NIOSHTIC-2 Publications Search
Real-time process control techniques for a system with unpredictable disturbances.
Instrument Society of America Conference, Toronto, Canada, 1995. Pittsburgh, PA: Instrument Society of America, 1995 Jan; 50(1):11-19
The design of control systems to ensure that an aerosol concentration in an environmental chamber remained constant was investigated. The aim was to use the system in cases where there was a low signal to noise ratio. A feedback based control system was developed for controlling the production of aerosols in the environmental chamber. The aerosol was generated by packing the bulk powder in a metal cylinder and then slowly rotating the cylinder across a metal blade contained within a Wright dust/feed (WDF). The scraped powder was entrained in an air stream passing through the WDF and was carried into the main dilution air stream entering the chamber. The target aerosol concentrations in the exposure chamber, which were measured by an electronic aerosol monitor placed in the center of the chamber, were determined by the speed of the aerosol generator. The aerosol generator speed was maintained by a stepper motor that was interfaced to a personal computer through a series of gears that was attached to the dust cylinder. For a given rotation speed and air flow rate, aerosol concentrations in the chamber could deviate from the target level due to changes in the proportion of dust lost in the system before reaching the monitor, changes in the density of the dust cake packed into the cylinder, or the sudden entrainment of settled dust back into the air stream going through the chamber. There was also some noise associated with the monitor and which resulted from incomplete mixing of the dust in the air stream. A computer assisted design program was used to evaluate a transfer function that would smooth the aerosol concentration readings near some established value and create control limits that vary over time. This produced a program that used a pseudo random binary sequence of inputs, a random sequence of 1s and -1s, that generated an input to the process which could be evaluated along with the measured output. This model reduced the number of control actions needed to maintain the aerosol concentration at the target value and avoided the necessity of having to make unnecessary corrections when sudden short lived disturbances affected the system.
NIOSH-Grant; Grants-other; Control-technology; Mathematical-models; Statistical-analysis; Automation; Aerosol-generators; Computer-models; Air-flow; Dust-analysis; Feedback-controls
Civil Engineering University of Vermont RM 226 Votey Bldg Burlington, VT 05405
Issue of Publication
Other Occupational Concerns; Grants-other
Instrument Society of America Conference, Toronto, Canada, 1995
University of Vermont & St Agric College, Burlington, Vermont
Page last reviewed: May 5, 2020
Content source: National Institute for Occupational Safety and Health Education and Information Division