Research on the Data-Driven Quality Control Method of Hydrological Time Series Data

Abstract

Ensuring the quality of hydrological data has become a key issue in the field of hydrology. Based on the characteristics of hydrological data, this paper proposes a data-driven quality control method for hydrological data. For continuous hydrological time series data, two combined forecasting models and one statistical control model are constructed from horizontal, vertical, and statistical perspectives and the three models provide three confidence intervals. Set the suspicious level based on the number of confidence intervals for data violations, control the data, and provide suggested values for suspicious and missing data. For the discrete hydrological data with large time-space difference, the similar weight topological map between the neighboring stations is established centering on the hydrological station under the test and it is adjusted continuously with the seasonal changes. Lastly, a spatial interpolation model is established to detect the data. The experimental results show that the quality control method proposed in this paper can effectively detect and control the data, find suspicious and erroneous data, and provide suggested values.

Numerical approximation of the hydrological time of concentration

“…Temez's Formula [6] = 0.3 0.76 −0.19 (8) This expression was born from time series data collected in basins in Spain with areas less than 3,000 2 . Eqs.…”

“…[1,2]. According to some authors, the time of concentration can be defined as the time that a single drop of water takes to travel from the most distant point in the basins to the point of exit [3][4][5][6][7][8]. Following [3], this parameter is related to the lines of equal time of flow to the outlet, called isochrones representing the grown of contributing area to the streamflow outlet after certain time.…”

“…This measure is considered after the maximum flood produced by the greatest possible rainfall. Other works, (see [8]) put forward the hydrological data quality method based on data-driven methods as an alternative to determine the hydrological data more efficiently and to provide an accurately reference for the hydrological workstations. functionality is also defined as the design of rainwater systems, hydraulic structures using rational methods and operations in hydraulic infrastructures [9].…”

Different Approaches to SCADA Data Completion in Water Networks

“…The problem of managing lost data is ubiquitous in many situations and is especially challenging when it manifests itself in long bursts. Dealing with incomplete data is very common in real-life cases, and it is usual to find such cases in water [1] and hydrological data management [2][3][4]. The current work used data from the Drinking Water Treatment Station (DWTS) of Aigües de Vic S.A. For this purpose, data from several DWTS sensors that are stored by the SCADA system in a database were analyzed.…”

Double Tensor-Decomposition for SCADA Data Completion in Water Networks

“…In practice, when processing this amount of information, the problem of incomplete or missing data has to be addressed. The management of data from water networks [1] or from hydrological resources [2][3][4] is no exception. The problem of data loss is especially challenging when it occurs in long bursts of consecutive values.…”