Autonomous Estimation and Control for Stochastic Nonlinear Chemical and Biological Processes

对加拿大化学和生物工业日益增长的能源、环境和质量要求使它们面临着很高的经济压力。因此，提高这些行业的全面效率对它们的经济福祉至关重要。虽然各种因素决定了它们的效率，但在本项目中，我们将重点开发通过自适应信息处理机制提高效率的新方法。具体而言，该项目的目标是：(1)开发一种非线性过程在线自适应建模机制，(2)开发一种自适应实验设计策略，(3)使用自适应建模机制生成可操作信息。

The ever increasing energy, environment and quality demands on Canadian chemical and biological industries are subjecting them to a high level of economic pressure. Hence, increasing the all-round efficiency of these industries is vital for their economic well-being. While a variety of factors determine their efficiency, in this project we will focus on development of novel methods to improve efficiency through a mechanism of adaptive information processing. In particular, the objectives of this project are (1) to develop a mechanism for online adaptive modeling of nonlinear processes, (2) to develop a strategy for adaptive experiment design and (3) to use the adaptive modeling mechanism for generating actionable information. A lot of information, in the form of data and a priori knowledge, is available from chemical and biological processes. This data are often generated through batch experiments and processed in a batch fashion to capture them in the form of models. The data are also used to identify discrepancies between the model and the measurements. However, industrial processes are often changing and batch processing of data is suboptimal. In this project, we will develop adaptive data processing techniques. In particular, nonlinear, noisy, and time varying nature of real processes will be accounted for in the mechanisms developed. The developed methods will enable Canadian economic drivers such as oil & gas, pulp & paper, and biotech industries to operate their processes at optimal level. The project will also help us train highly qualified personnel in an important field of engineering.