In recent years, more and more attention has been paid to the use of artificial neural networks (ANN) for the diagnostics of gas pumping units (GPU). Usually, ANN training is carried out on GPU workflow models, and generated sets of diagnostic data are used to simulate defect conditions. At the same time, the results obtained do not allow assessing the real state of the GPU. It is proposed to use the characteristics of the acoustic and vibration processes of the GPU as the input data of the ANN. A descriptive statistical analysis of real vibration and acoustic processes generated by the operation of the GPU type GTK-25-i (Nuovo Pignone, Italy) was carried out. The formation of packets of diagnostic features arriving at the input of the ANN was carried out. Diagnostic features are the five maximum amplitude components of the acoustic and vibration signals, as well as the value of the standard deviation for each sample. Diagnostic features are calculated directly in the ANN input data pipeline in real time for three technical states of the GPU. Using the frameworks TensorFlow, Keras, NumPy, pandas, in the Python 3 programming language, an architecture was developed for a deep fully connected feedforward ANN, trained on the backpropagation algorithm. The results of training and testing the developed ANN are presented. During testing, it was found that the signal classification precision for the "nominal" state of all 1,475 signal samples is 1,0000, for the "current" state, precision equals 0,9853, and for the "defective" state, precision is 0,9091. The use of the developed ANN makes it possible to classify the technical states of the GPU with an accuracy sufficient for practical use, which will prevent the occurrence of GPU failures. ANN can be used to diagnose GPU of any type and power.

Modern industrial technologies require raw materials of high purity from suppliers, especially due to the possible presence of inclusions of heterogeneous metals in them. The existence of metallic inclusions of heterogeneous metals in the raw materials leads to equipment failure, a decrease in the quality of output products, and, consequently, large financial losses. Since the main method of transporting raw materials in the industry is still a conveyor belt, this imposes additional conditions to control and remove metallic inclusions. For various reasons, current methods for removing metallic inclusions in the conveyor belt do not fully meet the needs of modern production. The main issue related to existing removal systems is the lack of intelligent interaction between these systems and the absence of information exchange between systems that detect and remove metallic inclusions. An alternative method for removing metallic inclusions of heterogeneous metals from loose medium has been proposed, which implies the tandem operation of the system that detects metallic inclusions and the system that removes them. The tandem operation of the two systems makes it possible to exchange information about a detected metallic inclusion and, as a result, to more flexibly use tools for the removal of metallic inclusion depending on the size and location of the metallic inclusion relative to the conveyor belt axis. At the same time, the control unit of the removal system makes it possible to control the conveyor belt itself, which allows the removal of complex metallic inclusions using the reverse of the electric drive of the belt, as well as enables a control check of the fact of removal. The developed algorithm of the removal system was implemented in the programming environment TIA-Portal. The introduction of this removal system could reduce the number of metallic inclusions in raw materials by 15 - 20 %; moreover, its application is not limited to only one sector of the national economy.