The algorithm for diagnosing leaks of the target product under conditions of uncertainty for the hydraulic system

Abstract

An algorithm for determining the presence or absence of leaks in functioning hydraulic systems is considered, based on existing methods for determining the location and volume of such leaks. Based on the collected data on the hydraulic parameters of such systems, it is proposed to create a statistical model with all the statistical parameters required for performing calculations. The emphasis is placed on the fact that it is necessary to take into account the features of the hydraulic system in question, depending on its purpose and the technologies used. Mathematical models and methods of leak diagnostics in hydraulic systems are considered, including the identification of the facts of the existence of a leak based on the use of mathematical models to determine the location and size of such leaks. In this article, the research focuses on the detection of leaks with unknown amplitudes based on the verification of two alternative hypotheses for a hydraulic system, taking into account interference from the stochasticity of the demand. It is noted that the transition from the task of verifying two alternative hypotheses to the task of verifying multiple alternative hypotheses or pattern recognition provides a deeper and more comprehensive analysis of leaks. This will allow for more accurate and efficient detection of unusual leaks and provide more reliable protection of hydraulic systems.