Deicing control method and system for wind power generation blade

Abstract

The embodiment of the invention provides a deicing control method and system for a wind power generation blade, and belongs to the technical field of wind power. The method comprises the steps that operation state information of the wind generating set is collected, feature extraction is carried out on the operation state information, and operation state features are obtained; calling a pre-trained deicing state prediction model based on the operation state characteristics, and determining a deicing prediction level in the current operation state; determining a control scheme of the gas-heat deicing device based on the deicing prediction grade; and executing the control scheme, and updating the deicing prediction grade of the wind generating set in real time in the execution process so as to execute control scheme correction of the corresponding hot deicing device when the deicing prediction grade changes. According to the scheme, it is ensured that the deicing process is more real-time and intelligent, the response speed and the energy consumption optimization level of the deicing system are effectively improved, and therefore the deicing efficiency of fan blades in the complex environment and the operation stability of a wind turbine generator are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and in particular to a deicing control method for a wind power blade and a deicing control system for a wind power blade. Background Art

[0002] When wind turbines operate in cold environments, ice easily forms on the surface of their blades, causing changes in aerodynamic characteristics, which in turn reduces power generation efficiency and may even cause blade damage or unit shutdown. Existing gas-heat deicing technology mainly relies on air blowing equipment to deliver heated air to the inner cavity of the blades to melt the external ice layer through heat conduction. However, traditional gas-heat deicing solutions still have many problems that affect their deicing effect and energy efficiency performance.

[0003] Most existing gas-heat deicing systems use a fixed start-stop strategy, relying on simple environmental parameter thresholds to determine whether to start the deicing device, which makes it difficult to adapt to complex and changeable meteorological conditions. For example, simply deciding to start or stop deicing based on the threshold of a parameter such as temperature, humidity or wind speed may cause deicing to start too early or too late, thereby affecting energy consumption control and deicing efficiency. In addition, since the ice coverage of various parts of the blade may vary significantly, the existing technology has failed to adopt fine-grained regulation for different ice coverage levels, and often uses full-power mode heating, which not only increases energy consumption, but may also cause local overheating and reduce the overall efficiency of the system.

[0004] The control method of the blower is relatively simple, usually using a fixed wind speed and constant heating mode, and failing to optimize and adjust the temperature distribution of different degrees of icing and different areas of the blades. Due to the complex shape of the blades, the hot air flow path may be uneven, and the existing system cannot ensure that all areas can be fully heated, resulting in insufficient ice melting or heat waste in local areas. At the same time, the operating state of the fan blades has an important impact on the deicing process. Factors such as blade angle and speed will change the airflow distribution and heat transfer efficiency, but the existing control scheme fails to effectively combine the fan operating state for dynamic optimization.

[0005] Current de-icing strategies generally adopt a passive approach, that is, the de-icing system is activated only when the ice coverage reaches the set threshold, and active prevention is not performed based on meteorological forecast data. When meteorological conditions have the risk of icing, if the anti-icing mode can be entered in advance and low-power preheating is performed appropriately, it will help reduce the formation of ice and reduce the energy consumption of subsequent de-icing. However, the existing technology lacks effective multi-modal data fusion methods and cannot fully utilize the comprehensive information of meteorological forecasts, real-time environmental parameters and ice coverage monitoring data to optimize the de-icing strategy.

[0006] In summary, the current gas-heat deicing scheme still has great limitations in terms of the intelligent degree of start-stop control, energy consumption optimization, adaptability of fan operation status, and predictive prevention capabilities, making it difficult to achieve an efficient, accurate, and energy-saving intelligent deicing control strategy. Therefore, an improved deicing control method is urgently needed to improve the overall deicing efficiency and energy consumption management level. Summary of the invention

[0007] The purpose of the embodiments of the present invention is to provide a de-icing control method and system for wind turbine blades, so as to at least solve the problem that the current gas-thermal de-icing scheme has great limitations in the intelligence level of start-stop control, energy consumption optimization, adaptability to wind turbine operating conditions, and predictive and preventive capabilities.

[0008] In order to achieve the above-mentioned objectives, the first aspect of the present invention provides a de-icing control method for wind turbine blades, the method comprising: collecting operating status information of the wind turbine generator set, and performing feature extraction on the operating status information to obtain operating status characteristics; calling a pre-trained de-icing status prediction model based on the operating status characteristics to determine the de-icing prediction level under the current operating state; determining a control scheme of a gas-thermal de-icing device based on the de-icing prediction level; executing the control scheme, and updating the de-icing prediction level of the wind turbine generator set in real time during the execution process, so as to execute the control scheme correction of the corresponding thermal de-icing device when the de-icing prediction level changes.

[0009] Optionally, the operating status information of the wind turbine generator set includes any one or more of: weather forecast information, real-time environmental parameter information, and real-time ice thickness information of wind turbine blades.

[0010] Optionally, performing feature extraction on the running status information to obtain running status features includes: performing preprocessing on the running status information; performing one-hot encoding on the preprocessed running status information to obtain corresponding numerical vectors of each preprocessed running status information; and performing feature fusion on the numerical vectors of each preprocessed running status information based on the AHP hierarchical analysis method to obtain a comprehensive feature vector as the running status feature.

[0011] Optionally, the deicing prediction level includes a prevention level and a deicing level; the deicing level includes multiple levels, and the higher the level, the thicker the corresponding deicing layer.

[0012] Optionally, the gas thermal deicing device includes: a heating module, a blower and a blade angle adjustment device; the control scheme of the gas thermal deicing device determined based on the deicing prediction level includes: determining the target operating state of each gas thermal deicing device based on the deicing prediction level; generating a corresponding control scheme based on the current operating state of each gas thermal deicing device and the corresponding target operating state.

[0013] Optionally, if the current deicing prediction level is a prevention level, the control scheme for determining the gas-thermal deicing device based on the deicing prediction level includes: predicting the freezing time based on the operating status information of the wind turbine generator set to obtain the predicted freezing time; determining the intervention time of each gas-thermal deicing device based on the predicted freezing time; and generating an intermittent air supply operation plan for each gas-thermal deicing device based on the intervention time and a preset intermittent air supply timing table as a control scheme for the gas-thermal deicing device.

[0014] Optionally, if the current de-icing prediction level is the de-icing level, the control scheme for determining the gas-thermal de-icing device based on the de-icing prediction level includes: determining the real-time ice thickness information of the wind turbine blades based on the operating status information of the wind turbine generator set, and predicting the development of the ice thickness; determining the thickness of the layer to be de-iced based on the real-time ice thickness information and the ice thickness development prediction results; simulating the reverse development information of the ice thickness toward the preset value of the safe ice thickness based on the thickness of the layer to be de-iced, generating a target operating state for each gas-thermal de-icing device, and generating a corresponding control scheme based on the current operating state of each gas-thermal de-icing device and the corresponding target operating state.

[0015] Optionally, the de-icing prediction level of the wind turbine generator set is updated in real time during the execution process, so as to execute the control scheme correction of the corresponding thermal de-icing device when the de-icing prediction level changes, including: during the execution process, real-time collection of operating status information of the wind turbine generator set during the execution process; performing feature extraction on the operating status information during the execution process to obtain operating status features during the execution process; calling a pre-trained de-icing status prediction model based on the operating status features during the execution process to determine the de-icing prediction level under the current operating state; if the de-icing prediction level under the current operating state is different from the de-icing prediction level corresponding to the control scheme being executed, then determining the control scheme of the gas-thermal de-icing device based on the de-icing prediction level under the current operating state as a new control scheme, and executing the new control scheme.

[0016] A second aspect of the present invention provides a de-icing control system for wind turbine blades, the system comprising: a collection unit, for collecting operating status information of the wind turbine generator set, and performing feature extraction on the operating status information to obtain operating status characteristics; a prediction unit, for calling a pre-trained de-icing status prediction model based on the operating status characteristics to determine the de-icing prediction level under the current operating state; a scheme generation unit, for determining a control scheme of a gas-thermal de-icing device based on the de-icing prediction level; an execution unit, for executing the control scheme and updating the de-icing prediction level of the wind turbine generator set in real time during the execution process, so as to execute the control scheme correction of the corresponding thermal de-icing device when the de-icing prediction level changes.

[0017] On the other hand, the present invention provides a computer-readable storage medium, on which instructions are stored, which, when executed on a computer, enable the computer to execute the above-mentioned deicing control method for wind turbine blades.

[0018] Through the above technical scheme, the scheme of the present invention collects the operating status information of the wind turbine in real time, extracts its features, and generates operating status features, thereby improving the accuracy of deicing decisions. Using the pre-trained deicing state prediction model, the method can intelligently evaluate the deicing prediction level according to the current operating state, avoiding the misjudgment or energy waste caused by the traditional method relying only on a single threshold judgment. Based on the prediction level, the strategy dynamically adjusts the control scheme of the gas-heat deicing device to achieve graded deicing and improve energy efficiency. At the same time, during the deicing process, the system can monitor the state changes of the wind turbine in real time, update the deicing prediction level, and modify the deicing control scheme according to the latest situation to make it more adaptable. Through closed-loop control, the method ensures that the deicing process is more real-time and intelligent, effectively improves the response speed and energy consumption optimization level of the deicing system, thereby improving the deicing efficiency of the fan blades in complex environments and the operating stability of the wind turbine.

[0019] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings: Figure 1 It is a flowchart of the steps of a deicing control method for wind turbine blades provided by an embodiment of the present invention; Figure 2 It is a system structure diagram of a de-icing control system for wind turbine blades provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0021] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.

[0022] Figure 1 FIG. 1 is a flow chart of a method for controlling deicing of wind turbine blades provided by an embodiment of the present invention. Figure 1 As shown, an embodiment of the present invention provides a deicing control method for a wind turbine blade, the method comprising: Step S10: collecting the operating status information of the wind turbine generator set, and performing feature extraction on the operating status information to obtain operating status features.

[0023] Specifically, the operating status information of the wind turbine generator set includes any one or more of: weather forecast information, real-time environmental parameter information, and real-time ice thickness information of wind turbine blades.

[0024] Furthermore, performing feature extraction on the running status information to obtain the running status feature includes: performing preprocessing on the running status information; performing one-hot encoding on the preprocessed running status information to obtain a numerical vector of each preprocessed running status information; and performing feature fusion on the numerical vector of each preprocessed running status information based on the AHP hierarchical analysis method to obtain a comprehensive feature vector as the running status feature.

[0025] In the embodiment of the present invention, the operating status information of the wind turbine generator set is composed of data of multiple dimensions, including but not limited to: 1) Weather forecast information: Weather forecast data for the area where the wind turbine is located, including temperature, humidity, wind speed, air pressure, precipitation, cloud thickness, etc. for a period of time in the future. These data can help predict whether there is a risk of icing on the wind turbine blades.

[0026] 2) Real-time environmental parameter information: The temperature, humidity, wind speed, wind direction, air pressure and other parameters monitored in real time by environmental sensors around the fan can reflect the real-time conditions of the environment in which the fan is located, providing a key basis for judging the formation of ice.

[0027] 3) Real-time ice thickness information of wind turbine blades: Use ice sensors, thermal imaging cameras or other detection methods to measure the actual thickness of the ice layer formed on the blade surface to obtain the most direct icing status data.

[0028] These data can be collected through meteorological stations, wind turbine sensors and external remote monitoring to ensure data diversity and real-time nature, and improve the ability to comprehensively judge the operating status of the wind turbine.

[0029] Furthermore, after obtaining the operating status information of the wind turbine, it is necessary to extract features from these data to form a numerical feature vector that can be used by the subsequent de-icing prediction model. Since the collected raw data may have problems such as noise, incompleteness, and inconsistent formats, data cleaning and preprocessing are required. Linear interpolation, mean filling, or machine learning prediction methods are used to supplement missing data. Box plot analysis or the 3σ principle is used to detect and remove outliers to prevent data from misleading decisions. Min-Max normalization or Z-score normalization is used to ensure that different data types have a unified scale during the feature extraction process and improve calculation stability.

[0030] In order to ensure the effective use of different categorical data, this method performs One-Hot encoding on the processed running status information, converts discrete categorical variables into computable numerical vectors, and avoids misleading due to the size of the category values. One-Hot encoding is a commonly used data preprocessing method, which is mainly used to convert discrete categorical variables into numerical vectors so that computers can understand and process these data more effectively. It is widely used in fields such as machine learning, deep learning, and natural language processing. In the original data, categorical variables are usually expressed in the form of text or integers, such as "sunny, cloudy, rainy" or "low, medium, high" classification information. However, directly converting these categories to numerical values ​​(such as 0, 1, 2) may introduce incorrect order relationships, causing the model to mistakenly believe that there is a size or weight relationship between categories. Therefore, One-Hot encoding provides a disordered and equal-distance conversion method so that different categories have the same weight and distance in terms of numerical values. After One-Hot encoding, all data are converted into binary vectors of a unified format to ensure that the data will not be affected by the numerical size when it is input into the model.

[0031] Furthermore, since the sources of wind turbine operating status information are diverse and each data source has a different degree of influence on the de-icing decision, AHP is used to calculate the importance weights of different data sources and perform feature fusion to form the final comprehensive feature vector.

[0032] In one possible implementation, the relative importance of each data source is set through expert experience and historical data analysis, including the weight of weather forecast information (α), the weight of real-time environmental parameter information (β), and the weight of blade ice thickness information (γ). α + β + γ = 1, and the relative importance of each feature is calculated using a pairwise comparison matrix, and the weight rationality is ensured through a consistency test (CR < 0.1). The comprehensive feature vector after weighted fusion can more accurately reflect the actual operating status of the wind turbine and eliminate the misjudgment that may be caused by a single data source.

[0033] Based on the scheme of the present invention, combined with weather forecasting, real-time environmental monitoring and blade icing monitoring, AHP is used to calculate the importance of features, so that the final feature vector is more reasonable and the model's adaptability to complex meteorological conditions is improved. Through data preprocessing + One-Hot encoding, the data format is unified, the impact of noise is reduced, the input stability of the deep learning model is improved, and the numerical value size misleads classification decisions. Compared with the traditional simple weighted method, the use of AHP to calculate the weighted feature vector can more scientifically integrate different data sources, improve the accuracy of deicing prediction, and make the start and stop of the deicing device more intelligent and energy-saving. Because the blade icing situation and external environmental changes can be accurately calculated, more efficient deicing control can be achieved, avoiding energy waste caused by excessive heating, and reducing the risk of blade damage due to excessive ice.

[0034] Step S20: calling a pre-trained deicing state prediction model based on the operating state characteristics to determine a deicing prediction level under the current operating state.

[0035] Specifically, the scheme of the present invention adopts a deep convolutional neural network (ResNet18) as the core algorithm of the deicing state prediction model to achieve accurate identification and intelligent decision-making of the operating status of wind turbines. As a deep residual network, ResNet18 can efficiently extract the features of input data and identify the complex patterns of the icing state of wind turbine blades through a hierarchical learning structure. In this scheme, the input data is a comprehensive feature vector, which is generated by feature extraction, encoding and feature fusion of the wind turbine's meteorological forecast information, real-time environmental parameter information and ice thickness information, and can fully reflect the current operating status of the wind turbine.

[0036] Furthermore, the deep convolutional neural network builds deep feature extraction capabilities through multiple convolutional layers, pooling layers, and residual connections. Among them, ResNet18 can effectively alleviate the gradient vanishing and gradient explosion problems in deep networks by introducing residual structures, enabling the model to learn deeper features. In this scheme, the comprehensive feature vector is used as the model input. First, local features are extracted through multiple convolution operations, and then deep feature fusion is performed through residual blocks to ensure that the complex patterns of wind turbine operating status can be fully captured. Finally, the fully connected layer of the network maps the extracted high-dimensional features to the classification output space, and uses Softmax for multi-classification prediction to determine the current de-icing prediction level.

[0037] Preferably, the deicing prediction level includes a prevention level and a deicing level; the deicing level includes multiple levels, and the higher the level, the thicker the corresponding deicing layer.

[0038] In an embodiment of the present invention, the deicing prediction level system of the scheme of the present invention includes two parts: prevention level and deicing level, so as to realize accurate identification and graded control of the ice coating state of wind turbine blades. The prevention level is mainly used to determine whether it is necessary to adopt low-power mode for anti-icing operation in advance, while the deicing level is further subdivided into multiple levels, and the corresponding deicing intensity and control strategy are determined according to the thickness of ice coating on the blade surface. The higher the level, the more serious the blade icing situation, the thicker the deicing layer to be deiced, and the need to take higher power and longer deicing measures.

[0039] In actual applications, the prevention level is mainly based on weather forecast information and real-time environmental parameter information to determine whether there is an ice risk. For example, when the future temperature is predicted to be close to the freezing point, the humidity is high and the wind speed is low, the low-power anti-icing mode is entered in advance, and intermittent air supply or small heating is used to reduce the possibility of ice formation and reduce the energy consumption of subsequent large-scale deicing. At this time, high-power deicing will not be started directly, but preventive measures will be taken to reduce ice accumulation and improve energy efficiency.

[0040] The de-icing level is dynamically adjusted according to the real-time ice thickness information to ensure accurate matching of different degrees of ice coverage. Specifically, the de-icing level can be divided into light de-icing, medium de-icing, heavy de-icing and other levels. For example, when the ice thickness on the blade surface is thin (such as 0.5mm - 1mm), it enters the light de-icing mode, and uses low wind speed and short-time heating to de-ice to avoid energy waste; when the ice thickness reaches more than 2mm, it will be upgraded to the medium de-icing mode, and the heating intensity will be enhanced and the heat distribution will be optimized in combination with the fan angle adjustment; if the ice thickness increases further (such as more than 3mm), it will enter the heavy de-icing mode, run at full power, and ensure that the ice layer is completely melted through high-temperature blowing and long-time heating to prevent ice from falling off and causing damage to the blades or units.

[0041] Based on the solution of the present invention, compared with the traditional simple start-stop control mode, more accurate energy consumption management can be achieved, unnecessary high-power deicing operations can be avoided, and the impact of icing on the operating efficiency of the wind turbine can be reduced. In addition, due to the introduction of the prevention level, corresponding measures can be taken before the ice layer has accumulated seriously, reducing the need for subsequent high-power deicing, and further improving the operating safety of the wind turbine and the overall energy efficiency of deicing. In summary, the deicing prediction level system makes the deicing strategy more flexible and efficient through intelligent hierarchical control, which helps to improve the long-term stability of wind turbines under complex environmental conditions.

[0042] Step S30: determining a control scheme of the gas-heat deicing device based on the deicing prediction level.

[0043] Specifically, the gas thermal deicing device includes: a heating module, a blower and a blade angle adjustment device; the control scheme of the gas thermal deicing device determined based on the deicing prediction level includes: based on the deicing prediction level, respectively determining the target operating state of each gas thermal deicing device; generating a corresponding control scheme based on the current operating state of each gas thermal deicing device and the corresponding target operating state.

[0044] In an embodiment of the present invention, the heating module, the blower and the blade angle adjustment device are precisely controlled according to different deicing prediction levels to adapt to different degrees of icing conditions, thereby improving deicing efficiency, reducing energy consumption, and ensuring the safe operation of the fan blades. A specific control scheme is generated based on the difference between the current operating state of each gas-heat deicing device and the corresponding target operating state. During the control process, the temperature of the fan blades, the deicing progress and the changes in environmental conditions are detected in real time. If the actual deicing progress does not achieve the expected effect, the heating temperature, wind speed or blade angle will be automatically adjusted to ensure that the deicing process is carried out efficiently. For example, if it is detected that the local ice layer has not completely melted, the control can adjust the blade direction to expose the area to a higher temperature airflow, or appropriately extend the deicing time to ensure that the ice layer is completely removed.

[0045] Furthermore, if the current deicing prediction level is a prevention level, the control scheme for determining the gas-thermal deicing device based on the deicing prediction level includes: predicting the freezing time based on the operating status information of the wind turbine generator set to obtain the predicted freezing time; determining the intervention time of each gas-thermal deicing device based on the predicted freezing time; and generating an intermittent air supply operation scheme for each gas-thermal deicing device based on the intervention time and a preset intermittent air supply timing table as a control scheme for the gas-thermal deicing device.

[0046] In an embodiment of the present invention, when the current de-icing prediction level is at the prevention level, it is first necessary to predict the freezing time based on the operating status information of the wind turbine generator set (including meteorological forecasts, real-time environmental parameters, blade temperature and humidity, wind speed, etc.). Specifically, a physical modeling-based method is used to combine historical data and real-time meteorological data to predict the time point when freezing may occur in the future. The freezing time prediction mainly depends on multiple variables such as ambient temperature, air humidity, wind speed, air pressure, precipitation, blade surface temperature, etc., and the formation time of the ice layer is predicted by multivariate regression analysis or deep learning algorithms (such as LSTM). For example, when meteorological data indicates that the temperature will drop to the freezing point within the next 30 minutes, the air humidity is higher than 80%, and the wind speed is low (lower than 3m / s, reducing the self-cleaning ability of the blades), it can be predicted that the blades will begin to freeze in 25-40 minutes, and anti-icing measures can be deployed in advance based on this prediction result.

[0047] Furthermore, after obtaining the predicted icing moment, it is necessary to intervene in advance with the gas-heat deicing device to reduce the probability of ice formation and reduce the energy consumption of subsequent deicing. The prediction result of the icing moment, for example, if it is predicted that the blade will freeze in 30 minutes, it may enter the low-power prevention mode 10-1 minute in advance. The temperature difference between the current temperature of the blade and the predicted icing threshold, if the surface temperature of the blade is low, it is necessary to intervene in the anti-icing mode earlier. Combined with the wind speed and ambient temperature, if the wind speed is low and the heat stays on the blade surface for a long time, the intervention time should be appropriately postponed; if the wind speed is high and the heat is easy to lose, it is necessary to intervene in advance. The optimization goal of the intervention time is to reduce energy consumption as much as possible while ensuring the anti-icing effect, and dynamically calculate the best intervention time point through an intelligent algorithm.

[0048] Furthermore, after determining the intervention time, a specific intermittent air supply operation plan needs to be formulated to ensure that ice formation is prevented under low energy consumption. A series of optional intermittent air supply plans are stored, and the optimal plan is selected according to different meteorological conditions and blade conditions. Low-power air supply (for example, air supply for 2 minutes every 10 minutes) can avoid rapid heat loss and improve heat utilization efficiency. Air supply in different regions, that is, heating the root of the blade first, and then extending to the middle and tip of the blade, improves the overall anti-icing effect. Adjust the wind speed. If the ambient wind speed is high, use high temperature and low wind speed to prevent heat from dissipating too quickly; if the wind speed is low, use low temperature and high wind speed mode to make the hot air evenly cover the blade surface.

[0049] Preferably, if the current de-icing prediction level is the de-icing level, the control scheme for determining the gas-thermal de-icing device based on the de-icing prediction level includes: determining the real-time ice thickness information of the wind turbine blades based on the operating status information of the wind turbine generator set, and predicting the development of the ice thickness; determining the thickness of the layer to be de-iced based on the real-time ice thickness information and the ice thickness development prediction results; simulating the reverse development information of the ice thickness toward the preset value of the safe ice thickness based on the thickness of the layer to be de-iced, generating a target operating state for each gas-thermal de-icing device, and generating a corresponding control scheme based on the current operating state of each gas-thermal de-icing device and the corresponding target operating state.

[0050] In the embodiment of the present invention, under the de-icing level state, it is first necessary to accurately measure the current ice thickness of the blade and predict the future development trend of the ice layer. The process includes the following technical steps: 1) Real-time data collection: Use ice sensors, thermal imaging cameras or ultrasonic thickness gauges to obtain ice thickness data on the blade surface in real time to ensure measurement accuracy.

[0051] 2) Analysis of environmental factors: Simultaneously monitor meteorological conditions (temperature, humidity, wind speed, precipitation), and combine them with the operating parameters of the wind turbine (speed, blade angle, power output, etc.) to build a comprehensive environmental impact model to accurately predict the formation rate of ice cover.

[0052] 3) Ice thickness development trend prediction: Based on historical data and machine learning models, the possible growth rate of ice under current meteorological conditions is calculated, and the blade ice thickness is predicted to change in the future without external intervention. For example, if the current blade ice thickness is 2.5 mm, it is predicted that in the next 30 minutes, if no measures are taken, the ice thickness may increase to 3.5 mm. The thickness of the ice layer that needs to be removed is calculated, and the de-icing strategy is adjusted.

[0053] Furthermore, after obtaining the real-time ice thickness data and the predicted ice development trend, it is necessary to calculate the thickness of the de-icing layer, that is, how much ice layer needs to be removed to restore the blade to a safe working state. Set a preset value for safe ice thickness (such as 0.5mm or no ice at all), and calculate the difference between the current ice thickness and the safe thickness. If the current ice thickness is much higher than the safety value (such as above 3.5mm), a full-power de-icing strategy is required; if the thickness slightly exceeds the safety threshold (such as 1.5mm), low-power de-icing or intermittent de-icing is adopted to reduce energy consumption. Combined with the distribution of ice in different areas of the blade, the de-icing process is regionalized and optimized, that is, the blade areas with severe icing are focused on heating, while lower-power heating is adopted for areas with lighter ice to reduce unnecessary energy consumption.

[0054] Furthermore, after determining the thickness of the de-icing layer, it is necessary to calculate the target operating state of the gas-heat de-icing device to ensure that the ice thickness can be effectively reduced to a safe range. This process includes: 1) Simulate the reverse development process of ice thickness: Based on the thermodynamic simulation model, calculate the melting rate of the ice layer under different heating power and wind speed conditions to derive the optimal heating strategy.

[0055] 2) Adjust the heating module: If the ice layer to be de-iced is thick (e.g. >3mm), the heating module will be set to high temperature mode to provide stronger heat input; if the ice layer is thin (e.g. 1~2mm), a lower temperature will be used to melt the ice to reduce unnecessary energy consumption.

[0056] 3) Adjust the blower speed: According to the power setting of the heating module, adjust the blower speed to optimize the circulation path of the hot air and improve the heating efficiency. For example, the high wind speed mode can quickly spread the hot air to the entire surface of the blade, while the low wind speed mode is suitable for more delicate local deicing.

[0057] 4) Adjust the blade angle: If the fan is in a lower wind speed environment, fine-tune the blade pitch angle so that the blade receives hot air at the optimal angle to ensure uniform deicing and improve heat transfer efficiency.

[0058] Furthermore, based on the current operating state of each gas-heat deicing device and the calculated target operating state, a specific deicing control plan is finally generated. When the ice thickness far exceeds the safety threshold (such as above 3.5mm), the highest power deicing mode is entered, all heating modules operate at full power, the blower maintains the highest wind speed, and the blade angle is adjusted to the optimal position to ensure rapid ice melting. If the ice thickness is high but not at the limit (such as 2mm~3mm), a progressive deicing strategy is adopted, first preheating at medium power, and then gradually increasing the power to prevent heat waste caused by overheating in a short time. If the ice coverage of different blade areas is quite different, the zone heating strategy will be followed, that is, the area with thicker ice will be heated, and the low power mode will be adopted for the area with thinner ice to optimize energy consumption. When the ice layer is thin (such as 1mm~2mm), the strategy of intermittent air supply and short-time heating is adopted to complete the deicing process with minimum power, reduce energy consumption and prevent blade overheating.

[0059] Step S40: executing the control scheme, and updating the deicing prediction level of the wind turbine generator set in real time during the execution process, so as to modify the control scheme of the corresponding thermal deicing device when the deicing prediction level changes.

[0060] Specifically, during the execution process, the operating status information of the wind turbine generator set during the execution process is collected in real time; feature extraction is performed on the operating status information during the execution process to obtain the operating status features during the execution process; based on the operating status features during the execution process, a pre-trained deicing state prediction model is called to determine the deicing prediction level under the current operating state; if the deicing prediction level under the current operating state is different from the deicing prediction level corresponding to the control scheme being executed, then the control scheme of the gas thermal deicing device is determined based on the deicing prediction level under the current operating state as a new control scheme, and the new control scheme is executed.

[0061] In the embodiment of the present invention, during the deicing process of the wind turbine generator set, the external environmental conditions and the ice coating of the blades will change continuously, so the deicing control scheme needs to have the ability to adjust in real time. In the process of executing the control scheme, the present invention continuously collects the operating status information of the wind turbine, including ambient temperature, humidity, wind speed, blade surface temperature, ice coating thickness, operating status of the deicing device, etc. In order to ensure the validity of the data, feature extraction is performed on this information, abnormal data is removed, and a deep learning model is used to analyze the operating status characteristics to evaluate the current deicing prediction level.

[0062] Furthermore, during the deicing process, the pre-trained deicing state prediction model is continuously called to update the deicing prediction level according to the real-time monitoring data. If the prediction level changes, for example, the deicing effect is not as good as expected, or the environmental conditions suddenly change and cause the ice to cover faster, the optimal deicing control strategy is immediately recalculated. For example, when the ice layer has not yet reached a safe thickness, the heating power is increased, the wind speed is increased, or the blade angle is adjusted to optimize the heat flow distribution; on the contrary, if it is detected that the ice layer has basically melted, the power is reduced and the blower operation time is reduced to save energy.

[0063] Figure 2 FIG. 1 is a system structure diagram of a wind turbine blade deicing control system provided by an embodiment of the present invention. Figure 2 As shown, an embodiment of the present invention provides a de-icing control system for wind turbine blades, the system comprising: a collection unit, used to collect operating status information of the wind turbine generator set, and perform feature extraction on the operating status information to obtain operating status characteristics; a prediction unit, used to call a pre-trained de-icing status prediction model based on the operating status characteristics to determine a de-icing prediction level under the current operating state; a scheme generation unit, used to determine a control scheme of a gas-thermal de-icing device based on the de-icing prediction level; an execution unit, used to execute the control scheme and update the de-icing prediction level of the wind turbine generator set in real time during the execution process, so as to execute the control scheme correction of the corresponding thermal de-icing device when the de-icing prediction level changes. The embodiment of the present invention further provides a computer-readable storage medium, on which instructions are stored, which, when executed on a computer, enable the computer to execute the above-mentioned deicing control method for wind turbine blades.

[0064] Those skilled in the art will understand that all or part of the steps in the method for implementing the above-mentioned embodiments can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including several instructions for making a single-chip microcomputer, a chip or a processor execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0065] The optional embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. However, the embodiments of the present invention are not limited to the specific details in the above embodiments. Within the technical concept of the embodiments of the present invention, the technical scheme of the embodiments of the present invention can be subjected to a variety of simple modifications, and these simple modifications all belong to the protection scope of the embodiments of the present invention. It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the embodiments of the present invention will not further describe various possible combinations.

[0066] In addition, various embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the embodiments of the present invention, they should also be regarded as the contents disclosed in the embodiments of the present invention.

Claims

1. A deicing control method for wind turbine blades, characterized in that: The method comprises: Collecting the operating status information of the wind turbine generator set, and performing feature extraction on the operating status information to obtain operating status features; Based on the operating state characteristics, a pre-trained deicing state prediction model is called to determine a deicing prediction level under the current operating state; determining a control scheme for an air-heat deicing device based on the deicing prediction level; The control scheme is executed, and the deicing prediction level of the wind turbine generator set is updated in real time during the execution process, so as to modify the control scheme corresponding to the thermal deicing device when the deicing prediction level changes.

2. The method according to claim 1, characterized in that The operating status information of the wind turbine generator set includes: Any one or more of weather forecast information, real-time environmental parameter information and real-time ice thickness information of wind turbine blades.

3. The method according to claim 1, characterized in that The performing feature extraction on the running status information to obtain the running status feature includes: Performing preprocessing on the operation status information; One-hot encoding is performed on the preprocessed running status information to obtain a numerical vector of each preprocessed running status information; Based on the AHP hierarchical analysis method, feature fusion is performed on the numerical vectors of each preprocessed operating status information to obtain a comprehensive feature vector as the operating status feature.

4. The method according to claim 1, characterized in that: The deicing prediction level includes a prevention level and a deicing level; The deicing level includes multiple levels, and the higher the level, the thicker the corresponding deicing layer.

5. The method according to claim 4, characterized in that The gas-heat deicing device comprises: Heating module, blower and blade angle adjustment device; The control scheme for determining the gas-heat deicing device based on the deicing prediction level includes: determining target operating states of each gas-heat deicing device based on the deicing prediction level; A corresponding control scheme is generated based on the current operating state of each gas-heat deicing device and the corresponding target operating state.

6. The method according to claim 5, characterized in that If the current deicing prediction level is a prevention level, the control scheme of the gas-heat deicing device is determined based on the deicing prediction level, including: Predicting the freezing time based on the operating status information of the wind turbine generator set to obtain the predicted freezing time; Determine the intervention time of each gas-heat deicing device based on the predicted icing time; Based on the intervention time and the preset intermittent air supply timing table, an intermittent air supply operation plan for each gas-heat deicing device is generated as a control plan for the gas-heat deicing device.

7. The method according to claim 5, characterized in that If the current deicing prediction level is the deicing level, the control scheme of the gas-heat deicing device is determined based on the deicing prediction level, including: Determine the real-time ice thickness information of wind turbine blades based on the operating status information of wind turbine generator sets, and predict the development of ice thickness; Determine the thickness of the layer to be de-iced based on the real-time ice thickness information and the prediction results of ice thickness development; Based on the thickness of the layer to be de-iced, the ice thickness is simulated to develop inversely toward the preset value of the safe ice thickness, and the target operating state of each gas-thermal de-icing device is generated. Based on the current operating state of each gas-thermal de-icing device and the corresponding target operating state, a corresponding control scheme is generated.

8. The method according to claim 1, characterized in that During the execution process, the deicing prediction level of the wind turbine generator set is updated in real time, so as to modify the control scheme of the corresponding thermal deicing device when the deicing prediction level changes, including: During the execution process, the operating status information of the wind turbine generator set is collected in real time; Perform feature extraction on the running state information during the execution process to obtain the running state features during the execution process; Based on the operating state characteristics during the execution, a pre-trained deicing state prediction model is called to determine a deicing prediction level under the current operating state; If the deicing prediction level under the current operating state is different from the deicing prediction level corresponding to the control scheme being executed, the control scheme of the gas thermal deicing device is determined based on the deicing prediction level under the current operating state as a new control scheme, and the new control scheme is executed.

9. A de-icing control system for wind turbine blades, characterized in that: The system comprises: A collection unit, used to collect the operating status information of the wind turbine generator set, and perform feature extraction on the operating status information to obtain operating status features; A prediction unit, configured to call a pre-trained deicing state prediction model based on the operating state characteristics to determine a deicing prediction level under a current operating state; a scheme generating unit, configured to determine a control scheme of the gas-thermal deicing device based on the deicing prediction level; The execution unit is used to execute the control scheme and update the deicing prediction level of the wind turbine generator set in real time during the execution process, so as to perform the control scheme correction of the corresponding thermal deicing device when the deicing prediction level changes.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the deicing control method for wind turbine blades according to any one of claims 1 to 8.

Images