E.M. Farhadzadeh, A.Z. Muradaliyev, T.K. Rafiyeva, A.A. Rustamova
Improving the system for the management of the efficiency of power engineering devices
A new method and algorithm for ensuring the homogeneity of standardized samples of technical and economic indicators of power units of thermal power plants is presented. Homogeneity and normalization are prerequisites for evaluating the integrated indicators characterizing the efficiency of power units. The method is based on a fiducial approach. The boundary values of the fiducial interval are traditionally calculated by the statistical distribution function and a given significance level. I.e. they are basically calculated "mechanically." Since the “mechanical” approach is valid for homogeneous statistical data, and technical and economic indicators are multidimensional data, the application of this approach to the statistical function of the fiducial distribution is associated with a high risk of an erroneous decision. Many possible implementations of the actual values of technical and economic indicators have implementations due to "gross" errors when entering data into automated systems or when performing individual calculations manually. Non-typical implementations are often found, for instance, when working with a light load for 10 days of the month. This data forms boundary intervals and are named boundins by the authors. Automated search and removal of boundins provides reliable comparison and ranking of integrated indicators. It is shown that the rate of variation of boundins is significantly less than the rate of variation of typical implementations of technical and economic indicators. This fact became the basis for the recognition of boundins.
Keywords: Reliability, Safety, Efficiency, Fiducial interval, Boundin, Methods, Algorithms
Improving the system for the management of the efficiency of power engineering devices
A new method and algorithm for ensuring the homogeneity of standardized samples of technical and economic indicators of power units of thermal power plants is presented. Homogeneity and normalization are prerequisites for evaluating the integrated indicators characterizing the efficiency of power units. The method is based on a fiducial approach. The boundary values of the fiducial interval are traditionally calculated by the statistical distribution function and a given significance level. I.e. they are basically calculated "mechanically." Since the “mechanical” approach is valid for homogeneous statistical data, and technical and economic indicators are multidimensional data, the application of this approach to the statistical function of the fiducial distribution is associated with a high risk of an erroneous decision. Many possible implementations of the actual values of technical and economic indicators have implementations due to "gross" errors when entering data into automated systems or when performing individual calculations manually. Non-typical implementations are often found, for instance, when working with a light load for 10 days of the month. This data forms boundary intervals and are named boundins by the authors. Automated search and removal of boundins provides reliable comparison and ranking of integrated indicators. It is shown that the rate of variation of boundins is significantly less than the rate of variation of typical implementations of technical and economic indicators. This fact became the basis for the recognition of boundins.
Keywords: Reliability, Safety, Efficiency, Fiducial interval, Boundin, Methods, Algorithms