FEEDBACK CONTROL OF PARTICLES IN FLUID MEDIA

随着精确的实时监测器的发展， 和气体浓度，存在发展的潜力， 自动操纵建筑物的反馈控制系统 与污染物水平相关的通风率。 拟议 研究关注这样一个系统的发展， 使用动物吸入室。该设备将允许 控制设置，其中污染物浓度和流速可以 保持在预定的水平。本研究的具体目的 将调查（l）室浓度水平的响应 粉尘发生器速度和空气流量的各种变化 率，（2）统计过程控制（SPC）技术的使用 实现浓度的反馈控制。初步研究 已经进行了开发设备，以自动化生产的 带有计算机控制的粉尘进料和腔室空气阀的方石英 执行器. 该系统已被证明能够生产和 保持室浓度水平等于期望值 由运营商选择。拟议的研究将扩大这一系统 结合基于发送到 实时气溶胶监测仪的数据我们假设 在发展这个系统时，需要克服的困难包括： 浓度水平对浓度变化的反应不一致 发电机速度或流速，以及需要准确预测平均值 当阅读显示高浓度时， 瞬时变化的程度。 我们将尝试使用SPC 指数加权移动平均（EWMA）技术， 两者均表示平均水平，并提供统计学推导的对照 EWMA导出值的限制，以缓冲控制系统 监测器读数出现尖峰，可能导致 控制系统 这项研究的结果将立即 对吸入毒理学领域的影响， 电离室浓度水平对剂量结果很重要- 反应关系。 我们相信这些信息也可能对 涉及到通风系统的工业企业家， 稀释空气，以控制污染物水平-在工厂，矿山和 写字楼

With the development of accurate real-time monitors to measure aerosol and gas concentrations, the potential exists for the development of a feedback control system to automatically manipulate building ventilation rates in relation to contaminant levels. The proposed studies are concerned with the development of such a system with the use of an animal inhalation chamber. This device will allow for a controlled setting in which contaminant concentration and flow rate can be maintained at predetermined levels. The specific aims of this study will be to investigate (l) the response of chamber concentration levels of cristobalite to various changes in dust generator speed and air flow rate and, (2) the use of statistical process control (SPC) techniques to implement feedback control of the concentration. Preliminary studies have been conducted to develop devices to automated the production of cristobalite with a computer controlled dust feed and chamber air valve actuator. This system has proven to be capable of producing and maintaining a chamber concentration level equivalent to a desired value chosen by the operator. The proposed studies will expand this system to incorporate feedback control based on information sent to the computer from a real-time aerosol monitor. We hypothesize the main difficulties to be overcome in the development of this system include: inconsistent response of the concentration level to a change in generator speed or flow rate and, a need to accurately predict the mean level of the chamber concentration when the reading exhibit a high degree of instantaneous variation. We will attempt to use an SPC technique known as an exponentially weighted moving average (EWMA) to both indicate the mean level and provide statistically derived control limits about the EWMA derived value to buffer the control system from spikes in the monitor readings which may cause an over-reaction of the control system. Results from this study will have immediate implications to the field of inhalation toxicology where strict control of chamber conCentration levels is important to results of dose- response relationships. We believe this information may also be of use to industrial hygienists involved with ventilation systems which use dilution air to control contaminant levels-in factories, mines and office buildings.