A System Behavioral Approach to Big Data-driven Nonlinear Process Control

该项目旨在开发一种新的过程控制方法，利用大过程数据来提高工业过程的成本效益。行业中现有的监控系统已经收集了大量的过程操作数据，但很少努力使用大过程数据来增强过程操作。基于系统行为方法和耗散理论，结合机器学习技术，该项目期望开发一种新的框架，用于使用大过程数据的数据驱动控制。这些结果有望使澳大利亚的过程工业受益，因为许多过程都是由不适当的逻辑控制器控制的，从而提高了它们的运行效率。

This project aims to develop a novel process control approach that utilises big process data to improve the cost-effectiveness of industrial processes. Existing monitoring systems in the industry have been collecting a tremendous amount of process operation data but little effort has been made to use the big process data to enhance process operations. Based on the system behavioural approach and dissipativity theory, integrated with machine learning techniques, this project expects to develop a novel framework for data-driven control using big process data. The outcomes are expected to benefit the Australian process industry, where many processes are controlled by inadequate logic controllers, by improving their operational efficiency.