Abstract
The study was conducted to identify the possibilities of implementing neural network technologies to optimise the control of the operating modes of the sucker-rod pump (SRP), which will help to increase the efficiency of oil production and reduce operating costs. The study used data analysis and adaptive management methods to optimise the operation of the SRP. As a result of the study, it was found that the introduction of neural network technologies in the SRP management system can significantly increase their efficiency. Analysis of data from the unit’s sensors using neural networks helped to identify optimal operating modes that ensure maximum production with minimal energy consumption. A forecasting model has been developed that can detect potential equipment failures in advance, which reduces the risks of emergencies and maintenance costs. The study also showed that adaptive control algorithms based on artificial intelligence can automatically adjust the operating modes of the SRP depending on variable conditions, such as pressure fluctuations or changes in the properties of the oil produced. Based on the integration with the Internet of Things, the system has the ability to perform real-time monitoring, which increases the efficiency of decision-making. As a result, the introduction of neural network technologies not only optimises mining processes, but also helps to reduce operating costs. In addition, the study revealed that the use of neural networks in control systems can significantly reduce the time required to configure and optimise processes, which increases the overall productivity of the SRP