Non-iterative architecture generating U-shaped forms and controlled multimodal transitions: extension toward Maxwell–Boltzmann-type profiles
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
We present a family of non-iterative functions (NIF) in which the controlled variation of a limited number of constants makes it possible to obtain : • Four distinct U-shaped distributions, • Twelve morphologically different distributions generated from a single parameter, • Multimodal series with a variable number of modes, generated from a single parameter, • Profiles close to physical distributions. The MB-like profile results from a transformation applied to a distribution derived from a modified Gaussian Unitary Ensemble (GUE), establishing a structural link between random matrix theory and thermodynamic distributions. Two constants emerge: the coefficient of variation (Ct1), and a second constant (Ct2), defined as the product of the mean gas value and the maximum peak value of the distribution (PIC). The observed transitions are continuous, reproducible, and reversible under inverse parametric variation. Calculations are performed over 10⁷ points, with computation times of around one second without specific optimization on an HP G9.
Fields of Study & Subjects
Statistical Mechanics and Entropy
Mathematical and Computational Methods
Mathematical functions and polynomials
Publication details
- Journal / source
- Applied Mathematical Sciences
- Volume
- 20
- Issue
- 3
- Pages
- 105–113
- Publication date
- 1 January 2026
- Language
- EN