Improving the Diagnostics of Underground Pipelines at Oil­and­gas Enterprises Based on Determining Hydrogen Exponent (PH) of the Soil Media Applying Neural Networks

摘要: A set of key parameters and information flows has been formed to simulate stages probing the outside surface underground metal pipelines (UMP) taking into account pH soil contacting with pipe metal. Specimens 17G1S steel placed in acid, alkaline neutral media were examined using a polarization potential meter complex contactless current meter. Principles application neural networks (NN) processing experimental results formulated. database developed. It meets initial conditions for controlling at boundary under real conditions. Elements optimization approach assessing coated UMP medium proposed. The is based on multiplicative qualimetric criterion quality section two groups coefficients. first group coefficients refers internal characterizes pipeline second external (i.e., electrolyte). An NN was presented "pipeline-coating" system, which: 1) is capable solving problem cluster analysis image classification; 2) makes it possible process data without their prior spectral transformation operating discrete counts signals. proposed type allows dynamically expand its own knowledge base types defects controlled objects (pipelines) operation. With help NN, assessed an three situations. above important improving methods oil-and-gas enterprise UMPs, particular, correct assessment anode density nonlinear character informative parameters.