<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.3 20210610//EN" "JATS-journalpublishing1-3.dtd">
<article article-type="research-article" dtd-version="1.3" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xml:lang="ru"><front><journal-meta><journal-id journal-id-type="publisher-id">novtexmech</journal-id><journal-title-group><journal-title xml:lang="ru">Мехатроника, автоматизация, управление</journal-title><trans-title-group xml:lang="en"><trans-title>Mekhatronika, Avtomatizatsiya, Upravlenie</trans-title></trans-title-group></journal-title-group><issn pub-type="ppub">1684-6427</issn><issn pub-type="epub">2619-1253</issn><publisher><publisher-name>Commercial Publisher «New Technologies»</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.17587/mau.25.345-353</article-id><article-id custom-type="elpub" pub-id-type="custom">novtexmech-1591</article-id><article-categories><subj-group subj-group-type="heading"><subject>Research Article</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="ru"><subject>СИСТЕМНЫЙ АНАЛИЗ, УПРАВЛЕНИЕ И ОБРАБОТКА ИНФОРМАЦИИ</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="en"><subject>SYSTEM ANALYSIS, CONTROL AND INFORMATION PROCESSING</subject></subj-group></article-categories><title-group><article-title>Синтез устойчивых нейросетевых регуляторов для объектов с ограничителями в условиях неполной информации</article-title><trans-title-group xml:lang="en"><trans-title>The Method of Synthesis of a Stable Closed-Loop Object Control System with Limiters</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Хапкин</surname><given-names>Д. Л.</given-names></name><name name-style="western" xml:lang="en"><surname>Khapkin</surname><given-names>D. L.</given-names></name></name-alternatives><bio xml:lang="ru"><p>мл. науч. сотр.</p></bio><email xlink:type="simple">dima-hapkin@ya.ru</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Феофилов</surname><given-names>С. В.</given-names></name><name name-style="western" xml:lang="en"><surname>Feofilov</surname><given-names>S. V.</given-names></name></name-alternatives><bio xml:lang="ru"><p>д-р техн. наук, проф.</p></bio><bio xml:lang="en"><p>Dr. of Eng. Sc., Professor</p></bio><email xlink:type="simple">svfeofilov@mail.ru</email><xref ref-type="aff" rid="aff-1"/></contrib></contrib-group><aff-alternatives id="aff-1"><aff xml:lang="ru"><institution>Тульский государственный университет</institution><country>Россия</country></aff><aff xml:lang="en"><institution>Tula State University</institution><country>Russian Federation</country></aff></aff-alternatives><pub-date pub-type="collection"><year>2024</year></pub-date><pub-date pub-type="epub"><day>18</day><month>07</month><year>2024</year></pub-date><volume>25</volume><issue>7</issue><fpage>345</fpage><lpage>353</lpage><permissions><copyright-statement>Copyright &amp;#x00A9; Commercial Publisher «New Technologies», 2024</copyright-statement><copyright-year>2024</copyright-year><copyright-holder xml:lang="ru">Commercial Publisher «New Technologies»</copyright-holder><copyright-holder xml:lang="en">Commercial Publisher «New Technologies»</copyright-holder><license xlink:href="https://mech.novtex.ru/jour/about/submissions#copyrightNotice" xlink:type="simple"><license-p>https://mech.novtex.ru/jour/about/submissions#copyrightNotice</license-p></license></permissions><self-uri xlink:href="https://mech.novtex.ru/jour/article/view/1591">https://mech.novtex.ru/jour/article/view/1591</self-uri><abstract><p>Современная теория автоматического управления сталкивается с проблемой сложности синтеза регуляторов для нелинейных объектов управления в условиях неполной информации. Существующие методы и подходы уже не могут удовлетворить запросы разработчиков автоматических систем управления сложными динамическими объектами. Во многих случаях объекты управления являются существенно нелинейными, нестационарными и требуют использования цифрового управления с заданны ми показателями качества. При этом получение точной математической модели не всегда возможно.Предлагается подход к решению этой проблемы с использованием регуляторов, основанных на искусственных нейронных сетях. Они могут быть эффективно применены в случае, когда отсутствует адекватная верифицированная и достаточно точная математическая модель объекта управления, но могут быть получены экспериментальные данные. Достоинством таких регуляторов является их способность к обучению и адаптации под объект на основе полученных данных.Кроме того, для замкнутых нейросетевых систем управления отсутствуют теоретические гарантии устойчивости, что существенно снижает возможности их применения в критически важных или опасных объектах. Для решения этой проблемы в работе предлагается метод синтеза нейрорегулятора, гарантирующего устойчивость замкнутого контура. В качестве объектов управления рассматриваются системы с наиболее часто встречающимися на практике нелинейностями (ограничители типа насыщение, ограничители типа жесткий механический упор и т. д.). В статье предлагаются теоретические подходы к решению обозначенных проблем, а также проводится сравнительный анализ с экспериментальными исследованиями для оценки эффективности предложенных методов</p></abstract><trans-abstract xml:lang="en"><p>The modern theory of automatic control is faced with the problem of complexity of synthesis of regulators for nonlinear control objects in conditions of incomplete information. The existing methods and approaches can no longer satisfy the needs of developers of automatic control systems for complex dynamic objects. In many cases, control objects are essentially nonlinear, nonstationary and require the use of digital control with specified quality indicators. In this case, obtaining an accurate mathematical model is not always possible. We propose an approach to solving this problem using regulators based on artificial neural networks. They can be effectively applied in the case when there is no adequate verified and sufficiently accurate mathematical model of the control object, but experimental data can be obtained. The advantage of such regulators is their ability to learn and adapt to the object based on the obtained data. In addition, there are no theoretical stability guarantees for closed-loop neural network control systems, which significantly reduces the possibility of their application in critical or hazardous facilities. To solve this problem, the paper proposes a method for synthesizing a neural controller that guarantees the stability of a closed loop. Systems with the most frequently encountered in practice nonlinearities (saturation type limiters, rigid mechanical stop type limiters, etc.) are considered as control objects. This paper proposes theoretical approaches to the solution of these problems, and also carries out a comparative analysis with experimental studies to assess the effectiveness of the proposed methods.</p></trans-abstract><kwd-group xml:lang="ru"><kwd>искусственная нейронная сеть</kwd><kwd>нейросетевой регулятор</kwd><kwd>система автоматического управления</kwd><kwd>динамический объект</kwd><kwd>нелинейный объект</kwd><kwd>устойчивость</kwd><kwd>нейросетевой регулятор</kwd><kwd>целочисленное программирование</kwd><kwd>функция Ляпунова</kwd><kwd>звенья с ограничителями</kwd></kwd-group><kwd-group xml:lang="en"><kwd>artificial neural network</kwd><kwd>neural network controller</kwd><kwd>automatic control system</kwd><kwd>dynamic object</kwd><kwd>nonlinear object</kwd><kwd>stability</kwd><kwd>neural network controller</kwd><kwd>integer programming</kwd><kwd>Lyapunov function</kwd><kwd>links with limiters</kwd></kwd-group><funding-group><funding-statement xml:lang="ru">Исследование выполнено за счет гранта Российского научного фонда № 23-29-00609, https://rscf.ru/project/23-29-00609/.</funding-statement><funding-statement xml:lang="en">The research was carried out at the expense of the grant of the Russian Science Foundation No. 23-29-00609</funding-statement></funding-group></article-meta></front><back><ref-list><title>References</title><ref id="cit1"><label>1</label><citation-alternatives><mixed-citation xml:lang="ru">Agrawal A., Amos B., Barratt S., Boyd S. Differentiable Convex Optimization Layers // Proceedings of 33rd Conference on Neural Information Processing Systems (NeurIPS 2019). Vancouver, Canada. 2019.</mixed-citation><mixed-citation xml:lang="en">Agrawal A., Amos B., Barratt S., Boyd S. Differentiable Convex Optimization Layers, Proceedings of 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver, Canada, 2019.</mixed-citation></citation-alternatives></ref><ref id="cit2"><label>2</label><citation-alternatives><mixed-citation xml:lang="ru">Cheng C.-H., Nührenberg G., Huang C.-H., Ruess H. Verification of Binarized Neural Networks via Inter-neuron Factoring // Proceedings of Verified Software. Theories, Tools, and Experiments Lecture Notes in Computer Science, Cham: Springer International Publishing, 2018. P. 279—290.</mixed-citation><mixed-citation xml:lang="en">Cheng C.-H., Nührenberg G., Huang C.-H., Ruess H. Verification of Binarized Neural Networks via Inter-neuron Factoring, Proceedings of Verified Software. Theories, Tools, and Experiments Lecture Notes in Computer Science, Cham: Springer International Publishing, 2018, pp. 279—290.</mixed-citation></citation-alternatives></ref><ref id="cit3"><label>3</label><citation-alternatives><mixed-citation xml:lang="ru">Maas A. L., Hannun A. Y., Ng A. Y. Rectifier nonlinearities improve neural network acoustic models // Proc. Icml. Atlanta, Georgia, USA. 2013. P. 3.</mixed-citation><mixed-citation xml:lang="en">Maas A. L., Hannun A. Y., Ng A. Y. Rectifier nonlinearities improve neural network acoustic models, Proc. icml, Atlanta, Georgia, USA, 2013, pp. 3.</mixed-citation></citation-alternatives></ref><ref id="cit4"><label>4</label><citation-alternatives><mixed-citation xml:lang="ru">Bunel R. R., Turkaslan I., Torr P., Kohli P., Mudigonda P. K. A Unified View of Piecewise Linear Neural Network Verification // Proceedings of Advances in Neural Information Processing Systems. Curran Associates, Inc., 2018.</mixed-citation><mixed-citation xml:lang="en">Bunel R. R., Turkaslan I., Torr P., Kohli P., Mudigonda P. K. A Unified View of Piecewise Linear Neural Network Verification, Proceedings of Advances in Neural Information Processing Systems, Curran Associates, Inc., 2018.</mixed-citation></citation-alternatives></ref><ref id="cit5"><label>5</label><citation-alternatives><mixed-citation xml:lang="ru">Dai H., Landry B., Pavone M., Tedrake R. Counter-example guided synthesis of neural network Lyapunov functions for piecewise linear systems // Proceedings of 59th IEEE Conference on Decision and Control (CDC). 2020.</mixed-citation><mixed-citation xml:lang="en">Dai H., Landry B., Pavone M., Tedrake R. Counterexample guided synthesis of neural network Lyapunov functions for piecewise linear systems, Proceedings of 59th IEEE Conference on Decision and Control (CDC), 2020.</mixed-citation></citation-alternatives></ref><ref id="cit6"><label>6</label><citation-alternatives><mixed-citation xml:lang="ru">Chen S., Fazlyab M., Morari M., Pappas G. J., Preciado V. M. Learning Lyapunov Functions for Hybrid Systems // Proceedings of 55th Annual Conference on Information Sciences and Systems (CISS). Baltimore, MD, USA: IEEE, 2021. P. 1—1.</mixed-citation><mixed-citation xml:lang="en">Chen S., Fazlyab M., Morari M., Pappas G. J., Preciado V. M. Learning Lyapunov Functions for Hybrid Systems, Proceedings of 55th Annual Conference on Information Sciences and Systems (CISS), Baltimore, MD, USA, IEEE, 2021, pp. 1—1.</mixed-citation></citation-alternatives></ref><ref id="cit7"><label>7</label><citation-alternatives><mixed-citation xml:lang="ru">Tjeng V., Xiao K., Tedrake R. Evaluating Robustness of Neural Networks with Mixed Integer Programming // arXiv:1711.07356. 2019.</mixed-citation><mixed-citation xml:lang="en">Tjeng V., Xiao K., Tedrake R. Evaluating Robustness of Neural Networks with Mixed Integer Programming, arXiv: 1711.07356, 2019.</mixed-citation></citation-alternatives></ref><ref id="cit8"><label>8</label><citation-alternatives><mixed-citation xml:lang="ru">Nair V., Hinton G. E. Rectified Linear Units Improve Restricted Boltzmann Machines // Proceedings of the 27th International Conference on International Conference on Machine Learning ICML’10. Madison, WI, USA: Omnipress, 2010. P. 807—814.</mixed-citation><mixed-citation xml:lang="en">Nair V., Hinton G. E. Rectified Linear Units Improve Restricted Boltzmann Machines, Proceedings of the 27th International Conference on International Conference on Machine Learning ICML’10, Madison, WI, USA, Omnipress, 2010, pp. 807—814.</mixed-citation></citation-alternatives></ref><ref id="cit9"><label>9</label><citation-alternatives><mixed-citation xml:lang="ru">Glorot X., Bordes A., Bengio Y. Deep Sparse Rectifier Neural Networks // Proceedings of Journal of Machine Learning Research. 2010.</mixed-citation><mixed-citation xml:lang="en">Glorot X., Bordes A., Bengio Y. Deep Sparse Rectifier Neural Networks, Proceedings of Journal of Machine Learning Research, 2010.</mixed-citation></citation-alternatives></ref><ref id="cit10"><label>10</label><citation-alternatives><mixed-citation xml:lang="ru">Dai H., Landry B., Yang L., Pavone M., Tedrake R. Lyapunov-stable neural-network control // arXiv:2109.14152. 2021.</mixed-citation><mixed-citation xml:lang="en">Dai H., Landry B., Yang L., Pavone M., Tedrake R. Lyapunov-stable neural-network control, arXiv:2109.14152, 2021.</mixed-citation></citation-alternatives></ref><ref id="cit11"><label>11</label><citation-alternatives><mixed-citation xml:lang="ru">Werbos P. J. Backpropagation through time: what it does and how to do it // Proceedings of the IEEE. 1990. Vol. 78, N. 10. P. 1550—1560.</mixed-citation><mixed-citation xml:lang="en">Werbos P. J. Backpropagation through time: what it does and how to do it, Proceedings of the IEEE, 1990, vol. 78, no. 10, pp. 1550—1560.</mixed-citation></citation-alternatives></ref><ref id="cit12"><label>12</label><citation-alternatives><mixed-citation xml:lang="ru">Феофилов С. В., Козырь А. В., Хапкин Д. Л. Структурно-параметрический синтез нейросетевых регуляторов для объектов управления с ограничителями // Мехатроника, автоматизация, управление. 2023. Т. 24, № 11. С. 563—572.</mixed-citation><mixed-citation xml:lang="en">Feofilov S. V., Kozyr A. V., Khapkin D. L. Structural and Parametric Synthesis of Neural Network Controllers for Control Objects with Limiters. Mekhatronika, Avtomatizatsiya, Upravlenie, 2023, vol. 24, no. 11, pp 563—572 (in Russian), doi:10.17587/mau.24.563-572.</mixed-citation></citation-alternatives></ref><ref id="cit13"><label>13</label><citation-alternatives><mixed-citation xml:lang="ru">Feofilov S. V., Khapkin D. L. Synthesis of neural network controllers for objects with non-linearity of the constraint type // Journal of Physics: Conference Series. 2021. Vol. 1958, N. 1. P. 012014.</mixed-citation><mixed-citation xml:lang="en">Feofilov S. V., Khapkin D. L. Synthesis of neural network controllers for objects with non-linearity of the constraint type, Journal of Physics: Conference Series, 2021, vol. 1958, no. 1, pp. 012014.</mixed-citation></citation-alternatives></ref><ref id="cit14"><label>14</label><citation-alternatives><mixed-citation xml:lang="ru">Wong E., Kolter Z. Provable defenses against adversarial examples via the convex outer adversarial polytope // Proceedings of International Conference on Machine Learning, PMLR. 2018. P. 5286—5295.</mixed-citation><mixed-citation xml:lang="en">Wong E., Kolter Z. Provable defenses against adversarial examples via the convex outer adversarial polytope, Proceedings of International Conference on Machine Learning, PMLR, 2018, pp. 5286—5295.</mixed-citation></citation-alternatives></ref><ref id="cit15"><label>15</label><citation-alternatives><mixed-citation xml:lang="ru">Квакернаак Х., Сиван Р., Васильев В. А., Николаев Ю. А., Петров Б. Н. Линейные оптимальные системы управления. М.: Мир, 1977. 650 с.</mixed-citation><mixed-citation xml:lang="en">Kvakernaak H., Sivan R., Vasiliev V. A., Nikolaev Yu. A., Petrov B. N. Linear optimal control systems, Moscow, Mir, 1977, 650 p. (in Russian).</mixed-citation></citation-alternatives></ref><ref id="cit16"><label>16</label><citation-alternatives><mixed-citation xml:lang="ru">Anderson B. D. O., Moore J. B., Molinari B. P. Linear Optimal Control // Proceedings of IEEE Transactions on Systems, Man, and Cybernetics. 1972. Vol. SMC-2, N. 4. P. 559—559.</mixed-citation><mixed-citation xml:lang="en">Anderson B. D. O., Moore J. B., Molinari B. P. Linear Optimal Control, Proceedings of IEEE Transactions on Systems, Man, and Cybernetics, 1972, vol. SMC-2, no. 4, pp. 559—559.</mixed-citation></citation-alternatives></ref><ref id="cit17"><label>17</label><citation-alternatives><mixed-citation xml:lang="ru">Хоанг Ч. К. Оптимизация объемных силовых следящих гидроприводов по быстродействию и по точности режима слежения: диссертация... кандидата технических наук: 05.13.01 Тула, 2006. 131 с.</mixed-citation><mixed-citation xml:lang="en">Hoang C. K. Optimization of volumetric power tracking hydraulic drives in terms of speed and accuracy of the tracking mode: dissertation... Candidate of Technical Sciences: 05.13.01, Tula, 2006, 131 p. (in Russian).</mixed-citation></citation-alternatives></ref></ref-list><fn-group><fn fn-type="conflict"><p>The authors declare that there are no conflicts of interest present.</p></fn></fn-group></back></article>
