Robust Control of Biomedical and Environmentally Sustainable Engineered Systems

点击翻译按钮获取中文摘要

The environmental sustainability of engineered systems and the development of biomedical systems for improved therapies will be critical to ensure a better quality of life for Canadians and to preserve the environment for future generations. The main goal of the proposed engineering research program is thus to develop innovative robust and sustainable control systems for: (a) environmentally sustainable engineered systems: electric vehicles (EV), plug-in hybrid electric vehicles (PHEV), energy management systems, building heating, ventilation and air conditioning (HVAC) systems using renewable energy, and (b) medical applications, mostly focusing on the artificial pancreas.******Electric (EV) and plug-in hybrid electric vehicles (PHEV) produce no or little emissions and are now widely seen as the future of sustainable transportation. We first propose to study the design, control and charging of EV and PHEV from the point of view of dynamic energy optimization. The research will focus on minimizing consumption of fossil fuel and electrical energy through advanced feedback control design. One of the fundamental issues in PHEV and renewable energy systems is how to use the electrical energy accumulator with respect to the dynamics of the system and the load in order to minimize fuel or energy consumption. Secondly, we propose to investigate the control of heating, ventilation and air conditioning systems of large buildings to maximize the use of energy accumulators, such as ice banks, and photovoltaic panels. ******The objective of the blood glucose control system research for diabetic patients is to design and implement a closed-loop device that regulates glucose levels for people with Type 1 diabetes. The closed-loop device (so-called artificial pancreas) measures blood glucose levels and uses these measurements via a control algorithm to properly dose insulin and glucagon using the subcutaneous insulin and glucagon pumps. Our recent collaborative work has developed a control algorithm for a dual-hormone artificial pancreas which proved to be successful in a clinical trial on 15 patients with type 1 diabetes. We wish to continue this research in the development of an advanced wearable artificial pancreas. Novelty of the proposed approach lies in the development of adaptive and robust MPC algorithms that adapt to changes in insulin sensitivity while being robust to uncertain dynamics and long diffusion delays. We also propose to look at cycle-to-cycle control using iterative learning as insulin sensitivity patterns follow circadian rhythms. ******This research program will provide multidisciplinary training to three Ph.D. students and three Master's students who will be attractive to Canadian organizations and institutions as effective agents of change.