Wind-induced internal pressure and net pressure rapid prediction method and system for curved roof holing

Abstract

The invention relates to a method and system for rapidly predicting wind-induced internal pressure and net pressure under a curved roof holing, and the method comprises the following steps: S1, obtaining the pneumatic information of a large-span curved roof building with an irregular shape through a wind tunnel test, and rapidly mastering the geometric and topological characteristics of the building; s2, introducing an average impact factor # imgabs0 # and a pulsation impact factor # imgabs1 # to construct a ratio relationship between wind-induced internal pressure and hole external pressure under the holing of the curved roof, and establishing a multivariable coupling dimensionless prediction framework for rapid prediction of the wind-induced internal pressure and the net pressure; s3, establishing a dimensionless multi-parameter prediction model related to the influence factors (# imgabs2 # and # imgabs3 #), embedding the dimensionless multi-parameter prediction model into the prediction framework for rapidly predicting the wind-induced internal pressure and the net pressure in the step S2, and rapidly obtaining statistical values of the internal surface wind pressure and the net pressure of the building by combining the external surface wind pressure information; the internal pressure and net pressure rapid prediction method provides load reference for hole design and enclosure structure design of a large-span curved roof building, and has important scientific research and engineering significance.

Description

Technical Field

[0001] The present invention belongs to the technical field of urban resilience and disaster assessment, and relates to a method and system for quickly predicting wind-induced internal pressure and net pressure under an opening in a curved roof. Background Art

[0002] To ensure sufficient safety redundancy in building structures exposed to wind-induced disasters, numerous experimental and simulation studies have been conducted domestically and internationally on the wind load distribution of long-span curved roofs. Building wind loads are composed of both external and internal surface wind pressures. The external pressure load is directly related to the building's shape and can be quickly acquired through wind tunnel testing or CFD simulations. The internal pressure load is directly related to the distribution of building openings. Larger openings on the building surface primarily arise from destructive openings caused by wind-induced failure of the roof envelope. Because wind-induced failure of the envelope is influenced by multiple factors, such as structural construction, wind direction, and wind speed, the distribution of destructive openings is complex. Relying on wind tunnel testing or CFD simulations requires a large number of samples, significantly increasing the time and cost of the wind-resistant design phase. Therefore, it is necessary to establish a method for rapidly calculating the wind-induced internal pressure of curved roof openings based on parameter analysis to reduce the design cost of characterizing the net wind pressure distribution of long-span curved roofs exposed to wind-induced disasters.

[0003] Currently, several experimental research results on the internal pressure of building openings have been published both domestically and internationally, such as formulas for predicting the average, pulsating, and peak wind-induced internal pressures under openings on windward walls. The internal-to-external pressure ratio formula established using existing methods can quickly determine the statistical value of wind-induced internal pressure under openings on windward walls. However, this prediction method does not account for changes in wind direction, and its accuracy is extremely dependent on the selection of the external pressure at the opening. Traditional methods for selecting the external pressure at the opening use a characteristic point around the opening as the external pressure. For example, on the windward wall, since the opening on the windward wall is perpendicular to the incoming flow, the characteristic point is typically located at the intersection of the opening's symmetry axis and the upper edge, based on the geometric symmetry of the opening. However, due to the irregular shape of a curved roof, the relative position of the opening and the incoming flow changes continuously with the wind direction. The position of the characteristic point is difficult to determine based on the geometric characteristics of the opening. In addition, there is a large wind pressure gradient around the opening. Selecting the external pressure at different geometric positions will cause a large prediction deviation, making it difficult to accurately obtain the true wind-induced internal pressure under different opening conditions. Moreover, the existing prediction methods do not consider the differences in the correlation between the wind-induced internal pressure and the external pressure in different areas of the building. The calculation of the net wind pressure also has obvious deviations, resulting in a lack of scientific load input basis for assessing the wind-induced response safety of large-span curved roof structures under wind-induced disasters.

[0004] Therefore, it is necessary to propose a rapid prediction method for the wind-induced internal pressure and net pressure of openings in curved roofs. By establishing a selection principle for the external pressure of openings that is not affected by wind direction and geometric position and an estimation method for the correlation between internal and external pressures, accurate input and evaluation of wind loads on large-span curved roof structures under wind-induced disasters can be achieved. Summary of the Invention

[0005] In view of this, the present invention provides a method and system for quickly predicting wind-induced internal pressure and net pressure under openings in curved roofs in order to solve the problems of lack of a unified selection principle for external pressure at openings due to the complex distribution of openings in curved roofs, and unscientific and inaccurate load results for net pressure loads on large-span curved roofs due to lack of consideration of the differences in the correlation distribution of internal and external pressures.

[0006] In order to achieve the above object, the present invention provides the following technical solutions: A method for rapidly predicting wind-induced internal pressure and net pressure under an opening in a curved roof comprises the following steps: S1. Obtain aerodynamic information of irregularly shaped, large-span curved roof buildings through wind tunnel testing, and quickly grasp the geometric and topological characteristics of the buildings; S2. Introducing the average impact factor and pulsation impact factor Construct the ratio of wind-induced internal pressure to external pressure under openings in curved roofs, and establish a multivariate coupled dimensionless prediction framework for rapid prediction of wind-induced internal pressure and net pressure. S3. Establish the impact factor ( , ) is embedded in the prediction framework of the rapid prediction of wind-induced internal pressure and net pressure in step S2, and the statistical values ​​of wind pressure and net pressure on the inner surface of the building are quickly obtained in combination with the external surface wind pressure information.

[0007] Furthermore, the aerodynamic information of the building in step S1 includes information such as the wind pressure distribution on the outer surface of the building, the flow velocity and flow field, and the geometric and topological information of the building includes relational features such as the length, width and height of the building, the location of the opening, the size of the opening, the internal volume, and the relative position of the opening and the incoming flow.

[0008] Furthermore, the wind-induced internal pressure under the opening of the curved roof in step S2 is calculated by the newly defined opening external pressure and the influencing factor ( , ) is measured, and the net pressure under the opening of the curved roof is measured by the newly defined opening external pressure and the influence factor ( , ) and the overall external pressure distribution of the building.

[0009] Furthermore, step S2 specifically includes the following steps: S21. Based on the basic principle of the covariance integral method, the response process of wind-induced internal pressure can be equivalent to the process of the structure generating dynamic response under the external pressure load of the opening. The whole process has the characteristics of stability, ergodicity, and randomness. The structural dynamic system can be simplified as the superposition of multiple discrete linear systems. The structural dynamic response is the wind-induced internal pressure ( ) response can be equivalent to the unit external load, that is, the unit opening external pressure ( The product of the structural response caused by the actual external load. ) is simplified to a circle around the hole. N The pressure measuring points are superimposed as shown in formula (1). This redefines the selection method of the external pressure of the cave entrance. The joint distribution characteristics of multiple measuring points around the cave entrance are used to replace the traditional single measuring point selection principle: (1) in Indicates the first j External pressure at each measuring point; Indicates the j The attached area of ​​each measuring point; is the reference wind pressure; Introducing impact factors The wind-induced internal pressure response caused by the external pressure load of a unit opening is expressed. The relationship between the internal and external pressures is established based on the covariance integration method according to the force balance principle. Since the wind-induced internal pressure of the building opening is uniformly distributed along the interior space of the building, the wind-induced internal pressure acting on the corresponding attached area is expressed by a unified value, as shown in formula (2): (2) The external pressure measurement points around the opening are assumed to be evenly distributed along the edge of the opening, that is, the attached areas of each measurement point are the same; the wind-induced internal pressure ( ) is simplified to the superposition of the average internal pressure and the pulsating internal pressure. The average internal pressure ( ) and the newly defined average external pressure at the opening using the average influence factor Expressed as formula (3); the relationship between the pulsating internal pressure and the newly defined pulsating external pressure at the hole is expressed by the pulsating influence factor Indicates that the pulsating internal pressure is the internal pressure pulsation time course ( ) squared, the newly defined pulsating external pressure at the opening ( ) It is also necessary to calculate the pulsation time history of each point around the hole ( ) is the average of the product, as shown in formula (4): (3) (4) The internal pressure peak is the product of the peak factor and the internal pressure pulsation statistics. Since the airflow state of the building internal pressure mainly depends on the transmission of the external pressure at the opening, the Gaussian distribution characteristics of the wind-induced internal pressure are similar to those of the external pressure at the opening, and the peak factor is similar in size, the peak internal pressure ( ) prediction requires the introduction of both average and pulsation factors ( , ), as shown in formula (5): (5) In the formula , Indicates the first j and k Pulsating external pressure time history of each measuring point; , Indicates the j and k Peak factor of external pressure at each measuring point; S22, net pressure is equal to the difference between external pressure and internal pressure, which satisfies the linear system assumption of covariance integration method. The influence factors of external pressure and internal pressure are both 1, so the building surface z The net pressure statistics of the point are calculated according to formulas (6) to (8): (6) (7) (8) In the formula The exterior surface of the building z Average net pressure of the point; The exterior surface of the building z The average external pressure of the point; is the average net pressure; The exterior surface of the building z Net pressure pulsation value at the point; The exterior surface of the building z External pressure pulsation value of the point; is the net pressure pulsation value; The exterior surface of the building z The correlation coefficient between the external pressure and the internal pressure at the point; the product of the internal pressure peak factor and the pulsation value is calculated according to formula (5); Based on the covariance integral method, the impact factor ( , ), an approximate linear relationship between the newly defined external pressure at the opening and the wind-induced internal pressure was constructed. The two have a high linear correlation, so the correlation between the internal pressure and the external pressure on the building surface ( ) uses the newly defined external pressure of the opening and the external pressure of the building surface to measure the pulsation influence factor ( ) is substituted into formula (9): (9).

[0010] Furthermore, step S3 of constructing the multi-parameter prediction model specifically includes the following steps: S31、Impact Factor( , ) is affected by many factors, among which the internal volume of the building ( V 0), opening area ( A ), incoming wind speed ( ) into a dimensionless variable ,in a s is the speed of sound, which is 340m / s; the influence of the incoming flow turbulence is measured by the turbulence integral scale ( ) and the dimensionless ratio of the opening area ( ) Consider and prepare the data of building aerodynamic information under different wind directions and opening positions; S32, the prediction model uses a generalized neural network (GRNN) structure, and the output of the prediction model selects the average impact factor ( ) and pulsation impact factor ( ) are trained separately; the prediction model input parameters include the building shape coordinates, external surface wind pressure distribution, opening size, opening location, wind direction, incoming wind speed and wind field information obtained in step S1, and the input and output have a clear corresponding relationship; S33, GRNN neural network relies on smoothing factor Adjust the prediction effect and generalization ability of the model, and define the error index shown in formula (10) to evaluate the prediction accuracy of the model. This index represents the relative error of the prediction result and eliminates the influence of the value itself.

[0011] (10)

[0012] In the formula is the mean square error between the predicted result and the expected result; is the standard deviation of the expected value.

[0013] Furthermore, in step S31, the wind direction and opening position are difficult to describe with a single variable due to the differences between different buildings, so it is necessary to use data mining technology to capture the characteristics of building openings; in order to ensure that the prediction model has good generalization ability, it is necessary to construct reasonable training samples; first, single opening and multiple opening samples are divided into two categories according to the number of openings; multiple openings can be simplified as the superposition of multiple single openings, so the sample construction mainly considers the rationality of single opening samples.

[0014] Furthermore, in step S31, considering that the wind-induced internal pressure is mainly affected by the external pressure at the opening and has a low correlation with the external pressure at other locations, the opening classification ignores the differences between buildings and is divided into four categories: roof leading edge, roof top, roof trailing edge, and others.

[0015] Furthermore, in step S31 , the wind direction is divided into three categories: parallel, vertical, and oblique according to the relative position relationship between the incoming flow and the shape and orientation of the opening.

[0016] A system for rapidly predicting wind-induced internal pressure and net pressure under openings in curved roofs includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for rapidly predicting wind-induced internal pressure and net pressure under openings in curved roofs is implemented.

[0017] The rapid prediction method for wind-induced internal pressure and net pressure under openings in curved roofs is also applicable to cylindrical shell roofs with single or multiple openings.

[0018] The beneficial effects of the present invention are: The invention discloses a method for rapidly predicting wind-induced internal pressure and net pressure under curved roof openings, proposes a principle for selecting the external pressure at the openings under complex curved roof openings, and simplifies the wind-induced internal pressure under curved roof openings, which has many influencing factors, into average and pulsating influencing factors ( , ) model, and proposed a correlation measurement method for the internal and external pressure of the opening, by establishing the influencing factor ( , ) neural network prediction model to predict the wind-induced internal pressure and net pressure of openings, and achieve accurate prediction of the internal pressure mean, pulsation value and peak value; it also proposes a net wind pressure calculation method that considers the influence of the correlation between internal and external pressures, and establishes a rapid prediction process for net wind pressure under the situation of complex openings in large-span curved roofs, providing a scientific load input basis for evaluating the wind load of large-span curved roof structures under wind-induced disasters. The entire processing process is simple, efficient, and convenient for engineering designers to use.

[0019] Other advantages, objects, and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art upon examination of the following description or may be learned from practice of the present invention. The objects and other advantages of the present invention may be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings, in which: Figure 1 This is a flow chart of the method for rapidly predicting wind-induced internal pressure and net pressure under a curved roof opening according to the present invention; Figure 2 This is a flow chart for constructing a multi-parameter prediction model in step S3 of the present invention; Figure 3 Schematic diagram of the structure of the GRNN neural network in step S3 of the present invention; Figure 4 This is a diagram of a building model according to an embodiment of the present invention; Figure 5A diagram showing the building dimensions and wind direction of a building model according to an embodiment of the present invention; Figure 6 For the present invention Figure 5 Schematic diagram of a single opening at the corner of the middle roof; Figure 7 For the present invention Figure 4 Distribution diagram of wind-induced internal pressure and net pressure of building model; Figure 8 For the present invention Figure 4 Renderings of wind-induced internal pressure and net pressure prediction verification for the building model. DETAILED DESCRIPTION

[0021] The following describes the embodiments of the present invention through specific examples. Those skilled in the art will readily understand the other advantages and benefits of the present invention from the disclosure herein. The present invention may also be implemented or applied through various other specific embodiments, and the details in this specification may be modified or altered based on different viewpoints and applications without departing from the spirit of the present invention.

[0022] By conducting a small number of wind tunnel pressure tests, we can obtain information such as the wind pressure distribution on the outer surface of large-span curved roof buildings, building opening design, flow velocity and flow field, and quickly grasp the relationship characteristics such as the building opening position, opening size, internal volume, and the relative position of the opening and incoming flow. The established wind-induced internal pressure and net pressure prediction model is input, and the statistical values ​​of the wind pressure and net wind pressure on the inner surface of the building are quickly obtained in combination with the external surface wind pressure information, which provides a load reference for the opening design and enclosure structure design of large-span curved roof buildings, which has important scientific research and engineering significance.

[0023] like Figure 1 The basic principle of a method for rapid prediction of wind-induced internal pressure and net pressure under openings in curved roofs is shown in Figure 1 As shown, the specific steps include: S1. Through wind tunnel testing, obtain information such as wind pressure distribution on the exterior surface of irregularly shaped, large-span curved roof buildings, building opening design, and flow velocity and flow field. Quickly understand the relationship between building opening location, opening size, internal volume, and the relative position of the opening and incoming flow. S2. Introducing the average impact factor and pulsation impact factor The ratio of wind-induced internal pressure to the external pressure at the opening of a curved roof is constructed; a prediction framework for rapid prediction of wind-induced internal pressure and net pressure is established. The wind-induced internal pressure at the opening of a curved roof is calculated by the newly defined external pressure at the opening and the influencing factor ( , ) is measured; the net pressure under the opening of the curved roof is measured by the newly defined opening external pressure and the influence factor ( , ) and the overall external pressure distribution of the building; The key and core issue in the rapid prediction of wind-induced internal pressure under curved roof openings lies in determining the transmission relationship between the external and internal pressures at the opening. The simplest method is to select a characteristic point around the opening as the external pressure and establish a ratio between the internal and external pressures. However, for irregular curved roofs, the opening orientation varies irregularly with the incoming wind direction, making the position of the characteristic point difficult to determine. Even if a unified selection principle (such as the windward edge of the opening) is adopted, the position of the characteristic point will still change with changes in the opening position and wind direction. Establishing the ratio between internal and external pressures using this method is not only inefficient but also prone to deviations due to inaccurate positioning of the characteristic point, making it impractical for engineering applications. Therefore, the present invention introduces the covariance integration method, which is widely used in linear structural systems, and constructs a transmission relationship between internal and external pressures based on a new principle for selecting the external pressure of the opening.

[0024] Specifically, the selection principle of S21 and the external pressure of the opening is based on the covariance integration method. The basic principle is as follows: the external pressure of the opening ( ) is equivalent to the external load input, and the wind-induced internal pressure ( ) is equivalent to the structural response output. The response process of wind-induced internal pressure can be equivalent to the process of the structure generating dynamic response under the action of external loads. The whole process has the characteristics of stability, ergodicity and randomness. The structural dynamic system can be simplified as the superposition of multiple discrete linear systems. The structural dynamic response is wind-induced internal pressure ( ) response can be equivalent to the unit external load, that is, the unit opening external pressure ( ) multiplied by the actual external load.

[0025] Since the external pressure around the opening of the curved roof is unevenly distributed and there is a large wind pressure gradient in the oblique wind direction, the actual external pressure load at the opening ( ) can be simplified as a circle around the hole. N The pressure measuring points are superimposed as shown in formula (1). This redefines the selection method of the external pressure of the cave entrance. The joint distribution characteristics of multiple measuring points around the cave entrance are used to replace the traditional single measuring point selection principle: (1) in Indicates the first j External pressure at each measuring point; Indicates the j The attached area of ​​each measuring point; is the reference wind pressure.

[0026] Introducing impact factors The wind-induced internal pressure response caused by the external pressure load of a unit opening is expressed. The relationship between the internal and external pressures is established based on the covariance integration method according to the force balance principle. Since the wind-induced internal pressure of the building opening is uniformly distributed along the interior space of the building, the wind-induced internal pressure acting on the corresponding attached area can be expressed by a unified value, as shown in formula (2): (2) The external pressure measurement points around the opening are assumed to be evenly distributed along the edge of the opening, that is, the attached area of ​​each measurement point is the same. ) can be simplified as the superposition of the average internal pressure and the pulsating internal pressure. The average internal pressure ( ) and the newly defined average external pressure at the opening can be expressed using the average influence factor Expressed as formula (3); the relationship between the pulsating internal pressure and the newly defined pulsating external pressure at the hole can be expressed using the pulsation influence factor Indicates that the pulsating internal pressure is the internal pressure pulsation time course ( ) squared, the newly defined opening external pressure ( ) It is also necessary to calculate the pulsation time history of each point around the hole ( ) is the average of the product, as shown in formula (4): (3) (4) The internal pressure peak is the product of the peak factor and the internal pressure pulsation statistics. Since the airflow state of the building internal pressure mainly depends on the transmission of the external pressure at the opening, the Gaussian distribution characteristics of the wind-induced internal pressure are similar to those of the external pressure at the opening, and the peak factor is similar in size, the peak internal pressure ( ) prediction requires the introduction of both average and pulsation factors ( , ), as shown in formula (5): (5) In the formula , Indicates the first j and k Pulsating external pressure time history of each measuring point; , Indicates the j and k The external pressure peak factor of each measuring point.

[0027] Based on the principle of covariance integration method, the average impact factor is introduced and pulsation impact factor The researchers established a relationship between the wind-induced internal pressure and the external pressure at the opening of a curved roof. They also redefined the selection of the external pressure based on the area surrounding the opening, rather than a specific characteristic point. This approach ensures accurate definition of the external pressure under different wind directions while remaining unaffected by factors such as the opening's shape and location. This approach has important scientific and engineering implications for rapidly predicting wind-induced internal pressure in large-span curved roofs with complex openings.

[0028] S22. The key and core issue of the rapid prediction method of the net pressure under the opening of the curved roof is how to measure the correlation between the wind-induced internal pressure and the wind pressure on the outer surface of the building. The traditional method can only obtain the correlation coefficient of the internal and external pressures by simultaneously obtaining the internal and external pressure time series through wind tunnel tests. Once the position of the opening changes due to design or failure of the enclosing structure, the time and economic cost of this method will increase exponentially, and there are many disadvantages in its application to engineering design. The covariance integration method introduced in the present invention can not only construct a relatively stable internal and external pressure transmission relationship, but also be used to measure the magnitude of the correlation between the internal pressure and the wind pressure on the outer surface of the building.

[0029] The net pressure is equal to the difference between the external pressure and the internal pressure, which satisfies the linear system assumption of the covariance integral method. The influence factors of the external pressure and the internal pressure are both 1, so the building surface z The net pressure statistics of a point can be calculated according to formulas (6) to (8): (6) (7) (8) In the formula The exterior surface of the building z Average net pressure of the point; The exterior surface of the building z The average external pressure of the point; is the average net pressure; The exterior surface of the building z Net pressure pulsation value at the point; The exterior surface of the building z External pressure pulsation value of the point; is the net pressure pulsation value; The exterior surface of the building z The correlation coefficient between the external pressure and the internal pressure at the point; the product of the internal pressure peak factor and the pulsation value is calculated according to formula (5).

[0030] This paper redefines the selection principle of the external pressure of the cave entrance through the covariance integration method, based on the influencing factors ( , ) constructed a newly defined linear relationship between the external pressure at the opening and the wind-induced internal pressure. The two have a high linear correlation, so the correlation between the internal pressure and the external pressure on the building surface ( ) can be measured by the external pressure around the opening and the external pressure on the building surface, and the pulsation influence factor ( ) is substituted into formula (9): (9) Based on the principle of covariance integration method, a fast prediction framework for wind-induced internal pressure and net pressure is established. The wind-induced internal pressure under the opening of the curved roof is calculated by the newly defined opening external pressure and the influencing factor ( , ) is measured; the net pressure under the opening of the curved roof is measured by the newly defined opening external pressure and the influence factor ( , ) and the overall external pressure distribution of the building.

[0031] S3. Establish impact factor ( , ) is embedded in the prediction framework of the rapid prediction of wind-induced internal pressure and net pressure in step S2, and the wind-induced internal pressure and net pressure under the roof opening of the large-span curved roof building are quickly estimated in combination with the external surface wind pressure information.

[0032] The rapid prediction framework of wind-induced internal pressure and net pressure under openings in curved roofs based on the covariance integral method requires influencing factors ( , ) predictive model driven by Figure 2 The process shown establishes the impact factor ( , ) is embedded in the prediction framework to estimate the wind-induced internal pressure and net pressure under the roof openings of large-span curved roof buildings, providing a reference for structural load design and optimization.

[0033] Specifically, S31, impact factor ( , ) The establishment of the prediction model requires a large number of roof opening wind-induced internal pressure test samples. , ) is affected by many factors, among which the internal volume of the building ( V 0), opening area ( A ), incoming wind speed ( ) can be aggregated into a dimensionless variable ,in a s is the speed of sound, which is 340 m / s; the influence of the incoming flow turbulence can be measured by the turbulence integral scale ( ) and the dimensionless ratio of the opening area ( ) considerations. The above factors and influencing factors ( , ) relationships can mostly be represented by piecewise function curves. However, even if the above variables are the same, the influencing factor ( , ) value will still change with wind direction. When the opening position changes, such as the opening on the roof and the front edge of the roof, the influence factor ( , ) values ​​will also vary. Wind direction and opening location vary from building to building, making it difficult to describe them with a single variable. Therefore, data mining techniques are needed to capture the characteristics of building openings.

[0034] To ensure the prediction model has good generalization capabilities, it is necessary to construct a reasonable training sample. Based on the number of openings, single-opening and multiple-opening samples are first divided into two categories. Multiple-opening samples can be simplified as the superposition of multiple single openings, so sample construction primarily considers the rationality of single-opening samples. Considering that wind-induced internal pressure is primarily affected by external pressure at openings and has a low correlation with external pressure at other locations, the opening classification can ignore differences between buildings and be divided into four categories: roof leading edge, roof top, roof trailing edge, and other. Wind direction is divided into three categories: parallel, perpendicular, and oblique, based on the relative position of the incoming flow and the opening shape and orientation.

[0035] S32, the prediction model uses Figure 3 The generalized neural network (GRNN) structure shown in the figure is suitable for learning and prediction with less sample data. The network structure is as follows Figure 3 As shown, the output of the model selects the average impact factor ( ) and pulsation impact factor ( ) are trained separately; the input parameters include building shape coordinates, external surface wind pressure distribution, opening size, opening location, wind direction, incoming wind speed and wind field information, and there is a clear one-to-one correspondence between input and output; S33, GRNN neural network relies on smoothing factor Adjust the prediction effect and generalization ability of the model, and define the error index shown in formula (10) to evaluate the prediction accuracy of the model. This index represents the relative error of the prediction result and eliminates the influence of the value itself.

[0036] (10)

[0037] In the formula is the mean square error between the predicted result and the expected result; is the standard deviation of the expected value.

[0038] Example

[0039] 1. Data Acquisition

[0040] Taking a coal yard closure renovation project in China as an example, based on local meteorological data, a Class B geomorphic flow field was used, and the building model was Figure 4As shown in the figure, a small number of wind tunnel tests were conducted to obtain the distribution of external pressure on the building surface in various wind directions. The building size and wind direction are arranged as shown in the figure. Figure 5 The building model is 724 mm wide and 732 mm high, with a scale ratio of 1:250. Sixteen different wind directions are applied to the building model, ranging from 0° to 360°, with one direction every 22.5°. These wind directions are used to simulate the impact of wind from different directions on the building model and to obtain the external pressure distribution of the building model under different wind directions.

[0041] 2. Cave entrance information

[0042] The building model mainly considers the failure of the enclosure structure. Two test conditions are designed: a single opening at the roof corner and multiple openings at the roof air tower. The opening distribution under the single opening at the roof corner is as follows: Figure 6 As shown, the rectangular opening at the corner of the roof is 50×30mm.

[0043] To verify the reliability of the prediction method, the wind-induced internal and external pressures under the openings of the building model were collected through wind tunnel tests and compared with the predicted results.

[0044] 3. Characteristics of wind-induced internal pressure and net pressure distribution

[0045] Figure 7 Shows the peak pressure coefficient and dimensionless parameters under different conditions The relationship between the black, red, blue and green represent the peak pressure coefficients of external pressure (External), internal pressure (Internal), net pressure (Net) and closed net pressure (Net-enclosed). These data points reflect the peak pressure coefficients of the external pressure (External), internal pressure (Internal), net pressure (Net) and closed net pressure (Net-enclosed). The changing trend of different types of pressure coefficients under different values. The wind-induced internal pressure and net pressure on the windward wall are predicted based on the prediction model under the given values. It can be seen that the wind-induced internal pressure caused by the opening in the roof corner is relatively large. When superimposed on the windward wall, it causes a sharp increase in net pressure, with the load being approximately twice that of the closed state, posing a significant threat to the enclosing structure on the windward wall, especially the doors and windows.

[0046] 4. Prediction method accuracy

[0047] The test values ​​of the net pressure at a measuring point on the roof in various wind directions were selected from the prediction data and compared with the prediction values ​​derived by the method of the present invention. Figure 8As shown. The horizontal axis in the figure represents the measured peak net pressure (Measured peak net pressure), and the vertical axis represents the estimated peak net pressure (Estimating peak net pressure). The data points (black squares) in the figure represent the measured values ​​of a single roof opening. The red straight line in the figure is the best fit line obtained by linear fitting, with a slope of 1.00 and an intercept of 0.17. R 2 The value is 0.996. R 2 The value is very close to 1, indicating a very high linear correlation between the estimated and measured values. That is, the predictions from the estimation model are very close to the actual measured results. As can be seen from the figure, the predicted net pressure values ​​agree well with the experimental measurements, with an accuracy of over 90%, demonstrating the accuracy of the proposed method.

[0048] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solutions, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for rapidly predicting wind-induced internal pressure and net pressure under a curved roof opening, characterized in that: The following steps are involved: S1. Obtain aerodynamic information of irregularly shaped, large-span curved roof buildings through wind tunnel testing, and quickly grasp the building's geometric and topological information; S2. Introducing the average impact factor and pulsation impact factor Construct the ratio of wind-induced internal pressure to external pressure under openings in curved roofs, and establish a multivariate coupled dimensionless prediction framework for rapid prediction of wind-induced internal pressure and net pressure. S3. Establish the impact factor ( , ) is embedded in the prediction framework of the rapid prediction of wind-induced internal pressure and net pressure in step S2, and the statistical values ​​of wind pressure and net pressure on the inner surface of the building are quickly obtained in combination with the external surface wind pressure information.

2. The method for rapid prediction of wind-induced internal pressure and net pressure according to claim 1, characterized in that: The aerodynamic information of the building in step S1 includes the wind pressure distribution and flow velocity and flow field information on the building's outer surface. The geometric and topological information of the building includes the building's length, width, and height, the location of the building's openings, the opening size, the internal volume, and the relative position of the openings and the incoming flow.

3. The method for rapid prediction of wind-induced internal pressure and net pressure according to claim 1, wherein: In step S2, the wind-induced internal pressure under the opening of the curved roof is calculated by the newly defined opening external pressure and the influencing factor ( , ) is measured, and the net pressure under the opening of the curved roof is measured by the newly defined opening external pressure and the influence factor ( , ) and the overall external pressure distribution of the building.

4. The method for rapid prediction of wind-induced internal pressure and net pressure according to claim 2, wherein: Step S2 specifically includes the following steps: S21. Based on the basic principle of covariance integration method, the response process of wind-induced internal pressure is equivalent to the process of dynamic response of the structure under the external pressure load of the opening. The structural dynamic system is simplified to the superposition of multiple discrete linear systems. The structural dynamic response is the wind-induced internal pressure ( ) response is equivalent to the unit external load, that is, the unit opening external pressure ( ) is the product of the structural response caused by the actual external load, the actual external pressure load force of the opening ( ) is simplified to a circle around the hole. N The pressure measuring points are superimposed as shown in formula (1). This redefines the selection method of the external pressure of the cave entrance. The joint distribution characteristics of multiple measuring points around the cave entrance are used to replace the traditional single measuring point selection principle: (1) in Indicates the first j External pressure at each measuring point; Indicates the j The attached area of ​​each measuring point; is the reference wind pressure; Introducing impact factors The wind-induced internal pressure response caused by the external pressure load of a unit opening is expressed. The relationship between the internal and external pressures is established based on the covariance integration method according to the force balance principle. Since the wind-induced internal pressure of the building opening is uniformly distributed along the interior space of the building, the wind-induced internal pressure acting on the corresponding attached area is expressed by a unified value, as shown in formula (2): (2) The external pressure measurement points around the opening are assumed to be evenly distributed along the edge of the opening, that is, the attached areas of each measurement point are the same; the wind-induced internal pressure ( ) is simplified to the superposition of the average internal pressure and the pulsating internal pressure. The average internal pressure ( ) and the newly defined average external pressure at the opening using the average influence factor Expressed as formula (3); the relationship between the pulsating internal pressure and the newly defined pulsating external pressure at the hole is expressed by the pulsating influence factor Indicates that the pulsating internal pressure is the internal pressure pulsation time course ( ) squared, the newly defined pulsating external pressure at the opening ( ) It is also necessary to calculate the pulsation time history of each point around the hole ( ) is the average of the product, as shown in formula (4): (3) (4) The internal pressure peak is the product of the peak factor and the internal pressure pulsation statistics. Since the airflow state of the building internal pressure mainly depends on the transmission of the external pressure at the opening, the Gaussian distribution characteristics of the wind-induced internal pressure are similar to those of the external pressure at the opening, and the peak factor is similar in size, the peak internal pressure ( ) prediction requires the introduction of both average and pulsation factors ( , ), as shown in formula (5): (5) In the formula , Indicates the first j and k Pulsating external pressure time history of each measuring point; , Indicates the j and k Peak factor of external pressure at each measuring point; S22, net pressure is equal to the difference between external pressure and internal pressure, which satisfies the linear system assumption of covariance integration method. The influence factors of external pressure and internal pressure are both 1, so the building surface z The net pressure statistics of the point are calculated according to formulas (6) to (8): (6) (7) (8) In the formula The exterior surface of the building z Average net pressure of the point; The exterior surface of the building z The average external pressure of the point; is the average net pressure; The exterior surface of the building z Net pressure pulsation value at the point; The exterior surface of the building z External pressure pulsation value of the point; is the net pressure pulsation value; The exterior surface of the building z The correlation coefficient between the external pressure and the internal pressure at the point; the product of the internal pressure peak factor and the pulsation value is calculated according to formula (5); Based on the covariance integral method, the impact factor ( , ), an approximate linear relationship between the newly defined external pressure at the opening and the wind-induced internal pressure was constructed. The two have a high linear correlation, so the correlation between the internal pressure and the external pressure on the building surface ( ) uses the newly defined external pressure of the opening and the external pressure of the building surface to measure the pulsation influence factor ( ) is substituted into formula (9): (9)。 5. The method for rapid prediction of wind-induced internal pressure and net pressure according to claim 4, characterized in that: Step S3: The construction of the multi-parameter prediction model specifically includes the following steps: S31、Impact Factor( , ) is affected by many factors, among which the internal volume of the building ( V 0), opening area ( A ), incoming wind speed ( ) into a dimensionless variable ,in a s is the speed of sound, which is 340 m / s; The influence of the incoming flow turbulence is measured by the turbulence integral scale ( ) and the dimensionless ratio of the opening area ( ) Consider; prepare the data of building aerodynamic information under different wind directions and opening positions; S32, the prediction model uses a generalized neural network (GRNN) structure, and the output of the prediction model selects the average impact factor ( ) and pulsation impact factor ( ) are trained separately; the prediction model input parameters include the building shape coordinates, external surface wind pressure distribution, opening size, opening location, wind direction, incoming wind speed and wind field information obtained in step S1, and the input and output have a clear corresponding relationship; S33, GRNN neural network relies on smoothing factor Adjust the prediction effect and generalization ability of the model, and define the error index shown in formula (10) to evaluate the prediction accuracy of the model. This index represents the relative error of the prediction result and eliminates the influence of the value itself. (10) In the formula is the mean square error between the predicted result and the expected result; is the standard deviation of the expected value.

6. The method for rapid prediction of wind-induced internal pressure and net pressure according to claim 5, characterized in that: In step S31, the wind direction and opening position are difficult to describe with a single variable due to the differences between different buildings. Therefore, it is necessary to use data mining technology to capture the characteristics of building openings. In order to ensure that the prediction model has good generalization ability, it is necessary to construct reasonable training samples. First, single opening and multiple opening samples are divided into two categories according to the number of openings. Multiple openings can be simplified as the superposition of multiple single openings. Therefore, the sample construction only considers the rationality of single opening samples.

7. The method for rapid prediction of wind-induced internal pressure and net pressure according to claim 6, characterized in that: In step S31, considering that the wind-induced internal pressure is mainly affected by the external pressure at the opening and has a low correlation with the external pressure at other locations, the openings are divided into four categories: the leading edge of the roof, the top of the roof, the trailing edge of the roof, and others.

8. The method for rapid prediction of wind-induced internal pressure and net pressure according to claim 7, characterized in that: In step S31, the wind direction is divided into three categories: parallel, vertical, and oblique according to the relative position relationship between the incoming flow and the shape and direction of the hole.

9. A rapid prediction system for wind-induced internal pressure and net pressure under openings in curved roofs, characterized by: The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for rapidly predicting wind-induced internal pressure and net pressure under an opening in a curved roof as claimed in any one of claims 1 to 8 is implemented.

10. The method for rapid prediction of wind-induced internal pressure and net pressure according to any one of claims 1 to 8, characterized in that: Applicable to local single opening and multiple openings on curved roofs.

Images