Indirect air cooling tower dust accumulation performance test method, device and equipment

Abstract

The embodiment of the application provides a kind of indirect air cooling tower radiator dust performance test method, device and equipment.This application relates to the field of indirect air cooling tower radiator dust performance test, which comprises the following steps: obtaining the reference thermal performance parameters of the radiator in the dust-free state based on the dusting device; simulate different sand and dust conditions based on the dusting device, and obtain the dusting thermal performance parameters and dusting morphological characteristic parameters of the radiator under each condition; based on the reference thermal performance parameters, the dusting thermal performance parameters and the dusting morphological characteristic parameters, establish the correlation between the radiator performance and the sand and dust conditions; based on the correlation, a radiator performance prediction model is constructed using a machine learning algorithm.The method is used to accurately predict the dust performance of the radiator.

Description

Technical Field

[0001] This application relates to the field of dust accumulation performance testing of indirect air cooling tower radiators, and in particular to a method, apparatus and equipment for testing the dust accumulation performance of indirect air cooling tower radiators. Background Technology

[0002] Indirect air-cooled towers are crucial heat exchange equipment in nuclear power, thermal power, and chemical industries. Their core component, the radiator, typically operates continuously in open-air environments, effectively dissipating the large amounts of heat generated during industrial cycles into the ambient air. In actual operation, the radiator fins are constantly exposed to various environmental factors such as wind, sand, dust, and moisture, gradually forming and accumulating a layer of dust. This dust layer not only significantly increases the thermal resistance along the heat transfer path, reducing heat dissipation efficiency, but also substantially increases air-side resistance due to flow channel blockage, ultimately leading to operational problems such as decreased heat dissipation capacity and increased energy consumption in the entire air-cooling system.

[0003] Currently, performance testing of radiators mainly focuses on their basic heat transfer and resistance characteristics under dust-free conditions. While these methods can accurately reflect the initial performance of radiators under ideal conditions, they are difficult to effectively simulate the dynamic performance degradation caused by dust accumulation during long-term service. Due to the lack of systematic testing methods, existing technologies cannot accurately quantify the impact of dust accumulation. Therefore, existing technologies suffer from the technical problem of not being able to accurately predict the performance of radiators under dust accumulation. Summary of the Invention

[0004] This application provides a method, apparatus, and equipment for testing the dust accumulation performance of indirect air cooling tower radiators, in order to improve the technical effect of accurately predicting the dust accumulation performance of radiators.

[0005] In a first aspect, embodiments of this application provide a method for testing the dust accumulation performance of an indirect air cooling tower radiator, including:

[0006] Based on the dust collection device, the baseline thermodynamic performance parameters of the radiator in a dust-free state are obtained;

[0007] Based on the dust accumulation device, different sand and dust conditions were simulated to obtain the dust accumulation thermodynamic performance parameters and dust accumulation morphology parameters of the radiator under each condition.

[0008] Based on the baseline thermodynamic performance parameters, dust accumulation thermodynamic performance parameters, and dust accumulation morphology parameters, the correlation between radiator performance and dust conditions is established.

[0009] Based on the correlation, a radiator performance prediction model is constructed using machine learning algorithms. The radiator performance prediction model is used to predict the performance degradation trend curve of the radiator based on actual operating parameters, and the performance degradation trend curve is used to generate maintenance early warning information for the radiator.

[0010] In one possible implementation, the baseline thermal performance parameters include dust-free pressure differential and dust-free heat dissipation, and the dust accumulation thermal performance parameters include dust accumulation pressure differential and dust accumulation heat dissipation. The dust-free pressure differential is used to characterize the airflow resistance of the radiator in a dust-free state, the dust-free heat dissipation is used to characterize the heat exchange capacity of the radiator in a dust-free state, the dust accumulation pressure differential is used to characterize the degree of obstruction of the ventilation channel by sand and dust adhesion to the radiator in a dust-accumulated state, and the dust accumulation heat dissipation is used to characterize the degree of weakening of heat dissipation performance by dust accumulation to the radiator in a dust-accumulated state.

[0011] In one possible implementation, the dust accumulation morphology characteristic parameters are obtained as follows:

[0012] The surface images of the radiator under various operating conditions are captured by a high-definition camera on the dust collection device.

[0013] Image recognition technology was used to calculate the gap blockage rate and average dust accumulation thickness of the radiator fins. The gap blockage rate was used to characterize the degree of reduction in the effective ventilation area between the radiator fins, and the average dust accumulation thickness was used to characterize the geometric scale of the dust layer attached to the radiator surface.

[0014] In one possible implementation, a correlation between radiator performance and dust conditions is established based on baseline thermodynamic performance parameters, dust accumulation thermodynamic performance parameters, and dust accumulation morphology characteristics, including:

[0015] The pressure difference increase rate is obtained based on the dust-free pressure difference in the baseline thermodynamic performance parameters and the dust accumulation pressure difference in the dust accumulation thermodynamic performance parameters.

[0016] The heat dissipation rate is obtained based on the heat dissipation of dust-free thermal performance parameters and the heat dissipation of dust-accumulated thermal performance parameters.

[0017] Based on the pressure difference increase rate and the gap blockage rate in the dust accumulation morphology characteristic parameters, the drag influence coefficient is obtained; whereby the drag influence coefficient is used to characterize the severity of the influence of dust accumulation morphology on the increase of surface roughness or local flow field distortion.

[0018] The effective area heat dissipation rate is obtained based on the heat dissipation attenuation rate and the gap blockage rate in the dust accumulation morphology characteristic parameters; whereby the effective area heat dissipation rate is used to characterize the actual heat dissipation capacity on the remaining effective heat exchange area after eliminating the factor of reduced ventilation area.

[0019] In one possible implementation, a heat sink performance prediction model is constructed using machine learning algorithms based on correlation relationships, including:

[0020] A training dataset is constructed, which includes multiple sets of training data. Each set of training data includes the operating condition type, the proportion of sand and dust particles, the current total amount of dust accumulation, the gap blockage rate, the average dust accumulation thickness, the resistance influence coefficient, the effective area heat dissipation rate, and the correlation relationship. Among them, the correlation relationship includes the pressure difference increase rate and the heat dissipation attenuation rate.

[0021] Each set of training data is input into the initial radiator performance prediction model to obtain the predicted correlation; the correlation includes the predicted differential pressure increase rate and the predicted heat dissipation attenuation rate.

[0022] The loss function value between the pre-defined loss function and the predicted loss function is calculated.

[0023] The initial radiator performance prediction model is trained based on the loss function value until the preset loss function converges, resulting in a well-trained radiator performance prediction model.

[0024] In one possible implementation, it also includes:

[0025] The actual dust particle distribution parameters of the test site environment and the estimated total dust accumulation obtained from the expected operating time are input into the radiator performance prediction model.

[0026] Output the pressure difference prediction curve and heat dissipation prediction curve for a future set time period;

[0027] A maintenance warning is triggered when the increase in differential pressure in the differential pressure prediction curve exceeds the preset differential pressure threshold, or the decrease in heat dissipation in the heat dissipation prediction curve exceeds the preset heat dissipation threshold.

[0028] In one possible implementation, it also includes:

[0029] Based on the operating conditions dominated by large particles, small particles, and mixed particles, the particle size distribution ratio of sand and dust in the dust collection device is configured respectively.

[0030] Among them, the large particle-dominated operating condition is used to simulate an environment in which physical obstruction is easily generated between fins; the small particle-dominated operating condition is used to simulate an environment in which a dense covering layer is easily formed on the fin surface; and the mixed particle operating condition is used to simulate a complex environment in which the spatial skeleton formed by large particles and the pore filling generated by small particles are superimposed, thereby causing a synergistic deterioration of airflow resistance and heat transfer performance.

[0031] Secondly, embodiments of this application provide a dust accumulation performance testing device for an indirect air cooling tower radiator, comprising:

[0032] The first processing module is used to obtain the baseline thermodynamic performance parameters of the radiator in a dust-free state based on the dust accumulation device.

[0033] The first processing module is also used to simulate different sand and dust conditions based on the dust accumulation device, and obtain the dust accumulation thermodynamic performance parameters and dust accumulation morphology parameters of the radiator under each condition.

[0034] The second processing module is used to establish the correlation between radiator performance and dust conditions based on the baseline thermodynamic performance parameters, dust accumulation thermodynamic performance parameters, and dust accumulation morphology characteristics parameters.

[0035] The third processing module is used to construct a radiator performance prediction model based on the correlation relationship using machine learning algorithms. The radiator performance prediction model is used to predict the performance degradation trend curve of the radiator based on actual operating parameters, and the performance degradation trend curve is used to generate maintenance early warning information for the radiator.

[0036] In one possible implementation, the baseline thermal performance parameters include dust-free pressure differential and dust-free heat dissipation, and the dust accumulation thermal performance parameters include dust accumulation pressure differential and dust accumulation heat dissipation. The dust-free pressure differential is used to characterize the airflow resistance of the radiator in a dust-free state, the dust-free heat dissipation is used to characterize the heat exchange capacity of the radiator in a dust-free state, the dust accumulation pressure differential is used to characterize the degree of obstruction of the ventilation channel by sand and dust adhesion to the radiator in a dust-accumulated state, and the dust accumulation heat dissipation is used to characterize the degree of weakening of heat dissipation performance by dust accumulation to the radiator in a dust-accumulated state.

[0037] In one possible implementation, the first processing module is further configured to:

[0038] The surface images of the radiator under various operating conditions are captured by a high-definition camera on the dust collection device.

[0039] Image recognition technology was used to calculate the gap blockage rate and average dust accumulation thickness of the radiator fins. The gap blockage rate was used to characterize the degree of reduction in the effective ventilation area between the radiator fins, and the average dust accumulation thickness was used to characterize the geometric scale of the dust layer attached to the radiator surface.

[0040] In one possible implementation, the second processing module is further configured to:

[0041] The pressure difference increase rate is obtained based on the dust-free pressure difference in the baseline thermodynamic performance parameters and the dust accumulation pressure difference in the dust accumulation thermodynamic performance parameters.

[0042] The heat dissipation rate is obtained based on the heat dissipation of dust-free thermal performance parameters and the heat dissipation of dust-accumulated thermal performance parameters.

[0043] Based on the pressure difference increase rate and the gap blockage rate in the dust accumulation morphology characteristic parameters, the drag influence coefficient is obtained; whereby the drag influence coefficient is used to characterize the severity of the influence of dust accumulation morphology on the increase of surface roughness or local flow field distortion.

[0044] The effective area heat dissipation rate is obtained based on the heat dissipation attenuation rate and the gap blockage rate in the dust accumulation morphology characteristic parameters; whereby the effective area heat dissipation rate is used to characterize the actual heat dissipation capacity on the remaining effective heat exchange area after eliminating the factor of reduced ventilation area.

[0045] In one possible implementation, the third processing module is further configured to:

[0046] A training dataset is constructed, which includes multiple sets of training data. Each set of training data includes the operating condition type, the proportion of sand and dust particles, the current total amount of dust accumulation, the gap blockage rate, the average dust accumulation thickness, the resistance influence coefficient, the effective area heat dissipation rate, and the correlation relationship. Among them, the correlation relationship includes the pressure difference increase rate and the heat dissipation attenuation rate.

[0047] Each set of training data is input into the initial radiator performance prediction model to obtain the predicted correlation; the correlation includes the predicted differential pressure increase rate and the predicted heat dissipation attenuation rate.

[0048] The loss function value between the pre-defined loss function and the predicted loss function is calculated.

[0049] The initial radiator performance prediction model is trained based on the loss function value until the preset loss function converges, resulting in a well-trained radiator performance prediction model.

[0050] In one possible implementation, the third processing module is further configured to:

[0051] The actual dust particle distribution parameters of the test site environment and the estimated total dust accumulation obtained from the expected operating time are input into the radiator performance prediction model.

[0052] Output the pressure difference prediction curve and heat dissipation prediction curve for a future set time period;

[0053] A maintenance warning is triggered when the increase in differential pressure in the differential pressure prediction curve exceeds the preset differential pressure threshold, or the decrease in heat dissipation in the heat dissipation prediction curve exceeds the preset heat dissipation threshold.

[0054] In one possible implementation, the first processing module is further configured to:

[0055] Based on the operating conditions dominated by large particles, small particles, and mixed particles, the particle size distribution ratio of sand and dust in the dust collection device is configured respectively.

[0056] Among them, the large particle-dominated operating condition is used to simulate an environment in which physical obstruction is easily generated between fins; the small particle-dominated operating condition is used to simulate an environment in which a dense covering layer is easily formed on the fin surface; and the mixed particle operating condition is used to simulate a complex environment in which the spatial skeleton formed by large particles and the pore filling generated by small particles are superimposed, thereby causing a synergistic deterioration of airflow resistance and heat transfer performance.

[0057] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0058] The memory stores computer-executed instructions;

[0059] The processor executes computer execution instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0060] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0061] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0062] The indirect air cooling tower radiator dust accumulation performance testing method, apparatus, and equipment provided in this application obtain the baseline thermodynamic performance parameters of the radiator under dust-free conditions using a dust accumulation device, establishing an initial reference for performance analysis. The dust accumulation device simulates diverse sand and dust conditions, dynamically acquiring the dust accumulation thermodynamic performance parameters of the radiator under different conditions and their corresponding dust morphology characteristics, thereby constructing an experimental dataset containing multi-dimensional information. Based on this, a comprehensive analysis is performed on the baseline thermodynamic performance parameters, dust accumulation thermodynamic performance parameters, and dust morphology characteristics to establish the correlation between radiator performance and sand and dust conditions. The established correlation is used as key feature input, and a machine learning algorithm is used to construct a radiator performance prediction model. This radiator performance prediction model can dynamically predict the radiator's performance degradation trend curve based on actual operating parameters such as wind speed, sand and dust concentration, and humidity in the target environment. This effectively improves the technical effect of accurately predicting the dust accumulation performance of radiators. Attached Figure Description

[0063] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0064] Figure 1 Flowchart of the dust accumulation performance test method for indirect air cooling tower radiators provided in this application Figure 1 ;

[0065] Figure 2 Flowchart of the dust accumulation performance test method for indirect air cooling tower radiators provided in this application Figure 2 ;

[0066] Figure 3 A schematic diagram of the structure of the dust accumulation performance testing device for the indirect air cooling tower radiator provided in this application;

[0067] Figure 4 A hardware schematic diagram of the dust accumulation performance testing device for the indirect air cooling tower radiator provided in this application.

[0068] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0069] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of methods and approaches consistent with some aspects of this application as detailed in the appended claims.

[0070] Currently, radiator performance testing primarily relies on cleanroom testing conducted in standard laboratory environments. These tests are typically performed in artificial climate chambers or wind tunnels, precisely controlling inlet air temperature, humidity, flow rate, and the temperature and flow rate of the circulating medium to determine the radiator's basic heat transfer performance and air-side resistance characteristics under conditions free of contamination. While these standardized initial performance tests can accurately and repeatedly assess the performance of different radiator designs under ideal conditions, providing a basis for product selection and comparative analysis, their testing conditions differ from actual operating environments. Therefore, they cannot reflect the true trajectory of radiator performance dynamic degradation as dust accumulation increases over long-term operation.

[0071] To compensate for the limitations of laboratory testing, existing technologies assess the impact of dust accumulation through on-site measurements. Common implementation methods include installing differential pressure sensors and infrared thermometers on the radiators of operating air-cooled towers, or conducting periodic heat dissipation capacity tests. However, the on-site environment is highly random and uncontrollable. Key influencing factors such as wind speed, wind direction, ambient temperature, humidity, and the concentration and composition of atmospheric pollutants are constantly changing and cannot be manually set. This makes it difficult to maintain consistent boundary conditions between any two measurements, resulting in large dispersion and poor repeatability of test data, making it difficult to extract universal and quantifiable performance degradation patterns.

[0072] Secondly, dust accumulation is a slow, long-term process measured in months or even years. Obtaining a statistically significant performance degradation curve requires continuous monitoring for several years, an extremely lengthy process. The timeliness of data acquisition is completely unacceptable for new product development, rapid optimization, or the formulation of short-term operation and maintenance strategies. Furthermore, on-site measurement itself faces significant technical challenges. In harsh industrial environments characterized by high temperature, high humidity, high vibration, and strong electromagnetic interference, the long-term stability, installation reliability, and signal accuracy of high-precision sensors are difficult to guarantee. Key parameters of the dust layer itself, such as thickness distribution, packing density, adhesion, and its non-uniform distribution on the heat sink surface, are difficult to measure accurately and in real-time online without interrupting operation or damaging the structure.

[0073] Therefore, on-site measurements can usually only obtain phenomenal performance change trends, but cannot accurately reveal the instantaneous and quantitative correlation between dust accumulation, dust accumulation morphology and local and overall performance degradation of the heat sink.

[0074] In summary, the two main implementation paths of existing technologies—laboratory cleanliness testing detached from reality, and on-site testing that is close to reality but highly uncontrollable, inaccurate, and time-consuming—result in a lack of an efficient method throughout the entire lifecycle of radiator design, selection, operation, and maintenance. This method can accurately simulate the dynamic dust accumulation process under controllable and repeatable experimental conditions, and simultaneously, continuously, and quantitatively measure the evolution of the radiator's thermal and resistance performance. Therefore, it is impossible to accurately predict the dust accumulation performance of radiators during long-term operation.

[0075] To address the aforementioned technical issues, this application provides a method, apparatus, and equipment for testing the dust accumulation performance of indirect air cooling tower radiators. Firstly, relying on a dust accumulation device capable of precisely controlling multiple physical field environments such as wind, dust, humidity, and heat, the entire process from baseline to dynamic dust accumulation is systematically tested under highly realistic controlled experimental conditions. Specifically, baseline thermodynamic performance parameters of the radiator in a dust-free state are obtained. Using the same dust accumulation device, various typical or extreme dust conditions are simulated by adjusting key variables such as dust concentration, particle size distribution, ambient humidity, and wind speed. Under these simulated conditions, key data in two dimensions are collected simultaneously and continuously: firstly, the dust accumulation thermodynamic performance parameters reflecting changes in radiator heat transfer and resistance; and secondly, the dust accumulation morphology parameters reflecting the dust accumulation development process, obtained through imaging and other technical means. Through this step, this method efficiently reproduces the complex dust accumulation process that a radiator might undergo in the field over months or even years, obtaining high spatiotemporal resolution performance and morphological evolution data.

[0076] After obtaining the above multi-dimensional and multi-condition experimental dataset, we conducted in-depth correlation analysis on the baseline thermal performance parameters, dust accumulation thermal performance parameters, and dust accumulation morphology parameters to establish the correlation between radiator performance and dust conditions. The aim was to reveal the intrinsic quantitative laws between the performance degradation rate, the final degradation degree, and dust concentration, humidity, time, and the physical morphology of the dust layer.

[0077] Based on correlations, a radiator performance prediction model is constructed using machine learning algorithms. This model uses key features revealed by experimentally established correlations as important prior knowledge or feature inputs, and employs a large amount of laboratory simulation data as its training set. Machine learning algorithms (such as neural networks and random forests) can autonomously learn from this high-value data and capture complex nonlinear mapping relationships between multiple variables. The fundamental function of the finally trained radiator performance prediction model is that, when inputting actual operating parameters monitored or predicted at the target site (such as local historical and future wind speed, dust concentration, and humidity time-series data), the model can extrapolate the performance degradation trend curve of the radiator at a specific location. This prediction curve dynamically shows the trajectory of radiator performance (such as heat transfer efficiency) decreasing over time or with accumulated pollution. Based on this curve, performance thresholds can be intelligently set. When the predicted performance approaches the maintenance critical point, the system can automatically generate maintenance warning information for the radiator, thereby achieving a fundamental shift from post-fault maintenance or periodic over-maintenance to state-based, precise predictive maintenance.

[0078] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0079] Figure 1 Flowchart of the dust accumulation performance test method for indirect air cooling tower radiators provided in this application Figure 1 ,like Figure 1 As shown, the method includes:

[0080] S101. Based on the dust collection device, obtain the reference thermal performance parameters of the radiator in a dust-free state.

[0081] In this embodiment, relying on the constructed integrated dust accumulation test device, the initial performance calibration of the radiator is carried out in a strictly dust-free environment. The radiator under test is cleanly installed in the wind tunnel test section, the automatic feeding system is turned off, and a series of standardized test conditions (such as different wind speeds and different inlet water temperature combinations) are established by adjusting the variable frequency fan, constant temperature water tank and other devices. High-precision sensors are used to measure the core thermo-hydraulic parameters of the radiator under each condition, such as heat exchange, heat transfer coefficient, and air-side resistance. These parameters are recorded as the baseline thermo-hydraulic performance parameters.

[0082] S102. Simulate different sand and dust conditions based on the dust accumulation device, and obtain the dust accumulation thermodynamic performance parameters and dust accumulation morphology parameters of the radiator under each condition.

[0083] In this embodiment, the same dust collection device is used to simulate various typical sandstorm conditions through the coordinated operation of its automatic feeding system, humidification system, and wind tunnel equipment. For example, it can simulate low-concentration, long-term environmental dust backgrounds, or high-concentration, short-term sandstorm processes. It can also be combined with the humidification system to simulate dust adhesion and caking under high humidity conditions. During these dynamic and controlled dust collection processes, two types of key data are continuously and synchronously collected: one type is the dust thermodynamic performance parameters reflecting the dynamic changes in the heat exchange and flow capacity of the radiator; the other type is the dust morphological characteristic parameters describing the development state of dust accumulation, such as dust coverage, average thickness, bulk density, and surface roughness, obtained through high-definition image acquisition, laser scanning, and other means.

[0084] S103. Based on the baseline thermodynamic performance parameters, dust accumulation thermodynamic performance parameters, and dust accumulation morphology parameters, establish the correlation between radiator performance and dust conditions.

[0085] In this embodiment, a deep correlation analysis is performed on the aforementioned multi-dimensional data to establish the correlation between radiator performance and dust conditions. Specifically, by comprehensively analyzing the baseline thermodynamic performance parameters, the dust accumulation thermodynamic performance parameters of each time series, and their corresponding dust accumulation morphology characteristics, the relationship between performance degradation and various influencing factors (such as cumulative dust accumulation, dust morphology, ambient humidity, and wind speed) can be quantified. For example, through multiple regression analysis or response surface modeling, specific correlation patterns can be derived, such as "under specific humidity conditions, when the dust morphology changes from loose to compacted and the coverage exceeds a certain threshold, the flow resistance will increase non-linearly and sharply."

[0086] S104. Based on the correlation, a radiator performance prediction model is constructed using machine learning algorithms.

[0087] In this embodiment, the radiator performance prediction model is used to predict the radiator's performance degradation trend curve based on actual operating parameters, and the performance degradation trend curve is used to generate radiator maintenance early warning information.

[0088] The dust accumulation performance testing method for indirect air cooling tower radiators provided in this application embodiment measures the baseline thermodynamic performance parameters of the radiator under dust-free conditions using a dust accumulation device. The same device system simulates different dust conditions, simultaneously acquiring the dust accumulation thermodynamic performance parameters of the radiator at different dust accumulation stages during the simulation, and recording the corresponding dust morphology characteristics using high-precision monitoring methods. By combining and analyzing the experimental data, a correlation between radiator performance and dust conditions can be established, revealing the quantitative laws governing performance degradation and dust concentration, humidity, time, and dust morphology. Based on this correlation, a radiator performance prediction model is further constructed using machine learning algorithms. This model can learn and internalize complex nonlinear relationships between multiple variables, thereby dynamically predicting the radiator's performance degradation trend curve based on actual operating parameters such as wind speed, dust concentration, and humidity at the target site. Through the radiator performance prediction model, and based on the performance degradation trend curve simulated using actual operating parameters, a cleaning plan can be scientifically formulated, and operational strategies optimized to avoid unplanned downtime, reduce energy consumption losses, and achieve optimal life-cycle cost and comprehensive improvement in operational safety of the air-cooled system.

[0089] Figure 2 Flowchart of the dust accumulation performance test method for indirect air cooling tower radiators provided in this application Figure 2 ,like Figure 2 As shown, in this embodiment... Figure 1 Based on the examples, a detailed description of the dust accumulation performance test method for indirect air cooling tower radiators is provided, wherein the method includes:

[0090] S201. Based on the dust collection device, obtain the reference thermal performance parameters of the radiator in a dust-free state.

[0091] In this embodiment, the baseline thermal performance parameters include dust-free pressure difference and dust-free heat dissipation, and the dust accumulation thermal performance parameters include dust accumulation pressure difference and dust accumulation heat dissipation. The dust-free pressure difference is used to characterize the airflow resistance of the radiator in a dust-free state, the dust-free heat dissipation is used to characterize the heat exchange capacity of the radiator in a dust-free state, the dust accumulation pressure difference is used to characterize the degree of obstruction of the ventilation channel by sand and dust adhesion to the radiator in a dust-accumulated state, and the dust accumulation heat dissipation is used to characterize the degree of weakening of heat dissipation performance by dust accumulation to the radiator in a dust-accumulated state.

[0092] Specifically, a cleanroom-free radiator under test is installed in the wind tunnel test section. Without any dust or sand being introduced, the wind tunnel speed, circulating water temperature, and flow rate are precisely adjusted to ensure operation under a series of standardized and stable conditions. During this process, a high-precision differential pressure sensor measures the static pressure difference before and after the airflow passes through the radiator; this is the cleanroom pressure difference, which characterizes the inherent resistance to airflow caused by the radiator fins and tube bundle structure in a clean state, reflecting its designed aerodynamic characteristics. Simultaneously, by measuring the temperature change and mass flow rate of the circulating fluid flowing through the radiator and calculating it using heat transfer formulas, the cleanroom heat dissipation under the current operating conditions can be obtained. This parameter directly quantifies the radiator's maximum heat exchange capacity under ideal conditions.

[0093] S202. Simulate different sand and dust conditions based on the dust collection device to obtain the dust accumulation thermodynamic performance parameters of the radiator under each condition; collect surface images of the radiator under each condition using a high-definition camera on the dust collection device; and calculate the gap blockage rate and average dust accumulation thickness of the radiator fins using image recognition technology.

[0094] In this embodiment, the gap blockage rate is used to characterize the degree of reduction in the effective ventilation area between the radiator fins; the average dust accumulation thickness is used to characterize the geometric scale of the dust layer attached to the radiator surface.

[0095] Specifically, based on the baseline thermodynamic performance parameters obtained under dust-free conditions, a dynamic dust accumulation simulation experiment is initiated using the automatic feeding, humidification, and airflow control system of the dust accumulation device. For example, a set of combined operating conditions can be set, including different dust concentrations (e.g., 0.1 g / m³ simulating light pollution, 1.0 g / m³ simulating dusty weather) and different ambient humidity levels (e.g., 30% RH dry conditions, 80% RH high humidity conditions). Under each operating condition, the device runs continuously for several hours according to a preset program. During this period, the dust accumulation thermodynamic performance parameters, namely dust accumulation pressure difference and dust accumulation heat dissipation, are continuously collected at high frequency, and the complete curves of dynamic decay of resistance and heat exchange capacity as dust accumulates are recorded in real time.

[0096] Meanwhile, a high-definition camera fixed outside the transparent observation window of the dust accumulation device test section automatically acquires high-definition images of the radiator fin surface at fixed time intervals (e.g., every 5 minutes). These image sequences completely record the entire physical deposition process of dust, from initial adhesion to gradual coverage and even potential hardening.

[0097] The acquired images are automatically processed using image recognition technology: First, threshold segmentation and edge detection algorithms are used to identify the fin area and the dust-covered area. Then, the proportion of the pixel area of ​​the fin gap covered by dust to the total pixel area of ​​the fin gap is calculated, thus obtaining the gap blockage rate. Simultaneously, by analyzing the shadow features of the dust deposition area in the image or using the principle of binocular vision, the average dust accumulation thickness can be estimated. This step essentially transforms the complex three-dimensional physical form of dust accumulation, which is difficult to measure directly, into quantifiable and traceable digital morphological characteristics.

[0098] S203. Based on the dust-free pressure difference in the reference thermodynamic performance parameters and the dust accumulation pressure difference in the dust accumulation thermodynamic performance parameters, the pressure difference increase rate is obtained; based on the dust-free heat dissipation in the reference thermodynamic performance parameters and the dust accumulation heat dissipation in the dust accumulation thermodynamic performance parameters, the heat dissipation attenuation rate is obtained; based on the pressure difference increase rate and the gap blockage rate in the dust accumulation morphology characteristic parameters, the resistance influence coefficient is obtained; based on the heat dissipation attenuation rate and the gap blockage rate in the dust accumulation morphology characteristic parameters, the effective area heat dissipation rate is obtained.

[0099] In this embodiment, the resistance influence coefficient is used to characterize the severity of the influence of dust accumulation morphology on the increase of surface roughness or local flow field distortion; the effective area heat dissipation rate is used to characterize the actual heat dissipation capacity on the remaining effective heat exchange area after eliminating the factor of reduced ventilation area.

[0100] Step S203 is designed based on the principle of performing in-depth mathematical transformation and physical decomposition on the test data obtained in the previous steps. This allows for the deriving of the impact of different factors on radiator performance from the macroscopic performance degradation, thereby revealing the intrinsic mechanism by which dust accumulation affects performance. Specifically, firstly, using the dust-free pressure difference obtained in S201 and the dust-accumulated pressure difference simultaneously collected in S202, the pressure difference increase rate is calculated by (dust-accumulated pressure difference - dust-free pressure difference) / dust-free pressure difference × 100%. This indicator eliminates the influence of differences in the structural resistance of different radiators and reflects the relative resistance increase caused by dust accumulation.

[0101] Similarly, the heat dissipation attenuation rate is calculated using the heat dissipation without dust and the heat dissipation with accumulated dust. This indicator objectively quantifies the percentage of heat exchange capacity loss caused by dust accumulation. Using the gap blockage rate (i.e., the loss rate of effective ventilation area) obtained through image recognition in step S202, the first two attenuation indicators are analyzed a second time. The pressure difference increase rate is divided by the gap blockage rate to obtain the resistance influence coefficient; the heat dissipation attenuation rate is divided by the gap blockage rate to obtain the effective area heat dissipation rate. These calculation processes essentially perform attribution analysis on performance degradation using mathematical methods, breaking down the total attenuation into the part directly caused by area loss and the residual part caused by other factors (such as roughness and thermal resistance).

[0102] Step S203 yields the drag influence coefficient and effective area heat dissipation rate. Taking data from a high-humidity sandstorm test as an example, assuming the measured pressure difference increase rate is 150%, the heat dissipation attenuation rate is 35%, and the gap blockage rate is 50%, the drag influence coefficient is calculated to be 150% / 50% = 3.0, and the effective area heat dissipation rate is 35% / 50% = 0.7. The emergence of these two derived parameters upgrades the dimension of data analysis from "what happened" to "why it happened."

[0103] The physical meaning of the drag effect coefficient lies in its measurement of the drag increase factor corresponding to the blockage rate per unit area. If the coefficient is much greater than 1, it strongly suggests that in addition to the simple reduction in flow area, dust accumulation has significantly increased the surface roughness of the fins or altered the local flow field structure, triggering severe eddies and flow separation. Conversely, if the coefficient is close to 1, it indicates that the increase in drag is mainly caused by area blockage, with less influence from morphology.

[0104] The effective area heat dissipation ratio reveals another reason for the degradation of heat dissipation performance. In the example above, an effective area heat dissipation ratio of 0.7 means that after eliminating the 50% area loss factor, the remaining 50% of the effective heat exchange area only maintains 70% of its original heat dissipation capacity per unit area. This directly proves that the dust accumulation layer itself, as a low thermal conductivity thermal resistance layer, severely hinders the transfer of heat from the fins to the air. If the effective area heat dissipation ratio is much less than 1, it indicates that the thermal resistance effect of dust accumulation is the dominant factor; if it is close to 1, it indicates that the heat dissipation degradation mainly stems from the reduction in the total amount of fluid (air) participating in heat exchange.

[0105] S204. Construct a training dataset, which includes multiple sets of training data. Each set of training data includes the operating condition type, the proportion of sand and dust particles, the current total amount of dust accumulation, the gap blockage rate, the average dust accumulation thickness, the resistance influence coefficient, the effective area heat dissipation rate, and the correlation. Input each set of training data into the initial radiator performance prediction model to obtain the predicted correlation. Calculate the loss function value between the correlation and the predicted correlation based on the preset loss function. Train the initial radiator performance prediction model according to the loss function value until the preset loss function converges to obtain the trained radiator performance prediction model.

[0106] In this embodiment, the correlation includes the pressure difference increase rate and the heat dissipation attenuation rate; the correlation also includes the predicted pressure difference increase rate and the predicted heat dissipation attenuation rate. First, a training dataset needs to be constructed. Each training data set in this dataset is a data package containing complete information from cause to effect. For example, a typical data package may include: operating condition type (e.g., "high humidity sandstorm"), dust particle ratio (e.g., PM2.5 to PM10 concentration ratio), current total dust accumulation, gap blockage rate, average dust accumulation thickness, resistance influence coefficient, effective area heat dissipation rate, and the correlation as the target output, namely, the pressure difference increase rate and the heat dissipation attenuation rate. This data comes from dust accumulation simulation experiments under various operating conditions in S201 to S203, covering the entire process from initial cleanliness to severe dust accumulation, and each data point corresponds to a precise morphology and performance state under a specific operating condition and a specific amount of dust accumulation. Subsequently, this structured training data is batch-input into the initial radiator performance prediction model. This initial model can choose algorithms suitable for processing structured data and temporal relationships, such as gradient boosting decision trees, long short-term memory networks, or graph neural networks. The model takes all other parameters except the correlation as input features, and its current task is to output the predicted correlation based on these features, namely the predicted rate of increase in pressure difference and the predicted rate of decrease in heat dissipation.

[0107] Step S204, the training process, is an iterative and optimized learning process. Each time the model calculates the predicted pressure drop increase rate and the predicted heat dissipation attenuation rate based on the input features, it immediately compares them with the actual pressure drop increase rate and heat dissipation attenuation rate measured and calculated experimentally in the data packet. This comparison is quantified using a preset loss function (such as the mean squared error loss function), and the calculated loss function value characterizes the magnitude of the model's current prediction error. The optimizer of the training system automatically adjusts tens of thousands of parameters (such as the weights and biases of the neural network) within the initial radiator performance prediction model based on this loss function value, using algorithms such as backpropagation, to make the predicted values ​​as close as possible to the true values. This process is repeated until the preset loss function converges on the validation dataset, meaning the error no longer decreases significantly. The model is then considered to have fully learned the patterns contained in the training data, at which point the trained radiator performance prediction model is obtained.

[0108] S205. Input the actual dust particle distribution parameters of the test site environment and the estimated total dust accumulation obtained from the expected operating time into the radiator performance prediction model; output the differential pressure prediction curve and heat dissipation prediction curve for the future set time period; when the differential pressure increase value in the differential pressure prediction curve exceeds the preset differential pressure threshold, or the heat dissipation decrease value in the heat dissipation prediction curve exceeds the preset heat dissipation threshold, trigger the maintenance warning.

[0109] In this embodiment, a data interface first needs to be established for the target site where a specific type of radiator is deployed (such as a thermal power plant located in a windy and sandy area). This interface continuously or periodically collects the actual dust particle distribution parameters of the environment at the site to be tested. Optionally, this can be obtained through an online particulate matter monitor installed near the air-cooling tower, which can provide time-series data on the concentration of particles of different sizes such as PM2.5 and PM10. At the same time, based on information such as on-site wind speed and operating time, the estimated total dust accumulation is obtained by integral calculation or empirical modeling from the expected operating time. These two types of data describe the characteristics of dust and the accumulation amount, respectively. These preprocessed feature data are input into the radiator performance prediction model trained in step S204. The model will perform inference calculations based on the learned complex laws about "what kind of dust and how much dust will cause what kind of performance degradation" on these on-site data, and finally dynamically output the pressure difference prediction curve and heat dissipation prediction curve for a future set time period.

[0110] Taking a power plant during its windy and dusty spring season as an example, by integrating 15-day weather forecast data (including wind speed and humidity) and typical local dust particle size distribution, the model can predict that within the next two weeks, the radiator's differential pressure prediction curve will steadily rise from its current value, accelerating upwards around day 10, potentially exceeding the maximum allowable back pressure of the fan design. Simultaneously, the heat dissipation prediction curve will show a slow but continuous downward trend. These two curves are no longer records of historical data, but rather scientific extrapolations of future performance trajectories based on physical mechanisms and data-driven approaches. The model pre-sets performance thresholds determined by engineering experience and safety regulations, such as a pre-set differential pressure threshold (e.g., 80% of the fan head margin) and a pre-set heat dissipation threshold (e.g., 85% of the rated heat dissipation). By comparing the predicted curves with these thresholds in real time, once the model-predicted increase in differential pressure (or decrease in heat dissipation) exceeds the pre-set differential pressure threshold (or pre-set heat dissipation threshold) at some future point in time, a maintenance warning will be immediately triggered. This warning is not based on a current failure, but on a prediction of potential performance exceeding limits in the coming days or even weeks.

[0111] In one possible implementation, the particle size distribution ratio of sand and dust in the dust collection device is configured based on three conditions: large particle-dominant, small particle-dominant, and mixed particle conditions. The large particle-dominant condition is used to simulate an environment where physical obstruction between fins is easily generated. The small particle-dominant condition is used to simulate an environment where a dense covering layer is easily formed on the fin surface. The mixed particle condition is used to simulate a complex environment where the spatial skeleton formed by large particles and the pore filling formed by small particles are superimposed, thereby causing a synergistic deterioration of airflow resistance and heat transfer performance.

[0112] In real-world polluted environments such as sandstorms and industrial dust, dust particles are not of a single size; their size distribution significantly influences deposition morphology and mechanisms. Therefore, this method categorizes the infinitely many possible particle size distributions into three typical and physically significant operating conditions: large particle-dominated, small particle-dominated, and mixed particle-dominated conditions. Under the large particle-dominated condition, sand and dust with a predominantly large particle size (e.g., >50μm) is introduced into the automatic feeding system of the dust collection device to simulate the environment of coarse sand and small gravel easily carried by strong winds in deserts and other areas. Due to their high inertia, these particles are more likely to directly impact the fins and remain at the fin gap inlet or surface when flowing through the radiator fin array. Their dominant effect is to create physical obstruction between the fin gaps, which primarily reduces the effective ventilation area geometrically.

[0113] In environments dominated by small particles, dust with a particle size of <10μm is configured to simulate the fine particulate matter environment of urban industrial emissions and long-distance transportation. These particles are easy to follow the airflow and form a dense coating layer on the fin surface mainly through mechanisms such as diffusion and electrostatic adsorption. Although this layer is thin, it is continuous and uniform, and mainly weakens heat transfer by increasing additional thermal resistance.

[0114] The mixed-particle scenario is the most complex and realistic, with large and small particles mixed in a certain proportion to simulate most real-world complex pollution environments. Under this scenario, large particles are deposited first; subsequently, fine particles are carried by the airflow and fill the pores of these skeletal structures, leading to increased deposition density and structural solidification. This not only macroscopically blocks the airflow channels but also makes their interiors extremely tortuous and rough, resulting in a synergistic deterioration in both physical obstruction and flow distortion.

[0115] By dividing the simulation into three typical dust storm conditions, the realism and scientific rigor of the laboratory simulation are enhanced, allowing the test results to better reflect the actual performance of the radiator in specific types of pollution environments. Secondly, a clear experimental path is provided for a deeper understanding of the dust accumulation mechanism. By comparing the test results under conditions dominated by large particles and those dominated by small particles, the differentiated impact weights of physical blockage and thermal resistance coverage on performance (especially the rate of pressure difference increase and heat dissipation attenuation) can be clearly distinguished. This embodiment, through standardized particle configuration, transforms chaotic natural phenomena into analytical, quantifiable, and predictable scientific experiments, providing a solid data and cognitive foundation for the precise design, adaptive selection, and intelligent operation and maintenance of radiators.

[0116] Figure 3 A schematic diagram of the dust accumulation performance testing device for indirect air cooling tower radiators provided in this application is shown below. Figure 3 As shown, the dust accumulation performance testing device 30 for indirect air cooling tower radiators provided in this embodiment includes:

[0117] The first processing module 301 is used to obtain the reference thermodynamic performance parameters of the radiator in a dust-free state based on the dust accumulation device.

[0118] The first processing module 301 is also used to simulate different sand and dust conditions based on the dust accumulation device, and obtain the dust accumulation thermodynamic performance parameters and dust accumulation morphology parameters of the radiator under each condition.

[0119] The second processing module 302 is used to establish the correlation between radiator performance and dust conditions based on the reference thermodynamic performance parameters, dust accumulation thermodynamic performance parameters and dust accumulation morphology characteristics parameters.

[0120] The third processing module 303 is used to construct a radiator performance prediction model based on the correlation relationship using machine learning algorithms; wherein, the radiator performance prediction model is used to predict the performance degradation trend curve of the radiator based on actual operating parameters, and the performance degradation trend curve is used to generate maintenance early warning information for the radiator.

[0121] In one possible implementation, the baseline thermal performance parameters include dust-free pressure differential and dust-free heat dissipation, and the dust accumulation thermal performance parameters include dust accumulation pressure differential and dust accumulation heat dissipation. The dust-free pressure differential is used to characterize the airflow resistance of the radiator in a dust-free state, the dust-free heat dissipation is used to characterize the heat exchange capacity of the radiator in a dust-free state, the dust accumulation pressure differential is used to characterize the degree to which dust accumulation on the ventilation channel obstructs the radiator in a dust-accumulated state, and the dust accumulation heat dissipation is used to characterize the degree to which dust accumulation weakens the heat dissipation performance of the radiator in a dust-accumulated state.

[0122] In one possible implementation, the first processing module 301 is further configured to:

[0123] The surface images of the radiator under various operating conditions are captured by a high-definition camera on the dust collection device.

[0124] Image recognition technology was used to calculate the gap blockage rate and average dust accumulation thickness of the radiator fins. The gap blockage rate was used to characterize the degree of reduction in the effective ventilation area between the radiator fins, and the average dust accumulation thickness was used to characterize the geometric scale of the dust layer attached to the radiator surface.

[0125] In one possible implementation, the second processing module 302 is further configured to:

[0126] The pressure difference increase rate is obtained based on the dust-free pressure difference in the baseline thermodynamic performance parameters and the dust accumulation pressure difference in the dust accumulation thermodynamic performance parameters.

[0127] The heat dissipation rate is obtained based on the heat dissipation of dust-free thermal performance parameters and the heat dissipation of dust-accumulated thermal performance parameters.

[0128] Based on the pressure difference increase rate and the gap blockage rate in the dust accumulation morphology characteristic parameters, the drag influence coefficient is obtained; whereby the drag influence coefficient is used to characterize the severity of the influence of dust accumulation morphology on the increase of surface roughness or local flow field distortion.

[0129] The effective area heat dissipation rate is obtained based on the heat dissipation attenuation rate and the gap blockage rate in the dust accumulation morphology characteristic parameters; whereby the effective area heat dissipation rate is used to characterize the actual heat dissipation capacity on the remaining effective heat exchange area after eliminating the factor of reduced ventilation area.

[0130] In one possible implementation, the third processing module 303 is further configured to:

[0131] A training dataset is constructed, which includes multiple sets of training data. Each set of training data includes the operating condition type, the proportion of sand and dust particles, the current total amount of dust accumulation, the gap blockage rate, the average dust accumulation thickness, the resistance influence coefficient, the effective area heat dissipation rate, and the correlation relationship. Among them, the correlation relationship includes the pressure difference increase rate and the heat dissipation attenuation rate.

[0132] Each set of training data is input into the initial radiator performance prediction model to obtain the predicted correlation; the correlation includes the predicted differential pressure increase rate and the predicted heat dissipation attenuation rate.

[0133] The loss function value between the pre-defined loss function and the predicted loss function is calculated.

[0134] The initial radiator performance prediction model is trained based on the loss function value until the preset loss function converges, resulting in a well-trained radiator performance prediction model.

[0135] In one possible implementation, the third processing module 303 is further configured to:

[0136] The actual dust particle distribution parameters of the test site environment and the estimated total dust accumulation obtained from the expected operating time are input into the radiator performance prediction model.

[0137] Output the pressure difference prediction curve and heat dissipation prediction curve for a future set time period;

[0138] A maintenance warning is triggered when the increase in differential pressure in the differential pressure prediction curve exceeds the preset differential pressure threshold, or the decrease in heat dissipation in the heat dissipation prediction curve exceeds the preset heat dissipation threshold.

[0139] In one possible implementation, the first processing module 301 is further configured to:

[0140] Based on the operating conditions dominated by large particles, small particles, and mixed particles, the particle size distribution ratio of sand and dust in the dust collection device is configured respectively.

[0141] Among them, the large particle-dominated operating condition is used to simulate an environment in which physical obstruction is easily generated between fins; the small particle-dominated operating condition is used to simulate an environment in which a dense covering layer is easily formed on the fin surface; and the mixed particle operating condition is used to simulate a complex environment in which the spatial skeleton formed by large particles and the pore filling generated by small particles are superimposed, thereby causing a synergistic deterioration of airflow resistance and heat transfer performance.

[0142] The indirect air cooling tower radiator dust accumulation performance testing device provided in this embodiment can perform the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0143] Figure 4 A hardware schematic diagram of the dust accumulation performance testing device for the indirect air cooling tower radiator provided in this application. Figure 4 As shown, the electronic device 40 provided in this embodiment includes at least one processor 401 and a memory 402. Optionally, the device 40 further includes a communication component 403. The processor 401, memory 402, and communication component 403 are connected via a bus 404.

[0144] In a specific implementation, at least one processor 401 executes computer execution instructions stored in memory 402, causing at least one processor 401 to perform the above-described method.

[0145] The specific implementation process of processor 401 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0146] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0147] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0148] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0149] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0150] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0151] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0152] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0153] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, methods, or units, and may be electrical, mechanical, or other forms.

[0154] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0155] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0156] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0157] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0158] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for testing the dust accumulation performance of an indirect air cooling tower radiator, characterized in that, include: Based on the dust collection device, the baseline thermodynamic performance parameters of the radiator in a dust-free state are obtained; Based on the dust accumulation device, different sand and dust conditions are simulated to obtain the dust accumulation thermodynamic performance parameters and dust accumulation morphology parameters of the radiator under each condition. Based on the aforementioned baseline thermodynamic performance parameters, dust accumulation thermodynamic performance parameters, and dust accumulation morphology parameters, a correlation between radiator performance and sandstorm conditions is established. Based on the aforementioned correlation, a radiator performance prediction model is constructed using machine learning algorithms; wherein, the radiator performance prediction model is used to predict the performance degradation trend curve of the radiator based on actual operating parameters, and the performance degradation trend curve is used to generate maintenance early warning information for the radiator.

2. The method according to claim 1, characterized in that, The reference thermal performance parameters include dust-free pressure difference and dust-free heat dissipation, and the dust-accumulated thermal performance parameters include dust-accumulated pressure difference and dust-accumulated heat dissipation. The dust-free pressure difference is used to characterize the airflow resistance of the radiator in the dust-free state, the dust-free heat dissipation is used to characterize the heat exchange capacity of the radiator in the dust-free state, the dust-accumulated pressure difference is used to characterize the degree of obstruction of the ventilation channel by sand and dust adhesion to the radiator in the dust-accumulated state, and the dust-accumulated heat dissipation is used to characterize the degree of weakening of heat dissipation performance of the radiator by dust accumulation in the dust-accumulated state.

3. The method according to claim 1, characterized in that, The dust accumulation morphology parameters are obtained in the following manner: The surface images of the radiator under various operating conditions are captured using a high-definition camera on the dust collection device. Image recognition technology is used to calculate the gap blockage rate and average dust accumulation thickness of the radiator fins; wherein, the gap blockage rate is used to characterize the degree of reduction in the effective ventilation area between the radiator fins; and the average dust accumulation thickness is used to characterize the geometric scale of the dust layer attached to the radiator surface.

4. The method according to claim 1, characterized in that, The establishment of the correlation between radiator performance and dust conditions based on the aforementioned baseline thermodynamic performance parameters, dust accumulation thermodynamic performance parameters, and dust accumulation morphology characteristics parameters includes: Based on the dust-free pressure difference in the reference thermodynamic performance parameters and the dust-accumulated pressure difference in the thermodynamic performance parameters, the pressure difference increase rate is obtained; The heat dissipation rate is obtained based on the dust-free heat dissipation in the reference thermodynamic performance parameters and the dust-accumulated heat dissipation in the dust-accumulated thermodynamic performance parameters. Based on the pressure difference increase rate and the gap blockage rate in the dust accumulation morphology characteristic parameters, a resistance influence coefficient is obtained; wherein, the resistance influence coefficient is used to characterize the severity of the influence of dust accumulation morphology on the increase of surface roughness or local flow field distortion. Based on the heat dissipation attenuation rate and the gap blockage rate in the dust accumulation morphology characteristic parameters, the effective area heat dissipation rate is obtained; wherein, the effective area heat dissipation rate is used to characterize the actual heat dissipation capacity on the remaining effective heat exchange area after eliminating the factor of reduced ventilation area.

5. The method according to claim 4, characterized in that, The step of constructing a radiator performance prediction model based on the aforementioned correlation using machine learning algorithms includes: A training dataset is constructed, which includes multiple sets of training data. Each set of training data includes the operating condition type, the proportion of sand and dust particles, the current total amount of dust accumulation, the gap blockage rate, the average dust accumulation thickness, the resistance influence coefficient, the effective area heat dissipation rate, and the correlation relationship. The correlation relationship includes the pressure difference increase rate and the heat dissipation attenuation rate. Each set of training data is input into the initial radiator performance prediction model to obtain the predicted correlation; wherein, the correlation includes the predicted differential pressure increase rate and the predicted heat dissipation attenuation rate; The loss function value between the correlation and the predicted correlation is calculated based on a preset loss function; The initial radiator performance prediction model is trained based on the loss function value until the preset loss function converges, thus obtaining the trained radiator performance prediction model.

6. The method according to claim 5, characterized in that, Also includes: The actual dust particle distribution parameters of the test site environment and the estimated total dust accumulation obtained from the expected operating time are input into the radiator performance prediction model. Output the pressure difference prediction curve and heat dissipation prediction curve for a future set time period; The maintenance warning is triggered when the pressure difference increase in the pressure difference prediction curve exceeds the preset pressure difference threshold, or the heat dissipation decrease in the heat dissipation prediction curve exceeds the preset heat dissipation threshold.

7. The method according to any one of claims 1-6, characterized in that, Also includes: Based on the operating conditions dominated by large particles, small particles, and mixed particles, the particle size distribution ratio of sand and dust in the dust collection device is configured respectively. The large particle-dominated operating condition is used to simulate an environment where physical obstruction between fins is easily generated; the small particle-dominated operating condition is used to simulate an environment where a dense covering layer is easily formed on the fin surface; and the mixed particle operating condition is used to simulate a complex environment where the spatial skeleton formed by large particles and the pore filling generated by small particles are superimposed, thereby causing a synergistic deterioration of airflow resistance and heat transfer performance.

8. A device for testing the dust accumulation performance of an indirect air cooling tower radiator, characterized in that, include: The first processing module is used to obtain the baseline thermodynamic performance parameters of the radiator in a dust-free state based on the dust accumulation device. The first processing module is also used to simulate different sandstorm conditions based on the dust accumulation device, and obtain the dust accumulation thermodynamic performance parameters and dust accumulation morphology parameters of the radiator under each condition. The second processing module is used to establish the correlation between radiator performance and sandstorm conditions based on the benchmark thermodynamic performance parameters, dust accumulation thermodynamic performance parameters, and dust accumulation morphology characteristic parameters. The third processing module is used to construct a radiator performance prediction model based on the correlation relationship using machine learning algorithms; wherein, the radiator performance prediction model is used to predict the performance degradation trend curve of the radiator based on actual operating parameters, and the performance degradation trend curve is used to generate maintenance early warning information for the radiator.

9. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-7.

Images