基于经验模态分解和灰色模型的水工结构健康监测数据分析

Data Analysis on Health Monitoring Based on Empirical Mode Decomposition and Hydraulic Structures of Grey Model

由于大坝结构的复杂性以及所处的环境,大坝坝体内部的情况往往难以探明,管理人员很难觉察到内部的细微变化。为了实现大坝结构安全运行,必须引入智能化的监测系统。基于此,文章对大坝的位移进行了监测,并基于经验模态分解法对位移数据进行处理,结合灰色线性回归组合模型对数据进行了拟合,发现该模型取得了较高的拟合效果,灰色线性回归组合模型由于不仅描述了原始序列的指数增长趋势,还有效描述了变量之间的线性关系,取得了较高的拟合效果,其平均相对误差最小,仅为1.610%,可以为大坝健康监测数据预测提供理论依据,确保大坝结构的安全运行。

Due to the complexity of the dam structure and the environment in which it is located, the internal conditions of the dam body are often difficult to detect. It is difficult for managers to detect subtle changes inside. In order to realize the safe operation of dam structure, intelligent monitoring system must be introduced. Based on this, the displacement of the dam is monitored in this paper.The displacement data were processed based on the Empirical Mode Decomposition Method. The data was also fitted with the Grey Linear Regression Combination Model. It is found that the model has achieved high fitting effect. The Gray Linear Regression Combination Model not only describes the exponential growth trend of the original series, but also effectively describes the linear relationship between variables. It achieved a high fitting effect. Its average relative error is the smallest, which is only 1.610%. The model can provide theoretical basis for dam health monitoring data prediction and ensure the safe operation of the dam structure.