Study on a mother wavelet optimization framework based on change-point detection of hydrological time series

Abstract. Hydrological time series (HTS) are the key basis of water conservancy project planning and construction. However, under influence climate change, human activities other factors, consistency HTS has been destroyed cannot meet requirements mathematical statistics. Series division wavelet transform effective methods to reuse analyse HTS. they limited by change-point detection mother (MWT) selection difficult apply promote in practice. To address these issues, we constructed a potential set based on cumulative anomaly method, Mann–Kendall test detection. Then, degree change before after point was calculated with Kolmogorov–Smirnov test, criteria were proposed. Finally, optimization framework proposed according accuracy MWT, continuous used evolution. We Pingshan station Yichang Yangtze River as study cases. The results show that (1) can quickly locate points, determine trajectory complete (2) MWT optimal select conforms characteristics ensure uniqueness transformation. This analyses evolution provides better for hydrological hydraulic calculation, which will improve design flood estimation operation scheme preparation.