SANOAT TEXNIK OBYEKTLARI UCHUN CHUQUR O‘RGANISHGA ASOSLANGAN NOSOZLIKLARNI ANIQLASH VA BASHORATLI TEXNIK XIZMAT KO‘RSATISH TIZIMI
Ключевые слова:
chuqur o‘rganish, nosozliklarni aniqlash, graf neyron tarmoqlari, ko‘p agentli mustahkamlash o‘rganishi, bashoratli texnik xizmat, issiqlik elektr stantsiyasi.Аннотация
Ushbu maqolada sanoat texnik obyektlari — xususan, issiqlik elektr stantsiyalari — uchun mo‘ljallangan chuqur o‘rganishga asoslangan nosozliklarni aniqlash va bashoratli texnik xizmat ko‘rsatish tizimi taqdim etiladi. Taklif qilingan tizim graf neyron tarmoqlari (GNN), ko‘p agentli mustahkamlash o‘rganishi (MARL), Proximal Policy Optimization (PPO) va Soft Actor-Critic (SAC) algoritmlarini hamda Shapley-asosli koordinatsiya mexanizmini birlashtiradi. Eksperimental natijalar GNN+MARL giperlarning nosozliklarni aniqlash aniqligini 98.3% ga, F1-ko‘rsatkichni esa 97.5% ga etkazganini ko‘rsatdi, bu esa mavjud usullardan sezilarli darajada ustun kelishini tasdiqlaydi. Bundan tashqari, tizim real vaqtda kechikishni 22.6 ms chegarasida saqlab qoladi, bu esa sanoat joylashtirishning amaliy talablariga javob beradi.
Скачивания
Библиографические ссылки
1. Zhang J. Rule-based fault detection limitations in complex industrial systems / J. Zhang, P. Wang, R. Yan // IEEE Transactions on Industrial Informatics. — 2022. — Vol. 18, № 4. — P. 2341–2352.
2. Hochreiter S. Long short-term memory / S. Hochreiter, J. Schmidhuber // Neural Computation. — 1997. — Vol. 9, № 8. — P. 1735–1780.
3. LeCun Y. Deep learning / Y. LeCun, Y. Bengio, G. Hinton // Nature. — 2015. — Vol. 521, № 7553. — P. 436–444.
4. Kingma D. P. Auto-encoding variational bayes / D. P. Kingma, M.Welling. — arXiv:1312.6114, 2013.
5. Zhang L. Sliding window-based fault detection from high-dimensional data streams / L. Zhang, J. Lin, R. Karim // IEEE Transactions on Systems. — 2019. — Vol. 49, № 3. — P. 723–737.
6. Jiang G. Multiscale convolutional neural networks for fault diagnosis of wind turbine gearbox / G. Jiang, H. He, J. Yan // IEEE Transactions on Industrial Electronics. — 2019. — Vol. 66, № 4. — P. 3196–3207.
7. Lei Y. Applications of machine learning to machine fault diagnosis: a review and roadmap / Y. Lei, B. Yang, X. Jiang // Mechanical Systems and Signal Processing. — 2020. — Vol. 138. — Article 106587.
8. Hasan M. J. Acoustic spectral imaging and transfer learning for reliable bearing fault diagnosis under variable speed conditions / M. J. Hasan, M. M. M. Islam, J. M. Kim // Measurement. — 2019. — Vol. 138. — P. 620–631.
9. Scarselli F. The graph neural network model / F. Scarselli, M. Gori, A. C. Tsoi // IEEE Transactions on Neural Networks. — 2009. — Vol. 20, № 1. — P. 61–80.
10. Wu Z. A comprehensive study on graph neural networks / Z. Wu, S. Pan, F. Chen // IEEE Transactions on Neural Networks and Learning Systems. — 2020. — Vol. 32, № 1. — P. 4–24.
11. Zhou K. Fault diagnosis of hydraulic system based on graph neural network / K. Zhou, J. Tang, L. Deng // Measurement. — 2021. — Vol. 174. — Article 109074.
12. Li Y. Gated graph sequence neural networks for industrial fault diagnosis / Y. Li, J. Bilmes, D. Tarlow // ICLR Workshop on Graph Representation Learning. — 2020.
13. Busoniu L. A comprehensive survey of multiagent reinforcement learning / L. Busoniu, R. Babuska, B. De Schutter // IEEE Transactions on Systems. — 2008. — Vol. 38, № 2. — P. 156–172.
14. Schulman J. Proximal policy optimization algorithms / J. Schulman, F.Wolski, P. Dhariwal. — arXiv:1707.06347, 2017.
15. Haarnoja T. Soft actor-critic: off-policy maximum entropy deep reinforcement learning / T. Haarnoja, A. Zhou, P. Abbeel // ICML. — 2018. — P. 1861–1870.
16. Lowe R. Multi-agent actor-critic for mixed cooperative-competitive environments / R. Lowe, Y. Wu, A. Tamar // NeurIPS. — 2017. — P. 6379–6390.
17. Shapley L. S. A value for n-person games / L. S. Shapley // Contributions to the Theory of Games. — 1953. — Vol. 2, № 28. — P. 307–317.
18. Sunehag P. Value-decomposition networks for cooperative multi-agent learning / P. Sunehag, G. Lever, A. Gruslys. — arXiv:1706.05296, 2018.
