Building structure analysis method and device

Abstract

The invention relates to the technical field of data analysis, in particular to a building structure analysis method and device. Obtaining building structure design data corresponding to multiple candidate designs of the target building; the building structure design data comprises a building structure type, a structure material type, a structure material size, a structure material consumption and structure node data; for each candidate design, according to the building structure design data, analyzing the structure safety level and the carbon emission standard level of the target building; aiming at each candidate design, based on the structure safety level and the carbon emission standard level, carrying out design scoring on the candidate design; and determining the candidate design with the highest design score as a final building structure design scheme of the target building. By acquiring and analyzing multi-dimensional data of building structure design, comprehensive evaluation of the safety level and the carbon emission standard level of a target building structure is realized, design scoring is performed based on an evaluation result, and an optimal building structure scheme is determined.

Description

Technical Field

[0001] The present application relates to the technical field of data analysis, and in particular to a method and device for analyzing building structures. Background Art

[0002] With the acceleration of urbanization and the continuous improvement of people's requirements for building quality, building structure design has become a core link in the field of construction engineering. Building structure design is not only related to the safety, stability and durability of buildings, but also directly affects their economy, environmental protection and aesthetics. Traditional building structure design methods often focus on single-dimensional considerations, such as only focusing on the safety or economy of the structure, while ignoring the comprehensive evaluation of other important factors. Once the building structure is selected and enters the construction preparation stage, any further adjustments will greatly affect the construction progress.

[0003] Therefore, there is an urgent need for a technology that can conduct a comprehensive and reasonable analysis when designing building structures to balance structural safety, economy, and environmental protection. Summary of the invention

[0004] In order to solve the problems in the background technology, the present application provides a building structure analysis method and device, which realizes a comprehensive assessment of the safety level and carbon emission compliance level of the target building structure by comprehensively acquiring and analyzing multi-dimensional data of the building structure design, and performs design scoring based on the evaluation results to ultimately determine the optimal building structure solution.

[0005] In a first aspect, the present application provides a building structure analysis method, comprising: Acquire building structure design data corresponding to multiple candidate designs of the target building; the building structure design data includes building structure type, structural material type, structural material size, structural material usage, and structural node data; For each candidate design, analyzing the structural safety level and carbon emission compliance level of the target building based on the building structure design data; For each candidate design, a design score is given to the candidate design based on the structural safety level and the carbon emission compliance level; The candidate design with the highest design score is determined as the final architectural structure design solution of the target building.

[0006] Optionally, analyzing the structural safety level and carbon emission compliance level of the target building according to the building structure design data includes: Determine the load type and corresponding load requirement of the target building according to the building type and building location; Conduct structural strength analysis based on building structure type, structural material type, structural material size, structural material usage and structural node data; Analyze the structural strength grade of the target building according to the structural strength and the load requirements corresponding to different load types; Combined with the instability factors, the second-order effect analysis is carried out, and the stability analysis is carried out to determine the critical load when the structure is unstable; Analyze the stability level of the target building according to the critical load when the structure is unstable and the load requirements corresponding to different load types; Determine the earthquake intensity and geological conditions of the target building location based on the building location; Analyzing the seismic resistance level of the target building according to the earthquake intensity and geological conditions; The structural safety grade of the target building is determined according to the structural strength grade, stability grade and earthquake resistance grade of the target building.

[0007] Optionally, for each design, analyzing the structural safety level and carbon emission compliance level of the target building according to the building structure design data includes: Acquire the building data and relevant carbon emission requirement data of the target building; the building data includes building age data, building structure data, building scale data, building life data, building location data, and building type data; Input the building data and relevant carbon emission requirement data into the shared input layer of the trained carbon emission phased prediction model; the carbon emission phased prediction model is trained based on historical building data, historical relevant carbon emission requirement data and corresponding historical carbon emission data; the historical carbon emission data includes carbon emission data at different stages of the building's life cycle; the carbon emission phased prediction model includes a main network and multiple sub-networks, the main network includes a shared input layer, and each of the sub-networks includes an independent hidden layer and an independent output layer, each corresponding to a stage in the life cycle; Outputting preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building in multiple independent output layers of the carbon emissions staged prediction model respectively; Based on the preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building, the predicted correlation between the carbon emission data of each stage is calculated; If the predicted correlation is not within the preset correlation range, the preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building are adjusted until the predicted correlation between the carbon emission data of each stage reaches within the preset correlation range, thereby obtaining the predicted values ​​of carbon emissions corresponding to different stages of the target building.

[0008] Optionally, if the predicted correlation is not within a preset correlation range, the preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building are adjusted until the predicted correlation between the carbon emission data of each stage reaches within a preset correlation range, thereby obtaining the predicted values ​​of carbon emissions corresponding to different stages of the target building, including: For the predicted correlation between carbon emission data of any two stages, calculate the deviation direction and deviation value of the predicted correlation and the preset correlation between the carbon emission data of the two stages; Determine the deviation stage according to the deviation direction and deviation value corresponding to the carbon emission data of each two stages; The preliminary predicted value of carbon emissions in the deviation stage is adjusted until the predicted correlation between the carbon emission data of each stage reaches a preset correlation range, thereby obtaining the predicted values ​​of carbon emissions corresponding to different stages of the target building.

[0009] Optionally, the training process of the carbon emission phased prediction model includes: Obtain the building data of historical buildings, historical relevant carbon emission requirements data and corresponding historical carbon emission data; Analyze the characteristics of the historical carbon emission data, determine the time step, and construct carbon emission time series sample data and carbon emission time series test data at different stages of the historical building's life cycle; The carbon emission time series sample data of the historical building at different stages of its life cycle are input into the constructed long-short term memory model for iterative training until the set stop condition is reached; Input the historical building data and historical relevant carbon emission requirement data in the carbon emission time series test data into the iteratively trained long short-term memory model, and compare the output results with the historical carbon emission data in the carbon emission time series test data set to verify the training effect; Repeat the iterative training and validation steps until the expected training effect is achieved and a trained carbon emission prediction model is obtained.

[0010] Optionally, analyzing the characteristics of the historical carbon emission data, determining the time step, and constructing the carbon emission time series sample data and carbon emission time series test data at different stages of the life cycle of the historical building include: For each stage of carbon emission data corresponding to each historical building, a time window is set to sample the carbon emission data of the stage, the initial size of the time window is the time span of the stage, and the step size of the time window is the size of the time window; For the sampled data in the time window, calculate the range value and the variance value thereof. If the range value is greater than a first preset threshold value, and / or the variance value is greater than a second preset threshold value, retain all the data in the time window, reduce the size of the time window, perform re-sampling, and repeat this step until the range value of the data in all time windows is not greater than the first preset threshold value, and the variance value is not greater than the second preset threshold value; The time step is determined according to the size of each time window at this time, and the carbon emission data corresponding to each time step is determined according to the mean of the data in each time window, so as to construct the carbon emission time series sample data and carbon emission time series test data of the historical building at different stages of its life cycle.

[0011] Optionally, reducing the size of the time window includes: If the range value of the sampled data in any two adjacent time windows is greater than the first preset threshold, and / or the variance value is greater than the second preset threshold, then the distance between the two maximum values ​​in the two adjacent time windows is calculated; If the time distance between the two maximum values ​​is less than or equal to the window size, the time window is reduced by 1 / 3.

[0012] Optionally, determining the deviation stage according to the deviation direction and deviation value corresponding to the carbon emission data of each two stages includes: For each stage, if the deviation value of the predicted correlation between the carbon emission data of this stage and the carbon emission data of at least two other stages relative to the preset correlation is greater than the preset value, and the deviation direction is consistent with the deviation direction between the preset correlation degrees corresponding to the carbon emission data of at least two other stages, then this stage is determined as a deviation stage.

[0013] Optionally, the method further includes: After each stage of the target building is completed, the energy consumption data of the stage is obtained, and the actual carbon emission data of the stage is determined based on the energy consumption data of the stage; The actual carbon emission data of the stage is compared with the carbon emission forecast value of the stage. If the data difference is greater than a third preset threshold, the carbon emission forecast value of the next stage is adjusted based on the data difference and the preset correlation to determine the carbon emission corrected forecast value.

[0014] In a second aspect, the present application provides a building structure analysis device, comprising: An acquisition module is used to acquire building structure design data corresponding to multiple candidate designs of a target building; the building structure design data includes building structure type, structural material type, structural material size, structural material usage, and structural node data; An analysis module, for analyzing the structural safety level and carbon emission compliance level of the target building according to the building structure design data for each candidate design; A design scoring module, for scoring each candidate design based on the structural safety level and the carbon emission compliance level; The design scheme determination module is used to determine the candidate design with the highest design score as the final building structure design scheme of the target building.

[0015] Optionally, the analysis module is specifically used to obtain the building data and relevant carbon emission requirement data of the target building; the building data includes building age data, building structure data, building scale data, building life data, building location data, and building type data; the building data and relevant carbon emission requirement data are input into the shared input layer of the trained carbon emission staged prediction model; the carbon emission staged prediction model is trained based on historical building data, historical relevant carbon emission requirement data and corresponding historical carbon emission data; the historical carbon emission data includes carbon emission data at different stages of the building's entire life cycle; the carbon emission staged prediction model includes a main network and multiple sub-networks, The main network includes a shared input layer, and each of the sub-networks includes an independent hidden layer and an independent output layer, each corresponding to a stage in the entire life cycle; the multiple independent output layers of the carbon emission staged prediction model respectively output the preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building; based on the preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building, the predicted correlation between the carbon emission data of each stage is calculated; if the predicted correlation is not within a preset correlation range, the preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building are adjusted until the predicted correlation between the carbon emission data of each stage reaches within the preset correlation range, thereby obtaining the carbon emission predicted values ​​corresponding to different stages of the target building.

[0016] Optionally, the preliminary prediction value adjustment module is specifically used to: For the predicted correlation between carbon emission data of any two stages, calculate the deviation direction and deviation value of the predicted correlation and the preset correlation between the carbon emission data of the two stages; Determine the deviation stage according to the deviation direction and deviation value corresponding to the carbon emission data of each two stages; The preliminary predicted value of carbon emissions in the deviation stage is adjusted until the predicted correlation between the carbon emission data of each stage reaches a preset correlation range, thereby obtaining the predicted values ​​of carbon emissions corresponding to different stages of the target building.

[0017] Optionally, a training module is also included to: Obtain the building data of historical buildings, historical relevant carbon emission requirements data and corresponding historical carbon emission data; Analyze the characteristics of the historical carbon emission data, determine the time step, and construct carbon emission time series sample data and carbon emission time series test data at different stages of the historical building's life cycle; The carbon emission time series sample data of the historical building at different stages of its life cycle are input into the constructed long-short term memory model for iterative training until the set stop condition is reached; Input the historical building data and historical relevant carbon emission requirement data in the carbon emission time series test data into the iteratively trained long short-term memory model, and compare the output results with the historical carbon emission data in the carbon emission time series test data set to verify the training effect; Repeat the iterative training and validation steps until the expected training effect is achieved and a trained carbon emission prediction model is obtained.

[0018] Optionally, when analyzing the characteristics of the historical carbon emission data, determining the time step, and constructing the carbon emission time series sample data and carbon emission time series test data at different stages of the life cycle of the historical building, the training module is specifically used to: For each stage of carbon emission data corresponding to each historical building, a time window is set to sample the carbon emission data of the stage, the initial size of the time window is the time span of the stage, and the step size of the time window is the size of the time window; For the sampled data in the time window, calculate the range value and the variance value thereof. If the range value is greater than a first preset threshold value, and / or the variance value is greater than a second preset threshold value, retain all the data in the time window, reduce the size of the time window, perform re-sampling, and repeat this step until the range value of the data in all time windows is not greater than the first preset threshold value, and the variance value is not greater than the second preset threshold value; The time step is determined according to the size of each time window at this time, and the carbon emission data corresponding to each time step is determined according to the mean of the data in each time window, so as to construct the carbon emission time series sample data and carbon emission time series test data of the historical building at different stages of its life cycle.

[0019] Optionally, when reducing the size of the time window, the training module is specifically used to: If the range value of the sampled data in any two adjacent time windows is greater than the first preset threshold, and / or the variance value is greater than the second preset threshold, then the distance between the two maximum values ​​in the two adjacent time windows is calculated; If the time distance between the two maximum values ​​is less than or equal to the window size, the time window is reduced by 1 / 3.

[0020] Optionally, when determining the deviation stage according to the deviation direction and deviation value corresponding to the carbon emission data of each two stages, the preliminary prediction value adjustment module is specifically used to: For each stage, if the deviation value of the predicted correlation between the carbon emission data of this stage and the carbon emission data of at least two other stages relative to the preset correlation is greater than the preset value, and the deviation direction is consistent with the deviation direction between the preset correlation degrees corresponding to the carbon emission data of at least two other stages, then this stage is determined as a deviation stage.

[0021] Optionally, the device further includes a prediction value correction module, configured to: After each stage of the target building is completed, the energy consumption data of the stage is obtained, and the actual carbon emission data of the stage is determined based on the energy consumption data of the stage; The actual carbon emission data of the stage is compared with the carbon emission forecast value of the stage. If the data difference is greater than a third preset threshold, the carbon emission forecast value of the next stage is adjusted based on the data difference and the preset correlation to determine the carbon emission corrected forecast value.

[0022] In a third aspect, the present application provides an electronic device, comprising: a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute the method of the first aspect.

[0023] In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and execute the method of the first aspect.

[0024] In a fifth aspect, the present application provides a computer program product, comprising: a computer program; when the computer program is executed by a processor, it implements the method described in any one of the first aspects.

[0025] This application provides a building structure analysis method and device. By comprehensively acquiring and analyzing multi-dimensional data of building structure design, a comprehensive assessment of the safety level and carbon emission compliance level of the target building structure is achieved, and a design score is performed based on the assessment results, and the optimal building structure solution is finally determined. This method has the following significant technical effects: 1. Comprehensiveness and accuracy: By obtaining multi-dimensional data of building structure design (including building structure type, structural material type, structural material size, structural material usage, structural node data, etc.), this method can comprehensively reflect the various performance indicators of the building structure and improve the accuracy and reliability of the evaluation.

[0026] 2. Comprehensive evaluation capability: This method not only focuses on the safety of the structure, but also takes into account the economy and environmental protection (such as the carbon emission compliance level), and realizes the comprehensive evaluation of the building structure design scheme. This comprehensive evaluation method helps designers find the best balance between multiple goals and design a building structure that is safe, economical and environmentally friendly.

[0027] 3. Efficiency and intelligence: Through automated data analysis and design scoring processes, this method significantly improves the efficiency of building structure design. At the same time, the design selection process based on the evaluation results also reflects the characteristics of intelligent decision-making, which helps to quickly determine the optimal building structure solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0029] Figure 1 A flow chart of a building structure analysis method provided in one embodiment of the present application; Figure 2 A flowchart of a construction process carbon emission prediction process provided by an embodiment of the present application; Figure 3 A flowchart of a training process of a carbon emission phased prediction model provided in an embodiment of the present application; Figure 4 A schematic diagram of the structure of a building structure analysis device provided in one embodiment of the present application; Figure 5 A schematic diagram of the structure of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are 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.

[0031] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0032] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0033] Figure 1 A flow chart of a building structure analysis method provided for this application, such as Figure 1 As shown, the method includes: S101: Acquire building structure design data corresponding to multiple candidate designs of a target building; the building structure design data includes building structure type, structural material type, structural material size, structural material usage, and structural node data.

[0034] For each candidate design of the target building, detailed building structure design data is collected. These data include building structure type (such as frame, shear wall, cylinder, etc.), structural material type (such as steel, concrete, wood, etc.), structural material size (such as beam cross-sectional size, plate thickness, etc.), structural material usage (total usage of various materials or usage per unit area), and structural node data (such as node connection method, material reinforcement at the node, etc.).

[0035] The dimensions of building structure design data are multifaceted, and the above are some of the main dimensions. Structural types include frame structure, truss structure, arch structure, cable suspension structure, etc. Common structural materials include steel, concrete, wood, stone, glass, etc. Each material has its own unique physical and chemical properties. Structural material dimensions include the span of beams, floor and column values, etc. Structural nodes include connection methods, components, dimensions and materials, etc.

[0036] S102: For each candidate design, analyzing the structural safety level and carbon emission compliance level of the target building according to the building structure design data.

[0037] For each candidate design, a structural safety analysis is conducted based on its building structure design data. At the same time, the carbon emissions of each candidate design need to be analyzed, including carbon emissions at each stage of the building life cycle, such as building material production, transportation, construction and building operation, to determine its carbon emission compliance level.

[0038] S103: For each candidate design, a design score is given to the candidate design based on the structural safety level and the carbon emission compliance level.

[0039] Each candidate design is scored based on the structural safety level and carbon emission compliance level. The score can be a comprehensive indicator that considers both structural safety and environmental performance. For example, a weight can be set to combine the structural safety level and carbon emission compliance level in a certain proportion to obtain a comprehensive score.

[0040] S104: Determine the candidate design with the highest design score as the final architectural structure design solution of the target building.

[0041] Among all candidate designs, the one with the highest design score is selected as the final architectural structural design of the target building. This design should achieve the best balance between structural safety and environmental performance.

[0042] By comprehensively acquiring and analyzing multi-dimensional data of building structure design, a comprehensive assessment of the safety level and carbon emission compliance level of the target building structure is achieved, and a design score is given based on the assessment results to ultimately determine the optimal building structure solution. This method has the following significant technical effects: 1. Comprehensiveness and accuracy: By obtaining multi-dimensional data of building structure design (including building structure type, structural material type, structural material size, structural material usage, structural node data, etc.), this method can comprehensively reflect the various performance indicators of the building structure and improve the accuracy and reliability of the evaluation.

[0043] 2. Comprehensive evaluation capability: This method not only focuses on the safety of the structure, but also takes into account the economy and environmental protection (such as the carbon emission compliance level), and realizes the comprehensive evaluation of the building structure design scheme. This comprehensive evaluation method helps designers find the best balance between multiple goals and design a building structure that is safe, economical and environmentally friendly.

[0044] 3. Efficiency and intelligence: Through automated data analysis and design scoring processes, this method significantly improves the efficiency of building structure design. At the same time, the design selection process based on the evaluation results also reflects the characteristics of intelligent decision-making, which helps to quickly determine the optimal building structure solution.

[0045] In some specific embodiments, the structural safety level and carbon emission compliance level of the target building are analyzed according to the building structure design data, including: determining the load type and corresponding load demand of the target building according to the building type and building location; performing structural strength analysis based on the building structure type, structural material type, structural material size, structural material usage and structural node data; analyzing the structural strength level of the target building according to the structural strength and the load demand corresponding to different load types; performing second-order effect analysis and stability analysis in combination with instability factors to determine the critical load when the structure is unstable; analyzing the stability level of the target building according to the critical load when the structure is unstable and the load demand corresponding to different load types; determining the earthquake intensity and geological conditions of the location of the target building according to the building location; analyzing the seismic resistance level of the target building according to the earthquake intensity and geological conditions; determining the structural safety level of the target building according to the structural strength level, stability level and seismic resistance level of the target building.

[0046] Structural strength analysis aims to evaluate the bearing capacity of building structures under normal use and extreme conditions, and ensure that the structure will not be damaged under various loads. According to the use function and geographical location of the building, determine the types of loads that may act on the structure, including dead loads (such as structural deadweight), live loads (such as crowds, furniture weight), wind loads, snow loads, earthquake loads, etc. A corresponding calculation model can be established. For example, professional structural analysis software (such as STAAD, ANSYS, etc.) can be used to establish an accurate structural calculation model based on the building structure type, material type, material size and node data. According to relevant specifications, different types of loads are combined to simulate the most unfavorable working conditions that the structure may withstand. Through software calculations, the stress, strain and displacement distribution of the structure under each load combination can be obtained to evaluate whether the strength of the structure meets the requirements.

[0047] Stability analysis focuses on the ability of a structure to maintain its original equilibrium state under a specific load to prevent the structure from becoming unstable overall or locally. Possible instability modes, such as overall overturning and local buckling, are identified based on the structure type and node connection method. Instability factors are added to the structural calculation model, and second-order effect analysis (such as P-Δ effect) is performed to more accurately evaluate the stability of the structure. The critical load when the structure becomes unstable is calculated through software, and compared with the design load to evaluate whether the stability of the structure meets the requirements.

[0048] Seismic performance analysis aims to evaluate the response and ability of building structures to resist damage under earthquakes, ensure that the structure can maintain overall stability during earthquakes, and reduce casualties and property losses. According to the earthquake intensity, site category and geological conditions of the building, select appropriate seismic motion parameters (such as acceleration time history curve) as input for seismic analysis. The software calculates the modal parameters such as the natural vibration period, vibration mode and damping ratio of the structure to understand the dynamic characteristics of the structure. Apply the seismic motion input to the structural calculation model, perform time history analysis, and obtain the response parameters such as displacement, velocity, acceleration and internal force of the structure under earthquake. According to the structural response parameters and failure criteria, evaluate whether the seismic performance of the structure meets the requirements of the specification. Focus on the ductility, energy dissipation capacity and reinforcement measures of the weak links of the structure.

[0049] In some embodiments, the process of determining the carbon emission compliance level includes: first predicting the carbon emission prediction values ​​corresponding to different stages of the target building, and then comparing the current relevant carbon emission requirements to determine the carbon emission compliance level.

[0050] Among them, the methods for predicting carbon emissions during the construction process are as follows: Figure 2 As shown, including: S201: Obtain building data and relevant carbon emission requirement data of a target building.

[0051] In this application, "building data" refers to building-related data, including but not limited to detailed data such as the age of the building, building structure, building scale, building life, building location, building type, and building style. The age of the building may include the time when the building was designed, the time when construction started, etc. Different eras have different design and construction methods, and the impact on carbon emissions is also different. The scale of the building may include the size of various types of building spaces. The building type may include commercial, residential, industrial and other building uses. The architectural style may include new Chinese, European, Italian and other styles.

[0052] Relevant carbon emission requirement data refers to the relevant data on normative requirements related to carbon emissions during the entire life cycle of a building. The promulgation and implementation of relevant specifications will also have varying degrees of impact on carbon emissions.

[0053] The target building may be in various stages of planning, design, construction, use, maintenance, demolition, etc. Therefore, its building data and related carbon emission requirements data may be complete or incomplete. Reasonable supplementation can be made for incomplete data.

[0054] These data can provide a data basis for subsequent carbon emission forecasts.

[0055] S202: Input the building data and relevant carbon emission requirement data into a shared input layer of a trained carbon emission phased prediction model.

[0056] S203: Outputting preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building in multiple independent output layers of the carbon emissions staged prediction model respectively.

[0057] In this application, the model basis of the carbon emission phased prediction model is the Long Short-Term Memory (LSTM) model. A main LSTM network is constructed, and the input layer of the network receives common features, namely building data and related carbon emission requirement data, which are related to each stage of the building life cycle. From this shared input layer, multiple independent LSTM sub-networks can be branched out, each of which corresponds to a specific carbon emission prediction stage, such as design, construction, use, maintenance, demolition, etc. The independent LSTM sub-network contains independent hidden layers and independent output layers, allowing the sub-network corresponding to each stage to independently learn its unique time-dependent pattern and output the carbon emission prediction value of the corresponding stage. According to the prediction target, a corresponding activation function is set for each output layer. During training, the loss functions of all sub-networks can be combined into a total loss, and then the entire model is optimized by back propagation. This architecture helps the model learn common features across stages, while making predictions in the specific context of each stage, ensuring the independence and accuracy of the predictions.

[0058] Among them, historical carbon emission data includes carbon emission data at different stages of the building’s life cycle.

[0059] The model is characterized by a main network and multiple sub-networks. The main network is responsible for receiving input data, while each sub-network corresponds to a stage in the building's life cycle and independently outputs a preliminary prediction of carbon emissions at that stage. This structure enables the model to consider the carbon emission characteristics of each stage of the building in a more detailed manner.

[0060] S204: Based on the preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building, the predicted correlation between the carbon emission data of each stage is calculated.

[0061] After obtaining the preliminary predicted values ​​of carbon emissions at each stage of the target building in multiple independent output layers of the model, the predicted correlation between these preliminary predicted values ​​is further calculated. This correlation reflects the degree of mutual influence between carbon emissions at each stage reflected in the carbon emission prediction data, and is an important indicator for evaluating the rationality of the prediction results.

[0062] The degree of association can be calculated using a correlation coefficient (such as the Pearson correlation coefficient).

[0063] S205: If the predicted correlation is not within the preset correlation range, the preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building are adjusted until the predicted correlation between the carbon emission data of each stage reaches within the preset correlation range, thereby obtaining the predicted values ​​of carbon emissions corresponding to different stages of the target building.

[0064] The preset correlation range is the correlation degree between different stages of the building determined based on the analysis of historical carbon emission data. Even for the same two stages, different types of buildings may correspond to different correlations. Therefore, the predicted correlation is more likely to be a range. When judging the predicted correlation, it can be judged whether it falls into the corresponding range, and the preset correlation range can be further subdivided according to the type of building to judge whether it falls into the range of the corresponding type of building.

[0065] If the calculated prediction correlation is not within the preset correlation range, it means that there may be logical inconsistency or impracticability between the preliminary prediction values. At this time, the preliminary prediction values ​​need to be adjusted until the prediction correlation between each stage reaches the preset range. This step ensures the accuracy and reliability of the prediction results.

[0066] After the above adjustments, the predicted carbon emissions of the target building at each stage are finally obtained. These predicted values ​​not only take into account the characteristics of the building itself, but also fully reflect the correlation between the various stages, so they have higher practical value.

[0067] Through the method of this embodiment, a carbon emission phased prediction model is constructed, the main LSTM is used to input building data and related carbon emission requirement data, and independent LSTM sub-networks are used to output the carbon emission prediction values ​​of the corresponding stages respectively, allowing the sub-network corresponding to each stage to independently learn its unique time-dependent pattern, and make more accurate predictions for each stage. At the same time, based on the correlation between different stages (preset correlation), the data of each stage is adjusted, so that the carbon emission prediction data not only meets the time-dependent characteristics of a single stage, but also meets the special correlation between stages, thereby improving the prediction accuracy of carbon emission data. It can provide strong support for carbon emission management and emission reduction strategy formulation.

[0068] For each candidate design, the process of scoring the design based on the structural safety level and the carbon emission compliance level can follow the following steps. For the structural safety level, it can be divided into different levels according to relevant building codes and standards, and each level can be assigned a corresponding score. For the carbon emission compliance level, different compliance levels can also be set according to environmental protection requirements and standards, and corresponding scores can be assigned. In the design scoring, the importance of the structural safety level and the carbon emission compliance level may be different. Therefore, different weights can be assigned to these two aspects to reflect their relative importance in the overall score. The allocation of weights can be adjusted according to the specific requirements and goals of the project. For example, if the project pays special attention to environmental performance, a higher weight can be given to the carbon emission compliance level. For each candidate design, find the corresponding score in the scoring criteria according to its structural safety level and carbon emission compliance level. Then, multiply these two scores by their respective weights to obtain a weighted score. Finally, add the weighted scores to obtain the total score of the candidate design. After calculating the total scores of all candidate designs, they can be compared. The candidate design with the highest total score will be regarded as the solution that achieves the best balance between structural safety and environmental performance.

[0069] In some implementations, if the predicted correlation is not within the preset correlation range, the preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building are adjusted until the predicted correlation between the carbon emission data of each stage reaches within the preset correlation range, and the predicted values ​​of carbon emissions corresponding to different stages of the target building are obtained, including: for the predicted correlation between the carbon emission data of any two stages, calculating the deviation direction and deviation value of the predicted correlation and the preset correlation between the carbon emission data of the two stages; determining the deviation stage according to the deviation direction and deviation value corresponding to the carbon emission data of each two stages; adjusting the preliminary predicted values ​​of carbon emissions in the deviation stage until the predicted correlation between the carbon emission data of each stage reaches within the preset correlation range, and obtaining the predicted values ​​of carbon emissions corresponding to different stages of the target building.

[0070] The deviation direction refers to whether the predicted correlation is higher or lower than the preset value, and the deviation value is the specific difference between them.

[0071] If the predicted correlation of a certain stage is inconsistent with the deviation direction and / or deviation value of the preset value, then this stage is marked as a "deviation stage". For the deviation stage, adjust the parameters of the model or use other methods to correct the carbon emission prediction value. This process may need to be iterated until the predicted correlation of all stages is within the preset acceptable range.

[0072] When determining the deviation stage, specific analysis can be performed for each stage. If the deviation values ​​of the predicted correlations corresponding to the carbon emission data of this stage and the carbon emission data of at least two other stages relative to the preset correlations are greater than the preset values, and the deviation directions are consistent, then this stage is determined as the deviation stage.

[0073] For example, if the deviation between the predicted correlation 1 and the preset correlation 1 between stage 1 and stage 2 is greater than the preset value, the deviation direction is positive; the deviation between the predicted correlation 2 and the preset correlation 2 between stage 1 and stage 3 is greater than the preset value, the deviation direction is positive; then stage 1 can be determined as a deviation stage. After adjusting stage 1, the predicted correlation 1 and the predicted correlation 2 can basically be corrected to an acceptable range.

[0074] like Figure 3 As shown in the figure, the training process of the carbon emission phased prediction model includes: S301: Acquire building data of historical buildings, historical relevant carbon emission requirement data and corresponding historical carbon emission data.

[0075] Collect architectural data of historical buildings, relevant carbon emission requirements data and corresponding historical carbon emission data. These data constitute the basic data set for model training.

[0076] Organize the corresponding historical carbon emission data to ensure the accuracy and completeness of the data. For example, pre-process the collected data, including data cleaning, missing value filling, outlier processing, etc.

[0077] S302: Analyze the characteristics of the historical carbon emission data, determine the time step, and construct carbon emission time series sample data and carbon emission time series test data at different stages of the entire life cycle of the historical building.

[0078] The characteristics of historical carbon emission data are analyzed to determine the appropriate time step, and then the carbon emission time series sample data and test data are constructed accordingly. The key to this step is to extract the time series characteristics that can reflect the law of carbon emission changes through reasonable data processing.

[0079] The characteristics of historical carbon emission data can be deeply analyzed, such as seasonality, periodicity, trend, etc. According to the characteristics of the data, the appropriate time step is determined to construct a time series sample. The carbon emission time series sample data and carbon emission time series test data of different stages of the life cycle of historical buildings are constructed. Ensure that the sample data can fully reflect the carbon emissions of buildings at different stages.

[0080] S303: Input the carbon emission time series sample data of the historical building at different stages of its life cycle into the constructed long-short term memory model, and perform iterative training until the set stop condition is reached.

[0081] The constructed time series sample data is input into the long short-term memory model for iterative training. Through continuous iterative optimization, the model can learn the implicit rules in the historical data.

[0082] The LSTM model has good performance in processing time series data.

[0083] First, initialize the model parameters, including weights, biases, etc. During the training process, use appropriate loss functions (such as mean square error MSE) and optimization algorithms (such as Adam, RMSprop, etc.) to optimize the model parameters. Set appropriate stop conditions, such as the number of training rounds, the rate at which the loss value decreases, etc. When the stop condition is reached, stop training.

[0084] S304: Input the historical building data and historical relevant carbon emission requirement data in the carbon emission time series test data into the iteratively trained long short-term memory model, and compare the output results with the historical carbon emission data in the carbon emission time series test data set to verify the training effect.

[0085] The trained model is validated using test data to evaluate its prediction performance. If the prediction effect does not meet expectations, iterative training and validation are repeated until the model performance reaches a satisfactory level.

[0086] The historical building data and historical carbon emission requirement data in the carbon emission time series test data are input into the iteratively trained LSTM model to obtain the prediction results. The prediction results are compared with the historical carbon emission data in the carbon emission time series test data set, and the evaluation indicators (such as accuracy, recall rate, F1 value, etc.) are calculated to verify the training effect of the model. According to the verification results, the model is tuned, including adjusting the model structure, parameters, optimization algorithm, etc.

[0087] S305: Repeat the iterative training and verification steps until the expected training effect is achieved, and a trained carbon emission prediction model is obtained.

[0088] Repeat the steps of model training and verification until the expected training effect is achieved. When the performance of the model meets the preset requirements, the model is determined as the final carbon emission phased prediction model. The trained model is deployed to the actual application scenario to predict the carbon emissions of the building at different stages of its life cycle. Based on the prediction results, scientific basis and suggestions are provided for energy saving and emission reduction of the building.

[0089] The above-mentioned analysis of the characteristics of the historical carbon emission data, determination of the time step, and construction of the carbon emission time series sample data and carbon emission time series test data at different stages of the life cycle of the historical building include: for the carbon emission data of each stage corresponding to each historical building, setting a time window, sampling the carbon emission data of the stage, the initial size of the time window is the time span of the stage, and the step size of the time window is the size of the time window; for the sampled data in the time window, calculating its range and variance value, if the range value is greater than the first preset threshold, and / or the variance value is greater than the second preset threshold, then retain all the data in the time window, reduce the size of the time window, and sample again, repeat this step until the range value of the data in all time windows is not greater than the first preset threshold, and the variance value is not greater than the second preset threshold; determine the time step according to the size of each time window at this time, determine the carbon emission data corresponding to each time step according to the mean value of the data in each time window, and construct the carbon emission time series sample data and carbon emission time series test data at different stages of the life cycle of the historical building.

[0090] For each stage of the life cycle of each historical building, set an initial time window. The initial size of this time window can be set to the time span of the stage. Set the step size of the time window, which can be set as a proportion of the initial size of the time window, such as 100%, 50% or 25%.

[0091] Sample carbon emission data in each time window. This can be achieved by selecting all data points in the time window or selecting representative data points. Calculate the range (difference between the maximum and minimum values) and variance of the sampled data in each time window. These two indicators can reflect the volatility and dispersion of the data.

[0092] If the range of the data in the time window is greater than the first preset threshold, or the variance is greater than the second preset threshold, it is considered that the data in the time window fluctuates too much or is too discrete, and contains many time series data features that need to be retained as much as possible. At this time, it is necessary to reduce the size of the time window, and resample and analyze the characteristics. Repeat this step until the data in all time windows meet the preset threshold conditions. At this point, the data in each window tends to be stable and homogenous, with fewer features, and the data volume can be compressed by selecting specific values ​​for representation.

[0093] Based on the size of each time window finally determined, the time step can be calculated. The time step is the time interval that the model considers each time when processing time series data. The choice of time step should be based on the characteristics of the data and the needs of the model. A shorter time step can capture more details, but may make the model too complicated; while a longer time step may ignore some important details.

[0094] According to the mean or other statistics (such as median, mode, etc.) of the data in each time window, the carbon emission data corresponding to each time step is determined. These data are arranged in chronological order to construct the carbon emission time series sample data of different stages of the life cycle of historical buildings. At the same time, a part of the original data is divided as carbon emission time series test data to verify the generalization ability and prediction accuracy of the model. That is, the data in a window is compressed into one data, which greatly reduces the amount of sample data, making the model training more efficient, and at the same time, the sample features are not lost, which does not affect the training effect of the model.

[0095] If the carbon emission data in certain stages are too scarce or unbalanced, data enhancement techniques can be used to generate more training samples to improve the stability and robustness of the model. For example, interpolation, resampling, noise addition and other methods can be used to enhance the training data.

[0096] In order to improve the training efficiency and prediction accuracy of the model, the time series data can be normalized or standardized. This can eliminate the dimensional differences and order of magnitude differences between different data, making the model easier to learn and predict.

[0097] Through the above steps, high-quality and representative carbon emission time series sample data and test data can be constructed to provide strong support for subsequent model training and verification.

[0098] Specifically, the above-mentioned reduction of the size of the time window includes: if the range value of the sampling data in any two adjacent time windows is greater than a first preset threshold, and / or the variance value is greater than a second preset threshold, then calculating the distance between the two maximum values ​​in the two adjacent time windows; if the time distance between the two maximum values ​​is less than or equal to the window size, then reducing the time window by 1 / 3.

[0099] For each stage of the life cycle of each historical building, an initial time window is set. The initial size of this time window is usually set to the time span of the stage.

[0100] Calculate the range (the difference between the maximum and minimum values) and variance of the data in each time window.

[0101] Check the data characteristics of adjacent time windows. If there are any two adjacent time windows, the range value of the sampled data within them is greater than the first preset threshold, or the variance value is greater than the second preset threshold, then perform the following steps.

[0102] Calculate the time distance between the maximum values ​​(maxima) in the two adjacent time windows. If the time distance between the two maximum values ​​is less than or equal to the current time window size, reduce the size of the time window by 1 / 3. This means that the new size of the time window will be 2 / 3 of the original size.

[0103] Repeat the above steps to continue checking the data characteristics within the new time window until all preset conditions are met.

[0104] This process can ensure the stability of time series data and adaptively adjust the time window size when the data fluctuates greatly, so as to more accurately capture the changing trend of carbon emissions. At the same time, this adaptive method can also reduce the impact of outliers on model training to a certain extent.

[0105] Optionally, the method also includes: after each stage of the target building is completed, obtaining the energy consumption data of the stage, and determining the actual carbon emission data of the stage based on the energy consumption data of the stage; comparing the actual carbon emission data of the stage with the carbon emission predicted value of the stage, if the data difference is greater than a third preset threshold, adjusting the carbon emission predicted value of the next stage based on the data difference and the preset correlation, and determining the corrected carbon emission predicted value.

[0106] When a certain stage of the target building is completed, the actual carbon emission data is compared with the predicted value, and the predicted value of the subsequent stage is adjusted according to the difference, which can improve the accuracy and practicality of the model. Specifically, the actual carbon emission data is collected first. After each stage of the target building is completed, the actual carbon emission data of that stage is collected. In the process, the actual energy consumption data of that stage can be recorded and counted. For example, electricity consumption data is recorded through an electric meter; gas consumption data is recorded through a gas meter, and so on. Then, based on certain statistical methods, the corresponding actual carbon emission data is calculated through energy consumption data. In some specific implementation methods, the existing statistical algorithm can be used for calculation, and the actual carbon emission data is calculated by averaging the calculation results of multiple algorithms or taking the median and taking the extreme value. The existing statistical algorithm can also be improved to calculate the actual carbon emission data with a new algorithm. It should be noted that no matter which algorithm is used, in order to ensure the consistency of the prediction results with the actual calculation data, the algorithm related to the carbon emission staged prediction model should be guaranteed as much as possible.

[0107] In addition, the data type of these data should be the same as the data type relied on when the model is predicted to ensure the accuracy of the comparison. Then compare the predicted value with the actual value. Compare the collected actual carbon emission data with the carbon emission predicted value of the corresponding stage. The difference or difference percentage between the two can be calculated. Further, determine whether the data difference is significant. A third preset threshold can be set to determine whether the difference between the actual data and the predicted value is significant. This threshold can be set according to the actual situation and model performance, usually a percentage or an absolute value. If the data difference is greater than the third preset threshold, it is considered that there is a significant difference between the predicted value and the actual value and adjustment is required. The cause of the difference can also be analyzed. After determining that the predicted value needs to be adjusted, analyze the reasons for the difference. This may include errors in the model itself, inaccurate input data, changes in certain factors in the actual construction stage, etc. Considering these causes of error, the predicted value can be further adjusted based on data differences and correlation. According to the difference between the actual data and the predicted value and the preset correlation, the carbon emission predicted value for the next stage is adjusted. The specific adjustment method can be to directly add a correction factor, or to indirectly affect the predicted value by adjusting the parameters of the model. The adjusted predicted value is the carbon emission corrected predicted value. This value will serve as the basis for carbon emission forecasts in subsequent stages and will be used to guide building operations and management. Throughout the life cycle of the building, actual carbon emission data will be continuously collected and compared with the revised forecast value. Based on the comparison results, the forecast model will be continuously iterated and adjusted to improve the accuracy and adaptability of the model.

[0108] Through this process, the effectiveness and reliability of the carbon emission prediction model in practical applications can be ensured, providing strong support for building energy conservation and emission reduction.

[0109] In some embodiments, the prediction results need to be compared with certain specific emission standards. In the event that the emission standards may be exceeded, the building data of a specific stage may need to be adjusted and re-predicted.

[0110] like Figure 4 As shown, the present application also provides a building structure analysis device 400, comprising: The acquisition module 401 is used to acquire the building structure design data corresponding to the multiple candidate designs of the target building; the building structure design data includes the building structure type, the structure material type, the structure material size, the structure material usage, and the structure node data; An analysis module 402 is used to analyze the structural safety level and carbon emission compliance level of the target building according to the building structure design data for each candidate design; A design scoring module 403 is used to score each candidate design based on the structural safety level and the carbon emission compliance level; The design scheme determining module 404 is used to determine the candidate design with the highest design score as the final architectural structure design scheme of the target building.

[0111] The analysis module 402 is specifically used for: Determine the load type and corresponding load requirement of the target building according to the building type and building location; Conduct structural strength analysis based on building structure type, structural material type, structural material size, structural material usage and structural node data; Analyze the structural strength grade of the target building according to the structural strength and the load requirements corresponding to different load types; Combined with the instability factors, the second-order effect analysis is carried out, and the stability analysis is carried out to determine the critical load when the structure is unstable; Analyze the stability level of the target building according to the critical load when the structure is unstable and the load requirements corresponding to different load types; Determine the earthquake intensity and geological conditions of the target building location based on the building location; Analyzing the seismic resistance level of the target building according to the earthquake intensity and geological conditions; The structural safety grade of the target building is determined according to the structural strength grade, stability grade and earthquake resistance grade of the target building.

[0112] The analysis module 402 is specifically used for: Acquire the building data and relevant carbon emission requirement data of the target building; the building data includes building age data, building structure data, building scale data, building life data, building location data, and building type data; Input the building data and relevant carbon emission requirement data into the shared input layer of the trained carbon emission phased prediction model; the carbon emission phased prediction model is trained based on historical building data, historical relevant carbon emission requirement data and corresponding historical carbon emission data; the historical carbon emission data includes carbon emission data at different stages of the building's life cycle; the carbon emission phased prediction model includes a main network and multiple sub-networks, the main network includes a shared input layer, and each of the sub-networks includes an independent hidden layer and an independent output layer, each corresponding to a stage in the life cycle; Outputting preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building in multiple independent output layers of the carbon emissions staged prediction model respectively; Based on the preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building, the predicted correlation between the carbon emission data of each stage is calculated; If the predicted correlation is not within the preset correlation range, the preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building are adjusted until the predicted correlation between the carbon emission data of each stage reaches within the preset correlation range, thereby obtaining the predicted values ​​of carbon emissions corresponding to different stages of the target building.

[0113] Optionally, the analysis module 402, when the predicted correlation is not within a preset correlation range, adjusts the preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building until the predicted correlation between the carbon emission data of each stage reaches within a preset correlation range, and obtains the predicted values ​​of carbon emissions corresponding to different stages of the target building, specifically for: For the predicted correlation between carbon emission data of any two stages, calculate the deviation direction and deviation value of the predicted correlation and the preset correlation between the carbon emission data of the two stages; Determine the deviation stage according to the deviation direction and deviation value corresponding to the carbon emission data of each two stages; The preliminary predicted value of carbon emissions in the deviation stage is adjusted until the predicted correlation between the carbon emission data of each stage reaches a preset correlation range, thereby obtaining the predicted values ​​of carbon emissions corresponding to different stages of the target building.

[0114] Optionally, a training module 405 is also included, which is used to: Obtain the building data of historical buildings, historical relevant carbon emission requirements data and corresponding historical carbon emission data; Analyze the characteristics of the historical carbon emission data, determine the time step, and construct carbon emission time series sample data and carbon emission time series test data at different stages of the historical building's life cycle; The carbon emission time series sample data of the historical building at different stages of its life cycle are input into the constructed long-short term memory model for iterative training until the set stop condition is reached; Input the historical building data and historical relevant carbon emission requirement data in the carbon emission time series test data into the iteratively trained long short-term memory model, and compare the output results with the historical carbon emission data in the carbon emission time series test data set to verify the training effect; Repeat the iterative training and validation steps until the expected training effect is achieved and a trained carbon emission prediction model is obtained.

[0115] Optionally, when analyzing the characteristics of the historical carbon emission data, determining the time step, and constructing the carbon emission time series sample data and carbon emission time series test data at different stages of the life cycle of the historical building, the training module 405 is specifically used to: For each stage of carbon emission data corresponding to each historical building, a time window is set to sample the carbon emission data of the stage, the initial size of the time window is the time span of the stage, and the step size of the time window is the size of the time window; For the sampled data in the time window, calculate the range value and the variance value thereof. If the range value is greater than a first preset threshold value, and / or the variance value is greater than a second preset threshold value, retain all the data in the time window, reduce the size of the time window, perform re-sampling, and repeat this step until the range value of the data in all time windows is not greater than the first preset threshold value, and the variance value is not greater than the second preset threshold value; The time step is determined according to the size of each time window at this time, and the carbon emission data corresponding to each time step is determined according to the mean of the data in each time window, so as to construct the carbon emission time series sample data and carbon emission time series test data of the historical building at different stages of its life cycle.

[0116] Optionally, when reducing the size of the time window, the training module 405 is specifically used to: If the range value of the sampled data in any two adjacent time windows is greater than the first preset threshold, and / or the variance value is greater than the second preset threshold, then the distance between the two maximum values ​​in the two adjacent time windows is calculated; If the time distance between the two maximum values ​​is less than or equal to the window size, the time window is reduced by 1 / 3.

[0117] Optionally, when determining the deviation stage according to the deviation direction and deviation value corresponding to the carbon emission data of each two stages, the analysis module 402 is specifically used to: For each stage, if the deviation value of the predicted correlation between the carbon emission data of this stage and the carbon emission data of at least two other stages relative to the preset correlation is greater than the preset value, and the deviation direction is consistent with the deviation direction between the preset correlation degrees corresponding to the carbon emission data of at least two other stages, then this stage is determined as a deviation stage.

[0118] Optionally, the device further includes a prediction value correction module 406, which is used to: After each stage of the target building is completed, the energy consumption data of the stage is obtained, and the actual carbon emission data of the stage is determined based on the energy consumption data of the stage; the actual carbon emission data of the stage is compared with the carbon emission forecast value of the stage, and if the data difference is greater than a third preset threshold, the carbon emission forecast value of the next stage is adjusted based on the data difference and the preset correlation to determine the carbon emission corrected forecast value.

[0119] The device of this embodiment can be used to execute the method of any of the above embodiments. Its implementation principle and technical effects are similar and will not be described in detail here.

[0120] Figure 5 A schematic diagram of the structure of an electronic device provided in one embodiment of the present application is shown in FIG. Figure 5 As shown, the electronic device 500 of this embodiment may include: a memory 501 and a processor 502 .

[0121] The memory 501 stores a computer program that can be loaded by the processor 502 and execute the method in the above embodiment.

[0122] The processor 502 and the memory 501 are connected, for example, via a bus.

[0123] Optionally, the electronic device 500 may further include a transceiver. It should be noted that in actual applications, the number of transceivers is not limited to one, and the structure of the electronic device 500 does not constitute a limitation on the embodiments of the present application.

[0124] The processor 502 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor 502 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0125] The bus may include a path to transmit information between the above components. The bus may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0126] The memory 501 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0127] The memory 501 is used to store the application code for executing the solution of the present application, and the execution is controlled by the processor 502. The processor 502 is used to execute the application code stored in the memory 501 to implement the contents shown in the above method embodiment.

[0128] The electronic devices include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), and fixed terminals such as digital TVs, desktop computers, etc. It can also be a server, etc. Figure 5 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0129] The electronic device of this embodiment can be used to execute the method of any of the above embodiments, and its implementation principle and technical effect are similar, which will not be described in detail here.

[0130] The present application also provides a computer-readable storage medium storing a computer program that can be loaded by a processor and execute the method in the above embodiment.

[0131] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

Claims

1. A building structure analysis method, characterized in that: include: Obtaining architectural structure design data corresponding to multiple candidate designs of a target building; The building structure design data includes building structure type, structural material type, structural material size, structural material usage, and structural node data; For each candidate design, analyzing the structural safety level and carbon emission compliance level of the target building based on the building structure design data; For each candidate design, a design score is given to the candidate design based on the structural safety level and the carbon emission compliance level; Determine the candidate design with the highest design score as the final architectural structure design solution of the target building; Analyzing the structural safety level and carbon emission compliance level of the target building according to the building structure design data includes: Determine the load type and corresponding load requirement of the target building according to the building type and building location; Conduct structural strength analysis based on building structure type, structural material type, structural material size, structural material usage and structural node data; Analyze the structural strength grade of the target building according to the structural strength and the load requirements corresponding to different load types; Combined with the instability factors, the second-order effect analysis is carried out, and the stability analysis is carried out to determine the critical load when the structure is unstable; Analyze the stability level of the target building according to the critical load when the structure is unstable and the load requirements corresponding to different load types; Determine the earthquake intensity and geological conditions of the target building location based on the building location; Analyzing the seismic resistance level of the target building according to the earthquake intensity and geological conditions; The structural safety grade of the target building is determined according to the structural strength grade, stability grade and earthquake resistance grade of the target building.

2. The building structure analysis method according to claim 1, characterized in that: For each design, the structural safety level and carbon emission compliance level of the target building are analyzed according to the building structure design data, including: Acquire the building data and relevant carbon emission requirement data of the target building; the building data includes building age data, building structure data, building scale data, building life data, building location data, and building type data; Input the building data and relevant carbon emission requirement data into the shared input layer of the trained carbon emission phased prediction model; the carbon emission phased prediction model is trained based on historical building data, historical relevant carbon emission requirement data and corresponding historical carbon emission data; the historical carbon emission data includes carbon emission data at different stages of the building's life cycle; the carbon emission phased prediction model includes a main network and multiple sub-networks, the main network includes a shared input layer, and each of the sub-networks includes an independent hidden layer and an independent output layer, each corresponding to a stage in the life cycle; Outputting preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building in multiple independent output layers of the carbon emissions staged prediction model respectively; Based on the preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building, the predicted correlation between the carbon emission data of each stage is calculated; If the predicted correlation is not within the preset correlation range, the preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building are adjusted until the predicted correlation between the carbon emission data of each stage reaches within the preset correlation range, thereby obtaining the predicted values ​​of carbon emissions corresponding to different stages of the target building.

3. The method according to claim 2, characterized in that If the predicted correlation is not within the preset correlation range, the preliminary predicted values ​​of carbon emissions corresponding to different stages of the target building are adjusted until the predicted correlation between the carbon emission data of each stage reaches the preset correlation range, and the predicted values ​​of carbon emissions corresponding to different stages of the target building are obtained, including: For the predicted correlation between carbon emission data of any two stages, calculate the deviation direction and deviation value of the predicted correlation and the preset correlation between the carbon emission data of the two stages; Determine the deviation stage according to the deviation direction and deviation value corresponding to the carbon emission data of each two stages; The preliminary predicted value of carbon emissions in the deviation stage is adjusted until the predicted correlation between the carbon emission data of each stage reaches a preset correlation range, thereby obtaining the predicted values ​​of carbon emissions corresponding to different stages of the target building.

4. The method according to claim 2 or 3, characterized in that: The training process of the carbon emission phased prediction model includes: Obtain the building data of historical buildings, historical relevant carbon emission requirements data and corresponding historical carbon emission data; Analyze the characteristics of the historical carbon emission data, determine the time step, and construct carbon emission time series sample data and carbon emission time series test data at different stages of the historical building's life cycle; The carbon emission time series sample data of the historical building at different stages of its life cycle are input into the constructed long-short term memory model for iterative training until the set stop condition is reached; Input the historical building data and historical relevant carbon emission requirement data in the carbon emission time series test data into the iteratively trained long short-term memory model, and compare the output results with the historical carbon emission data in the carbon emission time series test data set to verify the training effect; Repeat the iterative training and validation steps until the expected training effect is achieved and a trained carbon emission prediction model is obtained.

5. The method according to claim 4, characterized in that The analyzing the characteristics of the historical carbon emission data, determining the time step, and constructing the carbon emission time series sample data and carbon emission time series test data at different stages of the life cycle of the historical building include: For each stage of carbon emission data corresponding to each historical building, a time window is set to sample the carbon emission data of the stage, the initial size of the time window is the time span of the stage, and the step size of the time window is the size of the time window; For the sampled data in the time window, calculate the range value and the variance value thereof. If the range value is greater than a first preset threshold value, and / or the variance value is greater than a second preset threshold value, retain all the data in the time window, reduce the size of the time window, perform re-sampling, and repeat this step until the range value of the data in all time windows is not greater than the first preset threshold value, and the variance value is not greater than the second preset threshold value; The time step is determined according to the size of each time window at this time, and the carbon emission data corresponding to each time step is determined according to the mean of the data in each time window, so as to construct the carbon emission time series sample data and carbon emission time series test data of the historical building at different stages of its life cycle.

6. The method according to claim 5, characterized in that The reducing the size of the time window includes: If the range value of the sampled data in any two adjacent time windows is greater than the first preset threshold, and / or the variance value is greater than the second preset threshold, then the distance between the two maximum values ​​in the two adjacent time windows is calculated; If the time distance between the two maximum values ​​is less than or equal to the window size, the time window is reduced by 1 / 3.

7. The method according to claim 3, characterized in that Determining the deviation stage according to the deviation direction and deviation value corresponding to the carbon emission data of each two stages includes: For each stage, if the deviation value of the predicted correlation between the carbon emission data of this stage and the carbon emission data of at least two other stages relative to the preset correlation is greater than the preset value, and the deviation direction is consistent with the deviation direction between the preset correlation degrees corresponding to the carbon emission data of at least two other stages, then this stage is determined as a deviation stage.

8. The method according to claim 6 or 7, characterized in that: The method further comprises: After each stage of the target building is completed, the energy consumption data of the stage is obtained, and the actual carbon emission data of the stage is determined based on the energy consumption data of the stage; The actual carbon emission data of the stage is compared with the carbon emission forecast value of the stage. If the data difference is greater than a third preset threshold, the carbon emission forecast value of the next stage is adjusted based on the data difference and the preset correlation to determine the carbon emission corrected forecast value.

9. A building structure analysis device, characterized in that: include: An acquisition module, used to acquire architectural structure design data corresponding to multiple candidate designs of a target building; The building structure design data includes building structure type, structural material type, structural material size, structural material usage, and structural node data; An analysis module, for analyzing the structural safety level and carbon emission compliance level of the target building according to the building structure design data for each candidate design; A design scoring module, for scoring each candidate design based on the structural safety level and the carbon emission compliance level; A design scheme determination module, used to determine the candidate design with the highest design score as the final building structure design scheme of the target building; The analysis module is specifically used to: determine the load type and corresponding load demand of the target building according to the building type and building location; Perform structural strength analysis based on building structure type, structural material type, structural material size, structural material usage and structural node data; analyze the structural strength grade of the target building according to the structural strength and load requirements corresponding to different load types; Combined with the instability factors, a second-order effect analysis is performed, and a stability analysis is performed to determine the critical load when the structure is unstable; according to the critical load when the structure is unstable and the load requirements corresponding to different load types, the stability level of the target building is analyzed; Determine the earthquake intensity and geological conditions of the target building location based on the building location; The seismic resistance level of the target building is analyzed according to the earthquake intensity and geological conditions; the structural safety level of the target building is determined according to the structural strength level, stability level and seismic resistance level of the target building.

10. An electronic device comprising: A memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute the method according to any one of claims 1 to 8.

Images