利用卡式過濾理論進行線上外力識別

Abstract (In Chinese) 本研究介紹不同之外力識別法，以藉由量測所得之系統反應評估作用於系統之外力歷時。當施加之外力不影響量測方程式，亦即量測反應為位移或速度時，(1) RLS-KF Method、(2) Modified KF Input ID Using State Estimation、(3) Input Estimation through Direct Operation & Covariance Filtering 和 (4) Observer Technique for Unknown Input Estimation 等四種方法將應用於外力之計算。而 KF for Input ID Using Acceleration Responses 這個方法應用於外力同時影響到狀態與量測方程式的情況，例如量測的反應為加速度時。由數值模擬建立一個簡支梁結構系統，輸入隨機外力後運用狀態空間法建立系統反應後，以驗證上述的各種外力識別法。接著，當證明外力識別法的可行性後，將其實際應用於兩種不同的實驗。首先，設計一個兩端固定的梁並均勻裝設應變計做為量測反應之感應器。因為應變計的體積小、質量輕，對結構的質量影響很小，因此可以靈活地安裝在欲關心量測之區域，而且應變計之造價遠低於位移計、速度計和加速度計等傳感器。因此將嘗試直接以應變作為外力識別之量測反應的可行性。接著，將外力識別法應用於振動台之土壤─樁基礎互制實驗中。對土壤─樁基礎系統施予白噪音並量測系統反應，並以狀態空間法建立樁基礎之系統矩陣。最後，用介紹的外力識別法計算土壤─樁基礎系統之作用力大小。實際應用提出之外力識別法於上述兩種實驗後，將比較其識別外力結果並討論各種外力識別法之優缺點。

關鍵字

外力計算；系統識別法；應變量測；卡式過濾理論；遞迴式最小平方估測法

並列摘要

Abstract (In English) In this study several input force identification methods is presented using the direct response measurements. Consider the situation of without the unknown input term in the measurement equation, four input force identification methods, RLS-KF Method, Modified KF Input ID Using State Estimation, Input Estimation through Direct Operation & Covariance Filtering, and Observer Technique for Unknown Input Estimation, are used to deal with the above problem. The method of KF for Input ID Using Acceleration Responses is used when the measurement equation include both the state response and excitation term. The proposed input force identification methods were verified first using numerical simulation of a simple supported beam. After verified all of the above input force identification methods are practicable, the responses from two different experiments are used to identify the external forces. First, a simple beam structure with fixed end is designed and installs strain gauges on the bottom of the beam evenly. Because of the strain gauge can be easily installed onto structures and the price is not expensive, it is suggested to be the measurement sensor in the designed beam structure and provide a method to transfer the strain data to displacement data. Consequently, the external force can be identified from the strain response directly through the input estimation algorithm which are proposed. Second, data from the shaking table tests of soil-pile interaction are used to identify the interaction forces. For this large-scale shaking table test, the ambient vibration measurement of the pile is used to establish the structural system matrix. Based on the proposed identification methods in this study, soil-pile force is identified. After two experimental validations are conducted to confirm the applicability of the proposed input force identification methods. The identification results are compared and the different among each method are discussed.

並列關鍵字

Input estimation ； system identification ； Kalman filter ； strain measure ； recursive least-squares estimator ； strain measure

參考文獻

1. P. C. Tuan, C. C. Ji, L. W. Fong. W. T. Huang, “An input estimation approach to on-line two-dimensional inverse heat conduction problems,” Numerical Heat Transfer, Part B, 29, 345-363, (1996).

2. C. K. Ma, P. C. Tuan, D. C. Lin, “A study of inverse method for the estimation of impulse loads,” International Journal of System Science, 29(6), 663-672, (1998).

3. J. J. Liu, C. K. Ma, I. C. Kung, D. C. Liu, “Input force estimation of a cantilever plate by using a system identification technique,” Computer Methods in Applied Mechanics and Engineering, 190, 1309-1322, (2000).

4. C. K. Ma, P. C. Tuan, J. M. Chang, D. C. Lin, “Adaptive weighting inverse method for the estimation of input loads,” International Journal of System Science, 34(3), 181-194, (2003).

5. C. H. Loh, A. L. Wu, J. N. Yang, C. H. Chen, and T. S. Ueng, “Input force identification using kalman filter techniques: Application to soil-pile interaction,” Structural Control and Health Monitoring, 16, 223-240, (2009)

國際替代計量

利用卡式過濾理論進行線上外力識別

全文下載

主題瀏覽