TY  - JOUR
PY  - 2021//
TI  - Neural-network prediction of damage to accident-prone sections of oil and gas pipelines
JO  - Russian engineering research
A1  - Burdina, A.A.
A1  - Nekhrest-Bobkova, A.A.
SP  - 775
EP  - 778
VL  - 41
IS  - 8
N2  - The prediction of possible damage due to accidents on linear sections of oil and gas pipelines is considered. In the proposed mathematical model, the total economic loss includes components associated with damage to the pipeline itself, gas or oil leaks, and environmental pollution. A fully connected neural network is used to calculate the probability of an accident due to equipment wear, while a fully convolutional neural network is used to calculate the probability of an accident due to external mechanical factors. The result is a mechanism for predicting the possible economic loss on the basis of neural networks. © 2021, Allerton Press, published by Springer Nature    Keywords: Pipeline transportation<p />  <p>Language: en</p>
LA  - en
SN  - 1068-798X
UR  - http://dx.doi.org/10.3103/S1068798X21080104
ID  - ref1
ER  -