PSI - Issue 28

Evgeniia Georgievskaia et al. / Procedia Structural Integrity 28 (2020) 836–842 Evgeniia Georgievskaia/ Structural Integrity Procedia 00 (2020) 000–000

840

Fig. 3. Forecasting technology of the proposed predictive analytics system

The proposed solution transfers all hardship to the development of the individual module that collects information about the stress state of the unit’s components under every operating mode in a complete working range. These data are extracted from the results of multivariate multidisciplinary simulations for the digital model of the hydraulic turbine. The individual module is an n-dimensional matrix of equipment’s response to external influences at different operating points. It is obvious that the individual module will be significantly different for different units. The primary benefits of the proposed solution are: • it is suitable for all types and constructions of hydraulic turbines • it takes into account all individual design features and manufacturing techniques • it takes into account the actual and expected operating modes • it decides the task quickly and cost-effective, without heavy computer calculations and high computer resources • it allows saving money due to transition from scheduled maintenance to condition-based maintenance • simplicity and rapidity for users of PAS (the user of PAS can be any technical specialist of HPP or employee of managing company not only specialist on computer-aided engineering in hydropower). 5. The results The capabilities of the proposed predictive analytics approach as a way to smart maintenance of hydraulic turbines can be demonstrated by some examples. The solutions presented below were obtained using the proposed predictive