dc.contributorVallejo Velásquez, Mónica Ayde
dc.contributorRivadeneira Paz, Pablo Santiago
dc.creatorGoez Mora, Jhon Edison
dc.date.accessioned2022-03-07T15:23:13Z
dc.date.accessioned2022-09-21T19:10:59Z
dc.date.available2022-03-07T15:23:13Z
dc.date.available2022-09-21T19:10:59Z
dc.date.created2022-03-07T15:23:13Z
dc.date.issued2022-03
dc.identifierhttps://repositorio.unal.edu.co/handle/unal/81137
dc.identifierUniversidad Nacional de Colombia
dc.identifierRepositorio Institucional Universidad Nacional de Colombia
dc.identifierhttps://repositorio.unal.edu.co/
dc.identifier.urihttp://repositorioslatinoamericanos.uchile.cl/handle/2250/3413876
dc.description.abstractLos avances tecnológicos actuales han acercado a la realidad el proyecto de un páncreas artificial (PA) seguro, portátil y eficiente para personas con diabetes tipo 1. Entre las estrategias de control desarrolladas para la diabetes tipo 1, el control predictivo basado en modelo (MPC por sus siglas en inglés) se ha enfatizado en la literatura como un control prometedor para la regulación de la glucosa. Sin embargo, estas estrategias de control se diseñan comúnmente en un entorno simulado, independientemente de las limitaciones de un dispositivo portátil. En este trabajo, se evalúa el rendimiento de seis sistemas embebidos, tres paquetes de optimización de código abierto y cuatro formulaciones de MPC con un esquema de hardware en el ciclo (HIL por sus siglas en inglés), para encontrar la mejor combinación tomando como criterios de selección la temperatura del procesador, el tiempo de ejecución, el coeficiente de variación, el porcentaje de tiempo en normoglucemia, la energía consumida, la cantidad de eventos de hiperglucemia y la diferencia respecto a la evolución obtenida en MATLAB. Al escoger la mejor combinación se aplica la estrategia de eventos de activación con el fin de reducir el número de veces que se ejecuta el cálculo de la dosificación de insulina óptima permitiendo que durante los periodos de tiempo en los que no es necesario llevar a cabo acciones de control el dispositivo ahorre energía. Durante el desarrollo de las pruebas los controladores son expuestos a variaciones fisiológicas simuladas en los pacientes virtuales, a ingesta de carbohidratos y a ejecutarse con y sin anuncio de comida. Los primeros resultados muestran que la Raspberry pi 3 B, el paquete quadprog y la estrategia de eliminación de offset son la mejor combinación resaltando el bajo consumo energético del dispositivo. Con esta base se integra la estrategia de activación de eventos y se realizan las pruebas poblacionales encontrando una reducción significativa en el número de controles calculados, aunque se presenta una pérdida de desempeño en el controlador al elevarse los niveles promedios de glucemia. Por último, se realiza una emulación del PA con un paciente virtual en donde se implementa un sensor inteligente, un micromotor paso a paso como actuador y una batería con la que se determina el consumo real del dispositivo y su tiempo de autonomía con y sin el nuevo controlador basado en MPC. (Texto tomado de la fuente)
dc.description.abstractCurrent technological advances have brought closer to reality the project of a safe, portable, and efficient artificial pancreas for people with type 1 diabetes (T1D). Among the developed control strategies for T1D, model predictive control (MPC) has been emphasized in literature as a promising control for glucose regulation. However, these control strategies are commonly designed in a computer environment, regardless of the limitations of a portable device. When choosing the best combination, the event-triggering strategy is applied in order to reduce the number of times the calculation of the optimal insulin dosage is executed, allowing during the periods of time in which it is not necessary to carry out control actions the device save energy. During the development of the tests, the controllers are exposed to simulated physiological variations in virtual patients, carbohydrate intake and to execute with and without announced meals. The first results show that the Raspberry pi 3 B, the quadprog package, and the offset-free strategy are the best combination highlighting the low power consumption of the device. With this base, the event-triggering strategy is integrated and population tests are carried out, finding a significant reduction in the number of calculated controls, although there is a loss of performance in the controller as average blood glucose levels rise. Finally, an emulation of the artificial pancreas is carried out with a virtual patient where an intelligent sensor, a micro stepper motor as an actuator, and a battery in implemented with which the real consumption of the device and its autonomy time is determined whit and without the new controller base on MPC.
dc.languagespa
dc.publisherUniversidad Nacional de Colombia
dc.publisherMedellín - Minas - Maestría en Ingeniería - Automatización Industrial
dc.publisherDepartamento de Ingeniería Eléctrica y Automática
dc.publisherFacultad de Minas
dc.publisherMedellín, Colombia
dc.publisherUniversidad Nacional de Colombia - Sede Medellín
dc.relationAndersen, E., Roos, C., and Terlaky, T. (2003). On implementing a primal-dual interior-point method for conic quadratic optimization.Mathematical Programming, 95:249–277.
dc.relationAnderson, S. M., Dassau, E., Raghinaru, D., Lum, J., Brown, S. A.,Pinsker, J. E., Church, M. M., Levy, C., Lam, D., Kudva, Y. C., Buckingham, B., P., F. G.,Wadwa, R. P., Laffel, L., Doyle, F. J., HansDevries, J., Renard, E., Cobelli, C., Boscari,F., Del Favero, S., and Kovatchev, B. P. (2019). The international diabetes closed-loop study: Testing artificial pancreas component interoperability.Diabetes Technology and therapeutics, 2:73–80.
dc.relationARM (2011).ARM®Cortex™-A7 MPCore™Processor Technical ReferenceManual. https://developer.arm.com/ip-products/processors/cortex-a/cortex-a7.
dc.relationARM (2012a).ARM®Cortex®-A15 MPCore™Processor Technical Refe-rence Manual. https://developer.arm.com/ip-products/processors/cortex-a/cortex-a15
dc.relationARM (2012b).ARM®Cortex®-A17 MPCore™Processor Technical Refe-rence Manual. https://developer.arm.com/ip-products/processors/cortex-a/cortex-a17.
dc.relationARM (2012c).Arm®Cortex®-A53 MPCore Processor Technical ReferenceManual. https://developer.arm.com/ip-products/processors/cortex-a/cortex-a53.
dc.relationARM (2016a).ARM®Cortex®-A57 MPCore Processor Technical ReferenceManual. https://developer.arm.com/ip-products/processors/cortex-a/cortex-a57.
dc.relationARM (2016b).ARM®Cortex®-A72 MPCore Processor Technical ReferenceManual. https://developer.arm.com/ip-products/processors/cortex-a/cortex-a72.
dc.relationASUS(2018).TinkerBoardSTechspecs.https://www.asus.com/us/Motherboards-Components/Single-Board-Computer/All-series/Tinker-Board-S/.
dc.relationBergman, R., Ider, Y., Bowden, C., and Cobelli, C. (1979). Quanti-tative estimation of insulin sensitivity.The American journal of physiology, 236:67–77.
dc.relationBerian, J., Bravo, I., Gardel, A., L ́azaro, J., and Hernández, S. (2019).A wearable closed-loop insulin deliverysystem based on low-power socs.Electronics, 8.
dc.relationChakrabarty, A., Healey, E., Shi, D., Zavitsanou, S., J., D. F., andDassau, E. (2020). Embedded model predictive control for a wearable artificial pancreas.IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY, 28:2600–2607.
dc.relationChakrabarty, A., Zavitsanou, S., Doyle, F., and Dassau, E. (2018).Event-triggered model predictive control for embedded artificial pancreas systems.IEEETransactions on Biomedical Engineering, 65:575–586.
dc.relationCO,S.X.S.(2017).OrangePiPcPlusTechspecs.http://www.orangepi.org/orangepipcplus/.
dc.relationColmegna, P., Garelli, F., Battista, H. D., and Sánchez peña, R.(2018). Automatic regulatory control in type 1diabetes without carbohydrate counting.Control Engineering Practice, 74:22–32.
dc.relationCostHelper,I.(2013).InsulinPumpCost.https://health.costhelper.com/insulin-pump.html.
dc.relationDalla Man, C., Micheletto, F., Lv, D., Breton, M., Kovatchev, B.,and Cobelli, C. (2014). The uva/padova type 1 diabetes simulator: New features.Diabe-tesScience and Technology, 8:26–34.
dc.relationDe Alto Costo, C. (2018). Dia mundial de la diabetes. boletín de información técnica especializada. Technical report, Cuenta de Alto Costo, Fondo colombiano de enfermedades de alto costo.
dc.relationDe Alto Costo, C. (2020). Cuenta de alto costo asamblea general ordinaria 2020 anexo técnico. Technical report, Cuenta de Alto Costo, Fondo colombiano de enfermedades de alto costo.
dc.relationDexcom(2015).SensorDexcomG5.https://s3-us-west- .amazonaws.com/dexcompdf/Downloads+and+Guides+Updates/LBL013358+G5+Mobile+UG+OUS+ES+mgdL.pdf.
dc.relationDougGuan, D. A. E. T. C. L. (2020). Report n°: Sa2007199m 01001 li-ion polymer battery 955465. Technical report, DougGuan Anci Electronic Technology Co.Ltd.
dc.relationDoyle, F. J., Huyett, L. M., Lee, J. B., Z., C., H., and Dassau, E.(2014). Closed-loop artificial pancreas systems: Engineering the algorithms.DiabetesCare, 37:1191–1197.
dc.relationEMC, E. T. R. (2020). Report n°: Htt202003477er model pisugar2-pro. Tech-nical report, Shenzhen HTT Technology Co., Ltd.
dc.relationEqtami, A., Dimarogonas, D. V., and Kyriakopoulos, K. J. (2010).Event-triggered control for discrete-time systems.Proceedings of the 2010 American Control Conference.
dc.relationF., V.-T. M., Caicedo, M. A., and Rivadeneira, P. S. (2020). Offset-free mpc strategy for nonzero regulation of linear impulsive systems.ISA Transactions, 101:91–101.
dc.relationFOUNDATION, R. P. (2016).Raspberry Pi 3 Model B Techspecs.https://www.raspberrypi.org/products/raspberry-pi-3-model-b/.
dc.relationFOUNDATION, R. P. (2018).Raspberry Pi 4 Techspecs.https://www.raspberrypi.org/products/raspberry-pi-4-model-b/specifications/.
dc.relationGarcía, P. A. (2019). Sistemas Embebidos de Tiempo Real con aplicaciones en Bioingeniería. Tesis de doctorado, Universidad Nacional de la Plata.
dc.relationGiselsson, P. and Boyd, S. (2017). Linear convergence and metric selection for Douglas–rachford splitting and admm.IEEE Trans. Autom. Control,62:532–544.
dc.relationGoldfarb, D. and Idnani, A. (2011). A numerically stable dual method for solving strictly convex quadratic programs.Mathematical Programming, 27:1–33.
dc.relationGondhalekar, R., Dassau, E., and Doyle, F. J. (2016). Periodiczone-mpc with asymmetric costs for outpatient-ready safety of an artificial pancreas totreat type 1 diabetes.Automatica, 71:237–246.
dc.relationGondhalekar, R., Dassau, E., and Doyle, F. J. (2018). Velocity-weighting & velocity-penalty mpc of an artificialpancreas: Improved safety & performance.Automatica, 91:105–117.
dc.relationGrosman, B., Dassau, E., Zisser, H. C., and Jovanoviˇc, L.and Doyle,F. J. (2010). Zone model predictive control: a strategy to minimize hyper- and hypoglycemic events.J. Diabetes Sci. Technol., 4:961–975.
dc.relationHajizadeh, I., Rashid, M., Turksoy, K., Samadi, S., Feng, J., Frantz,N., Sevil, M., Cengiz, E., and Cinar, A. (2017). Plasma insulin estimation in people with type 1 diabetes mellitus.Industrial and Engineering ChemistryResearch, 56:9846–9857.
dc.relationHardKernel,L.(2015).ODROID-XU4Techspecs.https://wiki.odroid.com/odroid-xu4/hardware/hardware.
dc.relationHei, X., Du, X., Lin, S., Lee, I., and Sokolsky, O. (2015). Patient infusion pattern-based access control schemes for wireless insulin pump system.IEEE Transactions on Parallel and Distributed Systems26 pages 3108–3121.
dc.relationHoyos, J., Villa-Tamayo, M., Builes-Monta ̃no, C., Ramirez-Rincón, A.,Godoy, J., Carcia-Tirado, J., and Rivadeneira, P. (2021). Identifiability of control orientedglucose insulin linear models: review and analysis.IEEE ACCESS, pages 1–16.
dc.relationHuyett, L. M., Dassau, E., Zisser, H. C., and Doyle, F. J. (2015). De-sign and evaluation of a robust pid controllerfor a fully implantable artificial pancreas.Industrial and Engineering Chemistry Research, 54:10311–10321.
dc.relationI+D, E. (2015). Tarjeta para interfaz usb - serial. Technical report, ElectrónicaI+D, Medellín.
dc.relationIDF (2019). Idf diabetes atlas ninth edition. Technical report, InternationalDiabetes Federation.
dc.relationIntel (2020). New intel®core™processors with intel®hybrid technology.Technical report,©Intel Corporation.
dc.relationJohansen, T. A. (2017). Toward dependable embedded model predictivecontrol.IEEE Systems Journal, 11:1208–1219.
dc.relationJung, C. A.and Lee, S. J. (2016). Design of automatic insulin injection systemwith continuous glucose monitoring (cgm) signalss.3rd IEEE EMBS Int. Conf. Biomed.Heal. Informatics, pages 102–105.
dc.relationKouzoupis, D., Ferreau, H. J., Peyrl, H., and Diehl, M. (2015). First-order methods in embedded nonlinear model predictive control.2015 European ControlConference (ECC).
dc.relationLehmann, D., Henriksson, E., and Johansson, K. (2013). Event-triggered model predictive control of discrete-time linear systems subject to disturbances.European Control Conference (ECC), page 1156–1161.
dc.relationLi, H. and Shi, Y. (2014). Event-triggered robust model predictive controlof continuous-time nonlinear systems.Automatica, 50:1507–1513.
dc.relationLimited, F. T. D. I. (2011).Future Technology Devices International Ltd.FT232R USB UART IC Datasheet. https://www.raspberrypi.org/products/raspberry-pi-4-model-b/specifications/.
dc.relationLopez-Valencia, A. (2021). Construcción de un prototipo de infusor de insulina e integración con un sensor de glucosa en sangre freestyle libre mediante comunicación inalámbrica. Master’s thesis, Universidad Nacional de Colombia.
dc.relationMaciejowski, J. (2001). Predictive Control with Constraints. PearsonCollege Div 1 edition.
dc.relationMagdelaine, N., Chaillous, L., Guilhem, I., J.-Y., P., Krempf, M.,Moog, C. H., and Le Carpentier, E. (2015). A long-term model of the glucose-insulin dy-namics of type 1 diabetes.IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING,62:1546–1552.
dc.relationMathWorks(2021a).Hardware-In-The-LoopSimulation.https://la.mathworks.com/help/physmod/simscape/ug/what-is-hardware-in-the-loop-simulation.html.
dc.relationMathWorks (2021b). Optimization toolbox documentation, quadprog.Technical report, The MathWorks, Inc.
dc.relationMedtronic (2017). Minimed 640g system user guide. Technical report,The MathWorks, Inc, Los Angeles.
dc.relationMelexis microelectronics systems(2006).MLX90614 family Single and Dual Zone Infra Red Thermometer in TO-39.https://pdf1.alldatasheet.com/datasheet-pdf/view/218977/ETC2/MLX90614.html.
dc.relationNVIDIA(2019).JetsonNanoTechspecs.https://developer.nvidia.com/embedded/jetson-nano-2gb-developer-kit.
dc.relationOperating as Diabetes UK, T. B. D. A. (2018).Insulinpumps.https://www.diabetes.org.uk/guide-to-diabetes/managing-your-diabetes/treating-your-diabetes/insulin-pumps.
dc.relationPinsker, J. E., Lee, J. B., Dassau, E., Seborg, D. E., Bradley, P. K.,Gondhalekar, R., Bevier, W. C., Huyett, L., Z., C., H., and Doyle, F. J. (2016). Randomi-zed crossover comparison of personalized mpc and pid controlalgorithms for the artificialpancreas.Diabetes Care, 39:1135–1142.
dc.relationRadziuk, J. (2012). The artificial pancreas. Diabetes, 61:2221–2224.
dc.relationRivadeneira, P. S., Ferramosca, A., and González, A. H. (2018).Control strategies for non-zero set-point regulation of linear impulsive systems. IEEETransactions on Automatic Control, 63:2994–3001.
dc.relationRuan, Y., Wilinska, M. E., Thabit, H., and Hovorka, R. (2017). Modelingday-to-day variability of glucose-insulin regulation over 12-week home use of closed-loopinsulin delivery.IEEE Transactions on Biomedical Engineering, 64:1412–1419.
dc.relationS.F., P. (2021).Floating Point Arithmetic: Issues and Limitations.https://docs.python.org/3/tutorial/floatingpoint.html.
dc.relationShi, D., Dassau, E., and Doyle, F. J. (2019). Adaptive zone model predictive control of artificial pancreas based on glucose- and velocity-dependent control penalties.IEEE transactions on biomedical engineering, 66:1045–1054.
dc.relationSimon, D. (2006).Optimal State Estimation: Kalman, H Infinity, and Nonli-near Approaches. Wiley-Interscience 1st Edition
dc.relationSra, S., Nowozin, S., and Wright, J. (2011).Optimization for MachineLearning. Neural Information Processing series.
dc.relationStellato, B., Banjac, G., Goulart, P., Bemporad, A., and Boyd, S.(2020). OSQP: an operator splitting solver for quadratic programs.Mathematical Pro-gramming Computation, 12:637–672.
dc.relationStellato, B., Naik, V., Bemporad, A., Goulart, P., and Boyd, S. (2018).Embedded mixed-integer quadratic optimization using the osqp solver.European ControlConference, ECC, in press:1–6.
dc.relationthe Future, D. D. (2021).Gravity: I2C Digital Wattmeter.https://wiki.dfrobot.com/Gravity
dc.relationToffanin, C., Aiello, E. M., Del Favero, S., Cobelli, C., and Magni,L. (2019). Multiple models for artificial pancreas predictions identified from free-livingcondition data: A proof of concept study. Process Control, 77:29–37.
dc.relationVettoretti, M., Facchinetti, A., Sparacino, G., and Cobelli, C. (2018).Type-1 diabetes patient decision simula-tor for in silico testing safety and effectiveness ofinsulin treatments.IEEE Transactions on BiomedicalEngineering, 65:1281–1290.
dc.relationVilla-tamayo, M. (2021). Compensación del efecto de variaciones fisiológicas en la glucemia de pacientes diabéticos tipo 1 utilizando control predictivo con entradas impulsivas. Master’s thesis, Universidad Nacional de Colombia.
dc.relationVilla-Tamayo, M. F., Leon-Vargas, F., Garcia-Jaramillo, M.,and Rivadeneira, P. S. (2021a). Glycemic control strategy based on an impulsive mpcwith safety layer coupling for iob limitation.IEEE Control Systems Letters, 5(5):1669–1674.
dc.relationVilla-Tamayo, M. F., Leon-Vargas, F., Garcia-Jaramillo, M.,and Rivadeneira, P. S. (2021b). Glycemic control strategy based on an impulsive mpcwith safety layer couplingfor iob limitation.IEEE Control Systems Letters, pages 1669–1674.
dc.relationVilla-Tamayo, M. F. and Rivadeneira, P. S. (2020a).Adaptive impulsive offset-free mpc to handle parameter variations for type 1 diabetestreatment.Ind. Eng. Chem. Res., 59(13):5865–5876.
dc.relationVilla-Tamayo, M. F. and Rivadeneira, P. S. (2020b).Adaptive impulsive offset-freempc to handle parameter variations for type 1 diabetes treat-ment.Ind. Eng. Chem.Res, 59:5865–5876.
dc.relationZavitsanou, S., Chakrabarty, A., Dassau, E., and Doyle, F. J.(2016). Embedded control in wearable medical devices: Application to the artificial pan-creas.Processes, 4.
dc.rightsAtribución-NoComercial-SinDerivadas 4.0 Internacional
dc.rightshttp://creativecommons.org/licenses/by-nc-nd/4.0/
dc.rightsinfo:eu-repo/semantics/openAccess
dc.titleDesarrollo de estrategias para el uso eficiente de los recursos energéticos en un control predictivo basado en modelo implementado en un sistema embebido para el tratamiento de la diabetes mellitus tipo I
dc.typeTesis


Este ítem pertenece a la siguiente institución