A Recurrent Neural Network-Based Method for Dynamic Load Identification of Beam Structures

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.

Vibration behavior of an overhead conductor under updraft winds

At present, the wind-induced response analysis of an overhead conductor is mainly based on the action of horizontal normal wind. However, for crossing hillsides or extremely strong winds, such a conductor will bear the action of updraft wind, which will change the geometry of the conductor and make its structural dynamic characteristics nonlinear to some extent. In this work, the in-plane and out-of-plane two-dimensional nonlinear equations were established under the action of self-weight and updraft wind. Furthermore, the improved equations of conductor tension and sag were obtained, and the wind-induced vibration response was further investigated. The results showed that the updraft wind caused the nonlinearity of the tension and sag of the overhead conductor, and the nonlinear geometric change significantly affected its resonance response, which exceeded 25% if the wind speed was 50 m/s. In addition, because the proportion of the resonance response in the total wind-induced response was different, the influence of the wind attack angle calculated using the gust response factor method on the gust response factor was slightly larger than that calculated using the the American society of civil engineers method.

“Illness Representations in COVID-19”: Somatoperception During a Pandemic

Addressing the psychosomatic aspects of the pandemic gives rise to a wide range of questions. National experience in addressing these issues is analyzed here in the light of classical concepts of somatoperception and body experience recognition. The results, obtained from following COVID-19 patients samples have been taken into consideration: affected patients, recovered patients and those who have not come across this disease directly. To describe somatoperceptual characteristics in a pandemic, the term “Illness representations in COVID-19” is proposed. During the comparison of the degree of study in different levels, it was found that the sensitive level has not yet received sufficient lighting, unlike the other levels. The analysis of the literature data led to the identification of future research directions of illness representations in COVID-19. In addition to the features of the sensitive level, interlevel mutual influence and structural-dynamic characteristics of illness representations in COVID-19 depending on the person’s own experience in dealing with COVID-19 may be the subject of such studies.

Flutter Stability of a Long-Span Suspension Bridge During Erection in Mountainous Areas

In mountainous areas, more challenges are expected for the construction of long-span bridges. The flutter instability during erection is an outstanding issue due to flexible structural characteristics and strong winds with large angles of attack. Taking the suspension bridge as an example, the flutter stability of the bridge with different suspending sequences was investigated. First, the dynamic characteristics of the bridge during erection were computed by the finite element software ANSYS, along with the effects on flutter stability discussed. Then, different aerodynamic shapes of the bridge girder during erection were considered. The aerodynamic coefficients and the critical flutter state were determined by wind tunnel tests. Based on the above analysis, some structural measures are proposed for improving the flutter stability of the bridge during erection. The results show that the flutter stability of the bridge during erection is related to the suspending sequence and the aerodynamic shape of the girder. Owing to the structural dynamic characteristics, the bridge has better flutter stability when the girder segments are suspended symmetrically from the two towers to the mid-span. Considering the construction requirement that the bridge deck should be laid without intervals, this structural superiority is seriously weakened by the unfavorable aerodynamic shape of the girder. In order to improve the flutter stability of the bridge during erection, an effective way is to adopt some temporary structural strengthening measures.

High-vibration diagnosis of gas turbines: An experimental investigation

Rotordynamics is a very challenging field because of machine complexities. Many internal and external factors contribute toward change in the structural dynamic characteristics. One of these factors is broad-band high-vibration amplitudes. In this article, a similar high vibration issue on a gas turbine is investigated using bode, orbit, and shaft centerline plots. Data from proximity probes installed on turbine generator system are captured and analyzed against any factor contributing toward high vibration issue. Fish bone diagram was used for root cause investigation. Main components investigated for the root cause of high vibration issue include generator rotor and casing. Rotor behavior has been examined by capturing orbit and shaft centerline diagrams, whereas casing contribution has been investigated by conducting operating deflection shape analysis. A comparison is drawn between a machine suffering from high vibration issue and a normal machine. Resonance was identified as the root cause, and stiffness enhancement was recommended to change the natural frequency of casing. Based on investigations, recommendations are given and a final comparison is drawn after structural modification was done. In addition to early fault finding, reduction in maximum vibration was 38% after implementation of the fix that confirmed the accuracy of the root cause investigation process.