Bearing degradation is the primary cause of electrical machine failures, making reliable condition monitoring essential to prevent breakdowns. This paper presents a novel hybrid model for the detection of multiple faults in bearings, combining Long Short-Term Memory (LSTM) networks with random forest (RF) classifiers, further enhanced by the Grey Wolf Optimization (GWO) algorithm. The proposed approach is structured in three stages: first, time and frequency domain features are manually extracted from vibration signals; second, these features are processed by a dual-layer LSTM network, which is specifically designed to capture complex temporal relationships within the data; finally, the GWO algorithm is employed to optimize feature selection from the LSTM outputs, feeding the most relevant features into the RF classifier for fault classification. The model was rigorously evaluated using a dataset comprising six distinct bearing health conditions: healthy, outer race fault, ball fault, inner race fault, compounded fault, and generalized degradation. The hybrid LSTM-RF-GWO model achieved a remarkable classification accuracy of 98.97%, significantly outperforming standalone models such as LSTM (93.56%) and RF (98.44%). Furthermore, the inclusion of GWO led to an additional accuracy improvement of 0.39% compared to the hybrid LSTM-RF model without optimization. Other performance metrics, including precision, kappa coefficient, false negative rate (FNR), and false positive rate (FPR), were also improved, with precision reaching 99.28% and the kappa coefficient achieving 99.13%. The FNR and FPR were reduced to 0.0071 and 0.0015, respectively, underscoring the model's effectiveness in minimizing misclassifications. The experimental results demonstrate that the proposed hybrid LSTM-RF-GWO framework not only enhances fault detection accuracy but also provides a robust solution for distinguishing between closely related fault conditions, making it a valuable tool for predictive maintenance in industrial applications.

• Klíčová slova
• Bearing fault detection, Feature selection, Grey wolf optimization, Hybrid model, LSTM, Machine learning, Random forest, Vibration signals,
• Publikační typ
• časopisecké články MeSH

Many experiments have demonstrated that some cell lines are resistant to chemically induced apoptosis in vitro, and that apoptosis itself is far from being a homogenous phenomenon. Here we show that 10 microg/ml etoposide elicited only minor changes in Bowes human melanoma cells (temporary decrease in cell viability and proliferation, transient phospatidylserine externalization and caspase-3 activation), which weren't clearly capable to start apoptotic pathway in the entire treated population. On the other hand, potassium chromate at concentration of 150 microg/ml executed cell death bearing some features of apoptosis (cell blebbing, caspase-3 activation and cytoskeletal changes) but lacking or showing weakly others (DNA fragmentation and phospatidylserine externalization). Our results suggest that in detecting apoptosis several fault-proof detection systems are to be used to avoid misleading results and conclusions in each experimental setting.

• MeSH
• aktiny metabolismus   MeSH
• annexin A5 analýza   MeSH
• apoptóza * účinky léků   MeSH
• biologické markery analýza   MeSH
• buněčné dělení MeSH
• chromany farmakologie   MeSH
• DNA nádorová analýza   účinky léků   MeSH
• etoposid farmakologie   MeSH
• fluorescein-5-isothiokyanát MeSH
• fluorescenční barviva MeSH
• fosfatidylseriny analýza   MeSH
• fragmentace DNA účinky léků   MeSH
• kaspasa 3 MeSH
• kaspasy metabolismus   MeSH
• lidé MeSH
• melanom experimentální chemie   patologie   MeSH
• nádorové buňky kultivované chemie   patologie   MeSH
• sloučeniny draslíku farmakologie   MeSH
• videomikroskopie MeSH
• Check Tag
• lidé MeSH
• Publikační typ
• časopisecké články MeSH
• práce podpořená grantem MeSH
• Názvy látek
• aktiny MeSH
• annexin A5 MeSH
• biologické markery MeSH
• CASP3 protein, human MeSH Prohlížeč
• chromany MeSH
• DNA nádorová MeSH
• etoposid MeSH
• fluorescein-5-isothiokyanát MeSH
• fluorescenční barviva MeSH
• fosfatidylseriny MeSH
• kaspasa 3 MeSH
• kaspasy MeSH
• potassium chromate(VI) MeSH Prohlížeč
• sloučeniny draslíku MeSH

The monitoring of impeller blade vibrations is an important task in the diagnosis of turbomachinery, especially in terms of steam turbines. Early detection of potential faults is the key to avoid the risk of turbine unexpected outages and to minimize profit loss. One of the ways to achieve this is long-term monitoring. However, existing monitoring systems for impeller blade long-term monitoring are quite expensive and also require special sensors to be installed. It is even common that the impeller blades are not monitored at all. In recent years, the authors of this paper developed a new method of impeller blade monitoring that is based on relative shaft vibration signal measurement and analysis. In this case, sensors that are already standardly installed in the bearing pedestal are used. This is a significant change in the accessibility of blade monitoring for a steam turbine operator in terms of expenditures. This article describes the developed algorithm for the relative shaft vibration signal analysis that is designed to run in a long-term perspective as a part of a remote monitoring system to track the natural blade frequency and its amplitude automatically.

• Klíčová slova
• algorithm, diagnostics, impeller blade, monitoring, signal processing, steam turbine, vibration,
• MeSH
• algoritmy MeSH
• chirurgické nástroje MeSH
• pára * MeSH
• počítačové zpracování signálu MeSH
• vibrace * MeSH
• Publikační typ
• časopisecké články MeSH
• Názvy látek
• pára * MeSH