Nowadays, reduced demand led European process industry to decrease its productivity. The globalised market increases the competition on the acquisition of raw materials as well as on the product side, while more stringent regulations pressurise towards the reduction of environmental impact. Due to the traditional approach to industrial plant design, process industries are not optimised to face periods of reduced productivity and variable raw materials. An improved flexibility of the production cycles while respecting environmental constraints is the ambitious goal aimed but not yet achieved by process industry. The improvement in prognostic monitoring and control of complex industrial processes and the study of different scenarios can support the industry in achieving such target. This PhD thesis, which is focused on Modelling & Simulation of complex industrial processes, is inline with this objective, as in-depth investigations of common, uncommon or future scenarios allow evaluating the feasibility of process integration solutions and multi-objective optimizations to improve economic and environmental process sustainability. In addition, starting from the fact that man-machine interaction is essential to interface manpower and process, as man is the “brain”, the knowledge and the experience while the machine is the “brawn” (“If the man will be able to use it with a creative spirit, the machine will be the servant and the liberator of mankind”, Frank Lloyd Wright), the present PhD thesis is focused on an extension of the above concept toward a more efficient "man-process" interaction. A more agile, dynamic and flexible supervision is achieved on all the aspects of the production processes as well as on the interactions among different sub-processes to the final aim of maximizing resource efficiency and improve process sustainability. Noticeably, the concept of "resource" includes not only primary raw materials and energy but also by-products, wastes, wastewater and off-gases. In this context and considering that some industrial processes and related parameters can be difficult to directly monitor through sensors (due e.g. to harsh conditions), the combination of standard techniques with advanced tools, such as specialised modelling software (e.g. Aspen Plus®) and Artificial Intelligence techniques (e.g. neural networks, genetic algorithm) is here exploited in order to improve the man-process connection. So doing, trials and plant changes or revamping can be addressed only to the most promising solutions. The previous described approach is here applied to address real industrial cases studies related to the improvement of the sustainability of steelmaking industry. The investigated issues ranges from the development and validation through tests of a general-purpose approach for improving the use of wastewater and by-products to a simulation module that is a part of a Decision Support Tool evaluating the impact of common and uncommon scenarios in the form of easy-to-understand Key Performance Indicators. Moreover, the extension of real experimentations is faced through process simulation, in order to improve process knowledge, to find the best operating conditions and configurations of treatment processes and to maximize the reuse of valuable by-products. Finally advanced models forecasting gas production or use are presented, which are developed to the aim of optimizing their exploitation in gas-networks. Despite the specific application in steelmaking industry, the multidisciplinary approach, which is presented in this thesis, is compatible and transferable to other industrial context within process industry.
Modelling and simulation of industrial operations for prognostic monitoring and control, process integration and optimization.
2018
Abstract
Nowadays, reduced demand led European process industry to decrease its productivity. The globalised market increases the competition on the acquisition of raw materials as well as on the product side, while more stringent regulations pressurise towards the reduction of environmental impact. Due to the traditional approach to industrial plant design, process industries are not optimised to face periods of reduced productivity and variable raw materials. An improved flexibility of the production cycles while respecting environmental constraints is the ambitious goal aimed but not yet achieved by process industry. The improvement in prognostic monitoring and control of complex industrial processes and the study of different scenarios can support the industry in achieving such target. This PhD thesis, which is focused on Modelling & Simulation of complex industrial processes, is inline with this objective, as in-depth investigations of common, uncommon or future scenarios allow evaluating the feasibility of process integration solutions and multi-objective optimizations to improve economic and environmental process sustainability. In addition, starting from the fact that man-machine interaction is essential to interface manpower and process, as man is the “brain”, the knowledge and the experience while the machine is the “brawn” (“If the man will be able to use it with a creative spirit, the machine will be the servant and the liberator of mankind”, Frank Lloyd Wright), the present PhD thesis is focused on an extension of the above concept toward a more efficient "man-process" interaction. A more agile, dynamic and flexible supervision is achieved on all the aspects of the production processes as well as on the interactions among different sub-processes to the final aim of maximizing resource efficiency and improve process sustainability. Noticeably, the concept of "resource" includes not only primary raw materials and energy but also by-products, wastes, wastewater and off-gases. In this context and considering that some industrial processes and related parameters can be difficult to directly monitor through sensors (due e.g. to harsh conditions), the combination of standard techniques with advanced tools, such as specialised modelling software (e.g. Aspen Plus®) and Artificial Intelligence techniques (e.g. neural networks, genetic algorithm) is here exploited in order to improve the man-process connection. So doing, trials and plant changes or revamping can be addressed only to the most promising solutions. The previous described approach is here applied to address real industrial cases studies related to the improvement of the sustainability of steelmaking industry. The investigated issues ranges from the development and validation through tests of a general-purpose approach for improving the use of wastewater and by-products to a simulation module that is a part of a Decision Support Tool evaluating the impact of common and uncommon scenarios in the form of easy-to-understand Key Performance Indicators. Moreover, the extension of real experimentations is faced through process simulation, in order to improve process knowledge, to find the best operating conditions and configurations of treatment processes and to maximize the reuse of valuable by-products. Finally advanced models forecasting gas production or use are presented, which are developed to the aim of optimizing their exploitation in gas-networks. Despite the specific application in steelmaking industry, the multidisciplinary approach, which is presented in this thesis, is compatible and transferable to other industrial context within process industry.File | Dimensione | Formato | |
---|---|---|---|
Ismael_Matino_PhD_Thesis.pdf
Open Access dal 13/03/2021
Tipologia:
Altro materiale allegato
Dimensione
16.27 MB
Formato
Adobe PDF
|
16.27 MB | Adobe PDF | Visualizza/Apri |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/150154
URN:NBN:IT:SSSUP-150154