AP25794391 – Research and implementation of Machine Learning methods for monitoring the condition, diagnostics and predictive analytics of the oil producing wells

Objective of the project – Develop an intelligent system based on Machine Learning methods, which will allow monitoring the technical condition of oilwell equipment, accurately diagnosing potential faults and predicting their occurrence. The objective of the project is to create tools for early detection of potential well faults.

Relevance: The oil and gas industry is constantly evolving, generating large volumes of data for optimizing production processes. In the context of digital transformation, big data analysis has become a key factor in increasing enterprise efficiency. The use of artificial intelligence and Machine Learning methods enables more accurate and timely decision-making. Forecasting models help reduce equipment downtime risks and improve asset management. Therefore, the implementation of intelligent analytical systems in the oil and gas sector remains highly relevant.

Scientific supervisor:  Daur Akkonusovich Aktaukenov

Expected and achieved results: Within the project, a Machine Learning model was developed and trained for early detection of downhole equipment failures. The system allows timely planning of repair work, increases the reliability of well operation, and optimizes maintenance processes. The intelligent recommendation system is based on historical data and modern algorithms, enabling accurate prediction of equipment failure probability. The socioeconomic impact includes creating new jobs and increasing tax revenues. The environmental aspect reduces negative impacts on the environment. The scientific and technical novelty lies in the implementation of adaptive predictive models and artificial intelligence in the oil and gas sector. The international significance is that the results are of interest to the global scientific community and can be applied in other countries.