Forming the toolset for development of a system to control quality of operation of underground pipelines by oil and gas enterprises with the use of neural networks

摘要: A set of defining parameters for modeling stages a surface defect propagation in the outer metal pipeline taking into consideration fatigue strength has been formed. For section with defect, it was proposed to use an algorithm forecasting polarization potentials using means neural networks. procedure functioning testing elaborated estimating efficiency The includes appropriate training methods. According results analysis interconnected deformation and corrosion processes, elements methodology formation information support service life linear part underground pipelines have developed. Known estimation assumed nature rate. Relevant presented international national standards. Recent experimental studies shown that is advisable take nonlinear rate (BMP). BMP inspected aid potential meter together contactless current principles networks processing were formulated. An example actual considered analyzed pipe 17G1S grade steel its surface. This resulted revealed nonlinearity characterized by magnitude d=1.136. control method procedures help proposed. They make possible describe process depth wall physically sound mathematically more correct contrast standard procedures. important improving methods operated oil gas enterprises, particular, measurement evaluation anode currents insulation defects informative