Modular supervisory controller for complex systems. Lee, M. M. Y. Ph.D. Thesis, February, 2019. Accepted: 2019-04-01T20:28:14Z
Modular supervisory controller for complex systems [link]Paper  doi  abstract   bibtex   23 downloads  
Automation for the oil and gas industry is driven by the need to improve efficiency, productivity, consistency, and personnel safety, while reducing cost. Fully automated systems alleviate the physical toll on human operators and allow them to focus on monitoring unsafe well events and machinery maintenance. Complex systems like drilling rigs and snubbing units require supervisory controllers that can safely coordinate equipment and processes, overcome interoperability challenges and allow for functional scalability without sacrificing safety, security, and consistency of operations. The primary objective of this report is to explore the feasibility of developing a modular supervisory controller architecture which addresses these concerns by modifying and extending existing architectures. Such modifications include the use of non-homogeneous models in sub-system modules, including discrete event models for control and physics-based models for collision avoidance, addition of a system compilation module (Meta Module) to identify simple design errors, and implementation of an algorithm for synthesis of modules and filters to replace missing sub-systems. This report discusses the implementation results of the modular supervisory control architecture (modMFSM) on a simplified two-machine drilling system for assessment of design practices. Simulations for three test cases were executed to assess the ability of the controller to correctly perform error-free operations, detect and react to possible collisions, and adapt to missing equipment. The report then discusses the possibilities of extending the modMFSM architecture to control large complex systems such as drilling rigs, using snubbing operations as an example.
@phdthesis{lee_modular_2019,
	type = {Thesis},
	title = {Modular supervisory controller for complex systems},
	url = {https://repositories.lib.utexas.edu/handle/2152/73907},
	abstract = {Automation for the oil and gas industry is driven by the need to improve efficiency, productivity, consistency, and personnel safety, while reducing cost. Fully automated systems alleviate the physical toll on human operators and allow them to focus on monitoring unsafe well events and machinery maintenance. Complex systems like drilling rigs and snubbing units require supervisory controllers that can safely coordinate equipment and processes, overcome interoperability challenges and allow for functional scalability without sacrificing safety, security, and consistency of operations. The primary objective of this report is to explore the feasibility of developing a modular supervisory controller architecture which addresses these concerns by modifying and extending existing architectures. Such modifications include the use of non-homogeneous models in sub-system modules, including discrete event models for control and physics-based models for collision avoidance, addition of a system compilation module (Meta Module) to identify simple design errors, and implementation of an algorithm for synthesis of modules and filters to replace missing sub-systems. This report discusses the implementation results of the modular supervisory control architecture (modMFSM) on a simplified two-machine drilling system for assessment of design practices. Simulations for three test cases were executed to assess the ability of the controller to correctly perform error-free operations, detect and react to possible collisions, and adapt to missing equipment. The report then discusses the possibilities of extending the modMFSM architecture to control large complex systems such as drilling rigs, using snubbing operations as an example.},
	language = {en},
	urldate = {2020-05-10},
	author = {Lee, Melissa Mei Yun},
	month = feb,
	year = {2019},
	doi = {http://dx.doi.org/10.26153/tsw/1039},
	doi = {http://dx.doi.org/10.26153/tsw/1039},
	note = {Accepted: 2019-04-01T20:28:14Z},
}

Downloads: 23