【摘要】 土石坝的监测数据预测对于揭示坝体变形规律、保障坝体结构安全具有重要意义。基于分形理论,将改进变维分形模型(IVDF模型)应用于土石坝小数据量监测数据的预测,以铜山源土石坝为例,选取混凝土防渗墙的应变监测数据作为研究对象,对应变进行预测,并分别用灰色模型GM(1,1)和BP神经网络进行对比预测。计算结果证明,IVDF模型充分利用分形的自相似性特征,抗噪性强,预测精度高,适用于小数据量监测数据的预测,具有良好的应用前景。更多还原

【Abstract】 Forecast of earth-rock Dam’s monitoring data is very important in revealing the deformation law of dam as well as in the protection of its structural safety.Based on fractal theory,this paper take Tongshanyuan earth-rock dam as example and attempts to apply the Improved Variable Dimension Fractal Model(IVDF model) in the deformation forecast of earth-rock dam by small amount of monitoring data.The monitored strain data of concrete cut-off wall of Tongshanyuan earth-rock dam is selected as research object to forecast the wall strain the forecast result are compared with the results of gray model GM(1,1) and BP neural network.The results show that,as IVDF Model makes good use of the self-similarity of fractal theory,thus it has advantages of high forecast precision and noise immunity,suit for the forecast with small amount of monitored data and has a promising application prospect.更多还原