An Adaptive Operational Modal Analysis Method Using Encoder LSTM with Random Decrement Technique

A new parameter identification method under non-white noise excitation using transformer encoder and long short-term memory networks (LSTMs) is proposed in the paper. In this work, the random decrement technique (RDT) processing of the data is equivalent to eliminating the noise of the raw data. In general, the addition of the gate in LSTM allows the network to selectively store data, which avoids gradient disappearance and gradient explosion to a certain extent. It is worthwhile mentioning that the encoder can learn the essence of data, which reduces the burden for the LSTM. More specifically, establish as simple LSTM structure as possible to learn the data of this essence to achieve the best training effect. Finally, the proposed method is used for simulation and experimental verification, and the results show that the method has the advantages of high recognition accuracy, strong anti-noise ability, and fast convergence rate. Specially, the results indicated appropriate accuracy proposed by deep learning combined with traditional method for parameter identification as well as proper performance of the proposed method.

Identification of modal parameters from non-stationary responses of high-rise buildings

A key issue in the control, health monitoring, and condition assessment of civil structures is the estimation of structural modal parameters based on measured structural responses. However, field measurements of structural responses from civil structures under strong wind or earthquake excitations usually exhibit non-stationary feature and therefore cannot be adequately deal with by traditional modal identification methods. In this study, a novel procedure is integrated for modal parameter identification of civil structures from non-stationary structural responses on the basis of the variational mode decomposition (VMD) technique. First, the VMD algorithm is applied to decompose measured vibration signals into individual mode components. Then, the random decrement technique (RDT) is employed to obtain free vibration response of each mono component. Next, normalized Hilbert transform (NHT) is used to estimate modal natural frequency and damping ratio. The performance of the developed approach is evaluated using simulated non-stationary responses of a frame structure, and the identified results are validated. The effects of crucial factors such as levels of noise involved in structural response and data length on the modal parameter estimations are examined through detailed parametric study. Furthermore, the approach is applied to modal identification based on field measured non-stationary responses of a high-rise building during Typhoon Nida. The case study illustrates that the integrated method is an efficient tool for estimating the modal parameters of civil structures from non-stationary structural responses.

Structural health monitoring of two-way slabs based on random decrement technique

The current research attempts to explore the feasible use of a Structural Health Monitoring method for a two-way slab system through the effective vibration based damage diagnostic technique of Random Decrement (RD). Experimental investigations have been conducted on a total of four reinforced concrete two-way slab specimens. The slabs behaviour was examined under static loading. The results were presented in terms of load-deflection relationship at service and ultimate load, crack pattern and failure modes. At each stage of loading, the ambient vibration excitation test has been performed to investigate the extent of damage at the cracking, yield, and ultimate states through changes in dynamic parameters obtained from RD signatures. Additional applications of RD technique were performed on two-way slabs, first, to explore the location of damage by Multi-Channel Random Decrement using FBG sensor arrays. Secondly, RD technique was utilized to evaluate the extent of damage under successive equal dynamic impacts.

The use of random decrement technique for long term health monitoring of concrete structures

This investigation deals with the development of an advanced strategy for Structural Health Monitoring (SHM) of concrete beams and girders for important structures such as bridges, offshore platforms and nuclear power stations using smart monitoring systems, including an effective diagnostic approach for damage detection with a reliability-based performance ranking. The proposed strategy can be classified into four main sections: identification of the existence of damage, determination the localization of the damage, estimation the level of damage, and assessing the seriousness of the damage regarding the structure’s service life. Random Decrement (RD) is an accurate dynamic analysis diagnostic tool that has been used effectively for SHM; the technique has been used within various fields in mechanical, aerospace, and, recently, civil engineering. The damage detection approach is based on the RD technique. A theoretical, numerical, and experimental investigation has been conducted on concrete beams using the RD technique for damage detection in terms of changes in the dynamic properties and used at successive multiple points under certain leading point conditions to determine the location of damage through the development of the Multi-Channel Random Decrement (MCRD).

The use of random decrement technique for long term health monitoring of concrete structures

This investigation deals with the development of an advanced strategy for Structural Health Monitoring (SHM) of concrete beams and girders for important structures such as bridges, offshore platforms and nuclear power stations using smart monitoring systems, including an effective diagnostic approach for damage detection with a reliability-based performance ranking. The proposed strategy can be classified into four main sections: identification of the existence of damage, determination the localization of the damage, estimation the level of damage, and assessing the seriousness of the damage regarding the structure’s service life. Random Decrement (RD) is an accurate dynamic analysis diagnostic tool that has been used effectively for SHM; the technique has been used within various fields in mechanical, aerospace, and, recently, civil engineering. The damage detection approach is based on the RD technique. A theoretical, numerical, and experimental investigation has been conducted on concrete beams using the RD technique for damage detection in terms of changes in the dynamic properties and used at successive multiple points under certain leading point conditions to determine the location of damage through the development of the Multi-Channel Random Decrement (MCRD).

Structural health monitoring of two-way slabs based on random decrement technique

The current research attempts to explore the feasible use of a Structural Health Monitoring method for a two-way slab system through the effective vibration based damage diagnostic technique of Random Decrement (RD). Experimental investigations have been conducted on a total of four reinforced concrete two-way slab specimens. The slabs behaviour was examined under static loading. The results were presented in terms of load-deflection relationship at service and ultimate load, crack pattern and failure modes. At each stage of loading, the ambient vibration excitation test has been performed to investigate the extent of damage at the cracking, yield, and ultimate states through changes in dynamic parameters obtained from RD signatures. Additional applications of RD technique were performed on two-way slabs, first, to explore the location of damage by Multi-Channel Random Decrement using FBG sensor arrays. Secondly, RD technique was utilized to evaluate the extent of damage under successive equal dynamic impacts.

Using a Single-DOF Test Vehicle to Simultaneously Retrieve the First Few Frequencies and Damping Ratios of the Bridge

Bridge damping ratios are extracted via the skillful use of the single-degree-of-freedom (DOF) test vehicle for the first time in this paper. Central to the simultaneous retrieval of the first few frequencies and damping ratios from the contact (point) response of the bridge is the use of the variational mode decomposition (VMD) and random-decrement technique (RDT). Closed-form solutions are newly derived for the vehicle and contact responses of the damped bridge and validated later numerically. Using the proposed method, one calculates first the mono-component from the contact response by the VMD; then extracts the free-decay response for each mode by the RDT; and finally identifies the frequency and damping ratio by the Hilbert transform. The parametric study confirms that: (1) the contact response outperforms vehicle’s response in retrieving bridge frequencies and damping ratios; (2) the first few frequencies can be identified with robustness for reasonable levels of road roughness, vehicle speed, bridge damping and noise; (1) good result is obtained for the first damping ratio, in spite of the traditional uncertainty existing with damping; and (2) ongoing traffic can enhance the proposed method for bridge identification.