Deicing device starting method and device, storage medium and computer program product

Abstract

The invention discloses a starting method and device of a deicing device, a storage medium and a computer program product, and relates to the field of wind power generation, and the starting method of the deicing device comprises the steps: obtaining operation data of a wind turbine generator through a data collection system of the wind turbine generator, and extracting target features from the operation data through a feature selection method, the operation data comprises environment temperature, blade temperature, environment humidity, environment wind speed, blade angle and vibration condition of a cabin; training the target features through an integrated learning method to determine whether the blades of the wind turbine generator set are frozen or not; and under the condition that it is determined that the blade is frozen, a deicing device in the blade is started, and the deicing device is used for eliminating ice formed on the blade. By the adoption of the technical scheme, the problem that the deicing device is started inaccurately due to the fact that only temperature serves as the starting and stopping standard of the deicing device and other factors causing blade icing are not considered is solved.

Description

Technical Field

[0001] The present application relates to the field of wind power generation, and in particular to a method and device for starting a deicing device, a storage medium and a computer program product. Background Art

[0002] In the field of wind power generation, anemometers are key sensors for wind turbines, and their accuracy and reliability directly affect the normal operation and maintenance of wind turbines. However, in the cold winter, anemometer blades are prone to ice formation, resulting in distortion of measurement data and damage to equipment, thus affecting the normal operation and economic benefits of wind turbines.

[0003] In the related art, the start and stop control of the de-icing device mainly depends on the measurement data of the temperature sensor. However, only the temperature is used as the start and stop standard of the de-icing device without considering other factors that cause blade icing, and the start and stop of the de-icing device cannot be accurately controlled.

[0004] In the related art, only temperature is used as the start and stop standard of the de-icing device, and other factors that cause blade icing are not considered, resulting in inaccurate start-up of the de-icing device. No effective solution has been proposed so far.

[0005] Therefore, it is necessary to improve the related technology to overcome the above-mentioned defects in the related technology. Summary of the invention

[0006] The embodiments of the present application provide a method and device for starting a de-icing device, a storage medium and a computer program product, so as to at least solve the problem in the related art that only temperature is used as the start and stop standard of the de-icing device, and other factors that cause blade icing are not considered, resulting in inaccurate start-up of the de-icing device.

[0007] According to one aspect of an embodiment of the present application, a method for starting a de-icing device is provided, comprising: acquiring operating data of a wind turbine set through a data acquisition system of the wind turbine set, and extracting target features from the operating data through a feature selection method, wherein the operating data include: ambient temperature, blade temperature, ambient humidity, ambient wind speed, blade angle, and vibration of the cabin; training the target features through an integrated learning method to determine whether the blades of the wind turbine set are frozen; and when it is determined that the blades are frozen, starting a de-icing device inside the blades, wherein the de-icing device is used to remove ice formed on the blades.

[0008] In an exemplary embodiment, target features are extracted from the operating data by a feature selection method, including: calculating a first correlation coefficient between each feature data in the operating data and the icing state of the blade by a machine learning method; and determining that the feature data corresponding to the second correlation coefficient with the highest score in the first correlation coefficients is the target feature.

[0009] In an exemplary embodiment, after starting the de-icing device inside the blade, the method further includes: obtaining the temperature and humidity of the blade surface through a sensor of the blade; when the temperature of the blade surface reaches a first threshold and the humidity of the blade surface reaches a second threshold, determining that the ice on the blade has been completely removed, and turning off the de-icing device.

[0010] In an exemplary embodiment, after detecting whether the ice on the blade has been completely removed, the method further includes: determining that the ice removed from the blade has not reached a target threshold when the temperature of the blade surface has not reached a first threshold or the humidity of the blade surface has not reached a second threshold; and determining a current first heating gear of the de-icing device when the de-icing device uses a heating technology to remove ice from the blade, adjusting the heating gear of the de-icing device, and heating the blade at a second heating gear higher than the first heating gear.

[0011] In an exemplary embodiment, after determining that the ice removed from the blade has not reached a target threshold, the method further includes: when the de-icing device uses vibration technology to remove ice from the blade, determining a current first vibration level of the de-icing device, adjusting the vibration level of the de-icing device, and vibrating the blade at a second vibration level higher than the first vibration level.

[0012] In an exemplary embodiment, after acquiring the operating data of the wind turbine group through the data acquisition system of the wind turbine group, the method further includes: when it is determined that there are missing values ​​in the operating data, determining whether the number of missing values ​​in the operating data is less than a first preset value; when it is determined that the number of missing values ​​in the operating data is less than the first preset value and the number of operating data is greater than a second preset value, deleting all operating data at the time corresponding to the missing values ​​in the operating data.

[0013] In an exemplary embodiment, after determining whether the number of missing values ​​in the operating data is less than a first preset value, the method further includes: when it is determined that the number of missing values ​​in the operating data is greater than the first preset value, determining the data type corresponding to the missing value; calculating the median of the data type corresponding to the missing value through machine learning; and replacing the missing value with the median.

[0014] According to another aspect of an embodiment of the present application, a starting device for a de-icing device is also provided, including: an extraction module, used to obtain operating data of the wind turbine set through a data acquisition system of the wind turbine set, and extract target features from the operating data through a feature selection method, wherein the operating data includes: ambient temperature, blade temperature, ambient humidity, ambient wind speed, blade angle, and vibration of the cabin; a determination module, used to train the target features through an integrated learning method to determine whether the blades of the wind turbine set are frozen; a starting module, used to start a de-icing device inside the blade when it is determined that the blade is already frozen, wherein the de-icing device is used to remove ice formed on the blade.

[0015] According to another aspect of the embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned de-icing device startup method when running.

[0016] According to another aspect of an embodiment of the present application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the method for starting the de-icing device through the computer program.

[0017] According to another aspect of the embodiments of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps of the method described in each embodiment of the present application are implemented.

[0018] Through this application, the operation data of the wind turbine is obtained, and the target features are extracted from the operation data through the feature selection method; the target features are trained through the integrated learning method to determine whether the blades of the wind turbine are frozen; when it is determined that the blades are frozen, the deicing device inside the blades is started. This solves the problem in the related art that only temperature is used as the start and stop standard of the deicing device, and other factors that cause blade icing are not considered, resulting in inaccurate start of the deicing device. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0020] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0021] Figure 1 It is a hardware structure block diagram of a method for starting a deicing device according to an embodiment of the present application;

[0022] Figure 2 is a flow chart of a method for starting a deicing device according to an embodiment of the present application;

[0023] Figure 3 is a flow chart of a method for starting a deicing device according to an embodiment of the present application;

[0024] Figure 4 It is a structural block diagram of a starting device of a deicing device according to an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] The method embodiments provided in the embodiments of the present application can be executed in the blades of a wind turbine or a similar computing device. Taking the operation on the blades of a wind turbine as an example, Figure 1 FIG. 1 is a hardware structure diagram of a blade of a wind turbine generator set in a method for starting a deicing device according to an embodiment of the present application. Figure 1 As shown, the blades of the wind turbine generator may include one or more ( Figure 1Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor (Central Processing Unit, MCU) or a programmable logic device (Field Programmable Gate Array, FPGA)) and a memory 104 for storing data, wherein the blade of the above-mentioned wind turbine may also include a transmission device 106 for communication function and an input and output device 108. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the blades of the above-mentioned wind turbine. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0028] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the startup method of the deicing device in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the blades of the wind turbine through a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0029] The wireless network provided by the communication provider of the blades of the wind turbine. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (Radio Frequency, referred to as RF) module, which is used to communicate with the Internet wirelessly.

[0030] In this embodiment, a method for starting a deicing device is provided, which is applied to the blades of the above-mentioned wind turbine generator set. Figure 2 is a flow chart of a method for starting a deicing device according to an embodiment of the present application. Figure 2 As shown, the process includes the following steps:

[0031] Step S202, obtaining the operation data of the wind turbine generator set through a data acquisition system of the wind turbine generator set, and extracting target features from the operation data through a feature selection method, wherein the operation data includes: ambient temperature, blade temperature, ambient humidity, ambient wind speed, blade angle, and vibration of the nacelle;

[0032] Step S204, training the target feature by an integrated learning method to determine whether the blades of the wind turbine are frozen;

[0033] Step S206, when it is determined that the blade has been frozen, starting a de-icing device inside the blade, wherein the de-icing device is used to remove ice formed on the blade.

[0034] Through the above steps, the operation data of the wind turbine is acquired through the data acquisition system of the wind turbine, and the target features are extracted from the operation data through the feature selection method, wherein the operation data includes: ambient temperature, blade temperature, ambient humidity, ambient wind speed, blade angle, and vibration of the cabin; the target features are trained through the integrated learning method to determine whether the blades of the wind turbine are frozen; when it is determined that the blades are frozen, the deicing device inside the blade is started, wherein the deicing device is used to remove the ice on the blade. This solves the problem in the related art that only temperature is used as the start and stop standard of the deicing device, and other factors that cause blade icing are not considered, resulting in inaccurate start of the deicing device.

[0035] It should be clarified that the blade angle refers to the angle between the rotation plane of the blades in the wind turbine and the wind direction. When the wind speed is low, the blade angle is usually set to a smaller value to increase the wind capture area of ​​the blades and improve the utilization rate of wind energy. When the wind speed is high, in order to prevent the wind turbine from overloading or blade damage, the blade angle will be set to a larger value to reduce the wind capture area of ​​the blades to reduce the impact of wind on the wind turbine. When there is frost on the blades, the weight and shape of the frost will change the aerodynamic characteristics of the blades, resulting in a reduction in the adjustment effect of the blade angle, thereby affecting the output power and operating status of the wind turbine.

[0036] In an exemplary embodiment, target features are extracted from the operating data by a feature selection method, including: calculating a first correlation coefficient between each feature data in the operating data and the icing state of the blade by a machine learning method; and determining that the feature data corresponding to the second correlation coefficient with the highest score in the first correlation coefficients is the target feature.

[0037] Use machine learning methods, such as correlation coefficient, chi-square test, mutual information, recursive feature elimination, LASSO regression, feature importance, etc., to perform feature selection on the preprocessed operating data, and calculate the first correlation coefficient, i.e., the correlation score, between each feature data and the blade icing state to evaluate the degree of influence of the feature data on the blade icing state. Based on the score of the first correlation coefficient, determine the feature data corresponding to the second correlation coefficient with the highest score as the data feature most relevant to the blade icing state, i.e., the target feature. Among them, the target feature can be one or more, depending on the score distribution of the first correlation coefficient and the prediction requirements of the model.

[0038] Optionally, assume that the following characteristic data are extracted from the operation data of the wind turbine: ambient temperature, relative humidity, wind speed, duration, blade angle, cabin vibration, etc. Through calculation, the first correlation coefficient between each characteristic data and the blade icing state is obtained, which is 0.78 for ambient temperature, 0.83 for relative humidity, 0.65 for wind speed, 0.72 for duration, 0.88 for blade angle, and 0.75 for cabin vibration. Among them, the first correlation coefficient score of the blade angle is the highest, which is 0.88. Therefore, it can be determined that the blade angle is the data feature most relevant to the blade icing state, that is, the target feature.

[0039] In an exemplary embodiment, after starting the de-icing device inside the blade, the method further includes: obtaining the temperature and humidity of the blade surface through a sensor of the blade; when the temperature of the blade surface reaches a first threshold and the humidity of the blade surface reaches a second threshold, determining that the ice on the blade has been completely removed, and turning off the de-icing device.

[0040] Optionally, it is assumed that when the temperature of the blade surface reaches 0°C, the deicing device is started; when the temperature of the blade surface reaches 5°C and the humidity of the blade surface reaches 40%, the deicing device is turned off. During the deicing process, the following real-time data are obtained through the sensor on the blade surface: at 00:00, the temperature of the blade surface is -5°C and the humidity of the blade surface is 85%; at 00:10, the temperature of the blade surface is -2°C and the humidity of the blade surface is 80%; at 00:20, the temperature of the blade surface is 0°C and the humidity of the blade surface is 75%; at 00:30, the temperature of the blade surface is 2°C and the humidity of the blade surface is 70%; at 00:40, the temperature of the blade surface is 4°C and the humidity of the blade surface is 65%; at 00:50, the temperature of the blade surface is 6°C and the humidity of the blade surface is 42%. From the above data, it can be seen that from 00:20, the temperature of the blade surface reaches 0°C and the de-icing device is started; from 00:50, the temperature of the blade surface reaches 6°C and the humidity of the blade surface reaches 42%, which meets the conditions for shutting down the de-icing device. Therefore, the de-icing device is shut down and the de-icing process is stopped.

[0041] In an exemplary embodiment, after detecting whether the ice on the blade has been completely removed, the method further includes: determining that the ice removed from the blade has not reached a target threshold when the temperature of the blade surface has not reached a first threshold or the humidity of the blade surface has not reached a second threshold; and determining a current first heating gear of the de-icing device when the de-icing device uses a heating technology to remove ice from the blade, adjusting the heating gear of the de-icing device, and heating the blade at a second heating gear higher than the first heating gear.

[0042] Optionally, it is assumed that when the temperature of the blade surface reaches 5°C or the humidity of the blade surface reaches 40%, it is determined that there is no unremoved ice on the blade. During the deicing process, the following real-time data is obtained through the sensor on the blade surface: at 00:00, the temperature of the blade surface is -5°C, the humidity of the blade surface is 85%, and the current heating gear is low; at 00:10, the temperature of the blade surface is -2°C, the humidity of the blade surface is 75%, and the current heating gear is medium; at 00:20, the temperature of the blade surface is 1°C, the humidity of the blade surface is 60%, and the current heating gear is high; at 00:30, the temperature of the blade surface is 6°C, the humidity of the blade surface is 42%, and the deicing device is turned off. From the above data, it can be seen that starting from 00:20, in order to accelerate the elimination of frost, the heating gear is increased to high, until 00:30, the temperature of the blade surface reaches the first threshold and the humidity of the blade surface reaches the second threshold, at which time the deicing device is turned off.

[0043] In an exemplary embodiment, after determining that the ice removed from the blade has not reached a target threshold, the method further includes: when the de-icing device uses vibration technology to remove ice from the blade, determining a current first vibration level of the de-icing device, adjusting the vibration level of the de-icing device, and vibrating the blade at a second vibration level higher than the first vibration level.

[0044] Optionally, it is assumed that when the temperature of the blade surface reaches 5°C or the humidity of the blade surface reaches 40%, it is determined that there is no unremoved ice on the blade. In the case where the deicing device adopts vibration technology, if the temperature of the blade surface fails to reach 5°C or the humidity fails to reach 40%, the blade is vibrated at a gear higher than the current vibration gear by increasing the vibration gear. For example, at 00:00, the temperature of the blade surface is 0°C, the humidity of the blade surface is 65%, and the current vibration gear is low. The gear of the current vibration gear is increased and the blade is de-iced at the medium gear; at 00:10, the temperature of the blade surface is 6°C, the humidity of the blade surface is 42%, and the de-icing device is turned off. From the above data, it can be seen that starting from 00:00, in order to accelerate the elimination of frost, the vibration gear is increased from low gear to medium gear; until 00:10, the temperature of the blade surface reaches the first threshold and the humidity of the blade surface reaches the second threshold, at which time the de-icing device is turned off.

[0045] In an exemplary embodiment, after acquiring the operating data of the wind turbine group through the data acquisition system of the wind turbine group, the method further includes: when it is determined that there are missing values ​​in the operating data, determining whether the number of missing values ​​in the operating data is less than a first preset value; when it is determined that the number of missing values ​​in the operating data is less than the first preset value and the number of operating data is greater than a second preset value, deleting all operating data at the time corresponding to the missing values ​​in the operating data.

[0046] Optionally, assume that the wind turbine operating data is as follows:

[0047]

[0048] Assume that the first preset value is 2 and the second preset value is 5. In the above data, there are missing values ​​in the running data of 00:10 and 00:40. The number of missing values ​​is 2, which is less than the first preset value; the number of running data is 6, which is greater than the second preset value. At this time, delete all the running data of 00:10 and 00:40, and get the following complete data:

[0049]

[0050] In an exemplary embodiment, after determining whether the number of missing values ​​in the operating data is less than a first preset value, the method further includes: when it is determined that the number of missing values ​​in the operating data is greater than the first preset value, determining the data type corresponding to the missing value; calculating the median of the data type corresponding to the missing value through machine learning; and replacing the missing value with the median.

[0051] Optionally, assume that the wind turbine operating data is as follows:

[0052]

[0053] Assume that the first preset value is 2. In the above data, from 00:10 to 00:30, there are 3 missing values ​​in the humidity data of the blade surface, which exceeds the first preset value. At this time, it is necessary to determine the data type corresponding to the missing value, that is, the humidity of the blade surface. Then, the median of the humidity of the blade surface is calculated by machine learning. In the above data, the median of the humidity of the blade surface is 65%. Replace the missing values ​​with the median 65% to obtain the following complete data:

[0054]

[0055] In order to better understand the process of the above-mentioned deicing device startup method, the above-mentioned deicing device startup method is described below in combination with an optional embodiment, but it is not used to limit the technical solution of the embodiment of the present application.

[0056] Figure 3 is a schematic diagram of a method for starting a deicing device according to an embodiment of the present application, such as Figure 3 As shown, specifically including the following:

[0057] Step S302: Collect real-time or historical data from the supervisory control and data acquisition (SCADA) system of the wind turbine, including parameters such as ambient temperature, ambient humidity, ambient wind speed, cabin vibration, blade angle, etc. After collecting the operating data, preprocess the operating data to improve the performance of the model. The preprocessing steps include: missing value processing, that is, when the temperature, humidity and other information of certain data points are missing, interpolation methods (such as linear interpolation) or predictions based on other features can be used to estimate the missing values. Outlier detection, that is, when outliers appear in the operating data (such as sudden temperature changes), statistical methods (such as Z-score) can be used to detect and process these outliers. Normalization, that is, normalize all operating data so that all operating data are in the same numerical range to avoid some operating data from affecting model training due to excessive numerical ranges.

[0058] Step S304: Combine different types of sensor data (such as ambient temperature, ambient humidity, ambient wind speed, cabin vibration, etc.) into mixed features to capture more complex environmental factors that affect blade icing. For example, construct a feature to represent the combined effect of ambient temperature and ambient humidity, or a feature to represent the combined effect of ambient wind speed and cabin vibration. In this way, the model can not only learn individual environmental factors, but also learn the interactions between these factors.

[0059] Step S306: Use the Stacked-XGBoost algorithm to build a prediction model, which is an integrated learning method that can stack multiple different machine learning models, such as XGBoost, random forest, support vector machine (SVM), etc., arbitrarily arrange the prediction order of the machine learning model, and train the currently used machine learning model based on the operating data of the wind turbine and the prediction results of the previously used machine learning model on the operating data to improve the accuracy of the prediction. Optionally, assume that an XGBoost model is trained first, and then a random forest model is trained using the prediction results of the XGBoost model and the operating data of the wind turbine. Then, another model, such as SVM, is used to integrate the prediction results of the first two models. This process may be repeated multiple times to form a stacked model, in which each model is learned based on the prediction results of the previous model.

[0060] Step S308: During the model training process, the model parameters are adjusted to optimize the performance of the model. The model parameters include the number of trees, the maximum depth of the tree, the learning rate, etc. Optionally, assume that there are 100 trees (base learners) in the Stacked-XGBoost model. By increasing the number of trees to 200, the prediction performance of the model can be significantly improved. Alternatively, reducing the maximum depth of the tree from 6 to 3 can reduce the risk of overfitting of the model and improve the generalization ability of the model.

[0061] Step S310: After the model training and tuning are completed, the model is deployed on the actual wind turbine, and the blade icing state predicted by the model is compared with the actual blade icing state to evaluate the prediction performance of the model. Optionally, a wind turbine located in a cold zone is selected and the trained model is deployed to the SCADA system. Then, the model is allowed to predict the blade icing state for the next 24 hours, and the prediction results are compared with the actual blade icing state. If the prediction results of the model are highly consistent with the actual state, it means that the performance of the model is good; if the consistency is low, the model needs to be further tuned or more data is used for training. In order to have a clearer understanding of the prediction performance of the model, in the testing phase, various quantitative indicators such as accuracy, recall rate and area under the curve (Area Under the Curve, referred to as AUC) are used to comprehensively evaluate the prediction performance of the model. Optionally, it is assumed that the accuracy of the model in predicting blade icing is 90%, but the recall rate is only 80%. This means that the model is very accurate in predicting icing, but the missed detection rate is high when detecting all icing events. In order to balance the accuracy and recall rate, it is necessary to adjust the model threshold or use other models for stacking to improve the performance of the model.

[0062] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.

[0063] In this embodiment, a deicing device starting device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made are not repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0064] Figure 4 : is a structural block diagram of a starting device for a deicing device according to an embodiment of the present application, the device comprising:

[0065] The extraction module 40 is used to obtain the operation data of the wind turbine set through the data acquisition system of the wind turbine set, and extract the target features from the operation data through the feature selection method, wherein the operation data includes: ambient temperature, blade temperature, ambient humidity, ambient wind speed, blade angle, and vibration of the cabin;

[0066] A determination module 42, configured to train the target feature by an integrated learning method to determine whether the blades of the wind turbine are frozen;

[0067] The starting module 44 is used to start the deicing device inside the blade when it is determined that the blade has been iced, wherein the deicing device is used to remove ice formed on the blade.

[0068] Through the above device, the operation data of the wind turbine is obtained through the data acquisition system of the wind turbine, and the target features are extracted from the operation data through the feature selection method, wherein the operation data includes: ambient temperature, blade temperature, ambient humidity, ambient wind speed, blade angle, and vibration of the cabin; the target features are trained through the integrated learning method to determine whether the blades of the wind turbine are frozen; when it is determined that the blades are frozen, the deicing device inside the blade is started, wherein the deicing device is used to remove the ice on the blade. This solves the problem in the related art that only temperature is used as the start and stop standard of the deicing device, and other factors that cause blade icing are not considered, resulting in inaccurate start of the deicing device.

[0069] In an exemplary embodiment, the extraction module 40 is also used to calculate the first correlation coefficient between each characteristic data in the operating data and the icing state of the blade through a machine learning method; and determine that the characteristic data corresponding to the second correlation coefficient with the highest score in the first correlation coefficients is the target feature.

[0070] In an exemplary embodiment, the start module 44 is also used to obtain the temperature and humidity of the blade surface through the blade sensor; when the temperature of the blade surface reaches a first threshold and the humidity of the blade surface reaches a second threshold, it is determined that the ice on the blade has been completely removed and the de-icing device is turned off.

[0071] In an exemplary embodiment, the start module 44 is also used to determine that the ice removed from the blade has not reached a target threshold when the temperature of the blade surface has not reached a first threshold or the humidity of the blade surface has not reached a second threshold; when the de-icing device uses heating technology to remove ice from the blade, determine the current first heating gear of the de-icing device, adjust the heating gear of the de-icing device, and heat the blade at a second heating gear higher than the first heating gear.

[0072] In an exemplary embodiment, the start module 44 is also used to determine the current first vibration level of the de-icing device when the de-icing device adopts vibration technology to remove ice from the blades, and adjust the vibration level of the de-icing device to vibrate the blades at a second vibration level higher than the first vibration level.

[0073] In an exemplary embodiment, the extraction module 40 is also used to determine whether the number of missing values ​​in the operating data is less than a first preset value when it is determined that there are missing values ​​in the operating data; when it is determined that the number of missing values ​​in the operating data is less than the first preset value and the number of operating data is greater than a second preset value, delete all operating data at the time corresponding to the missing values ​​in the operating data.

[0074] In an exemplary embodiment, the extraction module 40 is also used to determine the data type corresponding to the missing value when it is determined that the number of missing values ​​in the operating data is greater than the first preset value; calculate the median of the data type corresponding to the missing value through machine learning; and replace the missing value with the median.

[0075] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when run.

[0076] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0077] S1, obtaining the operation data of the wind turbine set through a data acquisition system of the wind turbine set, and extracting target features from the operation data through a feature selection method, wherein the operation data includes: ambient temperature, blade temperature, ambient humidity, ambient wind speed, blade angle, and vibration of the cabin;

[0078] S2, training the target feature by an integrated learning method to determine whether the blades of the wind turbine are frozen;

[0079] S3, when it is determined that the blade has been frozen, starting a deicing device inside the blade, wherein the deicing device is used to remove ice formed on the blade.

[0080] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0081] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0082] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0083] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0084] S1, obtaining the operation data of the wind turbine set through a data acquisition system of the wind turbine set, and extracting target features from the operation data through a feature selection method, wherein the operation data includes: ambient temperature, blade temperature, ambient humidity, ambient wind speed, blade angle, and vibration of the cabin;

[0085] S2, training the target feature by an integrated learning method to determine whether the blades of the wind turbine are frozen;

[0086] S3, when it is determined that the blade has been frozen, starting a deicing device inside the blade, wherein the deicing device is used to remove ice formed on the blade.

[0087] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0088] An embodiment of the present application further provides a computer program product, comprising a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program product, and when the computer program is executed by a processor, the steps of the method described in each embodiment of the present application are implemented.

[0089] Optionally, in this embodiment, the above computer program may be configured to implement the following steps when executed by a processor:

[0090] S1, obtaining the operation data of the wind turbine set through a data acquisition system of the wind turbine set, and extracting target features from the operation data through a feature selection method, wherein the operation data includes: ambient temperature, blade temperature, ambient humidity, ambient wind speed, blade angle, and vibration of the cabin;

[0091] S2, training the target feature by an integrated learning method to determine whether the blades of the wind turbine are frozen;

[0092] S3, when it is determined that the blade has been frozen, starting a deicing device inside the blade, wherein the deicing device is used to remove ice formed on the blade.

[0093] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0094] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0095] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for starting a deicing device, characterized in that: include: Acquiring the operation data of the wind turbine generator set through a data acquisition system of the wind turbine generator set, and extracting target features from the operation data through a feature selection method, wherein the operation data includes: ambient temperature, blade temperature, ambient humidity, ambient wind speed, blade angle, and vibration of the cabin; Training the target feature by an integrated learning method to determine whether the blades of the wind turbine are frozen; When it is determined that the blade has been iced, a de-icing device inside the blade is started, wherein the de-icing device is used to remove ice formed on the blade.

2. The method for starting a deicing device according to claim 1, characterized in that: Extracting target features from the operation data by a feature selection method includes: Calculating a first correlation coefficient between each characteristic data in the operation data and the icing state of the blade by a machine learning method; The feature data corresponding to the second correlation coefficient with the highest score among the first correlation coefficients is determined as the target feature.

3. The method for starting a deicing device according to claim 1, characterized in that: After starting the deicing device inside the blade, the method further includes: Acquiring the temperature and humidity of the surface of the blade through the sensor of the blade; When the temperature of the blade surface reaches a first threshold and the humidity of the blade surface reaches a second threshold, it is determined that the ice on the blade has been completely removed, and the de-icing device is turned off.

4. The method for starting a deicing device according to claim 3, characterized in that: After detecting whether the ice on the blade has been completely removed, the method further includes: When the temperature of the blade surface does not reach a first threshold or the humidity of the blade surface does not reach a second threshold, determining that the ice removed from the blade does not reach a target threshold; In the case where the deicing device uses heating technology to remove ice from the blades, the current first heating gear of the deicing device is determined, and the heating gear of the deicing device is adjusted to heat the blades at a second heating gear higher than the first heating gear.

5. The method for starting a deicing device according to claim 4, characterized in that: After determining that the amount of ice removed from the blade has not reached a target threshold, the method further includes: In the case where the deicing device uses vibration technology to remove ice from the blades, the current first vibration level of the deicing device is determined, and the vibration level of the deicing device is adjusted to vibrate the blades at a second vibration level higher than the first vibration level.

6. The method for starting a deicing device according to claim 1, characterized in that: After acquiring the operating data of the wind turbine generator set through the data acquisition system of the wind turbine generator set, the method further includes: In the case where it is determined that there are missing values ​​in the operating data, determining whether the number of missing values ​​in the operating data is less than a first preset value; When it is determined that the number of missing values ​​in the operating data is less than the first preset value and the number of the operating data is greater than the second preset value, all operating data at the time points corresponding to the missing values ​​in the operating data are deleted.

7. The method for starting a deicing device according to claim 6, characterized in that: After determining whether the number of missing values ​​in the operating data is less than a first preset value, the method further includes: In a case where it is determined that the number of missing values ​​in the operating data is greater than the first preset value, determining the data type corresponding to the missing value; Calculate the median of the data type corresponding to the missing value through machine learning; The missing values ​​were replaced by the median.

8. A de-icing device starting device, characterized in that: include: An extraction module is used to obtain the operation data of the wind turbine set through the data acquisition system of the wind turbine set, and extract target features from the operation data through a feature selection method, wherein the operation data includes: ambient temperature, blade temperature, ambient humidity, ambient wind speed, blade angle, and vibration of the cabin; A determination module, used for training the target feature by an integrated learning method to determine whether the blades of the wind turbine are frozen; The starting module is used to start the deicing device inside the blade when it is determined that the blade has been frozen, wherein the deicing device is used to remove ice formed on the blade.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method according to any one of claims 1 to 7 when executed.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images