基于多源监测数据的短时雷电精准预警算法设

Design of Short-term Lightning Accurate Early Warning Algorithm Based on Multi-source Monitoring Data

为保证雷电预警的精准性,为气象局提供可靠的预警信息,提出基于多源监测数据的短时雷电精准预警算法。该算法利用多普勒天气雷达、闪电定位系统以及大气电场仪,采集雷电的多源监测数据;将采集的数据传送至数据管理模块中,采用联合卡尔曼滤波算法,融合多源监测数据;预警展示模块以融合后的数据为基础,通过多变量局部异常系数方法进行雷电的精准预测,并呈现雷电的预警结果。测试结果显示:该算法概率均值结果与临界成功指数结果分别在0.90与0.915以上,能够较好地完成多源监测数据融合,预警雷电发生范围,提升雷电预警精准性。

In order to ensure the accuracy of lightning early warning and provide reliable early warning information for the meteorological bureau,a short-term lightning accurate early warning algorithm based on multi-source monitoring data is proposed.The algorithm uses Doppler weather radar,lightning location system and atmospheric electric field instrument to collect multi-source monitoring data of lightning.The collected data is transmitted to the data management module,and the joint Kalman filtering algorithm is used to fuse multi-source monitoring data.The early warning display module is based on the fused data,carries out accurate prediction of lightning through the multivariable local anomaly coefficient method,and presents the early warning results of lightning.The test results show that the probability mean value and critical success index of the algorithm are above 0.90 and 0.915 respectively,which can better complete the multi-source monitoring data fusion,early warning the lightning occurrence range and improve the accuracy of lightning early warning.