Modeling Nutrient Dynamics in Wastewater Treatment Processes Using Hybrid Kinetic Model and Artifical Neural Network

Hao 9625183这是一个奖项，旨在为旨在开发一种利用悬浮生长生物反应器从废水中去除养分的方法进行建模的研究提供支持，该方法基于人工神经网络的使用与反应器内发生的潜在生物过程的知识相结合。调查人员计划使用一种混合建模方法，将简化的过程模型和神经网络结合起来，其目的是纠正使用简化模型遇到的任何错误。将串联结构和并联结构两种混合建模方法用于悬浮生长反应器模型和神经网络的集成。该方法将被探索用于序批式反应器和交替好氧和缺氧条件的系统中。这项研究的结果有望加强复杂生物过程的设计、控制和优化的程序。本项目将研究的拟议概念预计将全面加强过程控制，并导致工程设计在满足排放到环境之前从废水中去除营养物质的严格要求方面节省能源。

Hao 9625183 This is an award to provide support for research aimed at developing an approach to modeling nutrient removal from wastewater by use of suspended growth bioreactors based on coupling the use of artificial neural networks with knowledge of the underlying biological processes that take place within the reactors. The investigators plan on using a hybrid modeling approach that combines a simplified process model and a neural network, the purpose of which is to correct any errors encountered by use of the simplified model. Two hybrid modeling methods involving a series structure and parallel structure will be used to integrate a suspended growth reactor model and a neural network. The approach will be explored for potential use in both sequential batch reactors and in a system for alternating aerobic and anoxic conditions. Results of this research are expected to enhance the procedures by which complicated biological processes can be designed, controlled and optimized. The proposed concepts to be investigated in this project are expected to enhance process control in general and lead to engineering designs that save energy in meeting stringent requirements for removal of nutrients from wastewater prior to their discharge to the environment.