香港公共出租车转换站内污染物演变模式识别及预测

对香港公共出租车转换站(TTI)内的污染物扩散和分布情况进行了研究.影响TTI内污染物水平的因素有出租车流、环境条件、人们的活动以及TTI的结构等.对TTI内的悬浮粒子进行了现场监测,通过分析前述因素对悬浮粒子水平的影响,提出用人工神经网络来进行研究.基于所获得的不同时间段的数据,训练了适用于此类问题的神经网络模型,对该模型的性能进行了分析和验证.文中还对TTI内空气污染物的积聚模式及演变趋势进行了模拟预测和实测对比,对比结果表明,文中提出的TTI内污染物的预测方法是可行和可靠的.

Characteristics Recognition of Pollutant Levels in Public Taxi Transfer Interchanges in Hong Kong

An investigation on pollutant dispersion and distribution inside public taxi transfer interchange (TTI)is reported. The pollutant levels inside TTI are affected by many factors, e.g., taxi flow-rate, environmental con-ditions, human activities, and geometrical layout of TTI, etc.. A site measurement of respirable suspended parti-culate (RSP) level is carried out in a typical TTI in Hong Kong. After analyzing the effect of above factors onRSP level, the authors propose to use artificial neural network (ANN) to study such phenomena. The recordeddata within different time periods inside the selected TTIs are used as the test data set to train the proposed neuralnetwork model. The recovery performance of ANN model is analyzed and justified. In the study, the authorscompare the forecasting results and the measured data of RSP in morning and afternoon sessions respectively. Theresults show the feasibility and reliability of the proposed approach for forecasting pollutant levels inside TTI.