Reducing Size and Cost of Biowall Control System
Abstract
The control system of the Purdue Biowall (the botanical air filter) prototype has been targeted for replacement to improve the potential for commercialization. This thesis is focused on evaluating and improving the performance of a new control system based on a single board computer, which is slated to replace a much larger and more expensive Building Automation System that is currently being used. This analysis considers size, cost, accuracy of sensors, and data logging. The research was conducted in three phases. Initially, an evaluation of an existing control prototype was conducted and several critical hardware failures were identified. Some of the most significant malfunctions were caused by broken wires, incorrect power supply voltage, and a short circuit due to poor soldering. The second phase of work involved improving the hardware and software for the control platform. All the problems found were fixed to make the control system completely operational. Moreover, a Printed Circuit Board (PCB) was designed to replace the breadboard previously used. The third phase of work was evaluating the performance of the new control system. Evaluations at a component level (e.g. individual sensors) and the overall system (e.g. including Biowall control algorithm) were conducted. After a complete assessment, it was determined that a cheaper and smaller single board computer control system is able to substitute for the current Building Automation System. The accuracy of the sensors and the data collection were within the values expected. The physical size and cost of the controller was reduced by a factor of 30. The results obtained have identified several areas where further improvement is still needed. An efficient data logging code, replacement of wires, and PCB enhancement are still needed before deploying the new control solution into a building.
Degree
M.Sc.
Advisors
Hutzel, Purdue University.
Subject Area
Design|Atmospheric sciences|Energy|Information Technology
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.