A concentrated photovoltaic thermal component heat exchange structure optimization design method and related equipment

Abstract

The present application discloses a method for optimizing the design of the heat exchange structure of a concentrated photovoltaic and thermal assembly and related equipment, relating to the field of new energy technology. The method comprises: inputting a discrete model of the heat exchange area to be designed into a neural network operator proxy model to obtain corresponding evaluation indicators under different design parameters, wherein the network operator proxy model is obtained by performing a preset number of training on a multi-physical field coupled finite element simulation result set, and the design parameters include the shape characteristics of the flow area in the heat exchange area to be designed; performing multi-objective optimization and Pareto front solution generation operations on the discrete model of the heat exchange area to be designed and the corresponding evaluation indicators under all the corresponding design parameters to obtain a Pareto front solution set; obtaining comprehensive weight information of the evaluation indicators; and performing a comprehensive evaluation operation by combining TOPSIS and GRA based on the comprehensive weight information and the Pareto front solution set to determine the target flow area.

Description

Technical Field

[0001] This specification relates to the field of new energy technology, and more specifically, to a method for optimizing the design of heat exchange structures of concentrated photovoltaic and thermal components and related equipment. Background Art

[0002] Solar photovoltaic thermal technology, a derivative of photovoltaic technology, manages the thermal performance of solar cells, recovering waste heat while simultaneously reducing their temperature, effectively improving solar energy utilization efficiency. Among solar photovoltaic thermal technologies, concentrated photovoltaic thermal technology offers greater economic benefits, improving conversion efficiency and energy quality while reducing overall costs. However, the use of solar photovoltaic thermal technology under concentrated conditions faces certain risks of overheating and uneven temperature distribution, which are limited by the concentrator structure and installation level, compromising photovoltaic efficiency and lifespan. Optimizing the heat exchange structure used for thermal management can effectively improve component performance.

[0003] Current solar photovoltaic (PV) thermal heat exchanger design relies heavily on the designer's experience, requiring specialized technical expertise and subjectivity. Furthermore, the large search area for optimal design often makes the entire optimization process time-consuming and labor-intensive. Summary of the Invention

[0004] The Summary of the Invention introduces a series of simplified concepts that will be further described in the Detailed Description of the Invention. The Summary of the Invention of this application is not intended to limit the key features and essential technical features of the claimed technical solution, nor is it intended to determine the scope of protection of the claimed technical solution.

[0005] In the first aspect, this application proposes a method for optimizing the heat exchange structure of a concentrated photovoltaic and thermal assembly, comprising:

[0006] Inputting the discrete model of the heat exchange zone to be designed into a neural network operator proxy model to obtain corresponding evaluation indicators under different design parameters, wherein the network operator proxy model is obtained by performing a preset number of training on a multi-physics field coupled finite element simulation result set, and the design parameters include the shape characteristics of the flow area in the heat exchange zone to be designed;

[0007] The discrete model of the heat exchange zone to be designed and the corresponding evaluation indicators under all the corresponding design parameters are subjected to multi-objective optimization and Pareto front solution set generation operations to obtain the Pareto front solution set;

[0008] Obtain comprehensive weight information of the above evaluation indicators;

[0009] Based on the above comprehensive weight information and the above Pareto front solution set, a comprehensive evaluation operation is performed by combining TOPSIS and GRA to determine the target flow area.

[0010] In a feasible implementation, the specific steps of obtaining the above-mentioned neural network operator proxy model include:

[0011] Discretize the model of the heat exchange area to be designed to obtain a discrete model of the heat exchange area to be designed;

[0012] According to the discrete model of the heat exchange area to be designed, the corresponding geometric model is automatically generated according to the discretized design parameters;

[0013] Perform multi-physics field coupling finite element simulation based on the above geometric model to obtain simulation result data;

[0014] Performing data extraction and formatting output operations on the above simulation result data to obtain formatted data;

[0015] Performing feature extraction on the formatted data to obtain feature-extracted data;

[0016] The neural network operator proxy model is trained according to the heat exchange area model to be designed and the feature extraction data to achieve a preset number of training times or a preset accuracy to obtain the neural network operator proxy model.

[0017] In a feasible implementation, the multi-physics field coupled finite element simulation includes heat conduction simulation, fluid flow simulation, and current conduction simulation.

[0018] In a feasible implementation, the above-mentioned discrete model of the heat exchange zone to be designed and the corresponding evaluation indicators under all corresponding design parameters are subjected to multi-objective optimization and Pareto front solution set generation operations to obtain the Pareto front solution set, including:

[0019] Generate an initial population based on the discrete model of the heat exchange zone to be designed and the corresponding evaluation indicators under all the corresponding design parameters;

[0020] Perform fitness evaluation on the above initial population to obtain the fitness of the individuals in the population;

[0021] According to the individual fitness of the above population, the population evolution is guided by genetic operation and based on non-dominated sorting to obtain a new generation of population and its Pareto frontier solution set.

[0022] In a feasible implementation, the initial population is generated based on a Latin hypercube sampling method.

[0023] In a feasible embodiment, the objective function corresponding to the above fitness evaluation operation includes maximizing thermal power output, minimizing temperature gradient, and maximizing electrical power output.

[0024] In a feasible implementation manner, the above-mentioned obtaining of the comprehensive weight information of the above-mentioned evaluation indicators includes:

[0025] Determine the subjective weight information corresponding to different evaluation indicators based on the best-worst method;

[0026] Calculate objective weight information based on the CRITIC method;

[0027] The weight values ​​of the subjective weight information and the objective weight information are solved according to the Nash equilibrium solution method of game theory to obtain comprehensive weight information.

[0028] In a second aspect, the present application proposes a device for optimizing the heat exchange structure of a concentrated photovoltaic and thermal assembly, comprising:

[0029] a first acquisition unit, configured to input a discrete model of the heat exchange zone to be designed into a neural network operator proxy model to obtain corresponding evaluation indicators under different design parameters, wherein the network operator proxy model is obtained by performing a preset number of training on a multi-physics field coupled finite element simulation result set, and the design parameters include shape characteristics of the flow region in the heat exchange zone to be designed;

[0030] The second acquisition unit is used to perform multi-objective optimization and Pareto front solution set generation operations on the discrete model of the heat exchange zone to be designed and the corresponding evaluation indicators under all the corresponding design parameters, so as to obtain the Pareto front solution set;

[0031] A third obtaining unit is used to obtain comprehensive weight information of the above evaluation indicators;

[0032] The determination unit is used to perform a comprehensive evaluation operation using a combination of TOPSIS and GRA based on the comprehensive weight information and the Pareto front solution set to determine the target flow area.

[0033] In a third aspect, an electronic device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is configured to implement the steps of the method for optimizing the heat exchange structure of a concentrating photovoltaic and thermal component as described in any one of the first aspects above when executing the computer program stored in the memory.

[0034] In a fourth aspect, the present application further proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for optimizing the heat exchange structure of a concentrated photovoltaic and thermal component according to any one of the first aspects.

[0035] In summary, the traditional solar photovoltaic thermal heat exchange structure design relies on the designer's experience, and is designed and optimized through trial and error and manual adjustment. This method is not only time-consuming and labor-intensive, but the optimization results are greatly affected by the designer's personal experience and subjectivity, making it difficult to ensure the global optimality of the design. The present invention introduces a neural network operator proxy model to quickly and accurately predict performance indicators under different design parameters. The neural network operator proxy model is trained based on multi-physics field coupled finite element simulation data, replacing the complex numerical simulation process, greatly shortening the calculation time in the optimization process, reducing dependence on high computing power resources, and making the optimization design more efficient. By achieving rapid performance prediction through the proxy model, a larger design space can be explored in a short time, improving design efficiency and quality, and reducing the designer's subjective dependence. Existing photovoltaic thermal heat exchange structure designs can often only be optimized on a single objective, such as maximizing thermal power or minimizing temperature, but ignores other important performance indicators, resulting in the system design being biased towards a certain performance target and lacking a balance in overall performance. The present invention achieves a balance between multiple objective functions (such as maximizing thermal power output, minimizing temperature gradients, and maximizing electrical power output) through multi-objective optimization and Pareto front solution set generation. A non-dominated sorting genetic algorithm (PB-NSGA-III) based on reference directions is used for multi-objective optimization to generate a Pareto front solution set. The Pareto front solution set contains a series of optimal design solutions that achieve the best trade-offs between different design objectives, providing designers with a variety of design options. Through multi-objective optimization, overall system performance is improved, local optimality issues associated with single-objective optimization are avoided, and the design solutions achieve the best balance between various objectives. In existing design solutions, the selection of the final optimal solution often relies on the designer's judgment or a simple weighted average method. This approach lacks systematicity and objectivity and may not effectively evaluate the performance of each solution across multiple performance indicators. The present invention introduces a comprehensive evaluation method that combines TOPSIS with GRA (grey relational analysis). After generating the Pareto front solution set, the TOPSIS method is used to calculate the distance between each solution and the ideal solution and the negative ideal solution to evaluate the performance of each solution. The GRA method is also combined with the grey correlation method to further evaluate the similarity between the proposed solution and the ideal solution. This combined approach allows for a comprehensive, multi-faceted evaluation of the relative merits of each solution. Combining TOPSIS and GRA allows for a more comprehensive and objective assessment of the pros and cons of design solutions, ensuring that the optimal solution ultimately selected is more scientific and rational. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present description. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0037] Figure 1 A schematic diagram of a process for optimizing the design of heat exchange structures of concentrated photovoltaic and thermal assemblies provided in an embodiment of the present application;

[0038] Figure 2 A schematic diagram of a discrete model of a heat exchange zone to be designed provided in an embodiment of the present application;

[0039] Figure 3 A schematic diagram of a Pareto front solution set obtained for solving provided in an embodiment of the present application;

[0040] Figure 4 A schematic diagram of a device for optimizing the heat exchange structure of a concentrated photovoltaic and thermal assembly provided in an embodiment of the present application;

[0041] Figure 5 A schematic diagram of the electronic device structure for optimizing the heat exchange structure of a concentrated photovoltaic and thermal assembly provided in an embodiment of the present application. DETAILED DESCRIPTION

[0042] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or devices. The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the embodiments described are only part of the embodiments of the present application, not all of the embodiments.

[0043] See also Figure 1 , which is a schematic flow chart of a method for optimizing the heat exchange structure of a concentrated photovoltaic and thermal assembly provided in an embodiment of the present application, which may specifically include:

[0044] S110, inputting the discrete model of the heat exchange zone to be designed into a neural network operator proxy model to obtain corresponding evaluation indicators under different design parameters, wherein the network operator proxy model is obtained by performing a preset number of training on a multi-physics field coupled finite element simulation result set, and the design parameters include shape characteristics of the flow area in the heat exchange zone to be designed;

[0045] For example, Figure 2 Figure 2 shows a schematic diagram of a discrete model of a heat exchange zone to be designed. This discrete model is then input into a neural network operator proxy model for performance evaluation. This neural network operator proxy model is trained a preset number of times on a set of multi-physics coupled finite element simulation results. This proxy model can quickly predict performance indicators under different design parameters. Design parameters include, but are not limited to, the shape characteristics of the flow region within the heat exchange zone to be designed, such as the geometry, size, and curvature of the flow channel.

[0046] S120, performing multi-objective optimization and Pareto front solution set generation operations on the discrete model of the heat exchange zone to be designed and the corresponding evaluation indicators under all its corresponding design parameters to obtain a Pareto front solution set;

[0047] For example, a multi-objective optimization operation is performed on the discrete model of the heat exchange zone to be designed and the evaluation indicators under all its corresponding design parameters to generate a Pareto front solution set. The Pareto front solution set contains multiple optimal solutions that strike a balance between different design objectives. This step can be performed using an optimization algorithm (such as PB-NSGA-III), so that the solution set can effectively approximate the Pareto front in the multi-objective space.

[0048] Multi-objective optimization allows for simultaneous consideration of multiple objective functions, such as maximizing thermal power output, minimizing temperature gradients, and maximizing electrical power output. Using a genetic algorithm for iterative optimization, a series of Pareto front solutions representing multi-objective optimization are gradually obtained. These solutions represent the optimal balance between the various objectives, providing a foundation for subsequent comprehensive evaluation.

[0049] S130, obtaining comprehensive weight information of the above evaluation indicators;

[0050] For example, after generating the Pareto front solution set, the next step is to obtain the comprehensive weight information of each evaluation indicator. This weight information is determined by comprehensively considering the importance of each evaluation indicator to ensure that different indicators are reasonably reflected in the final evaluation.

[0051] S140. Perform a comprehensive evaluation operation using a combination of TOPSIS and GRA based on the comprehensive weight information and the Pareto front solution set to determine the target flow area.

[0052] For example, based on the comprehensive weight information and Pareto front solution set obtained above, a combination of TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution) and GRA (Grey Relational Analysis) is used to comprehensively evaluate the design solutions and ultimately determine the target flow area. This comprehensive evaluation method allows the selection of the design solution that performs best in various performance indicators from the Pareto front solution set.

[0053] The TOPSIS method evaluates the closest design solution to the ideal solution by calculating the distance between each design solution in the solution set and the ideal solution and the negative ideal solution. The TOPSIS method also evaluates the quality of each solution by calculating the distance between each solution and the ideal solution and the negative ideal solution. The ideal solution represents the optimal solution for all objectives, while the negative ideal solution represents the worst solution for all objectives. By calculating the relative distance of each solution to the ideal solution and the negative ideal solution, the overall approximation of each solution can be determined. Solutions with higher approximations—that is, solutions that are closer to the ideal solution—are considered to be more optimal. The GRA method uses grey correlation to evaluate the similarity of each solution to the ideal solution, providing a supplementary assessment of the quality of the solutions. The combined TOPSIS-GRA method leverages the strengths of both methods, evaluating the relative merits of each solution from multiple perspectives to ensure a more scientific and rational selection of the optimal solution.

[0054]

[0055]

[0056] sw ij =s ij wg j (12)

[0057] TG i is the comprehensive relative approximation value of the i-th Pareto solution; C i + and C i - is the comprehensive approximation value, C i + It reflects the closeness between the evaluation object and the ideal solution. The larger the value, the better the Pareto solution is. i - It reflects the closeness between the evaluation object and the ideal solution. The larger the value, the worse the Pareto solution. α and β are the approximation coefficients. D i + 、D i -、R i + and R i - is the grey correlation degree between the Pareto solution and the ideal solution; d i + d i - 、r i + and r i - The grey correlation degree between the Pareto solution and the ideal solution after dimensionless processing; R ij + and R ij - is the grey correlation coefficient of the Pareto solution and the positive and negative ideal solutions with respect to the j-th evaluation index; sw j best is the optimal value of the jth weighted evaluation index, that is, the positive ideal solution; sw j worst is the worst value of the jth weighted evaluation index, i.e., the negative ideal solution; ξ is the resolution coefficient; sw ij is the weighted index value of the jth evaluation index of the i-th Pareto solution; s ij is the index value of the jth evaluation index of the i-th Pareto solution; wg j is the combined weight of the j-th evaluation index.

[0058] In summary, the traditional solar photovoltaic thermal heat exchange structure design relies on the designer's experience, and is designed and optimized through trial and error and manual adjustment. This method is not only time-consuming and labor-intensive, but the optimization results are greatly affected by the designer's personal experience and subjectivity, making it difficult to ensure the global optimality of the design. The present invention introduces a neural network operator proxy model to quickly and accurately predict performance indicators under different design parameters. The neural network operator proxy model is trained based on multi-physics field coupled finite element simulation data, replacing the complex numerical simulation process, greatly shortening the calculation time in the optimization process, reducing dependence on high computing power resources, and making the optimization design more efficient. By achieving rapid performance prediction through the proxy model, a larger design space can be explored in a short time, improving design efficiency and quality, and reducing the designer's subjective dependence. Existing photovoltaic thermal heat exchange structure designs can often only be optimized on a single objective, such as maximizing thermal power or minimizing temperature, but ignores other important performance indicators, resulting in the system design being biased towards a certain performance target and lacking a balance in overall performance. The present invention achieves a balance between multiple objective functions (such as maximizing thermal power output, minimizing temperature gradients, and maximizing electrical power output) through multi-objective optimization and Pareto front solution set generation. A non-dominated sorting genetic algorithm (PB-NSGA-III) based on reference directions is used for multi-objective optimization to generate a Pareto front solution set. The Pareto front solution set contains a series of optimal design solutions that achieve the best trade-offs between different design objectives, providing designers with a variety of design options. Through multi-objective optimization, overall system performance is improved, local optimality issues associated with single-objective optimization are avoided, and the design solutions achieve the best balance between various objectives. In existing design solutions, the selection of the final optimal solution often relies on the designer's judgment or a simple weighted average method. This approach lacks systematicity and objectivity and may not effectively evaluate the performance of each solution across multiple performance indicators. The present invention introduces a comprehensive evaluation method that combines TOPSIS with GRA (grey relational analysis). After generating the Pareto front solution set, the TOPSIS method is used to calculate the distance between each solution and the ideal solution and the negative ideal solution to evaluate the performance of each solution. The GRA method is also combined with the grey correlation method to further evaluate the similarity between the proposed solution and the ideal solution. This combined approach allows for a comprehensive, multi-faceted evaluation of the relative merits of each solution. Combining TOPSIS and GRA allows for a more comprehensive and objective assessment of the pros and cons of design solutions, ensuring that the optimal solution ultimately selected is more scientific and rational.

[0059] In some examples, the specific steps of obtaining the neural network operator proxy model include:

[0060] Discretize the model of the heat exchange area to be designed to obtain a discrete model of the heat exchange area to be designed;

[0061] According to the discrete model of the heat exchange area to be designed, the corresponding geometric model is automatically generated according to the discretized design parameters;

[0062] Perform multi-physics field coupling finite element simulation based on the above geometric model to obtain simulation result data;

[0063] Performing data extraction and formatting output operations on the above simulation result data to obtain formatted data;

[0064] Performing feature extraction on the formatted data to obtain feature-extracted data;

[0065] The neural network operator proxy model is trained according to the heat exchange area model to be designed and the feature extraction data to achieve a preset number of training times or a preset accuracy to obtain the neural network operator proxy model.

[0066] For example, the heat exchange area to be designed is first discretized, with the goal of discretizing the continuous design space into a set of parameterized grids or cells. This process allows the complex design area to be described as discrete points, providing a foundation for subsequent geometric modeling and numerical simulation. The heat exchange area is divided into several small cells, each of which can have its own defined boundary conditions, material properties, and other characteristics, allowing for efficient numerical processing using mathematical modeling methods.

[0067] Based on the discretized design parameters, corresponding geometric models are automatically generated. This generation of geometric models is typically accomplished with computer-aided design (CAD) software, utilizing parametric design tools or scripting to rapidly and massively generate geometric models that conform to the discretized parameters. Geometric models may include heat exchange structures with varying flow channel shapes and sizes. By embedding the discretized parameters into the CAD design through scripting, these geometric models can be automatically generated in batches.

[0068] After generating the geometric model, the finite element method (FEM) is used to simulate multi-physics coupling. Multi-physics refers to the coupling effects of various physical phenomena, including heat transfer, fluid mechanics, and electricity. By solving governing equations such as the energy equation, momentum equation, and continuity equation, the system's physical field distribution (such as temperature and flow velocity) under different design parameters is simulated. For a specific heat exchange structure, the internal heat flow distribution and current flow path are simulated to obtain corresponding physical quantities such as temperature distribution and fluid velocity.

[0069] Extract valid data from the simulation results, including multiple physical quantities such as temperature, pressure, and flow rate. After extraction, this data needs to be standardized, de-noised, and output in a unified format to ensure smooth subsequent data analysis and model training. Extract the system's maximum temperature, average temperature, and pressure drop under different design conditions from the simulation results, then format this data to ensure consistency and comparability across datasets.

[0070] Feature extraction on formatted data aims to identify key physical quantities or design parameters from large amounts of simulation data. This feature data simplifies the data dimension while retaining the main factors affecting system performance, providing training data for subsequent neural network proxy models. For example, core indicators such as temperature gradient, fluid pressure loss, and electrical power can be extracted from simulation data and used as feature data.

[0071] The neural network operator proxy model is trained using feature-extracted data. By learning the complex relationship between inputs (such as design parameters) and outputs (such as system performance) in multi-physics simulation data, the neural network operator proxy model can quickly predict performance under different design parameters. The goal of the training process is to achieve a preset number of training times or meet specific accuracy requirements. By continuously adjusting the weights and biases of the neural network, it can predict the heat transfer performance under given design parameters with high accuracy, ultimately generating an efficient proxy model to replace complex numerical simulation processes.

[0072] Ultimately, through discretization, geometric modeling, multi-physics simulation, data extraction, and feature extraction, a neural network surrogate model was used to learn the input-output relationship in the simulation data, resulting in a surrogate model that can quickly predict performance. This surrogate model significantly reduces the computational cost of complex numerical simulations and accelerates the efficiency of design optimization.

[0073] In some examples, the multi-physics coupled finite element simulation includes heat conduction simulation, fluid flow simulation, and current conduction simulation.

[0074] For example, heat conduction simulation aims to predict the temperature distribution at different locations within and on the surface of a component. It simulates how heat is transferred within a solid structure, particularly through heat conduction through the component material. Fluid flow simulation simulates the flow behavior of fluids (such as air or coolant) within or around a component, primarily addressing fluid dynamics. It analyzes how coolant flows through flow channels and removes heat from the component. Current conduction simulation analyzes the current distribution within a component, particularly the electrical energy conduction process in photovoltaic systems. It describes how current is conducted through materials and analyzes the impact of different design parameters on the current distribution. These simulations are not isolated but rather interact with each other. For example, heat conduction and current conduction affect each other because changes in temperature alter the resistivity of the material, and Joule heating generated by the current affects the temperature distribution. Simultaneously, the cooling effect of the fluid flow also affects the overall thermal conductivity characteristics of the component. Therefore, these three physics fields need to be coupled together for simulation to accurately capture the interplay between the different physical phenomena.

[0075] For example, multi-physics field coupling finite element simulation can be performed using, but not limited to, the following formulas:

[0076] 1) Continuity equation

[0077]

[0078] 2) Momentum equation

[0079]

[0080] 3) Energy equation

[0081]

[0082] In the above formula, u, v, w are the velocity components in the x, y, and z directions respectively; ρ, v w , c p , k m are density, dynamic viscosity, specific heat capacity and thermal conductivity respectively; T is the temperature term; S h is the internal heat source term.

[0083] 4) Circuit equation

[0084] I=I ph -I D -I sh (18)

[0085]

[0086] I ph,ref =I sc,ref (27)

[0087]

[0088] In the above formula, I is the output current; I ph is the photocurrent; I D is the dark current; I0 ​​is the reverse saturation current; I sh is the current through the internal parallel resistor; q is the basic charge; U is the output voltage; R s is the internal series resistance; n is the ideality factor; k is the Boltzmann constant; T is the battery temperature; R sh is the internal parallel resistance; E g is the band gap width of the solar cell material; α ISC is the temperature coefficient of short-circuit current; α Rs is the temperature coefficient of the internal series resistance; I SC is the short-circuit current; V OC The subscript ref in the above formula indicates that the parameter value is the parameter value under the standard test conditions of solar cells.

[0089] In some examples, the above-mentioned discrete model of the heat exchange zone to be designed and the corresponding evaluation indicators under all corresponding design parameters are subjected to multi-objective optimization and Pareto front solution set generation operations to obtain a Pareto front solution set, including:

[0090] Generate an initial population based on the discrete model of the heat exchange zone to be designed and the corresponding evaluation indicators under all the corresponding design parameters;

[0091] Perform fitness evaluation on the above initial population to obtain the fitness of the individuals in the population;

[0092] According to the individual fitness of the above population, the population evolution is guided by genetic operation and based on non-dominated sorting to obtain a new generation of population and its Pareto frontier solution set.

[0093] In some examples, the initial population is generated based on a Latin hypercube sampling method.

[0094] In some examples, the objective functions corresponding to the fitness evaluation operation include maximizing thermal power output, minimizing temperature gradient, and maximizing electrical power output.

[0095] For example, Figure 4The figure shows a schematic diagram of the Pareto front solution set obtained by solving. In the first step of multi-objective optimization, generating the initial population is crucial. Each individual in the population represents a possible design scheme, and its characteristics are determined by the discrete model of the heat exchange zone to be designed and the corresponding design parameters. One of the methods for generating the initial population is Latin Hypercube Sampling (LHS), which can sample uniformly in the multidimensional design space, thereby ensuring the diversity of the population and covering the possible range of design parameters. For example: based on the geometric and physical characteristics of the discrete model of the heat exchange zone to be designed, the LHS method is used to generate the initial population. Each individual may represent a flow channel design scheme, with specific parameters such as flow channel width, length, curvature, etc.

[0096] After the initial population is generated, the fitness of each individual in the population is evaluated. The basis for fitness evaluation is a predefined objective function, which reflects the specific design requirements of multi-objective optimization. In the heat exchange structure design of photovoltaic thermal components, common objective functions include:

[0097] Maximize thermal power output: Ensure the design performs best in terms of thermal conversion efficiency.

[0098] Minimize temperature gradients: Reduce uneven temperature distribution in the system and prevent local overheating.

[0099] Maximize electrical power output: Improve the electrical power output performance of components.

[0100] For each design solution, a neural network operator proxy model is used to quickly predict the values ​​of these objective functions. A heat exchange zone design solution may perform well in some aspects but poorly in others, so multi-objective optimization can find a balance that suits different design requirements.

[0101] After completing the fitness evaluation, the population is evolved using three operations in the genetic algorithm:

[0102] Selection operation: Through methods such as roulette wheel selection or tournament selection, better individuals are selected based on fitness values ​​to enter the next generation.

[0103] Crossover operation: Through genetic recombination, the design parameters of two individuals are combined to generate new individuals and increase population diversity.

[0104] Mutation operation: By randomly changing the design parameters of some individuals, the population can explore new design spaces and avoid falling into local optimal solutions.

[0105] After each iteration, the population is guided by non-dominated sorting, which updates the Pareto front of the population. Non-dominated sorting evaluates each individual's performance in the multi-objective space and selects individuals that are not dominated by other individuals to be included in the Pareto front of the population. These solutions represent the optimal trade-off between the various objective functions.

[0106] Through genetic manipulation, we continuously optimize design options. One solution might excel in thermal power output but have lower electrical power output, while another offers a clear advantage in temperature gradient control. Through non-dominated sorting, we determine which individuals will enter the Pareto frontier solution set, gradually approaching the global optimal solution.

[0107] Multi-objective optimization generates new populations through multiple iterations and continuously updates the Pareto front solution set. The iterative process is typically terminated when a preset maximum number of iterations is reached or when the change in the optimization metric between generations of populations falls below a set convergence threshold.

[0108] Through the aforementioned multi-objective optimization process, particularly in the reference-direction-based non-dominated sorting genetic algorithm (PB-NSGA-III), the optimization process can find the optimal trade-offs between different design objectives (such as maximizing thermal power output, minimizing temperature gradients, and maximizing electrical power output) through genetic operations and non-dominated sorting, generating a Pareto front solution set. These solution sets provide designers with a variety of design options, ultimately allowing them to select the appropriate design based on actual needs. In this optimization method, the generation of the initial population (e.g., based on Latin hypercube sampling), the efficient implementation of fitness evaluation, and the evolutionary operation performed by the genetic algorithm jointly ensure the effectiveness and comprehensiveness of the optimization process.

[0109] In some examples, obtaining the comprehensive weight information of the evaluation indicators includes:

[0110] Determine the subjective weight information corresponding to different evaluation indicators based on the best-worst method;

[0111] Calculate objective weight information based on the CRITIC method;

[0112] The weight values ​​of the subjective weight information and the objective weight information are solved according to the Nash equilibrium solution method of game theory to obtain comprehensive weight information.

[0113] For example, the best-worst method (BWM) is a method for calculating the subjective weight of each evaluation indicator. BWM calculates the subjective weight of each evaluation indicator by constructing a one-to-many comparison matrix to compare the relative importance of each evaluation indicator with the most important and least important indicators. This method improves the efficiency and consistency of weight calculation by reducing the number of pairwise comparisons. The specific steps include: first, determining the most important and least important evaluation indicators; then, evaluating the relative importance of other indicators and these two extreme indicators to form a comparison matrix; finally, calculating the weight of each indicator by solving the optimization model. The advantage of the BWM method lies in its high efficiency and good stability, which can quickly obtain the subjective weight of the evaluation indicator without losing accuracy.

[0114] V BO =(v B1 ,v B2 ,…,v Bn ) (29)

[0115] V OW =(v 1W ,v 2W ,…,v nW ) (30)

[0116]

[0117] In the above formula, V BO is the importance vector of evaluation index; V OW is the unimportance vector of evaluation indicators; ξ is the consistency index; w j 、w B and w W are the weight of the jth evaluation index, the weight of the most important evaluation index, and the weight of the least important evaluation index; v Bj and v jW They are the preference degree of the most important evaluation indicator relative to the j-th evaluation indicator and the preference degree of the j-th evaluation indicator relative to the least important evaluation indicator.

[0118] Calculate objective weights using the CRITIC method: The CRITIC (Criteria Importance Through Intercriteria Correlation) method is used to calculate the objective weight of each evaluation indicator. It quantifies the importance of an indicator through the variability and correlation of the evaluation indicators. Specifically, the CRITIC method measures the amount of information and independence of each indicator in the overall evaluation system by calculating the standard deviation of each indicator and the correlation coefficient with other indicators. The standard deviation reflects the variability of the indicator, while the correlation coefficient reflects the redundancy between indicators. By comprehensively considering these two factors, the CRITIC method can effectively identify evaluation indicators that have both high variability and low correlation with other indicators, and assign them higher weights. This method focuses on the objective attributes of the data and can effectively reduce subjective bias in multi-objective comprehensive evaluations.

[0119]

[0120]

[0121] S is the normalized decision matrix; m is the number of Pareto solutions; n is the number of evaluation indicators; Z is the original decision matrix composed of Pareto solutions; and are the best and worst values ​​of the jth evaluation index respectively; w j is the weight of the jth evaluation index; σ j is the standard deviation of the jth criterion; r jj’ is the correlation coefficient between the two evaluation indicators.

[0122] Game theory combination weighting: After obtaining the subjective and objective weights, they need to be combined to form the final weight set. The weight values ​​of the subjective and objective weights are solved according to the game theory Nash equilibrium solution method. 。

[0123]

[0124] wg is the final combination weight; w i are the weight values ​​obtained from the subjective and objective perspectives respectively; a i is the weighting coefficient of the weight obtained from the subjective and objective perspectives respectively; a i * is the normalized weighting coefficient.

[0125] like Figure 4 As shown, this application proposes a device for optimizing the heat exchange structure of a concentrated photovoltaic and thermal assembly, comprising:

[0126] A first acquisition unit 21 is configured to input a discrete model of the heat exchange zone to be designed into a neural network operator proxy model to obtain corresponding evaluation indicators under different design parameters, wherein the network operator proxy model is obtained by performing a preset number of training on a multi-physics field coupled finite element simulation result set, and the design parameters include shape characteristics of the flow region in the heat exchange zone to be designed;

[0127] The second acquisition unit 22 is used to perform multi-objective optimization and Pareto front solution set generation operations on the discrete model of the heat exchange zone to be designed and the corresponding evaluation indicators under all the corresponding design parameters, so as to obtain a Pareto front solution set;

[0128] The third obtaining unit 23 is used to obtain the comprehensive weight information of the above evaluation indicators;

[0129] The determination unit 24 is configured to perform a comprehensive evaluation operation by combining TOPSIS and GRA based on the comprehensive weight information and the Pareto front solution set to determine the target flow area.

[0130] like Figure 5 As shown, an embodiment of the present application also provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored on the memory 310 and executable on the processor. When the processor 320 executes the computer program 311, the steps of any of the above-mentioned methods for optimizing the heat exchange structure of the concentrated photovoltaic and thermal components are implemented.

[0131] Since the electronic device introduced in this embodiment is a device used to implement a concentrated photovoltaic and thermal component heat exchange structure optimization design device in the embodiment of this application, based on the method introduced in the embodiment of this application, technical personnel in this field can understand the specific implementation of the electronic device of this embodiment and its various variations. Therefore, how the electronic device implements the method in the embodiment of this application will not be introduced in detail here. As long as the equipment used by technical personnel in this field to implement the method in the embodiment of this application falls within the scope of protection of this application.

[0132] In the specific implementation process, the computer program 311 can be implemented when executed by the processor Figure 1 Any implementation manner in the corresponding embodiments.

[0133] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0134] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0135] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0136] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0137] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0138] An embodiment of the present application also provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device executes the process of optimizing the heat exchange structure of the concentrated photovoltaic and thermal components in the corresponding embodiment.

[0139] A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)).

[0140] Those skilled in the art will 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.

[0141] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0142] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0143] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0144] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0145] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some 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 application.

Claims

1. A method for optimizing the heat exchange structure of a concentrated photovoltaic and thermal assembly, characterized in that: include: Inputting the discrete model of the heat exchange zone to be designed into a neural network operator proxy model to obtain corresponding evaluation indicators under different design parameters, wherein the neural network operator proxy model is obtained by performing a preset number of training on a multi-physics field coupled finite element simulation result set, and the design parameters include the shape characteristics of the flow area in the heat exchange zone to be designed; Perform multi-objective optimization and Pareto front solution set generation operations on the discrete model of the heat exchange zone to be designed and the corresponding evaluation indicators under all its corresponding design parameters to obtain a Pareto front solution set; Obtaining comprehensive weight information of the evaluation indicators; A comprehensive evaluation operation is performed using a combination of TOPSIS and GRA based on the comprehensive weight information and the Pareto front solution set to determine the target flow area; The obtaining of the comprehensive weight information of the evaluation index includes: Determine the subjective weight information corresponding to different evaluation indicators based on the best-worst method; Calculate objective weight information based on the CRITIC method; The subjective weight information and the objective weight information are weighted according to the Nash equilibrium solution method of game theory to obtain comprehensive weight information.

2. The method for optimizing the heat exchange structure of a concentrated photovoltaic and thermal assembly according to claim 1, characterized in that: The specific steps of obtaining the neural network operator proxy model include: Discretize the model of the heat exchange area to be designed to obtain a discrete model of the heat exchange area to be designed; Automatically generate a corresponding geometric model according to the discrete model of the heat exchange area to be designed and the discretized design parameters; Perform multi-physics field coupling finite element simulation according to the geometric model to obtain simulation result data; Performing data extraction and formatting output operations on the simulation result data to obtain formatted data; performing feature extraction on the formatted data to obtain feature-extracted data; The neural network operator proxy model is trained according to the heat exchange area model to be designed and the feature extraction data to achieve a preset number of training times or a preset accuracy to obtain the neural network operator proxy model.

3. The method for optimizing the heat exchange structure of a concentrated photovoltaic and thermal assembly according to claim 1, characterized in that: The multi-physics field coupled finite element simulation includes heat conduction simulation, fluid flow simulation and current conduction simulation.

4. The method for optimizing the heat exchange structure of a concentrated photovoltaic and thermal assembly according to claim 1, wherein: The multi-objective optimization and Pareto front solution set generation operations are performed on the discrete model of the heat exchange zone to be designed and the corresponding evaluation indicators under all corresponding design parameters to obtain the Pareto front solution set, including: Generate an initial population according to the discrete model of the heat exchange zone to be designed and the corresponding evaluation indicators under all the corresponding design parameters; Performing a fitness evaluation operation on the initial population to obtain individual fitness of the population; According to the fitness of the individuals in the population, the population is guided to evolve through genetic operations based on non-dominated sorting to obtain a new generation of population and its Pareto frontier solution set.

5. The method for optimizing the heat exchange structure of a concentrated photovoltaic and thermal assembly according to claim 4, characterized in that: The initial population is generated based on the Latin hypercube sampling method.

6. The method for optimizing the heat exchange structure of a concentrated photovoltaic and thermal assembly according to claim 4, characterized in that: The objective functions corresponding to the fitness evaluation operation include maximizing thermal power output, minimizing temperature gradient, and maximizing electrical power output.

7. A device for optimizing the heat exchange structure of a concentrated photovoltaic and thermal assembly, characterized in that: include: a first acquisition unit, configured to input a discrete model of the heat exchange zone to be designed into a neural network operator proxy model to obtain corresponding evaluation indicators under different design parameters, wherein the neural network operator proxy model is obtained by performing a preset number of training on a multi-physics field coupled finite element simulation result set, and the design parameters include shape characteristics of the flow area in the heat exchange zone to be designed; The second acquisition unit is used to perform multi-objective optimization and Pareto front solution set generation operations on the discrete model of the heat exchange zone to be designed and the corresponding evaluation indicators under all corresponding design parameters, so as to obtain a Pareto front solution set; A third obtaining unit is used to obtain comprehensive weight information of the evaluation index; a determination unit, configured to perform a comprehensive evaluation operation using a combination of TOPSIS and GRA based on the comprehensive weight information and the Pareto front solution set to determine a target flow area; The obtaining of the comprehensive weight information of the evaluation index includes: Determine the subjective weight information corresponding to different evaluation indicators based on the best-worst method; Calculate objective weight information based on the CRITIC method; The subjective weight information and the objective weight information are weighted according to the Nash equilibrium solution method of game theory to obtain comprehensive weight information.

8. An electronic device comprising: A memory and a processor, characterized in that the processor is used to implement the steps of the method for optimizing the heat exchange structure of a concentrating photovoltaic thermal component as described in any one of claims 1 to 6 when executing the computer program stored in the memory.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for optimizing the heat exchange structure of a concentrating photovoltaic and thermal assembly according to any one of claims 1 to 6 are implemented.

Images