Scholarly record

APPLICATION OF STOCHASTIC MODELING IN DETERMINING THE AVAILABLE WATER RESOURCE FOR DAM MANAGEMENT IN BULGARIA

A. Yordanova, I. Ilcheva, E. Bojilova

Abstract

In studying WRS that belong to complex probability systems, it is necessary first to make a mathematical description of the input process of the system - the process of changing the river flow. The use of the available series, which cover several decades of years (in Bulgaria about 30-70 years), cannot reflect the alternation of sequences of high-water and low-water years and the great diversity of the intra-annual runoff distribution, which characterizes the different water years. For these reasons, studies based on the direct use of insufficient available series are fraught with significant random errors in outflow parameters and their values have low stability. The Monte Carlo method, based on the statistical parameters and the distribution function, can more accurately reflect the patterns of runoff and water consumption in the general population. It is based, instead of the available time series, on the mathematically modeled long-term simulated realizations for the runoff and water consumption and through processing in a balanced way the realizations of the transformed process - the regulated outflow is obtained. By statistical processing of the transformed process the responsibility functions characteristics 1) of the runoff regulation, 2) of the overflowed water masses, 3) of the filling of the reservoir, 4) of the drained water masses and 5) of the deficits during the unreliable period are obtained. Without these characteristics, it is impossible to establish the optimal regulation parameters. In reservoirs planning or managing, we work with the monthly data on river runoff. The report presents a methodological approach (Methodology for determining volumes in dams under Annex ?1 of the Water Act, www.moew.government.bg) based on the use of modeled series in determining the available water resource and there are specified management rules for dams with desired water management indicators. Generating of 500 or 1000 annual sequences of monthly water volumes was performed by the method of the linear autoregression and the method of fragments. Based on the obtained series of realizations of the inflow to the dams, it can be estimated the required reservoir useful volume, the maximum volume that can be submitted for water consumption, the management functions are developed. Stochastic modeling models are applied in practice on a number of projects and contracts with different institutions. The application of the approach for the complex and significant dams of Bulgaria (Iskar dam, Topolnitsa Dam) is presented.

Publication Impact Profile

• Citations
• Scopus - Citation Indexes: 1

• Captures
• Mendeley - Readers: 2

PLUMX see details

DOI resolver: doi.org/10.5593/sgem2021/3.1/s12.05

Publication details

Back to publication list