加速度-位移关系的贝叶斯推理方法
Bayesian inference-based acceleration-displacement relation recognition method
-
摘要: 动力位移是地震工程、军事武器设计和结构健康监测等领域重要的物理量,但在实际测试过程中,通常能直 接量测的只有振动加速度信号。由于受环境等不确定性测试条件影响,加速度信号不可避免地含有低频和高频噪 声,导致在加速度积分过程中,速度和位移时程会产生较为明显的漂移现象。基于贝叶斯理论框架,构建了动力位 移贝叶斯学习识别方法,针对不同噪声工况(白噪声、人工噪声)反演获取了位移响应,识别出的动力位移与解析位 移基本一致;利用大型振动台试验数据,对比了不同性能加速度传感信号反演的位移,并分析了其不确定性。结果 表明:该动力位移贝叶斯学习识别方法在加速度?位移关系表征方面具备一定的优势,可不依赖对加速度信号的处 理实现位移求解,从而避免了噪声累积误差导致的位移积分失真。Abstract: Dynamic displacement is an important physical quantity in the fields of seismic engineering, military weapon design, and structural health monitoring. In the actual test process, the acceleration can usually be directly measured. Due to the uncertain test conditions such as the environment, the acceleration signal is unavoidable contains low-frequency and highfrequency noise, which causes a significant drift in velocity and displacement during the acceleration integration process. Based on the theoretical frame work of Bayesian inference, a Bayesian learning dynamic displacement identification method is constructed. The results show that, the displacement response obtained by inversion for different noise conditions (white noise, artificial noise) is basically consistent with the analytical displacement; the displacements of inversion of acceleration sensor signals with different performances are com pared by using a large shaking table test data, and their uncertainty is analyzed. The results show that this method has certain ad vantages in the characterization of the acceleration-displacement relationship, and can achieve the displacement solution without re lying on the processing of the acceleration signal, thereby avoiding the displacement integral distortion caused by the accumulated noise error.