
Partial Differential Equations in Applied Mathematics, Год журнала: 2024, Номер 13, С. 101012 - 101012
Опубликована: Дек. 11, 2024
Язык: Английский
Partial Differential Equations in Applied Mathematics, Год журнала: 2024, Номер 13, С. 101012 - 101012
Опубликована: Дек. 11, 2024
Язык: Английский
European Journal of Mechanics - A/Solids, Год журнала: 2025, Номер 111, С. 105547 - 105547
Опубликована: Янв. 2, 2025
Язык: Английский
Процитировано
2Applied Mathematical Modelling, Год журнала: 2025, Номер unknown, С. 116093 - 116093
Опубликована: Март 1, 2025
Язык: Английский
Процитировано
0Proceedings of the Royal Society A Mathematical Physical and Engineering Sciences, Год журнала: 2025, Номер 481(2313)
Опубликована: Май 1, 2025
Complex phenomena can be better understood when broken down into a limited number of simpler ‘components’. Linear statistical methods such as principal component analysis and its variants are widely used across various fields applied science to identify rank these components based on the variance they represent in data. These seen factorizations matrix collecting all data, assuming it consists time series sampled from fixed points space. However, data sampling locations vary over time, with mobile monitoring stations meteorology oceanography or particle tracking velocimetry experimental fluid dynamics, advanced interpolation techniques required project onto grid before factorization. This is often expensive inaccurate. work proposes method decompose scattered without interpolating. The approach employs physics-constrained radial basis function regression compute inner products space time. provides an analytical mesh-independent decomposition demonstrating higher accuracy. Our allows distilling most relevant ‘components’ even for measurements whose natural output distribution maintaining high accuracy mesh independence.
Язык: Английский
Процитировано
0Scientific Reports, Год журнала: 2025, Номер 15(1)
Опубликована: Май 25, 2025
Solving Ordinary Differential Equations (ODEs) is an essential and very challenging computational problem in many areas of science engineering like physics, biology, control system, economics. However, traditional numerical methods such as Euler's method the Runge-Kutta generally suffer from problems grid dependency, propagation errors not all are applicable to nonlinear systems that complex. In return, metaheuristic algorithms have become promising alternatives strongly transform solution process into optimization task. this paper, we introduce a new search algorithm called Thermodynamic Inspired Search Algorithm (TSA) for approximate linear ODEs (LODEs), (NLODEs) (SODEs). by thermodynamic processes, heat exchange, energy minimization entropy control, TSA employs on purpose balancing global exploration local exploitation throughout process. A mesh free ODE solver constructed with accurate approximation exact using Fourier periodic expansion basis function combined proposed weighted residual method. The population candidate solutions represent coefficients series. It comes up levels based associated fitness values, enabling adaptive through exchange between solutions. Entropy diversity indicator prevents converging prematurely promotion diversity. transformation occurs gradually temperature reduction formulated objective constraints, residuals boundary conditions penalty functions ensuring constraint satisfaction. performance further evaluated diverse benchmark suite twenty which demonstrates superior state-of-the-art optimizers, namely ADE, PSO, ABC accuracy, convergence rate robustness. Mesh nature enables us advantage (i) being dependent (iii) preservation Energy driven updates. shown experimental results achieve lower Root Mean Square Errors (RMSE) than existing both IVPs BVPs. addition, least-square approach improve precision approximations TSA's performance. gives robust, agile answer design framework unraveling complicated frameworks, attainable investigation PDE hybrid local-global optimal enhancement plans.
Язык: Английский
Процитировано
0Engineering Fracture Mechanics, Год журнала: 2024, Номер unknown, С. 110504 - 110504
Опубликована: Сен. 1, 2024
Язык: Английский
Процитировано
2Опубликована: Янв. 1, 2024
Язык: Английский
Процитировано
0Propulsion and Power Research, Год журнала: 2024, Номер unknown
Опубликована: Ноя. 1, 2024
Язык: Английский
Процитировано
0Partial Differential Equations in Applied Mathematics, Год журнала: 2024, Номер 13, С. 101012 - 101012
Опубликована: Дек. 11, 2024
Язык: Английский
Процитировано
0