Machine room modular design system and method and storage medium

Abstract

The invention discloses a machine room modular design system and method and a storage medium, and the system comprises a warehouse management module which is used for constructing a machine room material model and establishing a model coding rule to carry out the coding management of the machine room material model, and the machine room material model comprises a component model pipeline model and an excess material pipeline model; the algorithm module is used for executing a design algorithm and obtaining a machine room material model from the library management module according to design requirements to design an economic module of the machine room; and the learning optimization module learns the collected case model and optimizes the economic module design of the machine room to obtain the modular design of the machine room. According to the scheme, the complexity and risk of site construction can be reduced, meanwhile, a customized and extensible design scheme is formed, seamless joint of design and construction is achieved in response to the actual product component requirement of construction, and the machine room module design performance is evaluated while the machine room material use efficiency is improved.

Description

Technical Field

[0001] The present application relates to the application field of electromechanical engineering construction technology, and in particular to a modular design system, method and storage medium for a computer room. Background Art

[0002] With the continuous development of my country's economy and the rapid advancement of science and technology, super-high-rise and super-large buildings are rising rapidly. With the continuous improvement of the quality and progress control requirements of building mechanical and electrical installation, the innovation of mechanical and electrical installation technology has become an inevitable trend for the development and survival of the building mechanical and electrical installation industry. How to reduce the complexity and risks of on-site construction while forming customized and scalable design solutions to improve the flexibility and plasticity of computer room design IDE, and to achieve seamless connection between design and construction in response to the actual product component requirements of construction, improve the efficiency of computer room material use, evaluate the labor consumption of computer room installation and module design performance, directly link the accumulated cases for intelligent adjustment and optimization, improve the pertinence and practicality of the design solution, and further reduce material waste and labor costs, is a difficult problem that needs to be solved urgently. Summary of the invention

[0003] In response to the above problems, the purpose of this application is to provide a modular design system, method and storage medium for a computer room, which can realize the electromechanical system functions of the computer room by splicing multiple independent modules, reduce the complexity and risk of on-site construction, and form customized and scalable design solutions, thereby improving the flexibility and reusability of computer room design; respond to the actual product component requirements of construction to achieve seamless connection between design and construction, improve the efficiency of computer room material use, and evaluate the labor consumption of computer room installation and module design performance; directly link accumulated cases for intelligent adjustment and optimization, improve the pertinence and practicality of the design solution, and further reduce material waste and labor costs.

[0004] To achieve the above objectives, this application provides the following solutions:

[0005] In a first aspect, the present invention provides a computer room modular design system, comprising:

[0006] The warehouse management module is used to construct a computer room material model and establish a model coding rule to perform coding management on the computer room material model. The computer room material model includes: a component model, a pipeline model, and a surplus material pipeline model;

[0007] An algorithm module is used to execute a design algorithm to obtain a computer room material model from the warehouse management module according to design requirements for modular design;

[0008] The learning optimization module is used to learn the collected case models and optimize the economic module design of the computer room based on the learning results to obtain a modular design of the computer room.

[0009] Optionally, the warehouse management module includes:

[0010] A component library is used to process the acquired sample data to establish a data form, and to construct a component model consistent with the data in the data form;

[0011] A pipeline library, used for processing the acquired bill of materials to form a pipeline bill of materials, and building a corresponding pipeline model based on pipeline parameters of the pipeline bill of materials;

[0012] A residual material library is used to manage the pipeline models that do not meet the use requirements or the residual pipeline models of the remaining sections after the algorithm module completes the modular design;

[0013] The inventory model management unit is used to encode the models in the component library, the pipeline library and the residual material library to form a model code list.

[0014] Optionally, the algorithm module includes:

[0015] A standard module design unit, used for executing a standard connection algorithm to call the corresponding component model and pipeline model from the library management module to design a standard module design;

[0016] An economical module design unit, used for executing a surplus material replacement algorithm to identify a surplus material pipeline model corresponding to a pipeline model in the standard module design from the warehouse management module and replace the surplus material pipeline model to obtain an economical module design;

[0017] A module performance analysis unit, used for performing module performance analysis on the economic module design;

[0018] An assembly cost verification unit is used to verify the assembly cost of the economic module design that has passed the module performance analysis.

[0019] Optionally, the learning module includes:

[0020] A module design optimization unit, used for learning the collected module design case models to optimize the module design of the economic module design;

[0021] A module layout optimization unit, used to learn the collected layout design case models to optimize the layout design of the economic module design;

[0022] The module connection optimization unit is used to learn the collected connection design case models to optimize the pipeline connection of the economic module design.

[0023] In a second aspect, the present application provides a modular design method for a computer room, comprising:

[0024] Select corresponding component models and pipeline models based on received design requirements to carry out standard module design;

[0025] Identify and obtain a surplus pipeline model corresponding to the pipeline model in the standard module design to perform economic module design;

[0026] The economic module design is optimized based on machine learning to obtain a modular design of the computer room.

[0027] Optionally, the step of selecting corresponding component models and pipeline models for standard module design based on the received design requirements includes:

[0028] Analyze the design requirements and determine the corresponding computer room material model;

[0029] Based on the parameters of the computer room material model, the corresponding component model and pipeline model are called from the warehouse management module;

[0030] Executing a standard connection algorithm to arrange the component model and the pipeline model in a linear path;

[0031] Based on the linear path arrangement, the component models and pipeline models are arranged one by one according to specifications to obtain a standard module design.

[0032] Optionally, the step of identifying and acquiring the surplus material pipeline model corresponding to the pipeline model in the standard module design to perform economic module design includes:

[0033] Identifying selection parameters of the residual material pipeline model based on the model code of the pipeline model in the standard module design;

[0034] Based on the model code of the residual material pipeline model, identify and call the residual material pipeline model that meets the selected parameters;

[0035] Comparing the length of the residual material pipeline model with the length of the corresponding pipeline model in the standard module design;

[0036] The residual material pipeline model whose comparison result meets the comparison rules is replaced with the corresponding pipeline model to obtain an economical module design.

[0037] Optionally, the step of optimizing the economic module design based on machine learning to obtain a modular design of a computer room includes:

[0038] After learning the collected design case models, optimizing the module design of the economic module design;

[0039] After learning the collected layout design case models, the layout design of the economic module design is optimized;

[0040] After studying the collected connection design case models, the pipeline connection designed by the economic module is optimized.

[0041] Optionally, before the step of optimizing the economic module design based on machine learning to obtain a modular design of a computer room, the step further includes:

[0042] Conduct module performance analysis for economic module design;

[0043] Verify assembly cost of economic module design through module performance analysis;

[0044] When the assembly cost is verified, the economic module design is optimized to obtain a modular design of the computer room.

[0045] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-mentioned methods for modular design of a computer room.

[0046] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0047] The present application provides a modular design system, method and storage medium for a computer room, which realizes rapid modeling of complex electromechanical pipe fittings by utilizing BIM parametric software, realizes the electromechanical system functions of the computer room by splicing multiple independent modules, reduces the complexity and risk of on-site construction, forms a customized and scalable design scheme, and improves the flexibility and reusability of computer room design; responds to the actual product component requirements of construction to achieve seamless connection between design and construction, improves the material utilization efficiency of the computer room, and evaluates the labor consumption of computer room installation and module design performance; directly links accumulated cases for intelligent adjustment and optimization, improves the pertinence and practicality of the design scheme, and further reduces material waste and labor costs. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0049] Figure 1 A schematic diagram of the modular design system framework for a computer room provided in an embodiment of the present application;

[0050] Figure 2 A schematic diagram of the workflow of the modular design system for a computer room provided in an embodiment of the present application;

[0051] Figure 3The main component intention of the component library provided for the embodiment of the present application;

[0052] Figure 4 The main component intention of the component library provided for the embodiment of the present application;

[0053] Figure 5 A schematic diagram of a method for producing a residual material pipeline provided in an embodiment of the present application;

[0054] Figure 6 A schematic diagram of a module design library provided for an embodiment of the present application;

[0055] Figure 7 A schematic diagram of a modular design method for a computer room provided in an embodiment of the present application;

[0056] Figure 8 A modular design method for a computer room and a workflow diagram provided in an embodiment of the present application. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0058] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0059] refer to Figure 1 and Figure 2 As shown, the present application provides a modular design system for a computer room, comprising: a warehouse management module, an algorithm module and a learning optimization module, wherein the warehouse management module is used to construct models of various materials needed for the modular design of the computer room, such as component models, pipeline models, and surplus material pipeline models, and at the same time establish model coding rules, and set unique codes for the constructed component models, pipeline models, and surplus material pipeline models for management; wherein the algorithm module is used to execute the design algorithm to obtain the computer room material model from the warehouse management module according to the design requirements to design the economic module design of the computer room; the learning optimization module optimizes the economic module design of the computer room by learning the collected case models to obtain the modular design of the computer room.

[0060] Continue to refer Figure 1 As shown, in the embodiment of the present application, the warehouse management module includes: a component library, a pipeline library, a surplus material library and an inventory model management unit;

[0061] Among them, the component library needs to cover the sample range according to the needs, such as component types including common equipment in the computer room, pipe connectors, valves, equipment including water pumps, refrigeration units, heat exchangers, water processors, softening water devices, online cleaning devices and manifolds, etc.; such as pipe connectors including elbows, tees, crosses, reducers and flexible joints, etc., valves including butterfly valves, gate valves, filters, check valves, water flow indicators and stop valves, etc.; obtain detailed sample data from manufacturers, including drawings, technical specifications, performance parameters, etc. required for the establishment of the component library; classify and organize the collected data, divide them according to the properties such as component type, specification, material, etc., and establish corresponding data forms, and import the data forms into BIM software to establish various types of equipment, pipe connectors, and valves. During the modeling process, it is necessary to ensure that the properties such as the geometric shape, size, material, and performance information of the component are consistent with the sample data to form a component library, and thus complete the construction of the component library. Figure 3 , Figure 4 shown.

[0062] The pipeline library processes the bill of materials obtained from the material management department to form a pipeline material list, from which pipeline items for computer room construction are allocated. Pipeline items are usually the total length of various types of pipelines with different specifications. For example, 6m is used as 1 inventory to divide the finished product length of various types of pipelines according to specifications, and the part less than 6m is recorded as 1 inventory, thereby forming a pipeline material list containing pipeline types, specifications, and inventory quantities. Based on BIM software, an API endpoint is created by writing code, and by reading and calling the list data, pipeline models of various types and specifications with the same inventory quantity are created in the BIM software to form a pipeline library, thereby completing the construction of the pipeline library.

[0063] Among them, the surplus material library is used to manage the surplus material pipeline models of pipelines that do not meet the use requirements or the remaining sections after the design is completed. Specifically, the library management module provides the pipeline model of the pipeline library to the algorithm module for completing the standard module design. The algorithm module uses each 6m long pipeline module section by section during the standard module design process. When the remaining section is shorter than the use requirement, that is, it cannot meet the use requirement, or when there is still a surplus after the current standard module design is completed, the remaining section is returned to the library management module as a surplus material model and enters the surplus material library to form a surplus material library. In this embodiment, the surplus material library provides a surplus material pipeline model for the economic module design to replace the pipeline model intercepted by the standard pipeline provided by the pipeline library, thereby reducing the use of the standard pipeline model. The surplus material generation method is as follows: Figure 5 As shown:

[0064] The inventory model management unit is used to encode the models in the component library, the pipeline library and the surplus material library to form a model code list, wherein each model has a unique code. The model is coded and identified, entered and formed into a model code management list, so as to facilitate subsequent model management and retrieval. During the management process, if the model is called out of the warehouse management module, the warehouse management module removes the code corresponding to the called out model from the model code management list. If the surplus material is returned to the warehouse management module, the warehouse management module encodes the surplus material and enters it into the model code management list.

[0065] In the embodiment of the present application, the model coding structure is "KX.YZ001", where: "K" represents the table code, indicating that the code belongs to the model library. "X" represents the major category code, indicating the library classification to which the model belongs. There are three categories in total: GJ, GD, and YL. GJ corresponds to the component library, GD corresponds to the pipeline library, and YL corresponds to the residual material library. "Y" represents the medium category code, indicating the type to which the model belongs. There are five categories in total: SB, GJ, FM, GD, and YL. SB corresponds to equipment, GJ corresponds to pipe fittings, FM corresponds to valves, GD corresponds to pipelines, and YL corresponds to residual materials. "Z" represents the sub-category code, indicating the specifications of this type of model. There are eight categories: 50, 65, 100, 125, 150, 200, 250, and 300. 50 corresponds to diameter DN50, 65 corresponds to diameter DN65, 100 corresponds to diameter DN100, 125 corresponds to diameter DN125, 150 corresponds to diameter DN150, 200 corresponds to diameter DN200, 250 corresponds to diameter DN250, and 300 corresponds to diameter DN300. "001" represents the sub-category code, involving three categories: equipment, namely SB, pipe fittings, namely GJ, and valves, namely FM, indicating different individual models under this type of model specifications, starting from 001 and ending at 999; involving two categories: pipes, namely GD, and residual materials, namely YL, indicating different length models under this type of model specifications, counting in cm, ending at 001 to 600.

[0066] For example, the model coding example: the model code of the first water pump with an interface of DN200 in the component library in the library management module is K-GJ.SB.200.001. The pipe model with a diameter of DN100 (length 6m or 600cm) in the pipeline library in the library management module is K-GD.GD.100.600. The model code of the residual material module with a diameter of DN50 and a length of 320cm in the residual material module in the library management module is K-YL.YL.50.320. The model coding rules are shown in Table 1:

[0067]

[0068]

[0069] Table 1

[0070] Continue to refer Figure 1 As shown, the algorithm module includes: a standard module design unit, an economic module design unit, a module performance analysis unit and an assembly cost verification unit.

[0071] The standard module design unit is used to execute the standard connection algorithm to call the corresponding construction model and pipeline model from the library management module to design the standard module; the standard connection algorithm rules are as follows:

[0072] First, select the appropriate equipment model according to the functional requirements of the module design, determine the model and specifications of the equipment model based on the required functional parameters such as water flow, head, pressure, etc., and call the model from the warehouse management module based on the model code formed by the equipment, model and specifications. When calling, refer to the model code management list described in step 4 to call out the corresponding model.

[0073] Then, the reserved connection interface positions of the module are set to the two ends of the top of the module, and the set interface positions and the equipment interface positions are connected with a linear path, so that the valve model, pipe fitting model and pipeline model are arranged along the linear path.

[0074] Finally, according to the location and specifications of the inlet and outlet interfaces of the selected equipment model, the valve model and pipe fitting model are called one by one in accordance with the conventional design logic sequence. When calling, the corresponding model is called out according to the model code management list described in step 4 according to the model code. The valve model and pipe fitting model are arranged one by one according to the minimum spacing allowed by the specification.

[0075] The pipeline model is called according to the inlet and outlet specifications of the selected equipment model. When calling, the corresponding model is called out according to the model code control step 4 described in the model code management list. When arranging the pipeline model, the spacing between the valve model and the pipe fitting model is followed, and the layout is intercepted and arranged section by section from the pipeline model. At the same time, the corresponding code of the pipeline model is deleted from the model code management list.

[0076] The length of the pipeline model is 6m. When the remaining section during the interception process is shorter than the distance between the valve model and the pipe fitting model, the remaining section is returned to the residual material library of the warehouse management module as a residual material model as described in step 3, and the coding is marked and entered into the model coding management list as described in step 4. At the same time, a new pipeline model is called from the pipeline library of the warehouse management module according to the model code. When the standard module design is completed, there is still a surplus of the currently intercepted pipeline model. As described in step 3, the remaining section is returned to the residual material library of the warehouse management module as a residual material model, and the coding is marked and entered into the model coding management list as described in step 4. The standard module design is now completed.

[0077] The economic module design unit is used to execute the surplus material replacement algorithm to identify the surplus material pipeline model corresponding to the pipeline model in the standard module design from the warehouse management module and replace it to design the economic module; wherein the rules of the surplus material replacement algorithm are as follows:

[0078] First, the model code of the pipeline model called during the current module standard design is compared to identify the selected specifications of the surplus pipeline model, the length of the pipeline model cut section by section during the current standard module design is retrieved, and the selected length of the surplus model is identified.

[0079] Return the surplus material model code identification of the model code management list entered into the surplus material warehouse through the surplus material model, and verify the surplus material model codes that meet the surplus material model selection specifications one by one. The verification content is whether the small category code in the coding rule, that is, whether the specifications to which the surplus material model belongs are met. The surplus material models that do not meet the surplus material model selection specifications will continue to be retained in the surplus material warehouse for use, and the surplus material models that meet the surplus material model selection specifications will be transferred out of the surplus material warehouse for use.

[0080] Finally, the spare material model and the module standard are compared and called out to be used, and the length of the pipe model cut into sections is designed in rotation. The spare material model that meets the comparison rules is used to replace the pipe segment model to form the module. The replaced pipe segment model returns to the pipe model segment cutting process, and the corresponding pipe model cutting step is skipped.

[0081] The comparison process is as follows: compare the length of the first residual stock model with the first pipe model cut-off length, then compare the length of the first residual stock model with the second pipe model cut-off length, until the first residual stock model length completes the comparison of all pipe model lengths, then rotate the second residual stock model length to compare with the first pipe model cut-off length, until all pipe model cut-off lengths are compared. During the process, (1) if the length of the stand-by residual stock model is the same as the pipe model cut-off length, it is directly replaced, and the corresponding code of the residual stock model is deleted from the model code management list, and the current pipe model cut-off length exits the comparison; (2) if If the length of the unused stock model is less than the cut-off length of all pipeline models, the current unused stock model is returned to the stock library of the warehouse management module, and the model coding management list is not operated; (3) If the length of the unused stock model is longer than the cut-off length of the pipeline model, and the length of the extended part is less than 10 cm, the cut-off length of the pipeline model is cut from the stock model to replace the pipeline, and the corresponding code of the stock model is deleted from the model coding management list. At the same time, the remaining section of the stock model is returned to the stock library of the warehouse management module as the stock model, and the coding is marked and entered into the model coding management list as described in step 4. The current pipe model cut-off length exits the comparison; (4) If the unused stock model is longer than the cut-off length of the pipeline model, and the length of the extended part is less than 10 cm, the cut-off length of the pipeline model is cut from the stock model to replace the pipeline, and the corresponding code of the stock model is deleted from the model coding management list. At the same time, the remaining section of the stock model is returned to the stock library of the warehouse management module as the stock model, and the coding is marked and entered into the model coding management list as described in step 4. The current pipe model cut-off length exits the comparison; If the length of the stock model is longer than the cut-off length of the pipeline model, and the length of the extended part is greater than or equal to 10 cm and less than or equal to 20 cm, then the current length of the stock model is compared with the cut-off lengths of other pipeline models, and the cut-off length of the pipeline model closest to the current stock model length is selected. The cut-off length of the pipeline model is cut from the stock model to replace the pipeline, and the corresponding code of the stock model is deleted from the model coding management list. At the same time, the remaining section of the stock model is returned to the stock library of the library management module as the stock model, and the coding is marked and entered into the model coding management list as described in step 4. The current pipe model cut-off length exits the comparison; (5) If the unused stock model If the length is longer than the cut-off length of the pipeline model, and the length of the protruding part is greater than 20 cm, the length of the current residual material model is compared with the cut-off length of other pipeline models. The execution priority of the comparison process is as follows: if the comparison result shows that the length of the residual material model to be used is the same as the cut-off length of the pipeline model, execute (1); if the comparison result shows that the length of the protruding part is less than 10 cm, execute (3); if the comparison result shows that the length of the protruding part is greater than or equal to 10 cm and less than or equal to 20 cm, execute (4); if the comparison result shows that the protruding part is greater than 20 cm, the current residual material model to be used is returned to the residual material library of the library management module, and the model coding management list is not operated.

[0082] The module performance analysis unit is used to perform module performance analysis on the economic module. When performing module performance analysis, scientific simulation verification is adopted, and the formula preset by the algorithm module is automatically executed and the verification result is output. The preset formula is as follows:

[0083]

[0084] Where z is the module flow resistance loss, in meters. y is the module pipeline resistance loss, in meters. j is the resistance loss of the module components, in meters. f is the pipeline resistance coefficient, which is taken as 0.03 during simulation verification. 001 is the detailed category code of the model coding as described in step 4, that is, the length, ∑001 is the sum of the detailed category codes, that is, the sum of the lengths, in cm. v is the liquid flow velocity in the module, and its value is 2.0m / s. Z is the subcategory code of the model coding as described in step 4, that is, the specification, in mm. Y is the middle category code of the model coding as described in step 4, that is, the type, ∑Y is the sum of the corresponding values ​​of the middle category codes, that is, the sum of the corresponding values ​​of the types, and the corresponding values ​​are: SB, that is, the equipment value is 32, GJ, that is, the pipe fittings value is 1.5, and FM, that is, the valve value is 9.

[0085] Among them, when conducting scientific simulation verification, according to the model codes corresponding to the component models, pipeline models and surplus stock models contained in the current economic module design, the information contained in the model codes is extracted, such as: the sub-category code of the pipeline model and the surplus stock model, that is, the length, and the total length of the pipeline model and the surplus stock model used by the module is obtained by adding the sub-category codes; the sub-category code of the pipeline model and the surplus stock model, that is, the specification, and the diameter of the pipeline model and the surplus stock model is obtained by the sub-category code; the middle class code corresponding to the equipment, pipe fittings, and valves, that is, the type

[0086] The sum of the values ​​corresponding to the middle class code is obtained by adding the corresponding values ​​one by one. The module flow resistance loss is the sum of the module pipeline resistance loss and the module component resistance loss. For example, the flow resistance loss of the economic module is set to be no greater than 100. If it is greater than 100, the algorithm module will alarm. If it is less than or equal to 100, the analysis is passed, and the module performance analysis is completed.

[0087] The assembly cost verification unit is used to verify the assembly cost of the economic module that has passed the module performance analysis. In the embodiment of the present application, the assembly cost verification adopts the method of embedding manual quotas, that is, embedding the preset manual quotas into the algorithm module to verify the module assembly cost. The details are as follows:

[0088] First, the model codes of the component model, pipeline model and surplus material model of the current module are retrieved, and the information contained in the model code is extracted. The extracted information includes: obtaining the number of equipment, pipe fittings and valves through the detailed code; obtaining the length and number of pipes and surplus materials through the number of codes corresponding to the detailed code and the medium code; obtaining the number of assembly joints from the total number of equipment, pipe fittings, valves, pipes and surplus materials. Then the labor quota rules are input into the algorithm module. The labor quota rules are as follows: installer cost: 400 yuan / person*day, welder cost: 750 yuan / person*day, equipment assembly efficiency: 4 people*0.5 days / unit, pipe fitting assembly efficiency: 1 person*0.1 day / piece, valve assembly efficiency: 2 people*0.2 days / piece, pipe assembly efficiency: 2 people*1 day / 120 meters, surplus material assembly efficiency: 2 people*1 day / 120 meters, joint connection efficiency: 1 person*0.4 day / piece, equipment, pipe fittings, valves, pipes and surplus materials are assembled by installers, and joints are connected by welders. Finally, the module assembly cost calculation formula is established: module assembly cost = equipment assembly cost + pipe assembly cost + valve assembly cost + pipeline assembly cost + surplus material assembly cost + joint connection cost, among which equipment assembly cost = installer cost * equipment assembly efficiency * number of equipment, pipe assembly cost = installer cost * pipe assembly efficiency * number of pipes, valve assembly cost = installer cost * valve assembly efficiency * number of valves, pipeline assembly cost = installer cost * pipeline assembly efficiency * pipeline length, surplus material assembly cost = installer cost * surplus material assembly efficiency * surplus material length, joint connection cost = welding cost * joint connection efficiency * number of joints. For example, the assembly cost of the economic module is set to not be greater than 160,000. If it is greater than 160,000, the algorithm module will alarm, and if it is less than or equal to 160,000, the analysis will pass, and the assembly cost verification is completed.

[0089] Continue to refer Figure 1 As shown, the learning module includes: a module design optimization unit, a module layout optimization unit and a module connection optimization unit.

[0090] In the example of this application, in order to optimize the economic module to obtain the final modular design of the computer room, this application collects module design case models from multiple projects and multiple fields through various methods such as pre-self-construction, exchange, and purchase, collects various types of modular layout design case models of computer rooms through the company's accumulated project documents, and collects various types of connection design case models between computer room modules through the company's accumulated project documents, etc., further uses machine learning to learn the acquired module design case model, layout design case model, and connection design case model, and performs targeted processing on the economic module based on the learning results.

[0091] In this embodiment, the module design optimization unit learns the collected module design case models to optimize the module design of the economic module design. For example, the collected cases are classified and sorted according to the functional modules to form a module design library, and the module design library is connected to the machine learning. By calling the model coding rules of the library management module, the obtained module component models are encoded and included in the model coding management list. At the same time, the machine learning ensures that the module models contained in the electromechanical module design library meet the requirements by calling the module performance analysis and assembly cost verification of the algorithm module.

[0092] Furthermore, machine learning forms a table database by reading the information contained in the module model. The table data includes: the classification, type, specification and number of each component model of the module extracted from the model coding of each component component of the economic module, the position coordinates of each component component model of the module, and the starting point coordinates and end point coordinates of each component pipeline model and residual material model of the module.

[0093] Perform unique hot encoding on the category labels corresponding to the module functions to obtain the feature vector G , input into the Embedding layer, and get the label information embedded in W (1×N dimension). Construct three fully connected layers FC with independent parameters for the component model, pipeline model, and residual model respectively. i (N×M i Dimension), where M i is the number of classifications under the corresponding model type. The label information is embedded into the fully connected layer FC i (Calculation, calculation result R i The normalization process is completed through the Sigmod operation to obtain the category probability P i , compare the class probabilities P i Each element p ij With threshold T i The size of the output is greater than or equal to the corresponding threshold T i The predicted label P′ i , complete the selection of model categories in component models, pipeline models, and residual material models.

[0094] R i =FC i (Embedding(G))

[0095] P i =Sigmod(R i ), P i =[p i1 , p i2 , ..., p iMi ]

[0096] P′ i =[p′ i1 , p′i2 , ..., p′ iMi ]

[0097]

[0098] After the predicted label is one-hot encoded, it is concatenated with the label information embedding W to obtain the category-label encoding W′, which is input into several fully connected layers. The fully connected layers are connected using the ReLU() activation function to obtain the predicted values ​​of the specifications, number, coordinates and other parameters of the corresponding model type. The model selects the mean square error MSE() as the training loss function L, and sets the learning bias vector α, where the value for the component model is 0.6, the value for the pipeline model is 0.3, and the value for the surplus model is 0.1. The overall loss function L selected by the model 整体 =0.6*L 构件 +0.3*L 管道 +0.1*L 余料 For the prediction results, the cost verification algorithm in step 8 is called to obtain the assembly cost C under the model prediction assembly solution. The algorithm also optimizes L 整体 Together with the assembly cost C, a module combination solution with the lowest assembly cost is formed.

[0099] Machine learning also uses modular design libraries (such as Figure 6 As shown in the figure, new data is added to iteratively optimize the module composition design, and the module design optimization is completed.

[0100] In this embodiment, the collected layout design case models are learned by the module layout optimization unit to optimize the layout design of the economic module design to achieve maximum space utilization and maximum equipment maintenance convenience. The module layout optimization unit extracts features that affect the optimization target from the case model, such as the size of the machine room, the type and number of module functions adopted, the location and orientation of the module layout, the width and location of the reserved maintenance channel, etc., converts the influencing features into a table database, assigns a space utilization weight value of 0.7 and a equipment maintenance convenience weight value of 0.3, and reads the module function type contained in the above module design library. Based on the optimization target, a genetic algorithm model is adopted. The table database is randomly generated with real number coding to establish a basic layout plan group. Then, the shuffle crossover and continuous mutation methods are adopted. According to the optimization targets and corresponding weight values ​​of the maximum space utilization E and the maximum equipment maintenance convenience M, each layout plan in the layout plan group is evaluated to obtain the fitness value of each plan. The higher the fitness, the better the plan. Among them, the maximum space utilization E is the ratio of the area S corresponding to the effective activity space of the computer room to the total area A of the computer room. The maximum equipment maintenance convenience M is reflected in the fact that the sum of the distances D from each entrance p_i of the computer room to each equipment maintenance area q_j is the shortest, which is related to the module function type and quantity, the layout position and orientation of each module, and the width and position of the reserved maintenance channel. The width and position of the reserved maintenance channel must meet the maintenance activities of maintenance personnel.

[0101]

[0102] M:min(∑dis(p i ,q j ))

[0103] The constraints set include:

[0104] (1) The module layout is mainly constrained by the space of the computer room. It cannot exceed the boundaries of the computer room, and must meet the positional relationship between adjacent modules that have distance requirements.

[0105] (2) Modules in the computer room do not interfere with each other and cannot be intertwined.

[0106] (3) Ensure that each module does not affect the entry and exit of maintenance personnel.

[0107] When the optimal individual fitness or group fitness no longer increases, the model iteration is stopped. Finally, the machine learning selects the module layout optimization plan based on the fitness. At the same time, the machine learning also iteratively optimizes the module layout position by adding new project documents, thus completing the module layout optimization.

[0108] In an embodiment of the present application, the collected connection design case models are learned by the module connection optimization unit to optimize the pipeline connection of the economic module design. In this embodiment, the module connection optimization unit analyzes and identifies the pipeline routing selection principle and pipeline bend setting method in the connection design case model, and determines that the goal of module connection optimization is the shortest pipeline length and the least pipeline bends. The learning module extracts pipeline layout features that affect the optimization target from the engineering case, and the pipeline layout features include pipeline straight path, pipeline length, number of pipeline bends, etc.

[0109] First, we construct an N×N dimensional module connectivity matrix A, where N is the number of modules installed in the computer room and the matrix element a is ij A value of 0 indicates that module i and module j are not connected, and a value of 1 indicates that they need to be connected. Construct an N×M-dimensional module information matrix B, where N is the number of modules installed in the room, and M is the number of features of the module. The features include the three-dimensional spatial coordinates of the module layout, the module function type, etc. The matrix element b ij is the actual value of the corresponding feature of the module. The module connectivity matrix and the module information matrix are concat spliced ​​to form an N×L-dimensional input matrix, where L=N+M. The input matrix is ​​calculated through two fully connected layers. The first fully connected layer FC1 is L×N-dimensional, and the second fully connected layer FC2 is N×N-dimensional. The calculation results are processed by the ReLU() function and rounded down to obtain the prediction matrix R of the number of bends in the connecting pipelines between each module. The element r ij is the predicted number of bends between device i and device j.

[0110]

[0111]

[0112]

[0113] Among them, for a pair of modules, according to the predicted value r of the number of bends in the connecting pipe between the pair of modules ij And the module three-dimensional space information, input several layers of fully connected layers, output r ijThe three-dimensional space coordinates of the bends of each pipeline are predicted, and the three-dimensional space distances between the starting point, the bends and the end point between each device are calculated by summing up to obtain the total length of the pipeline route. The model selects the mean square error MSE() as the loss function, compares the actual value of the pipeline routing in the existing design case with the model prediction value, and finally completes the model training. The preset pipeline cost and pipeline bending cost are used as constants, and the total cost is used to reflect the effect of the solution generated by the learning module, which is used to consider the module connection optimization effect and provide reference data for the module connection implementation. At the same time, the learning module has the ability to iteratively optimize the module connection route by adding new project documents, thus completing the module connection optimization.

[0114] refer to Figure 7 and Figure 8 As shown, the present application provides a modular design method for a computer room, including:

[0115] Based on the received module design functional requirements, the corresponding component model and pipeline model are selected to perform standard module design. In this embodiment, the standard module design is performed according to the following steps:

[0116] Analyze the functional requirements of the module design to determine the corresponding equipment model;

[0117] Based on the parameters of the equipment model, the corresponding component model and pipeline model are called from the library management module;

[0118] Executing a standard connection algorithm to arrange the component model and the pipeline model in a linear path;

[0119] Based on the linear path arrangement, the component models and pipeline models are arranged one by one according to specifications to obtain a standard module design.

[0120] In an embodiment of the present application, after receiving the design function requirements, the system determines the model and specifications of the equipment model based on functional parameters such as water flow, head, pressure, etc. Further, according to the model code formed by the equipment name, model and specifications, the corresponding model is called from the library management module to perform standard module design.

[0121] In this embodiment, by setting the module reserved connection interface position to the two ends of the module top, the set interface position and the device interface position are connected by a linear path, so that the valve model, pipe fitting model and pipeline model are arranged along the linear path. Further, according to the selected equipment model inlet and outlet interface positions and specifications, the valve model and pipe fitting model are called one by one in accordance with the conventional design logic sequence. When calling, the corresponding model is called out according to the model code and the model code management list. When the valve model and pipe fitting model are arranged, they are arranged one by one according to the minimum spacing allowed by the specification. Further, according to the selected equipment model inlet and outlet specifications, the pipeline model is called. When calling, the corresponding model is called out according to the model code and the model code management list. When the pipeline model is arranged, the spacing between the valve model and the pipe fitting model is followed. The arrangement is intercepted section by section from the pipeline model, and the corresponding code of the pipeline model is deleted from the model code management list.

[0122] The length of the pipeline model is 6m. When the remaining section during the interception process is shorter than the distance between the valve model and the pipe fitting model, the remaining section is returned to the residual material library of the warehouse management module as a residual material model as described in step 3, and the remaining section is coded and identified and entered into the model coding management list. At the same time, a new pipeline model is called from the pipeline library of the warehouse management module according to the model code. When the standard module design is completed, there is still a surplus of the currently intercepted pipeline model. As described in step 3, the remaining section is returned to the residual material library of the warehouse management module as a residual material model, and the coding is identified and entered into the model coding management list as described in step 4, and the standard module design is completed.

[0123] According to the preset rules, the residual material pipeline model corresponding to the pipeline model in the standard module design is identified to perform economic module design; when performing the economic module design specifically, the following steps are performed:

[0124] Identifying the selection parameters of the residual material pipeline model based on the coding of the pipeline model in the standard module design;

[0125] Identify and call a residual material pipeline model that meets the selected parameters based on the residual material pipeline model code;

[0126] Comparing the length of the residual material pipeline model with the length of the corresponding pipeline model in the standard module design;

[0127] The residual material pipeline model whose comparison result meets the comparison rules is replaced with the corresponding pipeline model to obtain an economical module design.

[0128] In an embodiment of the present application, firstly, by comparing the model code of the pipeline model in the current standard module design, the selected specifications in the surplus pipeline model are identified, the length of the pipeline model is cut section by section during the standard module design, and the selected length of the surplus model is identified.

[0129] Return the surplus material model code identification of the model code management list entered into the surplus material warehouse through the surplus material model, and verify the surplus material model codes that meet the surplus material model selection specifications one by one. The verification content is whether the small category code in the coding rule, that is, whether the specifications to which the surplus material model belongs are met. The surplus material models that do not meet the surplus material model selection specifications will continue to be retained in the surplus material warehouse for use, and the surplus material models that meet the surplus material model selection specifications will be transferred out of the surplus material warehouse for use.

[0130] Finally, the spare material model and the length of the module standard designed pipeline model are called out for rotation comparison. The spare material model that meets the comparison rules is used to replace the pipe segment model to form the module. The replaced pipe segment model returns to the pipeline model segment cutting process, skipping the corresponding pipeline model segment cutting.

[0131] The comparison process is: compare the length of the first residual stock model with the cut length of the first pipeline model, then compare the length of the first residual stock model with the cut length of the second pipeline model, until the length of the first residual stock model completes the comparison of all pipeline models, then the length of the second residual stock model is rotated to compare with the cut length of the first pipeline model, until the comparison of the cut length of all pipeline models is completed.

[0132] (1) If the length of the spare stock model to be used is the same as the cut-off length of the pipeline model, it is directly replaced, and the corresponding code of the spare stock model is deleted from the model code management list, and the cut-off length of the current pipeline model is removed from the comparison;

[0133] (2) If the length of the unused residual material model is less than the intercepted length of all pipeline models, the current unused residual material model is returned to the residual material library of the library management module, and the model code management list is not operated;

[0134] (3) If the length of the spare stock model to be used is longer than the cut-off length of the pipeline model, and the length of the extended part is less than 10 cm, the cut-off length of the pipeline model is cut off from the spare stock model to replace the pipeline, and the corresponding code of the spare stock model is deleted from the model code management list. At the same time, the remaining section of the spare stock model is returned to the spare stock library of the library management module as the spare stock model. The spare stock pipeline model is coded and identified and entered into the model code management list, and the current cut-off length of the pipeline model exits the comparison;

[0135] (4) If the length of the spare stock model to be used is longer than the cut-off length of the pipeline model, and the length of the extended part is greater than or equal to 10 cm and less than or equal to 20 cm, then the current spare stock model length is compared with the cut-off lengths of other pipeline models, and the cut-off length of the pipeline model closest to the current spare stock model length is selected. The cut-off length of the pipeline model is cut from the spare stock model to replace the pipeline, and the corresponding code of the spare stock model is deleted from the model code management list. At the same time, the remaining section of the spare stock model is returned to the spare stock library of the library management module as the spare stock model, and the spare stock pipeline model is coded and identified and entered into the model code management list. The current pipeline model cut-off length exits the comparison;

[0136] (5) If the length of the spare material model to be used is longer than the cut-off length of the pipeline model, and the length of the protruding part is greater than 20 cm, then the length of the current spare material model is compared with the cut-off lengths of other pipeline models. The execution priority of the comparison process is as follows: if the comparison result shows that the length of the spare material model to be used is the same as the cut-off length of the pipeline model, execute (1); if the comparison result shows that the length of the protruding part is less than 10 cm, execute (3); if the comparison result shows that the length of the protruding part is greater than or equal to 10 cm and less than or equal to 20 cm, execute (4); if the comparison result shows that the protruding part is greater than 20 cm, the current spare material model to be used is returned to the spare material library of the library management module, and the model coding management list is not operated.

[0137] The economic module design is optimized based on machine learning to obtain a modular design of the computer room.

[0138] In this embodiment, in order to optimize the economic module to obtain the final modular design of the computer room, this application collects module design case models from multiple projects and multiple fields through various methods such as pre-built, exchanged, and purchased, collects various modular layout design case models of computer rooms through the company's accumulated project documents, and collects various connection design case models between computer room modules through the company's accumulated project documents, etc., and further uses machine learning to learn the acquired module design case models, layout design case models, and connection design case models, and optimizes the economic module according to the learning results. Specifically,

[0139] Optimizing the module design of the economic module design by learning the collected design case models;

[0140] Optimizing the layout design of the economic module design by learning the collected layout design case models;

[0141] The pipeline connection designed by the economic module is optimized after learning the collected connection design case models.

[0142] Before the step of optimizing the economic module design based on machine learning to obtain a modular design of a computer room, the step further includes:

[0143] Perform module performance analysis on economic design modules based on scientific simulation verification;

[0144] Verify assembly cost of economic module design through module performance analysis;

[0145] When the assembly cost is verified, the economic module design is optimized to obtain a modular design of the computer room.

[0146] In the embodiment of the present application, when performing module performance analysis, scientific simulation verification is adopted, and the formula preset by the algorithm module is automatically executed and the verification result is output, wherein the preset formula is as follows:

[0147]

[0148] Where z is the module flow resistance loss, in meters. y is the module pipeline resistance loss, in meters. j is the resistance loss of the module components, in meters. f is the pipeline resistance coefficient, which is taken as 0.03 during simulation verification. 001 is the detailed category code of the model coding as described in step 4, that is, the length, ∑001 is the sum of the detailed category codes, that is, the sum of the lengths, in cm. v is the liquid flow velocity in the module, and its value is 2.0m / s. Z is the subcategory code of the model coding as described in step 4, that is, the specification, in mm. Y is the middle category code of the model coding as described in step 4, that is, the type, ∑Y is the sum of the corresponding values ​​of the middle category codes, that is, the sum of the corresponding values ​​of the types, and the corresponding values ​​are: SB, that is, the equipment value is 32, GJ, that is, the pipe fittings value is 1.5, and FM, that is, the valve value is 9.

[0149] Among them, when conducting scientific simulation verification, according to the model codes corresponding to the component models, pipeline models and surplus stock models contained in the current economic module design, the information contained in the model codes is extracted, such as: the sub-category code of the pipeline model and the surplus stock model, that is, the length, and the total length of the pipeline model and the surplus stock model used by the module is obtained by adding the sub-category codes; the sub-category code of the pipeline model and the surplus stock model, that is, the specification, and the diameter of the pipeline model and the surplus stock model is obtained by the sub-category code; the middle class code corresponding to the equipment, pipe fittings, and valves, that is, the type

[0150] The sum of the values ​​corresponding to the middle class code is obtained by adding the corresponding values ​​one by one. The module flow resistance loss is the sum of the module pipeline resistance loss and the module component resistance loss. For example, the flow resistance loss of the economic module is set to be no greater than 100. If it is greater than 100, the algorithm module will alarm. If it is less than or equal to 100, the analysis is passed, and the module performance analysis is completed.

[0151] When the assembly cost of the economic module that has passed the module performance analysis is verified, in the embodiment of the present application, the assembly cost verification adopts the method of embedding manual quotas, that is, embedding the preset manual quotas into the algorithm module to verify the module assembly cost. The details are as follows:

[0152] First, the model codes of the component model, pipeline model and surplus material model of the current module are retrieved, and the information contained in the model code is extracted. The extracted information includes: obtaining the number of equipment, pipe fittings and valves through the detailed code; obtaining the length and number of pipes and surplus materials through the number of codes corresponding to the detailed code and the medium code; obtaining the number of assembly joints from the total number of equipment, pipe fittings, valves, pipes and surplus materials. Then the labor quota rules are input into the algorithm module. The labor quota rules are as follows: installer cost: 400 yuan / person*day, welder cost: 750 yuan / person*day, equipment assembly efficiency: 4 people*0.5 days / unit, pipe fitting assembly efficiency: 1 person*0.1 day / piece, valve assembly efficiency: 2 people*0.2 days / piece, pipe assembly efficiency: 2 people*1 day / 120 meters, surplus material assembly efficiency: 2 people*1 day / 120 meters, joint connection efficiency: 1 person*0.4 day / piece, equipment, pipe fittings, valves, pipes and surplus materials are assembled by installers, and joints are connected by welders. Finally, the module assembly cost calculation formula is established: module assembly cost = equipment assembly cost + pipe assembly cost + valve assembly cost + pipeline assembly cost + surplus material assembly cost + joint connection cost, where equipment assembly cost = installer cost * equipment assembly efficiency * number of equipment, pipe assembly cost = installer cost * pipe assembly efficiency * number of pipes, valve assembly cost = installer cost * valve assembly efficiency * number of valves, pipeline assembly cost = installer cost * pipeline assembly efficiency * pipeline length, surplus material assembly cost = installer cost * surplus material assembly efficiency * surplus material length, joint connection cost = welding cost * joint connection efficiency * number of joints. For example, the assembly cost of the economic module is set to be no greater than 160,000. If it is greater than 160,000, the algorithm module will alarm, and if it is less than or equal to 160,000, the analysis will be passed. At this point, the assembly cost verification is completed, and the economic module design of the computer room is finally obtained.

[0153] To summarize, the present application scheme realizes the electromechanical system functions of the computer room by splicing multiple independent modules, reduces the complexity and risks of on-site construction, forms customized and scalable design schemes, and improves the flexibility and reusability of computer room design; responds to the actual product component needs of construction to achieve seamless connection between design and construction, improves the efficiency of computer room material use, and evaluates the labor consumption of computer room installation and module design performance; directly links accumulated cases for intelligent adjustment and optimization, improves the pertinence and practicality of the design scheme, and further reduces material waste and labor costs.

[0154] The present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of any one of the above-mentioned methods for modular design of a computer room are implemented.

[0155] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0156] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0157] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0158] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0159] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0160] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0161] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0162] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A modular design system for a computer room, characterized in that: The modular design system of the computer room includes: The warehouse management module is used to construct a computer room material model and establish a model coding rule to perform coding management on the computer room material model. The computer room material model includes: a component model, a pipeline model, and a surplus material pipeline model; An algorithm module is used to execute a design algorithm to obtain a computer room material model from the warehouse management module according to design requirements for modular design; The learning optimization module is used to learn the collected case models and optimize the economic module design of the computer room based on the learning results to obtain a modular design of the computer room.

2. The modular design system for a computer room according to claim 1, characterized in that: The warehouse management module includes: A component library is used to process the acquired sample data to establish a data form, and to construct a component model consistent with the data in the data form; A pipeline library, used for processing the acquired bill of materials to form a pipeline bill of materials, and building a corresponding pipeline model based on pipeline parameters of the pipeline bill of materials; A surplus material library is used to manage the pipeline models that do not meet the use requirements or the surplus pipeline models of the remaining sections after the algorithm module completes the modular design of the machine room; The inventory model management unit is used to encode the models in the component library, the pipeline library and the residual material library to form a model code list.

3. The modular design system for a computer room according to claim 1, characterized in that: The algorithm module includes: A standard module design unit, used for executing a standard connection algorithm to call the corresponding component model and pipeline model from the library management module to design a standard module design; An economical module design unit, used for executing a surplus material replacement algorithm to identify a surplus material pipeline model corresponding to a pipeline model in the standard module design from the warehouse management module and replace the surplus material pipeline model to obtain an economical module design; A module performance analysis unit, used for performing module performance analysis on the economic module design; An assembly cost verification unit is used to verify the assembly cost of the economic module design that has passed the module performance analysis.

4. The modular design system for a computer room according to claim 1, characterized in that: The learning modules include: A module design optimization unit, used for learning the collected module design case models to optimize the module design of the economic module design; A module layout optimization unit, used to learn the collected layout design case models to optimize the layout design of the economic module design; The module connection optimization unit is used to learn the collected connection design case models to optimize the pipeline connection of the economic module design.

5. A modular design method for a computer room, characterized in that: The modular design method for a computer room includes: Select corresponding component models and pipeline models based on received design requirements to carry out standard module design; Identify and obtain a surplus pipeline model corresponding to the pipeline model in the standard module design to perform economic module design; The economic module design is optimized based on machine learning to obtain a modular design of the computer room.

6. The modular design method for a computer room according to claim 5, characterized in that: The step of selecting corresponding component models and pipeline models for standard module design based on the received design function requirements includes: Analyze the design requirements and determine the corresponding computer room material model; Based on the parameters of the computer room material model, the corresponding component model and pipeline model are called from the warehouse management module; Executing a standard connection algorithm to arrange the component model and the pipeline model in a linear path; Based on the linear path arrangement, the component models and pipeline models are arranged one by one according to specifications to obtain a standard module design.

7. The modular design method for a computer room according to claim 5, characterized in that: The step of identifying and obtaining the surplus material pipeline model corresponding to the pipeline model in the standard module design to perform economic module design includes: Identifying selection parameters of the residual material pipeline model based on the model code of the pipeline model in the standard module design; Based on the model code of the residual material pipeline model, identify and call the residual material pipeline model that meets the selected parameters; Comparing the length of the residual material pipeline model with the length of the corresponding pipeline model in the standard module design; The residual material pipeline model whose comparison result meets the comparison rules is replaced with the corresponding pipeline model to obtain an economical module design.

8. The modular design method for a computer room according to claim 5, characterized in that: The step of optimizing the economic module design based on machine learning to obtain a modular design of a computer room includes: After learning the collected design case models, optimizing the module design of the economic module design; After learning the collected layout design case models, the layout design of the economic module design is optimized; After studying the collected connection design case models, the pipeline connection designed by the economic module is optimized.

9. The modular design method for a computer room according to any one of claims 5 to 8, characterized in that: Before the step of optimizing the economic module design based on machine learning to obtain a modular design of a computer room, the following steps are also included: Conduct module performance analysis for economic module design; Verify assembly cost of economic module design through module performance analysis; When the assembly cost is verified, the economic module design is optimized to obtain a modular design of the computer room.

10. 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 computer room modular design method described in any one of claims 5 to 9 are implemented.

Images