The determination of structural dynamic characteristics can be challenging, especially for complex cases. This can be a major impediment for dynamic load identification in many engineering applications. Hence, avoiding the need to find numerous solutions for structural dynamic characteristics can significantly simplify dynamic load identification. To achieve this, we rely on machine learning. The recent developments in machine learning have fundamentally changed the way we approach problems in numerous fields. Machine learning models can be more easily established to solve inverse problems compared to standard approaches. Here, we propose a novel method for dynamic load identification, exploiting deep learning. The proposed algorithm is a time-domain solution for beam structures based on the recurrent neural network theory and the long short-term memory. A deep learning model, which contains one bidirectional long short-term memory layer, one long short-term memory layer and two full connection layers, is constructed to identify the typical dynamic loads of a simply supported beam. The dynamic inverse model based on the proposed algorithm is then used to identify a sinusoidal, an impulsive and a random excitation. The accuracy, the robustness and the adaptability of the model are analyzed. Moreover, the effects of different architectures and hyperparameters on the identification results are evaluated. We show that the model can identify multi-points excitations well. Ultimately, the impact of the number and the position of the measuring points is discussed, and it is confirmed that the identification errors are not sensitive to the layout of the measuring points. All the presented results indicate the advantages of the proposed method, which can be beneficial for many applications.

We introduce the augmented Tikhonov regularization method motivated by Bayesian principle to improve the load identification accuracy in seriously ill-posed problems. Firstly, the Green kernel function of a structural dynamic response is established; then, the unknown external loads are identified. In order to reduce the identification error, the augmented Tikhonov regularization method is combined with the Green kernel function. It should be also noted that we propose a novel algorithm to determine the initial values of the regularization parameters. The initial value is selected by finding a local minimum value of the slope of the residual norm. To verify the effectiveness and the accuracy of the proposed method, three experiments are performed, and then the proposed algorithm is used to reproduce the experimental results numerically. Numerical comparisons with the standard Tikhonov regularization method show the advantages of the proposed method. Furthermore, the presented results show clear advantages when dealing with ill-posedness of the problem.

Evaluating dynamic loads in real time is crucial for health monitoring, fault diagnosis and fatigue analysis in aerospace, automotive and earthquake engineering among other vibration related applications. Developing such algorithms can be vital for several safety and performance functionalities. Therefore, over the past few years the identification of dynamic loads has attracted a lot of attention; however, little literature on the online identification can be found. In this paper, we propose an online-identification method of structural dynamic loads so that the dynamic load is evaluated in real time and while the system response is still being measured. This is achieved by significantly improving the identification efficiency while retaining a high accuracy. The proposed method which is based on Kalman filter, is introduced in detail for a finite as well as an infinite number of degrees of freedom. Starting from an initial guess of the state vector we evaluate the error covariance, which then helps to identify the value of the excitation force using a weighted least square method and minimizing the covariance unbiased estimation. This is repeated at certain time intervals i.e., time steps where the state vector is updated in real time as acceleration measurements are updated. The feasibility of the method is validated using numerical simulations and an experimental verification where a detailed LabVIEW (National Instruments Ltd.) implementation is provided.