【正文】
om this point of view, therefore. Historical studies on underflow gates are available, and it is currently a mon belief that the discharge character is tics of vertical gates have been detailed in the past century. This is definitely not the case, because of the accuracy of discharge measurement, and the small hydraulic models often used. Well known approaches include those of Boileau (1848), Bornemann (1871, 1880), containing summaries of the experiments of Lesbros et al. Haberstroh (1890), Gibson (1920),Hurst and Watt (1925), Keutner (1932, 1935), Fawer (1937),Escande(1938), Gentilini(1941), and Smetana(1948). In these historical experimental studies, the exact geometrical configurations are often poorly specified, and the data are not always available. Details of gate fixation are also not described. The first modern study relating to free gate flow was conducted by Rajaratnam and Subramanya (1967). The coefficient of discharge was related to the difference of flow depths in the up and downstream sections hCa, where o c h approach flow depth, coefficient of contraction and o c agate opening. According to observations for both free and submerged flow C is exclusively a function of the relative gated opening a/h , and C increases slightly as a/h increases,o d o starting from . The effect of skin friction was stated d to be there as on for deviations between putations based on the potential flow theory and observations. Rajaratnam (1977) conducted a second study on vertical gates in a rectangular channel 311mm wide, with gate openings between 26 and 101 mm. The axial free surface profile downstream of the gate section was shown to be selfsimilar. Nout sopoulos and Fanariotis (1978) pointed at the significant scatter of data relating to both coefficients of contraction and discharge. The deviations between observations and theory were attributed to the spatial flow characteristics, and the channels too small often used in laboratories. Nago(1978) made ob