Damage detection based on modal parameter changes has become popular in the last few decades. Nowadays, there are robust and reliable mathematical relations available to predict natural frequency changes if damage parameters are known. Using these relations, it is possible to create databases containing a large variety of damage scenarios. Damage can be thus assessed by applying an inverse method. The problem is the complexity of the database, especially for structures with more cracks. In this paper, we propose two machine learning methods, namely the random forest (RF), and the artificial neural network (ANN), as search tools. The databases we developed contain damage scenarios for a prismatic cantilever beam with one crack and ideal and non-ideal boundary conditions. The crack assessment was made in two steps. First, a coarse damage location was found from the networks trained for scenarios comprising the whole beam. Afterwards, the assessment was made involving a particular network trained for the segment of the beam on which the crack was previously found. Using the two machine learning methods, we succeeded in estimating the crack location and severity with high accuracy for both simulation and laboratory experiments. Regarding the location of the crack, which was the main goal of the practitioners, the errors were less than 0.6%. Based on these achievements, we concluded that the damage assessment we propose, in conjunction with the machine learning methods, is robust and reliable.

Fall risk assessment is becoming an important concern, with the realization that falls, and more importantly fainting occurrences, in most cases require immediate medical attention and can pose huge health risks, as well as financial and social burdens. The development of an accurate inertial sensor-based fall risk assessment tool combined with machine learning algorithms could significantly advance healthcare. This research aims to investigate the development of a machine learning approach for falling and fainting detection, using wearable sensors with an emphasis on forward falls. In the current paper we address the problem of the lack of inertial time-series data to differentiate the forward fall event from normal activities, which are difficult to obtain from real subjects. To solve this problem, we proposed a forward dynamics method to generate necessary training data using the OpenSim software, version 4.5. To develop a model as close to the real world as possible, anthropometric data taken from the literature was used. The raw X and Y axes acceleration data was generated using OpenSim software, and ML fall prediction methods were trained. The machine learning (ML) accuracy was validated by testing with data acquired from six unique volunteers, considering the forward fall type.

In this paper, the main objective is to present the current status of the automotive specifications related to the crimping process, an in particular how these specifications were adopted by the suppliers of the crimping components (e.g. Terminals and cables). For this reason, we took as reference two of the most worldwide used standards form the automotive manufacturing industry: DIN (“Deutsches Institut für Normung”) with the corresponding Romanian standard SR EN 60352-2:2006 and SAE/USCAR-21 REVISION 3.

Structural health monitoring plays a crucial role in ensuring the integrity and safety of engineering structures such as steel beams. This research paper presents a comprehensive methodology for detecting transverse cracks in beams with a constant section and any boundary conditions. The proposed approach utilizes the normalized squared modal curvature of the beam, the damage severity, and the natural frequency of the undamaged beam. By analyzing the natural frequencies of both the undamaged and damaged states, Relative Frequency Shift (RFS) values are obtained. Subsequently, the Damage Location Coefficients (DLC) are calculated by normalizing the RFS values. These DLC values are then employed to establish a comprehensive database of known damage signatures, enabling the training of an artificial neural network (ANN) in MATLAB. The trained ANN can predict the locations of damages for new scenarios by utilizing DLC values obtained from measurements. To validate the effectiveness of the ANN, extensive simulations using Finite Element Method (FEM) and experimental measurements are conducted on a steel cantilever beam. The results demonstrate the ANN’s capability to accurately predict the locations of transverse cracks, showcasing its potential as a reliable tool for structural health monitoring of steel beams.

In this paper, we introduce a mathematical relation which allows prediction of frequency changes of beams if a damage producing an abrupt stiffness decrease occurs in the structure. This relation is contrived from the energies stored in the beam in the intact, respectively in the damaged state. First, the analysis was performed for the damage reducing the stiffness of the slice subjected to the biggest bending moment. From this analysis, we found the damage severity evolution, reflected in a frequency drop, with the damage depth. Afterward, we replaced this damage on several locations along the beam and quantified the effect upon the natural frequencies for each of them. Based on this analysis, we found out also the effect of damage position upon the natural frequencies. The final relation resulted by aggregating the effect of the severity with that of the position. To prove the reliability of this relation, we tested it against results obtained by means of the finite element method.

Real beams have non-ideal boundary conditions and it is necessary to use new models to determine the real modal parameters. Models that use ideal conditions do not fully reflect reality and can lead to unsatisfactory description of the dynamic behavior. The hinged – hinged boundary conditions, which is in the focus of the paper, are not analyzed as a single beam, but as a continuous beam with three spans, free at the ends. The continuous beam with three spans is analyzed for cases in which the intermediate supports can occupy any position along the length of the beam, by an analytical solution of the problem, with the example of cases when the intermediate supports are located very close at the free ends of the continuous beam, thus simulating the real case for an hinged beam at both ends; the situation in which the intermediate supports are very close to one of the ends of the beam, thus simulating the real case of the clamped beam, with an imperfect clamped end; and the situation in which the intermediate supports are very close located anywhere on the beam length, thus simulating the hypothetic case with a continuous beam free at the ends and fix on the hinged supports. The analytic results are compared with numerical results by using finite elements method.