter estimation focused on non-linear, unknown and time-variant systems. The proposal consists of an on-line learning algorithm to estimate the system equations, and a controller that uses this information to stabilize the plant. Our approach is to combine a non-linear observer and a parameter estimator using sparse optimization techniques. The resulting algorithm operates in two stages: the learning stage, where unknown functions are learned on-line, and the execution stage, where the plant is controlled with pre-established objectives. The proposed algorithm can autonomously decide what is its current stage by using a learning metric. We have validated our algorithm using numerical simulation in a roto-magnet plant., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Combined heat and power using high-temperature proton exchange membrane fuel cells for housing facilities

  • Sanz i López, Víctor
  • Lopez Diest, Guillermo
  • Costa Castelló, Ramon|||0000-0003-2553-5901
  • Batlle Arnau, Carles|||0000-0002-6088-6187
© 20xx IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Recently, new alternatives to conventional energy sources such as fossil fuels are arising due to global problems related to climate change effect and energy shortage. In this context, fuel cells and combined heat and power technologies appear as a possible solution due to their ability to provide both electrical and thermal energy more efficiently compared to traditional methods. Related to this, high-temperature proton exchange membrane fuel cells offer the possibility of implementing combined heat and power systems, and they are also considered an efficient technology that emits less greenhouse gases. In this article a model predictive control based energy management system for a specific house is presented. Simulation and control models of the system are presented, together with dimensions and energy profiles used. Finally, control objectives and the proposed control algorithm are detailed, and the results when trying to match residential heat and power demands are discussed., Peer Reviewed, Objectius de Desenvolupament Sostenible::7 - Energia Assequible i No Contaminant
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

On state-estimation in weakly-observable scenarios and implicitly regularized observers

  • Cecilia Piñol, Andreu|||0000-0002-5579-4157
  • Costa Castelló, Ramon|||0000-0003-2553-5901
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, This work proposes a framework to design observers for systems that present low observability. It is shown that, in these scenarios, the estimation problem becomes ill-posed, which drastically limits the performance of standard observers, specially in the presence of noise. Consequently, this paper presents a method to design an observer that optimizes some potential function to be defined by the designer. This allows to implicitly regularize the estimation and recover a well-posed problem. The proposed technique is validated in a set of weakly-observable systems and the performance is compared with a common Kalman filter-like observer., This work has been partially funded by the Spanish State Research Agency through the María de Maeztu Seal of Excellence to IRI (MDM-2016-0656) and by the project DOVELAR (ref. RTI2018-096001-B-C32)., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Library-based adaptive observation through a sparsity-promoting adaptive observer

  • Cecilia Piñol, Andreu
  • Costa Castelló, Ramon
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, This paper proposes an adaptive observer for a class of nonlinear system with linear parametrization. The main novelty of the technique is that the regressor vector is considered to be unknown. Instead, a library of candidate non-linear functions is implemented, which transforms the original parameter vector into a new one that is characterized by being sparse. In such problem, it is shown that standard adaptive observers cannot recover the original vector due to a lack of persistence of excitation. Instead, a parameter-adaptation with an implicit l1 regularization is implemented. It is shown that this new observer can recover the parameter vector under standard assumptions of sparse signal recovery. The results are validated in a numerical simulation., This work has been partially funded by the Spanish State Research Agency through the María de Maeztu Seal of Excellence to IRI (MDM-2016-0656), by the project DOVELAR (ref. RTI2018-096001-B-C32) and by the PTI FLOWBAT 2021 project (ref. 642 201980E101)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Flow controlling tuning for the voltage of a redox flow battery considering the effect of overpotentials

  • Clemente Escuin, Antonio
  • Costa Castelló, Ramon
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, This conference paper presents a comparison between a H8 control technique and a classical PID, applied on a redox flow battery system. The study presents a dynamic nonlinear electrochemical model that considers the effect of over-potentials losses in the computation of the voltage measurement. The controller is designed to regulate the output voltage of the battery. The H8 controller is designed using the classical weighting function approach while the PID controller is tuned using a particle swarm optimizer. Finally, a comparison between the designed H8 controller, and the classic PID is presented., This research was supported by the CSIC under the PTI FLOWBAT 2021 project (ref. 642 201980E101), the María de Maeztu Seal of Excellence to IRI (ref. MDM-2016-0656) and the Spanish Ministry of Economy and Competitiveness under Project DOVELAR (ref. RTI2018-096001-B-C32), Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Implementación y validación experimental del control de un sistema híbrido basado en pilas de combustible para vehículos eléctricos

  • Anderson, Jorge
  • Moré,, J.J.
  • Puleston, Pablo Federico
  • Roda Serrat, Vicente
  • Costa Castelló, Ramon
En este trabajo se presenta el diseño, implementación y validación experimental de un controlador por modos deslizantes de segundo orden aplicado a un módulo de generación híbrida basado en una pila de combustible y supercapacitores. El controlador desarrollado se encuentra basado en el algoritmo Super-Twisting. Para simplificar el diseño de las acciones de control del sistema, se emplea una transformación no lineal que permite independizar el diseño de las variables de deslizamiento. El sistema de control propuesto es ensayado experimentalmente en una plataforma que incorpora una pila de combustible de 1.2kW y supercapacitores de uso comercial. El módulo de generación híbrida controlado es entonces evaluado para satisfacer la demanda de potencia de un vehículo eléctrico en condiciones de conducción urbana, con la presencia de múltiples etapas de aceleración y desaceleración. Los resultados obtenidos muestran una buena regulación de la tensión del bus común y un seguimiento robusto de las referencias impuestas, aún frente a la presencia de fuertes variaciones y perturbaciones de la potencia de carga., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Stochastic neural radiance fields: quantifying uncertainty in implicit 3D representations

  • Shen, Jianxiong
  • Ruiz Ovejero, Adrià
  • Agudo Martínez, Antonio
  • Moreno-Noguer, Francesc
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, Neural Radiance Fields (NeRF) has become a popular framework for learning implicit 3D representations and addressing different tasks such as novel-view synthesis or depth-map estimation. However, in downstream applications where decisions need to be made based on automatic predictions, it is critical to leverage the confidence associated with the model estimations. Whereas uncertainty quantification is a long-standing problem in Machine Learning, it has been largely overlooked in the recent NeRF literature. In this context, we propose Stochastic Neural Radiance Fields (S-NeRF), a generalization of standard NeRF that learns a probability distribution over all the possible radiance fields modeling the scene. This distribution allows to quantify the uncertainty associated with the scene information provided by the model. S-NeRF optimization is posed as a Bayesian learning problem that is efficiently addressed using the Variational Inference framework. Exhaustive experiments over benchmark datasets demonstrate that S-NeRF is able to provide more reliable predictions and confidence values than generic approaches previously proposed for uncertainty estimation in other domains., This work is supported partly by the Chinese Scholarship Council (CSC) under grant (201906120031), by the Spanish government under project MoHuCo PID2020- 120049RB-I00, the ERA-Net Chistera project IPALM PCI2019-103386 and María de Maeztu Seal of Excellence MDM-2016-0656. Adria Ruiz acknowledges financial support from MICINN through the program Juan de la Cierva., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

SOC and diffusion rate estimation in redox flow batteries: An I&I-based high-gain observer approach

  • Clemente Escuin, Antonio
  • Cecilia Piñol, Andreu
  • Costa Castelló, Ramon
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, This paper presents an adaptive non-linear observer for the state of charge estimation in vanadium redox flow batteries. The study is based directly on the use of non-linear equations that describe the evolution of the species concentration inside the system, and a nonlinear cell voltage expression that takes into account the effect of overpotentials. Moreover, a more realistic approach is used which does not consider that the electrolyte concentration is the same in the catholyte and anolyte sides of the system. It is shown that the state of charge can be estimated through the measurement of the output voltage and a high-gain observer. Nonetheless, the accuracy of the estimation is affected by uncertainty in the system diffusion rates. For this reason, the observer is robustified by means of an immersive and invariance adaptive parameter estimation. The results are validated in a numerical simulation., This research was funded by the CSIC under the PTI FLOWBAT 2021 project (ref. 642 201980E101), the Spanish Ministry of Economy and Competitiveness under Project DOVELAR (ref. RTI2018-096001-B-C32), and by the Spanish State Research Agency through the María de Maeztu Seal of Excellence to IRI (MDM-2016-0656)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Control no lineal adaptativo con identificación dispersa

  • Domingo-Enrich, Roger
  • Cecilia Piñol, Andreu
  • Costa Castelló, Ramon
This work presents a controller with parameter estimation focused on non-linear, unknown and time-variant systems. The proposal consists of an on-line learning algorithm to estimate the system equations, and a controller that uses this information to stabilize the plant. Our approach is to combine a non-linear observer and a parameter estimator using sparse optimization techniques. The resulting algorithm operates in two stages: the learning stage, where unknown functions are learned on-line, and the execution stage, where the plant is controlled with pre-established objectives. The proposed algorithm can autonomously decide what is its current stage by using a learning metric. We have validated our algorithm using numerical simulation in a roto-magnet plant., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Combined heat and power using high-temperature proton exchange membrane fuel cells for housing facilities

  • Sanz i López, Víctor
  • Lopez Diest, Guillermo
  • Costa Castelló, Ramon
  • Batlle Arnau, Carles
© 20xx IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Recently, new alternatives to conventional energy sources such as fossil fuels are arising due to global problems related to climate change effect and energy shortage. In this context, fuel cells and combined heat and power technologies appear as a possible solution due to their ability to provide both electrical and thermal energy more efficiently compared to traditional methods. Related to this, high-temperature proton exchange membrane fuel cells offer the possibility of implementing combined heat and power systems, and they are also considered an efficient technology that emits less greenhouse gases. In this article a model predictive control based energy management system for a specific house is presented. Simulation and control models of the system are presented, together with dimensions and energy profiles used. Finally, control objectives and the proposed control algorithm are detailed, and the results when trying to match residential heat and power demands are discussed., Peer Reviewed, Objectius de Desenvolupament Sostenible::7 - Energia Assequible i No Contaminant, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

On state-estimation in weakly-observable scenarios and implicitly regularized observers

  • Cecilia Piñol, Andreu
  • Costa Castelló, Ramon
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, This work proposes a framework to design observers for systems that present low observability. It is shown that, in these scenarios, the estimation problem becomes ill-posed, which drastically limits the performance of standard observers, specially in the presence of noise. Consequently, this paper presents a method to design an observer that optimizes some potential function to be defined by the designer. This allows to implicitly regularize the estimation and recover a well-posed problem. The proposed technique is validated in a set of weakly-observable systems and the performance is compared with a common Kalman filter-like observer., This work has been partially funded by the Spanish State Research Agency through the María de Maeztu Seal of Excellence to IRI (MDM-2016-0656) and by the project DOVELAR (ref. RTI2018-096001-B-C32)., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Parameter fitting of a reversible solid oxide stack model

  • Lleonsí Carrillo, Ignasi
  • Bernadet, Lucile
  • Torrell, Marc
  • Serra, Maria
Ponència presentada a la European Hydrogen Energy Conference, celebrada a Madrid (España), del 18 al 20 de maig de 2022, In this work, the fitting of five parameters of a Reversible Solid Oxide (RSO) stack model is done based on data obtained at different conditions of operation. Two types of experimental curves are used: polarization curves and sweeps. The system suffered a degradation and the analysis has permitted to study which one of the parameters was affected, indicating which part of the stack was damaged., Project funded partially by Ajuntament de Barcelona, in the framework of Pla Barcelona Ciència and partially by the Spanish State Research Agency through the project DOVELAR (ref. RTI2018-096001-B-C32), Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

PEMFC state and parameter estimation through a high-gain based adaptive observer

  • Cecilia Piñol, Andreu
  • Serra, Maria
  • Costa Castelló, Ramon
This work presents the design of an adaptive observer to estimate the cathode catalytic layer's water content of a polymer electrolyte membrane fuel cell and some of the parameters related to its water dynamics. However, in general, existing adaptive observer algorithms require a certain relative degree condition which is not satisfied in the concerned fuel cell system. This conflict is solved by modifying the adaptive observer strategy with an auxiliary signal that does satisfy the relative degree condition. This signal is estimated through a high-gain observer. The viability of the presented observer is validated through numerical simulations., This work has been partially funded by the Spanish State Research Agency through the Maria de Maeztu Seal of Excellence to IRI (MDM-2016-0656), by the project DOVELAR (ref. RTI2018-096001-B-C32) and by AGAUR of Generalitat de Catalunya through theAdvanced Control Systems grant (SAC) (2017 SGR 482, Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

H8 interactive controller design for teaching purposes

  • Díaz Martínez, José Manuel
  • Dormido Bencomo, Sebastián
  • Nicolau, Bernat
  • Costa Castelló, Ramon
H8-based controller design is one of the most powerful methodologies for controller design in the frequency domain. Unfortunately, its use requires advanced knowledge of control theory. The use of interactive tools allows to a non-specialized user to use these techniques and to learn the fundamental concepts behind the H8 theory. In this work a graphical and completely interactive tool used to introduce students in H8 control is presented., Peer Reviewed, Postprint (published version)