Calculation method and system for non-fluid phase change molten salt heat exchange

Abstract

The invention provides a calculation method and system for non-fluid phase change molten salt heat exchange, and the method comprises the steps: constructing a self-adaptive thermodynamic model through a Bayesian optimization algorithm, and obtaining the thermodynamic characteristic description of a molten salt system based on the self-adaptive thermodynamic model; analyzing the thermodynamic characteristic description by applying a multi-scale modeling method, a discrete element method, a finite volume method and an implicit time integration scheme, and generating a target description of a heat transfer path in a non-fluid state; based on the target description, determining a mode recognition result by adopting an agent model technology and a deep neural network, and based on the mode recognition result, determining optimal heat exchange performance configuration; based on real-time fused salt temperature response data and optimal heat exchange performance configuration, a target operation strategy is generated by applying an evolutionary algorithm and a cooperative competition mechanism in a game theory framework, so that calculation of non-fluid phase change fused salt heat exchange is achieved; according to the method, the precision, efficiency and self-adaptability of fused salt heat exchange calculation are improved, and the requirements of efficient energy storage and conversion are met.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of computer simulation technology, and in particular to a calculation method and system for non-fluid phase-change molten salt heat exchange. Background Art

[0002] With the growing demand for efficient energy storage and conversion in renewable energy and industrial processes, non-fluid phase change molten salt heat exchange technology has shown great potential in many fields. Especially in application scenarios such as solar thermal power generation, industrial waste heat recovery, and power peak regulation, molten salt, as an efficient heat storage medium, can store a large amount of thermal energy under high temperature conditions and release it for power generation or heating when needed. However, these applications require accurate heat transfer calculation methods to ensure efficient operation and long-term stability of the system.

[0003] At present, the calculation methods for molten salt heat transfer mainly include methods based on empirical formulas and traditional numerical simulation. The existing methods based on empirical formulas rely on laboratory test data and engineering experience, and cannot accurately describe the dynamic behavior of molten salt under different temperature conditions; in addition, the existing methods generally have insufficient calculation accuracy, low calculation efficiency, and long calculation time, making it difficult to adapt to the needs of real-time monitoring and rapid adjustment; these methods are usually based on fixed parameter sets and assumptions, lacking adaptability and flexibility. Summary of the invention

[0004] The embodiments of the present invention provide a calculation method and system for non-fluid phase-change molten salt heat exchange, which are used to solve the problems of insufficient calculation accuracy, low efficiency, difficulty in adapting to the needs of real-time monitoring and rapid adjustment, and lack of adaptability and flexibility in the prior art.

[0005] In a first aspect, an embodiment of the present invention provides a method for calculating heat transfer of non-fluid phase-change molten salt, comprising:

[0006] A target state equation is selected by using a Bayesian optimization algorithm, and an adaptive thermodynamic model is constructed based on the target state equation, molten salt composition, and temperature conditions. Based on the adaptive thermodynamic model, a set of physical property parameters, latent heat changes, and thermodynamic behaviors of the molten salt at different temperatures are determined to obtain a thermodynamic characteristic description of the molten salt system;

[0007] Based on the thermodynamic property description, a multi-scale modeling method is applied to extract spatial distribution characteristics. Based on the spatial distribution characteristics, a discrete element method is used to generate a description of the kinetic behavior of the molten salt at the particle level. At the same time, a finite volume method is used to deal with the heat transfer problem to determine the temperature field distribution. In combination with the phase field theory and the thermodynamic property description, the evolution of the temperature field inside the molten salt and at the heat exchange interface between the molten salt and the outside world is processed through an implicit time integration scheme to obtain the temperature field evolution results. Based on the spatial distribution characteristics, the kinetic behavior description, the temperature field distribution and the temperature field evolution results, a target description of the heat transfer path under a non-fluid state is generated;

[0008] Based on the target description, the agent model technology is used to evaluate the influence of each phase change influencing factor to obtain the phase change influence evaluation result, the historical operation data is analyzed by deep neural network to obtain the historical data analysis result, and the pattern recognition result is determined by combining the historical data analysis result and the phase change influence evaluation result. Based on the pattern recognition result, the operating parameter configuration of the molten salt system is optimized to determine the optimal heat exchange performance configuration;

[0009] Based on the real-time molten salt temperature response data and the optimal heat exchange performance configuration, an evolutionary algorithm is applied to determine the target operating parameter set, and a topological structure is constructed based on the target operating parameter set. Based on the topological structure and the target operating parameter set, a cooperative competition mechanism in a game theory framework is adopted to generate a target operating strategy, which is used to guide the operation of the molten salt system to perform calculations on non-fluid phase change molten salt heat exchange.

[0010] Optionally, based on the thermodynamic property description, a multi-scale modeling method is applied to extract spatial distribution characteristics, based on the spatial distribution characteristics, a discrete element method is used to generate a description of the kinetic behavior of the molten salt particle level, and a finite volume method is used to deal with the heat transfer problem to determine the temperature field distribution, and the temperature field evolution inside the molten salt and at the heat exchange interface between the molten salt and the outside is processed by an implicit time integration scheme in combination with the phase field theory and the thermodynamic property description to obtain the temperature field evolution results, including:

[0011] Based on the thermodynamic characteristics description, microscopic molecular dynamics and macroscopic continuum mechanics, a multi-scale modeling method is applied to simulate and extract the spatial distribution characteristics of different scales in the molten salt system;

[0012] Setting material properties and contact mechanical behaviors of molten salt particles, applying discrete element method to simulate the interaction between molten salt particles and the influence of the interaction between molten salt particles on the molten salt system based on the material properties, contact mechanical behaviors and the thermodynamic characteristics description, and generating a description of the dynamic behavior of molten salt particles based on the interaction between molten salt particles and the influence of the interaction between molten salt particles on the molten salt system;

[0013] Based on the spatial distribution characteristics, an optimized grid is generated using an adaptive grid refinement technology and a topology optimization algorithm, and boundary conditions are set. Based on the optimized grid and the boundary conditions, a finite volume method is used to process the heat transfer problem to obtain simulation results. Based on the simulation results, the temperature field distribution in the molten salt system is finally determined. In the process of processing the heat transfer problem, a graph neural network is introduced to optimize the simulation results. The boundary conditions include grid boundary conditions and physical boundary conditions.

[0014] Combining the phase field theory and the description of the thermodynamic characteristics, the evolution of the temperature field inside the molten salt and at the heat exchange interface between the molten salt and the outside world is processed by an implicit time integration scheme to obtain the temperature field evolution results. In the process of the implicit time integration scheme, a nonlinear dynamic equation solver and an adaptive step control mechanism are introduced to obtain the change pattern of the temperature field over time.

[0015] Optionally, based on the spatial distribution characteristics, an optimized grid is generated using an adaptive grid refinement technique and a topology optimization algorithm, and boundary conditions are set; based on the optimized grid and the boundary conditions, a finite volume method is used to process the heat transfer problem to obtain a simulation result; based on the simulation result, the temperature field distribution in the molten salt system is finally determined; and a graph neural network is introduced in the process of processing the heat transfer problem to optimize the simulation result, including:

[0016] Based on the spatial distribution characteristics, the geometric structure and physical characteristics of the molten salt system are dynamically adjusted by using an adaptive mesh refinement technology to generate an initial mesh, a topology optimization algorithm is introduced to optimize the shape and distribution of the initial mesh, an optimized mesh is generated, and boundary conditions are set based on the optimized mesh;

[0017] Based on the optimized grid and the boundary conditions, a finite volume method is used to process the heat transfer problem to obtain a simulation result, wherein the simulation result includes a temperature field distribution to be optimized;

[0018] Based on the optimized grid, a graph neural network is constructed; based on the graph neural network, a connection pattern between nodes in the optimized grid is analyzed to identify heat transfer paths in a molten salt system and interactions between the heat transfer paths; based on the heat transfer paths and interactions between the heat transfer paths, a graph neural network model is generated;

[0019] The heat transfer path and the heat transfer efficiency of the heat transfer path are predicted based on the graph neural network model, and the heat transfer path and the heat transfer efficiency of the heat transfer path are optimized to obtain the heat transfer result, and the temperature field distribution to be optimized in the simulation result is optimized based on the heat transfer result, and finally the temperature field distribution in the molten salt system is determined.

[0020] Optionally, based on the target description, an agent model technology is used to evaluate the influence of each phase change influencing factor to obtain a phase change influence evaluation result, a deep neural network is used to analyze historical operation data to obtain a historical data analysis result, and a pattern recognition result is determined by combining the historical data analysis result and the phase change influence evaluation result. Based on the pattern recognition result, the operating parameter configuration of the molten salt system is optimized to determine the optimal heat exchange performance configuration, including:

[0021] A target proxy model is selected by using a proxy model technology, and a global sensitivity analysis method in the target proxy model is used to evaluate the influence of each phase change influencing factor on the overall heat exchange efficiency to obtain a phase change factor influence result. Based on the phase change factor influence result, a local sensitivity analysis method in the target proxy model is used to identify key factors that affect the overall heat exchange efficiency under specific operating conditions, and a phase change influence evaluation result is obtained based on the phase change factor influence result and the key factors. The phase change influencing factors include temperature change, latent heat release, and latent heat absorption. The specific operating conditions include high-temperature operating conditions and low-temperature starting conditions. The key factors include latent heat release rate and latent heat absorption rate.

[0022] Building a hybrid physical data-driven model based on physical laws and deep neural networks, analyzing historical operating data based on the hybrid physical data-driven model to obtain historical data analysis results;

[0023] Based on the historical data analysis results and the phase change impact assessment results, an advanced pattern recognition technology is used to identify the mutual relationship between the phase change influencing factors, and based on the key factors and the mutual relationship between the phase change influencing factors, a pattern recognition result is determined;

[0024] Based on the pattern recognition results, the Bayesian optimization algorithm is used to adjust the operating parameter configuration of the molten salt system to generate a preliminary optimized operating parameter configuration. A genetic algorithm is introduced to optimize the preliminary optimized operating parameter configuration to generate a final optimized operating parameter configuration. Based on the final optimized operating parameter configuration, the optimal heat exchange performance configuration is determined.

[0025] Optionally, a hybrid physical data-driven model is constructed based on physical laws and a deep neural network, and historical operation data is analyzed based on the hybrid physical data-driven model to obtain historical data analysis results, including:

[0026] Selecting a neural network structure, building a hybrid physical data-driven model combining prior physical knowledge based on the neural network structure, physical laws, and deep learning technology, analyzing historical operation data based on the hybrid physical data-driven model, and obtaining preliminary historical data analysis results;

[0027] Based on the preliminary historical data analysis results, using a long short-term memory network, time series prediction is performed on the historical operation data to obtain a time series prediction result;

[0028] Anomaly detection is performed on the preliminary historical data analysis results using an isolation forest algorithm to generate anomaly detection results, and based on the time series prediction results and the anomaly detection results, the preliminary historical data analysis results are optimized to obtain historical data analysis results.

[0029] Optionally, a target state equation is selected using a Bayesian optimization algorithm, and an adaptive thermodynamic model is constructed based on the target state equation, molten salt composition and temperature conditions. Based on the adaptive thermodynamic model, a set of physical property parameters, latent heat change and thermodynamic behavior of the molten salt at different temperatures are determined to obtain a thermodynamic characteristic description of the molten salt system, including:

[0030] Based on the Bayesian optimization algorithm and Gaussian process regression, the performance of different state equations is evaluated to obtain performance evaluation results, a target state equation is selected based on the performance evaluation results, and a physical information neural network is applied to construct an adaptive thermodynamic model based on the target state equation, molten salt composition and temperature conditions;

[0031] Based on the adaptive thermodynamic model, the physical properties of the molten salt at different temperatures are calculated to obtain a set of physical property parameters, wherein the physical properties include density, specific heat capacity and thermal conductivity;

[0032] Based on the physical property parameter set, simulating the latent heat release and latent heat absorption of the molten salt during the phase change process to calculate the latent heat change;

[0033] The thermodynamic behavior of the molten salt system is analyzed in combination with the physical property parameter set and the latent heat change, and a thermodynamic behavior description of the molten salt system is generated, wherein the thermodynamic behavior includes a phase change point, a phase change rate, and an energy conversion efficiency.

[0034] Optionally, combining the physical property parameter set and the latent heat change, analyzing the thermodynamic behavior of the molten salt system, and generating a thermodynamic behavior description of the molten salt system, the thermodynamic behavior including the phase change point, the phase change rate and the energy conversion efficiency, including:

[0035] Based on the physical property parameter set and the latent heat change, identifying a phase transition temperature point of the molten salt from solid to liquid or from liquid to solid;

[0036] Based on the phase change temperature point, the temperature change rate and the latent heat change amount, the speed of state conversion of the molten salt during the phase change process is calculated to obtain the phase change rate;

[0037] Based on the phase change rate, the energy transfer path and loss mechanism of the molten salt system are evaluated by using coupled fluid dynamics simulation and finite element analysis method to obtain energy transfer evaluation results;

[0038] Based on the energy transfer evaluation result and the phase change rate, the energy conversion efficiency of the molten salt system during the heat exchange process is evaluated, and a description of the thermodynamic behavior of the molten salt system is generated by combining the phase change rate, the energy transfer evaluation result and the energy conversion efficiency.

[0039] In a second aspect, an embodiment of the present invention provides a computing system for non-fluid phase-change molten salt heat exchange, comprising:

[0040] A construction module is used to select a target state equation using a Bayesian optimization algorithm, and to construct an adaptive thermodynamic model based on the target state equation, molten salt composition, and temperature conditions. Based on the adaptive thermodynamic model, a set of physical property parameters, a latent heat change, and a thermodynamic behavior of the molten salt at different temperatures are determined to obtain a thermodynamic characteristic description of the molten salt system;

[0041] A determination module is used to extract spatial distribution characteristics based on the thermodynamic characteristic description and apply a multi-scale modeling method. Based on the spatial distribution characteristics, a discrete element method is used to generate a description of the kinetic behavior of the molten salt particle level. At the same time, a finite volume method is used to process the heat transfer problem to determine the temperature field distribution. In combination with the phase field theory and the thermodynamic characteristic description, the temperature field evolution inside the molten salt and at the heat exchange interface between the molten salt and the outside world is processed through an implicit time integration scheme to obtain the temperature field evolution result. Based on the spatial distribution characteristics, the kinetic behavior description, the temperature field distribution and the temperature field evolution result, a target description of the heat transfer path under a non-fluid state is generated;

[0042] an optimization module, for evaluating the influence of each phase change influencing factor based on the target description by using a proxy model technology to obtain a phase change influence evaluation result, analyzing historical operation data by using a deep neural network to obtain a historical data analysis result, combining the historical data analysis result with the phase change influence evaluation result to determine a pattern recognition result, and optimizing the operating parameter configuration of the molten salt system based on the pattern recognition result to determine an optimal heat exchange performance configuration;

[0043] A generation module is used to determine a target operating parameter set based on real-time molten salt temperature response data and the optimal heat exchange performance configuration by applying an evolutionary algorithm, construct a topological structure based on the target operating parameter set, generate a target operating strategy based on the topological structure and the target operating parameter set by adopting a cooperative competition mechanism under a game theory framework, and realize the calculation of non-fluid phase change molten salt heat exchange based on the target operating strategy.

[0044] In a third aspect, an embodiment of the present invention provides a computing device, comprising a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute a calculation method for non-fluid phase-change molten salt heat exchange as described in any one of the first aspects.

[0045] In a fourth aspect, an embodiment of the present invention provides a computer storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement a calculation method for non-fluid phase-change molten salt heat exchange as described in any one of the first aspects.

[0046] In an embodiment of the present invention, a Bayesian optimization algorithm is used to select a target state equation, and an adaptive thermodynamic model is constructed based on the target state equation, molten salt composition and temperature conditions. Based on the adaptive thermodynamic model, a set of physical property parameters, latent heat change and thermodynamic behavior of the molten salt at different temperatures are determined to obtain a thermodynamic characteristic description of the molten salt system; based on the thermodynamic characteristic description, a multi-scale modeling method is applied to extract spatial distribution characteristics, and based on the spatial distribution characteristics, a discrete element method is used to generate a kinetic behavior description of the molten salt particle level, and a finite volume method is used to handle heat transfer problems to determine the temperature field distribution, and in combination with phase field theory and the thermodynamic characteristic description, an implicit time integration scheme is used to process the temperature field evolution inside the molten salt and at the heat exchange interface between the molten salt and the outside world to obtain a temperature field evolution result, based on the spatial distribution characteristics, the kinetic behavior description, the temperature field distribution and The temperature field evolution result generates a target description of the heat transfer path under the non-fluid state; based on the target description, the agent model technology is used to evaluate the influence of each phase change influencing factor to obtain the phase change impact evaluation result, and the historical operation data is analyzed by deep neural network to obtain the historical data analysis result, and the historical data analysis result and the phase change impact evaluation result are combined to determine the pattern recognition result, and the operating parameter configuration of the molten salt system is optimized based on the pattern recognition result to determine the optimal heat exchange performance configuration; based on the real-time molten salt temperature response data and the optimal heat exchange performance configuration, the evolutionary algorithm is used to determine the target operating parameter set, and the topological structure is constructed based on the target operating parameter set. Based on the topological structure and the target operating parameter set, the cooperative competition mechanism in the game theory framework is used to generate the target operating strategy, and the target operating strategy is used to guide the operation of the molten salt system to calculate the heat exchange of non-fluid phase change molten salt. The technical solution provided by the present invention significantly improves the accuracy, efficiency and adaptability of molten salt heat exchange calculation. These technical effects not only meet the needs of efficient energy storage and conversion in practical applications, but also provide strong support for future energy management and industrial process optimization.The prediction accuracy is improved by combining adaptive thermodynamic models, multi-scale modeling and discrete element method. The adaptive thermodynamic model can be dynamically adjusted to adapt to different temperature conditions and changes in molten salt composition. The generated kinetic behavior description ensures the comprehensive capture of complex microstructures and macroscopic heat transfer processes. The use of finite volume method, implicit time integration scheme method and agent model technology not only improves the accuracy of simulation, but also significantly reduces the calculation time, which helps to achieve real-time monitoring and rapid adjustment. Through deep neural network analysis, evolutionary algorithm and game theory framework, the long-term stability and optimal performance of the system are ensured, and it has a high degree of adaptability and can cope with complex dynamic environmental changes. Through the above comprehensive method, the target operation strategy is finally generated to achieve precise control and efficient management of the molten salt system, ensuring that the system can maintain the best heat transfer performance under various conditions, and ensuring that the calculation method can accurately and efficiently simulate the heat transfer process of the molten salt system, providing reliable technical support for practical applications. Among them, by providing a detailed description of the heat transfer path, the understanding of the internal heat transfer mechanism of the molten salt system is enhanced, providing data support for optimizing the heat transfer path; by predicting and optimizing the heat transfer path and heat transfer efficiency through the graph neural network model, a more uniform and accurate temperature field distribution is ultimately determined, which significantly improves the accuracy of the simulation and the heat exchange efficiency of the system.

[0047] These and other aspects of the present invention will become more apparent from the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0049] Figure 1 A flow chart of a method for calculating heat exchange of non-fluid phase-change molten salt provided in an embodiment of the present invention;

[0050] Figure 2 A schematic diagram of the structure of a calculation system for non-fluid phase-change molten salt heat exchange provided by an embodiment of the present invention;

[0051] Figure 3 A schematic diagram of the structure of a computing device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0053] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0054] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0055] Figure 1 A flow chart of a method for calculating heat exchange of non-fluid phase-change molten salt is provided for an embodiment of the present invention, such as Figure 1 As shown, the method includes:

[0056] In the field of non-fluid phase-change molten salt heat exchange, the existing technology still has many deficiencies in accurate simulation, real-time optimization and adaptive adjustment. In order to solve the limitations of the existing technology and provide a new solution for the efficient operation of complex molten salt systems, based on this, the present invention provides a calculation method for non-fluid phase-change molten salt heat exchange, such as Figure 1 ,include:

[0057] Step 101: Select a target state equation using a Bayesian optimization algorithm, construct an adaptive thermodynamic model based on the target state equation, molten salt composition and temperature conditions, and determine a set of physical property parameters, latent heat variation and thermodynamic behavior of the molten salt at different temperatures based on the adaptive thermodynamic model to obtain a thermodynamic characteristic description of the molten salt system;

[0058] In this step, the target state equation refers to a mathematical expression that describes the thermodynamic behavior of molten salt, such as the ideal gas state equation, the van der Waals equation, etc., which can reflect the physical properties of molten salt at different temperatures and pressures.

[0059] Suppose there is a solar thermal power generation system in which molten salt is used as a heat storage medium. Using the Bayesian optimization algorithm, the state equation that is most suitable as the target state equation is selected. First, a set of candidate state equations, such as the ideal gas state equation, the van der Waals equation, and the Dividend-Amsterdam equation, are defined. These equations are evaluated based on experimental data or historical operation data, and the best parameter combination is selected iteratively to finally determine the state equation that best describes the molten salt system. Based on the selected target state equation, molten salt composition, and temperature conditions, the molten salt composition is such as a mixture of sodium nitrate and potassium nitrate, and the temperature conditions are such as 300°C to 800°C. An adaptive thermodynamic model is constructed, which can dynamically adjust its internal parameters to adapt to different operating conditions. Using the adaptive thermodynamic model, the physical property parameter set, latent heat change, and thermodynamic behavior of the molten salt at different temperatures are calculated. The physical property parameter set is such as density, specific heat capacity, thermal conductivity, latent heat change such as melting latent heat, and thermodynamic behavior such as phase transition point. For example, the model can predict that the density of molten salt at 400°C is 1.7g / cm 3 , the specific heat capacity is 1.2 J / g·K, and the latent heat of fusion is 200 J / g. Based on the above results, the thermodynamic characteristics of the molten salt system are determined to describe

[0060] Step 102: Based on the thermodynamic property description, a multi-scale modeling method is applied to extract spatial distribution characteristics. Based on the spatial distribution characteristics, a discrete element method is used to generate a description of the kinetic behavior of the molten salt particle level. At the same time, a finite volume method is used to process the heat transfer problem to determine the temperature field distribution. In combination with the phase field theory and the thermodynamic property description, the temperature field evolution inside the molten salt and at the heat exchange interface between the molten salt and the outside is processed through an implicit time integration scheme to obtain a temperature field evolution result. Based on the spatial distribution characteristics, the kinetic behavior description, the temperature field distribution and the temperature field evolution result, a target description of the heat transfer path under a non-fluid state is generated;

[0061] This step applies multi-scale modeling methods to extract spatial distribution characteristics within the molten salt system from microscopic to macroscopic levels, such as particle-level structure at the microscopic level and overall temperature field distribution at the macroscopic level. For example, analyze the arrangement of molten salt particles in the container and its impact on overall heat transfer. Use discrete element method to simulate the interaction between molten salt particles and generate a particle-level description of dynamic behavior. For example, simulate how particles move and rearrange during heating, affecting the heat transfer path. Use finite volume method to deal with heat transfer problems and determine the temperature field distribution. For example, simulate the change of temperature of molten salt with time and position during heating. Combine phase field theory and thermodynamic property description to deal with the evolution of temperature field inside molten salt and at the external heat exchange interface through implicit time integration scheme. For example, simulate the dynamic change of temperature field during the transition of molten salt from solid to liquid. Combine the above information to generate a target description of heat transfer path in non-fluid state. For example, determine which areas have the highest heat transfer efficiency and which areas need further optimization.

[0062] Step 103: Based on the target description, the agent model technology is used to evaluate the influence of each phase change influencing factor to obtain a phase change influence evaluation result, the historical operation data is analyzed by a deep neural network to obtain a historical data analysis result, and the historical data analysis result and the phase change influence evaluation result are combined to determine a pattern recognition result, and based on the pattern recognition result, the operating parameter configuration of the molten salt system is optimized to determine the optimal heat exchange performance configuration;

[0063] In this step, the pattern recognition results refer to the optimization suggestions for operating parameters obtained through comprehensive analysis of historical data analysis and phase change impact assessment results.

[0064] This step uses proxy model technology to evaluate the impact of various phase change influencing factors, such as temperature change, latent heat release, etc. For example, evaluate the impact of latent heat release on the overall heat transfer efficiency under different temperature conditions. Use deep neural networks to analyze historical operating data and identify the relationship between phase change influencing factors. For example, analyze the operating data of the molten salt system under different operating conditions in the past year to find the best operating conditions. Combine the phase change impact assessment results and historical data analysis results to determine the pattern recognition results. For example, determine which operating parameter configuration can achieve the best heat transfer performance within a specific temperature range. Based on the pattern recognition results, optimize the operating parameter configuration of the molten salt system and determine the optimal heat transfer performance configuration. For example, adjust parameters such as heating rate and cooling rate to maximize heat transfer efficiency.

[0065] Step 104: Based on the real-time molten salt temperature response data and the optimal heat exchange performance configuration, an evolutionary algorithm is applied to determine a target operating parameter set, a topology structure is constructed based on the target operating parameter set, and based on the topology structure and the target operating parameter set, a cooperative competition mechanism in a game theory framework is adopted to generate a target operating strategy, wherein the target operating strategy is used to guide the operation of the molten salt system to perform calculations of non-fluid phase change molten salt heat exchange;

[0066] This step uses an evolutionary algorithm to determine the target operating parameter set based on the real-time molten salt temperature response data and the optimal heat transfer performance configuration. For example, parameters such as heating power and cooling rate are adjusted according to the current molten salt temperature and the historical optimal configuration. The topology of the molten salt system is constructed based on the target operating parameter set. For example, the optimal layout of heaters, coolers, and sensors is determined. Based on the topology and the target operating parameter set, the cooperative competition mechanism in the game theory framework is used to generate the target operating strategy. For example, a set of operating rules is formulated to ensure that the system can maintain optimal performance under different operating conditions.

[0067] The embodiments of the present invention can achieve the following beneficial effects through the above steps:

[0068] The Bayesian optimization algorithm is used to select the target state equation that best describes the specific molten salt system, ensuring that the basic data for subsequent model construction is accurate and reliable;

[0069] The adaptive thermodynamic model can dynamically adjust its internal parameters according to the composition and temperature of the molten salt, providing a high-precision description of the thermodynamic characteristics and ensuring accurate prediction of the physical properties, latent heat changes and thermodynamic behavior of the molten salt at different temperatures.

[0070] The application of each method improves the calculation accuracy and efficiency;

[0071] By optimizing the heat transfer path and operating parameters, the energy loss during the transmission process is reduced, the energy utilization rate of the entire system is improved, and a more environmentally friendly and economical operation mode is achieved;

[0072] The generated target operation strategy can flexibly adjust the operating parameters under different working conditions, adapt to complex dynamic environmental changes, and enhance the system's adaptive ability.

[0073] After constructing the thermodynamic characteristics description of the molten salt system, in order to further accurately simulate the complex heat transfer behavior in the non-fluid phase change process, based on this, the present invention provides a specific embodiment, step 102, based on the thermodynamic characteristics description, applying a multi-scale modeling method, extracting spatial distribution characteristics, based on the spatial distribution characteristics, using a discrete element method to generate a kinetic behavior description of the molten salt particle level, and using a finite volume method to deal with the heat transfer problem to determine the temperature field distribution, combining phase field theory and the thermodynamic characteristics description, and processing the temperature field evolution inside the molten salt and at the heat exchange interface between the molten salt and the outside through an implicit time integration scheme to obtain a temperature field evolution result, based on the spatial distribution characteristics, the kinetic behavior description, the temperature field distribution and the temperature field evolution result, Generate a target description of the heat transfer path under a non-fluid state, specifically including the following steps:

[0074] Step 210: Based on the thermodynamic property description, microscopic molecular dynamics, and macroscopic continuum mechanics, a multi-scale modeling method is applied to simulate and extract spatial distribution characteristics at different scales in the molten salt system;

[0075] This step uses microscopic molecular dynamics to simulate the interactions between molten salt particles and their contribution to the overall system. For example, simulate the vibration and movement behavior of a single molten salt particle under high temperature conditions. Use macroscopic continuum mechanics to analyze the heat transfer and flow characteristics of the molten salt system on a larger scale. For example, evaluate the temperature distribution and fluid flow throughout the molten salt storage tank. Combine the thermodynamic characterization, microscopic and macroscopic simulation results to extract the spatial distribution characteristics at different scales within the molten salt system. For example, determine the arrangement of molten salt particles in the container and its impact on the overall heat transfer.

[0076] Step 211: setting material properties and contact mechanical behaviors of molten salt particles, applying discrete element method to simulate the interaction between molten salt particles and the influence of the interaction between molten salt particles on the molten salt system based on the material properties, contact mechanical behaviors and the thermodynamic characteristics description, and generating a description of the dynamic behavior of molten salt particles based on the interaction between molten salt particles and the influence of the interaction between molten salt particles on the molten salt system;

[0077] In this step, the material properties and contact mechanical behavior of the molten salt particles are set according to the molten salt composition and experimental data. The material properties, such as density, are 1.7 g / cm 3 , elastic modulus of 50 GPa, contact mechanical behavior such as friction coefficient of 0.3; use discrete element method to simulate the interaction between molten salt particles and generate a description of dynamic behavior. For example, simulate how particles move and rearrange during heating and the impact of these changes on the heat transfer path. Combining the above information, generate a description of the dynamic behavior of molten salt particles. For example, describe how particles change their position and state under different temperature conditions, affecting the overall heat transfer efficiency.

[0078] Step 212: Based on the spatial distribution characteristics, an optimized grid is generated using an adaptive grid refinement technique and a topology optimization algorithm, and boundary conditions are set. Based on the optimized grid and the boundary conditions, a finite volume method is used to process the heat transfer problem to obtain simulation results. Based on the simulation results, the temperature field distribution in the molten salt system is finally determined. In the process of processing the heat transfer problem, a graph neural network is introduced to optimize the simulation results. The boundary conditions include grid boundary conditions and physical boundary conditions.

[0079] In this step, boundary conditions include mesh boundary conditions and physical boundary conditions, mesh boundaries such as fixed temperature or adiabatic boundaries, and physical boundaries such as heater positions and powers.

[0080] This step uses adaptive mesh refinement technology and topology optimization algorithms to generate optimized meshes and set boundary conditions. For example, fixed temperature boundary conditions are set at the inlet and outlet of the molten salt storage tank, and heating power boundary conditions are set at the heater position. Based on the optimized mesh and boundary conditions, the finite volume method is used to deal with the heat transfer problem and obtain simulation results. For example, the temperature change of molten salt during the heating process is simulated to determine the temperature field distribution. In the process of dealing with heat transfer problems, graph neural networks are introduced to optimize the simulation results. For example, graph neural networks are used to capture the complex relationship between molten salt particles and improve the accuracy of temperature field distribution prediction.

[0081] Step 213: combining the phase field theory and the thermodynamic characteristics description, processing the temperature field evolution inside the molten salt and at the heat exchange interface between the molten salt and the outside through an implicit time integration scheme to obtain the temperature field evolution result. In the process of implicit time integration scheme processing, a nonlinear dynamic equation solver and an adaptive step control mechanism are introduced to obtain the change law of the temperature field over time;

[0082] This step combines phase field theory and thermodynamic properties to process the evolution of the temperature field inside the molten salt and at the external heat exchange interface through an implicit time integration scheme. For example, the dynamic changes of the temperature field during the transition of the molten salt from solid to liquid are simulated. A nonlinear dynamic equation solver and an adaptive step control mechanism are introduced into the implicit time integration scheme to obtain the change law of the temperature field over time. For example, a smaller time step is ensured near the phase change point to capture the rapidly changing temperature field.

[0083] The embodiments of the present invention can achieve the following beneficial effects through the above steps:

[0084] By providing a description of particle-level kinetic behavior, it helps to understand the complex microstructure inside the molten salt system and its impact on heat transfer;

[0085] The accurate temperature field distribution is determined through adaptive mesh refinement technology, graph neural network and other algorithms, which improves the simulation accuracy and efficiency of heat transfer problems;

[0086] By processing the evolution of the temperature field, the variation law of the temperature field over time is obtained, ensuring the accurate simulation and prediction of the phase change process.

[0087] After extracting the spatial distribution characteristics of different scales in the molten salt system, in order to further accurately simulate the heat transfer process and optimize the calculation results, based on this, the present invention provides a specific embodiment, step 212, based on the spatial distribution characteristics, using adaptive grid refinement technology and topology optimization algorithm, generate an optimized grid, and set boundary conditions, based on the optimized grid and the boundary conditions, use the finite volume method to process the heat transfer problem, obtain simulation results, based on the simulation results, finally determine the temperature field distribution in the molten salt system, introduce a graph neural network in the process of processing the heat transfer problem to optimize the simulation results, specifically including the following steps:

[0088] Step 221: Based on the spatial distribution characteristics, the geometric structure and physical characteristics of the molten salt system are dynamically adjusted by using an adaptive mesh refinement technology to generate an initial mesh, a topology optimization algorithm is introduced to optimize the shape and distribution of the initial mesh, an optimized mesh is generated, and boundary conditions are set based on the optimized mesh;

[0089] In this step, the spatial distribution characteristics refer to the physical and geometric characteristics at different scales in the molten salt system extracted by the multi-scale modeling method. The initial grid refers to the basic grid generated based on the geometric structure and physical characteristics of the molten salt system.

[0090] This step uses adaptive mesh refinement technology to dynamically adjust the geometric structure and physical properties of the molten salt system based on the spatial distribution characteristics. For example, in key areas of the molten salt storage tank, such as near the heater, a finer mesh is used, and in areas with smaller temperature changes, a coarser mesh is used to reduce the amount of calculation and improve accuracy. A topology optimization algorithm is introduced to optimize the shape and distribution of the initial mesh to generate an optimized mesh. For example, by adjusting the position and connection method of the mesh nodes, it is ensured that the mesh can still maintain high quality in complex geometric structures while minimizing the consumption of computing resources. Boundary conditions are set based on the optimized mesh. For example, fixed temperature boundary conditions are set at the inlet and outlet of the molten salt storage tank, and heating power boundary conditions are set at the heater position to ensure the authenticity and reliability of the simulation.

[0091] Step 222: Based on the optimized grid and the boundary conditions, a finite volume method is used to process the heat transfer problem to obtain a simulation result, wherein the simulation result includes a temperature field distribution to be optimized;

[0092] This step uses the finite volume method to process the heat transfer problem based on the optimized grid and the boundary conditions to obtain simulation results. The simulation results include the temperature field distribution to be optimized, that is, the temperature field distribution obtained by preliminary simulation, which needs to be further optimized to improve its accuracy. For example, the temperature change of molten salt during the heating process is simulated to preliminarily determine the temperature field distribution to be optimized.

[0093] Step 223: constructing a graph neural network based on the optimized grid, analyzing the connection pattern between nodes in the optimized grid based on the graph neural network to identify heat transfer paths in the molten salt system and interactions between the heat transfer paths, and generating a graph neural network model based on the heat transfer paths and interactions between the heat transfer paths;

[0094] This step builds a graph neural network based on the optimized grid, analyzes the connection pattern between the nodes in the optimized grid, and identifies the heat transfer paths and the interactions between the heat transfer paths in the molten salt system. For example, identify which connections between nodes are the main heat transfer paths and which paths have significant interactions. Generate a graph neural network model based on the heat transfer paths and the interactions between the heat transfer paths. For example, build a graph neural network model that can predict heat transfer paths and interactions between paths to ensure that the model can capture complex heat transfer behaviors.

[0095] Step 224: predicting the heat transfer path and the heat transfer efficiency of the heat transfer path based on the graph neural network model, and optimizing the heat transfer path and the heat transfer efficiency of the heat transfer path to obtain a heat transfer result, optimizing the temperature field distribution to be optimized in the simulation result based on the heat transfer result, and finally determining the temperature field distribution in the molten salt system;

[0096] This step predicts the heat transfer path and the heat transfer efficiency of the heat transfer path based on the graph neural network model. For example, predict which heat transfer paths have the highest transfer efficiency and which paths need further optimization. Optimize the heat transfer path and the heat transfer efficiency of the heat transfer path according to the prediction results. For example, adjust the geometric structure or material properties of certain heat transfer paths to improve the overall heat transfer efficiency. Optimize the temperature field distribution to be optimized in the simulation results based on the heat transfer results, and finally determine the temperature field distribution in the molten salt system. For example, by optimizing the heat transfer path, the temperature field distribution is made more uniform and local overheating is reduced.

[0097] More specifically, the present invention provides a calculation formula for temperature field distribution to generate a target description of the heat transfer path in a non-fluid state. The specific calculation formula is as follows:

[0098] T final =OPT(T pre ,Q pred);

[0099] Among them, T final represents the final temperature field distribution after optimization; OPT represents the optimization function, which is used to adjust the temperature field distribution to be optimized to meet the optimized heat transfer path and heat transfer efficiency; T pre represents the temperature field distribution to be optimized; Q pred Indicates the heat transfer result;

[0100] The temperature field distribution to be optimized may have areas with uneven distribution or large energy loss. The above temperature field distribution formula can optimize these areas in a targeted manner by introducing the heat transfer path and heat transfer efficiency provided by the heat transfer result formula, improve the overall energy utilization efficiency, and effectively avoid local overheating or insufficient cooling. This formula ensures that heat is transferred along the optimal path, improves the overall efficiency of the system, and the optimized temperature field distribution is more uniform, reducing the stress concentration problem caused by the temperature gradient and extending the service life of the equipment.

[0101] The calculation formula of heat transfer result is as follows:

[0102] Q pred =GNN model (G opt ,C n ,HTP,HTPI);

[0103] Among them, Q pred Represents the heat transfer result, which is used to optimize the temperature field distribution to be optimized; GNN model represents the graph neural network model; G opt represents the optimized grid; C n represents the connection pattern between grid nodes; HTP represents the heat transfer path; HTPI represents the interaction between heat transfer paths;

[0104] The heat transfer process inside the molten salt system involves complex geometric structures and variable physical properties. Traditional numerical methods have difficulty in efficiently handling these complexities. The graph neural network model can capture the nonlinear relationship between nodes and adapt to different grid structures to provide more accurate simulations. The above heat transfer result formula introduces the graph neural network model to more accurately predict the heat transfer path and efficiency in the molten salt system, providing reliable basic data for subsequent optimization.

[0105] The calculation formula of heat transfer path is as follows:

[0106]

[0107] Where HTP represents the heat transfer path; argmin p Indicates that among all possible heat transfer paths p, select the one that can make ∑i,j w ij (T i -T j ) 2 The p that minimizes; ∑ i,j represents the sum of all relevant node pairs (i, j), where the relevant nodes refer to the node pairs that have the possibility of heat transfer physically; w ij is the weight coefficient from node i to node j; T i represents the temperature of node i; T j represents the temperature of node j; subject to p∈P means that the optimization process is subject to the restriction that the heat transfer path p finally selected must belong to the heat transfer path set P;

[0108] Minimizing energy loss is one of the important goals of thermal management. The heat transfer path formula can be used to find the most efficient heat transfer path from one node to another, that is, the one with the least energy loss. By minimizing the sum of the squares of the temperature differences between adjacent nodes, the formula ensures that heat flows in the direction of least resistance, which is in line with the laws of heat conduction in physics. The weight coefficient can be used to represent the thermal conductivity of different materials or paths, so the formula can be adjusted according to actual physical conditions.

[0109] The calculation formula of heat transfer path is as follows:

[0110] HTPI=∑ i,j (F ij ·n ij )exp(-α|T i -T j |);

[0111] Where HTPI represents the interaction between heat transfer paths; ∑ i,j represents the sum of all relevant node pairs (i, j), where the relevant nodes refer to the node pairs that have the possibility of heat transfer physically; F ij represents the heat transfer flux from node i to node j; n ij Represents the unit normal vector between nodes; exp(-α|T i -T j |) represents the exponential decay function, α represents the decay factor, |T i -T j | represents the absolute value of the temperature difference between nodes i and j.

[0112] In real applications, multiple heat transfer paths may affect each other, such as paths crossing due to space constraints or heat flow direction changing due to temperature gradient changes. The above heat transfer path formula takes these factors into account to more accurately simulate real-world phenomena. The formula introduces a decay factor, which reflects the phenomenon that the product of heat transfer flux and the unit normal vector gradually decreases as the temperature difference increases; this can not only simulate direct heat transfer effects, but also capture indirect effects such as thermal radiation or convection heat transfer.

[0113] The embodiments of the present invention can achieve the following beneficial effects through the above steps:

[0114] It provides high-quality and efficient initial mesh, laying the foundation for the subsequent treatment of heat transfer problems and ensuring the accuracy and efficiency of calculations;

[0115] The preliminary temperature field distribution was determined, providing basic data for subsequent optimization and improving the accuracy of the simulation;

[0116] Provides a detailed description of the heat transfer path, enhances the understanding of the internal heat transfer mechanism of the molten salt system, and provides data support for optimizing the heat transfer path;

[0117] By predicting and optimizing the heat transfer path and heat transfer efficiency through the graph neural network model, the precise temperature field distribution was ultimately determined, significantly improving the simulation accuracy and the heat transfer efficiency of the system.

[0118] After generating a target description of the heat transfer path under a non-fluid state, in order to further optimize the heat transfer performance of the molten salt system, based on this, the present invention provides a specific embodiment, step 103, based on the target description, using the proxy model technology to evaluate the influence of each phase change influencing factor, to obtain a phase change influence evaluation result, using a deep neural network to analyze historical operation data, to obtain a historical data analysis result, combining the historical data analysis result and the phase change influence evaluation result, determining a pattern recognition result, based on the pattern recognition result, optimizing the operating parameter configuration of the molten salt system to determine the optimal heat transfer performance configuration, specifically comprising the following steps:

[0119] Step 311: select a target proxy model using proxy model technology, use the global sensitivity analysis method in the target proxy model to evaluate the influence of each phase change influencing factor on the overall heat exchange efficiency, obtain the phase change factor influence result, based on the phase change factor influence result, use the local sensitivity analysis method in the target proxy model to identify the key factors affecting the overall heat exchange efficiency under specific working conditions, and obtain the phase change influence evaluation result based on the phase change factor influence result and the key factors, the phase change influencing factors include temperature change, latent heat release and latent heat absorption, the specific working conditions include high temperature operating conditions and low temperature starting conditions, and the key factors include latent heat release rate and latent heat absorption rate;

[0120] In this step, the phase change influencing factors refer to the factors that affect the heat transfer performance of the molten salt system, such as temperature change, latent heat release, and latent heat absorption.

[0121] This step uses the proxy model technology to select a target proxy model suitable for evaluating phase change influencing factors. For example, Gaussian process regression is selected as the target proxy model because it can handle nonlinear relationships well and provide uncertainty estimates. The global sensitivity analysis method in the target proxy model is used to evaluate the degree of influence of each phase change influencing factor on the overall heat transfer efficiency. For example, through global sensitivity analysis, it is found that latent heat release has a significant impact on the heat transfer efficiency under high temperature operating conditions. Based on the phase change factor impact results, the local sensitivity analysis method in the target proxy model is used to identify the key factors affecting the overall heat transfer efficiency under specific conditions. For example, under high temperature operating conditions, the latent heat release rate is the key factor; while under low temperature starting conditions, the latent heat absorption rate is more important. Combining the above analysis results, the phase change impact assessment results are obtained, and it is clear which phase change influencing factors have the greatest impact on the heat transfer efficiency under different conditions. For example, it is determined that the latent heat release rate should be focused on during high temperature operation, while the latent heat absorption rate should be optimized during low temperature startup.

[0122] Step 312: constructing a hybrid physical data-driven model based on physical laws and a deep neural network, and analyzing historical operation data based on the hybrid physical data-driven model to obtain historical data analysis results;

[0123] This step builds a hybrid physical data-driven model based on physical laws and deep neural networks. For example, the heat conduction equation is used as the basis of the model, and the deep neural network is used to capture the nonlinear relationship in the data. The hybrid physical data-driven model is used to analyze the historical operating data of the molten salt system to obtain historical data analysis results. For example, the operating data of the molten salt system under different operating conditions in the past year is analyzed to find out the optimal operating conditions and potential problems.

[0124] Step 313: Based on the historical data analysis results and the phase change impact assessment results, an advanced pattern recognition technology is used to identify the relationship between the phase change impact factors, and based on the key factors and the relationship between the phase change impact factors, a pattern recognition result is determined;

[0125] This step is based on the historical data analysis results and the phase change impact assessment results, and uses advanced pattern recognition technology to identify the relationship between the phase change influencing factors. For example, it is found that there is an obvious negative correlation between the latent heat release rate and the latent heat absorption rate. Based on the key factors and the relationship between the phase change influencing factors, the pattern recognition results are determined. For example, it is confirmed that under high temperature operating conditions, the increase in latent heat release rate will lead to a significant improvement in the overall heat exchange efficiency, while under low temperature startup conditions, the optimization of the latent heat absorption rate can effectively improve the system startup speed.

[0126] Step 314: Based on the pattern recognition result, the Bayesian optimization algorithm is used to adjust the operating parameter configuration of the molten salt system to generate a preliminary optimized operating parameter configuration, a genetic algorithm is introduced to optimize the preliminary optimized operating parameter configuration to generate a final optimized operating parameter configuration, and based on the final optimized operating parameter configuration, an optimal heat exchange performance configuration is determined;

[0127] This step uses the Bayesian optimization algorithm to adjust the operating parameter configuration of the molten salt system based on the pattern recognition results to generate a preliminary optimized operating parameter configuration. For example, adjust parameters such as heating power and cooling rate to maximize heat exchange efficiency. A genetic algorithm is introduced to optimize the preliminary optimized operating parameter configuration to generate a final optimized operating parameter configuration. For example, the specific values ​​of heating power and cooling rate are further optimized by a genetic algorithm to ensure that the system maintains optimal performance under various operating conditions. Based on the final optimized operating parameter configuration, the optimal heat exchange performance configuration is determined. For example, it is determined that under high temperature operating conditions, the heating power is XkW and the cooling rate is Y℃ / min, which can achieve the highest heat exchange efficiency.

[0128] The embodiments of the present invention can achieve the following beneficial effects through the above steps:

[0129] The target proxy model provides detailed phase change impact assessment results, providing a scientific basis for subsequent optimization and ensuring the rationality of operating parameter configuration;

[0130] Advanced pattern recognition technology provides comprehensive pattern recognition results, helps understand the complex relationship between factors affecting phase change, and provides a scientific basis for optimizing operating parameter configuration

[0131] The precise optimization of operating parameter configuration is achieved to ensure that the molten salt system can achieve optimal heat exchange performance under different operating conditions, thereby improving the overall efficiency and stability of the system.

[0132] After completing the evaluation of the phase change influencing factors and the identification of the key factors, in order to further improve the scientificity and accuracy of the molten salt system operating parameter configuration, based on this, the present invention provides a specific embodiment, step 312, based on physical laws and deep neural networks to build a hybrid physical data driven model, based on the hybrid physical data driven model to analyze the historical operation data, to obtain the historical data analysis results, specifically including the following steps:

[0133] Step 321: Select a neural network structure, and build a hybrid physical data-driven model that combines prior physical knowledge based on the neural network structure, physical laws, and deep learning technology, and analyze historical operation data based on the hybrid physical data-driven model to obtain preliminary historical data analysis results;

[0134] In this step, a suitable neural network structure is selected according to the characteristics of the molten salt system. For example, a convolutional neural network (CNN) or a recurrent neural network (RNN) containing multiple hidden layers is selected to process complex spatiotemporal data. Based on the neural network structure, physical laws and deep learning technology, a hybrid physical data-driven model combining prior physical knowledge is constructed. For example, the heat conduction equation is embedded in the neural network so that the model can better understand the physical process. The historical operating data of the molten salt system is analyzed using the hybrid physical data-driven model to obtain preliminary historical data analysis results. For example, the temperature changes, latent heat release and other parameters of the molten salt system under different operating conditions in the past year are analyzed to preliminarily identify the key factors affecting the heat exchange efficiency.

[0135] Step 322: Based on the preliminary historical data analysis results, use a long short-term memory network to perform time series prediction on the historical operation data to obtain a time series prediction result;

[0136] Based on the preliminary historical data analysis results, this step uses a long short-term memory network (LSTM) to perform time series prediction on the historical operation data to obtain time series prediction results. For example, the temperature change trend of the molten salt system in the next week is predicted to help adjust the operating parameters in advance to optimize performance.

[0137] Step 323: Use the isolation forest algorithm to perform anomaly detection on the preliminary historical data analysis result to generate an anomaly detection result, and optimize the preliminary historical data analysis result based on the time series prediction result and the anomaly detection result to obtain a historical data analysis result;

[0138] This step uses the isolation forest algorithm to perform anomaly detection on the preliminary historical data analysis results to generate anomaly detection results. For example, some abnormal fluctuations in the historical operation data are identified, such as sudden temperature rises or drops, which may be signs of equipment failure or operational errors. Based on the time series prediction results and the anomaly detection results, the preliminary historical data analysis results are optimized to obtain the final historical data analysis results. For example, after removing abnormal data points, the historical operation data is re-evaluated to ensure the accuracy and reliability of the analysis results.

[0139] The embodiments of the present invention can achieve the following beneficial effects through the above steps:

[0140] It provides an analytical framework that combines physical knowledge and data-driven, ensuring that the model can not only capture complex data patterns but also maintain the accuracy of physical meaning;

[0141] The use of long short-term memory networks improves the ability to predict future operating states, enhances the system's predictability and response speed, and helps prevent potential problems;

[0142] The historical data analysis results improve the accuracy of historical data analysis, remove the impact of abnormal data, and make subsequent optimization and decision-making more reliable.

[0143] After clarifying the spatial distribution characteristics and heat transfer paths of different scales in the molten salt system, in order to further accurately describe the thermodynamic behavior of the system, based on this, the present invention provides a specific embodiment, in which step 101 uses a Bayesian optimization algorithm to select a target state equation, and constructs an adaptive thermodynamic model based on the target state equation, molten salt composition and temperature conditions. Based on the adaptive thermodynamic model, the physical property parameter set, latent heat change and thermodynamic behavior of the molten salt at different temperatures are determined to obtain a thermodynamic characteristic description of the molten salt system, which specifically includes the following steps:

[0144] Step 111: Based on the Bayesian optimization algorithm and Gaussian process regression, the performance of different state equations is evaluated to obtain a performance evaluation result, a target state equation is selected based on the performance evaluation result, and a physical information neural network is applied to construct an adaptive thermodynamic model based on the target state equation, molten salt composition and temperature conditions;

[0145] In this step, the state equation refers to the mathematical expression that describes the physical properties of a substance under different temperature and pressure conditions. The performance evaluation results refer to the results obtained after evaluating different state equations, including their prediction accuracy, calculation efficiency and other indicators.

[0146] This step uses the Bayesian optimization algorithm and Gaussian process regression to evaluate the performance of multiple candidate state equations. For example, for a mixed molten salt system containing sodium nitrate and potassium nitrate, the performance of different state equations in simulating the physical properties of the molten salt is evaluated. The most suitable target state equation is selected based on the performance evaluation results. For example, after evaluation, it was found that the physical properties of the molten salt system can be most accurately described using the improved van der Waals equation, so this equation is selected as the target state equation. Based on the target state equation, molten salt composition and temperature conditions, a physical information neural network is applied to construct an adaptive thermodynamic model. For example, the heat conduction equation is embedded in the neural network so that the adaptive thermodynamic model can better understand the physical process and dynamically adjust the internal parameters according to the actual operating conditions.

[0147] Step 112: Based on the adaptive thermodynamic model, the physical properties of the molten salt at different temperatures are calculated to obtain a set of physical property parameters, where the physical properties include density, specific heat capacity, and thermal conductivity;

[0148] This step calculates the physical properties of the molten salt at different temperatures based on the adaptive thermodynamic model to obtain a set of physical property parameters. For example, the density of the molten salt at 400°C is calculated to be 1.7 g / cm 3 , specific heat capacity is 1.2J / g·K, and thermal conductivity is 0.5W / m·K.

[0149] Step 113: Based on the physical property parameter set, simulating the latent heat release and latent heat absorption of the molten salt during the phase change process to calculate the latent heat change;

[0150] This step simulates the latent heat release and latent heat absorption of the molten salt during the phase change process based on the physical property parameter set to calculate the latent heat change. For example, in the simulation of the molten salt changing from solid to liquid, the latent heat absorbed by each gram of molten salt is calculated to be 200 J / g.

[0151] Step 114: Analyze the thermodynamic behavior of the molten salt system in combination with the physical property parameter set and the latent heat variation, and generate a thermodynamic behavior description of the molten salt system, wherein the thermodynamic behavior includes a phase change point, a phase change rate, and an energy conversion efficiency;

[0152] This step combines the physical property parameter set and the latent heat change to analyze the thermodynamic behavior of the molten salt system and generate a thermodynamic behavior description of the molten salt system. For example, the phase change point of the molten salt is determined to be 450°C, the phase change rate is 0.5°C / min, and the energy conversion efficiency during the phase change process is 90%, thereby optimizing the operating parameter configuration of the system to ensure efficient operation.

[0153] The embodiments of the present invention can achieve the following beneficial effects through the above steps:

[0154] Ensure that the state equation that best describes the molten salt system is selected, which improves the accuracy of subsequent modeling and calculations;

[0155] Provides a high-precision adaptive thermodynamic model to ensure that the model can accurately reflect the behavior of the molten salt system under different conditions;

[0156] Provides a detailed set of physical property parameters to ensure that the basic data for subsequent analysis is accurate and reliable;

[0157] It provides accurate latent heat change, which helps to understand the energy conversion mechanism during phase change;

[0158] It provides a comprehensive description of thermodynamic behavior, helps optimize the operating parameter configuration of the molten salt system, and improves the overall performance and stability of the system.

[0159] After calculating the physical property parameter set of the molten salt at different temperatures and simulating the latent heat change during the phase change process, in order to further understand the overall thermodynamic behavior of the molten salt system, based on this, the present invention provides a specific embodiment, step 114, combining the physical property parameter set and the latent heat change, analyzing the thermodynamic behavior of the molten salt system, and generating a thermodynamic behavior description of the molten salt system, the thermodynamic behavior includes the phase change point, the phase change rate and the energy conversion efficiency, specifically comprising the following steps:

[0160] Step 121: Based on the physical property parameter set and the latent heat change, identifying the phase transition temperature point of the molten salt from solid to liquid or from liquid to solid;

[0161] In this step, the latent heat change refers to the total amount of heat absorbed or released during the phase change process. The phase change temperature point refers to the temperature point at which a substance changes from solid to liquid or from liquid to solid.

[0162] This step is based on the physical property parameter set and the latent heat change, and identifies the phase transition temperature point of the molten salt from solid to liquid or from liquid to solid by analyzing the thermodynamic behavior of the molten salt at different temperatures. For example, for a mixed molten salt system containing sodium nitrate and potassium nitrate, a significant latent heat absorption phenomenon is observed at 450°C, and this temperature is determined to be the phase transition temperature point of the system.

[0163] Step 122: Based on the phase change temperature point, the temperature change rate and the latent heat change amount, the speed of state conversion of the molten salt during the phase change process is calculated to obtain the phase change rate;

[0164] In this step, the temperature change rate refers to the rate of change of temperature over time.

[0165] This step calculates the speed of state conversion of the molten salt during the phase change process based on the phase change temperature point, the temperature change rate and the latent heat change, and obtains the phase change rate, the temperature change rate such as the heating or cooling rate. For example, assuming that the molten salt system heats up at a rate of 0.5°C / min near the phase change temperature point, and each gram of molten salt absorbs 200J / g of latent heat, the phase change rate calculated by the thermodynamic formula is 0.1g / s.

[0166] Step 122: Based on the phase change rate, using coupled fluid dynamics simulation and finite element analysis method, evaluate the energy transfer path and loss mechanism of the molten salt system to obtain an energy transfer evaluation result;

[0167] This step is based on the phase change rate, using coupled fluid dynamics simulation and finite element analysis methods to evaluate the energy transfer path and loss mechanism of the molten salt system and obtain the energy transfer evaluation results. For example, through simulation, it is found that the molten salt transfers heat to the container wall mainly through heat conduction and convection during the phase change process, and there is a certain amount of heat loss at the container wall. Further analysis shows that the material selection and thickness of the container wall have a significant impact on heat loss.

[0168] Step 122: Based on the energy transfer evaluation result and the phase change rate, the energy conversion efficiency of the molten salt system in the heat exchange process is evaluated, and the thermodynamic behavior description of the molten salt system is generated by combining the phase change rate, the energy transfer evaluation result and the energy conversion efficiency;

[0169] In this step, energy conversion efficiency refers to the effective utilization of energy during the phase change process.

[0170] This step evaluates the energy conversion efficiency of the molten salt system during the heat exchange process based on the energy transfer evaluation results and the phase change rate. For example, by analyzing the simulation results, it is found that the energy conversion efficiency of the molten salt during the phase change process is 90%, that is, 90% of the input energy is effectively used for phase change, while the remaining 10% is dissipated due to heat loss and other factors. Combining the phase change rate, energy transfer evaluation results and energy conversion efficiency, a thermodynamic behavior description of the molten salt system is generated. For example, the final generated thermodynamic behavior description includes a phase change point of 450°C, a phase change rate of 0.1g / s, and an energy conversion efficiency of 90%.

[0171] The embodiments of the present invention can achieve the following beneficial effects through the above steps:

[0172] The phase transition temperature point of the molten salt system was accurately identified, providing key data support for subsequent calculations;

[0173] It provides accurate phase change rate, which helps to understand the kinetic characteristics of the phase change process and optimize the operating conditions;

[0174] Provides detailed analysis of energy transfer paths and loss mechanisms to help optimize system design and reduce unnecessary energy losses;

[0175] It provides a comprehensive description of thermodynamic behavior, helps optimize the operating parameter configuration of the molten salt system, and improves the overall performance and stability of the system.

[0176] Figure 2 A schematic diagram of a calculation system for non-fluid phase-change molten salt heat exchange is provided for an embodiment of the present invention. Figure 2 As shown, the system includes:

[0177] A construction module 21 is used to select a target state equation using a Bayesian optimization algorithm, and to construct an adaptive thermodynamic model based on the target state equation, molten salt composition, and temperature conditions. Based on the adaptive thermodynamic model, a set of physical property parameters, a latent heat change, and thermodynamic behavior of the molten salt at different temperatures are determined to obtain a thermodynamic characteristic description of the molten salt system.

[0178] A determination module 22 is used to extract spatial distribution characteristics based on the thermodynamic characteristic description and apply a multi-scale modeling method. Based on the spatial distribution characteristics, a discrete element method is used to generate a description of the kinetic behavior of the molten salt particle level. At the same time, a finite volume method is used to process the heat transfer problem to determine the temperature field distribution. In combination with the phase field theory and the thermodynamic characteristic description, the temperature field evolution inside the molten salt and at the heat exchange interface between the molten salt and the outside is processed through an implicit time integration scheme to obtain a temperature field evolution result. Based on the spatial distribution characteristics, the kinetic behavior description, the temperature field distribution and the temperature field evolution result, a target description of the heat transfer path under a non-fluid state is generated;

[0179] The optimization module 23 is used to evaluate the influence of each phase change influencing factor based on the target description by using the agent model technology to obtain the phase change influence evaluation result, analyze the historical operation data by using the deep neural network to obtain the historical data analysis result, combine the historical data analysis result and the phase change influence evaluation result to determine the pattern recognition result, and optimize the operating parameter configuration of the molten salt system based on the pattern recognition result to determine the optimal heat exchange performance configuration;

[0180] A generation module 24 is used to determine a target operating parameter set based on the real-time molten salt temperature response data and the optimal heat exchange performance configuration by applying an evolutionary algorithm, construct a topological structure based on the target operating parameter set, and generate a target operating strategy based on the topological structure and the target operating parameter set by adopting a cooperative competition mechanism under the framework of game theory. Based on the target operating strategy, the calculation of non-fluid phase change molten salt heat exchange is realized.

[0181] Figure 2 The computing system for non-fluid phase change molten salt heat exchange can execute Figure 1 The implementation principle and technical effect of the calculation method of non-fluid phase-change molten salt heat exchange described in the embodiment are not repeated here. The specific way in which each module and unit performs operations in the calculation system of non-fluid phase-change molten salt heat exchange in the above embodiment has been described in detail in the embodiment of the method, and will not be elaborated here.

[0182] In one possible design, Figure 2 A computing system for non-fluid phase-change molten salt heat exchange in the illustrated embodiment can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0183] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0184] The processing component 32 is used to: select the target state equation using a Bayesian optimization algorithm, construct an adaptive thermodynamic model based on the target state equation, molten salt composition and temperature conditions, determine the physical property parameter set, latent heat change and thermodynamic behavior of the molten salt at different temperatures based on the adaptive thermodynamic model, so as to obtain a thermodynamic characteristic description of the molten salt system; based on the thermodynamic characteristic description, apply a multi-scale modeling method to extract spatial distribution characteristics, and based on the spatial distribution characteristics, use a discrete element method to generate a kinetic behavior description of the molten salt particle level, and use a finite volume method to process heat transfer problems to determine the temperature field distribution, combine phase field theory and the thermodynamic characteristic description, and process the temperature field evolution inside the molten salt and at the heat exchange interface between the molten salt and the outside world through an implicit time integration scheme to obtain a temperature field evolution result, based on the spatial distribution characteristics, the kinetic behavior description, and the temperature field distribution. and the temperature field evolution results, generate a target description of the heat transfer path under the non-fluid state; based on the target description, use the proxy model technology to evaluate the influence of each phase change influencing factor to obtain the phase change impact evaluation result, use the deep neural network to analyze the historical operation data to obtain the historical data analysis result, combine the historical data analysis result and the phase change impact evaluation result to determine the pattern recognition result, and optimize the operating parameter configuration of the molten salt system based on the pattern recognition result to determine the optimal heat exchange performance configuration; based on the real-time molten salt temperature response data and the optimal heat exchange performance configuration, use the evolutionary algorithm to determine the target operating parameter set, build a topological structure based on the target operating parameter set, and based on the topological structure and the target operating parameter set, use the cooperative competition mechanism in the game theory framework to generate a target operating strategy, which is used to guide the operation of the molten salt system to perform non-fluid phase change molten salt heat exchange calculations.

[0185] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0186] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0187] Of course, the computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0188] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0189] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0190] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0191] The embodiment of the present invention further provides a computer storage medium storing a computer program, which can achieve the above-mentioned Figure 1 A calculation method for non-fluid phase-change molten salt heat exchange in the illustrated embodiment.

[0192] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0193] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0194] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0195] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention 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 of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A calculation method for non-fluid phase change molten salt heat transfer, characterized in that: include: A target state equation is selected by using a Bayesian optimization algorithm, and an adaptive thermodynamic model is constructed based on the target state equation, molten salt composition, and temperature conditions. Based on the adaptive thermodynamic model, a set of physical property parameters, latent heat changes, and thermodynamic behaviors of the molten salt at different temperatures are determined to obtain a thermodynamic characteristic description of the molten salt system; Based on the thermodynamic property description, a multi-scale modeling method is applied to extract spatial distribution characteristics. Based on the spatial distribution characteristics, a discrete element method is used to generate a description of the kinetic behavior of the molten salt at the particle level. At the same time, a finite volume method is used to deal with the heat transfer problem to determine the temperature field distribution. In combination with the phase field theory and the thermodynamic property description, the evolution of the temperature field inside the molten salt and at the heat exchange interface between the molten salt and the outside world is processed through an implicit time integration scheme to obtain the temperature field evolution results. Based on the spatial distribution characteristics, the kinetic behavior description, the temperature field distribution and the temperature field evolution results, a target description of the heat transfer path under a non-fluid state is generated; Based on the target description, the agent model technology is used to evaluate the influence of each phase change influencing factor to obtain the phase change influence evaluation result, the historical operation data is analyzed by deep neural network to obtain the historical data analysis result, and the pattern recognition result is determined by combining the historical data analysis result and the phase change influence evaluation result. Based on the pattern recognition result, the operating parameter configuration of the molten salt system is optimized to determine the optimal heat exchange performance configuration; Based on the real-time molten salt temperature response data and the optimal heat exchange performance configuration, an evolutionary algorithm is applied to determine the target operating parameter set, and a topological structure is constructed based on the target operating parameter set. Based on the topological structure and the target operating parameter set, a cooperative competition mechanism in a game theory framework is adopted to generate a target operating strategy, which is used to guide the operation of the molten salt system to perform calculations on non-fluid phase change molten salt heat exchange.

2. The method according to claim 1, characterized in that Based on the thermodynamic characteristics description, a multi-scale modeling method is applied to extract the spatial distribution characteristics. Based on the spatial distribution characteristics, a discrete element method is used to generate a description of the kinetic behavior of the molten salt particle level. At the same time, a finite volume method is used to deal with the heat transfer problem to determine the temperature field distribution. Combining phase field theory and the thermodynamic characteristics description, the temperature field evolution inside the molten salt and at the heat exchange interface between the molten salt and the outside world is processed through an implicit time integration scheme to obtain the temperature field evolution results, including: Based on the thermodynamic characteristics description, microscopic molecular dynamics and macroscopic continuum mechanics, a multi-scale modeling method is applied to simulate and extract the spatial distribution characteristics of different scales in the molten salt system; Setting material properties and contact mechanical behaviors of molten salt particles, applying discrete element method to simulate the interaction between molten salt particles and the influence of the interaction between molten salt particles on the molten salt system based on the material properties, contact mechanical behaviors and the thermodynamic characteristics description, and generating a description of the dynamic behavior of molten salt particles based on the interaction between molten salt particles and the influence of the interaction between molten salt particles on the molten salt system; Based on the spatial distribution characteristics, an optimized grid is generated using an adaptive grid refinement technology and a topology optimization algorithm, and boundary conditions are set. Based on the optimized grid and the boundary conditions, a finite volume method is used to process the heat transfer problem to obtain simulation results. Based on the simulation results, the temperature field distribution in the molten salt system is finally determined. In the process of processing the heat transfer problem, a graph neural network is introduced to optimize the simulation results. The boundary conditions include grid boundary conditions and physical boundary conditions. Combining the phase field theory and the description of the thermodynamic characteristics, the evolution of the temperature field inside the molten salt and at the heat exchange interface between the molten salt and the outside world is processed by an implicit time integration scheme to obtain the temperature field evolution results. In the process of the implicit time integration scheme, a nonlinear dynamic equation solver and an adaptive step control mechanism are introduced to obtain the change pattern of the temperature field over time.

3. The method according to claim 2, characterized in that Based on the spatial distribution characteristics, an optimized grid is generated using adaptive grid refinement technology and a topology optimization algorithm, and boundary conditions are set. Based on the optimized grid and the boundary conditions, a finite volume method is used to process the heat transfer problem to obtain simulation results. Based on the simulation results, the temperature field distribution in the molten salt system is finally determined. In the process of processing the heat transfer problem, a graph neural network is introduced to optimize the simulation results, including: Based on the spatial distribution characteristics, the geometric structure and physical characteristics of the molten salt system are dynamically adjusted by using an adaptive mesh refinement technology to generate an initial mesh, a topology optimization algorithm is introduced to optimize the shape and distribution of the initial mesh, an optimized mesh is generated, and boundary conditions are set based on the optimized mesh; Based on the optimized grid and the boundary conditions, a finite volume method is used to process the heat transfer problem to obtain a simulation result, wherein the simulation result includes a temperature field distribution to be optimized; Based on the optimized grid, a graph neural network is constructed; based on the graph neural network, a connection pattern between nodes in the optimized grid is analyzed to identify heat transfer paths in a molten salt system and interactions between the heat transfer paths; based on the heat transfer paths and interactions between the heat transfer paths, a graph neural network model is generated; The heat transfer path and the heat transfer efficiency of the heat transfer path are predicted based on the graph neural network model, and the heat transfer path and the heat transfer efficiency of the heat transfer path are optimized to obtain the heat transfer result, and the temperature field distribution to be optimized in the simulation result is optimized based on the heat transfer result, and finally the temperature field distribution in the molten salt system is determined.

4. The method according to claim 1, characterized in that: Based on the target description, the influence of each phase change influencing factor is evaluated by using the agent model technology to obtain the phase change impact evaluation result, the historical operation data is analyzed by using the deep neural network to obtain the historical data analysis result, and the pattern recognition result is determined by combining the historical data analysis result and the phase change impact evaluation result. Based on the pattern recognition result, the operating parameter configuration of the molten salt system is optimized to determine the optimal heat exchange performance configuration, including: A target proxy model is selected by using a proxy model technology, and a global sensitivity analysis method in the target proxy model is used to evaluate the influence of each phase change influencing factor on the overall heat exchange efficiency to obtain a phase change factor influence result. Based on the phase change factor influence result, a local sensitivity analysis method in the target proxy model is used to identify key factors that affect the overall heat exchange efficiency under specific operating conditions, and a phase change influence evaluation result is obtained based on the phase change factor influence result and the key factors. The phase change influencing factors include temperature change, latent heat release, and latent heat absorption. The specific operating conditions include high-temperature operating conditions and low-temperature starting conditions. The key factors include latent heat release rate and latent heat absorption rate. Building a hybrid physical data-driven model based on physical laws and deep neural networks, analyzing historical operating data based on the hybrid physical data-driven model to obtain historical data analysis results; Based on the historical data analysis results and the phase change impact assessment results, an advanced pattern recognition technology is used to identify the mutual relationship between the phase change influencing factors, and based on the key factors and the mutual relationship between the phase change influencing factors, a pattern recognition result is determined; Based on the pattern recognition results, the Bayesian optimization algorithm is used to adjust the operating parameter configuration of the molten salt system to generate a preliminary optimized operating parameter configuration. A genetic algorithm is introduced to optimize the preliminary optimized operating parameter configuration to generate a final optimized operating parameter configuration. Based on the final optimized operating parameter configuration, the optimal heat exchange performance configuration is determined.

5. The method according to claim 4, characterized in that A hybrid physical data-driven model is constructed based on physical laws and deep neural networks, and historical operation data is analyzed based on the hybrid physical data-driven model to obtain historical data analysis results, including: Selecting a neural network structure, building a hybrid physical data-driven model combining prior physical knowledge based on the neural network structure, physical laws, and deep learning technology, analyzing historical operation data based on the hybrid physical data-driven model, and obtaining preliminary historical data analysis results; Based on the preliminary historical data analysis results, using a long short-term memory network, time series prediction is performed on the historical operation data to obtain a time series prediction result; Anomaly detection is performed on the preliminary historical data analysis results using an isolation forest algorithm to generate anomaly detection results, and based on the time series prediction results and the anomaly detection results, the preliminary historical data analysis results are optimized to obtain historical data analysis results.

6. The method according to claim 1, characterized in that The target state equation is selected by using the Bayesian optimization algorithm. Based on the target state equation, the molten salt composition and the temperature condition, an adaptive thermodynamic model is constructed. Based on the adaptive thermodynamic model, the physical property parameter set, latent heat change and thermodynamic behavior of the molten salt at different temperatures are determined to obtain a thermodynamic characteristic description of the molten salt system, including: Based on the Bayesian optimization algorithm and Gaussian process regression, the performance of different state equations is evaluated to obtain performance evaluation results, a target state equation is selected based on the performance evaluation results, and a physical information neural network is applied to construct an adaptive thermodynamic model based on the target state equation, molten salt composition and temperature conditions; Based on the adaptive thermodynamic model, the physical properties of the molten salt at different temperatures are calculated to obtain a set of physical property parameters, wherein the physical properties include density, specific heat capacity and thermal conductivity; Based on the physical property parameter set, simulating the latent heat release and latent heat absorption of the molten salt during the phase change process to calculate the latent heat change; The thermodynamic behavior of the molten salt system is analyzed in combination with the physical property parameter set and the latent heat change, and a thermodynamic behavior description of the molten salt system is generated, wherein the thermodynamic behavior includes a phase change point, a phase change rate, and an energy conversion efficiency.

7. The method according to claim 6, characterized in that In combination with the physical property parameter set and the latent heat change, the thermodynamic behavior of the molten salt system is analyzed to generate a thermodynamic behavior description of the molten salt system, wherein the thermodynamic behavior includes a phase change point, a phase change rate, and an energy conversion efficiency, including: Based on the physical property parameter set and the latent heat change, identifying a phase transition temperature point of the molten salt from solid to liquid or from liquid to solid; Based on the phase change temperature point, the temperature change rate and the latent heat change amount, the speed of state conversion of the molten salt during the phase change process is calculated to obtain the phase change rate; Based on the phase change rate, the energy transfer path and loss mechanism of the molten salt system are evaluated by using coupled fluid dynamics simulation and finite element analysis method to obtain energy transfer evaluation results; Based on the energy transfer evaluation result and the phase change rate, the energy conversion efficiency of the molten salt system during the heat exchange process is evaluated, and a description of the thermodynamic behavior of the molten salt system is generated by combining the phase change rate, the energy transfer evaluation result and the energy conversion efficiency.

8. A calculation system for non-fluid phase change molten salt heat exchange, characterized in that: include: A construction module is used to select a target state equation using a Bayesian optimization algorithm, and to construct an adaptive thermodynamic model based on the target state equation, molten salt composition, and temperature conditions. Based on the adaptive thermodynamic model, a set of physical property parameters, a latent heat change, and a thermodynamic behavior of the molten salt at different temperatures are determined to obtain a thermodynamic characteristic description of the molten salt system; A determination module is used to extract spatial distribution characteristics based on the thermodynamic characteristic description and apply a multi-scale modeling method. Based on the spatial distribution characteristics, a discrete element method is used to generate a description of the kinetic behavior of the molten salt particle level. At the same time, a finite volume method is used to process the heat transfer problem to determine the temperature field distribution. In combination with the phase field theory and the thermodynamic characteristic description, the temperature field evolution inside the molten salt and at the heat exchange interface between the molten salt and the outside world is processed through an implicit time integration scheme to obtain the temperature field evolution result. Based on the spatial distribution characteristics, the kinetic behavior description, the temperature field distribution and the temperature field evolution result, a target description of the heat transfer path under a non-fluid state is generated; an optimization module, for evaluating the influence of each phase change influencing factor based on the target description by using a proxy model technology to obtain a phase change influence evaluation result, analyzing historical operation data by using a deep neural network to obtain a historical data analysis result, combining the historical data analysis result with the phase change influence evaluation result to determine a pattern recognition result, and optimizing the operating parameter configuration of the molten salt system based on the pattern recognition result to determine an optimal heat exchange performance configuration; A generation module is used to determine a target operating parameter set based on real-time molten salt temperature response data and the optimal heat exchange performance configuration by applying an evolutionary algorithm, construct a topological structure based on the target operating parameter set, generate a target operating strategy based on the topological structure and the target operating parameter set by adopting a cooperative competition mechanism under a game theory framework, and realize the calculation of non-fluid phase change molten salt heat exchange based on the target operating strategy.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a calculation method for non-fluid phase-change molten salt heat exchange as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, a calculation method for non-fluid phase-change molten salt heat exchange as described in any one of claims 1 to 7 is implemented.

Images