论文部分内容阅读
利用时间序列动态数据处理与建模对结构有损状态进行线性和非线性过程处理，用这两种方法对结构的物理状态进行谱分析和模态频率的提取进行了比较，计算结果表明非线性 Ｓ Ｅ Ｔ Ａ Ｒ 模型在特征谱的识别与提取上比线性 Ａ Ｒ Ｍ Ａ Ｘ 模型更有效，对微小损伤更灵敏。说明了在参数识别中非线性 Ｓ Ｅ Ｔ Ａ Ｒ 模型对物体损伤情况下振动数据建模较为客观，更能反映振动体系的损伤信息，是损伤识别数学模型有效的方法。
The lossy state of the structure is processed by linear time and non-linear processes using time series dynamic data processing and modeling. Using these two methods, the physical state of the structure is compared with the spectrum analysis and the modal frequencies are extracted. The results show that the nonlinearity is nonlinear. The S E T A R model is more effective than the linear A R M A X model in recognizing and extracting the feature spectrum, and is more sensitive to micro damage. It is illustrated that the nonlinear S E T A R model is more objective in modeling the vibration data under the condition of object damage, and can better reflect the damage information of the vibration system. It is an effective method for the damage recognition mathematical model.