Hierarchical Control and Optimization Strategies Applied to Solar Membrane Distillation Facilities
- Autores: Juan Diego Gil Vergel
- Directores de la Tesis: Manuel Berenguel Soria (dir. tes.), Lidia Roca (codir. tes.)
- Lectura: En la Universidad de Almería ( España ) en 2020
- Idioma: inglés
- Número de páginas: 193
- Títulos paralelos:
- Estrategias de Control Jerárquico y Optimización Aplicadas a Plantas de Destilación por Membranas Alimentadas con Energía Solar
- Tribunal Calificador de la Tesis: Diego César Alarcón Padilla (presid.), José Luis Guzmán Sánchez (secret.), Antonio Visioli (voc.)
- Programa de doctorado: Programa de Doctorado en Informática por la Universidad de Almería
- Materias:
- Enlaces
- Tesis en acceso abierto en: riUAL
- Resumen
- español
La destilación por membranas es un proceso de desalación alternativo impulsado térmicamente a media-baja temperatura (generalmente inferior a 85 oC). Esta característica le permite ser fácilmente alimentada mediante energía solar o fuentes térmicas de baja entalpía, como calor residual, reduciendo así la huella de carbón de los procesos convencionales. Además, esta tecnología permite su uso descentralizado a pequeña escala y la realización de operaciones intermitentes, lo que la convierte en una solución adecuada para ser implantada en lugares remotos con buenas condiciones de irradiancia solar. De este modo, la destilación por membranas se erige como una de las tecnologías más convenientes, eficientes y sostenibles para expandir la gama de posibilidades de los procesos de desalación convencionales. Sin embargo, esta tecnología no se encuentra todavía implementada industrialmente y, en la actualidad, está en una etapa pre-comercial, siendo su gran consumo energético por unidad de destilado producida uno de los mayores impedimentos para su completa comercialización. Hasta la fecha, la mayoría de los trabajos de investigación en el campo de la tecnología de destilación por membranas se han centrado en la mejora del diseño interno del módulo y en la prueba de diferentes membranas, con el objetivo de minimizar el consumo térmico y aumentar la producción de destilado. Estos trabajos han provocado un gran avance en términos de consumo térmico, desencadenando que la tecnología entre en una fase de madurez avanzada, lo que se ha confirmado con la aparición de los primeros módulos a escala comercial y la instalación de las primeras plantas a escala piloto. Por lo tanto, la tecnología de destilación por membranas ha entrado en una nueva fase de investigación, en la cual, los trabajos deben estar dirigidos al desarrollo de estrategias de operación que optimicen el funcionamiento de los módulos entiempo real, especialmente cuando estos se alimentan mediante una fuente de energía con una naturaleza tan intermitente e impredecible como es el caso de la energía solar.
Esta tesis tiene como objetivo el desarrollo de estrategias de operación, mediante el uso de técnicas de modelado, control y optimización, para plantas de destilación por membranas. Los métodos de operación propuestos están centrados principalmente en la reducción del consumo térmico de los módulos y la maximización de su producción de destilado, dos de los puntos débiles de la tecnología, así como la minimización de los costes operativos tratando de reducir el precio del agua desalada. El desarrollo de la tesis se ha dividido en cuatro fases, siguiendo las etapas clásicas de un proyecto de ingeniería de control. En primer lugar, se realizó una etapa de modelado y diseño de controladores de bajo nivel para plantas de destilación por membranas alimentadas con energía solar. En los resultados obtenidos en esta etapa se demostró cómo, el tiempo para establecer una temperatura adecuada de operación a la entrada del módulo, se puede reducir entorno al 50 % en comparación con operaciones manuales llevadas a cabo por operadores cualificados. La segunda fase de desarrollo de la tesis se centró en el diseño de estrategias de control jerárquicas. Estos controladores tienen objetivos de alto nivel como la maximización de la eficiencia térmica, la producción de destilado, o la reducción de los costes operativos, los cuales se consiguen actuando sobre las referencias de los bucles de control de bajo nivel desarrollados en la etapa anterior. Los resultados de las pruebas realizadas en esta etapa evidenciaron cómo la producción de destilado se puede aumentar entorno a un 5-7 %, los costes operativos se pueden reducir en un 9-10 %, y el tiempo empleado en la fase de arranque se puede reducir también en un 11 % respecto a operaciones manuales. La tercera fase de desarrollo se focalizó en el diseño de estrategias de control para plantas de destilación por membranas a escala industrial. En estos casos, el principal objetivo consiste en gestionar eficientemente los múltiples módulos de destilación por membranas que componen la unidad de desalación, tratando de reducir su consumo de energía térmica al mismo tiempo que se aseguran la demanda de agua de un agente consumidor. Las pruebas realizadas con los algoritmos desarrollados en este a etapa mostraron cómo la cantidad de energía térmica consumida por la unidad de desalación se puede reducir significativamente, ahorrando hasta un 65 % de la energía requerida en una operación manual en días soleados. La última fase de desarrollo de la tesis consistió en la elaboración de un tutorial de técnicas de modelado y control aplicadas en la tecnología de destilación por membranas. Este tutorial recoge las principales técnicas utilizadas hasta el momento y describe el desarrollo tecnológico que se consigue con su aplicación. Además, propone ideas y recomendaciones para el desarrollo de trabajos futuros en este nuevo campo de investigación en la tecnología de destilación por membranas.
- English
Population growth coupled with industrial and agricultural activities is placing freshwater reserves under severe stress. Also, the effects of climate change are altering the natural water cycle through extreme meteorological phenomena such as droughts and floods, further aggravating the situation and causing aquifers worldwide to become depleted or polluted. Considering this panorama, humanity faces one of the greatest challenges in its history, which must be addressed urgently by modifying the current model of water use and supply.
Even though many countries around the world are developing policies aimed at achieving rational use of this resource, in most cases these are not enough and new sources of freshwater are needed to meet demands. In consequence, desalination has emerged as one of the most promising solutions for extending natural water resources, promoting many countries opt for its use. However, conventional desalination technologies present two main problems. On the one hand, although desalination processes have undergone major development in the last decades, these are still energy inefficient, so if they are powered with conventional energy sources, the water problem can turn into an energy problem. On the other hand, these processes cannot easily be reduced to a medium or small scale. In addition, they require the use of specific infrastructures and on-grid power connections to avoid discontinued operations. This fact prevents its implementation in rural or isolated areas, such as islands with low per capita consumption, where a large part of the population currently suffering from water shortages is concentrated.
Membrane distillation is an alternative thermally-driven desalination process powered at medium-low temperature (generally below 85 oC). This feature allows this technique to be easily powered by solar energy or low enthalpy heat sources such as waste heat, thus reducing the carbon footprint of conventional processes. Besides, this technology allows its decentralized use on a small scale and the intermittent operation, which makes it a suitable solution to be implanted in remote places with good conditions of solar irradiance. In this way, membrane distillation stands out as one of the most convenient, efficient and sustainable technologies to expand the range possibilities of conventional desalination processes. However, this technology is not yet industrially implemented and it is currently in a pre-commercial stage, being its high energy consumption per unit of distillate produced one of the major impediments to its complete commercialization.
So far the majority of research works in the field of membrane distillation technology have focused on improving the internal design of the module and on testing different membranes, intending to minimize thermal consumption and increase distillate production. These works have caused a great advance in terms of thermal consumption, allowing the technology to enter in an advanced stage of maturity, which was confirmed with the appearance of the first modules on a commercial scale and the installation of the first plants at pilot scale. Therefore, membrane distillation technology has entered in a new research phase, in which the efforts must be devoted to the development of operating strategies that optimize the operation of the modules in real-time, especially when these are powered by an energy source with such as intermittent and unpredictable nature as solar energy.
This thesis aims to develop operating strategies, through the use of modelling, control and optimization techniques for membrane distillation plants. The proposed operating methods are mainly focused on reducing the thermal consumption of the modules and on maximizing their distillate production, two of the main weak points of the technology, as well as minimizing operating costs trying to reduce the price of desalinated water. The development of the thesis has been divided into four phases, following the classic steps of a control engineering project. First, a stage of modelling and design of low-level controllers for membrane distillation plants powered by solar energy was performed. For this, dynamic models based on first principles already proposed in the literature were used to characterize the heat generation circuit, and a model based on experimental data for a commercial-scale membrane distillation module was developed.
These models served as the basis for the design of a complete low-level control architecture consisting of four loops governed by a reference generator, which allows a stable temperature to be maintained at the inlet of the membrane distillation module despite disturbances in solar irradiance. This strategy was experimentally tested in a pilot plant located at the Plataforma Solar of Almería validating the results obtained in simulation and evidencing how, the time for establishing an adequate operating temperature at the module inlet, can be reduced by 50 % compared to manual operations performed by qualify operators.
The second phase of the development of the thesis focused on the design of hierarchical control strategies. These controllers were tasked with high-level control objectives such as maximizing the thermal efficiency, distillate production, or reducing operating costs, which are achieved by acting on the references of the low-level control loops developed in the previous stage. In particular, two control architectures were developed, one for the phase of operation, which tries to optimize in real-time the objectives mentioned above, and another for the plant start-up procedure, which tries to reduce the time spent on said procedure. As in the previous stage, both controllers were experimentally tested, demonstrating, for example, how distillate production can be increased by around 5-7 %, operating costs can be reduced by 9-10 %, and the time spent in the start-up phase can be reduced by 11 % compared to manual operations.
The third development phase focused on the design of control strategies for membranes distillation plants at the industrial scale, where the control paradigm differs from previous approaches due to the presence of multiple membrane distillation modules and a water-consuming agent. The proposed control strategies aim to reduce the thermal consumption of the desalination unit while ensuring the variable water demand of the consuming agent, objectives that require contrary operating conditions in the fluid flow rate of the modules. The main challenge from a control point of view in these types of applications lies in the time spent to calculate optimal control actions, which grows exponentially as the number of distillation modules in the desalination unit increases. Thus, the first control architecture proposed was based on a distributed model predictive control technique. In this case, the controller was responsible for managing the feed flow rate of the different modules to achieve the abovementioned goals. The tests carried out showed how the specific thermal consumption of the desalination unit can be reduced by 5 % on average compared to manual operations, while the computing time was significantly reduced compared to centralized controllers. Additionally, a control algorithm was also proposed to manage, apart from the feed flow rate, the number of modules turned on at each sampling time according to the water demand of the consuming agent, turning the control problem into a nonlinear mixed integer problem. The proposed algorithm for its resolution was based on the generalized Benders decomposition. The tests showed how the computing time can be considerably reduced, for example, for a case with 64 modules, the time spent by a solver for the nonlinear mixed integer problem was around 1600 s, while that of the proposed algorithm was 5 s without significantly affecting the solution. Besides, the amount of thermal energy consumed by the desalination unit was also significantly reduced, saving up to 65 % of the energy required by manual operation on a sunny day.
The last development phase of the thesis consisted in the elaboration of a tutorial of modelling and control techniques applied to the membrane distillation technology. In this tutorial, all the techniques applied during the development of the thesis as well as other proposals in the literature are summarized, describing the technological development that can be achieved with its application.
To conclude, this summary ends by depicting the structure of this document, which has been divided into four parts according to those described in the University of Almería regulation for Ph.D. theses presented in the compendium modality:
• Chapter 1 describes the framework of the thesis and introduces the main methodologies used. In addition, this chapter describes the development structure of the thesis and indicates the publications dealing with each of the topics covered.
• Chapter 2 presents the scientific publications that support the work done.
• Chapter 3 summarizes the conclusions derived from the different publications as well as the recommendations for future work.
• Finally, Chapter 4 contains a list of other scientific contributions that are directly derived from the Ph.D. thesis.
- español
