Font Size: a A A

Frequency Analysis Of Nonstationary Hydrological Series In Loess Plateau

Posted on:2022-06-06Degree:MasterType:Thesis
Country:ChinaCandidate:C X LiFull Text:PDF
GTID:2480306512973639Subject:Hydraulic engineering
Abstract/Summary:PDF Full Text Request
In the middle reaches of the Yellow River basin,soil and water loss is serious,sediment deposition is frequent,and the ecological environment is fragile.Runoff and flood are sensitive to climate change and human activities,which lead to the loss of consistency of hydrological series.The application of traditional hydrological frequency analysis and calculation methods is limited.Therefore,it is necessary to carry out the nonstationary frequency analysis of hydrological series in the basin.In this paper,the frequency analysis method of inconsistent hydrological series under changing environment is studied,and the calculation results of design flood and design annual runoff of Qingyangcha,Jialu River and Xiaoli River basins in the Loess Plateau of the Yellow River basin are compared and analyzed,It is expected to provide decision-making reference for improving flood control safety and efficient utilization and management of water resources in the study area.The main research contents and results are as follows:(1)The nonstationary test of hydrological series.In this paper,linear trend analysis,Mann-Kendall test and Spearman test were used for trend analysis.The results showed that the annnal runoff series of Qingyangcha,Jialu River and Xiaoli River basins and the annual maximum flood peak series of Qingyangcha and Jialu River basins showed a significant downward trend except for the annual maximum flood peak series of Xiaoli River basin.By using ordered clustering method,sliding T test method,anomaly accumulation method and Lee-Heghinian method for mutation analysis,combined with simple hydro physical cause analysis,it was determined that the jump variation of annual maximum flood peak sequence occurred in 1970,and the jump variation of annual runoff sequence occurred in 1971 in Qingyangcha basin,The jump variation of annual maximum flood peak sequence and annual runoff sequence occurred in 1971 in Jialu River basin,and the jump variation point of annual runoff sequence occurred in 1970 in Xiaoli River basin.(2)The method of decomposition and composition was used to modify the consistency of hydrological series.Based on the trend and variability,the measured hydrological series was decomposed,and then the reduction series and the return series were composed.The consistency analysis and the calculation of the hydrological design value were carried out,and compared with the traditional design value.The results showed that the design value of the restoration sequence was generally larger than that of the traditional design value,and the design value of the present sequence was generally smaller than that of the traditional design value.(3)Based on the conditional probability distribution method,the hydrological frequency of the study area was analyzed.Based on the results of variability diagnosis,the parameters of conditional probability distribution model were obtained,and the hydrological frequency was analyzed.Through the fitting curve,goodness of fit test and evaluation criteria analysis,it could be seen that the fitting effect of conditional probability distribution model was better than the traditional method on the whole.(4)Based on GAMLSS model,the hydrological frequency of the study area was analyzed.The GAMLSS model with different covariates was established by introducing the precipitation on the first day and the first three days before the annual maximum flood peak,the annual precipitation,the annual average temperature and the effective runoff area controlled by the reservoir and dam project in the basin as covariates,and the hydrological design value was calculated.The results showed that the design value of gamlss model was an interval value,and the traditional hydrological design value was between the interval value.
Keywords/Search Tags:Nonstationary, Decomposition and composition, Conditional probability distribution, GAMLSS model
PDF Full Text Request
Related items