A design method and system for a trackside box-type intelligent machine room for 5T equipment on a vehicle

Abstract

The present invention provides a design method and system for a trackside box-type intelligent machine room for 5T equipment on a vehicle, which relates to the field of railway safety monitoring technology, including: obtaining first information, second information and third information; performing parameter optimization processing and performance simulation processing according to the first information to obtain fourth information, the fourth information being the performance characteristics under the current design requirements; performing deep learning processing on the fourth information and the second information based on a preset convolutional neural network mathematical model to obtain fifth information; performing multi-objective optimization processing according to the fifth information and the third information to obtain sixth information; performing graphic conversion processing according to the sixth information to obtain eighth information, the eighth information being an abstract diagram including the heat dissipation system structure, control strategy and component relationship; performing adaptive design processing and outputting design results according to the eighth information. The present invention can optimize the heat dissipation effect and humidity control, and ensure that the machine room can maintain the best operating state under various environmental conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of railway safety monitoring, and in particular to a design method and system for a trackside box-type intelligent machine room for 5T equipment on a vehicle. Background Art

[0002] With the rapid development of my country's railway industry, the 5T system plays a vital role in the safety monitoring of railway vehicles. The system uses intelligent, networked and information technologies to perform dynamic detection, networked operation, remote monitoring and information sharing on railway vehicles to enhance the safety prevention capabilities of railway vehicle operation. In the current field of vehicle 5T equipment trackside box-type intelligent machine room design, the existing technology mainly relies on basic design requirements and empirical rules. These rules cannot fully consider the performance goals of heat dissipation and humidity control, resulting in a lack of systematic and optimized design results. Traditional design methods often use manual parameter adjustment, which lacks scientificity and efficiency, and is easily limited by inexperienced designers, resulting in uncertainty in design results and waste of resources. In addition, current technology has limitations in dealing with complex intelligent machine room layout and control strategy optimization, and cannot effectively adapt to the needs of multi-objective performance optimization.

[0003] Based on the shortcomings of the existing technology, there is an urgent need for a design method and system for a trackside box-type intelligent machine room for 5T equipment on a vehicle. Summary of the invention

[0004] The purpose of the present invention is to provide a design method and system for a 5T trackside box-type intelligent machine room for vehicles to improve the above problems. In order to achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0005] In a first aspect, the present application provides a design method for a trackside box-type intelligent machine room for 5T equipment in a vehicle, comprising:

[0006] Acquire first information, second information and third information, wherein the first information includes design requirements of the intelligent computer room, the design requirements include performance targets of heat dissipation effect and humidity control, the second information includes meteorological data of the deployment location, and the third information includes historical computer room design solutions and corresponding performance data;

[0007] Perform parameter optimization processing according to the first information, and obtain fourth information by iteratively adjusting parameters and performing performance simulation processing, wherein the fourth information is a performance characteristic under the current design requirements;

[0008] Performing deep learning processing on the fourth information and the second information based on a preset convolutional neural network mathematical model to obtain fifth information, wherein the fifth information includes at least one arrangement scheme of the heat dissipation system and a corresponding control strategy;

[0009] Perform multi-objective optimization processing according to the fifth information and the third information, and obtain sixth information by performing an optimization process under at least two performance indicators in combination with historical performance data;

[0010] Performing a graph conversion process on the sixth information to obtain a node graph of the cooling system, and adjusting the node graph of the cooling system based on a preset graph neural network mathematical model to obtain eighth information, wherein the eighth information is an abstract graph including a cooling system structure, a control strategy, and a component relationship;

[0011] Adaptive design processing is performed according to the eighth information, and a design result is output by combining the actual site layout, system structure relationships and performance requirements. The design result includes all modules of the intelligent computer room, the configuration of the internal space and the corresponding heat dissipation control strategy.

[0012] In the second aspect, the present application also provides a design system for a 5T trackside box-type intelligent machine room for vehicles, including:

[0013] an acquisition module, configured to acquire first information, second information, and third information, wherein the first information includes design requirements of the intelligent computer room, the design requirements include performance targets of heat dissipation effect and humidity control, the second information includes meteorological data of the deployment location, and the third information includes historical computer room design solutions and corresponding performance data;

[0014] A first optimization module, configured to perform parameter optimization processing according to the first information, and obtain fourth information by iteratively adjusting parameters and performing performance simulation processing, wherein the fourth information is a performance characteristic under a current design requirement;

[0015] A first processing module, configured to perform deep learning processing on the fourth information and the second information based on a preset convolutional neural network mathematical model to obtain fifth information, wherein the fifth information includes at least one arrangement scheme of a heat dissipation system and a corresponding control strategy;

[0016] A second optimization module, configured to perform multi-objective optimization processing according to the fifth information and the third information, and obtain sixth information by performing an optimization process under at least two performance indicators in combination with historical performance data;

[0017] a conversion module, configured to perform a graph conversion process according to the sixth information to obtain a cooling system node graph, and adjust the cooling system node graph based on a preset graph neural network mathematical model to obtain eighth information, wherein the eighth information is an abstract graph including a cooling system structure, a control strategy, and a component relationship;

[0018] The second processing module is used to perform adaptive design processing according to the eighth information, and output design results by combining the actual site layout, system structure relationships and performance requirements. The design results include all modules of the intelligent computer room, the configuration of the internal space and the corresponding heat dissipation control strategy.

[0019] The beneficial effects of the present invention are:

[0020] The present invention can perform more accurate performance simulation and prediction by comprehensively considering the design requirements, meteorological data and historical design solutions of the intelligent computer room. Using genetic algorithms and fluid dynamics simulation, the present invention can optimize heat dissipation and humidity control to ensure that the computer room can maintain optimal operating conditions under various environmental conditions.

[0021] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 Schematic diagram of the design method of the vehicle 5T equipment trackside box-type intelligent machine room embodiment of the present invention;

[0024] Figure 2 This is a node diagram of the heat dissipation system described in an embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0026] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance. Embodiment 1:

[0027] This embodiment provides a design method for a trackside box-type intelligent machine room for 5T equipment on a vehicle.

[0028] See also Figure 1 , the figure shows that the method includes step S100, step S200, step S300, step S400, step S500 and step S600.

[0029] Step S100, obtaining first information, second information and third information, wherein the first information includes design requirements of the intelligent computer room, the design requirements include performance targets of heat dissipation effect and humidity control, the second information includes meteorological data of the deployment location, and the third information includes historical computer room design solutions and corresponding performance data;

[0030] It is understandable that at the beginning of the intelligent computer room design project, key information needs to be collected and integrated first, which will provide the basis for subsequent design optimization and decision-making. The first information, namely the design requirements of the intelligent computer room, is the core guiding principle in the design process. This includes specific performance targets for heat dissipation and humidity control, such as temperature range, humidity range, air flow rate and other parameters, which are usually determined by the use requirements of the computer room and the specifications of the equipment manufacturer. The second information involves meteorological data at the deployment site, which is crucial for evaluating the external environmental conditions that the computer room may face, such as annual average temperature, humidity, wind speed, etc. These factors directly affect the design and efficiency of the cooling system. The third information includes historical computer room design solutions and performance data, which can help evaluate the advantages and disadvantages of past solutions and provide reference for new designs.

[0031] Step S200, performing parameter optimization processing according to the first information, and obtaining fourth information by iteratively adjusting the parameters and performing performance simulation processing, where the fourth information is the performance characteristics under the current design requirements;

[0032] It should be noted that in the process of parameter optimization, the performance of the computer room under different parameter combinations is simulated through algorithms according to the performance goals of heat dissipation and humidity control. This process involves complex mathematical models, such as heat conduction-fluid flow models, to predict the actual operation of the computer room under specific design parameters. Through multiple iterations, continuous adjustment of parameters and simulation of performance, a set of optimal parameters is finally obtained, which can achieve the best heat dissipation effect and humidity control while meeting the design requirements.

[0033] Step S300: performing deep learning processing on the fourth information and the second information based on a preset convolutional neural network mathematical model to obtain fifth information, where the fifth information includes at least one arrangement scheme of the heat dissipation system and a corresponding control strategy;

[0034] In this step, the fourth information and the second information are provided as input values ​​to a preset convolutional neural network for deep learning processing. During the training process, the network learns the complex relationship between different design parameters, performance characteristics, and meteorological data. This enables the network to predict the optimal cooling system layout and control strategy under the current design requirements. Convolutional neural networks have advantages in processing complex relationships and multi-dimensional data, and can capture the nonlinear relationship between design parameters, thereby providing a more accurate cooling system layout and control strategy. It helps to achieve a more intelligent cooling system design, improve performance while reducing energy consumption.

[0035] Step S400, performing multi-objective optimization processing according to the fifth information and the third information, and obtaining sixth information by performing an optimization process under at least two performance indicators in combination with historical performance data;

[0036] It is understandable that a multi-objective optimization algorithm is used in this process, and the algorithm will find a balance between multiple design parameters and performance indicators, taking into account historical performance data to ensure that the optimized solution is more stable and feasible in practical applications.

[0037] Step S500: performing a graph conversion process according to the sixth information to obtain a node graph of the cooling system, and adjusting the node graph of the cooling system based on a preset graph neural network mathematical model to obtain eighth information, where the eighth information is an abstract graph including a cooling system structure, a control strategy, and a component relationship;

[0038] It should be noted that this abstract graph reflects the layout and interrelationships of the optimized cooling system at the component level, providing a basis for subsequent adaptive design. Preferably, graphics processing tools and graph neural network libraries, such as NetworkX and DGL, can be used to represent the cooling system layout and control strategy in the sixth information as a node graph. Then, by applying graph neural network algorithms such as GCN, the node graph is learned and adjusted to obtain the eighth information. This process needs to be adjusted in combination with the preset mathematical model and the actual situation of the current intelligent computer room to ensure that the structure of the graph is consistent with the real system.

[0039] Step S600, perform adaptive design processing according to the eighth information, and output design results by combining the actual site layout, system structure relationships and performance requirements. The design results include all modules of the intelligent computer room, the configuration of the internal space and the corresponding heat dissipation control strategy.

[0040] Specifically, this process includes: determining the placement of each module by considering the space limitations and characteristics of the actual site to ensure that the layout of the intelligent computer room in the actual site is reasonable; ensuring the optimal layout and connection method between the components of each system by considering the relationship between them, so as to effectively realize information sharing and network operation, which includes the relative position and connection method of components such as heat dissipation equipment, sensors, and control units; incorporating performance requirements into the design process to ensure that the design results achieve the expected performance level in terms of heat dissipation and humidity control, which includes adjusting the power of the heat dissipation equipment to improve the humidity control efficiency, etc.; and, by optimizing the layout of the internal space, ensuring that the placement of each module not only meets the performance requirements, but also facilitates maintenance and management; at the same time, generating corresponding heat dissipation control strategies to ensure that heat dissipation and humidity control can be effectively achieved under various working conditions.

[0041] It should be noted that step S200 includes step S210, step S220, step S230 and step S240.

[0042] Step S210: According to the performance targets of heat dissipation effect and humidity control in the first information, an initial design parameter set is obtained through multi-objective optimization genetic algorithm processing;

[0043] It is understandable that the genetic algorithm for multi-objective optimization is a tool that can handle multiple contradictory objectives. It simulates the evolutionary process, through the operations of selection, crossover and mutation, and gradually optimizes the design parameters so that it can achieve a better balance under multiple performance objectives. The initial design parameter set obtained can achieve a relatively superior performance balance in terms of heat dissipation effect and humidity control performance objectives. In this application, the initial design parameter set is a set of design parameters, including various properties of the heat dissipation system components, the location of the humidity sensor, and the type of control device. The combination of these parameters is optimized through multi-objective optimization to meet the heat dissipation effect and humidity control performance objectives of the intelligent computer room, while achieving a balance in multiple aspects.

[0044] It should be noted that step S210 includes step S211, step S212, step S213 and step S214.

[0045] Step S211, defining the performance target as a multi-objective problem;

[0046] Furthermore, the multi-objective problem is to minimize energy consumption and maximize humidity control efficiency, and ensure that a preset threshold is reached in terms of heat dissipation effect.

[0047] Step S212, using a genetic algorithm to optimize the multi-objective problem to obtain design parameters;

[0048] It should be noted that the genetic algorithm simulates the process of biological evolution and gradually improves the set of design parameters through the operations of selection, crossover and mutation. This process goes through multiple rounds of iterations, and each round generates a new set of design parameters.

[0049] Step S213: In each iteration, we need to define a fitness function, which will evaluate the performance of the current design parameter set in the multi-objective space.

[0050] In each iteration, a fitness function is defined, which will evaluate the performance of the current design parameter set in the multi-objective space.

[0051] Step S214: After multiple rounds of iterations, the genetic algorithm will generate a population, in which the individuals are the initial design parameter set.

[0052] It should be noted that the initial design parameter set represents a relatively superior design under multiple objective performance indicators.

[0053] Step S220, performing fluid dynamics simulation processing according to the initial design parameter set, and obtaining performance simulation results by solving a preset heat conduction-fluid flow mathematical model, wherein the performance simulation results include heat dissipation effect and humidity control performance under at least two parameter combinations;

[0054] It can be understood that in this step, the fluid dynamics simulation method is used to establish and solve the heat conduction-fluid flow mathematical model to evaluate the performance of the initial design parameter set under different working conditions, focusing on the heat dissipation effect and humidity control performance. Specifically, the heat conduction-fluid flow mathematical model in this application comprehensively considers multiple physical factors such as heat conduction, fluid flow and humidity transfer, couples the heat conduction equation and the fluid flow equation, and considers the humidity transfer process at the same time, so that it is closer to the actual scene and can comprehensively describe the thermal and humid environment in the intelligent computer room. The formulas involved are as follows:

[0055] Heat conduction equation (heat transfer):

[0056] ;

[0057] in, represents the temperature distribution, is the thermal diffusion coefficient, represents the gradient operator, represents the Laplace operator of the temperature distribution, β represents the system-specific heat transfer parameter, Indicates the heat generated by the heat dissipation device. represents heat loss, Indicates time.

[0058] Fluid flow equations:

[0059] ;

[0060] ;

[0061] in, represents the air density, represents the velocity field, Indicates time, represents the gradient operator, represents the pressure field, is the dynamic viscosity coefficient, Represents the acceleration due to gravity.

[0062] Moisture transfer equation:

[0063] ;

[0064] in, Represents humidity distribution, represents the gradient operator, represents the humidity diffusion coefficient, Indicates humidity loss, Indicates time, Represents the specific heat ratio of moist air.

[0065] Energy consumption equation:

[0066] ;

[0067] in, Indicates energy consumption, represents the gradient operator, is the thermal diffusion coefficient, represents the velocity field, Represents humidity distribution, represents the humidity diffusion coefficient, represents the volume of the fluid, represents the energy consumption contribution of the temperature gradient, represents the kinetic energy contribution of the velocity field, represents the volume element, Represents the wet energy contribution of the humidity gradient.

[0068] Step S230: According to the performance simulation results, the corresponding surface method is used to perform data fitting and optimization processing, and a function expression containing key performance characteristics is obtained by establishing a relationship model between heat dissipation effect, humidity control and design parameters;

[0069] Specifically, the heat dissipation effect (H) and humidity control (C) are compared with the design parameters The relationship between is modeled as a polynomial function expression. This function expression adopts an infinite order summation form, which includes the influence of different orders of each parameter. This modeling method can more comprehensively consider the influence of each parameter on the heat dissipation effect and humidity control, making the model more flexible and adaptable. The function expression is:

[0070] ;

[0071] ;

[0072] Among them, H represents the heat dissipation effect, C represents the humidity control, and n represents the order. Indicates the type, quantity and layout of heat dissipation equipment, Indicates the location of the humidity sensor and the type of control device, Indicates the overall structure and layout.

[0073] Step S240: Perform iterative optimization processing according to the function expression to obtain fourth information.

[0074] It can be understood that iterative optimization processing refers to adjusting the coefficients in the function expression through multiple iterations to gradually optimize the performance model so that it can more accurately reflect the actual situation.

[0075] It should be noted that step S240 includes step S241, step S242 and step S243.

[0076] Step S241: According to the function expression and the design requirements of the intelligent computer room, the objective function is constructed by minimizing the energy consumption of the cooling system and maximizing the humidity control efficiency;

[0077] The objective function is:

[0078] ;

[0079] in, represents the objective function, Represents the design space of the intelligent computer room. represents the set of design parameters, Indicates the heat dissipation effect. Indicates the humidity control efficiency. represents the overall function of heat dissipation and humidity control effectiveness, Represents the influence function of design parameters on heat dissipation effect, represents the influence function of design parameters on humidity control efficiency, represents the index value of the design parameter, and n represents the total number of design parameters.

[0080] Step S242, obtaining objective function gradient information according to the objective function calculation, where the objective function gradient information is the rate of change of the performance characteristic in the design parameter space;

[0081] It should be noted that the objective function gradient corresponds to the partial derivative of each design parameter in the objective function. It provides the direction and rate of change of the objective function when the design parameters change, and reflects the degree of influence of adjusting each parameter on the objective function in the design parameter space.

[0082] Step S243, performing iterative calculation according to the objective function, the objective function gradient information and the preset gradient descent mathematical model to obtain fourth information.

[0083] Gradient descent is an optimization algorithm used to find the minimum value of an objective function. The basic idea is to gradually reduce the value of the objective function by iteratively updating the design parameters. The iterative update rule of gradient descent in this application can be expressed as:

[0084] :

[0085] in, represents the number of iterations, represents the set of design parameters for the i-th iteration, represents the set of design parameters for the i+1th iteration, represents the learning rate, represents the gradient operator, represents the objective function.

[0086] It should be noted that step S300 includes step S310, step S320, step S330 and step S340.

[0087] Step S310: input the fourth information and the second information as input values ​​into a preset convolutional neural network for deep learning processing to construct a heat dissipation system layout prediction model;

[0088] It is understandable that the convolutional neural network is designed to learn and capture the complex relationships of the layout of the cooling system, so that it can predict the optimal layout of the cooling system given the design requirements and environmental data. In the present application, some specific improvements are made to the structure of the convolutional neural network to better meet the requirements of the design of intelligent computer rooms. These improvements mainly include the processing of input data and the optimization of the network architecture. Specifically, in the design of intelligent computer rooms, the input data involves the performance characteristics, design requirements and environmental data of the cooling system. In order to better capture these complex relationships, specific mathematical transformations are introduced, such as embedding the performance characteristics and design requirements to improve the network's perception of key features. Such mathematical transformations are achieved by applying appropriate nonlinear mapping or embedding methods to the input data. At the same time, considering that the design of intelligent computer rooms involves multiple data types, including performance characteristics and environmental data, the idea of ​​cross-modal information fusion is introduced. First, the features of different data types are fused to create a richer feature representation. Different parts of the input data, such as performance features and environmental data, are fused through splicing, addition, or other appropriate operations, so that the network can simultaneously consider the impact of different features and more comprehensively understand the relationship between design parameters and performance; design a multi-branch network structure, with each branch specializing in one data type. Each branch can have independent convolutional layers and pooling layers, and then feature fusion is performed at a higher level. This parallel processing helps the network learn the associations between different data types more effectively.

[0089] Step S320, performing prediction processing according to the heat dissipation system layout prediction model, and obtaining a predicted layout plan by inputting parameters and structural characteristics of the heat dissipation system;

[0090] It can be understood that the model performs prediction processing based on the input parameters and structural features, and outputs the optimal cooling system layout plan. This plan includes the location, quantity, layout of the cooling equipment, and related control strategies, such as fan speed adjustment, radiator working status, etc.

[0091] Step S330: Optimize the control strategy according to the prediction scheme and the preset deep deterministic policy gradient mathematical model, and obtain the control strategy by modeling the state and environment inside the intelligent computer room as a continuous state space and taking the control parameters to be adjusted as continuous actions;

[0092] It can be understood that the state space in this step includes state variables related to heat dissipation effect, humidity distribution, temperature distribution, etc., and the control parameters that need to be adjusted (such as heat dissipation device power, fan speed, etc.) are used as continuous actions. The deep deterministic policy gradient mathematical model is a deep learning model, usually in the form of a neural network, which is used to approximate the function of the control strategy of the intelligent computer room. The model receives the state of the intelligent computer room as input and outputs the probability distribution of continuous actions. In the optimization process, the optimization algorithm (such as gradient descent method, Adam optimizer, etc.) is used to adjust the parameters of the policy network so that the output control strategy can maximize the performance characteristics of the intelligent computer room. In this process, the policy network generates a probability distribution of continuous actions instead of discrete actions, and these action values ​​will be applied to the control parameters of the intelligent computer room, such as the power of the heat dissipation device, the speed of the fan, etc. Finally, by adjusting the parameters of the policy network, the intelligent computer room can make appropriate adjustments in the continuous state space, so as to achieve the goal of optimal performance characteristics.

[0093] Step S340: Obtain fifth information according to the predicted layout plan and control strategy.

[0094] It is understandable that the fifth information includes the specific configuration of control parameters such as the layout position of the heat dissipation device, the speed of the fan, the power of the heat dissipation device, etc. By comprehensively considering the performance characteristics such as heat dissipation effect and humidity control, it is ensured that the intelligent computer room can achieve efficient operation under various workloads and environmental conditions.

[0095] It should be noted that step S400 includes step S410, step S420 and step S430.

[0096] Step S410: construct at least two optimization objective functions according to the fifth information and the third information;

[0097] It is understandable that the setting of the optimization objective function will take into account the performance characteristics of the smart room, such as the improvement of heat dissipation effect, the accuracy of humidity control, etc. By constructing at least two optimization objective functions, multiple performance aspects of the smart room can be comprehensively considered, as well as the performance under different conditions. This can better guide the subsequent optimization process and ensure that the design of the smart room can achieve the ideal performance level in actual application.

[0098] Step S420: construct an optimization model using a multi-objective optimization algorithm according to all optimization objective functions;

[0099] Furthermore, the use of multi-objective optimization algorithms aims to find a set of design parameters and control strategies that make all optimization objective functions reach optimal or near-optimal states. When constructing an optimization model, it is necessary to consider the mutual influence and trade-off relationship between various optimization objectives, as well as the actual constraints of the intelligent computer room design. Through such an optimization model, a more comprehensive and effective guidance can be provided for the design and control of the intelligent computer room to achieve better performance.

[0100] Step S430: Initialize and solve the optimization model and historical performance data to obtain sixth information, where the sixth information is an optimized heat dissipation system layout and control strategy.

[0101] This step aims to maximize or minimize multiple optimization objective functions so that the cooling system layout scheme and control strategy can achieve the best performance in the intelligent computer room.

[0102] It should be noted that step S500 includes step S510, step S520 and step S530.

[0103] Step S510: Obtain a heat dissipation system node graph according to the sixth information conversion, where points in the node graph represent heat dissipation system components, and edges represent control relationships between heat dissipation system components;

[0104] It is understandable that if Figure 2 As shown in the figure, in the cooling system node graph, each node represents a component in the cooling system, such as a radiator, fan, sensor, etc. These components are connected by edges, which represent the interactions and dependencies between them. For example, there may be a connection between the radiator and the fan, indicating that the fan helps the radiator dissipate heat by blowing air. Similarly, the sensor may be connected to the control unit to monitor the temperature and humidity in real time and adjust the working state of the cooling system according to the monitoring results. By constructing a cooling system node graph, we can have a clearer understanding of the overall architecture of the cooling system and the relationship between the various components. This helps us identify the key components and bottlenecks in the system and provides guidance for further optimization and improvement of the system.

[0105] Step S520: performing feature extraction processing according to the cooling system node graph, mapping node attributes and capturing contextual relationships to obtain a feature set;

[0106] It should be noted that feature extraction is the process of converting a complex node graph into a numerical feature representation so that the computer algorithm can understand and process it. In this step, the attributes of each node are first mapped to convert the physical characteristics, functional characteristics and other information of the node into quantifiable feature values. For example, for the radiator node, its size, heat dissipation efficiency and other attributes are extracted as features; for the fan node, its speed, air volume and other attributes are extracted as features. In addition to the attributes of a single node, it is also necessary to capture the contextual relationship between nodes, that is, the connection and dependency between nodes. This can be achieved by calculating the edges in the graph. For example, the number of neighbor nodes of each node can be calculated to indicate the connection density of the node in the graph; or the path length between nodes can be calculated to indicate the efficiency of information transmission between nodes. Through feature extraction processing, the complex cooling system node graph can be converted into a structured and computable feature set, providing input data for subsequent machine learning models and algorithms.

[0107] Step S530: perform learning and node adjustment processing on the feature set according to the preset graph neural network mathematical model to obtain the eighth information.

[0108] Graph neural network is a deep learning model specially designed for processing graph structure data, which can effectively capture the relationship between nodes in the graph and the global topological structure. In this step, the feature set is used as input, and the graph neural network model is trained and learned to obtain a higher level of abstract representation, so as to better understand the operation mechanism and performance characteristics of the cooling system. Through the learning process of the graph neural network, the nodes of the cooling system can be adjusted and optimized. This involves operations such as updating node attributes and adjusting the connection relationship between nodes, so that the performance of the entire system is improved. Specifically, the importance of nodes in the cooling system can be learned through the graph neural network model, and the weighted adjustment of nodes can be performed accordingly to improve the performance indicators such as the cooling efficiency or humidity control accuracy of the system. The eighth information finally obtained contains an abstract graph of the cooling system structure, control strategy and component relationship after graph neural network learning and node adjustment. This abstract graph will provide an important reference for subsequent system optimization and improvement, and help better design and manage the operation of intelligent computer rooms.

[0109] It should be noted that step S600 includes step S610, step S620 and step S630.

[0110] Step S610: Perform model building processing according to the eighth information, and obtain a fuzzy logic model by converting the spatial distribution and relative positions of the components into fuzzy logic rules;

[0111] It should be noted that the use of fuzzy logic rules in this application to describe the spatial relationship between the components of the cooling system has unique advantages. Since the position and relative position of the components inside the intelligent room are often not strictly fixed, but are affected by many factors, such as the layout of the room, the installation method of the equipment, etc., it is difficult for precise logic to accurately describe the spatial relationship between them. Fuzzy logic rules can flexibly handle this ambiguity and uncertainty, and describe the spatial relationship between components, such as "closer", "farther", etc. through fuzzy sets and membership functions. Specifically, it is first necessary to convert the spatial distribution of the components of the cooling system into fuzzy logic rules, which includes determining the position, layout and distance relationship between each cooling component. By analyzing and abstracting this information, a set of fuzzy logic rules is established to describe the spatial relationship between different components, such as proximity, relative position, etc. Secondly, converting the control strategy into fuzzy logic rules involves converting the control strategy into a series of fuzzy rules to describe what cooling measures should be taken under different environmental conditions. For example, according to the changes in temperature and humidity, fuzzy rules are formulated to adjust control parameters such as fan speed and cooling device power. Finally, the spatial distribution and relative position of the components are combined with the control strategy to construct a complete fuzzy logic model. Based on fuzzy logic reasoning, this model can automatically adjust the operating status of the cooling system according to real-time environmental conditions and control strategies to achieve optimal cooling effect and energy consumption control.

[0112] Step S620, performing performance optimization processing according to the fuzzy logic model and the preset multi-objective particle swarm optimization mathematical model to obtain an optimization parameter set;

[0113] It is understandable that the fuzzy logic model converts the spatial distribution of components and control strategies of the intelligent computer room into fuzzy rules to form multiple optimization objectives. These objectives are then optimized with the help of the multi-objective particle swarm optimization algorithm. By continuously iteratively updating the position and speed of particles, the optimization algorithm can search for a set of optimal solutions in the solution space, so that the system can achieve the best state in terms of heat dissipation, energy consumption, humidity control and other aspects.

[0114] Step S630: adjusting the heat dissipation system node graph according to the optimization parameter set to obtain an optimized node graph, and combining the optimized node graph for transformation to obtain a design result.

[0115] Through this process, the cooling system can be adjusted and optimized by making full use of the optimization parameter set, the optimal node layout plan can be obtained, and it can be converted into actual design results, providing a feasible solution for improving the performance and reducing energy consumption of smart computer rooms. Embodiment 2:

[0116] This embodiment provides a design system for a 5T trackside box-type intelligent machine room for vehicles, the system comprising:

[0117] an acquisition module, configured to acquire first information, second information, and third information, wherein the first information includes design requirements of the intelligent computer room, the design requirements include performance targets of heat dissipation effect and humidity control, the second information includes meteorological data of the deployment location, and the third information includes historical computer room design plans and corresponding performance data;

[0118] A first optimization module is used to perform parameter optimization processing according to the first information, and obtain fourth information by iteratively adjusting the parameters and performing performance simulation processing, where the fourth information is the performance characteristics under the current design requirements;

[0119] A first processing module, configured to perform deep learning processing on the fourth information and the second information based on a preset convolutional neural network mathematical model to obtain fifth information, wherein the fifth information includes at least one arrangement scheme of a heat dissipation system and a corresponding control strategy;

[0120] A second optimization module is used to perform multi-objective optimization processing according to the fifth information and the third information, and obtain sixth information by performing an optimization process under at least two performance indicators in combination with historical performance data;

[0121] a conversion module, configured to perform a graph conversion process on the sixth information to obtain a node graph of the cooling system, and to adjust the node graph of the cooling system based on a preset graph neural network mathematical model to obtain eighth information, wherein the eighth information is an abstract graph including a cooling system structure, a control strategy, and a component relationship;

[0122] The second processing module is used to perform adaptive design processing according to the eighth information, and output design results by combining the actual site layout, system structure relationships and performance requirements. The design results include all modules of the intelligent computer room, the configuration of the internal space and the corresponding heat dissipation control strategy.

[0123] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0124] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A design method for a 5T trackside box-type intelligent machine room for a vehicle, characterized in that: include: Acquire first information, second information and third information, wherein the first information includes design requirements of the intelligent computer room, the design requirements include performance targets of heat dissipation effect and humidity control, the second information includes meteorological data of the deployment location, and the third information includes historical computer room design solutions and corresponding performance data; Perform parameter optimization processing according to the first information, and obtain fourth information by iteratively adjusting parameters and performing performance simulation processing, wherein the fourth information is a performance characteristic under the current design requirements; Performing deep learning processing on the fourth information and the second information based on a preset convolutional neural network mathematical model to obtain fifth information, wherein the fifth information includes at least one arrangement scheme of the heat dissipation system and a corresponding control strategy; Perform multi-objective optimization processing according to the fifth information and the third information, and obtain sixth information by performing an optimization process under at least two performance indicators in combination with historical performance data; Performing a graph conversion process on the sixth information to obtain a cooling system node graph, and adjusting the cooling system node graph based on a preset graph neural network mathematical model to obtain eighth information, wherein the eighth information is an abstract graph including a cooling system structure, a control strategy, and a component relationship; Performing adaptive design processing according to the eighth information, and outputting a design result by combining the actual site layout, the relationship between the system structure and the performance requirements, wherein the design result includes all modules of the intelligent computer room, the configuration of the internal space and the corresponding heat dissipation control strategy; Wherein, the fourth information and the second information are subjected to deep learning processing based on a preset convolutional neural network mathematical model to obtain the fifth information, including: Inputting the fourth information and the second information as input values ​​into a preset convolutional neural network for deep learning processing to construct a heat dissipation system layout prediction model; Performing prediction processing according to the heat dissipation system layout prediction model, and obtaining a prediction layout scheme by inputting parameters and structural characteristics of the heat dissipation system; The control strategy is optimized according to the predicted layout scheme and the preset deep deterministic policy gradient mathematical model, and the control strategy is obtained by modeling the state and environment inside the intelligent computer room as a continuous state space and taking the control parameters to be adjusted as continuous actions; Fifth information is obtained according to the predicted arrangement plan and the control strategy.

2. The design method of the trackside box-type intelligent machine room for 5T equipment in a vehicle according to claim 1 is characterized in that , performing parameter optimization processing according to the first information, and obtaining fourth information by iteratively adjusting parameters and performing performance simulation processing, including: According to the performance objectives of heat dissipation effect and humidity control in the first information, an initial design parameter set is obtained through a multi-objective optimization genetic algorithm process; Performing fluid dynamics simulation processing according to the initial design parameter set, and obtaining performance simulation results by solving a preset heat conduction-fluid flow mathematical model, wherein the performance simulation results include heat dissipation effect and humidity control performance under at least two parameter combinations; According to the performance simulation results, the corresponding surface method is used to perform data fitting and optimization processing, and a function expression containing key performance characteristics is obtained by establishing a relationship model between heat dissipation effect, humidity control and design parameters; The fourth information is obtained by performing iterative optimization processing according to the function expression.

3. The design method of the trackside box-type intelligent machine room for 5T equipment in a vehicle according to claim 2 is characterized in that , performing iterative optimization processing according to the function expression to obtain fourth information, including: According to the functional expression and the design requirements of the intelligent computer room, the objective function is constructed by minimizing the energy consumption of the cooling system and maximizing the humidity control efficiency; Obtain objective function gradient information according to the objective function calculation, wherein the objective function gradient information is a rate of change of performance characteristics in a design parameter space; Iterative calculation is performed according to the objective function, the objective function gradient information and a preset gradient descent mathematical model to obtain fourth information.

4. The design method of the trackside box-type intelligent machine room for 5T equipment in a vehicle according to claim 1 is characterized in that , performing multi-objective optimization processing according to the fifth information and the third information, and obtaining the sixth information by combining the historical performance data with the optimization process under at least two performance indicators, including: Constructing at least two optimization objective functions according to the fifth information and the third information; According to all the optimization objective functions, an optimization model is constructed using a multi-objective optimization algorithm; The optimization model and historical performance data are initialized and solved to obtain sixth information, where the sixth information is an optimized heat dissipation system layout and control strategy.

5. The design method of the trackside box-type intelligent machine room for 5T equipment in a vehicle according to claim 1 is characterized in that , performing a graph conversion process according to the sixth information to obtain a cooling system node graph, and adjusting the cooling system node graph based on a preset graph neural network mathematical model to obtain eighth information, including: A node graph of the cooling system is obtained according to the sixth information conversion, where points in the node graph represent cooling system components, and edges represent control relationships between cooling system components; Performing feature extraction processing according to the cooling system node graph, obtaining a feature set by mapping node attributes and capturing contextual relationships; The feature set is studied and node-adjusted according to a preset graph neural network mathematical model to obtain the eighth information.

6. The design method of the trackside box-type intelligent machine room for 5T equipment in a vehicle according to claim 1 is characterized in that , performing adaptive design processing according to the eighth information, and outputting design results by combining the actual site layout, system structure relationships and performance requirements, including: Performing model building processing according to the eighth information, obtaining a fuzzy logic model by converting the spatial distribution and relative positions of the components into fuzzy logic rules; Perform performance optimization processing according to the fuzzy logic model and a preset multi-objective particle swarm optimization mathematical model to obtain an optimization parameter set; The heat dissipation system node graph is adjusted according to the optimization parameter set to obtain an optimized node graph, and the optimized node graph is combined and transformed to obtain a design result.

7. A design system for a 5T vehicle equipment trackside box-type intelligent machine room, characterized in that: include: an acquisition module, configured to acquire first information, second information, and third information, wherein the first information includes design requirements of the intelligent computer room, the design requirements include performance targets of heat dissipation effect and humidity control, the second information includes meteorological data of the deployment location, and the third information includes historical computer room design solutions and corresponding performance data; A first optimization module, configured to perform parameter optimization processing according to the first information, and obtain fourth information by iteratively adjusting parameters and performing performance simulation processing, wherein the fourth information is a performance characteristic under a current design requirement; A first processing module, configured to perform deep learning processing on the fourth information and the second information based on a preset convolutional neural network mathematical model to obtain fifth information, wherein the fifth information includes at least one arrangement scheme of a heat dissipation system and a corresponding control strategy; A second optimization module, configured to perform multi-objective optimization processing according to the fifth information and the third information, and obtain sixth information by performing an optimization process under at least two performance indicators in combination with historical performance data; a conversion module, configured to perform a graph conversion process according to the sixth information to obtain a cooling system node graph, and adjust the cooling system node graph based on a preset graph neural network mathematical model to obtain eighth information, wherein the eighth information is an abstract graph including a cooling system structure, a control strategy, and a component relationship; a second processing module, configured to perform adaptive design processing according to the eighth information, and output a design result by combining the actual site layout, the relationship between the system structure and the performance requirements, wherein the design result includes all modules of the intelligent computer room, the configuration of the internal space and the corresponding heat dissipation control strategy; Wherein, the first processing module includes: Inputting the fourth information and the second information as input values ​​into a preset convolutional neural network for deep learning processing to construct a heat dissipation system layout prediction model; Performing prediction processing according to the heat dissipation system layout prediction model, and obtaining a prediction layout scheme by inputting parameters and structural characteristics of the heat dissipation system; The control strategy is optimized according to the predicted layout scheme and the preset deep deterministic policy gradient mathematical model, and the control strategy is obtained by modeling the state and environment inside the intelligent computer room as a continuous state space and taking the control parameters to be adjusted as continuous actions; Fifth information is obtained according to the predicted arrangement plan and the control strategy.

8. The design system of the trackside box-type intelligent machine room for 5T equipment in vehicles according to claim 7 is characterized in that: The first optimization module comprises: A first processing unit is used to obtain an initial design parameter set according to the performance targets of heat dissipation effect and humidity control in the first information through a multi-objective optimization genetic algorithm; A first simulation unit is used to perform fluid dynamics simulation processing according to the initial design parameter set, and obtain performance simulation results by solving a preset heat conduction-fluid flow mathematical model, wherein the performance simulation results include heat dissipation effect and humidity control performance under at least two parameter combinations; A first construction unit is used to perform data fitting and optimization processing using a corresponding surface method according to the performance simulation results, and obtain a function expression containing key performance characteristics by establishing a relationship model between heat dissipation effect, humidity control and design parameters; The first optimization unit is used to perform iterative optimization processing according to the function expression to obtain fourth information.

9. The design system of the trackside box-type intelligent machine room for 5T equipment in vehicles according to claim 8 is characterized in that: The first optimization unit comprises: A second construction unit is used to construct an objective function by minimizing the energy consumption of the heat dissipation system and maximizing the humidity control efficiency according to the function expression and the design requirements of the intelligent computer room; A first calculation unit, configured to calculate objective function gradient information according to the objective function, wherein the objective function gradient information is a rate of change of a performance characteristic in a design parameter space; The second calculation unit is used to perform iterative calculation according to the objective function, the objective function gradient information and a preset gradient descent mathematical model to obtain fourth information.

Images