噪声测量时点数量对8h等效声级估算误差影响的模拟研究  

Simulation of the Effect of Noise Measurement Time-points on Error of 8-hour Equivalent Continuous Sound Level Estimation

作　　者：吕旌乔[1] 曾琳[1] 李会娟[1] 成小如[1] 李玉秦[2] 赵一鸣[3] 

机构地区：[1]北京大学第三医院临床流行病学研究中心,北京100083 [2]郑州市疾病预防控制中心,河南郑州450053 [3]北京大学第三医院职业病研究中心,北京100083

出　　处：《工业卫生与职业病》2008年第3期149-152,共4页Industrial Health and Occupational Diseases

基　　金：卫生部科学研究基金(98-1-270)

摘　　要：目的探讨噪声测量时点数量与工作日噪声暴露（LAeq·8h）评估准确性的关系。方法采用SH126型记录式声级计，在某纺织厂测量59个8h工作班次的噪声水平，其中挡车班次25个、机修班次34个。对每一班次的测量均按10min间隔顺序记录48个10min等效连续A声级（LAeq·10min）。为模拟日常工作中根据特定时点短时噪声测量结果估算LAeq·8h的方法，应用区组随机抽样程序进行8轮抽样，分别从59组数据中抽取1、2、3、4、5、6、7、8个LAeq·10min，根据抽样结果估算8个LAeq·8h用估算LAeq·8h与真实LAeq·8h（根据全部48个LAeq·10min计算得出）之差的绝对值反映LAeq·8h估算误差，评价增加测量时点对LAeq估算误差的影响。结果挡车工、机修工噪声暴露方式明显不同，前者为典型稳态噪声暴露，而后者表现为非稳态噪声暴露。当测量时点为1时，挡车工暴露估算误差为（1．00±1．33）dB，而机修工为（8．62±11．90）dB。两类暴露LAeq·8h估算误差均随测量时点增加而下降，在测量时点数达3～4个时，下降曲线逐渐平坦。测量时点数目相等时，机修工暴露估算误差均大于挡车工。结论按单一时点的测量结果估算非稳态噪声LAeq·8h可能会导致较大误差，增加测量时点可减少估算误差。就稳态噪声暴露而言，增加测量时点去除LAeq·8h估算误差的意义较小。Objective To study the effect of noise measurement time-points in a shift on estimation accuracy for A-weighted 8-hour equivalent continuous sound level （LAeq·8h ）. Methods A-weighted noise levels in 59 work shifts in a textile mill, including 25 weaving shifts and 34 maintenance shifts were measured using SH126 personal noise dosimeter. During the measurement in each shift, LAeq for every 10 minutes （LAeq·10min）were saved in the dosimeter all together 48 LAeq·10min S were recorded in one shift. In order to simulate routine approach of LAeq estimation based on short-time noise measurement at fixed time-points, a block-randomized sampling procedure was developed to randomly select 1, 2, 3, 4, 5, 6, 7 and 8 LAeq· 10 min （s）from each of 59 measured noise data in 8 sampling procedure respectively. Based on sampled LAeq· 10 min （S） in each sampling, 8 estimated LAeq·8h（s）were calculated for each of 59 shifts. These estimated LAeq·8h were compared with corresponding actual LAeq·8h calculated from all 48 Leq· 10 min. respectively. The absolute values of differences in-between them were calculated to reflect the estimation error, which was used to evaluate the effect of measurement time-points increase on estimation error of LAeq. Results The pattern of noise exposure in weavers and maintenance workers in a shift was distinctively different. The former exposed to typical steady noise, while the latter exposed to intermittent unsteady noise. When the measurement time-point was set to 1, the estimation error for weavers was （1.00± 1.33） dB, and for maintenance workers was （8. 62 ±11.90） dB. For both weavers and maintenance workers, the estimation errors decreased as the measurement timepoints increased. When the measurement time-pointsreached to 3-4, the figure of estimation error became flat gradually. However, when the measurement timepoints were equal, the estimation errors for maintenance shifts were always higher than those of weaving shifts. Conclusions When estimating the unste

关 键 词：工业噪声 测量时点 8 h等效连续A声级 

分 类 号：TB53[理学—物理]