| Popis: |
results of evaluation and forecasting of reliability indicators of technological equipment are presented on the basis of a neural network model for calculating fault-free operation indicators taking into account safety indicators of the generating system. In the process of creating a neural network, a topology was developed, a mechanism for training and testing the model was determined. During the study, a sample of input data was created, an algorithm for functioning and data exchange was built. The analysis of statistical data on the joint influence of a number of parameters of technical condition of resource-determining functional units on their index of technical condition, on the general technical condition of the hydraulic unit, which showed that the most confirmed are the joint effect of increasing the pressure difference in the servomotor cavities and increasing the pressure difference in the servomotor cavitation erosion of the blades in the flow part of the hydraulic unit. Multivariate data bases formed by the neural network model, combining the effect of blade turning on the change in the probability of hydraulic power equipment failure, impact of shaft combat on the index of technical condition of the unit “turbine bearing and shaft,” influence of shaft vibration change on index of technical condition of hydraulic power equipment and turbine cover, and taking into account the park life, inter-repair period, accident rate, metal condition should be used to create prognostic models that provide approaches to extending the operable state of elements of hydropower equipment. Obtained results can be used in creation of modern systems of monitoring and diagnostics of power equipment, its separate elements required for collection, storage, archiving of data taking into account actual state of specific element of hydraulic unit for qualitative prediction of indicators of reliability of hydropower equipment in order to increase efficiency of operation of generating and power systems. |
