This paper reviews the evolution of neural network models designed to capture complex hysteretic systems. It details advancements from Prandtl and Preisach neural networks, which introduced memory and non-Masing hysteresis, to more sophisticated models like generalized Prandtl neural networks (GPNNs) and the neuro-Madelung model (NMM). The review highlights how these models incorporate features such as stiffness degradation, rate dependence, and various forms of hysteresis to accurately represent nonlinear physical systems. AI
IMPACT Advances in modeling hysteretic systems could improve the accuracy of simulations in fields like material science and structural engineering.
RANK_REASON The item is a research paper published on arXiv detailing advancements in neural network models for hysteretic systems. [lever_c_demoted from research: ic=1 ai=1.0]
- DPCNNs
- EPNN
- GPNNs
- Madelung
- Mehrdad Shafiei Dizaji
- Neuro-Madelung
- Neuroplastic
- Prandtl neural networks
- Preisach neural networks
- PreisachNNs
- probabilistic neural network
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →