Từ những nhận xét: "Cuốn sách này có chuyên môn cao là thú vị không chỉ vì vấn đề quan trọng của nó mà còn vì quan điểm ích kỷ của nó trong khi độc giả đánh giá cao toán học sẽ thấy rằng cuốn sách cung cấp một đi vào lĩnh vực đang phát triển của nghiên cứu trận tuyết lở hạt."(Iverson Richard M., Tạp chí Địa chất, Vol. 116, 2008) "Pudasaini Shiva . đã viết tóm ấn tượng này để tóm tắt hai mươi năm nghiên cứu trong các lĩnh vực của các luồng hạt Các. | With this in mind stations 4 and 8 match fairly closely between flume and numerical model. Station 4 in the flume would still have a greater difference between outer and inner wave than that predicted by the model. The difference might be a manifestation of a three-dimensional effect that the model cannot mimic. The overall timing and height comparisons are good. Figure 27 shows the spatial profile of the outer wall water surface elevation of the numerical model versus distance downstream from the dam. These distance measurements are in terms of the center-line distance. The two conditions are for CX of and . first- and second-order temporal derivative. Figure 27. Dam break case water surface elevations comparison of temporal representation for time of sec The nodes are delineated by the symbols along the lines. The overshoot of the second-order scheme and the damping of the first-order is obvious. Again it is probable that the overshoot is a numerical artifact even though this is much like what the flume would show. Case 3 2-D Lateral Transition This is the most geometrically general case that we test. The numerical model is compared to flume results. The flume data was reported in Ippen and Dawson 1951 . The tests were conducted for an approach Froude number of 4 upstream depth of ft m and a total discharge of fp sec m3 sec . The channel contracts from 2 ft m to 1 ft 40 Chapter 3 Testing m wide in a length of ft m . an angle of 6 deg on each side. The model resolution was increased until we were confident that the results no longer changed with greater resolution. The numerical model was set up with 10 evenly spaced elements laterally across the channel and 24 over the length of the transition. The model limits were extended some 40 ft m . The total number of nodes was 1661 with 1500 elements. As in the flume test the numerical model was set up to provide a uniform depth of ft m
