Abstract. This article considers ways of monitoring and predicting the residual life of pumps used in irrigation systems. The relevance of this research is due to the need to improve the reliability and efficiency of pumping equipment, reduce the probability of unexpected failures, and optimise maintenance costs.
This study aims to develop and apply a high accuracy method for predicting the remaining useful life (RUL) of an irrigation system. The method is based on the use of convolutional neural network (CNN).
Research Objectives:
Study of existing methods for monitoring and predicting the technical condition of pumping equipment.
Development of convolutional neural network (CNN) architecture with two parallel channels of data processing: temporal and time-frequency.
Creation of the software complex ‘AITUM’ for collection, processing and analysis of data coming from vibration and temperature sensors.
Development of recommendations on application of the method in pumping equipment maintenance systems.
The constructed method on the basis of deep learning shows high accuracy of RUL prediction of pumps, which significantly exceeds the generally accepted approaches. A series of experiments on real data of pumping systems were conducted, which helped to discover the degradation patterns of the equipment. Application of this technology will reduce the risk of emergency failures, reduce operating costs and increase the efficiency of irrigation systems due to timely maintenance of equipment.

Keywords. residual life prediction (RUL), IoT sensors, CNN, pump units of irrigation systems, data mining algorithm.