TEXNOLOGIK JARAYONLARNI AVTOMATLASHTIRISHDA NEYRON TARMOQLARNI O‘QITISH QOIDALARINI QO‘LLASH
Sabirov U. K.
Annotatsiya
Ushbu maqolada texnologik jarayonlarni avtomatlashtirishda sun’iy neyron tarmoqlarni o‘qitish qoidalarining qo‘llanilishi o‘rganilgan. Neyron tarmoqlarning asosiy o‘qitish algoritmlari — gradient tushish, xatolikni minimallashtirish, Hebb o‘qitish qoidasi hamda backpropagation algoritmi tahlil qilindi. Tadqiqotda texnologik jarayonlarni boshqarish misolida neyron tarmoqlarning ishlash prinsipi va ularni o‘qitish jarayoni ko‘rib chiqildi. Shuningdek, sanoat jarayonlarini modellashtirish va optimallashtirishda neyron tarmoqlarning samaradorligi ilmiy manbalar asosida baholandi. Tadqiqot natijalari shuni ko‘rsatadiki, neyron tarmoqlar murakkab va nochiziqli texnologik tizimlarni modellashtirish va boshqarishda samarali vosita hisoblanadi.
Adabiyotlar
- Hornik K., Stinchcombe M., White H. Multilayer feedforward networks are universal approximators. Neural Networks, 1989. https://doi.org/10.1016/0893-6080(89)90020-8 (ScienceDirect)
- Willis M., Montague G. Artificial neural networks in process estimation and control. Automatica, 1997. https://doi.org/10.1016/S0005-1098(96)00154-9
- Rumelhart D., Hinton G., Williams R. Learning representations by back-propagating errors. Nature, 1986. https://doi.org/10.1038/323533a0
- Lawrence N. Machine learning for industrial sensing and control. ScienceDirect. https://www.sciencedirect.com
- González C. Neural networks for fast optimization in model predictive control. arXiv. https://arxiv.org/abs/2003.04216
- Kiš K. Neural network based MPC for chemical reactor control. arXiv. https://arxiv.org/abs/2103.01542
- Yousefian R. Neural network approaches for stability control in energy systems. arXiv. https://arxiv.org/abs/2104.03054
- Vyatkin V. Intelligent automation systems in industry. MDPI. https://www.mdpi.com
- Emmanouilides C. Neural network control of industrial processes. Springer. https://link.springer.com