Multiperiod modelling planning and productivity and energy efficient assessment of an industrial gases facility
- Autores: David Fernández Linares
- Directores de la Tesis: Gonzalo Guillén Gosálbez (dir. tes.), Carlos Pozo Fernández (codir. tes.), Laureano Jiménez Esteller (codir. tes.)
- Lectura: En la Universitat Rovira i Virgili ( España ) en 2018
- Idioma: español
- Tribunal Calificador de la Tesis: Rubén Ruiz Femenia (presid.), Dieter Boer (secret.), Christoph Erdmann (voc.)
- Programa de doctorado: Programa de Doctorado en Nanociencia, Materiales e Ingeniería Química por la Universidad Rovira i Virgili
- Materias:
- Enlaces
- Tesis en acceso abierto en: TDX hdl.handle.net
- Resumen
The world is currently faced with challenges in all three dimensions of sustainable development: economic, social and environmental. Millions of people are still living in extreme poverty, income inequality among countries is steadily growing and unsustainable consumption and production patterns have resulted in high economic and social costs and may endanger humans welfare. The efforts to address sustainability will be crucial and would benefit both the society and the environment by identifying proactive pathways towards sustainability. In this framework, research based information facing such problems may play a major role in decision and policy-making support to accelerate the effective shift toward a sustainable development.
Industrial energy use accounts for one-third of global electrical usage and is a substantial contributor to CO2 emissions causing global warming. The possibilities for improving the energy efficiency of industrial facilities are notorious, even in mature industries and technologies. Industrial sites are very large individual users of energy which may change production volumes or schedules many times during the useful life of the factory. Furthermore, industries are currently striving to improve the efficiency of their processes in order to reduce the energy input per unit of economic output.
This thesis points towards one of the main structural transformation needed to reconnect the human development to sustained progress, the energy transformation, which supports the shift towards an environmentally friendly economy to prevent exceeding the limits of the Earth. Specifically, this thesis is focused in the cryogenic air separation technology, which is the most efficient technology for mass-production of air products such as oxygen, nitrogen and argon. A high number of industries such as steel, petrochemical, metallurgy, medical or food demand large amounts of these air products. Inherent to its operation, cryogenic air separation plants are in general energy intensive, with the power input being the main factor on which the ultimate production cost will depend.
This thesis aims to give decision and policy makers methods and tools to improve the energy efficiency in energy-intensive consumers, which would not only bring down the electrical cost per amount of product obtained, but also reduce the overall environmental impacts derived from the lower usage of energy resources. Despite there is a large number of approaches to solve the emerging problems, mathematical optimization/programming appears as an effective tool to find the best solution to them. For this reason, it has been widely used to aid decision-making in many scientific or engineering problems, since mathematical programming allows solving real problems by building a model based on equations, which are later solved with the proper solver alternatives. In this context, the mathematical models stated in this thesis represent real facilities and thus their formulation allows to optimize existing industrial activities and provide solutions for plant managers and/or decision makers.
In the cryogenic air separation field, the complexity of optimally managing this kind of facilities is very high due to the interactions between process variables (flow rates, levels, quality requirements, etc.), product prices, fluctuations in customers’ product demand, utilities price, electricity varying prices, etc. In order to deal with this complexity, a tailored multiperiod model is developed, taking the form of a mixed-integer linear programming (MILP) problem, and is applied to the Messer plant located in Tarragona. This model allows to determine the optimal production schedule of an industrial cryogenic air separation process so as to maximize the net profit by minimizing energy consumption. In the context of cryogenic air separation, some tools were presented to determine the optimal operating schedule depending on the power costs, as well as depending on demand, contractual obligations and variable electricity pricing, but these previous works shown some limitations. The work of this thesis, extends previous proposals increasing the granularity in the modeling of the electricity price pattern in order to account for hourly variations considering electricity markets peculiarities, and including real demand levels when optimizing the production schedule. Furthermore, the network boundaries of the air separation process are amplified by including the complete system: production, compression, liquefaction, storage and delivery. The model formulation is complemented with the possibility to purchase a certain amount of product from an external supplier by means of economical and power pricing contracts, thereby offering the possibility to achieve significant reductions in operational costs if properly managed. This tool assists engineers in their daily activities by effectively optimizing production planning, energy rules, sales and product stocks, while considering external constraints and dynamic market conditions Furthermore, additional tools are applied to identify further inefficiency sources arising from the design of the air separation units instead of the operation. To this end, two standard mathematical programming methods (Data Envelopment Analysis and Malmquist Productivity Index) are applied to compare the relative performance of a set of Air Separation Units (ASUs) according to energy efficiency and productivity criteria. The dataset considered in this case study includes some Messer plants worldwide. As demonstrated, this tool provides insight on how to improve the efficiency of inefficient facilities by identifying inefficiency sources (and establishing quantitative targets for them to become efficient) and reference facilities that could be used for benchmarking.
The tools developed in this thesis can be extended to other plants that the company has in the rest of the world and apply them as good practices to improve the management and efficiency of their processes. These methods can also be applied in a wide range of energy intensive industrial processes (chemical, automotive, metallurgy, etc.) to minimize energy losses in their process moving towards a more sustainable world. Particularly, the method developed to optimize the scheduling in an industrial air separation unit constitutes a promising alternative for any other energy-intensive industrial process where energy savings play an important role.
