Icing early warning method and device, electronic equipment and storage medium

Abstract

The invention provides an icing early warning method and device, electronic equipment and a storage medium, and relates to the technical field of wind power generation. The method comprises the steps that training data are obtained, the training data comprise historical operation data of the wind generating set and the corresponding icing state, and the historical operation data comprise at least one of the wind speed and the power; the feature vector and the icing state are input into an icing prediction model to obtain an icing prediction result, and the feature vector is obtained by performing feature extraction on historical operation data; according to the difference between the icing prediction result and the icing state, training an icing prediction model to obtain a trained icing prediction model; and inputting the obtained feature vector of the real-time operation data into the trained icing prediction model to obtain an icing early warning result. The problem that icing early warning of the wind generating set is not accurate and not timely is solved, and the icing risk in the operation process of the wind generating set is effectively reduced.

Description

Technical Field

[0001] The present application relates to the technical field of wind power generation, and in particular to an icing warning method, device, electronic equipment and storage medium. Background Art

[0002] As a clean and renewable energy form, wind power generation has received widespread attention and promotion. However, wind turbines face some challenges during operation, one of which is the icing problem. When the blades and anemometers are frozen, it will cause performance degradation, damage and even safety risks.

[0003] The current technologies and methods for solving the icing problem of wind turbines are highly dependent on sensors, which leads to delays and inaccuracies in data collection. When using the collected data for analysis, there are certain limitations and it is impossible to fully tap the operating data of the wind turbine itself, which limits the timeliness and accuracy of icing warnings. Summary of the invention

[0004] The present application aims to solve one of the technical problems in the related art at least to some extent.

[0005] To this end, the first purpose of the present application is to propose an icing warning method to reduce the icing risk during the operation of a wind turbine.

[0006] The second objective of the present application is to provide an icing warning device.

[0007] The third objective of the present application is to provide an electronic device.

[0008] A fourth objective of the present application is to provide a computer-readable storage medium.

[0009] A fifth object of the present application is to provide a computer program product.

[0010] To achieve the above-mentioned purpose, the first embodiment of the present application proposes an icing warning method, comprising:

[0011] Acquire training data, wherein the training data includes historical operation data of the wind turbine generator set and corresponding icing status, and the historical operation data includes at least one of wind speed and power;

[0012] Inputting the feature vector and the icing state into an icing prediction model to obtain an icing prediction result, wherein the feature vector is obtained by feature extraction of the historical operation data;

[0013] According to the difference between the icing prediction result and the icing state, the icing prediction model is trained to obtain a trained icing prediction model;

[0014] The feature vector of the acquired real-time operation data is input into the trained icing prediction model to obtain an icing warning result.

[0015] To achieve the above-mentioned purpose, the second embodiment of the present application proposes an icing warning device, including:

[0016] An acquisition module, configured to acquire training data, wherein the training data includes historical operation data of the wind turbine generator set and corresponding icing status, and the historical operation data includes at least one of wind speed and power;

[0017] An input module, used for inputting a feature vector and the icing state into an icing prediction model to obtain an icing prediction result, wherein the feature vector is obtained by feature extraction of the historical operation data;

[0018] A training module, used for training the icing prediction model according to the difference between the icing prediction result and the icing state, so as to obtain a trained icing prediction model;

[0019] The early warning module is used to input the feature vector of the acquired real-time operation data into the trained icing prediction model to obtain an icing early warning result.

[0020] To achieve the above-mentioned purpose, the third aspect of the present application provides an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0021] The memory stores computer-executable instructions;

[0022] The processor executes the computer-executable instructions stored in the memory to implement an icing warning method proposed in the first aspect embodiment.

[0023] To achieve the above-mentioned purpose, the fourth aspect embodiment of the present application proposes a computer-readable storage medium, in which computer execution instructions are stored. When the computer execution instructions are executed by a processor, they are used to implement an icing warning method proposed in the above-mentioned first aspect embodiment.

[0024] To achieve the above-mentioned purpose, the fifth aspect of the present application proposes a computer program product, including a computer program, which, when executed by a processor, implements an icing warning method proposed in the first aspect of the present application.

[0025] The icing warning method, device, electronic device and storage medium provided by the present application obtain training data, wherein the training data includes historical operation data of the wind turbine and the corresponding icing state, and the historical operation data includes at least one of wind speed and power; input the feature vector and the icing state into the icing prediction model to obtain the icing prediction result, wherein the feature vector is obtained by feature extraction of the historical operation data; according to the difference between the icing prediction result and the icing state, the icing prediction model is trained to obtain the trained icing prediction model; the feature vector of the acquired real-time operation data is input into the trained icing prediction model to obtain the icing warning result. The problem of inaccurate and untimely icing warning of wind turbines is solved, and the icing risk of blades and anemometers is warned in time, which effectively reduces the icing risk during the operation of wind turbines, reduces the performance degradation and damage of wind turbines caused by icing, improves the safety and reliability of wind turbines, and promotes the further development and application of renewable energy.

[0026] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0028] Figure 1 A schematic diagram of a flow chart of an icing warning method provided in an embodiment of the present application;

[0029] Figure 2 A schematic diagram of a flow chart of another icing warning method provided in an embodiment of the present application;

[0030] Figure 3 A schematic structural diagram of an icing warning device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0031] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0032] The icing warning method and device according to the embodiments of the present application are described below with reference to the accompanying drawings.

[0033] Figure 1 A schematic flow chart of an icing warning method provided in an embodiment of the present application.

[0034] At present, there are some relevant technologies and methods for the icing problem of wind turbines, but there are the following problems:

[0035] 1) Sensor dependency: Many existing technologies require the use of additional sensors to monitor blade and anemometer icing conditions. This increases equipment and installation costs and can lead to inaccuracies and delays in data collection.

[0036] 2) Lack of effective icing warning methods: Existing technologies often cannot provide accurate blade and anemometer icing warnings, and cannot identify icing risks in a timely manner so that appropriate measures can be taken.

[0037] 3) Data analysis limitations: Some existing technologies have limitations in data analysis and cannot fully exploit the operating data of the wind turbine itself, thus limiting the ability to accurately assess and warn of icing risks.

[0038] In view of the above problems, the present application provides an icing warning method to reduce the icing risk during the operation of the wind turbine generator set. Figure 1 As shown, the icing warning method includes the following steps:

[0039] Step 101 : acquiring training data, wherein the training data includes historical operation data of the wind turbine generator set and corresponding icing status, and the historical operation data includes at least one of wind speed and power.

[0040] In the embodiment of the present application, during the historical operation of the wind turbine generator set, the historical operation data is monitored by a data monitoring system of the wind turbine generator set, and the historical operation data of the wind turbine generator set can be detected at every moment.

[0041] Among them, the wind turbine generator set can be various types of wind turbine generator sets in the wind power generation industry, including large-scale commercial wind farms, small-scale distributed wind power generation systems, etc.

[0042] It is understandable that during the historical operation process, it is possible to determine whether the blades and anemometers of the wind turbine generator set are frozen at any time, that is, it is possible to obtain the freezing state at any historical time.

[0043] Step 102, inputting the feature vector and the icing state into an icing prediction model to obtain an icing prediction result, wherein the feature vector is obtained by feature extraction of historical operation data.

[0044] In an embodiment of the present application, the historical operation data is divided into different time periods, and feature extraction is performed on the historical operation data to obtain a feature vector in any time period. The icing state in the corresponding time period is used as the corresponding label, which is input into the icing prediction model and output as the icing prediction result.

[0045] The icing prediction model may be a classifier model, such as a support vector machine (SVM) model.

[0046] Step 103: training an icing prediction model according to the difference between the icing prediction result and the icing state to obtain a trained icing prediction model.

[0047] Optionally, the icing prediction result may be the same as the label or may be different. The icing state in any time period is known, that is, the label represents the actual icing conditions in the corresponding time period. By comparing the icing prediction result with the label, the icing prediction model is corrected so that the icing prediction model outputs the correct icing state, thereby obtaining a trained icing prediction model.

[0048] Step 104 : Input the feature vector of the acquired real-time operation data into the trained icing prediction model to obtain an icing warning result.

[0049] In an embodiment of the present application, the trained icing prediction model can accurately predict the icing state of the wind turbine, obtain the real-time operation data of the wind turbine, extract feature vectors from the real-time operation data, input them into the trained icing prediction model, and output the corresponding icing state. When the output icing state is icing, an early warning message is issued so that the operation and maintenance personnel can take corresponding measures.

[0050] In this embodiment, by acquiring training data, wherein the training data includes historical operation data of the wind turbine and the corresponding icing state, the historical operation data includes at least one of wind speed and power; inputting the feature vector and the icing state into the icing prediction model to obtain the icing prediction result, wherein the feature vector is obtained by feature extraction of the historical operation data; according to the difference between the icing prediction result and the icing state, the icing prediction model is trained to obtain the trained icing prediction model; the feature vector of the acquired real-time operation data is input into the trained icing prediction model to obtain the icing warning result. The problem of inaccurate and untimely icing warning of wind turbines is solved, and the icing risk of blades and anemometers is warned in time, which effectively reduces the icing risk during the operation of wind turbines, reduces the performance degradation and damage of wind turbines caused by icing, improves the safety and reliability of wind turbines, and promotes the further development and application of renewable energy.

[0051] This embodiment provides another icing warning method. Figure 2 A schematic flow chart of another icing warning method provided in an embodiment of the present application.

[0052] like Figure 2 As shown, the icing warning method may include the following steps:

[0053] Step 201 : acquiring training data, wherein the training data includes historical operation data of the wind turbine generator set and corresponding icing status, and the historical operation data includes at least one of wind speed and power.

[0054] In an embodiment of the present application, during the historical operation of the wind turbine generator set, the data monitoring system of the wind turbine generator set monitors the operating data such as wind speed, power, and unit vibration signal at each moment, and collects historical operating data within a certain period of time and the corresponding icing status as training data.

[0055] The certain period of time may be three months, and each moment may be every second.

[0056] Optionally, the operating data may also include the rotation speed of the wind turbine generator set.

[0057] Optionally, the freezing state corresponding to the historical operating data at any moment is the label at the corresponding moment, and the label can indicate whether the blade or anemometer is frozen at the corresponding moment by a binary value, for example, 1 indicates freezing and 0 indicates non-freezing.

[0058] Furthermore, after collecting the historical operation data, the historical operation data is preprocessed, for example, data cleaning is performed on the historical operation data to eliminate the influence of noise, and the cleaned data is normalized to eliminate the influence of dimension.

[0059] Furthermore, the preprocessed training data is divided into a training set and a test set, wherein the training set is used to train the icing prediction model, and the test set is used to verify the icing prediction model.

[0060] The embodiment of the present application collects historical operation data through the data monitoring system of the wind turbine generator set, does not rely on additional sensors, and only uses the existing monitoring data of the wind turbine generator set, thereby reducing equipment installation and maintenance costs.

[0061] Step 202: traverse the historical operation data using a sliding window, wherein the historical operation data is sorted according to time series; and obtain a feature vector of the historical operation data in any sliding window.

[0062] In an embodiment of the present application, a sliding window method is used to perform time series analysis on historical operation data. The size of the sliding window is preset, and the sliding window is traversed on the historical operation data sorted according to the time series according to a preset step size. The sliding window contains a certain amount of continuous data, and the feature vector is extracted by analyzing the data in the sliding window.

[0063] The length of the sliding window may be the length of 10 minutes of historical operating data, and the step length may be the length of 1 minute of historical operating data.

[0064] The method for obtaining the feature vector may be:

[0065] The wind speed change rate, power fluctuation degree, and correlation coefficient between wind speed and power in any sliding window are obtained; and the corresponding eigenvector is determined based on the wind speed change rate, power fluctuation degree, and correlation coefficient between wind speed and power in any sliding window.

[0066] Optionally, the wind speed change rate can be determined by calculating the wind speed difference between adjacent moments. The wind speed change rate is used to describe the change of wind speed. When the wind speed changes too much, the environment in which the wind turbine generator set is located may have changed, such as a drop in temperature, an increase in humidity, etc., which may cause icing.

[0067] As a possible implementation, the wind speed difference between any two adjacent moments in the sliding window is calculated, and the average value of each wind speed difference is calculated as the wind speed change rate in the sliding window in the corresponding time period.

[0068] Optionally, the power fluctuation degree can be determined by calculating the standard deviation of the power. The power fluctuation degree is used to describe the stability of the output power of the wind turbine generator set. When the power fluctuation degree is too large, the output power of the wind turbine generator set may be affected by external factors and icing may occur.

[0069] As a possible implementation, the standard deviation of the power at each moment in the sliding window is calculated as the power fluctuation degree of the sliding window in the corresponding time period.

[0070] Optionally, the correlation coefficient between wind speed and power is used to describe the performance status of the wind turbine. Under normal circumstances, there should be a strong correlation between wind speed and power. In theory, as wind speed increases, the power output of the wind turbine will also increase, but this increase will tend to saturate after a certain wind speed. When the correlation coefficient between wind speed and power decreases, the performance of the wind turbine may be affected and icing may occur.

[0071] In the embodiment of the present application, the wind speed interval method (BIN method) is used to analyze the correlation between wind speed and power to obtain the correlation coefficient. The specific steps are as follows:

[0072] 1. Data preprocessing.

[0073] Removing outliers: Remove obvious outliers of wind speed or power based on practical experience or statistical methods.

[0074] Time filter: Select one or more time periods for analysis, such as analyzing only daytime data or excluding data when the fan is shut down.

[0075] 2.BIN value setting.

[0076] Determine the range of the BIN: For example, you can set the wind speed range to 0-25m / s.

[0077] Determine the size of BIN: According to the density of the data and specific needs, set the interval length (i.e. the value of BIN), for example, 0.5m / s, to divide the wind speed into multiple intervals.

[0078] Generate BIN: For each interval, calculate the average wind speed within the interval.

[0079] 3. Calculate the average power in each interval.

[0080] Data classification: According to the wind speed interval classification, each data point is classified into the corresponding interval.

[0081] Calculate the average: For each interval, calculate the average power within that interval.

[0082] 4. Generate wind speed-power curve.

[0083] Plot: Plot the average wind speed of the wind speed range on the horizontal axis and the corresponding average power on the vertical axis.

[0084] Under normal circumstances, the wind speed-power curve has the following shape:

[0085] Starting wind speed: Initially, when the wind speed increases, the power increases slowly. This is the start-up stage of the fan.

[0086] Before rated wind speed: As wind speed increases, power rises sharply.

[0087] After rated wind speed: After reaching a certain wind speed, the power increase will slow down and eventually reach saturation (i.e. the rated power of the fan).

[0088] 5. Calculation of correlation coefficient.

[0089] By comparing the actual wind speed-power curve with the design curve of the wind turbine, the correlation coefficient between wind speed and power is determined. For example, the similarity between the actual wind speed-power curve and the design curve of the wind turbine is calculated as the correlation coefficient between wind speed and power in the corresponding period.

[0090] If the actual wind speed-power curve of the wind turbine deviates significantly from historical data or similar models, the correlation coefficient will decrease, indicating that the wind turbine may have some degradation or damage.

[0091] Furthermore, the wind speed change rate, power fluctuation degree, and correlation coefficient between wind speed and power are standardized to obtain corresponding key features; the key features are combined to construct a feature vector.

[0092] By standardizing the wind speed change rate, power fluctuation degree, and the correlation coefficient between wind speed and power, the dimensional differences between different features are eliminated, so as to improve the generalization ability of the model in the subsequent model training process.

[0093] As a possible implementation method, the standardization method can be to subtract the corresponding feature mean from any feature, and then divide it by the corresponding standard deviation to obtain the key feature of the corresponding feature, and combine the three standardized key features. For example, the three key features are sorted in the order of wind speed change rate, power fluctuation degree, and correlation coefficient between wind speed and power to form a vector, which is the feature vector in the sliding window under the corresponding time period.

[0094] Furthermore, the speed stability can be used as a feature, and the speed stability can be obtained by calculating the standard deviation of the speed. The speed stability is used to describe the performance status of the wind turbine. When the speed stability is too large, the speed of the wind turbine may be affected by external factors, and icing may occur.

[0095] Furthermore, statistical features can be obtained from the unit vibration signal as features, such as the mean, variance and kurtosis of the unit vibration signal. The statistical features of the unit vibration signal are used to describe the performance status of the wind turbine generator set. When the statistical features of the vibration signal are too large, the wind turbine generator set may be affected by external factors and may have ice.

[0096] When the statistical characteristics of the unit vibration signal and the speed stability are used as features, the wind speed change rate, the power fluctuation degree, the correlation coefficient between wind speed and power, the statistical characteristics of the unit vibration signal, and the speed stability are standardized to obtain the corresponding key features; the key features are combined to construct a feature vector, and the feature vector at this time can be composed of five key features.

[0097] As another possible implementation, key features may be screened, for example, by using correlation analysis, mutual information and other methods to analyze the degree of correlation between each key feature and the icing risk, thereby screening out the key features most relevant to the icing risk, and the screened key features are combined to form a feature vector.

[0098] Through key feature screening, effective feature information can be extracted for the icing prediction model, thereby improving the output accuracy of the icing prediction model.

[0099] Step 203: input the feature vector and the icing state into an icing prediction model to obtain an icing prediction result.

[0100] In the embodiment of the present application, a feature vector corresponds to the sliding window in any time period, and the icing state in the corresponding time period is used as the corresponding label, which is input into the icing prediction model and output as the icing prediction result.

[0101] The icing prediction model may be a classifier model, such as a support vector machine (SVM) model.

[0102] Step 204: training an icing prediction model according to the difference between the icing prediction result and the icing state to obtain a trained icing prediction model.

[0103] Optionally, the SVM model is trained using the feature vectors in the training set and the corresponding icing state labels, and a corresponding threshold is set for any key feature. When the value of any key feature other than the correlation coefficient is greater than the corresponding threshold, the output icing prediction result is 1 (icing), otherwise the output icing prediction result is 0 (not icing); or, in the case where the correlation coefficient is used as the key feature, when the correlation coefficient is less than the corresponding threshold, the output icing prediction result is 1 (icing), otherwise the output icing prediction result is 0 (not icing). Based on the difference between the output icing prediction result and the label (icing state), the icing prediction model is trained and the corresponding threshold is adjusted until the model converges.

[0104] During the training process, the parameters of the SVM (such as kernel function, penalty coefficient, etc.) can be adjusted to improve the prediction accuracy of the model.

[0105] Furthermore, the performance of the trained SVM model is evaluated using the test set. Based on the verification results, the model can be adjusted and optimized. After the verification is completed, a trained icing prediction model is obtained.

[0106] Step 205: input the feature vector of the acquired real-time operation data into the trained icing prediction model to obtain an icing warning result.

[0107] Optionally, the trained icing prediction model can accurately predict the icing state of the wind turbine, obtain the real-time operation data of the wind turbine, extract feature vectors from the real-time operation data, input them into the trained icing prediction model, and output the corresponding icing state. When the output icing state is 1 (icing), an early warning message is issued so that the operation and maintenance personnel can take corresponding measures.

[0108] Through icing risk warning, the operation strategy of wind turbines can be adjusted in time to maintain the normal working state of wind turbines to the greatest extent and improve power generation efficiency.

[0109] Furthermore, based on the real-time operation data, the training data is updated to update the trained icing prediction model to maintain the accuracy and adaptability of the model.

[0110] For example, new wind turbine operation data can be collected regularly, the training set can be updated, and then the icing prediction model can be retrained within a certain period.

[0111] Optionally, the parameters and thresholds of the model may be adjusted according to actual application scenarios and changes in wind turbine characteristics to ensure that the early warning method has good adaptability in different scenarios.

[0112] Optionally, the embodiments of the present application can also be applied to other industrial fields that require early warning of icing risks. For example, engineering facilities such as aviation, aerospace, communication towers, bridges, etc., which are easily affected by freezing in cold environments and may cause safety hazards. By using the method of the embodiments of the present application, the operating data of the facilities can be monitored in real time, and early warning of icing risks can be issued to reduce potential safety risks.

[0113] In this embodiment, by acquiring training data, wherein the training data includes historical operation data of the wind turbine generator set and the corresponding icing state, the historical operation data includes at least one of wind speed and power; using a sliding window to traverse the historical operation data, wherein the historical operation data is sorted in time series; acquiring the feature vector of the historical operation data in any sliding window; inputting the feature vector and the icing state into the icing prediction model to obtain the icing prediction result; training the icing prediction model according to the difference between the icing prediction result and the icing state to obtain the trained icing prediction model; inputting the feature vector of the acquired real-time operation data into the trained icing prediction model to obtain the icing warning result. The problem of inaccurate and untimely icing warning of the wind turbine generator set is solved, the icing risk of the blades and anemometer is warned in time, the icing risk during the operation of the wind turbine generator set is effectively reduced, the performance degradation and damage of the wind turbine generator set caused by icing are reduced, the safety and reliability of the wind turbine generator set are improved, and the further development and application of renewable energy are promoted.

[0114] In order to implement the above embodiment, the present application also proposes an icing warning device.

[0115] Figure 3 A schematic structural diagram of an icing warning device provided in an embodiment of the present application.

[0116] like Figure 3 As shown, the icing warning device 300 includes: an acquisition module 301 , an input module 302 , a training module 303 and a warning module 304 .

[0117] An acquisition module 301 is used to acquire training data, wherein the training data includes historical operation data of the wind turbine generator set and corresponding icing status, and the historical operation data includes at least one of wind speed and power;

[0118] An input module 302 is used to input the feature vector and the icing state into an icing prediction model to obtain an icing prediction result, wherein the feature vector is obtained by extracting features from historical operation data;

[0119] A training module 303 is used to train an icing prediction model according to the difference between the icing prediction result and the icing state to obtain a trained icing prediction model;

[0120] The warning module 304 is used to input the feature vector of the acquired real-time operation data into the trained icing prediction model to obtain an icing warning result.

[0121] Furthermore, in a possible implementation of the embodiment of the present application, the input module 302 is further configured to:

[0122] Use a sliding window to traverse historical operation data, where the historical operation data is sorted according to time series;

[0123] Get the feature vector of historical running data in any sliding window.

[0124] Obtain the wind speed change rate, power fluctuation degree, and correlation coefficient between wind speed and power within any sliding window;

[0125] Based on the wind speed change rate, power fluctuation degree, and correlation coefficient between wind speed and power within any sliding window, the corresponding eigenvector is determined.

[0126] The wind speed change rate, power fluctuation degree, and the correlation coefficient between wind speed and power are standardized to obtain the corresponding key features;

[0127] Combine key features to construct feature vectors.

[0128] Furthermore, in a possible implementation of the embodiment of the present application, the acquisition module 301 is further configured to:

[0129] The training data is divided into a training set and a test set, where the training set is used to train the icing prediction model and the test set is used to verify the icing prediction model.

[0130] Clean and normalize historical operation data.

[0131] Furthermore, in a possible implementation of the embodiment of the present application, the device is also used for:

[0132] Based on the real-time operation data, the training data is updated to update the trained icing prediction model.

[0133] In this embodiment, by acquiring training data, wherein the training data includes historical operation data of the wind turbine generator set and the corresponding icing state, the historical operation data includes at least one of wind speed and power; using a sliding window to traverse the historical operation data, wherein the historical operation data is sorted in time series; acquiring the feature vector of the historical operation data in any sliding window; inputting the feature vector and the icing state into the icing prediction model to obtain the icing prediction result; training the icing prediction model according to the difference between the icing prediction result and the icing state to obtain the trained icing prediction model; inputting the feature vector of the acquired real-time operation data into the trained icing prediction model to obtain the icing warning result. The problem of inaccurate and untimely icing warning of the wind turbine generator set is solved, the icing risk of the blades and anemometer is warned in time, the icing risk during the operation of the wind turbine generator set is effectively reduced, the performance degradation and damage of the wind turbine generator set caused by icing are reduced, the safety and reliability of the wind turbine generator set are improved, and the further development and application of renewable energy are promoted.

[0134] It should be noted that the above explanations of the embodiment of the icing warning method are also applicable to the icing warning device of this embodiment, and will not be repeated here.

[0135] In order to implement the above embodiments, the present application also proposes an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method provided by the above embodiments.

[0136] In order to implement the above embodiments, the present application also proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the methods provided by the above embodiments.

[0137] In order to implement the above embodiments, the present application also proposes a computer program product, including a computer program, which implements the methods provided by the above embodiments when executed by a processor.

[0138] The collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in this application are in compliance with relevant laws and regulations and do not violate public order and good morals.

[0139] It should be noted that personal information from users should be collected for legitimate and reasonable purposes and should not be shared or sold outside of these legitimate uses. In addition, such collection / sharing should be carried out after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign the agreement / authorization including authorization of relevant user information before the user uses the function. In addition, any necessary steps should be taken to protect and safeguard access to such personal information data and ensure that others who have access to personal information data comply with its privacy policy and procedures.

[0140] The present application is expected to provide an implementation scheme for users to selectively block the use or access of personal information data. That is, the present disclosure is expected to provide hardware and / or software to prevent or block access to such personal information data. Once the personal information data is no longer needed, the risk can be minimized by limiting data collection and deleting the data. In addition, when applicable, such personal information is de-identified to protect the privacy of the user.

[0141] In the description of the aforementioned embodiments, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0142] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0143] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0144] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0145] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0146] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0147] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0148] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. An icing warning method, characterized in that: The following steps are involved: Acquire training data, wherein the training data includes historical operation data of the wind turbine generator set and corresponding icing status, and the historical operation data includes at least one of wind speed and power; Inputting the feature vector and the icing state into an icing prediction model to obtain an icing prediction result, wherein the feature vector is obtained by feature extraction of the historical operation data; According to the difference between the icing prediction result and the icing state, the icing prediction model is trained to obtain a trained icing prediction model; The feature vector of the acquired real-time operation data is input into the trained icing prediction model to obtain an icing warning result.

2. The icing warning method according to claim 1, characterized in that: The feature vector is obtained by extracting features from the historical operation data, including: Using a sliding window to traverse the historical operation data, wherein the historical operation data is sorted according to time series; Obtain a feature vector of the historical operating data in any sliding window.

3. The icing warning method according to claim 2, characterized in that: The obtaining of the feature vector of the historical operation data in any sliding window includes: Obtain the wind speed change rate, power fluctuation degree, and correlation coefficient between wind speed and power within any sliding window; Based on the wind speed change rate, the power fluctuation degree, and the correlation coefficient between wind speed and power within any one of the sliding windows, a corresponding eigenvector is determined.

4. The icing warning method according to claim 3, characterized in that: The determining of the corresponding characteristic vector based on the wind speed change rate, the power fluctuation degree, and the correlation coefficient between wind speed and power in any one of the sliding windows includes: Standardizing the wind speed change rate, the power fluctuation degree, and the correlation coefficient between the wind speed and the power to obtain corresponding key features; The key features are combined to construct the feature vector.

5. The icing warning method according to any one of claims 1 to 4, characterized in that: The method further comprises: Based on the real-time operation data, the training data is updated to update the trained icing prediction model.

6. The icing warning method according to any one of claims 1 to 4, characterized in that: After obtaining the training data, the method further includes: The training data is divided into a training set and a test set, wherein the training set is used to train the icing prediction model, and the test set is used to verify the icing prediction model.

7. The icing warning method according to any one of claims 1 to 4, characterized in that: After obtaining the training data, the method further includes: The historical operation data is cleaned and normalized.

8. An icing warning device, characterized in that: include: An acquisition module, configured to acquire training data, wherein the training data includes historical operation data of the wind turbine generator set and corresponding icing status, and the historical operation data includes at least one of wind speed and power; An input module, used for inputting a feature vector and the icing state into an icing prediction model to obtain an icing prediction result, wherein the feature vector is obtained by feature extraction of the historical operation data; A training module, used for training the icing prediction model according to the difference between the icing prediction result and the icing state, so as to obtain a trained icing prediction model; The early warning module is used to input the feature vector of the acquired real-time operation data into the trained icing prediction model to obtain an icing early warning result.

9. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 7.

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

Images