Seismic Analysis and Measurement of a Intake tower of reservoir.

(1)地震测量：1995年2月，在高42m的钢筋混凝土坝进水塔闸门操作室安装了地震仪。从4月初到7月，地震仪共记录到10次地震。其中有8个项目的最大记录加速度小于10Gal和M&lt；5.0。发生于‘95、7、3年的Sagami海湾外地震，最大水平加速度为42伽耳，震级为5.6级，震中距为60公里。此记录用于塔楼的反应分析。(2)将坝基加速度作为塔楼的输入运动：由于没有合适的设备来指定塔楼的输入地震运动，因此Nagara坝基地震仪采用了替代记录。为此，对“水利开发协作组”提供的各种帮助表示赞赏。(3)塔体力学模型的验证：分别采用常规梁柱有限元模型和能够反映基础摇摆的改进梁模型进行响应分析。由于材料力学常数通常不能最好地表示结构的观测自振周期，因此选择这些常量是为了使模型分析最好地表达观测自振周期。在此条件下，正常模型与观测到的加速度过程在统计意义上具有较好的一致性。但在一定的材料参数组合下，摇摆振动模型的响应值较大。

(1) Earthquake Measurement : In February 1995, the seismometer was installed in the gate operation room of the "NAGARA" dam intake tower which is RC and is 42m high. From the beginnings of April to July, 10 events were recorded by the seismometer. The maximum recorded acceleration of 8 events among them were smaller than 10 gal and M<5.0. The off-SAGAMI bay event occurred in '95,7,3 took the maximum horizontal acceleration of 42 gal, and of which magnitude is 5.6 and epicentral distance is 60 km. This record was used for the response analysis of the tower.(2) Use of acceleration in the dam foundation as a input motion to the tower : As there are no suitable devices which specify the input earthquake motion to the tower, an alternative record was adopted by the "NAGARA" dam foundation seismometer. To use this, various aids by "Water Resources Development Cooperation" were appreciated.(3) Approval of the mechanical model of the tower : Response analyzes using both normal beam-column FEM model and modified beam model which is able to express rocking of the foundation were carried out. Because the material mechanical constants are generally unknown to best expressing the observed natural period of the structure, these constants were chosen so that the model analyzes best express the observed natural periods. Under these conditions, normal model gave considerably good coincidence in the statistical sense comparing to the observed acceleration process. But under some combination of material constants, rocking vibration model gives a larger response value.