Construction method and device of phase change energy storage multi-phase-field multi-scale calculation model and storage medium

Abstract

The invention provides a construction method and device of a phase change energy storage multi-phase-field multi-scale calculation model and a storage medium, and belongs to the technical field of phase change energy storage. The method comprises the following steps: dividing the phase change energy storage system into a micro scale, a mesoscale and a macro scale, and dividing different computational domains based on different scales; constructing physical models of each computational domain, wherein the physical models comprise a micro-scale physical model, a mesoscale physical model and a macro-scale macro continuous model; the physical models of all the computational domains are subjected to parallel calculation, the physical models of all the computational domains are coupled in the parallel calculation process, and real-time transmission of information of all the computational domains is achieved; aiming at the defects of the existing research, a microscopic molecular dynamics algorithm, a mesoscopic lattice Boltzmann algorithm and a macroscopic finite element model are mutually coupled in parallel, and the cross-scale multi-phase field coupling research from a microscopic phenomenon to a macroscopic phenomenon in an energy storage system is realized.

Description

Technical Field

[0001] The present application relates to the field of phase change energy storage technology, and in particular to a method and device for constructing a multi-phase field and multi-scale calculation model for phase change energy storage, as well as a storage medium. Background Art

[0002] The world's demand for cooling energy is rapidly increasing due to the rapid growth of population, buildings, comfort requirements, and electronics. This has led to a conflict between electricity supply and demand. Meeting electricity demand by utilizing decentralized energy within the existing infrastructure and optimally allocating energy among various consumers is a straightforward and effective solution.

[0003] Latent heat storage has become an attractive energy conversion technology and an excellent choice for renewable energy utilization, waste heat recovery, air conditioning control, and other industrial applications. Phase change materials are favored due to their high storage density, light weight, and small size. In addition, phase change cold storage can provide a constant or near-constant temperature cooling source.

[0004] In recent years, the country has implemented a series of electricity policies, including tiered electricity usage and peak-valley electricity pricing, creating a favorable environment for the advancement of cold storage technology. Ice, as an inorganic phase change material with high latent heat, non-toxicity, and low cost, offers significant advantages in large-scale energy systems for commercial buildings. Ice storage technology is considered an effective means of shifting electricity loads from peak to off-peak periods. This not only improves equipment efficiency, extends equipment life during off-peak periods, and reduces the installed capacity of refrigeration equipment, but also alleviates the imbalance between electricity demand and supply, effectively regulating the imbalance between peak and valley power consumption in the power grid, achieving the goal of shifting peak loads and filling valleys, thereby balancing the power grid. Furthermore, by leveraging the nationally mandated peak-valley electricity price differential, operating costs can be reduced, creating better environmental and economic benefits for users. To fully utilize existing electricity resources and alleviate energy crises and environmental issues, research on heat and mass conversion and transmission within ice storage systems is of vital practical significance.

[0005] Currently, experimental research on ice storage technology focuses on improving the heat transfer coefficient and enhancing heat exchange, focusing on improvements in cold storage devices, phase change material additives, and phase change material packaging structures. Based on these experimental studies, numerical methods are used to investigate different cold storage systems or packaging structures, focusing on the heat transfer during solidification and melting. However, during the charging and discharging processes, factors affecting ice solidification and its microstructure directly affect the heat transfer coefficient and heat transfer performance, and there is also a certain degree of coupling between the multiphases within the cold storage device. Currently, research in this area is very limited, with few reports on experimental and numerical methods. Understanding the underlying mechanisms and laws is also very limited, and further research is urgently needed. Experimental research is limited by visualization, experimental equipment, and precision, necessitating the use of numerical methods. However, traditional numerical methods struggle to accurately reflect the internal microstructure and transport processes. Therefore, it is necessary to construct multiscale computational models to conduct cross-scale studies of the internal microstructure of ice, multiphase coupling mechanisms, flow, and heat and mass transfer mechanisms and properties.

[0006] Phase change energy storage systems involve energy transfer and conversion at multiple scales. Traditional research is mainly based on macroscopic energy-mass conversion and multiphase motion. However, microscopic phenomena and energy-mass transfer, such as interphase heat and mass transfer, phase interface motion characteristics, and crystal nucleus growth, during the phase change process in multiphase energy storage systems have a significant impact on the performance of the system. The difficulty in multiscale research lies in the mutual coupling of molecular-scale, mesoscopic-scale, and macroscopic-scale algorithms. Currently, no effective technical means have been proposed in existing technologies to solve the above problems. Summary of the Invention

[0007] In order to solve the above technical problems, this application proposes a method and device for constructing a multi-phase field multi-scale computational model of phase change energy storage, as well as a storage medium. By studying thermophysical problems such as intermolecular interactions, phase change nucleation and crystal growth, interface motion characteristics, and energy-to-mass conversion in multiphase energy storage systems from microscopic to macroscopic perspectives, the energy transport and conversion mechanism in the phase change energy storage system is revealed from multiple dimensions, the energy storage efficiency is improved, and a theoretical basis, performance prediction, and optimization are provided for the computational model of the multiphase phase change energy storage system.

[0008] The technical solution adopted in this application is: a method for constructing a multi-phase field multi-scale calculation model of phase change energy storage, comprising the following steps:

[0009] S1: The phase change energy storage system is divided into microscale, mesoscale and macroscale, and different computational domains are divided based on different scales. The computational domain of the microscale is the phase interface computational domain, the computational domain of the mesoscale is the phase interface and its vicinity computational domain, and the computational domain of the macroscale is each phase.

[0010] S2: Construct the physical models of each computational domain, namely the physical model of the microscopic scale, the physical model of the mesoscopic scale, and the macroscopic continuum model of the macroscopic scale;

[0011] S3: The physical models of each computational domain are calculated in parallel, and the physical models of each computational domain are coupled during the parallel calculation process to realize the real-time transmission of information of each computational domain, including: transmitting micro-scale information to meso-scale and meso-scale information to macro-scale through relaxation method and time coarse-graining method; transmitting macro-scale information to meso-scale and meso-scale information to micro-scale through time and space interpolation / extrapolation algorithm, realizing the real-time transmission of information of different scales in the phase change process.

[0012] Furthermore, two corresponding physical models are used for simultaneous and parallel coupling in the overlapping calculation domains of the microscale and mesoscale or the overlapping calculation domains of the mesoscale and macroscale. Three physical models are used for simultaneous and parallel coupling in the overlapping calculation domains of the microscale, mesoscale and macroscale. During parallel coupling, the physical models at each scale are calculated independently, and real-time information transmission is achieved through periodic data exchange.

[0013] Furthermore, the physical model at the microscopic scale is implemented using a molecular dynamics algorithm to simulate intermolecular interaction forces, nucleation and crystal growth. A multiphase phase transition molecular dynamics model is constructed at the microscopic scale and the parameters are initialized. The potential function and boundary conditions are determined, and the phase transition point, nucleation and crystal growth are determined by free energy calculation. Then, the accuracy is judged. When the calculation accuracy meets the requirements, the macroscopic thermodynamic parameters are calculated by statistical averaging, and the macroscopic thermodynamic parameters calculated by statistical averaging are transferred to the physical model at the mesoscopic scale through relaxation method and time coarse-graining amplification. When the accuracy does not meet the requirements, the potential function and boundary conditions are re-determined, and the phase transition point, nucleation and crystal growth are re-determined.

[0014] Furthermore, the mesoscopic physical model is implemented using the lattice Boltzmann algorithm to simulate the liquid / gas fluid flow under complex boundary conditions between melt flow and dendrites. A multiphase phase transition lattice Boltzmann model is constructed at the mesoscopic scale and the parameters are initialized to determine the equilibrium distribution function. Then, the evolution calculation of the flow and collision steps is performed, and the complex structure boundaries and moving boundaries are processed. Then, the accuracy is judged. When the calculation accuracy meets the requirements, the macroscopic parameters within the phase interface and its vicinity are calculated by statistical averaging, and the macroscopic parameters calculated by statistical averaging are transferred to the macroscopic physical model through relaxation method and time coarse-graining amplification. The macroscopic parameters are transferred to the multiphase phase transition molecular dynamics model through time and space interpolation / extrapolation algorithm. When the accuracy does not meet the requirements, the equilibrium distribution function is re-determined, and then the evolution calculation of the flow and collision steps is performed, and the complex structure boundaries and moving boundaries are processed.

[0015] Furthermore, the macro-continuum model at the macro scale is implemented using a macro-finite element / finite volume algorithm to simulate the macroscopic parameters of the flow field and temperature field within the phase, and a multiphase phase change energy storage model is constructed at the macro scale. The model is imported into the calculation software by compiling the code, and the compiled code is coupled with the finite element / finite volume phase change model to determine the boundary conditions, and then the accuracy is judged. When the calculation accuracy meets the requirements, the macroscopic flow field, temperature field, and phase field parameters are obtained, and the above-mentioned macroscopic flow field, temperature field, and phase field parameters are transferred to the multiphase phase change lattice Boltzmann model through time and space interpolation / extrapolation algorithms. If the calculation accuracy does not meet the requirements, the compiled code is re-coupled with the finite element / finite volume phase change model, and the boundary conditions are determined.

[0016] Furthermore, in areas or time periods where the phase change is drastic or rapid, dynamic adaptation is achieved by increasing the frequency of data information exchange.

[0017] Furthermore, at the boundary of the phase interface, the data information transmission between the mesoscale and microscale is realized by using the irregular boundary non-equilibrium state extrapolation method in the lattice Bolzmann algorithm.

[0018] Furthermore, a buffer layer is set at the boundary where different physical models are coupled in the overlapping computational domain, and the buffer layer is meshed. The boundary mesh node information of the buffer layer is not transmitted, and flux matching is forced in the buffer layer to make the flux in the algorithm used by the physical models at different scales consistent.

[0019] A computer device comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method.

[0020] A computer-readable storage medium stores a computer program / instruction thereon, which implements the steps of the method when executed by a processor.

[0021] The beneficial effects of this application compared to the existing technology are as follows: the existing research on phase change energy storage systems is mainly based on experimental and theoretical research on macroscopic working conditions, but there are few research methods for micro-mesoscopic mechanisms. This application mainly addresses the shortcomings of existing research and proposes a multi-phase field cross-scale mathematical model for phase change energy storage systems. It couples the microscopic molecular dynamics algorithm, the mesoscopic lattice Boltzmann algorithm, and the macroscopic finite element model in parallel to each other, realizing cross-scale multi-phase field coupling research from microscopic to macroscopic phenomena in phase change energy storage systems, revealing the energy storage phase change mechanism from different scales, improving performance, and thus providing certain guidance and predictions for practical engineering applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The present application will be further described below with reference to the accompanying drawings:

[0023] Figure 1 A flowchart of multi-scale multi-phase field coupling provided in an embodiment of the present application;

[0024] Figure 2 Different computational domain division diagrams provided for embodiments of the present application;

[0025] Figure 3 A flowchart of the calculation and coupling of different scales in the multi-scale mathematical model provided in the embodiment of the present application. DETAILED DESCRIPTION

[0026] like Figures 1 to 3 As shown, this application provides a method for constructing a multi-phase field and multi-scale computational model for phase change energy storage. This method couples the molecular dynamics method (MD), the mesoscopic lattice Boltzmann method (LBM), and the macroscopic finite volume (FVM) / finite element method (FEM) to establish a multi-scale mathematical model for a multiphase (solid, liquid, and gas) phase change energy storage system. The molecular dynamics method operates at the molecular scale and is used to study microscopic mechanisms such as intermolecular interactions, phase transition nucleation, and interface dynamics within the phase change energy storage system. The lattice Boltzmann algorithm operates at the mesoscale and, based on fluid dynamics, is used to study mesoscopic phenomena within the phase change energy storage system, such as multiphase flow, crystal nucleation, and phase interface motion characteristics. The macroscopic algorithm is used to study macroscopic mechanisms such as flow, heat transfer, and energy-to-mass conversion within the macroscopic continuous medium of the phase change energy storage system.

[0027] The above-mentioned algorithms of different scales can accurately obtain information of various dimensions within the phase change energy storage system. This application couples multi-scale algorithms to establish information communication channels between algorithms of each scale, realizes real-time communication and exchange of multi-level information, and more realistically reveals the phase change, nucleation, crystal growth phase interface movement, interphase energy and mass transfer, and multi-phase complex flow mechanisms of the phase change energy storage system from microscopic to macroscopic levels.

[0028] The mutual coupling of microscopic, mesoscopic and macroscopic algorithms is a key issue in the construction of multi-scale mathematical models, and is also the bottleneck and technical difficulty in the multi-scale research of phase change energy storage systems. The multi-scale coupling process of this application will run different algorithms simultaneously and exchange data in real time at different dimensions and algorithms, such as Figure 1 To achieve the above purpose, the specific steps of the method proposed in this application are as follows:

[0029] 1) Decompose the computational domain

[0030] In order to consider both the acquisition of detailed information and the computational cost, the computational domain is divided into three major regions: single phase (region A), phase interface (region C), and near phase interface (region B). Figure 2 As shown, an appropriate algorithm is set for each area.

[0031] 2) Determine the algorithms for each computational domain

[0032] Different calculation methods are determined in different calculation domains, and multi-phase energy storage physical models of each algorithm are established. Multiple algorithms in overlapping calculation domains are coupled with each other to realize real-time information exchange.

[0033] 3) Determine mathematical models and research content at different scales

[0034] The physical model in the phase interface domain is a microscopic molecular dynamics model, which uses a molecular dynamics algorithm and writes computer language program codes to simulate the intermolecular interaction force, crystal nucleation and crystal growth.

[0035] The phase interface and the two regions near the phase interface are mesoscopic Boltzmann models. The lattice Boltzmann algorithm is used to simulate the liquid / gas phase flow under the complex boundary conditions between melt flow and dendrites by writing computer language program code.

[0036] The phase is a macro-continuous model of macroscopic scale. The macroscopic finite element / finite volume algorithm is used with the help of calculation software such as FLUENT to simulate the macroscopic parameters such as flow field and temperature field in the phase.

[0037] 4) Coupling of algorithms at different scales to achieve real-time transmission of information in each computational domain.

[0038] 5) Data processing from microscopic to macroscopic computational data to obtain the phase field, temperature field and performance curves within the phase change energy storage system.

[0039] 6) Based on the performance parameters of the phase change energy storage system obtained, the actual process results are predicted.

[0040] The specific construction process is described in detail below.

[0041] (1) Determine the algorithm for each computational domain, and divide the computational domain into Figure 2 As shown, Figure 2 A represents the phase, B represents the phase interface and the vicinity of the phase interface, and C represents the phase interface domain. Figure 2 (a) shows the division of computational domain. Figure 2 (b) shows the schematic diagram of the extrapolation construction method of the mesoscopic particle distribution function near the phase interface, and Figure 2 (b) are fluid particles, For solid phase particles, is the discrete velocity vector of the particle in the i direction, is the discrete velocity vector of the particle in the opposite direction of i, are the particles at the interface, is the distance between adjacent particles, Figure 2(c) Schematic diagram of grid division and interface nodes at different scales;

[0042] Microscopic algorithms are computationally intensive, while the mesoscopic lattice Boltzmann algorithm is relatively efficient. Macroscopic algorithms are more efficient for larger phase change energy storage systems. This application distinguishes computational domains, employing molecular dynamics methods only in critical areas such as phase interfaces, and employing the lattice Boltzmann algorithm for phase interfaces and surrounding complex interfaces. A macroscopic model is used for the entire phase change energy storage system, with multiple algorithms coupled in parallel in overlapping regions to conserve computational resources while ensuring computational accuracy. The following are the execution strategies for algorithms in different computational domains.

[0043] 1) Molecular dynamics algorithms are used to simulate the intermolecular interaction forces, nucleation and crystal growth in the phase interface domain.

[0044] Figure 3 The meso- and micro-scale calculations involve the implementation of molecular dynamics algorithms. The key to molecular dynamics simulations is establishing a mathematical model to describe phase transitions and determining an appropriate potential function that accurately describes the properties of each phase and the interactions between molecules in different phases. Furthermore, the thermodynamic conditions for phase transitions are determined by calculating free energy.

[0045] The potential function of molecular dynamics phase transition includes the following situations:

[0046] ① For gas / liquid phase, the potential function of its phase transition The expression is:

[0047] ;

[0048] Where, 、 is the potential energy parameter, For particles With particles spacing;

[0049] ②For metals, the potential function of their phase transition The expression is:

[0050] ;

[0051] Where, For embedding energy, is the electron density, For particles With particles The spacing, For the double body position, 、 is the bond angle.

[0052] Periodic boundary conditions are selected to avoid size effects.

[0053] Based on Gibbs free energy calculation, the phase transition point is determined by order parameters (such as density, crystal orientation, etc.) to describe the nucleation process and the evolution of the phase interface.

[0054] Macroscopic thermodynamic parameters are obtained through statistical averaging, and multi-algorithm coupling is performed in step (2) to achieve mutual information transfer.

[0055] Develop and write code to implement execution strategies for molecular dynamics algorithms.

[0056] 2) The lattice Boltzmann algorithm is used in the computational domain B at and near the phase interface to simulate the melt flow and liquid / gas fluid flow under complex boundary conditions between dendrites, and to study the mesoscopic energy and mass transfer mechanisms within the energy storage phase change system, such as multi-physics field coupling within the energy storage, high-precision phase interface tracking, and crystal nucleus growth.

[0057] Figure 3 The intermediate view scale calculation part is the implementation process of the lattice Boltzmann algorithm.

[0058] The most critical part of lattice Boltzmann algorithm calculation is to establish a phase transition model and compile it into code.

[0059] The evolution equation of the lattice Boltzmann algorithm is:

[0060] ;

[0061] Where: is the time step, for time The particle at the position has a discrete velocity The distribution function in the direction, is the dimensionless relaxation time, is the equilibrium distribution function, is an external force source term used to describe multiphase interaction forces or phase change driving forces.

[0062] External force source The expression is as follows:

[0063] ;

[0064] Where, is the interaction force strength, For phase separation, is the density, is the phase distribution function.

[0065] In lattice Boltzmann calculations, the phase transition process simulation requires coupling the energy equation and the phase transition dynamics. The phase transition is achieved through the mass source term. accomplish:

[0066] ;

[0067] Where, is the phase change rate coefficient, is the density, is the saturation density.

[0068] During a phase transition, the phase interface changes as the transition progresses. In this application, moving boundary conditions are used to accurately track these interfaces. Mesh refinement is performed for interphase regions with significant information changes (e.g., density). Using a constructed multi-scale physical model (e.g., a microscopic molecular dynamics model, a mesoscopic Boltzmann model, and a macroscopic fluid dynamics model), information between regions of varying mesh density is transferred through multi-module spline interpolation, ensuring both computational accuracy and cost savings.

[0069] Through the evolution of the above-mentioned lattice Boltzmann algorithm, the macroscopic parameters (such as velocity, temperature, density, force, etc.) in the computational domain B are obtained using statistical averaging.

[0070] Develop code to implement the execution strategy of the Lattice Boltzmann algorithm.

[0071] 3) The entire computational domain A uses the macroscopic finite element / finite volume algorithm to simulate the macroscopic parameters such as the flow field and temperature field within the phase.

[0072] Different from the microscopic and mesoscopic algorithms, the macroscopic algorithm is based on the continuity assumption and simulates the macroscopic phase change process through finite element / finite volume numerical calculation methods and coupled phase change models.

[0073] Figure 3 The meso-macroscale calculation part is the macro-algorithm implementation process.

[0074] Set the phase equilibrium conditions:

[0075] ;

[0076] Where, For pressure, is the temperature, is the latent heat of phase change, is the difference in molar volumes between the two phases.

[0077] The condition for phase transition to occur is that the Gibbs free energy G of the two phases is equal, that is, .

[0078] The macroscopic flow, heat transfer and phase field distribution within the phase are obtained through macroscopic calculation.

[0079] Two or three algorithms can be run in parallel within the overlapping region, exchanging data in real time. While mature computational software is available for macroscopic computational models, coupling with microscopic and mesoscopic algorithms is still understudied. This application aims to achieve multi-scale and multi-phase field integration by developing a UDF code (see step (2)).

[0080] (2) Multi-phase field and multi-scale coupling

[0081] like Figure 2 As shown, the algorithm coupling of this application adopts parallel coupling. The calculation domains A, B, and C overlap with each other. Multiple algorithms calculate independently in the overlapping area and exchange data periodically to achieve real-time information transmission. The implementation process is as follows Figure 3 Different algorithms have different time and space scales, so the calculation parameters need to be interpolated in time and space to achieve multi-scale coupling of microscopic, mesoscopic, and macroscopic algorithms.

[0082] 1) Scale coupling

[0083] In a phase change energy storage system, phase change at the microscopic level is due to the release of latent heat due to the breaking or formation of molecular / atomic bonds. This phase change latent heat can be transferred to the lattice Boltzmann algorithm as a heat source term, and then the mapping of microscopic to mesoscopic information is achieved through the lattice Boltzmann temperature / energy evolution equation.

[0084] Different algorithms are not synchronized on the time scale, so a multi-time step strategy is needed to achieve asynchronous coupling, which is expressed as follows:

[0085] ;

[0086] Where, is the number of time steps, is the time step of the molecular dynamics algorithm, is the time step of the lattice Boltzmann algorithm.

[0087] Molecular dynamics runs with multiple time steps (at least one time step for the lattice Bolzmann algorithm / macroscopic continuum model) Finally, thermodynamic parameters such as velocity, interfacial energy, diffusion coefficient, latent heat of phase change (such as interfacial tension, etc.) are extracted in real time from molecular dynamics calculations and mapped to the lattice Boltzmann algorithm and macroscopic continuous model through relaxation method / time coarse-graining method.

[0088] Microscopic to mesoscopic information transmission:

[0089] when When, such as Figure 2 , the overlapping computational domain B can use the time coarse-graining method to map information. Taking particle velocity as an example, the average molecular velocity obtained by molecular dynamics through statistics is extracted. , and at the same time the number of molecules in the molecular dynamics calculation domain Adjust to the computational domain particles of the lattice Boltzmann algorithm and set the average molecular velocity Mapping to particle velocities in the lattice Boltzmann algorithm :

[0090] .

[0091] The same is true for other macro-quantity conversions.

[0092] For areas or time periods with drastic or rapid phase changes, the frequency of data information exchange needs to be increased to achieve dynamic adaptation. For data exchange, when compiling the code, the microscopic molecular dynamics model (see Figure 3 ) As a sub-loop of the lattice Boltzmann algorithm, LB extracts the latest molecular dynamics data in each iterative loop to achieve real-time information transmission.

[0093] 2) Bidirectional coupling

[0094] Information transmission from microscopic to mesoscopic:

[0095] The molecular / atomic information of the computational domain C is used to obtain the phase change latent heat by statistical averaging. Mapping to the lattice Boltzmann algorithm as the source term Computational domain B coupled to the lattice Boltzmann algorithm.

[0096] Then the lattice Boltzmann temperature field evolution equation is expressed as:

[0097] ;

[0098] Where, for time The particle at the position has a discrete velocity The temperature distribution function in the direction, is the corresponding dimensionless relaxation time, is the distribution function describing the temperature field, is the equilibrium temperature distribution function, is the weight coefficient.

[0099] The conversion of microscopic to mesoscopic heat source terms is achieved through the time coarse-graining method:

[0100] .

[0101] Information feedback from mesoscopic to microscopic:

[0102] The computational domain B of the lattice Boltzmann algorithm obtains macroscopic information such as the flow field, temperature field, and phase field of the phase change energy storage system, which serves as the external field for molecular dynamics calculations and affects the motion of molecules / atoms.

[0103] Fluid stress obtained by lattice Boltzmann calculation , which is converted into the external force term of particles in molecular dynamics through model coupling :

[0104] ;

[0105] Where, is the grid step size of the lattice Boltzmann algorithm, is the direction vector.

[0106] The mutual conversion coupling mechanism of other macro parameters is also the same, realizing the bidirectional coupling of the computational model.

[0107] 3) Boundary coupling

[0108] The lattice Boltzmann algorithm needs to match the dynamic behavior of the molecular dynamics region at the phase interface. Through appropriate time interpolation / extrapolation algorithms and compiled code, all parameters of the intermediate grid nodes on the phase interface (such as stress, heat flow, mass flow, etc.) are obtained. These parameters are passed to the molecular dynamics algorithm, realizing a bidirectional coupling mechanism for real-time information transmission and ensuring more accurate calculations.

[0109] Due to the complex boundary structure at the phase interface, the irregular boundary non-equilibrium state extrapolation method is used in the lattice Bolzmann algorithm:

[0110] Construct solid phase nodes adjacent to the phase interface Distribution function for:

[0111] ;

[0112] Where, 、 They are time The equilibrium and non-equilibrium distribution functions of particles at positions

[0113] like Figure 2 As shown in (b), macroscopic quantities (such as speed ) and the non-equilibrium part Through the fluid nodes adjacent to the phase interface 、 The non-equilibrium extrapolation of is obtained:

[0114] ;

[0115] ;

[0116] Where:

[0117] ;

[0118] ;

[0119] ;

[0120] For construction The speed at which the

[0121] Then the equilibrium state function of the solid phase node adjacent to the phase interface is for:

[0122] ;

[0123] Where, for The density of the structure, is the particle speed.

[0124] Therefore, the distribution function of the solid phase node adjacent to the phase interface after the collision step is obtained for:

[0125] ;

[0126] Where, for time The equilibrium distribution function constructed at .

[0127] Set up a buffer layer at the boundary of multi-algorithm coupling, such as Figure 2 As shown in (a), at the boundary of the computational domains A and B and The grid layer is a buffer layer. When the information of computational domain B is transferred to computational domain A, the buffer layer 、 Grid node information is not transmitted, ( : ) Real-time transmission of grid node information, taking speed as an example:

[0128] ;

[0129] In the buffer layer, flux matching is forced (such as mass, heat flux, etc. are constant values) to make the fluxes in algorithms of different dimensions consistent, avoiding energy leakage during information transmission and ensuring the continuity and conservation of physical quantities at the interface.

[0130] In addition, when computing domain A transfers information to computing domain B, the spatial scales are different, and the information on the smaller grid nodes in computing domain B needs to be obtained through interpolation. Figure 2 As shown in (c), taking the ratio of the grid step size of the computational domains A and B as 2:1 as an example, the grid node information of the overlapping parts of the computational domains A and B can be obtained through real-time exchange, but the information of the intermediate nodes of the computational domain B is unknown and needs to be interpolated in time and space. For example, the intermediate node " " can be obtained by spatial interpolation using the information of node "●" in the computational domain A. In order to eliminate the asymmetry in space, spline interpolation is used to obtain Moment Node "Post-collision distribution function at:

[0131] ;

[0132] Where, is the post-collision distribution function, 、 They are The first and second derivative functions of 、 、 、 is the coefficient, , , , Available , , The boundary conditions are obtained; For the boundary The number of nodes.

[0133] In addition, the evolution time of the computational domains A and B is not synchronized, so time interpolation is required to obtain (Right now ) at the moment of computing the information of all nodes in domain B. For example, it is necessary to obtain these unknown information through time interpolation. , , The node information at the moment is obtained by time interpolation Time post-collision distribution function :

[0134] ;

[0135] Where, for The post-collision distribution function at time .

[0136] 4) The coupling of macroscopic continuous models with microscopic and mesoscopic algorithms is a key and challenging aspect of multiphase field multiscale research. Based on the macroscopic model software, this application will integrate microscopic and mesoscopic mathematical models from open-port UDFs. By developing UDF code and coupling the interpolation and other coupling mechanisms in steps 1)-3) with macroscopic algorithm debugging, this will enable information transfer from microscopic and mesoscopic scales to the macroscopic computational model.

[0137] This application mainly constructs a multi-scale mathematical model for a multiphase energy storage phase change system, couples multiple physical fields such as phase change dynamics, heat transfer, and fluid flow within the phase change energy storage system, and couples microscopic, mesoscopic, and macroscopic algorithms in parallel in multiple computational domains, realizing real-time transmission of information from microscopic to macroscopic and macroscopic to microscopic. While retaining the microscopic phase change mechanism of the phase change energy storage system, it can efficiently simulate the macroscopic flow, heat transfer, and phase change process, providing a high-assurance computational model for the design and optimization of the phase change energy storage system.

[0138] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for constructing a multi-phase field multi-scale calculation model for phase change energy storage, characterized by: The following steps are involved: S1: The phase change energy storage system is divided into microscale, mesoscale and macroscale, and different computational domains are divided based on different scales. The computational domain of the microscale is the phase interface computational domain, the computational domain of the mesoscale is the phase interface and its vicinity computational domain, and the computational domain of the macroscale is each phase. S2: Construct the physical models of each computational domain, namely the physical model of the microscopic scale, the physical model of the mesoscopic scale, and the macroscopic continuum model of the macroscopic scale; S3: Parallel computation of the physical models of each computational domain, coupled during the parallel computation process, enables real-time information transfer between the computational domains. This includes transferring microscopic to mesoscopic and vice versa via relaxation and temporal coarse-graining methods. Through time and space interpolation / extrapolation algorithms, macroscopic scale information is transferred to mesoscopic scale, and mesoscopic scale information is transferred to microscopic scale, thus realizing real-time transmission of information at different scales in the phase change process.

2. The method for constructing a multi-phase field multi-scale calculation model for phase change energy storage according to claim 1, characterized in that: In the overlapping calculation domains of the microscale and mesoscale or the overlapping calculation domains of the mesoscale and macroscale, two corresponding physical models are used for simultaneous parallel coupling. In the overlapping calculation domains of the microscale, mesoscale and macroscale, three physical models are used for simultaneous parallel coupling. During parallel coupling, the physical models at each scale are calculated independently, and real-time information transmission is achieved through periodic data exchange.

3. The method for constructing a multi-phase field multi-scale calculation model for phase change energy storage according to claim 2, characterized in that: The physical model at the microscopic scale is implemented using a molecular dynamics algorithm to simulate intermolecular interaction forces, nucleation and crystal growth. A multiphase phase transition molecular dynamics model is constructed at the microscopic scale and the parameters are initialized. The potential function and boundary conditions are determined, and the phase transition point, nucleation and crystal growth are determined by free energy calculation. Then, the accuracy is judged. When the calculation accuracy meets the requirements, the macroscopic thermodynamic parameters are calculated by statistical averaging, and the macroscopic thermodynamic parameters calculated by statistical averaging are transferred to the mesoscopic scale physical model through relaxation method and time coarse-graining amplification. When the accuracy does not meet the requirements, the potential function and boundary conditions are re-determined, and the phase transition point, nucleation and crystal growth are re-determined.

4. The method for constructing a multi-phase field multi-scale calculation model for phase change energy storage according to claim 3, characterized in that: The physical model at the mesoscopic scale is implemented using the lattice Boltzmann algorithm, which is used to simulate the liquid / gas fluid flow under complex boundary conditions between melt flow and dendrites. A multiphase phase transition lattice Boltzmann model is constructed at the mesoscopic scale and the parameters are initialized to determine the equilibrium distribution function. Then, the evolution calculation of the flow and collision steps is performed, and the complex structure boundaries and moving boundaries are processed. Then, the accuracy is judged. When the calculation accuracy meets the requirements, the macroscopic parameters within the phase interface and its vicinity are calculated by statistical averaging, and the macroscopic parameters calculated by statistical averaging are transferred to the macroscopic physical model through relaxation method and time coarse-graining amplification. The macroscopic parameters are transferred to the multiphase phase transition molecular dynamics model through time and space interpolation / extrapolation algorithm. When the accuracy does not meet the requirements, the equilibrium distribution function is re-determined, and then the evolution calculation of the flow and collision steps is performed, and the complex structure boundaries and moving boundaries are processed.

5. The method for constructing a multi-phase field multi-scale calculation model for phase change energy storage according to claim 4, characterized in that: The macro-continuum model at the macro scale is implemented using a macro-finite element / finite volume algorithm to simulate the macroscopic parameters of the flow field and temperature field within the phase. A multiphase phase change energy storage model is constructed at the macro scale and imported into the calculation software through compiled code. The compiled code is coupled with the finite element / finite volume phase change model to determine the boundary conditions and then perform an accuracy judgment. When the calculation accuracy meets the requirements, the macroscopic flow field, temperature field, and phase field parameters are obtained and transferred to the multiphase phase change lattice Boltzmann model through time and space interpolation / extrapolation algorithms. If the calculation accuracy does not meet the requirements, the compiled code is re-coupled with the finite element / finite volume phase change model and the boundary conditions are determined.

6. The method for constructing a multi-phase field multi-scale calculation model for phase change energy storage according to any one of claims 1 to 5, characterized in that: In areas or time periods where phase changes are drastic or rapid, dynamic adaptation is achieved by increasing the frequency of data information exchange.

7. The method for constructing a multi-phase field multi-scale calculation model for phase change energy storage according to claim 4, characterized in that: At the boundary of the phase interface, the data information transmission between the mesoscale and microscale is realized by using the irregular boundary non-equilibrium state extrapolation method in the lattice Bolzmann algorithm.

8. The method for constructing a multi-phase field multi-scale calculation model for phase change energy storage according to claim 5, characterized in that: A buffer layer is set at the boundary where different physical models are coupled in the overlapping computational domain. The buffer layer is meshed, and the boundary mesh node information of the buffer layer is not transmitted. In addition, flux matching is forced in the buffer layer to ensure that the flux within the algorithm used by the physical models at different scales is consistent.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory, wherein: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images