Application of Linear Random Processes to Construction of Diagnostic System for Power Engineering Equipment
Résumé
Some mathematical models of informational signals and methods of the diagnostic system construction is considered. Different type of linear random processes is recommended for application. Some properties of the random processes are discussed. Application of the models depends on many factors. Such processes are used for classification of stochastic informational signals in the case of non-Gaussian distribution realizations of the signals, which energy spectra (within the framework of the first two moments) possess is often not enough for reliable recognition and classification of such signals. It is expedient to use information on higher moments or statistical characteristics of distributing such signals. It is possible to use the higher-order moments of such signals realization as the diagnostic features and provided the increase in accuracy and reliability of Power Engineering Equipment diagnostics. Attributes at the power engineering equipment diagnostic system construction are above mentioned justified diagnostic features, as well as informational diagnostic signals having been justified by the investigation of the appropriate mathematical models. Obtaining numerical estimation of chosen diagnostic features is first of all necessary for forming teaching complexes, and, further, for conducting the diagnostics itself according to the formed teaching complexes. Some peculiarities of constructing the multilevel diagnostic systems for resilience operation of electric power facilities are considered. An example of 3-level diagnostic system development for a Power station is represented.
Origine | Accord explicite pour ce dépôt |
---|