雷击森林火灾烟流污染含量动态监测方法研究

Study on Dynamic Monitoring Method of Smoke Pollution Content in Forest Fire Caused by Lightning

为降低火灾烟流污染含量监测结果误差,针对雷击森林火灾引起的烟流污染,提出新型污染含量动态监测方法。对雷击森林火灾烟流污染现状及其对环境及人体的影响进行分析,基于鲁棒单边对齐域自适应检测算法分析数据,定位火灾烟流污染源。以大气变层高度、水汽含量、温度、相对湿度为解释变量,建立多元回归分析模型。结合遥感数据,得到烟流污染实时含量值。运用软测量校准技术校验污染含量估算值,输出监测结果。实验结果表明:应用所提方法得到的动态监测结果,平均绝对百分比误差仅为2.13%,保证了动态监测准确性的大幅提升。

To reduce the measurement error of smoke pollution in fire,a new dynamic monitoring method for pollutants was proposed specifically for smoke pollution caused by lightning-induced forest fires.The current status of smoke pollution from lightning-induced forest fires and its impact on the environment and human health were analyzed.Based on the robust unilateral alignment domain adaptive detection algorithm,the data were analyzed to locate the source of smoke pollution from fires.A multiple regression analysis model was established using atmospheric inversion height,water vapor content,temperature,and relative humidity as explanatory variables.By combining remote sensing data,real-time measurements of smoke pollution concentration were obtained.The pollution estimates were validated using soft measurement calibration techniques,and the monitoring results were outputted.Experimental results demonstrated that the dynamic monitoring results obtained using the proposed method had an average absolute percentage error of only 2.13%,greatly improving the accuracy of dynamic monitoring.