Microscopic limits of PDEs modeling macroscopic heat conduction
Ronald Mickens, Talitha Washington · 2025
Read Microscopic limits of PDEs modeling macroscopic heat conduction by Ronald Mickens, Talitha Washington (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-21299.txt.
Introduction
Ronald Mickens and Talitha Washington run the usual modeling path in reverse for heat flow. Most work starts at the tiny scale and zooms out to broad laws. Their 2025 paper starts from the broad law, the partial differential equation for heat spread in one dimension. From it, they build fine-scale discrete equations that act at the level of small parts.
The build shows the two views can be tied in both ways. Heat flow is the test case here. The authors sketch how the same trick may fit other systems too. A two-way bridge could help check broad laws against fine behavior. It could also seed new schemes for computing heat spread.
Catalog record
| CREATOR | Ronald Mickens, Talitha Washington |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 23 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-21299
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Microscopic limits of PDEs modeling macroscopic heat conduction Ronald Mickens, Talitha Washington Abstract . We show that it is possible to construct microscopic-level discrete equations from macroscopic modeling PDEs for heat conduction in one space dimension. The significance of this result is that, in general, one starts from microscopic theories and then take their continuum limits to obtain the corresponding macroscopic PDEs, whereas here it is demonstrated that the reverse procedure is also possible. While our focus is on heat conduction, we discuss the applicability of our methodology to other physical systems. KEYWORDS : Continuum limits; microscopic limits; discretization; nonstandard finite difference scheme; heat conduction; PDEs; positivity. † † footnotetext: AMS CLASSIFICATION: 35B09; 35G16; 35K05; 65M06; 80A10; 1. Introduction A physical system involving simple heat conduction may be examined from either macroscopic or microscopic perspectives [ 3 , 5 , 14 ] . At the macroscopic level (MAL), the space-time variables are continuous, and the evolution equations are partial differential equations (PDEs) However, at the microscopic level (MIL), the space-time variables are generally discrete, and the governing equations are either difference equations (DE) or systems of coupled ordinary differential equations (ODE). For the heat transfer problem, a standard procedure is the assume the existence of a micro-level mathematical model and then take a “continuum limit” to determine the macro-level PDEs [ 3 , 10 , 12 , 14 ] . However, the reverse process, i.e., beginning with the MAL PDEs and using them to derive the MIL evolutionary DEs is generally thought to be either impossible or not able to provide a unique set of discrete MIL equations. A thorough search of the published research literature turns up no publications on this issue. The major goal of this article is to demonstrate that from various MAL equations of simple heat transfer, it is possible to derive an essentially unique set of MIL equations. The existence of this result may provide insights as to whether this type of methodology can be applied to other important physical systems, such as wave propagation and fluid dynamics [ 3 , 5 , 10 , 14 ] . In outline, the paper proceeds as follows: Section 2 provides definitions and discussions related to both the MAL and MIL limit equations and how to determine one given the other. In Section 3, we start with three MAL PDEs and show how to select finite difference discretizations such that the MIL DEs correspond to random walk equations. Finally, Section 5 summarizes our results and discusses their implications for other physical systems. It should be indicated that all of our current investigations, as they relate to this paper, refer to various versions of heat conduction for only one space dimension. Further, all parameters appearing in the associated PDEs are assumed to be constant. The dependent variable, u , is taken to represent the absolute Kelvin temperature, which has the property that it can not be negative. Throughout this paper, we will have the opportunity to use the following abbreviations: • CL: continuum limit • DE: discrete equation • MAL: macroscopic level • MIL: microscopic level • MM: mathematical model • ODE: ordinary differential equation • PDE: partial differential equation 2. Background For the purposes of this paper, a mathematical model of a physical system is an abstract representation of the system, which is expressed in terms of discrete equations (DE), ordinary differential equations (ODEs), or partial differential equations (PDEs). The dependent variables in these mathematical structures correspond to the dependent variables of the physical system, and it is assumed that the independent variables can be related to the usual space and time coordinates. Microscopic or micro-level (MIL) mathematical models are generally formulated within a context in which an “atomic substructure” of the physical system is assumed to exist. The mathematical model equations will be dependent on space, time, and system variables, which may be discrete and/or continuous-valued. Also, parameters may appear in these equations, and they will be assumed to be constant. (The parameters provide a characterization of the overall physical properties and behaviors of the MIL modeling system.) It should be noted that the MIL equations are usually DEs or systems of coupled ODEs [ 5 , 10 , 14 ] . Macroscopic or macro-level (MAL) equations for a physical system are usually derived by taking the continuum limit (CL) of the MIL equations [ 5 , 10 ] . However, for a given set of MIL equations, this procedure may not yield unique mathematical expressions for the MAL equations, which in some cases are PDEs [ 10 , 14 ] . The CLs are obtained by letting some or all of the parameters in the MIL equations go to special limiting values, such that particular combinations of the parameters are held fixed [ 5 , 10 , 14 ] . These combinations may usually be identified with measurable macroscopic physical properties of the system. We now illustrate these concepts by considering the case of simple heat conduction in a one-space dimension [ 14 ] . For this situation, a way to start is to assume that the MIC MM is a random walk equation (1) u_{m}^{k+1}=\left(\frac{1}{2}\right)u_{m+1}^{k}+\left(\frac{1}{2}\right)u_{m-1}^{k} where (k,m) refers to discrete time and space variables that are associated with parameters (\tau,a) , which have the physical units, respectively, of time and distance. See Zaudere [ [ 14 ] , Chapter 1] for full explanatory details. If we make the definitions (2a) \displaystyle k \displaystyle\rightarrow t_{k}=\tau k,k=(0,1,2,…), (2b) \displaystyle m \displaystyle\rightarrow x_{m}=am,m=\text{ integers}, (2c) \displaystyle u_{m}^{k} \displaystyle\rightarrow u(x_{m},t_{k}) then the continuum generalization of Eq. ( 1 ) is (3) u(x,t+\tau)=\left(\frac{1}{2}\right)u(x+a,t)+\left(\frac{1}{2}\right)u(x-a,t). After some algebraic manipulations, this last equation can be rewritten to the form (4) \frac{u(x,t+\tau)-u(x,t)}{\tau}=\left(\frac{a^{2}}{2\tau}\right)\left[\frac{u(x+a,t)-2u(x,t)+u(x-a,t)}{a^{2}}\right]. If now we make the (really big) assumption that u(x,t) has a Taylor series and take the limits (5) a\rightarrow 0,\quad\tau\rightarrow 0,\text{ with }\frac{a^{2}}{2\tau}=D=\text{ fixed,} then from Eq. ( 4 ), the following result is obtained (6) u_{t}=Du_{xx},\quad\partial_{t}=\partial/\partial t,\quad\partial_{xx}=\partial^{2}/\partial x^{2} The constant D has the physical units of (7) [D]=\frac{\text{Length}^{2}}{\text{Time}}. and is called the thermal diffusivity [ 10 , 14 ] . The just presented calculation is an example of how the macroscopic heat conduction PDE, Eq. ( 6 ), can be “derived” from the microscopic-based random walk equation. However, a number of other PDEs have been put forth to model heat conduction at the macroscopic level. Equations ( 11b ) and ( 11c ) are two such cases. As stated previously, the central goal of this paper is to show that the inverse problem is not meaningless, i.e., the answer to the question, “Given a MAL PDE modeling heat conduction, can we construct or derive a suitable MIL mathematical representation?”, is yes. Our general procedure is as follows: (1) First, select a MAL PDE. (2) Using finite differences, construct a discretization of the PDE. This is a critical step and will require both insight and innovation into the various possibilities for discretizing derivatives. (3) Rewrite the scheme as a linear combination of terms involving u_{m}^{k+1},u_{m+1}^{k+1},u_{m-1}^{k+1},u_{m}^{k-1} , etc. The coefficients of these terms will depend on the parameters appearing in the PDE and the space and time step-sizes, \Delta x and \Delta t . (4) Except for the terms involving u_{m}^{k+1},u_{m+1}^{k},u_{m}^{k},u_{m-1}^{k} ,