http://hdl.handle.net/2117/1126 Wed, 19 Jun 2013 03:56:24 GMT 2013-06-19T03:56:24Z webmaster.bupc@upc.edu Universitat Politècnica de Catalunya. Servei de Biblioteques i Documentació no http://hdl.handle.net/2117/19206 Title: Negotiation and learning in distributed MPC of large scale systems Authors: Javalera Rincón, Valeria; Morcego Seix, Bernardo; Puig Cayuela, Vicenç Abstract: A key issue in distributed MPC control of Large Scale Systems (LSS) is how shared variables among the different MPC controller in charge of controlling each system partition (subsystems) are handled. When these connections represent control variables, the distributed control has to be consistent for both subsystems and the optimal value of these variables will have to accomplish a common goal. In order to achieve this, the present work combines ideas from Distributed Artificial Intelligence (DAI), Reinforcement Learning (RL) and Model Predictive Control (MPC) in order to provide an approach based on negotiation, cooperation and learning techniques. Results of the application of this approach to a small drinking water network show that the resulting trajectories of the levels in tanks (control variables) can be acceptable compared to the centralized solution. The application to a real network (the Barcelona case) is currently under development. Tue, 14 May 2013 12:34:16 GMT http://hdl.handle.net/2117/19206 2013-05-14T12:34:16Z Javalera Rincón, Valeria; Morcego Seix, Bernardo; Puig Cayuela, Vicenç no predictive control PARAULES AUTOR:cooperative systems, distributed control, model predictive control, multi agent systems, negotiation, reinforcement learning A key issue in distributed MPC control of Large Scale Systems (LSS) is how shared variables among the different MPC controller in charge of controlling each system partition (subsystems) are handled. When these connections represent control variables, the distributed control has to be consistent for both subsystems and the optimal value of these variables will have to accomplish a common goal. In order to achieve this, the present work combines ideas from Distributed Artificial Intelligence (DAI), Reinforcement Learning (RL) and Model Predictive Control (MPC) in order to provide an approach based on negotiation, cooperation and learning techniques. Results of the application of this approach to a small drinking water network show that the resulting trajectories of the levels in tanks (control variables) can be acceptable compared to the centralized solution. The application to a real network (the Barcelona case) is currently under development. http://hdl.handle.net/2117/18544 Title: Self-video recording for the integration and assessment of generic competencies Authors: Masip Álvarez, Albert; Hervada Sala, Carme; Pàmies Gómez, Teresa; Arias Pujol, Antoni; Jaen Fernandez, Carles; Rodríguez Sorigué, M. Cristina; Romero Durán, David; Nejjari Akhi-Elarab, Fatiha; Álvarez del Castillo, María Dolores; Roca Lefler, Marta; Hernández Farràs, Jorge; Montferrer Liñan, Raul Abstract: This document presents the use of a self-video-recording service for the integration and assessment of certain generic competences (oral and written communication, teamwork and third (English) language) in all degree programs related to the Industrial area at the Terrassa School of Engineering (EET). The resulting service allows to carry out the assessment through active and collaborative learning methodologies using ICT as a support, becoming an educational resource where students are encouraged to evaluate their own activity, individually and in groups (co-evaluation), and with which the faculty can enhance certain sections of the syllabus while also evaluating generic competences. This paper outlines the rationale and design of the system based on self-recording-video that assists work on some generic competences. A preliminary experience has taken place and it has been carried out by volunteer students and some professors from the promoting team. It consisted of following all steps to obtain the video-recording as a result and fulfil the further assessment. Preparing the defence of the final project career, explaining the way a problem is faced are examples of use of the service endowed to assist the assessment of some generic competences. A good number of students and professors have participated in this preliminary experiment carried out on some subjects and in which oral effective communication will be assessed. A specific rubric for the assessment of generic competences has been used both for students and faculty. There is a summary of qualitative feedback from both students and instructors, and where several issues related to improving the self-assessment and self-video-recording system, are discussed. Finally, a survey is proposed, whose results will be helpful for its improvement. Tue, 02 Apr 2013 14:02:57 GMT http://hdl.handle.net/2117/18544 2013-04-02T14:02:57Z Masip Álvarez, Albert; Hervada Sala, Carme; Pàmies Gómez, Teresa; Arias Pujol, Antoni; Jaen Fernandez, Carles; Rodríguez Sorigué, M. Cristina; Romero Durán, David; Nejjari Akhi-Elarab, Fatiha; Álvarez del Castillo, María Dolores; Roca Lefler, Marta; Hernández Farràs, Jorge; Montferrer Liñan, Raul no This document presents the use of a self-video-recording service for the integration and assessment of certain generic competences (oral and written communication, teamwork and third (English) language) in all degree programs related to the Industrial area at the Terrassa School of Engineering (EET). The resulting service allows to carry out the assessment through active and collaborative learning methodologies using ICT as a support, becoming an educational resource where students are encouraged to evaluate their own activity, individually and in groups (co-evaluation), and with which the faculty can enhance certain sections of the syllabus while also evaluating generic competences. This paper outlines the rationale and design of the system based on self-recording-video that assists work on some generic competences. A preliminary experience has taken place and it has been carried out by volunteer students and some professors from the promoting team. It consisted of following all steps to obtain the video-recording as a result and fulfil the further assessment. Preparing the defence of the final project career, explaining the way a problem is faced are examples of use of the service endowed to assist the assessment of some generic competences. A good number of students and professors have participated in this preliminary experiment carried out on some subjects and in which oral effective communication will be assessed. A specific rubric for the assessment of generic competences has been used both for students and faculty. There is a summary of qualitative feedback from both students and instructors, and where several issues related to improving the self-assessment and self-video-recording system, are discussed. Finally, a survey is proposed, whose results will be helpful for its improvement. http://hdl.handle.net/2117/18367 Title: Real time diagnosis for a large gas turbine based on a deep model of the controller Authors: Guasch Petit, Antonio; Sarrate Estruch, Ramon Abstract: The ESPRIT project 6862 TIGER addresses the development of a real-time knowledge based diagnosis system for two gas turbines: a large scale industrial gas turbine and an auxiliary power unit for aviation. The project will combine the qualitative model based with the rule based diagnosis approach. This paper describes the rule based approach taken in the development of a real-time diagnostic system for a large scale gas turbine. Deep models will be automatically extracted from the deterministic knowledge embedded within the controller ladder logic diagrams. Heuristic knowledge will be extracted from the domain expert and engineering manuals. Mon, 18 Mar 2013 12:30:17 GMT http://hdl.handle.net/2117/18367 2013-03-18T12:30:17Z Guasch Petit, Antonio; Sarrate Estruch, Ramon no The ESPRIT project 6862 TIGER addresses the development of a real-time knowledge based diagnosis system for two gas turbines: a large scale industrial gas turbine and an auxiliary power unit for aviation. The project will combine the qualitative model based with the rule based diagnosis approach. This paper describes the rule based approach taken in the development of a real-time diagnostic system for a large scale gas turbine. Deep models will be automatically extracted from the deterministic knowledge embedded within the controller ladder logic diagrams. Heuristic knowledge will be extracted from the domain expert and engineering manuals. http://hdl.handle.net/2117/17990 Title: Multi-rate decentralized MPC strategy for drinking water networks: application to the Barcelona case study Authors: Ocampo-Martínez, Carlos; Puig Cayuela, Vicenç; Montes de Oca, Saúl Abstract: In this paper, a multi - rate decentralised model predictive control (MDMPC) approach is proposed for drinking water networks (DWN). The upper temporal layer works with a daily time scale and is in charge of achieving the global objectives , i.e., optimal selection of the water sources and the path to the reservoirs. On the other hand, the lower temporal layer is in charge of manipulating the set - point of the actuators to satisfy the local objectives, i.e., the minimisation of the energy needed for pumping water to the reservoirs. Once obtained the system decomposition into subsystems, some bidirectional flows of information between the MPC controllers that control the subsystems appea r . Hence, the proposed DMPC strategy co pes with them , resulting in a hierarchical - like scheme. Results obtained when used selected simulation scenarios over a partitioned model of the Barcelona DWN show the effectiveness of the control strategy in terms of system modularity, reduced computational burden and, at the same time, the admissible loss of performance in contrast to a centralised MPC (CMPC) strategy. Tue, 26 Feb 2013 19:14:41 GMT http://hdl.handle.net/2117/17990 2013-02-26T19:14:41Z Ocampo-Martínez, Carlos; Puig Cayuela, Vicenç; Montes de Oca, Saúl no control theory optimisation social aspects of automation In this paper, a multi - rate decentralised model predictive control (MDMPC) approach is proposed for drinking water networks (DWN). The upper temporal layer works with a daily time scale and is in charge of achieving the global objectives , i.e., optimal selection of the water sources and the path to the reservoirs. On the other hand, the lower temporal layer is in charge of manipulating the set - point of the actuators to satisfy the local objectives, i.e., the minimisation of the energy needed for pumping water to the reservoirs. Once obtained the system decomposition into subsystems, some bidirectional flows of information between the MPC controllers that control the subsystems appea r . Hence, the proposed DMPC strategy co pes with them , resulting in a hierarchical - like scheme. Results obtained when used selected simulation scenarios over a partitioned model of the Barcelona DWN show the effectiveness of the control strategy in terms of system modularity, reduced computational burden and, at the same time, the admissible loss of performance in contrast to a centralised MPC (CMPC) strategy. http://hdl.handle.net/2117/17983 Title: Automatic control to improve the seaworthiness conditions in inland navigation networks: application to a channel in the North of France Authors: Blesa Izquierdo, Joaquim; Duviella, Eric; Sayeb-Mouchaweb, Mohamar; Puig Cayuela, Vicenç; Chuquet, Karine Tue, 26 Feb 2013 13:52:19 GMT http://hdl.handle.net/2117/17983 2013-02-26T13:52:19Z Blesa Izquierdo, Joaquim; Duviella, Eric; Sayeb-Mouchaweb, Mohamar; Puig Cayuela, Vicenç; Chuquet, Karine no http://hdl.handle.net/2117/17982 Title: FTC design for polytopic LPV systems subject to actuator saturations Authors: Rotondo, Damiano; Nejjari Akhi-Elarab, Fatiha; Puig Cayuela, Vicenç Abstract: In this paper, a Fault Tolerant Control (FTC) scheme using virtual actuators for discrete-time polytopic Linear Parameter Varying (LPV) systems subject to actuator saturations is proposed. The plant with the faulty actuator is augmented to take into account the saturations and modified adding the virtual actuator block that masks the fault. The elements of the FTC control loop are designed using Linear Matrix Inequalities (LMIs) in order to achieve the Pole Placement and the H2 norm specifications. To assess the performance of the proposed approach an aeronautical application is used. Tue, 26 Feb 2013 13:29:46 GMT http://hdl.handle.net/2117/17982 2013-02-26T13:29:46Z Rotondo, Damiano; Nejjari Akhi-Elarab, Fatiha; Puig Cayuela, Vicenç no In this paper, a Fault Tolerant Control (FTC) scheme using virtual actuators for discrete-time polytopic Linear Parameter Varying (LPV) systems subject to actuator saturations is proposed. The plant with the faulty actuator is augmented to take into account the saturations and modified adding the virtual actuator block that masks the fault. The elements of the FTC control loop are designed using Linear Matrix Inequalities (LMIs) in order to achieve the Pole Placement and the H2 norm specifications. To assess the performance of the proposed approach an aeronautical application is used. http://hdl.handle.net/2117/17981 Title: Parity space hybrid system diagnosis under model uncertainty Authors: Vento Maldonado, Jorge; Puig Cayuela, Vicenç; Sarrate Estruch, Ramon Abstract: In this paper, diagnosis for hybrid systems using a parity space approach that considers model uncertainty is proposed. The hybrid diagnoser is composed of modules which carry out the mode recognition and diagnosis tasks both based on residuals generated using a model. Both tasks interact each other since the diagnosis module adapts himself according to the current mode of the hybrid system. Moreover, the methodology takes into account the parameter uncertainty using a passive robust strategy. An adaptive threshold for residual evaluation is generated and the parity space approach is used to design a set of residuals for each mode. The proposed fault diagnosis approach for hybrid systems is illustrated on a part of the Barcelona sewer network. Tue, 26 Feb 2013 13:12:59 GMT http://hdl.handle.net/2117/17981 2013-02-26T13:12:59Z Vento Maldonado, Jorge; Puig Cayuela, Vicenç; Sarrate Estruch, Ramon no In this paper, diagnosis for hybrid systems using a parity space approach that considers model uncertainty is proposed. The hybrid diagnoser is composed of modules which carry out the mode recognition and diagnosis tasks both based on residuals generated using a model. Both tasks interact each other since the diagnosis module adapts himself according to the current mode of the hybrid system. Moreover, the methodology takes into account the parameter uncertainty using a passive robust strategy. An adaptive threshold for residual evaluation is generated and the parity space approach is used to design a set of residuals for each mode. The proposed fault diagnosis approach for hybrid systems is illustrated on a part of the Barcelona sewer network. http://hdl.handle.net/2117/17958 Title: Fault detection and isolation of a real PEM fuel cell using interval LPV observers Authors: Lira Ramírez, Salvador de; Puig Cayuela, Vicenç; Quevedo Casín, Joseba Jokin Abstract: The purpose of this paper is to design a Fault Detection and Isolation (FDI) system based on interval LPV observers with application to a real fuel cell. The interval LPV observer is used to generate an adaptive threshold to enhance the robustness of the fault detection test. Real field data and system identification techniques are used to identify the nominal model as well as its uncertainty. Since fuel cells are highly non-linear systems when operating in their whole range of operation, a LPV model is used. Finally, fault isolation is based on an algorithm that uses the residual fault sensitivity. Several fault scenarios are used to show the performance of the proposed approach. Mon, 25 Feb 2013 12:56:09 GMT http://hdl.handle.net/2117/17958 2013-02-25T12:56:09Z Lira Ramírez, Salvador de; Puig Cayuela, Vicenç; Quevedo Casín, Joseba Jokin no The purpose of this paper is to design a Fault Detection and Isolation (FDI) system based on interval LPV observers with application to a real fuel cell. The interval LPV observer is used to generate an adaptive threshold to enhance the robustness of the fault detection test. Real field data and system identification techniques are used to identify the nominal model as well as its uncertainty. Since fuel cells are highly non-linear systems when operating in their whole range of operation, a LPV model is used. Finally, fault isolation is based on an algorithm that uses the residual fault sensitivity. Several fault scenarios are used to show the performance of the proposed approach. http://hdl.handle.net/2117/17945 Title: Fault estimation and virtual actuator FTC approach for LPV systems Authors: Rotondo, Damiano; Nejjari Akhi-Elarab, Fatiha; Puig Cayuela, Vicenç Abstract: In this paper, a Fault Tolerant Control (FTC) strategy using a virtual actuator for Linear Parameter Varying (LPV) systems is proposed. The main idea of this FTC method is to reconfigure the control loop such that the nominal controller could still be used without need of retuning it. The plant with the faulty actuator is modified adding the virtual actuator block that masks the fault. The suggested strategy is an active FTC strategy that reconfigures the virtual actuator on-line taking into account faults and operating point changes. In order to implement the virtual actuator approach, a fault estimation is required. In this work, the fault estimation is formulated as a parameter estimation problem. The LPV virtual actuator is designed using polytopic LPV techniques and Linear Matrix Inequalities (LMIs). To assess the performance of the proposed approach an aeronautical application is used. Fri, 22 Feb 2013 14:00:37 GMT http://hdl.handle.net/2117/17945 2013-02-22T14:00:37Z Rotondo, Damiano; Nejjari Akhi-Elarab, Fatiha; Puig Cayuela, Vicenç no In this paper, a Fault Tolerant Control (FTC) strategy using a virtual actuator for Linear Parameter Varying (LPV) systems is proposed. The main idea of this FTC method is to reconfigure the control loop such that the nominal controller could still be used without need of retuning it. The plant with the faulty actuator is modified adding the virtual actuator block that masks the fault. The suggested strategy is an active FTC strategy that reconfigures the virtual actuator on-line taking into account faults and operating point changes. In order to implement the virtual actuator approach, a fault estimation is required. In this work, the fault estimation is formulated as a parameter estimation problem. The LPV virtual actuator is designed using polytopic LPV techniques and Linear Matrix Inequalities (LMIs). To assess the performance of the proposed approach an aeronautical application is used. http://hdl.handle.net/2117/17942 Title: Fault detection and isolation of hybrid systems using diagnosers that reason on components Authors: Vento Maldonado, Jorge; Puig Cayuela, Vicenç; Sarrate Estruch, Ramon; Travé-Massuyés, Louise Abstract: In this paper, a design methodology and implementation architecure for diagnosers in the framework of hybrid systems is proposed. The design methodo logy is based on the hybrid automata model that represents the system behavior, in which each mode relates to a set of components. The architecture includes a set of modules which achieve mode recognitionand diagnosis tasks both based on residuals generated by structural analysis. The two tasks interact each other since the diagnosis module adapts accordingly to the current mode of the hybrid system. Mode recognition involves detecting mode change and identifying the set of residuals that are consistent with the current mode of the hybrid system. On the other hand, diagnosis involves detecting and isolating the fault by interlinking the components underlying the inconsistencies reported by the residuals, along the DX approach. As an application case study to illustrate the proposed fault diagnosis approach for hybrid systems a partial model of the Barcelona sewer network is used. Fri, 22 Feb 2013 13:34:14 GMT http://hdl.handle.net/2117/17942 2013-02-22T13:34:14Z Vento Maldonado, Jorge; Puig Cayuela, Vicenç; Sarrate Estruch, Ramon; Travé-Massuyés, Louise no In this paper, a design methodology and implementation architecure for diagnosers in the framework of hybrid systems is proposed. The design methodo logy is based on the hybrid automata model that represents the system behavior, in which each mode relates to a set of components. The architecture includes a set of modules which achieve mode recognitionand diagnosis tasks both based on residuals generated by structural analysis. The two tasks interact each other since the diagnosis module adapts accordingly to the current mode of the hybrid system. Mode recognition involves detecting mode change and identifying the set of residuals that are consistent with the current mode of the hybrid system. On the other hand, diagnosis involves detecting and isolating the fault by interlinking the components underlying the inconsistencies reported by the residuals, along the DX approach. As an application case study to illustrate the proposed fault diagnosis approach for hybrid systems a partial model of the Barcelona sewer network is used. http://hdl.handle.net/2117/17938 Title: Set-membership parity space approach for fault detection in linear uncertain dynamic systems Authors: Blesa Izquierdo, Joaquim; Puig Cayuela, Vicenç; Saludes Closa, Jordi Abstract: In this paper, a set-membership parity space approach for linear uncertain dynamic systems is proposed. First, a set of parity relations derived from the parity space approach are obtained. Then, the parameter identification of those relations is performed using a set-membership algorithm. In particular, in this work the parametric uncertainty is bounded by a zonotope. Fault detection is then based on checking, at every sampling time, the non existence of a parameter value in the parameter uncertainty set such that the model is consistent with all the system measurements. The proposed approach will be applied to a case study based in a two-tank system in order to evaluate its effectiveness. Fri, 22 Feb 2013 12:55:53 GMT http://hdl.handle.net/2117/17938 2013-02-22T12:55:53Z Blesa Izquierdo, Joaquim; Puig Cayuela, Vicenç; Saludes Closa, Jordi no In this paper, a set-membership parity space approach for linear uncertain dynamic systems is proposed. First, a set of parity relations derived from the parity space approach are obtained. Then, the parameter identification of those relations is performed using a set-membership algorithm. In particular, in this work the parametric uncertainty is bounded by a zonotope. Fault detection is then based on checking, at every sampling time, the non existence of a parameter value in the parameter uncertainty set such that the model is consistent with all the system measurements. The proposed approach will be applied to a case study based in a two-tank system in order to evaluate its effectiveness. http://hdl.handle.net/2117/17904 Title: Multiplatform virtual laboratory for engineering education Authors: Villar Zafra, Aitor; Zarza Sánchez, Sergio; Lázaro Villa, José Antonio; Fernández Canti, Rosa M. Abstract: An autonomous and multi platform virtual laboratory for educational purposes is presented. The implemented platform includes a server with a SSH(Secure SHeil)connection and a separated repository containing the virtual experiments. The programming of the experiments is implemented in Java language based tool, the EJS (Easy Java Simulation) and uses an external computation engine, for example MatIab. The virtual laboratory provides control system experimentsat University level. Two application examples are described, namely, a magnetic levitator and an inverted pendulum­ cart system. The virtual laboratory has been successfully used for education and training of Electronics Engineering students. A discussion of the results of this e-learning experience is also presented. Thu, 21 Feb 2013 10:24:01 GMT http://hdl.handle.net/2117/17904 2013-02-21T10:24:01Z Villar Zafra, Aitor; Zarza Sánchez, Sergio; Lázaro Villa, José Antonio; Fernández Canti, Rosa M. no An autonomous and multi platform virtual laboratory for educational purposes is presented. The implemented platform includes a server with a SSH(Secure SHeil)connection and a separated repository containing the virtual experiments. The programming of the experiments is implemented in Java language based tool, the EJS (Easy Java Simulation) and uses an external computation engine, for example MatIab. The virtual laboratory provides control system experimentsat University level. Two application examples are described, namely, a magnetic levitator and an inverted pendulum­ cart system. The virtual laboratory has been successfully used for education and training of Electronics Engineering students. A discussion of the results of this e-learning experience is also presented. http://hdl.handle.net/2117/17708 Title: A service reliability model predictive control with dynamic safety stocks and actuators health monitoring for drinking water networks Authors: Grosso Pérez, Juan Manuel; Ocampo-Martínez, Carlos; Puig Cayuela, Vicenç Abstract: This paper presents a model predictive control strategy to assure reliability in drinking water networks given a customer service level and a forecasting demand. The underlying idea concerns a two-layer hierarchical control structure. The upper layer performs a local steady-state optimization to set up an inventory replenishment policy based on dynamic safety stocks for each tank in the network. At the same stage, actuators health is revised to set up their next maximum allowable degradation in order to efficiently distribute overall control effort and guarantee system availability. In the lower layer, a model predictive control algorithm is implemented to compute optimal control set-points to minimize a multiobjective cost function. Simulation results in the Barcelona drinking water network have shown the effectiveness of the dynamic safety stocks allocation and the actuators health monitoring to assure service reliability and optimizing network operational costs. Wed, 13 Feb 2013 11:07:35 GMT http://hdl.handle.net/2117/17708 2013-02-13T11:07:35Z Grosso Pérez, Juan Manuel; Ocampo-Martínez, Carlos; Puig Cayuela, Vicenç no control theory linear programming optimisation predictive control Author keywords: dynamic safety stocks, Actuators health monitoring, Model predictive control This paper presents a model predictive control strategy to assure reliability in drinking water networks given a customer service level and a forecasting demand. The underlying idea concerns a two-layer hierarchical control structure. The upper layer performs a local steady-state optimization to set up an inventory replenishment policy based on dynamic safety stocks for each tank in the network. At the same stage, actuators health is revised to set up their next maximum allowable degradation in order to efficiently distribute overall control effort and guarantee system availability. In the lower layer, a model predictive control algorithm is implemented to compute optimal control set-points to minimize a multiobjective cost function. Simulation results in the Barcelona drinking water network have shown the effectiveness of the dynamic safety stocks allocation and the actuators health monitoring to assure service reliability and optimizing network operational costs. http://hdl.handle.net/2117/17668 Title: Methodology to detect and isolate water losses in water hydraulic networks: Application to Barcelona water network Authors: Quevedo Casín, Joseba Jokin; Pérez Magrané, Ramon; Pascual, Josep; Puig Cayuela, Vicenç; Cembrano Gennari, Gabriela; Peralta, Antoni Abstract: This paper presents a methodology to detect real water losses in urban water networks. The proposed method is based on the combined use of the DMA economical water balance confidence intervals for a period of time longer than a month (a trimester, semester or year) with appropriated night flow ratio intervals. This method allows distinguishing between real and apparent water losses in a given DMA. The use of uncertainty models allows taking into account measurement uncertainty of the flow data sensors. The use of different periods of time allows the stabilization of the results. This methodology is illustrated in a DMA of the Barcelona water distribution network using real data. Tue, 12 Feb 2013 15:40:30 GMT http://hdl.handle.net/2117/17668 2013-02-12T15:40:30Z Quevedo Casín, Joseba Jokin; Pérez Magrané, Ramon; Pascual, Josep; Puig Cayuela, Vicenç; Cembrano Gennari, Gabriela; Peralta, Antoni no Control system analysis computing This paper presents a methodology to detect real water losses in urban water networks. The proposed method is based on the combined use of the DMA economical water balance confidence intervals for a period of time longer than a month (a trimester, semester or year) with appropriated night flow ratio intervals. This method allows distinguishing between real and apparent water losses in a given DMA. The use of uncertainty models allows taking into account measurement uncertainty of the flow data sensors. The use of different periods of time allows the stabilization of the results. This methodology is illustrated in a DMA of the Barcelona water distribution network using real data. http://hdl.handle.net/2117/17560 Title: Adaptive multilevel neuro-fuzzy model predictive control for drinking water networks Authors: Grosso, J.M.; Ocampo-Martínez, Carlos; Puig Cayuela, Vicenç Abstract: This paper presents a constrained Model Predictive Control (MPC) strategy enriched with soft-control techniques as neural networks and fuzzy logic, to incorporate self-tuning capabilities and reliability aspects for drinking water transport network management. The system is a multilevel controller with three hierarchical layers: neural level, fuzzy level, MPC level. Results in the Barcelona Water Network have shown that the quasi-explicit nature of the proposed predictive controller leads to improve the computational time, especially when the complexity of the problem structure can vary while tuning the receding horizons. Fri, 01 Feb 2013 14:07:35 GMT http://hdl.handle.net/2117/17560 2013-02-01T14:07:35Z Grosso, J.M.; Ocampo-Martínez, Carlos; Puig Cayuela, Vicenç no This paper presents a constrained Model Predictive Control (MPC) strategy enriched with soft-control techniques as neural networks and fuzzy logic, to incorporate self-tuning capabilities and reliability aspects for drinking water transport network management. The system is a multilevel controller with three hierarchical layers: neural level, fuzzy level, MPC level. Results in the Barcelona Water Network have shown that the quasi-explicit nature of the proposed predictive controller leads to improve the computational time, especially when the complexity of the problem structure can vary while tuning the receding horizons.