Cooling tower surface high-risk area identification method based on wind tunnel pressure measurement test

Abstract

The invention provides a cooling tower surface high-risk area identification method based on a wind tunnel pressure measurement test, and aims to provide a cooling tower surface high-risk area identification method which is objective, quantitative and integrates average and fluctuating wind pressure effects. Comprising the steps of data acquisition, gridding, negative pressure area preliminary screening, comprehensive risk index calculation and result output. The core lies in the definition of a comprehensive risk index, the index combines an average local shape coefficient and a pulsation local shape coefficient, and the comprehensive risk index is formed through a mathematical function. According to the method, the defects that in the prior art, subjective experience is relied on, and fluctuating wind pressure contribution is neglected are overcome, so that automatic and precise recognition of the wind-induced high-risk area on the surface of the cooling tower is achieved, and a more scientific basis is provided for wind resistance design.

Description

Technical Field

[0001] This invention belongs to the field of wind engineering technology in civil engineering structures, specifically relating to a method for identifying high-risk areas in the wind-resistant design of large cooling tower structures. Background Technology

[0002] In the wind-resistant design of cooling towers, accurately identifying the weak points on their surface under wind loads is crucial. Existing technologies mainly focus on the following aspects:

[0003] Overall wind load analysis: Traditional methods mainly focus on the overall wind-induced response of the cooling tower, such as base bending moment and shear force, and design based on the overall shape coefficient provided by the code (Code for Design of Building Structures (GB 50009-2012)). This method ensures the overall stability of the structure, but cannot reveal the complex wind pressure distribution and potential damage risks on local surfaces.

[0004] Study on mean wind pressure distribution: With the development of wind tunnel testing and CFD (Computational Fluid Dynamics) technology, researchers have begun to map the mean wind pressure distribution cloud map (i.e., the mean local shape coefficient μ_s,mean distribution) on the surface of cooling towers. This method can visually display the area with the largest mean negative pressure (usually located on the leeward and crosswind sides) and is currently a more advanced analytical tool.

[0005] Extreme wind pressure estimation: Some studies use extreme wind pressure estimation methods based on the Hermite moment model, which estimate the extreme wind pressure under a certain guarantee rate by using average and fluctuating wind pressure, and use it to guide the design of building envelope.

[0006] The closest existing technical solution can be found in the literature on the method for analyzing the surface wind pressure distribution of cooling towers based on wind tunnel tests.

[0007] Implementation method description:

[0008] Data acquisition: In the boundary layer wind tunnel, a rigid pressure measurement model of the cooling tower was constructed according to the similarity criterion, and a large number of pressure measurement points were arranged on the surface of the model.

[0009] Operating condition simulation: Simulates various operating conditions such as different wind speeds (U), wind direction angles (θ), and different spacings (S) under multi-tower interference.

[0010] Data processing: Collect wind pressure time history data for each pressure measurement point, and calculate the average wind pressure coefficient (Cp_mean) and the root variance of the fluctuating wind pressure coefficient (Cp_rms) for each point.

[0011] Visualization analysis: The Cp_mean and Cp_rms data of discrete measurement points are transformed into continuous two-dimensional cloud maps through interpolation algorithms (such as Kriging, radial basis functions, etc.), and plotted on the unfolded surface of the cooling tower or a certain cross-section.

[0012] Result determination: By observing the cloud map, engineers subjectively judge which areas have a large average negative pressure (large negative Cp_mean value), and these areas are usually considered to be areas that need to be strengthened.

[0013] Examples of relevant literature:

[0014] The paper "Experimental Study on Wind Load Interference Effect of Twin-Tower Structure" (Liu Qingkuan, Ma Wenyong, Sheng Yongqing, et al. Engineering Mechanics, 2011, 28(S1):75-79) usually provides the distribution diagram of average wind pressure coefficient under different working conditions. "Wind Tunnel Test Study on Wind Load Characteristics of Ultra-Large Cooling Tower" (Li Pengfei, Zhao Lin, Ge Yaojun, et al. Engineering Mechanics, 2008, (06):60-67) mainly focuses on how to use synchronous pressure measurement wind tunnel test data to obtain the predicted extreme wind pressure distribution of the pulsating wind pressure distribution on the inner and outer surfaces.

[0015] The closest existing technical solution mentioned above has the following main drawbacks:

[0016] 1. Reliance on subjective experience and lack of objective quantitative standards: Determining whether an area is "high-risk" heavily depends on the engineer's experience. For areas with high average negative pressure but low pulsation, or moderate average negative pressure but extremely high pulsation, the lack of unified, quantitative judgment standards can easily lead to misjudgment or omission.

[0017] 2. Neglecting the contribution of fluctuating wind pressure: Wind-induced damage to structures (such as fatigue failure of the overlay layer and localized tearing) is often the result of the combined effects of mean wind pressure and fluctuating wind pressure. Existing methods mainly focus on the distribution of mean wind pressure and fail to incorporate the systematic impact of fluctuating wind pressure into the comprehensive evaluation system for high-risk areas.

[0018] 3. Inability to effectively identify "high-risk" areas: A region with a large but very stable average negative pressure, and a region with a slightly lower average negative pressure but experiencing violent fluctuations; the latter may be more destructive due to fatigue effects. Existing methods struggle to effectively identify this type of "hidden" high-risk area.

[0019] 4. Low efficiency and difficulty in handling comparisons of multiple operating conditions: When there are dozens or even hundreds of wind tunnel test conditions, manually interpreting cloud maps one by one is extremely inefficient and makes it difficult to conduct objective and quantitative comparative analysis between different operating conditions to determine the most unfavorable operating condition. Summary of the Invention

[0020] This invention aims to address the aforementioned shortcomings by providing an objective, quantitative, and comprehensive method for identifying high-risk areas on cooling tower surfaces based on averaging and the effect of pulsating wind pressure. The core technical problem this invention addresses is how to overcome the deficiencies of existing technologies that rely on subjective experience and ignore the contribution of pulsating wind pressure, thereby achieving automated and accurate identification of wind-induced high-risk areas on cooling tower surfaces and providing a more scientific basis for wind-resistant design.

[0021] The specific technical solution is as follows:

[0022] A method for identifying high-risk areas on the surface of a cooling tower based on wind tunnel pressure testing includes the following steps:

[0023] Step 1: Obtain discrete data through a wind tunnel test using a rigid model of the cooling tower;

[0024] Step 2: Use an interpolation function to grid the discrete data to obtain regular grid data;

[0025] Step 3: Using the aforementioned regular grid data, complete the initial screening of the negative pressure zone and the analysis of the connected regions to achieve quantitative zoning;

[0026] Step 4: Calculate the comprehensive risk index for each zone to complete the identification of high-risk areas; the comprehensive risk index is used to comprehensively evaluate the risk based on the static destructive potential energy, fatigue damage potential, and size of the negative pressure area to be analyzed.

[0027] Step 5: Visualize and output the results.

[0028] Furthermore, the discrete data includes the coordinates (X,Y) of all measuring points on the surface of the rigid model of the cooling tower under multiple operating conditions, the average wind pressure coefficient Cp_mean of all measuring points, and the root variance Cp_rms of the fluctuating wind pressure coefficient of all measuring points; the multiple operating conditions cover different incoming wind speeds U, wind direction angles θ, and spacing ratios S under multi-tower interference.

[0029] Furthermore, step two includes: using an interpolation function to interpolate the discrete data onto a predefined fine two-dimensional grid, generating matrix-form grid coordinates X_grid, Y_grid, mean wind pressure coefficient matrix Cp_mean_grid, and fluctuating wind pressure coefficient matrix Cp_rms_grid.

[0030] Furthermore, the initial screening of the negative pressure zone and the analysis of the connected regions include:

[0031] Set threshold: Use the set average wind pressure coefficient threshold to binarize the gridded average wind pressure coefficient;

[0032] Connected component labeling: Using the connected component analysis algorithm, the above-mentioned binarized negative pressure area is labeled to identify several independent negative pressure areas that are not connected to each other;

[0033] Regional feature extraction: For each independent connected region, calculate its area (Area), the average value of the average wind pressure coefficient (Mean_Cp), and the coordinates of the region's center (Centroid).

[0034] Furthermore, the comprehensive risk index R i It is expressed as follows:

[0035] R i =[α*|Mean_Cp_i|+β*Mean_Cp_rms_i]*(A_i / A_ref)γ

[0036] Where α and β are weighting coefficients, Mean_Cp_i is the average value of the average wind pressure coefficient of the i-th connected negative pressure region, Mean_Cp_rms_i is the average value of the root variance of the fluctuating wind pressure coefficient corresponding to the i-th connected negative pressure region extracted from the fluctuating wind pressure coefficient matrix Cp_rms_grid, A_i is the area of ​​the i-th connected negative pressure region; A_ref is the reference area used for dimensionless transformation; γ is the area index, when γ = 1, it indicates that the risk is linearly proportional to the area, when γ = 0.5, it indicates that the risk is proportional to the square root of the area.

[0037] Furthermore, α and β are determined based on the design strength and fatigue strength of the actual engineering cladding material, and their ratio relationship is as follows:

[0038] α / β=k*([σ] / Δσ f )

[0039] Where k is an adjustment coefficient, [σ] is the allowable material stress determined according to the concrete structure design code; Δσ f The allowable value of the fatigue stress amplitude of steel bars is determined according to the concrete structure design standard;

[0040] Furthermore, the weighting coefficients α and β can also be determined using any of the following methods:

[0041] Depending on the engineering design focus, the assumptions are preset based on experience, provided that α+β=1 is met; when the focus is on assessing the risk of static failure, α>β; when the focus is on assessing the risk of fatigue failure, α<β.

[0042] The reliability analysis and calibration are used to ensure that the identification results of high-risk areas match the target reliability index.

[0043] Furthermore, the comprehensive risk index R i It can also be expressed as follows:

[0044] Product form: R i =|Mean_Cp_i|*Mean_Cp_rms_i;

[0045] Based on extreme value statistics: R i =|Mean_Cp_i|+K*Mean_Cp_rms_i, where K is the peak factor.

[0046] Furthermore, the method for identifying high-risk areas is as follows: a comprehensive risk threshold is set, and all areas exceeding the comprehensive risk threshold are determined to be high-risk areas;

[0047] The method for identifying high-risk areas is as follows: For all areas, R... i Sort the values ​​and assign R to them. i The top N regions with the highest values ​​are identified as high-risk regions.

[0048] Furthermore, the visualization and output of the results include:

[0049] Draw a wind pressure distribution map: Generate a cloud map containing Cp_mean_grid and Cp_rms_grid;

[0050] High-risk areas are marked: On the cloud map, all identified high-risk areas are marked by boundary lines or color fills.

[0051] Generate a report: List the area, center location, average wind pressure, fluctuating wind pressure, and calculated comprehensive risk value R for each high-risk area. i .

[0052] By introducing a comprehensive risk index, this invention can solve the technical problems mentioned in the background art and bring the following significant beneficial effects:

[0053] 1. Achieves objective quantitative evaluation and reduces subjective dependence: This invention uses a clearly defined mathematical indicator R. i By establishing a threshold for determining risk levels, the identification of high-risk areas is transformed from a subjective process relying on experience into an objective calculation process based on data. The results are repeatable and verifiable, greatly improving the scientific rigor and reliability of the design.

[0054] 2. It provides a more comprehensive reflection of the mechanisms of wind-induced damage and improves identification accuracy: Because the comprehensive risk index considers both the static damage effect of average wind pressure and the dynamic fatigue effect of fluctuating wind pressure, it can identify high-risk areas that might be overlooked by average wind pressure cloud maps alone (e.g., areas with moderate average negative pressure but extremely large fluctuations, which are prone to fatigue failure). This makes the identification results more consistent with actual engineering damage patterns, and can more accurately guide local reinforcement design, avoiding premature structural failure.

[0055] 3. Significantly improves the design efficiency of multi-condition analysis: The method flow of this invention can be programmed (such as the provided MATLAB code) to achieve batch and automated processing of massive wind tunnel test conditions. Designers no longer need to manually consult hundreds of cloud maps; the program can automatically calculate and output a list of high-risk areas and their risk values ​​for all conditions, quickly identifying the most unfavorable condition and greatly improving design efficiency and accuracy.

[0056] 4. Added mechanistic explanations – core mechanistic explanations: a paradigm shift from “phenomenal observation” to “mechanism fusion”.

[0057] Existing technologies remain at the level of observing a single physical phenomenon of wind pressure fields (mean pressure), and their fundamental deficiency lies in failing to grasp the dual mechanical nature of wind-induced damage. This invention reveals for the first time that wind-induced damage to cooling towers is the result of the coupled effect of "static failure" and "fatigue accumulation effect," and establishes a precise quantitative model corresponding to these effects: Static failure (dominated by mean wind pressure): The local membrane stress in the shell caused by extreme negative pressure exceeds the tensile strength of concrete or the yield strength of steel reinforcement, leading to brittle cracking or buckling. This effect is quantified by |Mean_Cp_i|; Fatigue accumulation effect (dominated by pulsating wind pressure): High-frequency pulsating wind pressure induces alternating stress cycles in the steel reinforcement, resulting in irreversible damage accumulation within its microstructure, ultimately leading to sudden fatigue fracture at stress levels far below the static strength. This effect is quantified by Mean_Cp_rms_i, which is essentially a measure of the damage accumulation rate.

[0058] The breakthrough in the mechanism of this invention lies in the discovery that the two effects mentioned above do not exist independently, but are coupled and superimposed in time and space. A region may be inherently deficient due to a high average negative pressure, while simultaneously experiencing accelerated lifespan reduction due to intense pulsation. Traditional methods only recognize the "inherent deficiencies" and completely ignore the more insidious and dangerous destructive mode of "accelerated lifespan reduction." Therefore, the comprehensive risk index R proposed in this invention... i It is not a simple mathematical weighting, but a mathematical representation of the above-mentioned coupling disruption mechanism.

[0059] The breakthrough significance of this formula lies in the fact that it directly links the aerodynamic characteristics of wind pressure fields with the failure mechanics mechanism of structural materials through a calculable index for the first time, achieving a fundamental leap from "describing wind pressure" to "predicting risk".

[0060] 5. Enhanced Expected Results – An Industrial Transformation from “Chart Interpretation Assistance” to “Intelligent Decision Making”

[0061] The intended effect of this invention goes far beyond "more accurate identification," bringing a completely new and internationally leading digital decision-making system to the wind-resistant design of cooling towers.

[0062] 6. Achieved "precise targeting" identification of high-risk areas.

[0063] Expected results: This invention can accurately pinpoint "high-risk hidden areas" (i.e., areas with moderate average negative pressure but extremely strong pulsation) that are undetectable by traditional methods. For example, under dual-tower interference conditions, this invention can clearly reveal the impact area of ​​periodic detachment of wake vortices, which has an extremely high risk of fatigue damage but appears flat on the average wind pressure cloud map.

[0064] Breakthrough significance: It solves the problem of early identification of "hidden fatigue failure" that has plagued the international wind engineering community for many years, and provides an unprecedentedly accurate target for targeted reinforcement design (such as increasing the reinforcement ratio locally and using fatigue-resistant connection nodes).

[0065] 7. It pioneered the paradigm of "performance-based wind-resistant design".

[0066] Expected outcome: Designers will no longer be satisfied with merely ensuring the overall structure remains intact, but can instead leverage R... i Indicators are used to quantitatively design and verify the expected performance level of local components. For example, specifying R for all areas on the tower surface. i It must be below a certain limit in order to control the risk of damage from the design stage.

[0067] Breakthrough significance: This advancement advances the wind-resistant design of cooling towers from a "rough design" based on overall safety factors to a new stage of "refined design" based on local performance guarantees. This represents the international future development direction for wind-resistant design of tall structures.

[0068] 8. An "AI brain" for optimal decision-making under multiple operating conditions has been constructed.

[0069] Expected results: Through fully automated batch processing of massive wind tunnel test conditions, R... i The program calculates and sorts data, instantly outputting a list of the "most unfavorable risk conditions" and their risk values. Design no longer relies on engineers' subjective comparisons of complex cloud maps, but rather on scientific decision-making based on objective data.

[0070] Breakthrough significance: This invention provides an intelligent decision support tool for post-processing core wind tunnel test data, greatly improving design efficiency and reliability, and providing a key risk assessment kernel for engineering digitization.

[0071] Instruction manual illustrations

[0072] Figure 1 Flowchart of the method described in this invention (Flowchart of a method for identifying high-risk areas on the surface of a cooling tower based on wind tunnel pressure testing);

[0073] Figure 2aAverage wind pressure cloud map of the outer surface of the cooling tower obtained from wind tunnel tests;

[0074] Figure 2b 1. Fluctuating wind pressure cloud map of the outer surface of the cooling tower obtained from wind tunnel tests;

[0075] Figure 3a Initial screening of negative pressure zones based on regular grid average wind pressure data;

[0076] Figure 3b Initial screening of negative pressure zones based on regular grid-based pulsating wind pressure data;

[0077] Figure 3c High-risk areas identified based on comprehensive risk indicators; Detailed Implementation Plan

[0078] This invention provides a method for accurately identifying high-risk areas of a structural surface prone to wind-induced damage by quantitatively analyzing local average and fluctuating wind pressure based on wind tunnel test data. The flowchart of this invention's method is shown below. Figure 1 As shown, it mainly includes five core steps, and the detailed description of each step is as follows:

[0079] Step 1: Inputting Wind Tunnel Test Data

[0080] Obtain data from wind tunnel tests using a rigid cooling tower model. The data should include the coordinates (X, Y), mean wind pressure coefficient (Cp_mean), and root variance of fluctuating wind pressure coefficient (Cp_rms) of all measuring points on the model surface under various operating conditions. These operating conditions should cover different incoming wind velocities U, wind direction angles θ, and spacing ratios S under multi-tower interference.

[0081] Step 2: Data Preprocessing and Gridding

[0082] Because the measurement points are sparse and irregular, the data needs to be interpolated onto a regular grid to facilitate subsequent analysis.

[0083] Procedure: Using interpolation functions such as griddata or scatteredInterpolant, the discrete (X,Y,Cp_mean,Cp_rms) data is interpolated onto a predefined nx×ny fine two-dimensional grid, generating matrix-form grid coordinates X_grid, Y_grid, mean wind pressure coefficient matrix Cp_mean_grid, and fluctuating wind pressure coefficient matrix Cp_rms_grid.

[0084] Related: The regularized grid data generated in this step forms the basis for all subsequent analyses.

[0085] Step 3: Initial screening in the negative pressure zone and analysis of connected regions

[0086] Based on the average wind pressure, potential negative pressure areas of concern are initially identified.

[0087] Method:

[0088] Set a threshold: According to engineering experience, set a threshold (e.g., -0.65) for the average wind pressure coefficient. Areas where Cp_mean_grid < threshold are determined as "significant negative pressure areas".

[0089] Mark connected regions: Use the connected component analysis algorithm in image processing (such as the bwconncomp function) to mark the binarized negative pressure areas above, and identify several non-connected independent negative pressure areas.

[0090] Extract regional features: For each independent connected region, calculate its area (Area), the average value of the average wind pressure coefficient (Mean_Cp), and the centroid coordinates of the region (Centroid).

[0091] Improvement point: This step converts the subjective "looking at the cloud diagram" into an objective "quantitative partitioning", and extracts the basic features of each region, providing an object for subsequent refined analysis.

[0092] Association: Each extracted connected region is the basic unit for in-depth analysis in Step Four.

[0093] Step Four: Calculation of comprehensive risk indicators and identification of high-risk areas (core improvement point)

[0094] This is the core algorithm step of the present invention. It goes beyond the limitation of only focusing on the average wind pressure and creatively introduces the influence of fluctuating wind pressure.

[0095] Method:

[0096] ① For each connected region i identified in Step Three, not only record its average wind pressure feature Mean_Cp_i, but also extract the average value of the root mean square of the fluctuating wind pressure coefficient corresponding to this region (Mean_Cp_rms_i) from Cp_rms_grid.

[0097] ② Define the comprehensive risk indicator (R i )

[0098] To overcome the defect that the existing extreme wind pressure method only reflects the instantaneous effect of a point and cannot reflect the cumulative risk of the region, the present invention deeply integrates fracture mechanics and probability statistics theory, introduces the characteristic region area parameter A_i to reflect the "scale effect" and "cumulative effect" of damage, and defines the comprehensive risk indicator as follows:

[0099] R i=[α*|Mean_Cp_i|+β*Mean_Cp_rms_i]*(A_i / A_ref) γ

[0100] The model posits that the total risk of a region is a combination of its risk intensity and scale.

[0101] Wherein, α and β are weighting coefficients, which can be determined based on the design strength and fatigue strength of the actual engineering surface material, and their preferred ratio is:

[0102] α / β=k*([σ] / Δσ f )

[0103] in,

[0104] [σ] represents the allowable stress of the material as determined by the concrete structure design code. For reinforced concrete shells, it can be taken as the allowable tensile stress of the reinforcing steel; for HRB400 grade reinforcing steel, it can be taken as [σ]. s ] = 360 N / mm 2 .

[0105] Δσ f This refers to the allowable fatigue stress amplitude of reinforcing steel bars as determined according to the "Standard for Design of Concrete Structures" GB / T 50010-2010. For HRB400 grade reinforcing steel bars, Δσ can be taken as... f =162N / mm 2 (Fatigue stress ratio is 0.1).

[0106] k is an adjustment coefficient, which can be taken as 0.8 to 1.2 depending on the importance of the project.

[0107] The weighting coefficients can then be determined as follows:

[0108] α=[σ] / ([σ]+Δσ f = 360 / (360+162) ≈ 0.69

[0109] β=Δσ f / ([σ]+Δσ f = 162 / (360+162) ≈ 0.31

[0110] A_i: The area of ​​the i-th connected negative pressure region.

[0111] A_ref: Reference area, used for dimensionless transformation; can be the entire outer surface area of ​​the cooling tower, A_total.

[0112] γ: Area index, a key innovation coefficient; when γ = 1, it means that risk is linearly proportional to area; when γ = 0.5, it means that risk is proportional to the square root of area, which is more in line with the statistical law of the "weakest link" model (i.e., large areas have a higher probability of containing fatal defects).

[0113] This indicator defines the concept of total risk, quantifying the static destructive potential energy and fatigue damage potential inherent in the region. Extreme wind pressure calculations measure the maximum pressure that a point may experience at a single instant, failing to reflect the cumulative and scale effects over a region. This formula, however, calculates the overall threat to structural safety posed by a region throughout the entire wind load process. This represents a paradigm shift from "instantaneous point analysis" to "spatiotemporal regional analysis," aligning with the latest international concept of "holistic risk assessment." These improvements enable the assessment method of this invention to leap from predicting instantaneous extreme values ​​at a 'point' to predicting the long-term risk of a 'region,' overcoming the industry bottleneck of existing technologies' inability to scientifically quantify the scale of risk. This provides an internationally leading core tool for the performance-based wind-resistant design of major infrastructure.

[0114] In practical engineering applications—innovation and superiority at the engineering guidance level

[0115] 1) Solved the design dilemma of "high point not being high enough" and "low point not being low enough".

[0116] Scenario: The surface of the cooling tower has two areas:

[0117] Area A (high risk): very small area (1㎡), with very high extreme wind pressure (Cp_peak=-2.0).

[0118] Area B (low risk): large area (50㎡), moderate extreme wind pressure (Cp_peak=-1.2), but intense pulsation.

[0119] Extreme wind pressure guidance: Design resources (such as thickening slabs and increasing reinforcement) will be fully allocated to area A because it has the highest peak value. Area B is considered to have lower risk and will be ignored or treated only slightly.

[0120] Risk exposure guidance: Calculations show that the R_i value of region B is much larger than that of region A. This is because the overall risk exposure of region B, due to its large area and pulsation, far exceeds that of the isolated "high point".

[0121] Engineering significance: It avoids design blind spots. Traditional methods may miss the area with the greatest potential for damage (Area B), while the risk exposure index can accurately identify it as the primary reinforcement target, fundamentally eliminating the design error of "losing big for small" in the short term.

[0122] 2) Guiding differentiated and refined strengthening strategies

[0123] For regions with high [α·|Mean|+β·RMS] and low A_i, strategies to increase material strength should be adopted, such as using higher strength concrete or steel, to resist extremely high local stress.

[0124] For regions with medium [α·|Mean|+β·RMS] and large A_i, strategies to improve structural toughness and fatigue resistance should be adopted, such as optimizing reinforcement design (increasing the density of distributed reinforcement) and using fatigue-resistant structural details to suppress the initiation and propagation of cracks over a large area.

[0125] Engineering significance: It realizes the design concept of on-demand allocation and precise implementation. Limited engineering materials and economic costs are invested in the places where they are most needed, and the most appropriate reinforcement methods are adopted, thereby improving the safety and economy of the structure.

[0126] The risk exposure index proposed in this invention represents a theoretical leap from the traditional extreme wind pressure method of 'finding isolated high points' to 'assessing the total risk of a region'. It is no longer a simple aerodynamic load parameter, but an engineering risk assessment metric that deeply integrates wind pressure intensity, regional scale, and failure mechanism.

[0127] The innovation of this indicator lies in its precise revelation that structural failure stems not only from extreme load values ​​but also from the total amount of risk exposed spatially. Based on this, the engineering design paradigm can be transformed from a rudimentary model of 'passively resisting the highest point load' to a precise model of 'actively managing the overall risk distribution.' This allows for the scientific allocation of design resources to areas with the greatest destructive potential and the development of differentiated reinforcement strategies, thereby significantly improving economic efficiency while maintaining the same level of safety, or substantially enhancing structural reliability while maintaining the same level of economic efficiency. This risk assessment concept based on 'surface' represents the international forefront of wind-resistant design for tall structures.

[0128] ③ High-risk area determination: Set a comprehensive risk threshold R_threshold (e.g., 0.5), or apply R_threshold to all areas. i Sort the values ​​and assign R to them. i The N regions with the highest values ​​(or all regions exceeding the threshold) are identified as "high-risk regions".

[0129] ④ Core algorithm improvement: The fluctuating wind pressure (Cp_rms) is introduced as a core parameter into the evaluation system and weighted and integrated with the mean wind pressure (Cp_mean) to form a novel comprehensive risk index (R) that simultaneously reflects the potential destructive effects of static and dynamic wind loads. i This step represents a leap from "subjective judgment based on a single parameter" to "objective quantitative evaluation based on multiple parameters".

[0130] Step 5: Result Visualization and Output

[0131] The analysis results will be output in graphical and data-driven formats.

[0132] practice:

[0133] ① Draw a wind pressure distribution map: Generate a cloud map containing Cp_mean_grid and Cp_rms_grid (as shown in the background technology).

[0134] ② Mark high-risk areas: On the cloud map, clearly mark all high-risk areas identified in step four by using prominent boundary lines (such as contour lines) or color filling.

[0135] ③ Generate a report: Output a data report listing the area, center location, average wind pressure, fluctuating wind pressure, and calculated comprehensive risk value R for each high-risk area. i .

[0136] Related: This step presents the analysis results from the previous steps to the designers in a visual way, serving as a direct basis for their design decisions.

[0137] Figure 3a This diagram illustrates the distribution of the average wind pressure coefficient on the cooling tower surface under specific wind speeds (10 m / s), wind directions (20°), and spacing (0.5D) (the core of traditional analysis). The lined areas in the diagram represent the regions with the highest average negative pressure, which are traditionally considered the "most dangerous" areas. If only this diagram is considered, designers might subjectively concentrate resources on these lined areas, assuming they are the only high-risk points. This approach completely ignores the contribution of fluctuating wind pressure.

[0138] Under the same working conditions, Figure 3b The distribution of local shape coefficients during pulsations is shown (a new key dimension). The lined areas in the figure represent regions of severe wind pressure fluctuations, where the risk of fatigue damage is extremely high. One of the breakthroughs of this invention is... Figure 3b Incorporating core analytical dimensions. This invention recognizes that a mean negative pressure is moderate (in... Figure 3a (Inconspicuous) but with a strong pulse (in) Figure 3b The area marked with a line (in the center) may also be a high-risk area. This is a "hidden" risk that traditional methods cannot detect.

[0139] Figure 3c This invention demonstrates the identification of high-risk areas (an innovative achievement of this invention). Based on the comprehensive risk index proposed in this invention, it identifies high-risk areas... Figure 3a and Figure 3b The data is fused and calculated to automatically identify the final high-risk areas (the areas with the darkest color in the image). This is the second breakthrough of the invention. Figure 3c High-risk areas are Figure 3a and Figure 3bThe intelligent fusion of technologies includes areas with average wind pressure that is not extreme but exhibits very intense fluctuations, thus achieving a more comprehensive and accurate risk assessment. Furthermore, the identification of high-risk areas no longer relies on subjective human judgment based on visual descriptions, but is based on explicit algorithms and quantitative indicators, ensuring repeatable, verifiable, and comparable results. Finally, Figure 3c The output provides clear objectives for the design. Engineers can clearly identify which areas should be prioritized for reinforcement (such as increasing reinforcement or optimizing skin thickness), thus achieving a paradigm shift from "fuzzy perception" to "precise targeting" in design.

[0140] In summary, the difference of this invention lies in:

[0141] Traditional method = only analysis Figure 3a →Relying on human experience to make judgments may lead to missed assessments of fatigue risks.

[0142] This invention = comprehensive analysis Figure 3a + Figure 3b →Generated by algorithm quantization Figure 3c → It automatically, objectively, and comprehensively identifies all types of high-risk areas, providing a direct and reliable basis for guiding refined wind-resistant design.

[0143] The above technical solution also has the following alternatives:

[0144] 1. Comprehensive Risk Indicator (R) i Alternative formula for )

[0145] Multi-parameter fusion model: Introducing more parameters, such as the area of ​​the region (Area_i), to form a more complex index: R i = f(|Mean_Cp_i|,Mean_Cp_rms_i,Area_i), the larger the area, the higher the risk.

[0146] 2. Alternatives to the threshold determination method:

[0147] The threshold and R_threshold can be non-fixed to a certain value, but can be used as a relative percentage, for example: always keep R... i The top 5% of areas are classified as high-risk areas.

[0148] 3. Alternatives to data acquisition methods:

[0149] This invention is applicable not only to wind tunnel test data, but also to cooling tower surface wind pressure data obtained through computational fluid dynamics (CFD) numerical simulation. CFD simulation is another equivalent technique for obtaining wind pressure data.

[0150] 4. Alternatives to gridding and region analysis techniques:

[0151] (1) Grid interpolation can use various interpolation algorithms such as triangulation and kriging, and is not limited to the method used in the code.

[0152] (2) Connectivity analysis can use different neighborhood definitions (4-connected or 8-connected) or other clustering algorithms (such as DBSCAN) to cluster negative pressure points.

Claims

1. A method for identifying high-risk areas on the surface of a cooling tower based on wind tunnel pressure testing, characterized in that: Includes the following steps: Step 1: Obtain discrete data through a wind tunnel test using a rigid model of the cooling tower; Step 2: Use an interpolation function to grid the discrete data to obtain regular grid data; Step 3: Using the aforementioned regular grid data, complete the initial screening of the negative pressure zone and the analysis of the connected regions to achieve quantitative zoning; Step 4: Calculate the comprehensive risk index for each zone to complete the identification of high-risk areas; the comprehensive risk index is used to comprehensively evaluate the risk based on the static destructive potential energy, fatigue damage potential, and size of the negative pressure area to be analyzed. Step 5: Visualize and output the results.

2. The method for identifying high-risk areas on the surface of a cooling tower based on wind tunnel pressure testing according to claim 1, characterized in that: The discrete data includes the coordinates (X,Y) of all measuring points on the surface of the rigid model of the cooling tower under multiple operating conditions, the average wind pressure coefficient Cp_mean of all measuring points, and the root variance of the fluctuating wind pressure coefficient Cp_rms of all measuring points; the multiple operating conditions cover different incoming wind speeds U, wind direction angles θ, and spacing ratios S under multi-tower interference.

3. The method for identifying high-risk areas on the surface of a cooling tower based on wind tunnel pressure testing according to claim 2, characterized in that: Step two includes: using an interpolation function to interpolate the discrete data onto a predefined fine two-dimensional grid, generating matrix-form grid coordinates X_grid, Y_grid, mean wind pressure coefficient matrix Cp_mean_grid, and fluctuating wind pressure coefficient matrix Cp_rms_grid.

4. The method for identifying high-risk areas on the surface of a cooling tower based on wind tunnel pressure testing according to claim 3, characterized in that: The initial screening of the negative pressure zone and the analysis of the connected regions include: Set threshold: Use the set average wind pressure coefficient threshold to binarize the gridded average wind pressure coefficient; Connected component labeling: Using the connected component analysis algorithm, the above-mentioned binarized negative pressure area is labeled to identify several independent negative pressure areas that are not connected to each other; Regional feature extraction: For each independent connected region, calculate its area (Area), the average value of the average wind pressure coefficient (Mean_Cp), and the coordinates of the region's center (Centroid).

5. The method for identifying high-risk areas on the surface of a cooling tower based on wind tunnel pressure testing according to claim 4, characterized in that: The comprehensive risk index R i It is expressed as follows: R i =[α*|Mean_Cp_i|+β*Mean_Cp_rms_i]*(A_i / A_ref)γ Where α and β are weighting coefficients, Mean_Cp_i is the average value of the average wind pressure coefficient of the i-th connected negative pressure region, Mean_Cp_rms_i is the average value of the root variance of the fluctuating wind pressure coefficient corresponding to the i-th connected negative pressure region extracted from the fluctuating wind pressure coefficient matrix Cp_rms_grid, A_i is the area of ​​the i-th connected negative pressure region; A_ref is the reference area used for dimensionless transformation; γ is the area index, when γ = 1, it indicates that the risk is linearly proportional to the area, when γ = 0.5, it indicates that the risk is proportional to the square root of the area.

6. The method for identifying high-risk areas on the surface of a cooling tower based on wind tunnel pressure testing according to claim 5, characterized in that: The values ​​of α and β are determined based on the design strength and fatigue strength of the actual engineering cladding material, and their ratio is as follows: α / β=k*([σ] / Δσ f ) Where k is an adjustment coefficient, [σ] is the allowable material stress determined according to the concrete structure design code; Δσ f This refers to the allowable value of the fatigue stress amplitude of steel bars as determined according to the concrete structure design standard.

7. The method for identifying high-risk areas on the surface of a cooling tower based on wind tunnel pressure testing according to claim 5, characterized in that: The weighting coefficients α and β can also be determined in any of the following ways: (1) Based on the engineering design focus, the assumptions are made based on experience under the condition that α+β=1 is met; when the focus is on assessing the risk of static failure, α>β; when the focus is on assessing the risk of fatigue failure, α<β. (2) The reliability analysis is used to calibrate and determine the results so that the identification results of high-risk areas match the target reliability index.

8. The method for identifying high-risk areas on the surface of a cooling tower based on wind tunnel pressure testing according to claim 5, characterized in that: The comprehensive risk index R i It can also be expressed as follows: Product form: R i =|Mean_Cp_i|*Mean_Cp_rms_i; Based on extreme value statistics: R i =|Mean_Cp_i|+K*Mean_Cp_rms_i, where K is the peak factor.

9. The method for identifying high-risk areas on the surface of a cooling tower based on wind tunnel pressure testing according to any one of claims 1-8, characterized in that: The method for identifying high-risk areas is as follows: a comprehensive risk threshold is set, and all areas exceeding the comprehensive risk threshold are identified as high-risk areas; The method for identifying high-risk areas is as follows: For all areas, R... i Sort the values ​​and assign R to them. i The top N regions with the highest values ​​are identified as high-risk regions.

10. The method for identifying high-risk areas on the surface of a cooling tower based on wind tunnel pressure testing according to claim 9, characterized in that: The visualization and output of the results include: Draw a wind pressure distribution map: Generate a cloud map containing Cp_mean_grid and Cp_rms_grid; High-risk areas are marked: On the cloud map, all identified high-risk areas are marked by boundary lines or color fills. Generate a report: List the area, center location, average wind pressure, fluctuating wind pressure, and calculated comprehensive risk value R for each high-risk area. i .

Images