Method and system for predicting salt mist airflow field in marine atmospheric environment

Abstract

The invention discloses a salt mist airflow field prediction method and system in an ocean atmospheric environment, and relates to the technical field of ocean wind power plant salt mist corrosion trend prediction.The system divides a wind power plant into a plurality of salt mist monitoring areas, collects environmental data, constructs a salt mist airflow field model by adopting three-dimensional modeling and CFD simulation, and obtains the salt mist corrosion trend of the wind power plant. A digital twinborn model is generated through multi-physics field coupling, and dynamic deduction is carried out; constructing an AI corrosion prediction model by using historical data and simulated deduction data, extracting a corrosion risk feature vector FCV, and performing dynamic training to predict a corrosion trend; salt mist concentration and wind speed data are monitored in real time, a corrosion coefficient is calculated and compared with a threshold value for analysis, and a corresponding protection strategy is formulated; in combination with real-time data and digital twinborn deduction, evaluating a wind-driven salt mist coupling deposition risk, triggering early warning and optimizing a strategy; and a test report is automatically generated, future test parameters are optimized through an AI prediction model, and the corrosion resistance and the operation efficiency of the wind turbine are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of salt spray corrosion trend prediction for marine wind farms, and in particular to a method and system for predicting a salt spray airflow field in a marine atmospheric environment. Background Art

[0002] With the rapid development of offshore wind power, the impact of the marine atmosphere on wind turbines is becoming increasingly prominent, particularly the corrosive effects of salt spray. This salt spray not only accelerates the corrosion of metal components, severely impacting the long-term stable operation of wind turbines, but also increases maintenance costs and the frequency of equipment replacement. In offshore wind farms, the impact of salt spray concentration and climate change is particularly significant. Therefore, accurately predicting the salt spray flow field in the marine atmosphere has become crucial for ensuring the long-term stable operation of offshore wind turbines.

[0003] Currently, salt spray monitoring for offshore wind farms relies primarily on traditional meteorological monitoring methods and empirical models. However, these methods fail to fully account for the complex environmental factors and multiple impacts of wind turbine equipment, resulting in significant errors in salt spray corrosion assessments. Furthermore, the dynamic changes in the salt spray airflow field and the complex interactions with wind turbine equipment make traditional single-site monitoring and analysis methods difficult to provide a comprehensive and accurate corrosion risk assessment.

[0004] Therefore, there is an urgent need to develop a salt spray airflow field prediction system that integrates multiple sensing technologies, digital twin modeling and artificial intelligence prediction. Summary of the Invention

[0005] In view of the deficiencies of the prior art, the present invention provides a method and system for predicting salt fog airflow fields in a marine atmospheric environment to solve the problems mentioned in the background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a salt fog airflow field prediction system in a marine atmospheric environment, including a data acquisition module, a digital twin modeling module, an AI corrosion prediction model construction module, a salt fog concentration intelligent monitoring module, a wind-driven salt fog coupled deposition intelligent monitoring module, and an automated feedback tuning module;

[0007] The data acquisition module is used to divide the offshore wind farm into a number of salt fog monitoring areas, monitor the salt fog monitoring areas, and collect environmental data;

[0008] The digital twin modeling module is used to construct a salt fog airflow field model in the marine atmosphere based on environmental data using three-dimensional modeling and CFD simulation. It combines salt fog distribution, concentration gradient, and airflow humidity characteristics to simulate the deposition and corrosion effects of salt fog on wind power equipment. It integrates multiple physical fields to form a coupled digital twin model, enabling dynamic deduction under different climate conditions and obtaining simulation data.

[0009] The AI ​​corrosion prediction model construction module is used to build a convolutional neural network initial model CNN-Init based on marine atmospheric environmental data and simulation deduction data. It uses historical salt spray data, temperature and humidity parameters, and corrosion rate as input to extract spatiotemporal features and build an initial model with spatial diffusion and corrosion perception capabilities. Through joint training of multi-source coupled data, it extracts the corrosion risk feature vector FCV to characterize the nonlinear relationship between salt spray, climate, and material response. The FCV is input into the AI-Adaptive-Corrosion-Predictor for dynamic training to predict corrosion trends.

[0010] The salt spray concentration intelligent monitoring module is used to monitor the changes in the salt spray concentration in the environment in real time, combine the simulation data and the actual environmental data, calculate the salt spray concentration corrosion coefficient SCCX, and compare and analyze it with the first threshold Q1 to determine whether the salt spray concentration is normal. If it is abnormal, a strategy is given;

[0011] The wind-driven salt spray coupled deposition intelligent monitoring module is used to calculate the wind-driven salt spray coupled deposition coefficient WDCX based on the wind speed, wind direction angle, wind field viscosity parameters, equipment surface roughness and salt spray concentration data obtained by real-time monitoring, combined with the simulation deduction data output by the digital twin modeling module, and compare and analyze it with the second threshold Q2 to determine whether the wind turbine equipment has the risk of abnormal corrosion, and provide a strategy if there is a risk;

[0012] The automated feedback tuning module is used to automatically generate a test report after the test, based on simulation data, actual monitoring data, and salt spray concentration analysis, including test parameters, salt spray concentration, climate trends, and corrosion assessments. It also establishes an AI prediction model through historical test data and conducts model training to achieve pattern recognition, trend analysis, and anomaly detection of the salt spray airflow field, predict the impact of future environmental changes on wind turbine performance, and optimize test parameters.

[0013] Preferably, the data acquisition module includes a region division unit and a collection unit; the region division unit is used to obtain three-dimensional spatial geographic information, historical sea condition information and wind power equipment information of the offshore wind farm, draw an electronic distribution map of the offshore wind farm, and divide the offshore wind farm into a number of salt fog monitoring areas, and mark the several salt fog monitoring areas on the electronic distribution map as: Jcq1, Jcq2, ..., Jcqn; n represents the number of salt fog monitoring areas;

[0014] The acquisition unit is used to monitor the salt fog monitoring area in real time, and collects wind speed and direction by installing an ultrasonic anemometer on the top of the wind turbine tower; collects salt fog concentration by installing a corrosion environment salinity sensor on the edge of the wind turbine blade; and collects environmental humidity values ​​by installing a humidity sensor on the windward side of the wind turbine; and establishes an environmental data set.

[0015] Preferably, the digital twin modeling module is used to utilize environmental data set data to reconstruct the salt fog airflow field in the marine atmospheric environment through three-dimensional modeling; CFD computational fluid dynamics is used to simulate air flow, and the salt fog distribution and concentration gradient unique to the marine environment are combined to construct a salt fog airflow field model; by considering the humidity of the airflow, the condensation and diffusion characteristics of salt particles, the effect of salt fog on the deposition and corrosion of metal parts and wind turbine blade surfaces is simulated; and the airflow and corrosion simulations are integrated to construct a digital twin model based on the coupling of multiple physical fields of airflow, salt fog and humidity, so as to realize the dynamic deduction of the salt fog airflow field of offshore wind turbines under different climatic conditions and obtain simulation deduction data.

[0016] Preferably, the AI ​​corrosion prediction model construction module is used to construct a convolutional neural network initial model CNN-Init based on the environmental data of the marine atmosphere and the simulation deduction data; the CNN-Init model uses historical salt spray concentration change data, measured temperature and humidity parameters and equipment surface corrosion rate data as input samples, and uses multi-layer convolution kernels to extract spatiotemporal distribution characteristics to construct an initial model with spatial diffusion and corrosion dynamic perception capabilities; then, a data input structure based on multi-source coupling is introduced, including atmospheric corrosion level data and equipment material property data, and the CNN-Init model is jointly trained and tested. During the model training process, the intermediate feature layer output of the convolutional neural network is extracted as the corrosion risk feature vector Feature-Corrosion-Vector, FCV, which describes the nonlinear mapping relationship between salt spray deposition pattern, climate corrosion factor and material response; the FCV feature vector is input into the AI-Adaptive-Corrosion-Predictor, and after multiple rounds of dynamic training and regression testing, it is used as an AI corrosion prediction model to predict corrosion trends.

[0017] Preferably, the salt spray concentration intelligent monitoring module includes a first calculation unit and a first analysis unit;

[0018] The first calculation unit is used to monitor the change of salt spray concentration in the environment in real time, and calculate the salt spray concentration corrosion coefficient SCCX after dimensionless processing by combining simulation data and actual environmental data. The formula is as follows:

[0019]

[0020] Where C trepresents the salt spray concentration at time t, C0 represents the reference salt spray concentration, obtained by simulation, T t represents the time-varying trend factor of salt spray concentration at time t, T0 represents the initial trend factor of historical salt spray concentration, obtained by simulation deduction, and H t represents the ambient humidity value at time t, H0 represents the reference ambient humidity value, obtained by simulation, and V t represents the wind speed value at time t, V0 represents the reference environmental wind speed value, which is obtained by simulation, L represents the wind turbine blade area, which is obtained by wind turbine parameters, and D t It represents the corrosion resistance score of the wind turbine at time t, which is obtained through historical data. α represents the adjustment coefficient, which is obtained through experiments.

[0021] Preferably, the first analysis unit is used to preset a first threshold value Q1 in advance, and compare and analyze the salt spray concentration corrosion coefficient SCCX with the first threshold value Q1, and obtaining the first evaluation result includes:

[0022] When the salt spray concentration corrosion coefficient SCCX ≤ the first threshold Q1, it means that the concentration of salt spray is normal and there is no risk of accelerated corrosion of wind turbines, and continuous monitoring is required;

[0023] When the salt spray concentration corrosion coefficient SCCX is greater than the first threshold Q1, it indicates that the salt spray concentration is abnormal and there is a risk of accelerated corrosion of the wind turbine. This triggers the first warning instruction and generates the first strategy: adding anti-corrosion coating to the wind turbine equipment and increasing the frequency of regular maintenance by 50%.

[0024] Preferably, the wind-driven salt spray coupled deposition intelligent monitoring module includes a second calculation unit and a second analysis unit;

[0025] The second calculation unit is used to calculate the wind-driven salt spray coupling deposition coefficient WDCX based on the wind speed, wind direction angle, wind field viscosity parameters, equipment surface roughness and salt spray concentration data obtained through real-time monitoring, combined with the simulation deduction data output by the digital twin modeling module, after dimensionless processing. The formula is as follows:

[0026]

[0027] Where V t represents the wind speed value at time t, V0 represents the reference environment wind speed value, θ t It represents the angle between the wind direction at time t and the normal line of the surface of the wind turbine equipment, R s Represents the surface roughness parameter of the wind turbine, R0 represents the roughness reference value, C t represents the salt spray concentration at time t, C0 represents the reference salt spray concentration, U trepresents the wind field viscosity coefficient at time t, which is obtained by simulation deduction, U0 represents the wind field viscosity benchmark value, γ represents the empirical adjustment coefficient, which is obtained by training the AI ​​corrosion prediction model, w1, w2, w3, and w4 represent weight coefficients, which are obtained by training the AI ​​corrosion prediction model.

[0028] Preferably, the second analysis unit is used to preset a second threshold value Q2 in advance, and compare and analyze the wind-driven salt spray coupling deposition coefficient WDCX with the second threshold value Q2, and obtaining the second evaluation result includes:

[0029] When the wind-driven salt spray coupling deposition coefficient WDCX is less than the second threshold Q2, it indicates that the surface load of the wind turbine equipment is within the normal range, there is no risk of abnormal corrosion, and continuous monitoring is required;

[0030] When the wind-driven salt spray coupling deposition coefficient WDCX ≥ the second threshold Q2, it indicates that the surface attachment load of the wind turbine equipment is not within the normal range and there is a risk of abnormal corrosion. The second early warning instruction is triggered and the second strategy is generated: the self-cleaning material replacement rate is increased to 60%, the micro-nano structure roughness is adjusted by 20%, the structural orientation is fine-tuned by ±15°, and the frequency of regular maintenance is increased by 60%.

[0031] Preferably, the automated feedback tuning module is used to automatically generate a detailed test report after the test is completed, based on the simulation data, actual monitoring data and salt spray concentration analysis results, which includes test parameters, salt spray concentration and climate change trends, equipment corrosion assessment and optimization strategies; establish an AI prediction model based on historical test data through a deep learning model, and train the model through test report data to achieve pattern recognition, trend analysis and anomaly detection of the salt spray airflow field, predict the impact of future marine environment changes on wind turbine performance, and optimize the test parameter combination.

[0032] Preferably, a method for predicting a salt fog airflow field in a marine atmospheric environment comprises the following steps:

[0033] Step 1: Divide the offshore wind farm into several salt spray monitoring areas, monitor the salt spray monitoring areas, and collect environmental data;

[0034] Step 2: Based on environmental data, a salt fog airflow field model in the marine atmosphere is constructed using 3D modeling and CFD simulation. The effects of salt fog on the deposition and corrosion of wind turbines are simulated by combining salt fog distribution, concentration gradient, and airflow humidity characteristics. Multi-physics fields are integrated to form a coupled digital twin model, enabling dynamic deduction under different climate conditions and obtaining simulation data.

[0035] Step 3: Based on marine atmospheric environmental data and simulation data, the initial convolutional neural network model CNN-Init is constructed. Historical salt spray data, temperature and humidity parameters, and corrosion rate are used as input to extract spatiotemporal features and construct an initial model with spatial diffusion and corrosion perception capabilities. Through joint training of multi-source coupled data, the corrosion risk feature vector FCV is extracted to characterize the nonlinear relationship between salt spray, climate, and material response. The FCV is input into the AI-Adaptive-Corrosion-Predictor for dynamic training to predict corrosion trends.

[0036] Step 4: Monitor the changes in salt spray concentration in the environment in real time. Combine the simulation data with the actual environmental data to calculate the salt spray concentration corrosion coefficient SCCX, and compare and analyze it with the first threshold Q1 to determine whether the salt spray concentration is normal. If it is abnormal, a strategy is given;

[0037] Step 5: Based on the wind speed, wind direction, wind field viscosity parameters, equipment surface roughness, and salt spray concentration data obtained through real-time monitoring, combined with the simulation data output by the digital twin modeling module, the wind-driven salt spray coupling deposition coefficient WDCX is calculated and compared with the second threshold Q2 to determine whether the wind turbine equipment has the risk of abnormal corrosion. If there is a risk, a strategy is implemented.

[0038] Step 6. After the test, a test report is automatically generated based on simulation data, actual monitoring data and salt spray concentration analysis, including test parameters, salt spray concentration, climate trends and corrosion assessment; an AI prediction model is established through historical test data, and model training is performed to achieve pattern recognition, trend analysis and anomaly detection of the salt spray airflow field, predict the impact of future environmental changes on wind turbine performance, and optimize test parameters.

[0039] The present invention provides a method and system for predicting salt fog airflow in a marine atmosphere. It has the following beneficial effects:

[0040] (1) This method and system for predicting salt fog airflow fields in a marine atmosphere accurately simulate and predict the dynamic changes of salt fog airflow fields by integrating multiple sensing technologies, digital twin modeling, and a convolutional neural network (CNN) prediction model. This method can adjust the influence of environmental factors such as salt fog concentration, wind speed, and humidity in real time under different climatic conditions, thereby providing accurate corrosion risk assessment for wind power equipment. Compared with traditional single monitoring and empirical models, the prediction results are more accurate, effectively reducing equipment failures and maintenance costs caused by salt fog corrosion.

[0041] (2) A method and system for predicting salt fog airflow fields in a marine atmospheric environment, a salt fog concentration intelligent monitoring module, and a wind-driven salt fog coupled deposition intelligent monitoring module provide real-time monitoring of salt fog concentration and equipment surface corrosion risk. By calculating the salt fog concentration corrosion coefficient SCCX and the wind-driven salt fog coupled deposition coefficient WDCX and comparing them with the set thresholds, the system can immediately trigger an early warning when an anomaly occurs and automatically generate targeted protection strategies (such as adding coatings, increasing maintenance frequency, etc.), effectively ensuring the safety and stability of wind power equipment.

[0042] (3) This method and system for predicting salt fog airflow fields in a marine atmospheric environment is based on multiple factors such as salt fog concentration, wind speed, and equipment surface roughness, combined with digital twins and AI corrosion prediction models, to provide personalized protection strategies for wind power equipment. When the system detects abnormal salt fog concentration or increased corrosion risk, it can automatically adjust protective measures (such as adjusting material corrosion resistance, increasing the replacement rate of self-cleaning materials, optimizing equipment surface structure, etc.) based on real-time data and prediction results. This customized protection strategy can effectively extend the service life of equipment, reduce maintenance costs, and improve the operating efficiency and reliability of wind farms.

[0043] (4) A method and system for predicting salt fog airflow fields in a marine atmospheric environment. The automated feedback tuning module can automatically generate a detailed test report after the actual monitoring and testing, and optimize the prediction model based on historical data and simulation results. Through the training of the deep learning model, the system can continuously improve the pattern recognition, trend analysis, and anomaly detection capabilities of the salt fog airflow field. This dynamic tuning mechanism ensures that the system can continuously adapt to new environmental changes during long-term operation, thereby improving the flexibility and reliability of the prediction system. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a block diagram and flow chart of a system for predicting salt fog airflow field in a marine atmosphere environment according to the present invention;

[0045] Figure 2 The figure is a schematic diagram of the steps of a method for predicting salt fog airflow field in a marine atmospheric environment according to the present invention. DETAILED DESCRIPTION

[0046] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0047] Example 1

[0048] See also Figure 1 , the present invention provides a salt fog airflow field prediction system in a marine atmospheric environment, including a data acquisition module, a digital twin modeling module, an AI corrosion prediction model construction module, a salt fog concentration intelligent monitoring module, a wind-driven salt fog coupled deposition intelligent monitoring module and an automated feedback tuning module;

[0049] The data acquisition module is used to divide the offshore wind farm into a number of salt fog monitoring areas, monitor the salt fog monitoring areas, and collect environmental data;

[0050] The digital twin modeling module is used to construct a salt fog airflow field model in the marine atmosphere based on environmental data using three-dimensional modeling and CFD simulation. It combines salt fog distribution, concentration gradient, and airflow humidity characteristics to simulate the deposition and corrosion effects of salt fog on wind power equipment. It integrates multiple physical fields to form a coupled digital twin model, enabling dynamic deduction under different climate conditions and obtaining simulation data.

[0051] The AI ​​corrosion prediction model construction module is used to build a convolutional neural network initial model CNN-Init based on marine atmospheric environmental data and simulation deduction data. It uses historical salt spray data, temperature and humidity parameters, and corrosion rate as input to extract spatiotemporal features and build an initial model with spatial diffusion and corrosion perception capabilities. Through joint training of multi-source coupled data, it extracts the corrosion risk feature vector FCV to characterize the nonlinear relationship between salt spray, climate, and material response. The FCV is input into the AI-Adaptive-Corrosion-Predictor for dynamic training to predict corrosion trends.

[0052] The salt spray concentration intelligent monitoring module is used to monitor the changes in the salt spray concentration in the environment in real time, combine the simulation data and the actual environmental data, calculate the salt spray concentration corrosion coefficient SCCX, and compare and analyze it with the first threshold Q1 to determine whether the salt spray concentration is normal. If it is abnormal, a strategy is given;

[0053] The wind-driven salt spray coupled deposition intelligent monitoring module is used to calculate the wind-driven salt spray coupled deposition coefficient WDCX based on the wind speed, wind direction angle, wind field viscosity parameters, equipment surface roughness and salt spray concentration data obtained by real-time monitoring, combined with the simulation deduction data output by the digital twin modeling module, and compare and analyze it with the second threshold Q2 to determine whether the wind turbine equipment has the risk of abnormal corrosion, and provide a strategy if there is a risk;

[0054] The automated feedback tuning module is used to automatically generate a test report after the test, based on simulation data, actual monitoring data, and salt spray concentration analysis, including test parameters, salt spray concentration, climate trends, and corrosion assessments. It also establishes an AI prediction model through historical test data and conducts model training to achieve pattern recognition, trend analysis, and anomaly detection of the salt spray airflow field, predict the impact of future environmental changes on wind turbine performance, and optimize test parameters.

[0055] In this embodiment, by integrating modules such as data acquisition, digital twin modeling, AI-powered corrosion prediction, salt spray concentration monitoring, wind-driven salt spray-coupled deposition monitoring, and automated feedback optimization, it is possible to provide comprehensive, real-time, and accurate corrosion risk prediction and management for offshore wind farms. In particular, based on real-time monitoring and dynamic simulation, this system intelligently assesses salt spray concentration, wind speed, and corrosion risk based on environmental data and simulation results. It also automatically generates optimization strategies based on historical data, such as anti-corrosion coating optimization and maintenance frequency adjustment. This significantly improves equipment safety and operational stability, extends the service life of wind turbines, and effectively reduces maintenance and replacement costs.

[0056] Example 2. This example is an explanation of Example 1. Specifically, the data acquisition module includes a region division unit and an acquisition unit; the region division unit is used to obtain three-dimensional spatial geographic information, historical sea condition information, and wind power equipment information of the offshore wind farm, draw an electronic distribution map of the offshore wind farm, and divide the offshore wind farm into a number of salt fog monitoring areas. The salt fog monitoring areas are marked on the electronic distribution map as: Jcq1, Jcq2, ..., Jcqn; n represents the number of salt fog monitoring areas;

[0057] The acquisition unit is used to monitor the salt fog monitoring area in real time, and collects wind speed and direction by installing an ultrasonic anemometer on the top of the wind turbine tower; collects salt fog concentration by installing a corrosion environment salinity sensor on the edge of the wind turbine blade; and collects environmental humidity values ​​by installing a humidity sensor on the windward side of the wind turbine; and establishes an environmental data set.

[0058] In this embodiment, through the collaborative work of the regional division unit and the acquisition unit in the data acquisition module, the system can efficiently divide the offshore wind farm into multiple salt fog monitoring areas and monitor key environmental data such as wind speed, wind direction, salt fog concentration, and humidity in real time within each area. This precise environmental data acquisition method provides accurate basic data support for subsequent salt fog airflow field modeling and corrosion prediction, ensuring accurate assessment of the corrosion risk of wind turbine equipment and providing a data basis for real-time adjustment of anti-corrosion strategies.

[0059] Example 3. This example is an explanation of Example 2. Specifically, the digital twin modeling module is used to use environmental data set data to reconstruct the salt fog airflow field in the marine atmospheric environment through three-dimensional modeling; CFD computational fluid dynamics is used to simulate air flow, and the salt fog distribution and concentration gradient unique to the marine environment are combined to construct a salt fog airflow field model; by considering the humidity of the airflow, the condensation and diffusion characteristics of salt particles, the effect of salt fog on the deposition and corrosion of metal parts and wind turbine blades is simulated; and the airflow and corrosion simulations are integrated to construct a digital twin model based on the coupling of multiple physical fields of airflow, salt fog and humidity, so as to realize the dynamic deduction of the salt fog airflow field of offshore wind turbines under different climatic conditions and obtain simulation deduction data.

[0060] In this embodiment, by combining a digital twin modeling module with environmental datasets and CFD fluid dynamics simulations, the system accurately reconstructs the salt fog flow field in the marine atmosphere and simulates the effects of salt fog on wind turbine equipment deposition and corrosion. This model takes into account airflow humidity, salt particle condensation, and diffusion characteristics, enabling dynamic simulation of salt fog flow fields under different climatic conditions and providing accurate simulation data for corrosion risk assessment of wind turbine equipment. This technology can effectively improve the operational reliability of wind turbine equipment in marine environments, optimize anti-corrosion strategies, and predict the impact of salt fog atmosphere changes on the long-term stability of equipment, thereby reducing maintenance costs and extending equipment life.

[0061] Example 4. This example is explained in Example 8. Specifically, the AI ​​corrosion prediction model construction module is used to construct a convolutional neural network initial model CNN-Init based on marine atmospheric environmental data and simulation deduction data; the CNN-Init model uses historical salt spray concentration change data, measured temperature and humidity parameters, and equipment surface corrosion rate data as input samples, and uses multi-layer convolution kernels to extract spatiotemporal distribution characteristics to construct an initial model with spatial diffusion and corrosion dynamic perception capabilities; then, a data input structure based on multi-source coupling is introduced, including atmospheric corrosion level data and equipment material property data, and the CNN-Init model is jointly trained and tested. During the model training process, the intermediate feature layer output of the convolutional neural network is extracted as the corrosion risk feature vector Feature-Corrosion-Vector, FCV, which describes the nonlinear mapping relationship between salt spray deposition pattern, climate corrosion factor and material response; the FCV feature vector is input into the AI-Adaptive-Corrosion-Predictor, and after multiple rounds of dynamic training and regression testing, it is used as an AI corrosion prediction model to predict corrosion trends.

[0062] In this embodiment, by constructing an AI corrosion prediction model based on a convolutional neural network (CNN), the system can accurately extract the spatiotemporal characteristics of multi-dimensional data such as historical salt spray concentration, temperature and humidity changes, and equipment surface corrosion rate, and construct a model with spatial diffusion and corrosion dynamic perception capabilities. By introducing atmospheric corrosion levels and equipment material property data for multi-source coupling training, the model can accurately capture the complex nonlinear relationship between salt spray deposition, climate factors, and material response. By extracting the corrosion risk feature vector (FCV) and through multiple rounds of dynamic training and regression testing, the AI ​​corrosion prediction model can predict the corrosion trend of wind power equipment in real time.

[0063] Example 5, this example is an explanation of Example 4. Specifically, the salt spray concentration intelligent monitoring module includes a first calculation unit and a first analysis unit;

[0064] The first calculation unit is used to monitor the change of salt spray concentration in the environment in real time, and calculate the salt spray concentration corrosion coefficient SCCX after dimensionless processing by combining simulation data and actual environmental data. The formula is as follows:

[0065]

[0066] Where C t represents the salt spray concentration at time t, C0 represents the reference salt spray concentration, obtained by simulation, T t represents the time-varying trend factor of salt spray concentration at time t, T0 represents the initial trend factor of historical salt spray concentration, obtained by simulation deduction, and H t represents the ambient humidity value at time t, H0 represents the reference ambient humidity value, obtained by simulation, and V t represents the wind speed value at time t, V0 represents the reference environmental wind speed value, which is obtained by simulation, L represents the wind turbine blade area, which is obtained by wind turbine parameters, and D t It represents the corrosion resistance score of the wind turbine at time t, which is obtained through historical data. α represents the adjustment coefficient, which is obtained through experiments.

[0067] In this example, by monitoring changes in salt spray concentration in real time and combining simulations with actual environmental data, the salt spray concentration corrosion coefficient (SCCX) is accurately calculated. This calculation method comprehensively considers environmental parameters such as salt spray concentration, humidity, and wind speed, as well as the corrosion tolerance of wind turbine blades, enabling a dynamic assessment of the salt spray corrosion risk to wind turbine equipment. By introducing dimensionless processing and a time-varying trend factor, the monitoring results are more accurate and adaptable, reflecting the degree of corrosion in real time under different climate conditions. This provides a scientific basis for wind turbine operation, maintenance, and protective measures, and provides early warning of potential corrosion risks.

[0068] Example 6: This example is an explanation of Example 5. Specifically, the first analysis unit is used to preset a first threshold value Q1 in advance, and compare and analyze the salt spray concentration corrosion coefficient SCCX with the first threshold value Q1. Obtaining a first evaluation result includes:

[0069] When the salt spray concentration corrosion coefficient SCCX ≤ the first threshold Q1, it means that the concentration of salt spray is normal and there is no risk of accelerated corrosion of wind turbines, and continuous monitoring is required;

[0070] When the salt spray concentration corrosion coefficient SCCX is greater than the first threshold Q1, it indicates that the salt spray concentration is abnormal and there is a risk of accelerated corrosion of the wind turbine. This triggers the first warning instruction and generates the first strategy: adding anti-corrosion coating to the wind turbine equipment and increasing the frequency of regular maintenance by 50%.

[0071] In this embodiment, accurate corrosion risk assessment can be achieved by comparing and analyzing the salt spray concentration corrosion coefficient SCCX with a preset first threshold value Q1. When SCCX exceeds the threshold, the system promptly triggers an early warning instruction and generates targeted protection strategies, such as adding anti-corrosion coatings and increasing maintenance frequency. This intelligent monitoring and early warning mechanism can effectively prevent excessive corrosion of wind turbines, improve the operating life and safety of equipment, reduce unplanned downtime and maintenance costs, and ensure the continued stable operation of wind farms, as shown in the following table:

[0072]

[0073] Example 7, this example is explained in Example 8. Specifically, the wind-driven salt spray coupled deposition intelligent monitoring module includes a second calculation unit and a second analysis unit;

[0074] The second calculation unit is used to calculate the wind-driven salt spray coupling deposition coefficient WDCX based on the wind speed, wind direction angle, wind field viscosity parameters, equipment surface roughness and salt spray concentration data obtained through real-time monitoring, combined with the simulation deduction data output by the digital twin modeling module, after dimensionless processing. The formula is as follows:

[0075]

[0076] Where V t represents the wind speed value at time t, V0 represents the reference environment wind speed value, θ t It represents the angle between the wind direction at time t and the normal line of the surface of the wind turbine equipment, R s Represents the surface roughness parameter of the wind turbine, R0 represents the roughness reference value, C t represents the salt spray concentration at time t, C0 represents the reference salt spray concentration, U trepresents the wind field viscosity coefficient at time t, which is obtained by simulation deduction, U0 represents the wind field viscosity benchmark value, γ represents the empirical adjustment coefficient, which is obtained by training the AI ​​corrosion prediction model, w1, w2, w3, and w4 represent weight coefficients, which are obtained by training the AI ​​corrosion prediction model.

[0077] In this example, by integrating multiple real-time data points, such as wind speed, wind direction, wind field viscosity, equipment surface roughness, and salt spray concentration, and combining them with simulation results from the digital twin modeling module, the wind-driven salt spray coupling deposition coefficient (WDCX) is accurately calculated. This calculation dynamically reflects the impact of factors such as wind speed and salt spray concentration on salt spray deposition on the wind turbine surface, helping to assess the potential corrosion risk of the equipment in real time. This refined monitoring and analysis can provide data support for wind turbine equipment protection and maintenance strategies, optimize anti-corrosion measures, and extend the equipment's service life.

[0078] Example 8: This example is an explanation of Example 7. Specifically, the second analysis unit is used to preset a second threshold value Q2 in advance, and compare and analyze the wind-driven salt spray coupling deposition coefficient WDCX with the second threshold value Q2. Obtaining the second evaluation result includes:

[0079] When the wind-driven salt spray coupling deposition coefficient WDCX is less than the second threshold Q2, it indicates that the surface load of the wind turbine equipment is within the normal range, there is no risk of abnormal corrosion, and continuous monitoring is required;

[0080] When the wind-driven salt spray coupling deposition coefficient WDCX ≥ the second threshold Q2, it indicates that the surface attachment load of the wind turbine equipment is not within the normal range and there is a risk of abnormal corrosion. The second early warning instruction is triggered and the second strategy is generated: the self-cleaning material replacement rate is increased to 60%, the micro-nano structure roughness is adjusted by 20%, the structural orientation is fine-tuned by ±15°, and the frequency of regular maintenance is increased by 60%.

[0081] In this embodiment, by comparing the wind-driven salt spray coupling deposition coefficient WDCX with a preset second threshold value Q2, it is possible to promptly identify whether there is an abnormal corrosion risk on the surface of the wind turbine equipment. When WDCX exceeds the set threshold, the system automatically triggers an early warning and implements targeted protection strategies, such as increasing the self-cleaning material replacement rate, adjusting the roughness of the micro-nanostructure, and fine-tuning the structural orientation. These measures effectively reduce the corrosive effects of salt spray on the equipment, enhance the equipment's corrosion resistance, and ensure long-term stable operation by increasing maintenance frequency. This process can significantly extend the service life of wind turbines and reduce maintenance costs, as shown in the following table:

[0082]

[0083] Example 9. This example is an explanation of Example 8. Specifically, the automated feedback tuning module is used to automatically generate a detailed test report after the test is completed based on simulation data, actual monitoring data and salt spray concentration analysis results, including test parameters, salt spray concentration and climate change trends, equipment corrosion assessment and optimization strategies; establish an AI prediction model based on historical test data through a deep learning model, and train the model through test report data to achieve pattern recognition, trend analysis and anomaly detection of the salt spray airflow field, predict the impact of future marine environment changes on wind turbine performance, and optimize the test parameter combination.

[0084] In this embodiment, the automated feedback tuning module combines simulation data, actual monitoring data, and salt spray concentration analysis results to automatically generate a detailed test report after the test, comprehensively assessing the impact of salt spray on wind turbines and providing targeted optimization strategies. Through deep learning model training and prediction, the system can identify patterns in salt spray airflow fields, analyze climate trends, detect abnormal changes, and predict the potential impact of future marine environmental changes on wind turbine performance. This process not only provides accurate predictions and decision-making support for future testing and equipment maintenance, but also optimizes test parameter combinations.

[0085] Example 10, a method for predicting salt fog airflow field in marine atmospheric environment, please refer to Figure 2 , including the following steps:

[0086] Step 1: Divide the offshore wind farm into several salt spray monitoring areas, monitor the salt spray monitoring areas, and collect environmental data;

[0087] Step 2: Based on environmental data, a salt fog airflow field model in the marine atmosphere is constructed using 3D modeling and CFD simulation. The effects of salt fog on the deposition and corrosion of wind turbines are simulated by combining salt fog distribution, concentration gradient, and airflow humidity characteristics. Multi-physics fields are integrated to form a coupled digital twin model, enabling dynamic deduction under different climate conditions and obtaining simulation data.

[0088] Step 3: Based on marine atmospheric environmental data and simulation data, the initial convolutional neural network model CNN-Init is constructed. Historical salt spray data, temperature and humidity parameters, and corrosion rate are used as input to extract spatiotemporal features and construct an initial model with spatial diffusion and corrosion perception capabilities. Through joint training of multi-source coupled data, the corrosion risk feature vector FCV is extracted to characterize the nonlinear relationship between salt spray, climate, and material response. The FCV is input into the AI-Adaptive-Corrosion-Predictor for dynamic training to predict corrosion trends.

[0089] Step 4: Monitor the changes in salt spray concentration in the environment in real time. Combine the simulation data with the actual environmental data to calculate the salt spray concentration corrosion coefficient SCCX, and compare and analyze it with the first threshold Q1 to determine whether the salt spray concentration is normal. If it is abnormal, a strategy is given;

[0090] Step 5: Based on the wind speed, wind direction, wind field viscosity parameters, equipment surface roughness, and salt spray concentration data obtained through real-time monitoring, combined with the simulation data output by the digital twin modeling module, the wind-driven salt spray coupling deposition coefficient WDCX is calculated and compared with the second threshold Q2 to determine whether the wind turbine equipment has the risk of abnormal corrosion. If there is a risk, a strategy is implemented.

[0091] Step 6. After the test, a test report is automatically generated based on simulation data, actual monitoring data and salt spray concentration analysis, including test parameters, salt spray concentration, climate trends and corrosion assessment; an AI prediction model is established through historical test data, and model training is performed to achieve pattern recognition, trend analysis and anomaly detection of the salt spray airflow field, predict the impact of future environmental changes on wind turbine performance, and optimize test parameters.

[0092] In this embodiment, by implementing the above steps, the system can achieve a full range of dynamic marine wind farm corrosion risk assessment and prediction. First, through precise regional division and real-time data collection, a reliable environmental data foundation is provided for subsequent models; then, through digital twin modeling and CFD simulation, a detailed salt fog airflow field model is constructed, which can accurately simulate the impact of salt fog on wind power equipment and ensure early warning of corrosion. Then, through the training of the convolutional neural network model and the coupling of multi-source data, the system can dynamically predict corrosion trends, effectively assess potential risks, and respond quickly to changes in salt fog concentration and wind-driven salt fog deposition, providing optimized protection strategies for equipment. Finally, the automated feedback tuning module generates detailed test reports, combines AI model training and trend analysis, and achieves accurate predictions of future environmental changes, optimizes the test parameter combination, and improves the operating efficiency, corrosion resistance and equipment life cycle of wind turbines.

[0093] The threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technicians in this field for each set of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

Claims

1. A salt fog airflow field prediction system in a marine atmospheric environment, characterized in that: It includes data acquisition module, digital twin modeling module, AI corrosion prediction model construction module, salt spray concentration intelligent monitoring module, wind-driven salt spray coupled deposition intelligent monitoring module and automated feedback tuning module; The data acquisition module is used to divide the offshore wind farm into a number of salt fog monitoring areas, monitor the salt fog monitoring areas, and collect environmental data; The digital twin modeling module is used to construct a salt fog airflow field model in the marine atmosphere based on environmental data using three-dimensional modeling and CFD simulation. It combines salt fog distribution, concentration gradient, and airflow humidity characteristics to simulate the deposition and corrosion effects of salt fog on wind power equipment. It integrates multiple physical fields to form a coupled digital twin model, enabling dynamic deduction under different climate conditions and obtaining simulation data. The AI ​​corrosion prediction model construction module is used to build a convolutional neural network initial model CNN-Init based on marine atmospheric environmental data and simulation deduction data. It uses historical salt spray data, temperature and humidity parameters, and corrosion rate as input to extract spatiotemporal features and build an initial model with spatial diffusion and corrosion perception capabilities. Through joint training of multi-source coupled data, it extracts the corrosion risk feature vector FCV to characterize the nonlinear relationship between salt spray, climate, and material response. The FCV is input into the AI-Adaptive-Corrosion-Predictor for dynamic training to predict corrosion trends. The salt spray concentration intelligent monitoring module is used to monitor the changes in the salt spray concentration in the environment in real time, combine the simulation data and the actual environmental data, calculate the salt spray concentration corrosion coefficient SCCX, and compare and analyze it with the first threshold Q1 to determine whether the salt spray concentration is normal. If it is abnormal, a strategy is given; The wind-driven salt spray coupled deposition intelligent monitoring module is used to calculate the wind-driven salt spray coupled deposition coefficient WDCX based on the wind speed, wind direction angle, wind field viscosity parameters, equipment surface roughness and salt spray concentration data obtained by real-time monitoring, combined with the simulation deduction data output by the digital twin modeling module, and compare and analyze it with the second threshold Q2 to determine whether the wind turbine equipment has the risk of abnormal corrosion, and provide a strategy if there is a risk; The automated feedback tuning module is used to automatically generate a test report after the test, based on simulation data, actual monitoring data, and salt spray concentration analysis, including test parameters, salt spray concentration, climate trends, and corrosion assessments. It also establishes an AI prediction model through historical test data and conducts model training to achieve pattern recognition, trend analysis, and anomaly detection of the salt spray airflow field, predict the impact of future environmental changes on wind turbine performance, and optimize test parameters.

2. The salt fog airflow field prediction system in a marine atmospheric environment according to claim 1, wherein: The data acquisition module includes a region division unit and a collection unit; the region division unit is used to obtain three-dimensional spatial geographic information, historical sea condition information and wind power equipment information of the offshore wind farm, draw an electronic distribution map of the offshore wind farm, and divide the offshore wind farm into a number of salt fog monitoring areas. The salt fog monitoring areas are marked on the electronic distribution map as: Jcq1, Jcq2, ..., Jcqn, where n represents the number of salt fog monitoring areas; The acquisition unit is used to monitor the salt fog monitoring area in real time, and collects wind speed and direction by installing an ultrasonic anemometer on the top of the wind turbine tower; collects salt fog concentration by installing a corrosion environment salinity sensor on the edge of the wind turbine blade; and collects environmental humidity values ​​by installing a humidity sensor on the windward side of the wind turbine; and establishes an environmental data set.

3. The salt fog airflow field prediction system in a marine atmospheric environment according to claim 2, wherein: The digital twin modeling module is used to utilize environmental data set data to reconstruct the salt fog airflow field in the marine atmospheric environment through three-dimensional modeling; CFD computational fluid dynamics is used to simulate air flow, and the salt fog distribution and concentration gradient unique to the marine environment are combined to construct a salt fog airflow field model; by considering the humidity of the airflow, the condensation and diffusion characteristics of salt particles, the impact of salt fog on the deposition and corrosion of metal components and wind turbine blade surfaces is simulated; and the airflow and corrosion simulations are integrated to construct a digital twin model based on the coupling of multiple physical fields of airflow, salt fog and humidity, so as to realize the dynamic deduction of the salt fog airflow field of offshore wind turbines under different climatic conditions and obtain simulation deduction data.

4. The salt fog airflow field prediction system in a marine atmospheric environment according to claim 3, wherein: The AI ​​corrosion prediction model construction module is used to build a convolutional neural network initialization model CNN-Init based on marine atmospheric environmental data and simulation deduction data. The CNN-Init model uses historical salt spray concentration change data, measured temperature and humidity parameters, and equipment surface corrosion rate data as input samples, and uses multi-layer convolution kernels to extract spatiotemporal distribution characteristics to build an initial model with spatial diffusion and corrosion dynamic perception capabilities. Subsequently, a data input structure based on multi-source coupling was introduced, including atmospheric corrosion grade data and equipment material property data, and the CNN-Init model was jointly trained and tested. During the model training process, the intermediate feature layer output of the convolutional neural network was extracted as the corrosion risk feature vector Feature-Corrosion-Vector (FCV). The feature vector characterizes the nonlinear mapping relationship between salt spray deposition pattern, climate corrosion factor and material response; the FCV feature vector is input into the AI-Adaptive-Corrosion-Predictor. After multiple rounds of dynamic training and regression testing, it is used as an AI corrosion prediction model to predict corrosion trends.

5. The salt fog airflow field prediction system in a marine atmospheric environment according to claim 4, wherein: The salt spray concentration intelligent monitoring module includes a first calculation unit and a first analysis unit; The first calculation unit is used to monitor the change of salt spray concentration in the environment in real time, and calculate the salt spray concentration corrosion coefficient SCCX after dimensionless processing by combining simulation data and actual environmental data. The formula is as follows: Where, represents the salt spray concentration at time t, Represents the benchmark salt spray concentration, obtained by simulation, Represents the time-varying trend factor of salt spray concentration at time t, The initial trend factor representing the historical salt spray concentration is obtained by simulation. Represents the ambient humidity value at time t, Indicates the reference ambient humidity value, obtained by simulation. represents the wind speed value at time t, represents the reference environmental wind speed value, which is obtained by simulation deduction; L represents the wind turbine blade area, which is obtained by the wind turbine parameters; Represents the corrosion resistance score of the wind turbine at time t, obtained through historical data, Represents the adjustment coefficient, which is obtained through experiments.

6. The salt fog airflow field prediction system in a marine atmospheric environment according to claim 5, characterized in that: The first analysis unit is used to preset a first threshold value Q1 in advance, and compare and analyze the salt spray concentration corrosion coefficient SCCX with the first threshold value Q1, and obtain a first evaluation result including: When the salt spray concentration corrosion coefficient SCCX ≤ the first threshold Q1, it means that the concentration of salt spray is normal and there is no risk of accelerated corrosion of wind turbines, and continuous monitoring is required; When the salt spray concentration corrosion coefficient SCCX is greater than the first threshold Q1, it indicates that the salt spray concentration is abnormal and there is a risk of accelerated corrosion of the wind turbine. This triggers the first warning instruction and generates the first strategy: add anti-corrosion coating to the wind turbine equipment and increase the frequency of regular maintenance by 50%.

7. The salt fog airflow field prediction system in a marine atmospheric environment according to claim 4, characterized in that: The wind-driven salt spray coupled deposition intelligent monitoring module includes a second calculation unit and a second analysis unit; The second calculation unit is used to calculate the wind-driven salt spray coupling deposition coefficient WDCX based on the wind speed, wind direction angle, wind field viscosity parameters, equipment surface roughness and salt spray concentration data obtained through real-time monitoring, combined with the simulation deduction data output by the digital twin modeling module, after dimensionless processing. The formula is as follows: Where, represents the wind speed value at time t, Indicates the reference ambient wind speed value, It represents the angle between the wind direction at time t and the normal line of the surface of the wind turbine equipment. represents the surface roughness parameter of the wind turbine, Indicates the roughness reference value, represents the salt spray concentration at time t, Indicates the reference salt spray concentration, represents the wind field viscosity coefficient at time t, obtained by simulation, represents the reference value of wind field viscosity, represents the empirical adjustment coefficient, which is obtained by training the AI ​​corrosion prediction model. w1, w2, w3, and w4 represent weight coefficients, which are obtained by training the AI ​​corrosion prediction model.

8. The salt fog airflow field prediction system in a marine atmospheric environment according to claim 7, characterized in that: The second analysis unit is used to preset a second threshold value Q2 in advance, and compare and analyze the wind-driven salt spray coupling deposition coefficient WDCX with the second threshold value Q2 to obtain a second evaluation result including: When the wind-driven salt spray coupling deposition coefficient WDCX is less than the second threshold Q2, it indicates that the surface load of the wind turbine equipment is within the normal range, there is no risk of abnormal corrosion, and continuous monitoring is required; When the wind-driven salt spray coupling deposition coefficient WDCX ≥ the second threshold Q2, it indicates that the surface attachment load of the wind turbine equipment is not within the normal range and there is a risk of abnormal corrosion. The second early warning instruction is triggered and the second strategy is generated: the self-cleaning material replacement rate is increased to 60%, the micro-nano structure roughness is adjusted by 20%, the structure orientation is fine-tuned by ±15°, and the frequency of regular maintenance is increased by 60%.

9. The salt fog airflow field prediction system in a marine atmospheric environment according to claim 8, characterized in that: The automated feedback tuning module is used to automatically generate a detailed test report after the test is completed based on simulation data, actual monitoring data, and salt spray concentration analysis results, including test parameters, salt spray concentration and climate change trends, equipment corrosion assessment, and optimization strategies; An AI prediction model is established based on historical test data through a deep learning model, and the model is trained using test report data to achieve pattern recognition, trend analysis and anomaly detection of salt spray airflow fields, predict the impact of future marine environment changes on wind turbine performance, and optimize the test parameter combination.

10. A method for predicting a salt fog airflow field in a marine atmosphere environment, comprising a salt fog airflow field prediction system in a marine atmosphere environment according to any one of claims 1 to 9, characterized in that: The following steps are involved: Step 1: Divide the offshore wind farm into several salt spray monitoring areas, monitor the salt spray monitoring areas, and collect environmental data; Step 2: Based on environmental data, a salt fog airflow field model in the marine atmosphere is constructed using 3D modeling and CFD simulation. The effects of salt fog on the deposition and corrosion of wind turbines are simulated by combining salt fog distribution, concentration gradient, and airflow humidity characteristics. Multi-physics fields are integrated to form a coupled digital twin model, enabling dynamic deduction under different climate conditions and obtaining simulation data. Step 3: Based on marine atmospheric environmental data and simulation data, the initial convolutional neural network model CNN-Init is constructed. Historical salt spray data, temperature and humidity parameters, and corrosion rate are used as input to extract spatiotemporal features and construct an initial model with spatial diffusion and corrosion perception capabilities. Through joint training of multi-source coupled data, the corrosion risk feature vector FCV is extracted to characterize the nonlinear relationship between salt spray, climate, and material response. The FCV is input into the AI-Adaptive-Corrosion-Predictor for dynamic training to predict corrosion trends. Step 4: Monitor the changes in salt spray concentration in the environment in real time. Combine the simulation data with the actual environmental data to calculate the salt spray concentration corrosion coefficient SCCX, and compare and analyze it with the first threshold Q1 to determine whether the salt spray concentration is normal. If it is abnormal, a strategy is given; Step 5: Based on the wind speed, wind direction, wind field viscosity parameters, equipment surface roughness, and salt spray concentration data obtained through real-time monitoring, combined with the simulation data output by the digital twin modeling module, the wind-driven salt spray coupling deposition coefficient WDCX is calculated and compared with the second threshold Q2 to determine whether the wind turbine equipment has the risk of abnormal corrosion. If there is a risk, a strategy is implemented. Step 6. After the test, a test report is automatically generated based on simulation data, actual monitoring data and salt spray concentration analysis, including test parameters, salt spray concentration, climate trends and corrosion assessment; an AI prediction model is established through historical test data, and model training is performed to achieve pattern recognition, trend analysis and anomaly detection of the salt spray airflow field, predict the impact of future environmental changes on wind turbine performance, and optimize test parameters.

Images