Let's consider mathematical models and methods for diagnosing leaks in hydraulic pipeline systems. As part of the task, two subtasks can be distinguished: identifying the fact of a leak based on the use of a two-alternative hypothesis, as well as mathematical models for determining the location and size of the leak. The task of determining the location and size of the leak has been investigated in sufficient detail. Attention in this work is paid to the development of algorithms for diagnosing hydraulic systems for the presence or absence of leakage. It is proposed to solve both problems as a complex problem of leak diagnostics for hydraulic pipeline systems. The software created based on the results of the research can be used for automated control and monitoring of hydraulic systems. Based on the collected data on the parameters of the hydraulic system, such as pressure, consumption of the target product, etc., a model can be created that can include probability distributions, expected values for parameter variations. The solution to the specific problem of leak detection may vary depending on the type of hydraulic system. The main focus of this work is on leak detection based on a two-alternative hypothesis to determine the existence of a leak in a hydraulic system, taking into account the noise generated by the random selection of the target product by consumption. Leaks are classified as natural, resulting from emergency situations, and leaks also include unsanctioned selections of the target product.
Read full abstract