University of Central Florida
Recurrent Neural Network Modeling of a Developed Multi-Nozzle, Piezoelectric-Based, Spray Cooling Testbed
Abstract
dc:description.abstractTo model and examine the thermal fluid phenomena involved in high-pressure, multi-nozzle spray cooling, a testbed is developed which includes a heating subsystem and an accumulator to pressurize common rail based piezoelectric injectors. Compared to conventional platforms, the implemented testbed allows for an abundance of layout arrangements and settings that provide a greater range of functionality. The volumetric flow rate of the testbed is modeled by a recurrent neural network trained from time-sequential obtained through experiments. The fidelity of the model, as well as the testbed's hardware, software, functionalities, and shortcomings are discussed.
Degree
thesis:*- Grantor dc:publisher
- University of Central Florida
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Fordon, Andrew
- Contributors dc:contributor
-
- Xu, Yunjun
Subjects
dc:subject × 4- Thermal fluid phenomena; Recurrent neural network; Spray cooling; Testbed development; Multi-nozzle system
- Aerospace Engineering
- Space Vehicles
- Nozzles--Fluid dynamics--Mathematical models; Cooling--Mathematical models; Nozzles--Testing; Spray nozzles--Testing; Fluids--Thermal properties--Mathematical models
Rights
- Language dc:language
- English
Identifiers
dc:identifier.*- Identifier
- CFE0009725; DP0027832