Mostrar el registro sencillo del ítem

dc.contributor.advisorOtero, Fernando
dc.contributor.authorPaba Argote, Harry Joséspa
dc.contributor.authorNiño Cossio, Eudilsonspa
dc.date.accessioned2020-06-26T21:32:13Z
dc.date.available2020-06-26T21:32:13Z
dc.date.issued2000
dc.identifier.urihttp://hdl.handle.net/20.500.12749/3293
dc.description.abstractEn este trabajo se desarrolló un sensor virtual para predecir el índice de viscosidad de un proceso de la industria petroquímica denominado Extracción de Aceites Lubricantes con Fenol. Para ello se utiliza una red neuronal entrenada por el método de Activación de Pesos Aleatorios, técnica que a diferencia de otras no es iterativa y por ello resulta ser mucho más rápida que otros métodos tradicionales, alcanzando objetivos de error suficientemente bajos para reemplazar el sensor real. Como resultado, la herramienta de software elaborada para tal fin puede ser utilizada en el entrenamiento de cualquier sistema (entradas-salida) que se pretenda resolver aplicando redes neuronales de este tipo. Cuenta con todos los pasos intermedios requeridos como la asistencia en la selección de variables por métodos estadísticos que utilizan la matemática requerida para el tratamiento de esta clase de procesos estocásticos, pruebas de posible linealidad, tratamiento de señal para filtrar ruido y eliminar datos falsos, entrenamiento-validación y pruebas por simulación fuera de línea y en línea con el proceso real. Se utiliza la medición del Error RMS y el Máximo Error encontrado, principalmente en la fase de validación para ser usado como el parámetro de comparación que permita evaluar el desempeño del modelo obtenido. La técnica de entrenamiento utilizada es suficientemente rápida para implementar funciones de reentrenamiento en línea en caso de ser requerido por el sistema lo cual se registra gráficamente para atender este requerimiento.spa
dc.description.sponsorshipInstituto Tecnológico de Estudios Superiores de Monterrey ITESMspa
dc.description.tableofcontentsMARCO TEDRICO DISEÑO DE UN SENSOR INTELIGENTE PARA INFERIR EL IV EN LA PLANTA DE EXTRACCIÓN DE ACEITES LUBRICANTES CON FENOL PRUEBAS CONCLUSIONES Y RECOMENDACIONES BIBLIOGRAFIA
dc.format.mimetypeapplication/pdfspa
dc.language.isospaspa
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/2.5/co/*
dc.titleImplementación de sensores inteligentes utilizando redes neuronales aplicados en procesos de refinación del petróleospa
dc.title.translatedImplementation of smart sensors using neural networks applied in oil refining processeseng
dc.degree.nameMagíster en Ciencias Computacionalesspa
dc.coverageBucaramanga (Colombia)spa
dc.publisher.grantorUniversidad Autónoma de Bucaramanga UNABspa
dc.rights.localAbierto (Texto Completo)spa
dc.publisher.facultyFacultad Ingenieríaspa
dc.publisher.programMaestría en Ciencias Computacionalesspa
dc.description.degreelevelMaestríaspa
dc.type.driverinfo:eu-repo/semantics/masterThesis
dc.type.localTesisspa
dc.type.coarhttp://purl.org/coar/resource_type/c_bdcc
dc.subject.keywordsArtificial intelligenceeng
dc.subject.keywordsNeural Networks (Computers)eng
dc.subject.keywordsOil refiningeng
dc.subject.keywordsInformation systemseng
dc.subject.keywordsComputer scienceeng
dc.subject.keywordsInvestigationseng
dc.subject.keywordsAnalysiseng
dc.identifier.instnameinstname:Universidad Autónoma de Bucaramanga - UNABspa
dc.identifier.reponamereponame:Repositorio Institucional UNABspa
dc.type.hasversioninfo:eu-repo/semantics/acceptedVersion
dc.rights.accessrightsinfo:eu-repo/semantics/openAccessspa
dc.rights.accessrightshttp://purl.org/coar/access_right/c_abf2spa
dc.relation.referencesPaba Argote, Harry José, Niño Cossio, Eudilson (2000). Implementación de sensores inteligentes utilizando redes neuronales aplicados en procesos de refinación del petróleo. Bucaramanga (Colombia) : Universidad Autónoma de Bucaramanga UNAB, Instituto Tecnológico y de Estudios Superiores de Monterrey ITESMspa
dc.relation.referencesBABUSCA, R. y Verbruggen, H. B., Applied Fuzzy Modeling. Applied Fuzzy Modeling, Proceedings of the IFAC Symposium on Artificial Intelligence in Real Time Control, 1994. 61-66p.
dc.relation.referencesBABUSCA, R. y Verbruggen, H.B., Identification of composite linear models via fuzzy clustering, Proceedings of the ECC'95, 1995. 1207-1212p.
dc.relation.referencesBALDI, P., Gradient descent learning algorithm overview: a general dynamical systems perspective, IEEE Transactions on Neural Networks, 1995. 6:182-195p.
dc.relation.referencesBARHEN. J.. Toomarian. N. and Fijany, A., Learning without local minima, Proceedings of the lEEE International Conference on Neural Networks (ICNN), IEEE World Congress on Computational Intelligence, 1994. 71:4592-4596p.
dc.relation.referencesBARNET, M. W., Adaptative Neural Networks for Intrelligent Operation of the Activaded Sludge Process, Gensym Corp., 2000. 9p.
dc.relation.referencesBASHEER, A. l. And Najjar, Y.M., Predicting Dynamic Response of Adsorption Columns with Neural Nets, Journal of Computing in Civil Engineering, 1996. 10(1): 31-39p.
dc.relation.referencesBARRON, A. R., Universal approximation bounds for superpositions of sigmoidal function. IEEE Transactions on Information Theory, 1993. 39, 930-945p.
dc.relation.referencesBARTO, R. S. Sutton, and C. J. C. H. Watkins, Learning and sequential decision making. Technical report, COINS. 1989.
dc.relation.referencesBATES, J. M. and C.W. J. Granger, The combination of forecast, Operations Research Quarterly, 1969. 20:319-325p.
dc.relation.referencesBATTIKHA, N.E., The Condensed Handbook of Measurement and Control. Instrument Society of America. 1997.
dc.relation.referencesBATTITI, R., Usual mutual information for selecting features in supervised neural network learning. IEEE Transactions on Neural Networks. 1995.
dc.relation.referencesBHAT, N. and McAvoy, T.J., Use of Neural Nets for Dynamic Modeling and Control of Chemical Process Systems, Computers Chem.Engng., 1990. 14(4/5): 573-583p.
dc.relation.referencesBILLIGS S. A., Jamaluddin H. B. and Chen S. A comparison of the backpropagation and recursive prediction error algorithms for training neural networks. Mech. Syst. Signal Process. 1991, 5, 233-255p.
dc.relation.referencesBRAAKE, H. A. B. and van Straten, G. Random Activation Weight neural net (RAVWN) for fast non-iterative training. Engineering Applications of Artificial Intelligence, 1995. 8, 71- 80p.
dc.relation.referencesBRAAKE, H. A. B., van Can, H. J. L., van Straten, G. and Verbruggen, H. B.. Regulated Activation Weights Neural Network (RAWN). Proceedings of the European Symposium on Artificial Neural Networks, ESANN'96, Brugge, 1996. 19-24p.
dc.relation.referencesBRAJESH, K.J., Tambe, S.S. and Kulkarni, B.D., Estimating Diffusion Coefficients of a Micellar System Using an Artificial Neural Network, J. Colloid 8, Interface Science, 1995, 170:392-398p.
dc.relation.referencesCHAO, L.L., Estadistica para las Ciencias Administrativas. McGraw-Hill. 1975.
dc.relation.referencesCHERRE, D., Use of Artificial Neural Networks in Process Control, Master of Science Thesis, University of South Florida, Tampa. 1998.
dc.relation.referencesCHEN S., Cowan C. F. N., Billings S. S and Grant P. M. Parallel recursive prediction error algorithm for training layered neural networks. Int. J. Control. 1990. 51, 1215-1228p.
dc.relation.referencesCORRIPIO, A., Smith, C., Control Automatico de Procesos. Limusa. 1996.
dc.relation.referencesCYBENKO G. Approximation by superpositions of a sigmoidal function. Math. Control Signals Syst. 1989.2, 303-314p.
dc.relation.referencesDABNEY, James. Mastering Simulink 2, Upper Saddle River, NJ: Prentice Hall, 1998. 345p.
dc.relation.referencesDALLAS, E. J. Métodos Multivariados Aplicados al Analisis de Datos. Mexico, 1998. 566p.
dc.relation.referencesDELGADO, J.A., Elementos de Redes Neuronales y Algoritmos Genéticos, Neuronales Ltda. 1992.
dc.relation.referencesDAVE, R. N. Boundary Detection through Fuzzy Clustering. Proceedings of the IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), San Diego, 1992. 127-134p.
dc.relation.referencesDAYHOFF, J. Neural Networks Architectures: An Introduction, New York, NY: Van Nostrand Reinhold, 1990. 320p.
dc.relation.referencesDIEDERICH, J. Artificial. Neural Networks: Concept Learning, Los Alamitos, CA: Computer Society Press, 1990. 160p
dc.relation.referencesDRAGO G. P. and Ridella S. Statistically controlled activation weight initialization (SCAWI). IEEE Trans. Neural Networks 1992. 3, 627-631p
dc.relation.referencesFORTUNA, L., Graciani, S., Lo Presti, M. and Muscato, G. Improving backpropagation learning using auxiliary neural networks. International, Journal of Control, 1992. 55, 793- 907p.
dc.relation.referencesFUNAHASHI, K.!., On the aproximate realization of continuous mapping by neural networks. Neural Networks, 1989. 2, 183-192p.
dc.relation.referencesGOLUB, G. H. and van Loan, C. F. An Analysis of the Total Least Squares Problem. SIAM Journal on Numerical Analusus, 1980. 17, 883-893p.
dc.relation.referencesHAGAN, M.T. and Menhaj, M., Training Feedforward Networks with the Marquadt Algorithm, IEEE Transactions on Neural Networks, 1994, 5(6):989-993p.
dc.relation.referencesHANSELMAN, D., Littlefield, B., Mastering Matlab 5. Prentice Hall. 1998.
dc.relation.referencesHASSOUN, M. H. Fundamentals of Artificial Neural Networks. The Mit Press, Cambridge, MA. 1995.
dc.relation.referencesHAYKIN, S., Neural Networks. A Comprehensive Foundation, Macmillan College Publishing Company, N.Y. 1994
dc.relation.referencesHONEYWELL Hi-Spec Solutions, Lube Oil Extraction, Hidrocarbon Processing, Phoenix. 1999. 99:125p.
dc.relation.referencesHUNT, K. J. Sbarbaro, D., Zbikowki, R and Gawthrop, P. J., Neural Network for control systems a survey. Automatica, 1992. 28:1083-1112p.
dc.relation.referencesIGOR, Aleksander. Neural Computing Architectures, Cambridge, MA: MIT Press, 1989. 240p.
dc.relation.referencesJANG, J. S. Neuro-Fuzzy and Soft Computing. Upper Saddle River, NJ: Prentice Hall, 1997. 614p.
dc.relation.referencesJOHNSON, D.E., Metodos Multivariados Aplicados al Analisis de Datos. Thomson Editores. 2000.
dc.relation.referencesK-PATTENTS, Inc. Instruction Manual for PR-01-S (-EX/FM) Version 6.0. 1996,
dc.relation.referencesKOSKO, B. Neural networks and fuzzy system; a dynamical systems approach to machine intelligence. Prentice Hall, Englewood Cliffs, Nj. 1992.
dc.relation.referencesKREYSZIG E. Introductory Mathematical Statistics; Principles and Methods. Wiley, New York .1970,
dc.relation.referencesLEVIN, Mijail. Automatización de Procesos Piro e Hidrometalúrgicos, Saint Petersburgo, Rusia: Editorial Ciencia y Técnica, 1989.
dc.relation.referencesLIN, Chin-Teng. Neural Fuzzy Systems, Upper Saddle River, NJ: Prentice Hall, 1995. 797p.
dc.relation.referencesLITTLEFIELD, Bruce. Mastering Matlab 5, Upper Saddle River, NJ: Prentice Hall, 1998. 639p.
dc.relation.referencesMARE, Harston. Handbook of Neural Computing Applications, San Diego, CA: Academic Press, 1990. 245p.
dc.relation.referencesMASTERS, Timothy. Practical Neural Network Recipes in C++. San Diego, CA: Academic Press. 1993. 493p.
dc.relation.referencesMINSKY, M. L. Perceptrons (Expanded Edtition), Cambridge, MA: MIT Press, 1988. 230p.
dc.relation.referencesMYERS, R. H., Classical and Modern Regression with Applications, Duxbury Press. 1986.
dc.relation.referencesOGATA, K., Ingenieria de Control Moderna. Prentice Hall. 1980.
dc.relation.referencesORDOÑEZ, O., Aplicación de una metodología estadística para estimar beneficios de implantar sistema de control avanzado a la unidad de extracción de bases lubricantes con fenol en el Complejo Industrial de Barrancabermeja ECOPETROL. Universidad Insustrial de Santander. Escuela de Ingeniería Química. Bucaramanga. 1995. 252 p.
dc.relation.referencesOTERO, F., Use of Neural Networks in Process Engineering. CTSF, 1998. 49-64p.
dc.relation.referencesPAO Y-H., Phillips S. M. and Sobajic D. J. Neural-net computing and the intelligent control of systems. Int. J. Control. 1992. 56, 263-289p
dc.relation.referencesPEREZ, R., Artificial Neural Networks, Computer Science and Eng. Dept., University of South Florida. 1998.
dc.relation.referencesQUINTERO, A., Sistemas Unitarios de Control. Universidad Pontificia Bolivariana, Facultad de Ingenieria Electrica y Electrónica, Medellín. 1993.
dc.relation.referencesRAMESH, K. Tock, R. W. and Narayan, R.S., Prediction of solvent Activity in Polymer System with Neural Networks, Ind. Eng. Chem. Res., 1995. 34:3974-3980p.
dc.relation.referencesSHINSKEY, F.G., Process Control System. McGraw-Hill. 1988.
dc.relation.referencesSIMPSON, Patrick K. Artificial Neural Systems, New York, NY: Pergamon Press, 1990. 162p.
dc.relation.referencesSINGHAL S. and Wu L. Training feed-forward networks with the extended Dalman algorithm. IEEE Proc. 1989. 1187-1190p.
dc.relation.referencesSCALERO R. S. and Tependelenlioglu N. A fast new algorithm for training feedforward neural networks. IEEE Trans. Signal Process. 1992. 40, 202-210p.
dc.relation.referencesSOVKOVIC-Stevanovic,J., Neural Network for Process Analysis and Optimization: Modeling and Applications, Computers Chem. Engng., 1994. 18(11/12):1149-1155p.
dc.relation.referencesSTEPHEN, J. Neural Network Design and The Complexity of Learning, Cambridge, MA, 1990. 305p.
dc.relation.referencesSWINGLER, Kevin. Applying Neural Networks: A practical Guide, San Diego, CA: Academic Press, 1996. 303p.
dc.relation.referencesTE BRAAKE, H.A.B., Van Can, H.J.L., Van Straten, G. and Verbrugger, H.B., Two-step Approach in Training of Regulated Activation Weight Networks (RAWN), Engng. Applic.Artif.Intell., 1997, 10(2):157-170p.
dc.relation.referencesTE BRAAKE, H.A.B., Van Straten, G. Random Activation Weight Neural Net (RAWN) for Fast Non-Interactive Training, Engng. Applic. Artif. Intell. 1995. 8(1):71-80p.
dc.relation.referencesTHAM, M.T., A. Montague, et al., Soft-sensors for process estimation and inferential control, J, Proc. Cont., 1991. 1:3-14p.
dc.relation.referencesTHOMPSON, W., How Neural Network Modeling Methods Complement Those of Physical Modeling, NPRA Computer Conference. 1996.
dc.relation.referencesVALENCIA, J.H., Sistemas Automáticos de Contro, Universidad Ponficia Bolivariana, Facultad de Ingenieria Electrica y Electrónica, Medellín. 1994.
dc.relation.referencesVAN CAN, H. J. L.., Braake Te, H.A.B.. Hellings, C., Krijgsman, A, J.. Verbruggen, H. B., Luyben, K, Ch. A. M. and Heijnen, J.J., Desing and real time testing of a neural model predictive controller for a nonlinear system. Chemical Engineering Science , 1994, 50,2419-2430p.
dc.relation.referencesVEMURI, V. Artificial Neural Networks: Theoretical Concepts, Los Alamitos, CA: Computer Society Press, 1990, 125p.
dc.relation.referencesVILLALOBOS, E., Aplicacion de Redes Neuronales al Desarrollo de Sensores Virtuales, Postgrado en Control de Procesos, Universidad del Zulia. 1999.
dc.relation.referencesWASSERMAN, P.D. Neural Computing, New York, NY: Van Nostrand-Reinhold, 1989. 202p.
dc.relation.referencesYU, J., Zhong, W., Improve nonlinear soft sensing modeling by combining multiple models, Hydrocarbon Processing, 2000. 108-112p.
dc.relation.referencesProcesos Estocásticos http://www.itlp.edu.mx/publica/tutoriales/investoper2/tema44.htm
dc.relation.referencesA Real-Time Neural Network-Based Intelligent Performance and Emissions Prediction System for on-board Diagnostics and Engine Control. hitp://www.cemr.wvu.edu/-virtsens/
dc.relation.referencesVirtual Sensors http://www.chaotic.com/chaos/virtual.html
dc.relation.referencesJournal sobre Sensores http://www.mcb.co.uk/sr.htm
dc.relation.referencesMCB University Press. Centro Para Tecnología de Sensores http://www. ee.upenn.edu/center. cst. html http://pender.ee.upenn edu/-jan/
dc.relation.referencesViscosidad hitp://www.kimble.com/html/Viscosity-254.html
dc.relation.referencesViscosímetros de Laboratorio Saybolt Viscosimeter Bath http://www.geneq.com/catalog/en/asphalt/saybolt viscosimeter bath.html
dc.relation.referencesViscosímetros en Línea http://www. ige.ethz.ch/pel/annrep95/p1 15.html
dc.relation.referencesGENSYM CORP. http://www.gensym.com
dc.relation.referencesNEURODIMENSION INC. http://www. nd. com/
dc.relation.referencesZ-SOLUTIONS. Neural Networks: Advanced Technology That Can Change The Way You Work http://www. zsolutions.com/index.htm
dc.relation.referencesAn Introduction to Neural Networks http://www, cs.stir. ac. uk/-Iss/NN Intro/InvSlides. html
dc.relation.referencesIntroducción a redes neuronales http://www. cs. stir. ac. uk/-Iss/NNIntro/InvSlides.html
dc.relation.referencesElementos de una red neuronal http://www. ka.net/sacineural.html
dc.relation.referencesPublicaciones de Inteligencia Artificial del Departamento de Ciencias Experimentales e Ingeniería de la Universidad Rey Juan Carlos. http://www. escet urjc.es/-¡a/grupo/publicaciones.html
dc.relation.referencesNeural network based virtual sensor http://www, cemr.wvu.edu/-virtsens/
dc.subject.lembInteligencia artificialspa
dc.subject.lembRedes neuronales (Computadores)spa
dc.subject.lembRefinación del petróleospa
dc.subject.lembSistemas de informaciónspa
dc.subject.lembCiencias computacionalesspa
dc.subject.lembInvestigacionesspa
dc.subject.lembAnálisisspa
dc.description.abstractenglishIn this work, a virtual sensor was developed to predict the viscosity index of a process in the petrochemical industry called Extraction of Lubricating Oils with Phenol. For this, a neural network trained by the Random Weights Activation method is used, a technique that, unlike others, is not iterative and therefore turns out to be much faster than other traditional methods, reaching error objectives low enough to replace the real sensor. . As a result, the software tool developed for this purpose can be used in the training of any system (input-output) that is intended to be solved by applying neural networks of this type. It has all the intermediate steps required such as assistance in the selection of variables by statistical methods that use the mathematics required for the treatment of this type of stochastic processes, tests for possible linearity, signal treatment to filter noise and eliminate false data, training -validation and testing by simulation offline and in line with the real process. The measurement of the RMS Error and the Maximum Error found is used, mainly in the validation phase to be used as the comparison parameter that allows evaluating the performance of the obtained model. The training technique used is fast enough to implement online retraining functions if required by the system, which is recorded graphically to meet this requirement.eng
dc.type.redcolhttp://purl.org/redcol/resource_type/TM
dc.rights.creativecommonsAtribución-NoComercial-SinDerivadas 2.5 Colombia*
dc.coverage.campusUNAB Campus Bucaramangaspa
dc.description.learningmodalityModalidad Presencialspa


Ficheros en el ítem

Thumbnail

Este ítem aparece en la(s) siguiente(s) colección(ones)

Mostrar el registro sencillo del ítem

Atribución-NoComercial-SinDerivadas 2.5 Colombia
Excepto si se señala otra cosa, la licencia del ítem se describe como Atribución-NoComercial-SinDerivadas 2.5 Colombia