University of Zagreb. Faculty of Mining, Geology and Petroleum Engineering. Department of Geology and Geological Engineering.
Identification of impact of river dam near powerplant Zagreb on surrounding groundwater level : master's thesis
2017
U ovom radu identificiran je utjecaj praga TE – TO na podzemne vode neposrednog zaobalja. Analizirani su podaci iz razdoblja od 1. siječnja 1994. do 31. prosinca 2000. godine. Za utvrđivanje veze korištene su statističke metode korelacije, regresije i kroskorelacije. Obrada podataka izvršena je u programu Microsoft Excel, korištenjem funkcije za korelaciju, primjenom odabranih regresijskih modela te namjenskog VBA programa za kroskorelaciju. Metodom korelacije dobiveni su koeficijenti koji ukazuju na pozitivnu korelaciju, odnosno da je porast visine vodostaja praćen porastom razine podzemne vode. Metodom linearne regresije dobiveni su koeficijenti determinacije. U jednom slučaju utvrđeno je kako regresijski model dobro predstavlja mjerene vrijednosti, dok u preostala tri slučaja regresijski model ne predstavlja dobro mjerene vrijednosti. Metodom kroskorelacije dobiveni su koeficijenti korelacije na temelju kojih su određeni vremenski podaci, odnosno zaostajanje jedne varijable za drugom.In this thesis, the impact of the river dam near Thermal powerplant Zagreb on surrounding groundwater level had been identified. Data ranging from January 1st 1994 to December 31st 2000 was analysed. In order to determine the data relation, statistical methods correlation, regression and cross correlation had been used. Data analysis had been done in Microsoft Excel Software using correlation command, applying chosen regression models and with appropriate VBA code. Using correlation methods, correlation coefficients were determined. They show positive correlation, thus confirming water level growth followed by groundwater level growth. Using linear regression, coefficients of determination were determined. It had been shown with one result that the regression model represents valued measures well. In the rest of the cases, it had been shown that regression model represents valued measures poorly. Cross correlation method enabled calculating correlation coefficients which determined the time displacement, showing one variable lag in relation to another variable
Partners
Subscribe to repository