论文部分内容阅读
为了压低干扰和有效地提取信息,本文采用Kalman滤波方法,处理了京津张唐地区17口深水井1970～1981年间的观测资料。结果表明,1976年7月28日的唐山大地震前,天津地区6口水井水位有2—3年的异常下降,下降的开始时间以3—4公里/月的速度传播。天津的6口井位于大震破裂面的西南端。大震后至今,水位无明显的异常变化,与6级以上的强余震无明显关系。北京地区的11口水井水位与大震似无明显关系。本文所采用的Kalman滤波中,每一步都采用了自回归方法和递归滤波方法计算了一步转移矩阵的元素和控制项,因而转移矩阵和控制项是随时间变化的,更适于地震前兆资料的适时处理。
In order to suppress interference and extract information effectively, Kalman filtering method was used in this paper to process the observed data of 17 deep wells in the Tang and Tang Dynasties in Beijing, Tianjin and Yangzhou from 1970 to 1981. The results show that before the Tangshan earthquake on July 28, 1976, the water level of 6 wells in Tianjin area experienced an abnormal decline of 2-3 years, with the start of decline at 3-4 km / month. The 6 wells in Tianjin are located on the southwestern tip of the rupture. Since the earthquake, there is no obvious abnormal change in water level, which is not significantly related to the strong aftershocks above Grade 6. There are no obvious relations between the water level of 11 wells in Beijing and the earthquake. The Kalman filter used in this paper adopts the autoregressive method and recursive filtering method to calculate the elements and control items of the one-step transfer matrix in each step. Therefore, the transfer matrix and control items change with time, which is more suitable for seismic precursory data Timely processing.