Autoencoder-based anomaly detector for gear tooth bending fatigue cracks

The article reports on anomaly detection performance of data-driven models based on a few selected autoencoder topologies and compares them to the performance of a set of popular classical vibration-based condition indicators. The evaluation of these models employed data that consisted of baseline gearbox runs and the associated runs with seeded bending cracks in the root of the gear teeth for eight different gear pairings. The analyses showed that the data-driven models, trained on a subset of baseline data outperformed classical CIs as anomaly detectors.

Characteristics of the Donkey’s Dorsal Profile in Relation to Its Functional Body Condition Assessment

As the breeding of donkeys has increased due to different types of use, welfare evaluation importance increases. This equid’s welfare state has been described using body condition indicators and the geometric morphometrics method. However, the dorsal profile has not yet been assessed in donkeys. In this study, the body condition score (BCS), fatty neck score (FNS), dental condition score (DCS), sex, and breed were used as criteria of dorsal profile deformations. Photographs of 40 donkeys were analyzed using geometric morphometrics. Within the entire set of dorsal profiles, the variance of the first three principal components (PCs) was PC1 = 37.41%, PC2 = 23.43%, and PC3 = 13.34%. The dorsal profiles displayed deformation as an effect of FNS and BCS on size (FNS p = 0.012; BCS p = 0.024) and shape (FNS p < 0.0001; BCS p < 0.0001), rather than as an effect of DCS (p < 0.0001), sex (p = 0.0264), and breed (p < 0.0001) only on shape. The highest distances among the categories (Mahalanobis distances: MD ≥ 13.26; Procrustes distances: PD ≥ 0.044) were noted for FNS. The lowest distances were noted between jennets and males (MD = 4.58; PD = 0.012) and between BCS 1 and BCS 2 (MD = 4.70; PD = 0.018). Donkeys’ body condition affects their dorsal profile and both FNS and BCS measurements should be considered when a donkey’s dorsal profile is investigated.

Dynamics of general physical fitness indicators in the system of physical therapy of 11–13-year-old women weightlifters after traumatic injuries of the musculoskeletal system

The aim of the work to determine the dynamic characteristics of overall physical condition indicators in 11–13-year-old female weightlifters after injuries of the musculoskeletal system at the training stage of physical therapy. Materials and methods. The experimental population comprised 45 female weightlifters (aged 11–13 years, qualification ‒ I, II, III junior categories) from SСYSSOR “Spartak” of the Zaporizhzhia regional council, CYSS “Kolos” of Kamiansko-Dniprovskyi district of Zaporizhzhia region, Berdiansk CYSS of Zaporizhzhia Region City Council, and students of Kharkiv Regional Higher School of Physical Culture and Sports majoring in weightlifting. Patients after musculoskeletal injuries were divided into two groups: main (MG, n = 22) and control (CG, n = 23). Patients of the control group underwent standard treatment (generally prescribed in medical establishments). For main group patients in the training period, the program of physical therapy has been developed and implemented, with the current control of overall physical condition indicators. Research methods. Analysis of literature and sources of information presented on the Internet; clinical observations; pedagogical testing; methods of mathematical statistics. Results. Analysis of overall physical condition indicators of athletes of 11–13-year-old weightlifters of the main and control groups during the recovery and training periods allows us to state that main group athletes have higher growth rates, and dynamic characteristics of their overall physical condition are linear with constant focus on improvement. According to the indicators of “running 30 m, s” test, the following growth indicators were recorded: main group athletes (-0.48), control group athletes (-0.29), the difference (-0.19). According to the indicators of “standing long jump, cm” test, the following growth indicators were recorded: main group athletes (+14), control group athletes (+5), difference (+9). According to the results of the test “flexion-extension of the arms in the supine position, n” the following indicators were obtained: main group athletes (+4.44), control group athletes (+2.86), difference (+1.58). Conclusions. Intensification of the training process and forcing the training of junior athletes in weightlifting are the leading factors that result in injuries of the musculoskeletal system, multisystem pathologies and the development of pathomorphic phenomena. Indicators of overall physical condition of weightlifters aged 11–13 allow us to state that main group athletes have a higher difference in growth, and the tendency of their overall physical condition has a linear focus on improvement. Indicators of overall physical condition of control group athletes are nonlinear, according to “running 30 m, s” and “standing long jump, cm” the test results.

Improved damage detection in Pelton turbines using optimized condition indicators and data-driven techniques

The health condition of hydraulic turbines is one of the most critical factors for the operation safety and financial benefits of a hydro power plant. After the massive entrance of intermittent renewable energies, hydropower units have to regulate their output much more frequently for the balancing of the power grid. Under these conditions, the components of the machine have to withstand harsher excitation forces, which are more likely to produce damage and eventual failure in the turbines. To ensure the reliability of these machines, improved condition monitoring techniques are increasingly demanded. In this article, the feasibility of upgrading condition monitoring of Pelton turbines using novel vibration indicators and data-driven techniques is discussed. The new indicators are selected after performing a detailed analysis of the dynamic behavior of the turbine using numerical models and field measurements. After that, factor analysis is carried out in order to assess which are the most informative indicators and to reduce the dimension of the input data. For the validation of the proposed method, monitoring data from an actual Pelton turbine that suffered from an important fatigue failure due to a crack propagation on the buckets have been used. The novel condition indicators as well as classical indicators based on the spectrum and harmonics levels have been obtained while the machine was in good operation, during different stages of damage and after repair. All of these have been used to train an artificial neural network model in order to predict the evolution of the crack until failure occurs. The results show that using the improved monitoring methodology enhances the ability to predict the appearance of damage in comparison to typical condition indicators.