Adaptive optimization and estimation of complex dynamical systems

点击翻译按钮获取中文摘要

Global market pressures, environmental constraints and energy costs are having a significant impact on the management of companies in many sectors of the Canadian economy including chemical and petrochemical plants. Canadian companies must consider a holistic approach that combines science and technology into business strategies for producing services and products for a dynamic world market. These sudden and drastic changes in management culture have led to the development of alternative strategies that offer the flexibility required to handle economic pressures while reducing emissions and energy costs. Such an integrated approach to decision-making has pushed control engineers to think globally about plant behavior and to consider the full complexity of plant operation. The objective of this project is to develop process control techniques that are suitable for application in complex systems. Complex processes pose a number of technical challenges. They cannot be handled directly by the engineering tools that are commonly used to solve standard process control problems. In this research program, we will develop techniques to reduce the complexity of large systems by producing simpler descriptions that retain the essential features of complex systems. These simple descriptions provide a manageable platform to design control and monitoring systems for complex dynamic processes. The proposed research program will provide innovative techniques for the synthesis of estimation and control systems that allow the implementation of global objectives of the complex system architecture while providing maximum performance of each of its components. The research program tackles three specific areas of applications that address specific features of complex systems. The first area of application considers fluid flow estimation in building systems and fuel cell systems. The second area of application considers the design of control architectures suitable for large-scale system control such as chemical plants. The last area of application considers the design of drug delivery control systems.