Green low-carbon building curtain wall ecological effect evaluation system and method thereof

Abstract

The application discloses a kind of green low-carbon building curtain wall ecological effect evaluation system and method, it is related to ecological effect evaluation field, the method includes: the indoor temperature and outdoor temperature of the multiple aluminum structure points on building curtain wall are collected, curtain wall temperature difference field is constructed;The wind pressure field of building curtain wall is obtained, the temperature difference influence analysis of the multiple aluminum structure points is carried out, and temperature difference influence parameter field is obtained;According to the curtain wall temperature difference field, low-carbon energy-saving prediction is carried out, and basic low-carbon energy-saving parameter is obtained;The curtain wall temperature difference field and temperature difference influence parameter field are verified, and verification coefficient is obtained, the basic low-carbon energy-saving parameter is corrected, and low-carbon energy-saving parameter is obtained as ecological effect evaluation result.The application is by a kind of green low-carbon building curtain wall ecological effect evaluation system and method, solves the problem that current curtain wall green low-carbon evaluation is not accurate, error is big.

Description

Technical Field

[0001] This invention relates to the field of ecological effect evaluation, specifically to a green and low-carbon building curtain wall ecological effect evaluation system and method. Background Technology

[0002] Ecological effect assessment can be used to understand and estimate the current status and changing trends of environmental quality. It is the theoretical basis for environmental quality biological monitoring and biological evaluation, and has theoretical and practical significance for pollution prevention and environmental protection.

[0003] Traditional building curtain wall structures have the problem of high energy consumption, and the current ecological assessment methods for curtain wall renovations are too simplistic and cannot adapt to the complex environment of modern building curtain walls, resulting in poor environmental adaptability and inaccurate ecological effect assessments.

[0004] Therefore, there is an urgent need for an evaluation system and method for the ecological effects of green and low-carbon building curtain walls, so as to conduct efficient and accurate evaluation of the ecological effects of green and low-carbon building curtain walls and improve the evaluation efficiency of the ecological effects of green and low-carbon building curtain walls. Summary of the Invention

[0005] This application provides a green and low-carbon building curtain wall ecological effect evaluation system and method, which addresses the problem of low accuracy in the ecological effect evaluation of traditional building curtain wall structures in the prior art.

[0006] In view of the above problems, this application provides an ecological effect evaluation system and method for green and low-carbon building curtain walls.

[0007] Firstly, this application provides an ecological effect evaluation system for green and low-carbon building curtain walls, the system comprising:

[0008] The curtain wall temperature difference field construction module is used to collect the indoor and outdoor temperatures of multiple aluminum structural points on the building curtain wall and construct the curtain wall temperature difference field.

[0009] The temperature difference influence analysis module is used to obtain the wind pressure field of the building curtain wall, perform temperature difference influence analysis on the multiple aluminum structure points, and obtain the temperature difference influence parameter field.

[0010] The low-carbon energy-saving prediction module is used to predict low-carbon energy saving based on the temperature difference field of the curtain wall and obtain basic low-carbon energy-saving parameters.

[0011] The ecological benefit evaluation module is used to verify the temperature difference field and temperature difference influence parameter field of the curtain wall, obtain the verification coefficient, and correct the basic low-carbon energy-saving parameters to obtain low-carbon energy-saving parameters as the ecological effect evaluation result.

[0012] Secondly, this application provides a method for evaluating the ecological effects of green and low-carbon building curtain walls, the method comprising:

[0013] The indoor and outdoor temperatures at multiple aluminum structural points on the building curtain wall are collected to construct a temperature difference field for the curtain wall.

[0014] Obtain the wind pressure field of the building curtain wall, perform temperature difference influence analysis on the multiple aluminum structure points, and obtain the temperature difference influence parameter field;

[0015] Based on the temperature difference field of the curtain wall, low-carbon energy saving prediction is performed to obtain basic low-carbon energy saving parameters;

[0016] The temperature difference field and temperature difference influence parameter field of the curtain wall are verified to obtain the verification coefficient. The basic low-carbon and energy-saving parameters are corrected to obtain low-carbon and energy-saving parameters, which are used as the ecological effect evaluation results.

[0017] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0018] This application proposes a green and low-carbon building curtain wall ecological effect evaluation system and method. Through machine learning analysis and verification, correction, and optimization measures, it conducts intelligent, highly adaptive, and practical ecological effect evaluations, improving the reliability of ecological effect prediction and evaluation, and enhancing the evaluation effect of building curtain walls. Simultaneously, by employing the green and low-carbon building curtain wall ecological effect evaluation system, a high-precision ecological effect evaluation system is constructed to achieve energy conservation and carbon reduction effects. Through intelligent ecological effect prediction, the ecological effects of building curtain walls are evaluated efficiently and accurately. Compared with existing technologies, this application improves the accuracy and reliability of the evaluation, achieves dynamically adaptive ecological effect evaluation, and solves the current problems of high energy consumption, low accuracy, and unsatisfactory ecological effect evaluation results in building curtain walls. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of a green and low-carbon building curtain wall ecological effect evaluation system.

[0021] Figure 2 This is a flowchart illustrating a method for evaluating the ecological effects of green and low-carbon building curtain walls.

[0022] In the attached diagram, the labels represent the following: Curtain wall temperature difference field construction module 11; Temperature difference influence analysis module 12; Low-carbon energy saving prediction module 13; Ecological benefit evaluation module 14. Detailed Implementation

[0023] This application provides a green and low-carbon building curtain wall ecological effect evaluation system and method, which addresses the problem of inaccurate evaluation of the ecological effects of green and low-carbon building curtain walls.

[0024] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0025] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0026] The present invention will now be described in detail with reference to the accompanying drawings.

[0027] Example 1, as Figure 1 As shown, this application discloses a green and low-carbon building curtain wall ecological effect evaluation system, the system comprising:

[0028] The curtain wall temperature difference field construction module 11 is used to collect the indoor and outdoor temperatures of multiple aluminum structural points on the building curtain wall and construct the curtain wall temperature difference field.

[0029] In this embodiment, the indoor and outdoor temperatures of multiple aluminum structural points on the building curtain wall are collected, and the ratio of the outdoor temperature to the indoor temperature is calculated as multiple curtain wall temperature differences to construct a curtain wall temperature difference field.

[0030] In one embodiment, the curtain wall temperature difference field construction module 11 is further used for:

[0031] Collect indoor and outdoor temperatures at multiple aluminum structural points on the building curtain wall;

[0032] Calculate the ratio of outdoor temperature to indoor temperature at each aluminum structure point to obtain multiple curtain wall temperature differences and construct a curtain wall temperature difference field.

[0033] Specifically, temperature acquisition devices such as temperature sensors are used to collect indoor and outdoor temperatures at multiple aluminum structural points on the building curtain wall. The temperature sensors can be directly fixed to the building curtain wall, or they can be moved and monitored individually. For example, a temperature acquisition sensor can wirelessly transmit temperature data to a receiving unit. By placing the temperature sensors at multiple aluminum structural points on the building curtain wall, the indoor and outdoor temperatures at multiple points can be detected. Meanwhile, considering that the reflective and heat-absorbing properties of the curtain wall are related to the material properties of the curtain wall itself, for structures exposed to the outdoors or structures under construction, the impact of solar radiation should be considered based on the structure's orientation and surface heat absorption properties.

[0034] For example, select a point A on the aluminum frame of a commercial building's curtain wall. The indoor temperature at point A is 20°C and the outdoor temperature is 35°C.

[0035] Furthermore, the ratio of outdoor temperature to indoor temperature at each aluminum structural point is calculated to obtain multiple curtain wall temperature differences, which are then used to construct a curtain wall temperature difference field. This ratio reflects the relative thermal gradient; a ratio greater than 1 indicates heat inflow, meaning a higher outdoor temperature, while a ratio less than 1 indicates heat outflow, meaning a higher indoor temperature. To avoid bias in absolute temperature values, the ratio is standardized to mitigate temperature effects, making comparisons more comparable across different seasons or climates. The aluminum structural points of the building curtain wall, including the frame and connection points, are key areas for heat exchange. Collecting the indoor and outdoor temperatures at these points reflects the overall thermal performance of the curtain wall. The curtain wall temperature difference field is not a temperature difference value, but a spatial field that can identify thermal bridging effects or areas with weak insulation.

[0036] For example, five points (AE) are selected on the aluminum frame of the curtain wall of a commercial building. The measured temperatures and curtain wall temperature differences are as follows: Point A: Indoor temperature 20°C, outdoor temperature 35°C, ratio = 35 / 20 = 1.75; Point B: Indoor temperature 22°C, outdoor temperature 33°C, ratio = 33 / 22 = 1.50; Point C: Indoor temperature 21°C, outdoor temperature 30°C, ratio = 30 / 21 ≈ 1.43; Point D: Indoor temperature 19°C, outdoor temperature 36°C, ratio = 36 / 19 ≈ 1.89; Point E: Indoor temperature 23°C, outdoor temperature 32°C, ratio = 32 / 23 ≈ 1.39. These ratios form a curtain wall temperature difference field. Point D has the highest ratio, indicating the best insulation effect at that point.

[0037] Compared to existing technologies, this application obtains multiple curtain wall temperature differences by collecting and calculating the ratio of outdoor temperature to indoor temperature at each aluminum structural point on the building curtain wall, and constructs a curtain wall temperature difference field. This helps to quantify the heat loss or gain of the curtain wall, providing a basis for energy-saving analysis. Furthermore, by utilizing the characteristics of the aluminum structural points, the curtain wall temperature difference field is constructed, and data is collected in areas with high incidence of thermal bridge effects, which is beneficial for energy-saving analysis.

[0038] The temperature difference influence analysis module 12 is used to obtain the wind pressure field of the building curtain wall, perform temperature difference influence analysis on the multiple aluminum structure points, and obtain the temperature difference influence parameter field.

[0039] In this embodiment of the application, the wind pressure parameters, such as wind pressure values, of multiple test points outside the building curtain wall are first tested. Then, a wind pressure field covering all locations of the curtain wall is generated. Next, the key wind pressure parameters of the aluminum structure points are indexed from the wind pressure field, and the temperature difference influence analysis is performed to obtain the temperature difference influence parameter field.

[0040] In one embodiment, the temperature difference influence analysis module 12 is further used for:

[0041] The wind pressure parameters at multiple test points outside the building's curtain wall were obtained during the test.

[0042] Based on the wind pressure parameters of multiple test points, interpolation is performed to obtain the wind pressure field, which includes the wind pressure parameters of all locations outside the building curtain wall.

[0043] Based on the wind pressure field, multiple key wind pressure parameters of the multiple aluminum structure points are indexed, and the temperature difference influence analysis of the multiple aluminum structure points is performed to obtain the temperature difference influence parameter field.

[0044] Specifically, multiple wind pressure parameter test points are constructed outside the building curtain wall. These test points may be distributed around the curtain wall, and wind pressure parameters at multiple test points are obtained using barometers.

[0045] For example, 15 test points are set on the outside of the curtain wall to measure wind pressure: p1 wind pressure = 20Pa, p2 = 15Pa, p3 = 25Pa, etc.

[0046] Furthermore, based on the wind pressure parameters from multiple test points, inverse distance weighted interpolation or Kriging (a geographical interpolation statistical method) is used for interpolation processing to transform discrete points into a continuous field. This ensures that the wind pressure data covers the entire curtain wall surface, including untested points, thereby obtaining a wind pressure field that includes wind pressure parameters at all locations outside the building's curtain wall. The obtained wind pressure field exhibits spatial variability, with higher wind pressure at corners and lower wind pressure at the center, and each pixel in the wind pressure field has a corresponding wind pressure value.

[0047] For example, inverse distance weighted interpolation is used to generate a wind pressure field image, where each pixel has a wind pressure value, such as 10-30 Pa.

[0048] Furthermore, based on the wind pressure field index of multiple key wind pressure parameters at multiple aluminum structure points, a temperature difference impact analysis is performed at these points to obtain a temperature difference impact parameter field. Through ensemble learning analysis, the temperature difference impact parameters are specifically defined as the variation in curtain wall temperature under the given wind pressure parameters relative to when the wind pressure parameter is zero. The key wind pressure parameter in the temperature difference impact analysis is the wind pressure value corresponding to each aluminum structure point. Simultaneously, the temperature difference impact analysis can analyze how wind pressure changes the temperature difference based on physical models or data-driven models. For example, increased wind pressure may increase air infiltration, leading to a higher temperature difference ratio. This analysis integrates environmental dynamics, making the ecological effect assessment more comprehensive. The obtained temperature difference impact parameter field can represent the correction factor for the temperature difference at each point's wind pressure, which can be used for subsequent adjustments in energy-saving predictions.

[0049] For example, test points p1-p5 are set on the exterior of the curtain wall to measure wind pressure. The wind pressure at point p1 is 20 Pa, at point p2 it is 15 Pa, at point p3 it is 25 Pa, etc. Interpolation is performed to generate a wind pressure field image. Then, the positions of points AE on the aluminum structure are indexed on the wind pressure field: point A has a wind pressure of 18 Pa, point B has 12 Pa, point C has 22 Pa, point D has 28 Pa, and point E has 14 Pa. When performing temperature difference influence analysis, the wind pressure values ​​of points AE on the aluminum structure are input into the pre-trained model, which can output temperature difference influence parameters.

[0050] Therefore, to obtain the temperature difference influence parameter field, it is first necessary to obtain the temperature difference influence analysis branch group and calculate the ratio of the outdoor temperature to the maximum outdoor temperature at each aluminum structure point, which is used as the temperature difference influence analysis coefficient. Based on this coefficient, the number of analysis branches used for each point is determined, the key wind pressure parameters are selected for input on the corresponding branches, and the average value of the branch outputs is taken to finally obtain the temperature difference influence parameter field.

[0051] Among them, based on multiple key wind pressure parameters at multiple aluminum structure points, a temperature difference influence analysis was conducted at multiple aluminum structure points to obtain the temperature difference influence parameter field, including:

[0052] Obtain the branch group for temperature difference influence analysis;

[0053] Calculate the ratio of the outdoor temperature to the maximum outdoor temperature at each aluminum structure point, and use it as a coefficient for multiple temperature difference influence analysis.

[0054] The number of temperature difference analysis branches is determined according to the multiple temperature difference influence analysis coefficients. The corresponding number of temperature difference influence analysis branches are selected respectively. The key wind pressure parameters are input, and the outputs of several temperature difference influence analysis branches are averaged to obtain multiple temperature difference influence parameters, which are used as the temperature difference influence parameter field.

[0055] Specifically, the first step is to obtain a temperature difference influence analysis branch group, that is, to construct multiple temperature difference influence analysis branches, thereby constructing a temperature difference influence analysis branch group. Each branch in this branch group can independently analyze the influence of wind pressure on temperature difference. Using the temperature difference influence analysis branch group can improve the stability of the ecological effect evaluation system and reduce uncertainty.

[0056] Furthermore, the ratio of the outdoor temperature to the maximum outdoor temperature at each aluminum structure point is calculated as multiple temperature difference influence analysis coefficients. These coefficients, based on the ratio of the outdoor temperature to the maximum outdoor temperature, reflect the heat load level at each aluminum structure point. A coefficient closer to 1 indicates a higher outdoor temperature, potentially requiring more branches for refined analysis. Through temperature difference influence coefficient analysis, adaptive temperature difference influence analysis is achieved, ensuring that points with high thermal risk are prioritized, thus contributing to low-carbon energy conservation and improving the effectiveness of ecological impact assessment.

[0057] For example, at point F of the aluminum structure, the outdoor temperature is 35°C, and the maximum outdoor temperature of the entire curtain wall is 40°C. The temperature difference influence analysis coefficient is 35 / 40 = 0.875. At point G of the aluminum structure, the outdoor temperature is 37°C, and the maximum outdoor temperature of the entire curtain wall is 40°C. The temperature difference influence analysis coefficient is 37 / 40 = 0.925. The temperature difference influence coefficient of point G is close to 1, indicating that the higher the outdoor temperature and the greater the heat inflow at this point, the more likely it is to require stronger insulation. Therefore, more branches of temperature difference analysis are used for prediction.

[0058] Furthermore, the number of temperature difference analysis branches is determined according to multiple temperature difference influence analysis coefficients. For example, the product of each temperature difference influence analysis coefficient and the total number of temperature difference influence analysis branches is calculated and then rounded to obtain the number of temperature difference analysis branches. Temperature difference influence analysis branches corresponding to the specified number of branches are then selected. Key wind pressure parameters are input, and the outputs of several temperature difference influence analysis branches are averaged to smooth prediction errors. Finally, the obtained multiple temperature difference influence parameters are used as the temperature difference influence parameter field. By averaging the output results of the temperature difference influence analysis branches, the stability of the temperature difference influence parameter field can be ensured, improving its robustness.

[0059] For example, the key wind pressure parameter of 18 Pa at point A is input into four temperature difference analysis branches M1, M2, M3, and M4. The output influence parameter is 0.6 for M1, 0.7 for M2, 0.65 for M3, and 0.75 for M4. The mean is (0.6 + 0.7 + 0.65 + 0.75) / 4 = 0.675. The mean of 0.675 is used as the temperature difference influence parameter for point A. All points are processed in the same way to form a temperature difference influence parameter field.

[0060] This application embodiment collects a set of sample wind pressure parameters for sample aluminum structure points from building curtain wall operation and maintenance data, as well as the variation of curtain wall temperature difference relative to zero wind pressure under different wind pressures. Multiple analysis branches are constructed based on machine learning, and each branch is randomly divided multiple times for training and testing. After all branches pass the tests, a group of temperature difference influence analysis branches can be obtained.

[0061] Among them, the branch group for obtaining the influence of temperature difference includes:

[0062] Based on the operation and maintenance data of the building curtain wall, a set of sample wind pressure parameters was collected for sample aluminum structure points, and the change range of curtain wall temperature difference under different sample wind pressure parameters relative to the curtain wall temperature difference when the wind pressure parameter is 0 was collected and marked as sample temperature difference influence parameters, thus obtaining a set of sample temperature difference influence parameters.

[0063] Based on machine learning, multiple branches for analyzing the effects of temperature difference are constructed.

[0064] The sample wind pressure parameter set and the sample temperature difference influence parameter set are randomly divided multiple times. Supervised training and testing are performed on multiple temperature difference influence analysis branches. After all tests are passed, the temperature difference influence analysis branch group is obtained.

[0065] Specifically, based on the operation and maintenance data of the building curtain wall, a set of sample wind pressure parameters was collected for sample aluminum structure points. The variation range of the curtain wall temperature difference under different sample wind pressure parameters relative to the temperature difference when the wind pressure parameter was 0 was also collected and labeled as sample temperature difference influence parameters, thus obtaining a set of sample temperature difference influence parameters. The operation and maintenance data includes historical wind pressure measurements and corresponding temperature difference changes. Simultaneously, the impact of wind pressure was assessed by quantifying the variation range, and labeled data suitable for supervised learning was provided.

[0066] For example, if the wind pressure is 10 Pa and the temperature difference ratio is reduced by 5% compared to when the wind pressure parameter is 0, then the sample temperature difference influence parameter is 0.05.

[0067] Furthermore, based on machine learning, a backpropagation (BP) neural network is used to construct multiple branches for analyzing the impact of temperature differences. Deep features are extracted from the sample wind pressure parameter set, resulting in N branches for temperature difference analysis, where N is a positive integer. The BP neural network is then used to accurately predict the temperature difference impact parameters through regression. The BP neural network model is a feedforward neural network trained using backpropagation and is commonly used to predict continuous values.

[0068] For example, with N=10, 10 branches for analyzing the effects of temperature difference are constructed, resulting in a group of branches for analyzing the effects of temperature difference.

[0069] Secondly, the sample wind pressure parameter set and the sample temperature difference influence parameter set are randomly divided multiple times. From the original sample wind pressure parameter set and sample temperature difference influence parameter set, N groups of samples are selected by Bootstrap random sampling with replacement. That is, the random sampling and replacement process is repeated N times to obtain N sample sets.

[0070] The sample temperature difference influence parameter set is used as the supervision target parameter for the branch. Based on control accuracy and adaptability, the randomly partitioned sample temperature difference influence parameter set is used as the supervision target. Simultaneously, weights are assigned to the branch predictors in the constructed temperature difference influence analysis branch. An initial learning rate and weights are set, and the error between the prediction result and the sample temperature difference influence parameter set is calculated using a gradient descent function. Weight adjustments and calculations are performed, iterating repeatedly with the goal of minimizing the prediction error until the error is <5%, thus obtaining the temperature difference influence analysis branch. Similarly, N temperature difference influence analysis branches are trained simultaneously on N sample sets, resulting in a temperature difference influence analysis branch cluster.

[0071] For example, overfitting is prevented by multiple random partitions, ensuring the model's generalization ability. A random sampling ratio of 0.8 is set, and 10 sample sets are drawn. Simultaneously, 10 branches for temperature difference influence analysis are trained, resulting in a temperature difference influence analysis branch cluster.

[0072] In this embodiment, the temperature difference influence parameter field of the building curtain wall is obtained through temperature difference influence analysis, ensuring a more stable temperature difference influence parameter field and improving its robustness. Furthermore, a branch group for temperature difference influence analysis is constructed using a decision tree model, ensuring the reliability and accuracy of the branch group and making it applicable to various curtain wall types and environmental conditions. Simultaneously, using the temperature difference influence analysis branch group can improve the stability of the ecological effect evaluation system and reduce model uncertainty.

[0073] The low-carbon energy-saving prediction module 13 is used to perform low-carbon energy-saving prediction based on the temperature difference field of the curtain wall and obtain basic low-carbon energy-saving parameters.

[0074] In this embodiment, the average curtain wall temperature difference is calculated based on the curtain wall temperature difference field, and then input into a low-carbon energy-saving predictor to predict the energy consumption for maintaining the building's indoor temperature under this curtain wall temperature difference field, outputting basic low-carbon energy-saving parameters. The low-carbon energy-saving predictor is trained using historical operation and maintenance data, which can serve as the basis for energy-saving evaluation.

[0075] In one embodiment, the low-carbon energy-saving prediction module 13 is further used for:

[0076] The average curtain wall temperature difference is calculated based on the curtain wall temperature difference field.

[0077] The average curtain wall temperature difference is input into the low-carbon energy-saving predictor, and the basic low-carbon energy-saving parameters are output. The low-carbon energy-saving predictor is trained using a set of sample average curtain wall temperature differences and a set of sample low-carbon energy-saving parameters from the building curtain wall operation and maintenance data. Each sample low-carbon energy-saving parameter includes the energy consumption for maintaining the building's internal temperature.

[0078] Specifically, based on the curtain wall temperature difference field, the average curtain wall temperature difference is calculated. This average temperature difference is the mean of the temperature differences at all aluminum structural points on the building's curtain wall. The temperature difference field is simplified using an average ratio, and its value represents the overall thermal performance of the curtain wall. A smaller value indicates better insulation and potentially lower energy consumption.

[0079] For example, five points are selected on the aluminum frame of the curtain wall of a commercial building. The temperature differences of the curtain wall are measured as follows: point A 1.75, point B 1.50, point C 1.43, point D 1.89, point E 1.39. The average ratio is (1.75 + 1.50 + 1.43 + 1.89 + 1.39) / 4 = 1.592. A larger value indicates better thermal insulation.

[0080] Furthermore, the average curtain wall temperature difference is input into the low-carbon energy-saving predictor, which outputs basic low-carbon energy-saving parameters. The low-carbon energy-saving predictor is trained using a set of sample average curtain wall temperature differences and a set of sample low-carbon energy-saving parameters from the building curtain wall operation and maintenance data. By learning the relationship between temperature difference and energy consumption, it predicts and outputs basic low-carbon energy-saving parameters. Each sample low-carbon energy-saving parameter includes the energy consumption for maintaining the building's internal temperature.

[0081] A low-carbon energy-saving predictor is constructed using a backpropagation (BP) neural network, a type of multilayer feedforward neural network trained using an error backpropagation algorithm. This network is capable of data classification and prediction.

[0082] The construction process of the low-carbon energy-saving predictor model is as follows:

[0083] Model Structure: The BP neural network structure includes the number of nodes in the input layer, hidden layers, and output layer. The network weights and biases are initialized simultaneously. Model Training: A low-carbon energy-saving predictor model is constructed using a 7:2:1 ratio of training set, validation set, and test set. The model uses the sample average curtain wall temperature difference set and the sample low-carbon energy-saving parameter set as inputs. A dynamic adjustment strategy is employed, using the Sigmoid function as the activation function and the MSE loss function to calculate the error between the sample average curtain wall temperature difference set and the sample low-carbon energy-saving parameter set. Forward propagation predicts the results, and backpropagation updates the parameters. Performance is evaluated using the validation set after each training epoch. A decrease in MSE loss of <1e on the training set for five consecutive epochs is required. -5When the MSE loss on the validation set stabilizes below 0.01, the model is considered converged. Training is stopped, and the final network parameters are saved, resulting in a supervised-trained low-carbon energy-saving predictor. Ten training cycles are performed using the validation set, with a test conducted after each cycle to ultimately obtain the low-carbon energy-saving predictor.

[0084] For example, the average curtain wall temperature difference field is 1.592. The low-carbon energy-saving predictor is trained on historical data: when the average ratio is 1.5, the weekly energy consumption is 900 kWh; when the average ratio is 1.6, the weekly energy consumption is 1000 kWh. With an input average ratio of 1.592, the predictor outputs a basic low-carbon energy-saving parameter of 1100 kWh per week.

[0085] Compared to existing technologies, this application's embodiments construct a low-carbon energy-saving predictor using a convolutional neural network. It obtains the average curtain wall temperature difference through the curtain wall temperature difference field, transforming the complex temperature difference field into a single indicator for rapid prediction, while also improving accuracy through historical data. Simultaneously, the average curtain wall temperature difference is input into the low-carbon energy-saving predictor to predict the energy consumption required to maintain the building's indoor temperature under the corresponding curtain wall temperature difference field, serving as the basis for energy-saving evaluation.

[0086] The ecological benefit evaluation module 14 is used to verify the temperature difference field and temperature difference influence parameter field of the curtain wall, obtain the verification coefficient, and correct the basic low-carbon energy-saving parameters to obtain low-carbon energy-saving parameters as the ecological effect evaluation result.

[0087] In this embodiment, multiple curtain wall temperature differences within the curtain wall temperature difference field and multiple temperature difference influence parameters within the temperature difference influence parameter field are sorted in descending order and ascending order, respectively, to obtain a curtain wall temperature difference sequence and a temperature difference influence parameter sequence. Then, the similarity of the sorted positions of multiple aluminum structure points within the curtain wall temperature difference sequence and the temperature difference influence parameter sequence is calculated to obtain a verification coefficient and a correction coefficient related to the verification coefficient. Correction calculations are performed based on the correction coefficients, and the final low-carbon energy-saving parameters are used as the ecological effect evaluation result.

[0088] In one embodiment, the ecological benefit evaluation module 14 is also used for:

[0089] The multiple curtain wall temperature differences in the curtain wall temperature difference field are sorted in descending order to obtain a curtain wall temperature difference sequence. The multiple temperature difference influence parameters in the temperature difference influence parameter field are sorted in ascending order to obtain a temperature difference influence parameter sequence.

[0090] Calculate the similarity of the sorting positions of multiple aluminum structure points within the curtain wall temperature difference sequence and temperature difference influence parameter sequence to obtain the verification coefficient;

[0091] Based on the verification coefficient, a correction coefficient is calculated, and the basic low-carbon and energy-saving parameters are corrected to obtain low-carbon and energy-saving parameters, which serve as the ecological effect evaluation results.

[0092] Specifically, multiple curtain wall temperature differences within the curtain wall temperature difference field are sorted in descending order to obtain a curtain wall temperature difference sequence. Similarly, multiple temperature difference influence parameters within the temperature difference influence parameter field are sorted in descending order to obtain a temperature difference influence parameter sequence. Points with larger temperature differences correspond to points with larger heat inflow and thus larger wind pressure influence parameters. However, in the curtain wall temperature difference sequence from largest to smallest, larger temperature differences rank higher, while in the temperature difference influence parameter sequence from smallest to largest, smaller influence parameters rank higher, demonstrating a negative correlation between the two sequences.

[0093] For example, consider three points located at different positions on the curtain wall: H, I, and J. The temperature difference at point H is 1.8, ranking 1st; at point I, it is 1.5, ranking 3rd; and at point J, it is 1.6, ranking 2nd. The temperature difference influence parameter field is as follows: at point H, the influence parameter is 0.9, ranking 3rd; at point I, it is 0.5, ranking 1st; and at point J, it is 0.7, ranking 2nd. The resulting curtain wall temperature difference sequence is [1.8, 1.6, 1.5]; the temperature difference influence parameter sequence is [0.5, 0.7, 0.9]. The ranking values ​​of the curtain wall temperature difference sequence and the temperature difference influence parameter sequence for points H, I, and J are [1, 3, 2] and [3, 1, 2], respectively.

[0094] Furthermore, the similarity of the ordered positions of multiple aluminum structural points within the curtain wall temperature difference sequence and the temperature difference influence parameter sequence is calculated to obtain the verification coefficient. The positional similarity of each aluminum structural point within the two sequences is calculated, and the mean of all positional similarities is used as the verification coefficient. A similarity value close to 1 indicates a high negative correlation, meaning the corresponding aluminum structural points within the two sequences have high positional similarity; a similarity value close to 0 indicates no correlation, meaning the corresponding aluminum structural points within the two sequences have low positional similarity. The verification coefficient ranges from 0 to 1 in absolute value.

[0095] For example, the similarity of the multiple sorted positions of points H, I, and J located at different positions on the curtain wall within the curtain wall temperature difference sequence and temperature difference influence parameter sequence is calculated. The similarity of the sorted positions of point H is [1 - (|1 - 3|) / 3] ≈ 0.33, the similarity of the sorted positions of point I is [1 - (|3 - 1|) / 3] ≈ 0.66, and the similarity of the sorted positions of point J is [1 - (|2 - 2|) / 3] = 0. The verification coefficient is (0.33 + 0.66 + 0) / 3 = 0.33. The verification coefficient is relatively small, indicating that the positional similarity of the aluminum structure points within the two sequences is low.

[0096] Furthermore, based on the verification coefficient, a correction coefficient is calculated, and the basic low-carbon energy-saving parameters are corrected accordingly. The resulting low-carbon energy-saving parameters are used as the ecological effect evaluation results. A higher verification coefficient, closer to 1, indicates a smaller correction degree; a lower verification coefficient, closer to 0, indicates a larger correction degree. A larger verification coefficient indicates better insulation performance and a smaller low-carbon energy-saving parameter. The correction coefficient is 1 minus the verification coefficient. The low-carbon energy-saving parameter is the sum of the basic low-carbon energy-saving parameter multiplied by 1 and the correction coefficient. This corrects the basic low-carbon energy-saving parameter, compensating for energy consumption increases caused by errors. The low-carbon energy-saving parameter reflects the magnitude of energy consumption; the smaller the value, the lower the energy consumption.

[0097] For example, the basic low-carbon energy saving parameter per week is 1100 kWh, the verification coefficient is 0.33, the correction coefficient is 1-0.33=0.67, and the low-carbon energy saving parameter per week is (1+0.67)×1100=1837 kWh.

[0098] Compared to existing technologies, the embodiments of this application calculate the correction coefficients of basic low-carbon energy-saving parameters to obtain the required correction degree for basic low-carbon energy saving, ensuring the consistency of the curtain wall temperature difference field and the temperature difference influence parameter field, and improving the reliability of prediction. Simultaneously, corrections are made through statistical verification, reducing the uncertainty in low-carbon energy-saving parameter predictions. This makes it applicable to curtain wall evaluation in dynamic environments, improving the adaptability and practicality of ecological effect evaluation.

[0099] The embodiments of this application, through the specific implementation methods described above, achieve the following technical effects:

[0100] This application utilizes a green and low-carbon building curtain wall ecological effect evaluation system. First, by collecting and calculating the ratio of outdoor temperature to indoor temperature at each aluminum structural point on the building curtain wall, multiple curtain wall temperature differences are obtained, constructing a curtain wall temperature difference field. This helps quantify the heat loss or gain of the curtain wall, providing a basis for energy-saving analysis. Utilizing the characteristics of the aluminum structural points, the curtain wall temperature difference field is constructed, with data collected in areas prone to thermal bridging, facilitating energy-saving analysis. Second, the wind pressure field of the building curtain wall, i.e., the spatial distribution of wind pressure, is obtained, and a decision tree model is used to construct a branch group for temperature difference influence analysis. This ensures the reliability and accuracy of the branch group, making it applicable to various curtain wall types and environmental conditions. Simultaneously, using the temperature difference influence analysis branch group improves the stability of the ecological effect evaluation system and reduces model uncertainty.

[0101] Furthermore, a low-carbon energy-saving predictor is constructed using a convolutional neural network. The average curtain wall temperature difference is obtained from the curtain wall temperature difference field, transforming the complex temperature difference field into a single indicator for rapid prediction. Historical data is used to improve accuracy. The average curtain wall temperature difference is then input into the predictor to forecast the energy consumption for maintaining indoor temperature under the corresponding curtain wall temperature difference field. This estimated energy consumption for maintaining indoor temperature serves as a basis for energy-saving assessment. Finally, by calculating correction coefficients for the basic low-carbon energy-saving parameters, the required correction level for basic low-carbon energy saving is obtained, ensuring the consistency between the curtain wall temperature difference field and the temperature difference influence parameter field, thus improving the reliability of the prediction. Statistical verification is also used for correction, reducing the uncertainty in low-carbon energy-saving parameter prediction. This method is applicable to curtain wall evaluation in dynamic environments, improving the adaptability and practicality of ecological effect assessment.

[0102] Compared to existing technologies, this application presents a green and low-carbon building curtain wall ecological effect evaluation system and an intelligent energy-saving effect assessment technology system. It solves the problems of high energy consumption and poor ecological effect evaluation in current building curtain wall ecological effect evaluations, systematically constructs a full-chain innovation system for green and low-carbon evaluation models, effectively fills the technological gap in the field of building curtain wall ecological effect evaluation, and forms an adaptive, efficient, and energy-saving system applicable to curtain wall evaluation under dynamic environments.

[0103] Example 2, as Figure 2 As shown, this application provides a method for evaluating the ecological effects of green and low-carbon building curtain walls, the method comprising:

[0104] S10: Collect indoor and outdoor temperatures at multiple aluminum structural points on the building curtain wall to construct a curtain wall temperature difference field;

[0105] Step S10 in the method provided in this application embodiment includes:

[0106] Collect indoor and outdoor temperatures at multiple aluminum structural points on the building curtain wall;

[0107] Calculate the ratio of outdoor temperature to indoor temperature at each aluminum structure point to obtain multiple curtain wall temperature differences and construct a curtain wall temperature difference field.

[0108] S20: Obtain the wind pressure field of the building curtain wall, perform temperature difference influence analysis on the multiple aluminum structure points, and obtain the temperature difference influence parameter field;

[0109] Step S20 in the method provided in this application embodiment includes:

[0110] The wind pressure parameters at multiple test points outside the building's curtain wall were obtained during the test.

[0111] Based on the wind pressure parameters of multiple test points, interpolation is performed to obtain the wind pressure field, which includes the wind pressure parameters of all locations outside the building curtain wall.

[0112] Based on the wind pressure field, multiple key wind pressure parameters of the multiple aluminum structure points are indexed, and the temperature difference influence analysis of the multiple aluminum structure points is performed to obtain the temperature difference influence parameter field.

[0113] Specifically, based on the wind pressure field, multiple key wind pressure parameters of the multiple aluminum structure points are indexed, and a temperature difference influence analysis of the multiple aluminum structure points is performed to obtain a temperature difference influence parameter field, including:

[0114] Obtain the branch group for temperature difference influence analysis;

[0115] Calculate the ratio of the outdoor temperature to the maximum outdoor temperature at each aluminum structure point, and use it as a coefficient for multiple temperature difference influence analysis.

[0116] The number of temperature difference analysis branches is determined according to the multiple temperature difference influence analysis coefficients. The corresponding number of temperature difference influence analysis branches are selected respectively. The key wind pressure parameters are input, and the outputs of several temperature difference influence analysis branches are averaged to obtain multiple temperature difference influence parameters, which are used as the temperature difference influence parameter field.

[0117] The acquisition of the temperature difference influence analysis branch group includes:

[0118] Based on the operation and maintenance data of the building curtain wall, a set of sample wind pressure parameters was collected for sample aluminum structure points, and the change range of curtain wall temperature difference under different sample wind pressure parameters relative to the curtain wall temperature difference when the wind pressure parameter is 0 was collected and marked as sample temperature difference influence parameters, thus obtaining a set of sample temperature difference influence parameters.

[0119] Based on machine learning, multiple branches for analyzing the effects of temperature difference are constructed.

[0120] The sample wind pressure parameter set and the sample temperature difference influence parameter set are randomly divided multiple times. Supervised training and testing are performed on multiple temperature difference influence analysis branches. After all tests are passed, the temperature difference influence analysis branch group is obtained.

[0121] S30: Based on the temperature difference field of the curtain wall, perform low-carbon energy saving prediction and obtain basic low-carbon energy saving parameters;

[0122] In this embodiment of the application, step S30 includes:

[0123] The average curtain wall temperature difference is calculated based on the curtain wall temperature difference field.

[0124] The average curtain wall temperature difference is input into the low-carbon energy-saving predictor, and the basic low-carbon energy-saving parameters are output. The low-carbon energy-saving predictor is trained using a set of sample average curtain wall temperature differences and a set of sample low-carbon energy-saving parameters from the building curtain wall operation and maintenance data. Each sample low-carbon energy-saving parameter includes the energy consumption for maintaining the building's internal temperature.

[0125] S40: Verify the temperature difference field and temperature difference influence parameter field of the curtain wall, obtain the verification coefficient, correct the basic low-carbon energy-saving parameters, and obtain the low-carbon energy-saving parameters as the ecological effect evaluation result.

[0126] In this embodiment of the application, S40 includes:

[0127] The multiple curtain wall temperature differences in the curtain wall temperature difference field are sorted in descending order to obtain a curtain wall temperature difference sequence. The multiple temperature difference influence parameters in the temperature difference influence parameter field are sorted in ascending order to obtain a temperature difference influence parameter sequence.

[0128] Calculate the similarity of the sorting positions of multiple aluminum structure points within the curtain wall temperature difference sequence and temperature difference influence parameter sequence to obtain the verification coefficient;

[0129] Based on the verification coefficient, a correction coefficient is calculated, and the basic low-carbon and energy-saving parameters are corrected to obtain low-carbon and energy-saving parameters, which serve as the ecological effect evaluation results.

[0130] The embodiments of this application, through the specific implementation methods described above, achieve the following technical effects:

[0131] This application first collects the indoor and outdoor temperatures at multiple aluminum structural points on the building curtain wall, calculates the ratio between the two, obtains multiple curtain wall temperature differences, and constructs a curtain wall temperature difference field. Second, it acquires the wind pressure field of the building curtain wall, performs temperature difference influence analysis, obtains a temperature difference influence parameter field, and uses the sample wind pressure parameter set and the variation range of the curtain wall temperature difference as the sample temperature difference influence parameter set. Based on the random forest machine algorithm, it constructs multiple temperature difference influence analysis branches, performs supervised training and testing, and obtains a temperature difference influence analysis branch group. By using multiple temperature difference influence analysis coefficients, selecting the temperature difference influence analysis branch input, and performing mean processing, a temperature difference influence parameter field is obtained. Next, based on the curtain wall temperature difference field, the average curtain wall temperature difference is obtained for low-carbon energy saving prediction. The average curtain wall temperature difference is input into the low-carbon energy saving predictor, and the basic low-carbon energy saving parameters are output. Finally, by different sortings of the curtain wall temperature difference sequence and the temperature difference influence parameter sequence, verification is performed. The similarity of the sorting positions of multiple aluminum structure points within the two sequences is calculated to obtain verification coefficients. The basic low-carbon energy saving parameters are then corrected to obtain the low-carbon energy saving parameters, which are used as the ecological effect evaluation result.

[0132] In summary, this application's embodiments utilize machine learning analysis and verification correction to develop an intelligent ecological effect evaluation technology for green and low-carbon building curtain walls. Through design optimization measures, a highly accurate, adaptive, and practical ecological effect evaluation system is obtained. By comprehensively considering environmental factors, the accuracy and reliability of the evaluation are improved, achieving dynamically adaptive ecological effect evaluation. This presents a green and low-carbon building curtain wall ecological effect evaluation system and method, solving the current problems of high energy consumption, low accuracy, and unsatisfactory ecological effect evaluation results in building curtain walls.

[0133] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0134] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A green and low-carbon building curtain wall ecological effect evaluation system, characterized in that, The system includes: The curtain wall temperature difference field construction module is used to collect indoor and outdoor temperatures at multiple aluminum structural points on the building curtain wall to construct the curtain wall temperature difference field, including: Collect indoor and outdoor temperatures at multiple aluminum structural points on the building curtain wall; Calculate the ratio of outdoor temperature to indoor temperature at each aluminum structure point to obtain multiple curtain wall temperature differences and construct a curtain wall temperature difference field; The temperature difference influence analysis module is used to acquire the wind pressure field of the building curtain wall, perform temperature difference influence analysis on the multiple aluminum structure points, and obtain the temperature difference influence parameter field, including: The wind pressure parameters at multiple test points outside the building's curtain wall were obtained during the test. Based on the wind pressure parameters of multiple test points, interpolation is performed to obtain the wind pressure field, which includes the wind pressure parameters of all locations outside the building curtain wall. Based on the wind pressure field, multiple key wind pressure parameters of the multiple aluminum structure points are indexed, and a temperature difference influence analysis of the multiple aluminum structure points is performed to obtain the temperature difference influence parameter field, including: Obtain the branch group for the analysis of the effect of temperature difference, including: Based on the operation and maintenance data of the building curtain wall, a set of sample wind pressure parameters was collected for sample aluminum structure points, and the change range of curtain wall temperature difference under different sample wind pressure parameters relative to the curtain wall temperature difference when the wind pressure parameter is 0 was collected and marked as sample temperature difference influence parameters, thus obtaining a set of sample temperature difference influence parameters. Based on machine learning, multiple branches for analyzing the effects of temperature difference are constructed. The sample wind pressure parameter set and the sample temperature difference influence parameter set are randomly divided multiple times. Supervised training and testing are performed on multiple temperature difference influence analysis branches. After all tests are passed, the temperature difference influence analysis branch group is obtained. Calculate the ratio of the outdoor temperature to the maximum outdoor temperature at each aluminum structure point, and use it as a coefficient for multiple temperature difference influence analysis. The number of multiple temperature difference analysis branches is determined according to the multiple temperature difference influence analysis coefficients. The corresponding number of temperature difference influence analysis branches are selected respectively. The key wind pressure parameters are input, and the outputs of several temperature difference influence analysis branches are averaged to obtain multiple temperature difference influence parameters, which are used as temperature difference influence parameter fields. The low-carbon energy-saving prediction module is used to predict low-carbon energy saving based on the temperature difference field of the curtain wall and obtain basic low-carbon energy-saving parameters. The ecological benefit evaluation module is used to verify the temperature difference field and temperature difference influence parameter field of the curtain wall, obtain the verification coefficient, and correct the basic low-carbon energy-saving parameters to obtain low-carbon energy-saving parameters as the ecological effect evaluation result.

2. The green and low-carbon building curtain wall ecological effect evaluation system according to claim 1, characterized in that, Based on the temperature difference field of the curtain wall, low-carbon energy saving prediction is performed to obtain basic low-carbon energy saving parameters, including: The average curtain wall temperature difference is calculated based on the curtain wall temperature difference field. The average curtain wall temperature difference is input into the low-carbon energy-saving predictor, and the basic low-carbon energy-saving parameters are output. The low-carbon energy-saving predictor is trained using a set of sample average curtain wall temperature differences and a set of sample low-carbon energy-saving parameters from the building curtain wall operation and maintenance data. Each sample low-carbon energy-saving parameter includes the energy consumption for maintaining the building's internal temperature.

3. The green and low-carbon building curtain wall ecological effect evaluation system according to claim 1, characterized in that, The temperature difference field and temperature difference influence parameter field of the curtain wall are verified to obtain verification coefficients. The basic low-carbon and energy-saving parameters are corrected to obtain low-carbon and energy-saving parameters, which serve as the ecological effect evaluation results, including: The multiple curtain wall temperature differences in the curtain wall temperature difference field are sorted in descending order to obtain a curtain wall temperature difference sequence. The multiple temperature difference influence parameters in the temperature difference influence parameter field are sorted in ascending order to obtain a temperature difference influence parameter sequence. Calculate the similarity of the sorting positions of multiple aluminum structure points within the curtain wall temperature difference sequence and temperature difference influence parameter sequence to obtain the verification coefficient; Based on the verification coefficient, a correction coefficient is calculated, and the basic low-carbon and energy-saving parameters are corrected to obtain low-carbon and energy-saving parameters, which serve as the ecological effect evaluation results.

4. A method for evaluating the ecological effects of green and low-carbon building curtain walls, characterized in that, The method, applied to the ecological effect evaluation system for green and low-carbon building curtain walls according to any one of claims 1-3, comprises: Indoor and outdoor temperatures were collected at multiple aluminum structural points on the building curtain wall to construct a curtain wall temperature difference field, including: Collect indoor and outdoor temperatures at multiple aluminum structural points on the building curtain wall; Calculate the ratio of outdoor temperature to indoor temperature at each aluminum structure point to obtain multiple curtain wall temperature differences and construct a curtain wall temperature difference field; Obtain the wind pressure field of the building curtain wall, perform temperature difference influence analysis on the multiple aluminum structure points, and obtain the temperature difference influence parameter field, including: The wind pressure parameters at multiple test points outside the building's curtain wall were obtained during the test. Based on the wind pressure parameters of multiple test points, interpolation is performed to obtain the wind pressure field, which includes the wind pressure parameters of all locations outside the building curtain wall. Based on the wind pressure field, multiple key wind pressure parameters of the multiple aluminum structure points are indexed, and a temperature difference influence analysis of the multiple aluminum structure points is performed to obtain the temperature difference influence parameter field, including: Obtain the branch group for the analysis of the effect of temperature difference, including: Based on the operation and maintenance data of the building curtain wall, a set of sample wind pressure parameters was collected for sample aluminum structure points, and the change range of curtain wall temperature difference under different sample wind pressure parameters relative to the curtain wall temperature difference when the wind pressure parameter is 0 was collected and marked as sample temperature difference influence parameters, thus obtaining a set of sample temperature difference influence parameters. Based on machine learning, multiple branches for analyzing the effects of temperature difference are constructed. The sample wind pressure parameter set and the sample temperature difference influence parameter set are randomly divided multiple times. Supervised training and testing are performed on multiple temperature difference influence analysis branches. After all tests are passed, the temperature difference influence analysis branch group is obtained. Calculate the ratio of the outdoor temperature to the maximum outdoor temperature at each aluminum structure point, and use it as a coefficient for multiple temperature difference influence analysis. The number of multiple temperature difference analysis branches is determined according to the multiple temperature difference influence analysis coefficients. The corresponding number of temperature difference influence analysis branches are selected respectively. The key wind pressure parameters are input, and the outputs of several temperature difference influence analysis branches are averaged to obtain multiple temperature difference influence parameters, which are used as temperature difference influence parameter fields. Based on the temperature difference field of the curtain wall, low-carbon energy saving prediction is performed to obtain basic low-carbon energy saving parameters; The temperature difference field and temperature difference influence parameter field of the curtain wall are verified to obtain the verification coefficient. The basic low-carbon and energy-saving parameters are corrected to obtain low-carbon and energy-saving parameters, which are used as the ecological effect evaluation results.

Images