Subventions et des contributions :

Subvention ou bourse octroyée s'appliquant à plus d'un exercice financier. (2017-2018 à 2022-2023)

The increase in the complexity of industrial plants combined with an increased emphasis on more environmentally friendly and sustainable processes has created a need for the development of methods that can increase the degree of process integration and automation. For chemical plants, this emphasis requires the development and implementation of a holistic process control strategy, that can, not only provide acceptable control, but can also handle process disturbances, as well as internal and external changes, such as fouling and ambient temperature changes. This can be achieved by emphasising the long-term need to develop accurate and effective control and modelling strategies that can handle a continually changing plant. In the short to medium term, there is a need to develop methods that can perform process optimisation in the face of changing conditions, be they arising from changes in plant components, modules, or objectives.

Thus, this research programme will consider 2 key independent objectives. Firstly, the proposal will develop a framework for advanced soft sensors so that it can handle uncertain and potentially conflicting information that may not be available as often as required, for example, data that is required every second is only available every hour. Secondly, the proposal will examine data-driven process monitoring methods in the case where there are periodic changes in the system.
The main goal of this research programme is the development of a soft sensor framework that can handle uncertain and unreliable process information to provide the best, most accurate estimate of the current process conditions. Furthermore, data-driven process monitoring, which will be obtained by combining orthogonal function decomposition with standard monitoring metrics, will provide a window on the process irrespective of the process changes that may be occurring. Overall, this research programme seeks to provide the tools necessary to obtain a better understanding of the chemical plant leading to improvements in safety and reductions in environmental impact and costs, such as those associated with lowered tailings emissions in the Canadian oil sands industry.

This proposed research programme will also place a strong emphasis on student training. It is expected that 3 master’s students and 2 doctoral students will gain highly sought-after industrial skills, such as data mining, fault detection and isolation, and process monitoring. These skills will allow the students to pursue rewarding careers both in Canadian industries, such as pharmaceuticals, petrochemicals, and steel industries, and in academics.