Computational Methods for Design, Estimation and Real-Time Control of PDE Systems with Applications to Mobile Sensor Networks
Abstract
The primary objectives of this work are the construction of a rigorous mathematical framework and the corresponding computational science tools that can be used to address problems of parameter identification, real-time tracking and estimation for spatially dependent systems. This includes determining optimized sensor/actuator locations for complex hybrid spatial systems to enhance tracking, estimation, information and effectiveness while limiting energy consumption. Reduced-order modeling techniques are implemented as an efficient way to compute the functional gains. We illustrate how a small number of strategically placed sensor/actuator is sufficient to stabilize the flow while inappropriate placement of these sensors could destabilize the flow. Additionally we consider information delays present in the sensor/actuator network. The models are complex multi-scale systems of coupled partial and delay differential equations. We show that under suitable conditions, the coupled delay PDE systems are well posed and we use this corresponding abstract formulation to construct efficient numerical methods for control design.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 14, 2013
- Accession Number
- ADA589255
Entities
People
- Eugene M. Cliff
- John A. Burns
- Lizette Zietsman
Organizations
- Virginia Tech