启发式分割算法在高速铁路桥墩沉降评估中的应用OA

The settlement monitoring data of high-speed railway bridge piers have the characteristics of"small values and large fluctuations",which makes it difficult for the traditional curve regression method to accurately evaluate the settlement stability.To address this issue,based on the continuously measured settlement monitoring data of a certain high-speed rail-way bridge pier for 61 periods,this paper conducted the fitting analysis and settlement trend stabilization treatment research by using the hyperbola method,the three-point exponential curve method,and the heuristic segmentation algorithm.The results show that when the settlement stability cannot be evaluated by the curve regression method,the heuristic segmentation algorithm stabilizes the non-stationary sequence.By using the mean change point detection,it is found that from the 31st period to the last period,the mean value of the monitoring data has remained unchanged for 31 consecutive periods,and the number of stable periods exceeds half of the total number of measurement periods.It is determined that the bridge pier is cur-rently in a stable state.It is consistent with the evaluation result of the empirical method.The research results show that when the minimum segmentation scale and statistical significance are determined,the heuristic segmentation algorithm can more accurately reflect the current settlement deformation by stabilizing the non-stationary settlement data.It is suitable for the settlement evaluation of high-speed railway bridge piers and can be used as a supplement to the existing settlement evaluation methods.