COMPARATIVE ANALYSIS OF MACHINE LEARNING MODELS WHEN SOLVING PROBLEM OF DETECTING ABNORMAL ERRORS IN PARAMETERS MEASUREMENTS OF PIPELINE SYSTEMS’ OPERATIONAL MODES
UDC: 681.5:622.279
DOI: 10.33285/0132-2222-2021-1(570)-55-60
Authors:
GOLUBYATNIKOV EVGENY ALEXANDROVICH 1,
LEONOV DMITRY GENNADIEVICH 1
1 National University of Oil and Gas "Gubkin University", Moscow, Russian Federation
Keywords: gas transportation systems, Gross measurement error, machine learning models, linear regression, Stochastic gradient boosting, neural networks
Annotation:
The paper is a continuation of the research dedicated to the applicability of machine learning methods for solving the problem of identifying and filtering gross errors in measurements of parameters characterizing the operational modes of pipeline systems (PS). In the previous part of the research an original method for detecting gross measurement errors was proposed. The method is based on the regression reconstruction with help of machine learning models. Stochastic gradient boosting of decision-making trees model was used for solving regression tasks. The paper provides a comparative analysis of the models applicable for the regression estimation of the pipeline modes measurements. Possible alternatives are considered as a key step in the method for solving the problem of detecting gross errors when measuring real pipelines mode parameters.
Bibliography:
1. Цифровая экономика нефтяного производства / Г.И. Шмаль, Л.И. Григорьев, В.Я. Кершенбаум, Д.Г. Леонов // Нефт. хоз-во. - 2019. - № 1. - С. 100-103. - DOI: 10.24887/0028-2448-2019-1-100-103
2. Голубятников Е.А. Применение моделей машинного обучения для выявления аномальных измерений параметров эксплуатационных режимов трубопроводных систем // Автоматизация, телемеханизация и связь в нефтяной промышленности. - 2020. - № 1. - С. 40-47. - DOI: 10.33285/0132-2222-2020-1(558)-40-47
3. Леонов Д.Г. Методы, модели и технологии разработки и интеграции распределенных гетерогенных программно-вычислительных комплексов в транспорте газа. - М.: Издат. центр РГУ нефти и газа (НИУ) им. И.М. Губкина, 2017. - 196 с.
4. Goodfellow I., Bengio Y., Courville A. Deep Learning. - The MIT Press, 2016. - 800 p. - (Adaptive Computation and Machine Learning series).
5. Tibshirani R. Regression Shrinkage and Selection via the Lasso // J. of the Royal Statistical Society. Series B (Methodological). - 1996. - Vol. 58, Issue 1. - P. 267-288. - DOI: 10.1111/J.2517-6161.1996.TB02080.X
6. Natekin A., Knoll A. Gradient boosting machines, a tutorial // Frontiers in Neurorobotics. - 2013. - Vol. 7. - Article No 21. - DOI: 10.3389/fnbot.2013.00021
7. Сарданашвили С.А. Расчетные методы и алгоритмы (трубопроводный транспорт газа). - М.: Изд-во "Нефть и газ" РГУ нефти и газа им. И.М. Губкина, 2005. - 577 с.
8. Scikit-learn. Machine Learning in Python. - URL: https://scikit-learn.org (дата обращения 14.10.2020).
9. XGBoost Python Package. - URL: https://xgboost.readthedocs.io/en/latest/python/ (дата обращения 14.10.2020).
10. Keras: the Python deep learning API. - URL: https://keras.io (дата обращения 14.10.2020).
Back to issue