Exploitation of waste heat could achieve economic and environmental benefits, while at the same time increase energy efficiency in process sites. Diverse commercialised technologies exist to recover useful energy from waste heat. In addition, there are multiple on-site and offsite end-uses of recovered energy. The challenge is to find the optimal mix of technologies and end-uses of recovered energy taking into account the quantity and quality of waste heat sources, interactions with interconnected systems and constraints on capital investment. Explicit models for waste heat recovery technologies that are easily embedded within appropriate process synthesis frameworks are proposed in this work. A novel screening tool is also proposed to guide selection of technology options. The screening tool considers the deviation of the actual performance from the ideal performance of technologies, where the actual performance takes into account irreversibilities due to finite temperature heat transfer. Results from applying the screening tool show that better temperature matching between heat sources and technologies reduces the energy quality degradation during the conversion process. A ranking criterion is also proposed to evaluate end-uses of recovered energy. Applying the ranking criterion shows the use to which energy recovered from waste heat is put determines the economics and potential to reduce CO2 emissions when waste heat recovery is integrated in process sites. This thesis also proposes a novel methodological framework based on graphical and optimization techniques to integrate waste heat recovery into existing process sites. The graphical techniques are shown to provide useful insights into the features of a good solution and assess the potential in industrial waste heat prior to detailed design. The optimization model allows systematic selection and combination of waste heat source streams, selection of technology options, technology working fluids, and exploitation of interactions with interconnected systems. The optimization problem is formulated as a Mixed Integer Linear Program, solved using the branch-and-bound algorithm. The objective is to maximize the economic potential considering capital investment, maintenance costs and operating costs of the selected waste heat recovery technologies. The methodology is applied to industrial case studies. Results indicate that combining waste heat recovery options yield additional increases in efficiency, reductions in CO2 emissions and costs. The case study also demonstrates that significant benefits from waste heat utilization can be achieved when interactions with interconnected systems are considered simultaneously. The thesis shows that the methodology has potential to identify, screen, select and combine waste heat recovery options for process sites. Results suggest that recovery of waste heat can improve the energy security of process sites and global energy security through the conservation of fuel and reduction in CO2 emissions and costs. The methodological framework can inform integration of waste heat recovery in the process industries and formulation of public policies on industrial waste heat utilization.

Ce travail contribue aux efforts de l'Union Européenne pour réduire les émissions de CO2. Son objectif est d'aider les industries produisant de la chaleur fatale à récupérer cette énergie perdue, d'augmenter sa température et de la réutiliser in situ. Les transformateurs de chaleur (Absorption Heat Transformers ou AHT), machines à absorption consommant très peu d'électricité, sont alors ici étudiés. Les AHTs existants rencontrent des problèmes comme la corrosion, la cristallisation, la toxicité et les niveaux de pression éloignés de la pression atmosphérique. Ceux-ci sont causés par les fluides conventionnels (Eau/LiBr et Ammoniaque/Eau) et s'aggravent à des températures supérieures à 120°C. Des modèles de conception ainsi que des solutions techniques, applicables avec tous mélanges de fluides organiques, sont alors proposés dans cette thèse. Ces modèles sont validés avec des données de la littérature et implémentés dans des outils d'aide à la décision.Tout d'abord, un modèle de sélection de paires de fluides organiques (parmi une liste de fluides) est développé. Les contraintes prises en compte sont, entre autres, les types et les profils de températures des sources et puits de chaleur, et les propriétés du fluide. Pour chaque type de fluide, la méthode la plus adaptée au calcul des propriétés physiques des fluides est choisie.En second lieu, pour effectuer la séparation des 2 constituants du mélange de fluides organiques, le générateur (composant recevant la chaleur fatale) et le condenseur de l'AHT sont fusionnés pour former une colonne de distillation. Un modèle d'une colonne de distillation nommée « hybride » est alors développé en adaptant la méthode de Ponchon-Savarit et en la combinant avec la méthode ETD (Equal Thermodynamic Distance). Cette colonne associe les avantages des 2 types de colonnes adiabatiques et diabatiques. Elle allie réduction de production d'entropie et meilleure exploitation des sources de chaleur à températures glissantes. La conception mécanique de la colonne hybride est aussi incluse.Troisièmement, pour atteindre la température théorique maximale du mélange de fluide déjà choisi, l'absorbeur de l'AHT (où la chaleur à haute température est libérée) est divisé en sections adiabatiques suivies par des sections diabatiques. De plus, les modèles détaillés des colonnes à bulles (fonctionnant en co-courant ou en contre-courant) ainsi que de la colonne à garnissage sont présentés et comparés entre eux.Les principaux résultats de ces travaux consistent en une nouvelle méthodologie de choix de fluides organiques pouvant remplacer les mélanges classiques surtout à températures élevées (supérieures à 130 °C). En ce qui concerne la colonne de distillation, il est montré que la colonne adiabatique constitue un meilleur choix lorsqu'une source de chaleur latente est disponible tandis qu'avec une source de chaleur sensible, la colonne hybride engendre moins de pertes exergétiques. En passant à l'absorbeur, le nouveau mode d'opération de celui-ci permet à l'utilisateur d'atteindre des températures plus élevées que celles réalisées avec les technologies actuellement disponibles. Enfin, les modèles développés permettent de choisir les technologies de distillation (adiabatique, diabatique ou hybride) et d'absorption (colonne à bulles ou à garnissage) les plus appropriées en s'adaptant à différentes problématiques industrielles. / This work is part of the European union efforts to reduce its CO2 emissions. It aims to assist any waste heat producing industry in recuperating this lost thermal energy, pumping it to higher temperature levels and reusing it on site. Absorption Heat Transformers (AHT), that consume little electricity, are used for this task. Current AHT problems such as corrosion, crystallization, toxicity and inconvenient pressure levels are caused by conventionally used H2O/LiBr and NH3/ H2O working fluids and get worse at temperatures exceeding 120°C. Potential solutions are thus suggested. According to them, models are developed; they are all able to operate with any organic mixture and are customized to accompany the industrialist from start to finish. These solutions were validated by comparing them with literature data and are implemented into several tools.Firstly, a model selects the optimal organic binary mixture -among a list of fluids- in terms of the real case application's constraints: Heat transfer fluids used, Heat source's and heat sink's types and temperature profiles, mixtures transport properties among other parameters. Suitable thermodynamic model is selected for different fluid group types.Secondly, in order to separate the 2 components of the chosen mixture of organic compounds, the AHT generator (component which receives waste heat) is merged with the AHT condenser thus forming a distillation column. A “hybrid column” is designed by modifying the Ponchon-Savarit method and combining it with the Equal Thermodynamic Distance (ETD) method. This new column associates the best features of the two columns. It reduces entropy production rates and best exploits temperature gliding heat sources. Mechanical design for the hybrid column is also included.Thirdly, to ensure that the maximum theoretical temperature of the working fluid is reached, the AHT absorber (where high temperature heat is released) is divided into consecutive adiabatic parts followed by diabatic ones. Detailed Models for co-current and counter-current bubble columns as well as packing columns are presented and compared.Main results consist in a selection methodology of organic compounds mixtures, capable of replacing conventional ones specially at temperatures higher than 130 °C. It's also shown that adiabatic columns are better options when latent type heat sources are available while hybrid columns lose less exergy when used with sensible heat sources. As for the absorber, the new operating mode provides the user with higher temperatures than currently reached by available technologies. Finally, using the developed models, tailored and most suitable distillation (adiabatic, diabatic or hybrid columns) and absorber (bubble or packing columns) technologies can be proposed depending on the industrial specific cases and requirements.