Construction material optimization selection method for low-carbon building

Abstract

The invention discloses a low-carbon building-oriented construction material optimization selection method, which comprises the following steps of: inputting a construction component size parameter and a construction process code and a construction material code of a jth construction scheme into a construction material consumption prediction model, and calculating the construction material consumption of the jth construction scheme. Taking the minimum weighted sum of the carbon emission and the construction cost as a target function, and selecting an optimal construction scheme; the method comprehensively considers the carbon emission and the cost, selects the optimal construction scheme, is simple and convenient in calculation, plays a role in predicting and controlling the carbon emission in advance, and has important significance in reducing greenhouse gas emission and protecting the environment in the building industry.

Description

Technical Field

[0001] The present invention relates to the field of building construction, and in particular to a method for optimizing the selection of construction materials for low-carbon construction. Background Art

[0002] Construction projects consume vast amounts of resources and energy, as well as generate substantial amounts of solid waste and greenhouse gas emissions during their construction and operation. The large amounts of greenhouse gases emitted by building material production and construction machinery are a significant contributor to global warming and climate change.

[0003] At present, most of the research on construction scheme evaluation and optimization is qualitative research, and the evaluation systems are all established from the aspects of environmental protection, resource conservation, cost control, etc. There is no construction scheme selection method based on carbon emissions. However, the optimization results of previous construction scheme optimization methods cannot provide guidance for energy conservation and emission reduction in the construction industry. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for optimizing the selection of construction materials for low-carbon construction, comprising the following steps:

[0005] 1) Build construction plan database A and material replacement database B respectively;

[0006] The construction plan database A is as follows:

[0007] (1)

[0008] Where, It is the construction process data sub-library of the nth type of building; The classification code for the nth type of building; Code the first, second, and third level classifications for the nth type of building;

[0009] The construction plan data sub-library for the nth type of building is as follows:

[0010] (2)

[0011] (3)

[0012] Where, Classify and code the mth type of construction component for the nth type of building; is the construction plan matrix of the mth construction component of the nth building; The mth construction component of the nth building construction schemes; For construction plan Construction technology, construction materials, construction material characteristics; construction material characteristics ; are material density, strength grade, elastic modulus, unit carbon emission factor, and unit cost;

[0013] Material substitution database B is as follows:

[0014] (4)

[0015] Where, For construction materials A candidate replacement material vector for ;

[0016] 2) Obtain the name of the building to be built and the building drawings; generate a 3D building model based on the building drawings, and assign a three-level classification code to the building name, which is recorded as the classification code of the building to be built ;

[0017] 3) Split the 3D building model to obtain h construction component models, and assign classification codes for the h construction components to be constructed, namely:

[0018] (5)

[0019] Where, Classify and code the h-th construction component model and the h-th construction component model;

[0020] 4) Based on the classification code of the building to be constructed , Construction component classification code ( ), a construction scheme database A, generates multiple candidate construction schemes for h construction component models, and constructs a candidate construction scheme set;

[0021] Among them, the set of candidate construction schemes for the h-th construction component model is as follows:

[0022] (6)

[0023] Where, is the hth construction component model A selection of construction options; For the selected construction scheme Construction technology, construction materials, and construction material characteristics;

[0024] 5) Constructing a construction material consumption prediction model based on an RBF neural network; the construction material consumption prediction model includes an input layer, a hidden layer, and an output layer;

[0025] The input layer includes the construction process code, construction material code, and construction component size in the selected construction scheme;

[0026] The output of the output layer is the amount of construction materials corresponding to the current construction plan to be selected;

[0027] 6) Extract the construction materials in each candidate construction scheme of the i-th construction component model , and match it with the material replacement database B to obtain the construction material One or more candidate replacement materials, namely:

[0028] (7)

[0029] Where, For construction materials A candidate replacement material vector for ;

[0030] 7) Replace the corresponding construction material with each candidate replacement material in the candidate replacement material vector, thereby adding one or more candidate construction plans, and write the newly added candidate construction plans into the candidate construction plan set to obtain the updated candidate construction plan set. ,Right now:

[0031] (8)

[0032] 8) extracting the size parameters of the construction components in the i-th construction component model;

[0033] 9) Set of construction schemes to be selected The construction process and construction materials of the j-th construction plan are coded to obtain the construction process code and construction material code; the initial value of j is 1;

[0034] Input the construction component size parameters and the construction process code and construction material code of the j-th construction plan into the construction material consumption prediction model, calculate the construction material consumption of the j-th construction plan, write the construction material consumption into the corresponding construction plan, and update the set of candidate construction plans to obtain:

[0035] (9)

[0036] Where, The amount of construction materials used;

[0037] 10) Judgment j≥ Is it true? If so, go to step 11), otherwise, set j=j+1 and return to step 9);

[0038] 11) Calculate the set of construction options separately Carbon emissions and construction costs of each construction option;

[0039] Among them, the carbon emissions of the jth construction plan are and construction costs They are as follows:

[0040] (10)

[0041] (11)

[0042] Where, 、 for unit carbon emissions and unit construction costs;

[0043] 12) Taking the weighted sum of carbon emissions and construction costs as the minimum objective function, select the optimal construction plan for the i-th construction component model;

[0044] Among them, the objective function is as follows:

[0045] (12)

[0046] Where, 、 is the weight;

[0047] 13) Set i = i + 1 and return to step 6) until the optimal construction plan for all construction components is generated.

[0048] Furthermore, in step 1), the first-level building classification includes residential buildings, public buildings, industrial buildings, agricultural buildings, and special buildings;

[0049] The secondary classification of residential buildings includes residences, dormitories, hotels, and other residential buildings;

[0050] The three-level classification of housing includes detached houses, townhouses, and unit houses;

[0051] The three-level classification of dormitories includes student dormitories, staff dormitories;

[0052] The three-level classification of hotels includes hotels, guesthouses, hostels, resorts, and B&Bs;

[0053] The secondary classification of public buildings includes office buildings, educational buildings, scientific research buildings, cultural buildings, performance buildings, sports buildings, medical buildings, commercial buildings, transportation buildings, communication and broadcasting buildings, service buildings, memorial buildings, garden buildings, and other public buildings;

[0054] The three-level classification of office buildings includes administrative office buildings, commercial office buildings;

[0055] The three-level classification of educational buildings includes nurseries, kindergartens, primary and secondary schools, universities, vocational schools, and training centers;

[0056] The three-level classification of scientific research buildings includes research institutes, laboratories, and R&D centers;

[0057] The three-level classification of cultural buildings includes libraries, museums, archives, art galleries, exhibition halls, and cultural centers;

[0058] The three-level classification of performance buildings includes theaters, concert halls, and cinemas;

[0059] The three-level classification of sports buildings includes stadiums, gymnasiums, natatoriums, and fitness centers;

[0060] The three-level classification of medical buildings includes general hospitals, specialized hospitals, clinics, nursing homes, emergency centers, and disease control centers;

[0061] The three-level classification of commercial buildings includes department stores, shopping malls, supermarkets, wet markets, restaurant buildings, and financial buildings;

[0062] The three-level classification of transportation buildings includes bus terminals, railway stations, airport terminals, subway stations, and port passenger terminals;

[0063] The three-level classification of communication and broadcasting buildings includes telecommunication buildings, post offices, radio stations, television stations, and data centers;

[0064] The three-level classification of service buildings includes community service centers, police stations, nursing homes, and child welfare;

[0065] The three-level classification of memorial buildings includes memorial halls and mausoleum buildings;

[0066] The three-level classification of garden architecture includes pavilions, terraces, towers, pavilions, terraces, and corridors;

[0067] The secondary classification of industrial buildings includes production plants, power buildings, storage buildings, transportation buildings, and other industrial buildings;

[0068] The production plant includes workshops;

[0069] The three-level classification of power buildings includes power stations, substations, compressed air stations, and gas stations;

[0070] The third level classification of storage buildings includes warehouses;

[0071] The three-level classification of transportation buildings includes car garages, engine garages, and battery garages;

[0072] The secondary classification of agricultural buildings includes buildings used for agricultural production, processing, storage, and management, and other agricultural buildings;

[0073] The third-level classification of buildings used for agricultural production, processing, storage, and management includes greenhouses, feedlots, agricultural and sideline product processing plants, granaries, agricultural machinery stations, and seed stations;

[0074] The secondary classification of special buildings includes religious buildings, disaster prevention buildings, and other special buildings;

[0075] Furthermore, in step 2), a three-dimensional building model is generated using BIM software.

[0076] Furthermore, in step 3), when splitting the three-dimensional building model, the construction-oriented principle, the material consistency principle, the geometric continuity principle, and the process uniformity principle are followed;

[0077] The construction-oriented principle means that the split boundaries are consistent with the actual construction process;

[0078] The principle of material consistency means that: the materials of individual construction components are the same;

[0079] The principle of geometric continuity means that: a single construction component is a closed geometric body without interruption;

[0080] The principle of process uniformity means that the construction process of individual construction components is the same.

[0081] Furthermore, in step 4), the step of obtaining multiple candidate construction schemes for h construction component models includes:

[0082] 4.1) Match the first-level classification code of the building to be constructed with the first-level classification code in the construction plan database. If the classification code is the same, proceed to step 4.3). Otherwise, proceed to step 4.2);

[0083] 4.2) Match the classification codes of the construction components to be constructed with the classification codes of all construction components in the construction plan database;

[0084] If the construction component classification codes are the same, the construction schemes for the same construction components in the construction scheme database are used as candidate construction schemes, and / or multiple construction schemes manually input through a human-computer interaction device are used as candidate construction schemes;

[0085] If the classification codes of the construction components are different, all the construction plans in the construction plan database are used as candidate construction plans for each construction component, and / or multiple construction plans manually input through the human-computer interaction device are used as candidate construction plans;

[0086] 4.3) Match the second-level classification code of the building to be constructed with the second-level classification code in the construction plan database. If the classification code is the same, proceed to step 4.5). Otherwise, proceed to step 4.4);

[0087] 4.4) Match the classification code of the construction component to be constructed with the classification code of the construction component with the same first-level classification code;

[0088] If the construction component classification codes are the same, the construction schemes for the same construction components in the construction scheme database are used as candidate construction schemes, and / or multiple construction schemes manually input through a human-computer interaction device are used as candidate construction schemes;

[0089] If the classification codes of the construction components are different, all construction plans with the same first-level classification code are used as candidate construction plans for each construction component, and / or multiple construction plans manually input through a human-computer interaction device are used as candidate construction plans;

[0090] 4.5) Match the third-level classification code of the building to be constructed with the third-level classification code in the construction plan database. If the classification code is the same, proceed to step 4.7. Otherwise, proceed to step 4.6.

[0091] 4.6) Match the classification code of the construction component to be constructed with the classification code of the building construction component that has the same first-level and second-level classification codes;

[0092] If the construction component classification codes are the same, the construction schemes for the same construction components in the construction scheme database are used as candidate construction schemes, and / or multiple construction schemes manually input through a human-computer interaction device are used as candidate construction schemes;

[0093] If the classification codes of the construction components are different, all construction plans with the same first-level and second-level classification codes are used as candidate construction plans for each construction component, and / or multiple construction plans manually input through a human-computer interaction device are used as candidate construction plans;

[0094] 4.7) Match the classification code of the construction component to be constructed with the classification code of the building construction component that has the same first-level, second-level, and third-level classification codes;

[0095] If the construction component classification codes are the same, the construction schemes for the same construction components in the construction scheme database are used as candidate construction schemes, and / or multiple construction schemes manually input through a human-computer interaction device are used as candidate construction schemes;

[0096] If the construction component classification codes are different, all construction plans with the same first-level, second-level, and third-level classification codes are used as candidate construction plans for each construction component, and / or multiple construction plans manually input through a human-computer interaction device are used as candidate construction plans.

[0097] Furthermore, the human-computer interaction device includes a computer and a mobile phone.

[0098] Furthermore, in step 5), the steps of constructing a construction material consumption prediction model based on an RBF neural network include:

[0099] 5.1) Construct the RBF neural network model and determine the expected output G, that is:

[0100] ; (13)

[0101] Where P is the regression matrix; W is the weight matrix between the hidden layer and the output layer; I is the identity matrix;

[0102] 5.2) Obtain multiple training samples, including construction process codes, construction material codes, construction component dimensions, and corresponding construction material quantities;

[0103] Calculate the regression factor of each training sample to construct the regression matrix P;

[0104] Among them, the regression factor of the sth training sample is As shown below:

[0105] ; (14)

[0106] Where, is the expansion constant of the radial basis function; is the input in the training sample; is the center of the radial basis function r;

[0107] 4) Orthogonalize the regression matrix P to obtain the orthogonal matrix V and the orthogonal matrix U, namely:

[0108] ; (15)

[0109] Where V is the dimension The upper triangular matrix of , the main diagonal elements are 1; U is a matrix with orthogonal columns;

[0110] 5) Calculate the intermediate matrix u based on the orthogonal matrix U and the expected output G, that is:

[0111] ; (16)

[0112] 6) Based on the intermediate matrix u, calculate the weight W from the hidden layer to the output layer;

[0113] VW=u; (17)

[0114] 7) Based on the weight W from the hidden layer to the output layer, update the RBF neural network model, that is:

[0115] (18)

[0116] 8) Use the validation set to validate the updated RBF neural network model. If the validation passes, output the RBF neural network model. If the validation fails, return to step 5.2).

[0117] Furthermore, in step 9), the step of calculating the amount of construction materials used in the j-th construction plan includes:

[0118] 9.1) Take the construction component size parameters and the construction process code and construction material code of the j-th construction plan as the input vector X, and calculate the Euclidean distance between the input vector X and the weight vector, that is:

[0119] ; (19)

[0120] Where t is the weight vector; R is the total number of hidden nodes; i is any hidden node; is the input vector; is the center of the radial basis function r;

[0121] 9.2) Input the input vector X into the construction material consumption prediction model based on the RBF neural network to obtain the construction material consumption, that is:

[0122] ; (20)

[0123] ; (twenty one)

[0124] Where, is the weight from the hidden layer to the output layer; h is the number of radial basis functions; i is any hidden node; is the radial basis function.

[0125] Furthermore, the architecture, construction technology, construction materials, and construction components are all one-hot encoded.

[0126] Further, in step 9), after obtaining the set of candidate construction schemes Finally, the feasibility of each candidate construction plan is verified, and the candidate construction plans that fail the verification are deleted;

[0127] The steps to verify the feasibility of each candidate construction plan include:

[0128] S1) Discretize the construction component model into a finite element model;

[0129] S2) Input the material properties and construction process of the current construction plan to be selected;

[0130] S3) Define loads and boundary conditions;

[0131] S4) simulating the construction process and calculating structural safety indicators; if the structural safety indicators do not meet preset requirements, the current selected construction plan is not feasible; the structural safety indicators include stress level, deformation index, stability coefficient, and crack risk index.

[0132] The technical effect of the present invention is unquestionable. The present invention comprehensively considers carbon emissions and costs, selects the optimal construction plan, and is simple to calculate. It plays a role in predicting and controlling carbon emissions in advance, and is of great significance to reducing greenhouse gas emissions in the construction industry and protecting the environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0133] Figure 1 Flow chart of the method. DETAILED DESCRIPTION

[0134] The present invention will be further described below with reference to the following examples, but it should not be understood that the scope of the present invention is limited to the following examples. Without departing from the above technical ideas of the present invention, various substitutions and modifications can be made according to common technical knowledge and customary means in the art, and all should be included in the scope of protection of the present invention.

[0135] Example 1:

[0136] See also Figure 1 A method for optimizing the selection of construction materials for low-carbon construction comprises the following steps:

[0137] 1) Build construction plan database A and material replacement database B respectively;

[0138] The construction plan database A is as follows:

[0139] (1)

[0140] Where, It is the construction process data sub-library of the nth type of building; The classification code for the nth type of building; Code the first, second, and third level classifications for the nth type of building;

[0141] The construction plan data sub-library for the nth type of building is as follows:

[0142] (2)

[0143] (3)

[0144] Where, Classify and code the mth type of construction component for the nth type of building; is the construction plan matrix of the mth construction component of the nth building; The mth construction component of the nth building construction schemes; For construction plan Construction technology, construction materials, construction material characteristics; construction material characteristics ; are material density, strength grade, elastic modulus, unit carbon emission factor, and unit cost;

[0145] Material substitution database B is as follows:

[0146] (4)

[0147] Where, For construction materials A candidate replacement material vector for ;

[0148] 2) Obtain the name of the building to be built and the building drawings; generate a 3D building model based on the building drawings, and assign a three-level classification code to the building name, which is recorded as the classification code of the building to be built ;

[0149] 3) Split the 3D building model to obtain h construction component models, and assign classification codes for the h construction components to be constructed, namely:

[0150] (5)

[0151] Where, Classify and code the h-th construction component model and the h-th construction component model;

[0152] 4) Based on the classification code of the building to be constructed , Construction component classification code ( ), a construction scheme database A, generates multiple candidate construction schemes for h construction component models, and constructs a candidate construction scheme set;

[0153] Among them, the set of candidate construction schemes for the h-th construction component model is as follows:

[0154] (6)

[0155] Where, is the hth construction component model A selection of construction options; For the selected construction scheme Construction technology, construction materials, and construction material characteristics;

[0156] 5) Constructing a construction material consumption prediction model based on an RBF neural network; the construction material consumption prediction model includes an input layer, a hidden layer, and an output layer;

[0157] The input layer includes the construction process code, construction material code, and construction component size in the selected construction scheme;

[0158] The output of the output layer is the amount of construction materials corresponding to the current construction plan to be selected;

[0159] 6) Extract the construction materials in each candidate construction scheme of the i-th construction component model , and match it with the material replacement database B to obtain the construction material One or more candidate replacement materials, namely:

[0160] (7)

[0161] Where, For construction materials A candidate replacement material vector for ;

[0162] 7) Replace the corresponding construction material with each candidate replacement material in the candidate replacement material vector, thereby adding one or more candidate construction plans, and write the newly added candidate construction plans into the candidate construction plan set to obtain the updated candidate construction plan set. ,Right now:

[0163] (8)

[0164] 8) extracting the size parameters of the construction components in the i-th construction component model;

[0165] 9) Set of construction schemes to be selected The construction process and construction materials of the j-th construction plan are coded to obtain the construction process code and construction material code; the initial value of j is 1;

[0166] Input the construction component size parameters and the construction process code and construction material code of the j-th construction plan into the construction material consumption prediction model, calculate the construction material consumption of the j-th construction plan, write the construction material consumption into the corresponding construction plan, and update the set of candidate construction plans to obtain:

[0167] (9)

[0168] Where, The amount of construction materials used;

[0169] 10) Judgment j≥ Is it true? If so, go to step 11), otherwise, set j=j+1 and return to step 9);

[0170] 11) Calculate the set of construction options separately Carbon emissions and construction costs of each construction option;

[0171] Among them, the carbon emissions of the jth construction plan are and construction costs They are as follows:

[0172] (10)

[0173] (11)

[0174] Where, 、 for unit carbon emissions and unit construction costs;

[0175] 12) Taking the weighted sum of carbon emissions and construction costs as the minimum objective function, select the optimal construction plan for the i-th construction component model;

[0176] Among them, the objective function As shown below:

[0177] (12)

[0178] Where, 、 is the weight;

[0179] 13) Set i = i + 1 and return to step 6) until the optimal construction plan for all construction components is generated.

[0180] In step 1), the first-level building classification includes residential buildings, public buildings, industrial buildings, agricultural buildings, and special buildings;

[0181] The secondary classification of residential buildings includes residences, dormitories, hotels, and other residential buildings;

[0182] The three-level classification of housing includes detached houses, townhouses, and unit houses;

[0183] The three-level classification of dormitories includes student dormitories, staff dormitories;

[0184] The three-level classification of hotels includes hotels, guesthouses, hostels, resorts, and B&Bs;

[0185] The secondary classification of public buildings includes office buildings, educational buildings, scientific research buildings, cultural buildings, performance buildings, sports buildings, medical buildings, commercial buildings, transportation buildings, communication and broadcasting buildings, service buildings, memorial buildings, garden buildings, and other public buildings;

[0186] The three-level classification of office buildings includes administrative office buildings, commercial office buildings;

[0187] The three-level classification of educational buildings includes nurseries, kindergartens, primary and secondary schools, universities, vocational schools, and training centers;

[0188] The three-level classification of scientific research buildings includes research institutes, laboratories, and R&D centers;

[0189] The three-level classification of cultural buildings includes libraries, museums, archives, art galleries, exhibition halls, and cultural centers;

[0190] The three-level classification of performance buildings includes theaters, concert halls, and cinemas;

[0191] The three-level classification of sports buildings includes stadiums, gymnasiums, natatoriums, and fitness centers;

[0192] The three-level classification of medical buildings includes general hospitals, specialized hospitals, clinics, nursing homes, emergency centers, and disease control centers;

[0193] The three-level classification of commercial buildings includes department stores, shopping malls, supermarkets, wet markets, restaurant buildings, and financial buildings;

[0194] The three-level classification of transportation buildings includes bus terminals, railway stations, airport terminals, subway stations, and port passenger terminals;

[0195] The three-level classification of communication and broadcasting buildings includes telecommunication buildings, post offices, radio stations, television stations, and data centers;

[0196] The three-level classification of service buildings includes community service centers, police stations, nursing homes, and child welfare;

[0197] The three-level classification of memorial buildings includes memorial halls and mausoleum buildings;

[0198] The three-level classification of garden architecture includes pavilions, terraces, towers, pavilions, terraces, and corridors;

[0199] The secondary classification of industrial buildings includes production plants, power buildings, storage buildings, transportation buildings, and other industrial buildings;

[0200] The production plant includes workshops;

[0201] The three-level classification of power buildings includes power stations, substations, compressed air stations, and gas stations;

[0202] The third level classification of storage buildings includes warehouses;

[0203] The three-level classification of transportation buildings includes car garages, engine garages, and battery garages;

[0204] The secondary classification of agricultural buildings includes buildings used for agricultural production, processing, storage, and management, and other agricultural buildings;

[0205] The third-level classification of buildings used for agricultural production, processing, storage, and management includes greenhouses, feedlots, agricultural and sideline product processing plants, granaries, agricultural machinery stations, and seed stations;

[0206] The secondary classification of special buildings includes religious buildings, disaster prevention buildings, and other special buildings;

[0207] In step 2), a three-dimensional building model is generated using BIM software.

[0208] In step 3), when splitting the 3D building model, follow the construction-oriented principles, material consistency principles, geometric continuity principles, and process uniformity principles;

[0209] The construction-oriented principle means that the split boundaries are consistent with the actual construction process;

[0210] The principle of material consistency means that: the materials of individual construction components are the same;

[0211] The principle of geometric continuity means that: a single construction component is a closed geometric body without interruption;

[0212] The principle of process uniformity means that the construction process of individual construction components is the same.

[0213] In step 4), the step of obtaining multiple candidate construction schemes for h construction component models includes:

[0214] 4.1) Match the first-level classification code of the building to be constructed with the first-level classification code in the construction plan database. If the classification code is the same, proceed to step 4.3). Otherwise, proceed to step 4.2);

[0215] 4.2) Match the classification codes of the construction components to be constructed with the classification codes of all construction components in the construction plan database;

[0216] If the construction component classification codes are the same, the construction schemes for the same construction components in the construction scheme database are used as candidate construction schemes, and / or multiple construction schemes manually input through a human-computer interaction device are used as candidate construction schemes;

[0217] If the classification codes of the construction components are different, all the construction plans in the construction plan database are used as candidate construction plans for each construction component, and / or multiple construction plans manually input through the human-computer interaction device are used as candidate construction plans;

[0218] 4.3) Match the second-level classification code of the building to be constructed with the second-level classification code in the construction plan database. If the classification code is the same, proceed to step 4.5). Otherwise, proceed to step 4.4);

[0219] 4.4) Match the classification code of the construction component to be constructed with the classification code of the construction component with the same first-level classification code;

[0220] If the construction component classification codes are the same, the construction schemes for the same construction components in the construction scheme database are used as candidate construction schemes, and / or multiple construction schemes manually input through a human-computer interaction device are used as candidate construction schemes;

[0221] If the classification codes of the construction components are different, all construction plans with the same first-level classification code are used as candidate construction plans for each construction component, and / or multiple construction plans manually input through a human-computer interaction device are used as candidate construction plans;

[0222] 4.5) Match the third-level classification code of the building to be constructed with the third-level classification code in the construction plan database. If the classification code is the same, proceed to step 4.7. Otherwise, proceed to step 4.6.

[0223] 4.6) Match the classification code of the construction component to be constructed with the classification code of the building construction component that has the same first-level and second-level classification codes;

[0224] If the construction component classification codes are the same, the construction schemes for the same construction components in the construction scheme database are used as candidate construction schemes, and / or multiple construction schemes manually input through a human-computer interaction device are used as candidate construction schemes;

[0225] If the classification codes of the construction components are different, all construction plans with the same first-level and second-level classification codes are used as candidate construction plans for each construction component, and / or multiple construction plans manually input through a human-computer interaction device are used as candidate construction plans;

[0226] 4.7) Match the classification code of the construction component to be constructed with the classification code of the building construction component that has the same first-level, second-level, and third-level classification codes;

[0227] If the construction component classification codes are the same, the construction schemes for the same construction components in the construction scheme database are used as candidate construction schemes, and / or multiple construction schemes manually input through a human-computer interaction device are used as candidate construction schemes;

[0228] If the construction component classification codes are different, all construction plans with the same first-level, second-level, and third-level classification codes are used as candidate construction plans for each construction component, and / or multiple construction plans manually input through a human-computer interaction device are used as candidate construction plans.

[0229] The human-computer interaction device includes a computer and a mobile phone.

[0230] In step 5), the steps of constructing a construction material consumption prediction model based on RBF neural network include:

[0231] 5.1) Construct the RBF neural network model and determine the expected output G, that is:

[0232] ; (13)

[0233] Where P is the regression matrix; W is the weight matrix between the hidden layer and the output layer; I is the identity matrix;

[0234] 5.2) Obtain multiple training samples, including construction process codes, construction material codes, construction component dimensions, and corresponding construction material quantities;

[0235] Calculate the regression factor of each training sample to construct the regression matrix P;

[0236] Among them, the regression factor of the sth training sample is As shown below:

[0237] ; (14)

[0238] Where, is the expansion constant of the radial basis function; is the input in the training sample; is the center of the radial basis function r;

[0239] 4) Orthogonalize the regression matrix P to obtain the orthogonal matrix V and the orthogonal matrix U, namely:

[0240] ; (15)

[0241] Where V is an upper triangular matrix of dimension I×I with all diagonal elements being 1; U is a matrix with orthogonal columns;

[0242] 5) Calculate the intermediate matrix u based on the orthogonal matrix U and the expected output G, that is:

[0243] ; (16)

[0244] 6) Based on the intermediate matrix u, calculate the weight W from the hidden layer to the output layer;

[0245] VW=u; (17)

[0246] 7) Based on the weight W from the hidden layer to the output layer, update the RBF neural network model, that is:

[0247] (18)

[0248] 8) Use the validation set to validate the updated RBF neural network model. If the validation passes, output the RBF neural network model. If the validation fails, return to step 5.2).

[0249] In step 9), the steps of calculating the amount of construction materials for the j-th construction plan include:

[0250] 9.1) Take the construction component size parameters and the construction process code and construction material code of the j-th construction plan as the input vector X, and calculate the Euclidean distance between the input vector X and the weight vector, that is:

[0251] ; (19)

[0252] Where t is the weight vector; R is the total number of hidden nodes; i is any hidden node; is the input vector; is the center of the radial basis function r;

[0253] 9.2) Input the input vector X into the construction material consumption prediction model based on the RBF neural network to obtain the construction material consumption, that is:

[0254] ; (20)

[0255] ; (twenty one)

[0256] Where, is the weight from the hidden layer to the output layer; h is the number of radial basis functions; i is any hidden node; is the radial basis function.

[0257] Buildings, construction techniques, construction materials, and construction components all use one-hot encoding.

[0258] In step 9), after obtaining the set of construction schemes to be selected Finally, the feasibility of each candidate construction plan is verified, and the candidate construction plans that fail the verification are deleted;

[0259] The steps to verify the feasibility of each candidate construction plan include:

[0260] S1) Discretize the construction component model into a finite element model;

[0261] S2) Input the material properties and construction process of the current construction plan to be selected;

[0262] S3) Define loads and boundary conditions;

[0263] S4) simulating the construction process and calculating structural safety indicators; if the structural safety indicators do not meet preset requirements, the current selected construction plan is not feasible; the structural safety indicators include stress level, deformation index, stability coefficient, and crack risk index.

[0264] Example 2:

[0265] A method for optimizing the selection of construction materials for low-carbon construction comprises the following steps:

[0266] 1) Build construction plan database A and material replacement database B respectively;

[0267] The construction plan database A is as follows:

[0268] (1)

[0269] Where, It is the construction process data sub-library of the nth type of building; The classification code for the nth type of building; Code the first, second, and third level classifications for the nth type of building;

[0270] The construction plan data sub-library for the nth type of building is as follows:

[0271] (2)

[0272] (3)

[0273] Where, Classify and code the mth type of construction component for the nth type of building; is the construction plan matrix of the mth construction component of the nth building; The mth construction component of the nth building construction schemes; For construction plan Construction technology, construction materials, construction material characteristics; construction material characteristics ; are material density, strength grade, elastic modulus, unit carbon emission factor, and unit cost;

[0274] Material replacement database B is as follows:

[0275] (4)

[0276] Where, For construction materials A candidate replacement material vector for ;

[0277] 2) Obtain the name of the building to be built and the building drawings; generate a 3D building model based on the building drawings, and assign a three-level classification code to the building name, which is recorded as the classification code of the building to be built ;

[0278] 3) Split the 3D building model to obtain h construction component models, and assign classification codes for the h construction components to be constructed, namely:

[0279] (5)

[0280] Where, Classify and code the h-th construction component model and the h-th construction component model;

[0281] 4) Based on the classification code of the building to be constructed , Construction component classification code ( ), a construction scheme database A, generates multiple candidate construction schemes for h construction component models, and constructs a candidate construction scheme set;

[0282] Among them, the set of candidate construction schemes for the h-th construction component model is as follows:

[0283] (6)

[0284] Where, is the hth construction component model A selection of construction options; For the selected construction scheme Construction technology, construction materials, and construction material characteristics;

[0285] 5) Constructing a construction material consumption prediction model based on an RBF neural network; the construction material consumption prediction model includes an input layer, a hidden layer, and an output layer;

[0286] The input layer includes the construction process code, construction material code, and construction component size in the selected construction scheme;

[0287] The output of the output layer is the amount of construction materials corresponding to the current construction plan to be selected;

[0288] 6) Extract the construction materials in each candidate construction scheme of the i-th construction component model , and match it with the material replacement database B to obtain the construction material One or more candidate replacement materials, namely:

[0289] (7)

[0290] Where, For construction materials A candidate replacement material vector for ;

[0291] 7) Replace the corresponding construction material with each candidate replacement material in the candidate replacement material vector, thereby adding one or more candidate construction plans, and write the newly added candidate construction plans into the candidate construction plan set to obtain the updated candidate construction plan set. ,Right now:

[0292] (8)

[0293] 8) extracting the size parameters of the construction components in the i-th construction component model;

[0294] 9) Set of construction schemes to be selected The construction process and construction materials of the j-th construction plan are coded to obtain the construction process code and construction material code; the initial value of j is 1;

[0295] Input the construction component size parameters and the construction process code and construction material code of the j-th construction plan into the construction material consumption prediction model, calculate the construction material consumption of the j-th construction plan, write the construction material consumption into the corresponding construction plan, and update the set of candidate construction plans to obtain:

[0296] (9)

[0297] Where, The amount of construction materials used;

[0298] 10) Judgment j≥ Is it true? If so, go to step 11), otherwise, set j=j+1 and return to step 9);

[0299] 11) Calculate the set of construction options separately Carbon emissions and construction costs of each construction option;

[0300] Among them, the carbon emissions of the jth construction plan are and construction costs They are as follows:

[0301] (10)

[0302] (11)

[0303] Where, 、 for unit carbon emissions and unit construction costs;

[0304] 12) Taking the weighted sum of carbon emissions and construction costs as the minimum objective function, select the optimal construction plan for the i-th construction component model;

[0305] Among them, the objective function is as follows:

[0306] (12)

[0307] Where, 、 is the weight;

[0308] 13) Set i = i + 1 and return to step 6) until the optimal construction plan for all construction components is generated.

[0309] Example 3:

[0310] A method for optimizing the selection of construction materials for low-carbon construction, with the same technical content as in Example 2, wherein, in step 1), the first-level building classification includes residential buildings, public buildings, industrial buildings, agricultural buildings, and special buildings;

[0311] The secondary classification of residential buildings includes residences, dormitories, hotels, and other residential buildings;

[0312] The three-level classification of housing includes detached houses, townhouses, and unit houses;

[0313] The three-level classification of dormitories includes student dormitories, staff dormitories;

[0314] The three-level classification of hotels includes hotels, guesthouses, hostels, resorts, and B&Bs;

[0315] The secondary classification of public buildings includes office buildings, educational buildings, scientific research buildings, cultural buildings, performance buildings, sports buildings, medical buildings, commercial buildings, transportation buildings, communication and broadcasting buildings, service buildings, memorial buildings, garden buildings, and other public buildings;

[0316] The three-level classification of office buildings includes administrative office buildings, commercial office buildings;

[0317] The three-level classification of educational buildings includes nurseries, kindergartens, primary and secondary schools, universities, vocational schools, and training centers;

[0318] The three-level classification of scientific research buildings includes research institutes, laboratories, and R&D centers;

[0319] The three-level classification of cultural buildings includes libraries, museums, archives, art galleries, exhibition halls, and cultural centers;

[0320] The three-level classification of performance buildings includes theaters, concert halls, and cinemas;

[0321] The three-level classification of sports buildings includes stadiums, gymnasiums, natatoriums, and fitness centers;

[0322] The three-level classification of medical buildings includes general hospitals, specialized hospitals, clinics, nursing homes, emergency centers, and disease control centers;

[0323] The three-level classification of commercial buildings includes department stores, shopping malls, supermarkets, wet markets, restaurant buildings, and financial buildings;

[0324] The three-level classification of transportation buildings includes bus terminals, railway stations, airport terminals, subway stations, and port passenger terminals;

[0325] The three-level classification of communication and broadcasting buildings includes telecommunication buildings, post offices, radio stations, television stations, and data centers;

[0326] The three-level classification of service buildings includes community service centers, police stations, nursing homes, and child welfare;

[0327] The three-level classification of memorial buildings includes memorial halls and mausoleum buildings;

[0328] The three-level classification of garden architecture includes pavilions, terraces, towers, pavilions, terraces, and corridors;

[0329] The secondary classification of industrial buildings includes production plants, power buildings, storage buildings, transportation buildings, and other industrial buildings;

[0330] The production plant includes workshops;

[0331] The three-level classification of power buildings includes power stations, substations, compressed air stations, and gas stations;

[0332] The third level classification of storage buildings includes warehouses;

[0333] The three-level classification of transportation buildings includes car garages, engine garages, and battery garages;

[0334] The secondary classification of agricultural buildings includes buildings used for agricultural production, processing, storage, and management, and other agricultural buildings;

[0335] The third-level classification of buildings used for agricultural production, processing, storage, and management includes greenhouses, feedlots, agricultural and sideline product processing plants, granaries, agricultural machinery stations, and seed stations;

[0336] The secondary classification of special buildings includes religious buildings, disaster prevention buildings, and other special buildings;

[0337] Example 4:

[0338] A method for optimizing the selection of construction materials for low-carbon construction, with the same technical content as any one of Examples 2-3, further comprising: in step 2), using BIM software to generate a three-dimensional building model.

[0339] Example 5:

[0340] A method for optimizing the selection of construction materials for low-carbon construction, having the same technical content as any one of Examples 2-4, wherein, in step 3), when splitting the three-dimensional building model, the construction-oriented principle, the material consistency principle, the geometric continuity principle, and the process uniformity principle are followed;

[0341] The construction-oriented principle means that the split boundaries are consistent with the actual construction process;

[0342] The principle of material consistency means that: the materials of individual construction components are the same;

[0343] The principle of geometric continuity means that: a single construction component is a closed geometric body without interruption;

[0344] The principle of process uniformity means that the construction process of individual construction components is the same.

[0345] Example 6:

[0346] A method for optimizing the selection of construction materials for low-carbon construction, with the same technical content as any one of Examples 2-5, further comprising the step of obtaining multiple candidate construction schemes for h construction component models in step 4) comprising:

[0347] 4.1) Match the first-level classification code of the building to be constructed with the first-level classification code in the construction plan database. If the classification code is the same, proceed to step 4.3). Otherwise, proceed to step 4.2);

[0348] 4.2) Match the classification codes of the construction components to be constructed with the classification codes of all construction components in the construction plan database;

[0349] If the construction component classification codes are the same, the construction schemes for the same construction components in the construction scheme database are used as candidate construction schemes, and / or multiple construction schemes manually input through a human-computer interaction device are used as candidate construction schemes;

[0350] If the classification codes of the construction components are different, all the construction plans in the construction plan database are used as candidate construction plans for each construction component, and / or multiple construction plans manually input through the human-computer interaction device are used as candidate construction plans;

[0351] 4.3) Match the second-level classification code of the building to be constructed with the second-level classification code in the construction plan database. If the classification code is the same, proceed to step 4.5). Otherwise, proceed to step 4.4);

[0352] 4.4) Match the classification code of the construction component to be constructed with the classification code of the construction component with the same first-level classification code;

[0353] If the construction component classification codes are the same, the construction schemes for the same construction components in the construction scheme database are used as candidate construction schemes, and / or multiple construction schemes manually input through a human-computer interaction device are used as candidate construction schemes;

[0354] If the classification codes of the construction components are different, all construction plans with the same first-level classification code are used as candidate construction plans for each construction component, and / or multiple construction plans manually input through a human-computer interaction device are used as candidate construction plans;

[0355] 4.5) Match the third-level classification code of the building to be constructed with the third-level classification code in the construction plan database. If the classification code is the same, proceed to step 4.7. Otherwise, proceed to step 4.6.

[0356] 4.6) Match the classification code of the construction component to be constructed with the classification code of the building construction component that has the same first-level and second-level classification codes;

[0357] If the construction component classification codes are the same, the construction schemes for the same construction components in the construction scheme database are used as candidate construction schemes, and / or multiple construction schemes manually input through a human-computer interaction device are used as candidate construction schemes;

[0358] If the classification codes of the construction components are different, all construction plans with the same first-level and second-level classification codes are used as candidate construction plans for each construction component, and / or multiple construction plans manually input through a human-computer interaction device are used as candidate construction plans;

[0359] 4.7) Match the classification code of the construction component to be constructed with the classification code of the building construction component that has the same first-level, second-level, and third-level classification codes;

[0360] If the construction component classification codes are the same, the construction schemes for the same construction components in the construction scheme database are used as candidate construction schemes, and / or multiple construction schemes manually input through a human-computer interaction device are used as candidate construction schemes;

[0361] If the construction component classification codes are different, all construction plans with the same first-level, second-level, and third-level classification codes are used as candidate construction plans for each construction component, and / or multiple construction plans manually input through a human-computer interaction device are used as candidate construction plans.

[0362] Example 7:

[0363] A method for optimizing the selection of construction materials for low-carbon construction, the technical content of which is the same as any one of Examples 2-6, further, the human-computer interaction device includes a computer and a mobile phone.

[0364] Example 8:

[0365] A method for optimizing the selection of construction materials for low-carbon construction, with the same technical content as any one of Examples 2-7, further comprising the step of constructing a construction material consumption prediction model based on an RBF neural network in step 5), comprising:

[0366] 5.1) Construct the RBF neural network model and determine the expected output G, that is:

[0367] ; (13)

[0368] Where P is the regression matrix; W is the weight matrix between the hidden layer and the output layer; I is the identity matrix;

[0369] 5.2) Obtain multiple training samples, including construction process codes, construction material codes, construction component dimensions, and corresponding construction material quantities;

[0370] Calculate the regression factor of each training sample to construct the regression matrix P;

[0371] Among them, the regression factor of the sth training sample is As shown below:

[0372] ; (14)

[0373] Where, is the expansion constant of the radial basis function; is the input in the training sample; is the center of the radial basis function r;

[0374] 4) Orthogonalize the regression matrix P to obtain the orthogonal matrix V and the orthogonal matrix U, namely:

[0375] ; (15)

[0376] Where V is an upper triangular matrix of dimension I×I with all diagonal elements being 1; U is a matrix with orthogonal columns;

[0377] 5) Calculate the intermediate matrix u based on the orthogonal matrix U and the expected output G, that is:

[0378] ; (16)

[0379] 6) Based on the intermediate matrix u, calculate the weight W from the hidden layer to the output layer;

[0380] VW=u; (17)

[0381] 7) Based on the weight W from the hidden layer to the output layer, update the RBF neural network model, that is:

[0382] (18)

[0383] 8) Use the validation set to validate the updated RBF neural network model. If the validation passes, output the RBF neural network model. If the validation fails, return to step 5.2).

[0384] Example 9:

[0385] A method for optimizing the selection of construction materials for low-carbon construction, with the same technical content as any one of Examples 2-8, further comprising, in step 9), calculating the amount of construction materials for the j-th construction plan comprises:

[0386] 9.1) Take the construction component size parameters and the construction process code and construction material code of the j-th construction plan as the input vector X, and calculate the Euclidean distance between the input vector X and the weight vector, that is:

[0387] ; (19)

[0388] Where t is the weight vector; R is the total number of hidden nodes; i is any hidden node; is the input vector; is the center of the radial basis function r;

[0389] 9.2) Input the input vector X into the construction material consumption prediction model based on the RBF neural network to obtain the construction material consumption, that is:

[0390] ; (20)

[0391] ; (twenty one)

[0392] Where, is the weight from the hidden layer to the output layer; h is the number of radial basis functions; i is any hidden node; is the radial basis function.

[0393] Example 10:

[0394] A method for optimizing the selection of construction materials for low-carbon construction, with the same technical content as any one of Examples 2-9, further, the building, construction process, construction materials, and construction components all adopt one-hot encoding.

[0395] Example 11:

[0396] A method for optimizing the selection of construction materials for low-carbon construction, the technical content of which is the same as any one of embodiments 2-3, further, in step 9), after obtaining a set of candidate construction schemes, Finally, the feasibility of each candidate construction plan is verified, and the candidate construction plans that fail the verification are deleted;

[0397] The steps to verify the feasibility of each candidate construction plan include:

[0398] S1) Discretize the construction component model into a finite element model;

[0399] S2) Input the material properties and construction process of the current construction plan to be selected;

[0400] S3) Define loads and boundary conditions;

[0401] S4) simulating the construction process and calculating structural safety indicators; if the structural safety indicators do not meet preset requirements, the current selected construction plan is not feasible; the structural safety indicators include stress level, deformation index, stability coefficient, and crack risk index.

[0402] Example 12:

[0403] A method for optimizing the selection of construction materials for low-carbon construction, with the same technical content as any of Examples 2-11, wherein the construction material replacement database is periodically updated. Update methods include: manual input and crawling of materials with similar properties to the construction material from public literature.

[0404] Example 13:

[0405] A method for optimizing the selection of construction materials for low-carbon construction, with the same technical content as any one of Examples 2-11. When using a validation set to validate an updated RBF neural network model, the mean square error, root mean square error, and mean absolute error are used as validation indicators; if the mean square error, root mean square error, and mean absolute error are less than a preset threshold, the validation passes; otherwise, the validation fails.

Claims

1. A method for optimizing the selection of construction materials for low-carbon construction, characterized in that: The following steps are involved: 1) Build construction plan database A and material replacement database B respectively; The construction plan database A is as follows: （1） Where, It is the construction process data sub-library of the nth type of building; The classification code for the nth type of building; Code the first, second, and third level classifications for the nth type of building; The construction plan data sub-library for the nth type of building is as follows: （2） （3） Where, Classify and code the mth type of construction component for the nth type of building; is the construction plan matrix of the mth construction component of the nth building; The mth construction component of the nth building construction schemes; For construction plan Construction technology, construction materials, construction material characteristics; construction material characteristics ; are material density, strength grade, elastic modulus, unit carbon emission factor, and unit cost; Material replacement database B is as follows: （4） Where, For construction materials A candidate replacement material vector for ; 2) Obtain the name of the building to be built and the building drawings; generate a 3D building model based on the building drawings, and assign a three-level classification code to the building name, which is recorded as the classification code of the building to be built ; 3) Split the 3D building model to obtain h construction component models, and assign classification codes for the h construction components to be constructed, namely: （5） Where, Classify and code the h-th construction component model and the h-th construction component model; 4) Based on the classification code of the building to be constructed , Construction component classification code ( ), a construction scheme database A, generates multiple candidate construction schemes for h construction component models, and constructs a candidate construction scheme set; Among them, the set of candidate construction schemes for the h-th construction component model is as follows: （6） Where, is the hth construction component model A selection of construction options; For the selected construction scheme Construction technology, construction materials, and construction material characteristics; 5) Constructing a construction material consumption prediction model based on an RBF neural network; the construction material consumption prediction model includes an input layer, a hidden layer, and an output layer; The input layer includes the construction process code, construction material code, and construction component size in the selected construction scheme; The output of the output layer is the amount of construction materials corresponding to the current construction plan to be selected; 6) Extract the construction materials in each candidate construction scheme of the i-th construction component model , and match it with the material replacement database B to obtain the construction material One or more candidate replacement materials, namely: （7） Where, For construction materials A candidate replacement material vector for ; 7) Replace the corresponding construction material with each candidate replacement material in the candidate replacement material vector, thereby adding one or more candidate construction plans, and write the newly added candidate construction plans into the candidate construction plan set to obtain the updated candidate construction plan set. ,Right now: （8） 8) extracting the size parameters of the construction components in the i-th construction component model; 9) Set of construction schemes to be selected The construction process and construction materials of the j-th construction plan are coded to obtain the construction process code and construction material code; the initial value of j is 1; Input the construction component size parameters and the construction process code and construction material code of the j-th construction plan into the construction material consumption prediction model, calculate the construction material consumption of the j-th construction plan, write the construction material consumption into the corresponding construction plan, and update the set of candidate construction plans to obtain: （9） Where, The amount of construction materials used; 10) Judgment j≥ Is it true? If so, go to step 11), otherwise, set j=j+1 and return to step 9); 11) Calculate the set of construction options separately Carbon emissions and construction costs of each construction option; Among them, the carbon emissions of the jth construction plan are and construction costs They are as follows: （10） （11） Where, 、 for unit carbon emissions and unit construction costs; 12) Taking the weighted sum of carbon emissions and construction costs as the minimum objective function, select the optimal construction plan for the i-th construction component model; Among them, the objective function is as follows: （12） Where, 、 is the weight; 13) Set i = i + 1 and return to step 6) until the optimal construction plan for all construction components is generated.

2. The method for optimizing construction material selection for low-carbon construction according to claim 1, characterized in that: In step 1), the first-level building classification includes residential buildings, public buildings, industrial buildings, agricultural buildings, and special buildings; The secondary classification of residential buildings includes residences, dormitories, hotels, and other residential buildings; The three-level classification of housing includes detached houses, townhouses, and unit houses; The three-level classification of dormitories includes student dormitories, staff dormitories; The three-level classification of hotels includes hotels, guesthouses, hostels, resorts, and B&Bs; The secondary classification of public buildings includes office buildings, educational buildings, scientific research buildings, cultural buildings, performance buildings, sports buildings, medical buildings, commercial buildings, transportation buildings, communication and broadcasting buildings, service buildings, memorial buildings, garden buildings, and other public buildings; The three-level classification of office buildings includes administrative office buildings, commercial office buildings; The three-level classification of educational buildings includes nurseries, kindergartens, primary and secondary schools, universities, vocational schools, and training centers; The three-level classification of scientific research buildings includes research institutes, laboratories, and R&D centers; The three-level classification of cultural buildings includes libraries, museums, archives, art galleries, exhibition halls, and cultural centers; The three-level classification of performance buildings includes theaters, concert halls, and cinemas; The three-level classification of sports buildings includes stadiums, gymnasiums, natatoriums, and fitness centers; The three-level classification of medical buildings includes general hospitals, specialized hospitals, clinics, nursing homes, emergency centers, and disease control centers; The three-level classification of commercial buildings includes department stores, shopping malls, supermarkets, wet markets, restaurant buildings, and financial buildings; The three-level classification of transportation buildings includes bus terminals, railway stations, airport terminals, subway stations, and port passenger terminals; The three-level classification of communication and broadcasting buildings includes telecommunication buildings, post offices, radio stations, television stations, and data centers; The three-level classification of service buildings includes community service centers, police stations, nursing homes, and child welfare; The three-level classification of memorial buildings includes memorial halls and mausoleum buildings; The three-level classification of garden architecture includes pavilions, terraces, towers, pavilions, terraces, and corridors; The secondary classification of industrial buildings includes production plants, power buildings, storage buildings, transportation buildings, and other industrial buildings; The production plant includes workshops; The three-level classification of power buildings includes power stations, substations, compressed air stations, and gas stations; The third level classification of storage buildings includes warehouses; The three-level classification of transportation buildings includes car garages, engine garages, and battery garages; The secondary classification of agricultural buildings includes buildings used for agricultural production, processing, storage, and management, and other agricultural buildings; The third-level classification of buildings used for agricultural production, processing, storage, and management includes greenhouses, feedlots, agricultural and sideline product processing plants, granaries, agricultural machinery stations, and seed stations; The secondary classification of special buildings includes religious buildings, disaster prevention buildings, and other special buildings.

3. The method for optimizing construction material selection for low-carbon construction according to claim 1, characterized in that: In step 2), a three-dimensional building model is generated using BIM software.

4. The method for optimizing construction material selection for low-carbon construction according to claim 1, characterized in that: In step 3), when splitting the 3D building model, follow the construction-oriented principles, material consistency principles, geometric continuity principles, and process uniformity principles; The construction-oriented principle means that the split boundaries are consistent with the actual construction process; The principle of material consistency means that: the materials of individual construction components are the same; The principle of geometric continuity means that: a single construction component is a closed geometric body without interruption; The principle of process uniformity means that the construction process of individual construction components is the same.

5. The method for optimizing construction material selection for low-carbon construction according to claim 1, characterized in that: In step 4), the step of obtaining multiple candidate construction schemes for h construction component models includes: 4.1) Match the first-level classification code of the building to be constructed with the first-level classification code in the construction plan database. If the classification code is the same, proceed to step 4.3). Otherwise, proceed to step 4.2); 4.2) Match the classification codes of the construction components to be constructed with the classification codes of all construction components in the construction plan database; If the construction component classification codes are the same, the construction schemes for the same construction components in the construction scheme database are used as candidate construction schemes, and / or multiple construction schemes manually input through a human-computer interaction device are used as candidate construction schemes; If the classification codes of the construction components are different, all the construction plans in the construction plan database are used as candidate construction plans for each construction component, and / or multiple construction plans manually input through the human-computer interaction device are used as candidate construction plans; 4.3) Match the second-level classification code of the building to be constructed with the second-level classification code in the construction plan database. If the classification code is the same, proceed to step 4.5). Otherwise, proceed to step 4.4); 4.4) Match the classification code of the construction component to be constructed with the classification code of the construction component with the same first-level classification code; If the construction component classification codes are the same, the construction schemes for the same construction components in the construction scheme database are used as candidate construction schemes, and / or multiple construction schemes manually input through a human-computer interaction device are used as candidate construction schemes; If the classification codes of the construction components are different, all construction plans with the same first-level classification code are used as candidate construction plans for each construction component, and / or multiple construction plans manually input through a human-computer interaction device are used as candidate construction plans; 4.5) Match the third-level classification code of the building to be constructed with the third-level classification code in the construction plan database. If the classification code is the same, proceed to step 4.

7. Otherwise, proceed to step 4.

6. 4.6) Match the classification code of the construction component to be constructed with the classification code of the building construction component that has the same first-level and second-level classification codes; If the construction component classification codes are the same, the construction schemes for the same construction components in the construction scheme database are used as candidate construction schemes, and / or multiple construction schemes manually input through a human-computer interaction device are used as candidate construction schemes; If the classification codes of the construction components are different, all construction plans with the same first-level and second-level classification codes are used as candidate construction plans for each construction component, and / or multiple construction plans manually input through a human-computer interaction device are used as candidate construction plans; 4.7) Match the classification code of the construction component to be constructed with the classification code of the building construction component that has the same first-level, second-level, and third-level classification codes; If the construction component classification codes are the same, the construction schemes for the same construction components in the construction scheme database are used as candidate construction schemes, and / or multiple construction schemes manually input through a human-computer interaction device are used as candidate construction schemes; If the construction component classification codes are different, all construction plans with the same first-level, second-level, and third-level classification codes are used as candidate construction plans for each construction component, and / or multiple construction plans manually input through a human-computer interaction device are used as candidate construction plans.

6. The method for optimizing construction material selection for low-carbon construction according to claim 5, characterized in that: The human-computer interaction device includes a computer and a mobile phone.

7. The method for optimizing construction material selection for low-carbon construction according to claim 1, characterized in that: In step 5), the steps of constructing a construction material consumption prediction model based on RBF neural network include: 5.1) Construct the RBF neural network model and determine the expected output G, that is: ； （13） Where P is the regression matrix; W is the weight matrix between the hidden layer and the output layer; I is the identity matrix; 5.2) Obtain multiple training samples, including construction process codes, construction material codes, construction component dimensions, and corresponding construction material quantities; Calculate the regression factor of each training sample to construct the regression matrix P; Among them, the regression factor of the sth training sample is As shown below: ； （14） Where, is the expansion constant of the radial basis function; is the input in the training sample; is the center of the radial basis function r; 4) Orthogonalize the regression matrix P to obtain the orthogonal matrix V and the orthogonal matrix U, namely: ； （15） Where V is the dimension The upper triangular matrix of , the main diagonal elements are 1; U is a matrix with orthogonal columns; 5) Calculate the intermediate matrix u based on the orthogonal matrix U and the expected output G, that is: ； （16） 6) Based on the intermediate matrix u, calculate the weight W from the hidden layer to the output layer; VW=u; (17) 7) Based on the weight W from the hidden layer to the output layer, update the RBF neural network model, that is: （18） 8) Use the validation set to validate the updated RBF neural network model. If the validation passes, output the RBF neural network model. If the validation fails, return to step 5.2).

8. The online calculation method of probability power flow based on RBF neural network according to claim 1 is characterized in that: In step 9), the steps of calculating the amount of construction materials for the j-th construction plan include: 9.1) Take the construction component size parameters and the construction process code and construction material code of the j-th construction plan as the input vector X, and calculate the Euclidean distance between the input vector X and the weight vector, that is: ； （19） Where t is the weight vector; R is the total number of hidden nodes; i is any hidden node; is the input vector; is the center of the radial basis function r; 9.2) Input the input vector X into the construction material consumption prediction model based on the RBF neural network to obtain the construction material consumption, that is: ； （20） ； （21） Where, is the weight from the hidden layer to the output layer; h is the number of radial basis functions; i is any hidden node; is the radial basis function.

9. The method for online calculation of probabilistic power flow based on RBF neural network according to claim 1, characterized in that: Buildings, construction techniques, construction materials, and construction components all use one-hot encoding.

10. The method for online calculation of probabilistic power flow based on RBF neural network according to claim 1, characterized in that: In step 9), after obtaining the set of construction schemes to be selected Finally, the feasibility of each candidate construction plan is verified, and the candidate construction plans that fail the verification are deleted; The steps to verify the feasibility of each candidate construction plan include: S1) Discretize the construction component model into a finite element model; S2) Input the material properties and construction process of the current construction plan to be selected; S3) Define loads and boundary conditions; S4) simulating the construction process and calculating structural safety indicators; if the structural safety indicators do not meet preset requirements, the current selected construction plan is not feasible; the structural safety indicators include stress level, deformation index, stability coefficient, and crack risk index.

Images