Real-time detection method, system, equipment and medium for abnormal wind power data

Abstract

The disclosed embodiments provide a method, system, device and medium for real-time detection of abnormal wind power data, which belongs to the field of data processing technology, and specifically includes: selecting target data during normal operation from historical wind power data corresponding to the wind turbine; obtaining a completely random felling tree according to the target data, and establishing and initializing a normal data structure model of the wind turbine according to the completely random felling tree; collecting data to be detected to update the normal data structure model to obtain a target detection model; inputting the data to be detected into the target detection model to obtain a detection result. Through the scheme disclosed in the present invention, data during normal operation is selected to establish a stable normal data structure model, and the structure of normal data is learned to update the model, and then the obtained target detection model is used to detect the data to be detected, thereby improving detection efficiency, operation convenience and detection accuracy.

Description

Technical Field

[0001] The embodiments of the present disclosure relate to the field of data processing technology, and in particular to a method, system, device and medium for real-time detection of abnormal wind power data. Background Art

[0002] At present, the data acquisition and monitoring control system collects data from different sensors of wind turbines (WT) in real time. The data acquisition and monitoring control data can be used for power curve modeling, wind speed and power prediction, WT control and wind farm performance evaluation. The reliability of the data directly affects the analysis and evaluation of the WT status. However, in the actual operation process, due to the influence of factors such as measurement, transmission, control and wind abandonment, the data acquisition and monitoring control system will always collect a large amount of abnormal data. If these abnormal data cannot be processed in time, it will not only affect the real-time analysis of the WT status, but also accumulate in the historical data over time. Compared with other data, wind power data is more random in time and abnormal data is more difficult to detect. Therefore, it is urgent to study related methods.

[0003] At present, the research on wind power data anomaly detection mainly focuses on historical data, which includes scattered data and time series data. Such methods have high requirements on the quality of data sets. For real-time detection of abnormal data, their data cost or detection time will be greatly increased. In the case of insufficient research on real-time detection of abnormal wind power data, some research on real-time detection of abnormal data has been conducted in other fields. The mainstream method is to judge the deviation of abnormal data based on the trend of data flow. However, when the data fluctuates greatly within a certain period, the data distribution in the window becomes dispersed, the detection performance is reduced, and when the randomness of the data becomes larger, it may be difficult to judge whether the high-dimensional data is abnormal.

[0004] It can be seen that there is an urgent need for an efficient, convenient and highly accurate real-time detection method for abnormal wind power data. Summary of the invention

[0005] In view of this, the embodiments of the present disclosure provide a method, system, device and medium for real-time detection of abnormal wind power data, which at least partially solve the problems of complex operation and poor detection efficiency and accuracy in the prior art.

[0006] In a first aspect, an embodiment of the present disclosure provides a method for real-time detection of abnormal wind power data, comprising:

[0007] Select target data during normal operation from historical wind power data corresponding to the wind turbine generator set;

[0008] Obtaining a completely random felling tree according to the target data, and establishing and initializing a normal data structure model of the wind turbine generator set according to the completely random felling tree;

[0009] Collecting the data to be detected to update the normal data structure model to obtain a target detection model;

[0010] The data to be detected is input into the target detection model to obtain a detection result.

[0011] According to a specific implementation of the embodiment of the present disclosure, the step of selecting target data during normal operation from historical wind power data corresponding to the wind turbine generator set includes:

[0012] The characteristic values ​​of a plurality of characteristics in the historical wind power data at corresponding moments are selected to form the target data.

[0013] According to a specific implementation of the embodiment of the present disclosure, the step of obtaining a completely random felling tree according to the target data, and establishing and initializing a normal data structure model of the wind turbine according to the completely random felling tree includes:

[0014] Complete segmentation of all the target data to obtain the completely random felled tree;

[0015] The left node in the completely random cut tree is set to 0, and the right node is set to 1, to obtain the position vectors of all nodes in the completely random cut tree;

[0016] Calculate the depth of each layer of nodes in the completely random felling tree;

[0017] Calculate the index of each node according to the position vector and depth corresponding to each node;

[0018] The normal data structure model is established and initialized according to all the indexes.

[0019] According to a specific implementation of the embodiment of the present disclosure, the step of collecting the data to be detected, updating the normal data structure model, and obtaining the target detection model includes:

[0020] Selecting a brother node corresponding to the data to be detected from the normal data structure model;

[0021] Calculating the feature difference between the data to be detected and the brother node;

[0022] The normal data structure model is updated according to the feature difference to obtain the target detection model.

[0023] According to a specific implementation of the embodiment of the present disclosure, the step of inputting the data to be detected into the target detection model to obtain the detection result includes:

[0024] Sampling the data to be detected and inserting the sampled data into the normal data structure model and calculating the complexity change of the target detection model;

[0025] Whether all the data in the data to be detected are abnormal is determined according to the complexity change and the change threshold, and the detection result is output.

[0026] According to a specific implementation of the embodiment of the present disclosure, after the step of inputting the data to be detected into the target detection model to obtain the detection result, the method further includes:

[0027] Eliminate abnormal data in the data to be detected according to the detection result;

[0028] The characteristic difference between the data in the normal data structure model that is most similar to the remaining data is calculated, and it is determined whether to keep the remaining data in the normal data structure model according to the characteristic difference and the characteristic threshold.

[0029] In a second aspect, an embodiment of the present disclosure provides a real-time detection system for abnormal wind power data, including:

[0030] A selection module is used to select target data during normal operation from historical wind power data corresponding to the wind turbine;

[0031] An establishment module is used to obtain a completely random felling tree according to the target data, and to establish and initialize a normal data structure model of the wind turbine according to the completely random felling tree;

[0032] An acquisition module, used for acquiring the data to be detected, updating the normal data structure model, and obtaining a target detection model;

[0033] The detection module is used to input the data to be detected into the target detection model to obtain a detection result.

[0034] In a third aspect, an embodiment of the present disclosure further provides an electronic device, the electronic device comprising:

[0035] at least one processor; and,

[0036] a memory communicatively connected to the at least one processor; wherein,

[0037] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the real-time detection method for abnormal wind power data in the aforementioned first aspect or any implementation of the first aspect.

[0038] In a fourth aspect, an embodiment of the present disclosure further provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the computer to execute the real-time detection method for abnormal wind power data in the aforementioned first aspect or any implementation method of the first aspect.

[0039] In a fifth aspect, an embodiment of the present disclosure further provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the real-time detection method for abnormal wind power data in the aforementioned first aspect or any implementation method of the first aspect.

[0040] The real-time detection scheme for abnormal wind power data in the disclosed embodiment includes: selecting target data during normal operation from historical wind power data corresponding to the wind turbine; obtaining a completely random felling tree according to the target data, and establishing and initializing a normal data structure model of the wind turbine according to the completely random felling tree; collecting data to be detected to update the normal data structure model to obtain a target detection model; and inputting the data to be detected into the target detection model to obtain a detection result.

[0041] The beneficial effects of the embodiments of the present disclosure are as follows: through the scheme of the present disclosure, data during normal operation is selected to establish a stable normal data structure model, and the structure of the normal data is learned to update the model, and then the obtained target detection model is used to detect the data to be detected, thereby improving detection efficiency, operational convenience and detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0043] Figure 1 A schematic diagram of a flow chart of a method for real-time detection of abnormal wind power data provided by an embodiment of the present disclosure;

[0044] Figure 2 A schematic diagram of a real-time detection framework of a method for real-time detection of abnormal wind power data provided by an embodiment of the present disclosure;

[0045] Figure 3 A structural schematic diagram of a real-time detection system for abnormal wind power data provided by an embodiment of the present disclosure;

[0046] Figure 4 A schematic diagram of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0047] The embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0048] The following describes the embodiments of the present disclosure through specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present disclosure.

[0049] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein may be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present disclosure, it should be understood by those skilled in the art that an aspect described herein may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number of aspects described herein may be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein may be used to implement this device and / or practice this method.

[0050] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present disclosure. The drawings only show components related to the present disclosure rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0051] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects described may be practiced without these specific details.

[0052] The embodiment of the present disclosure provides a method for real-time detection of abnormal wind power data, which can be applied to the real-time detection process of abnormal wind power data in a wind power generation scenario.

[0053] See also Figure 1 , is a flow chart of a method for real-time detection of abnormal wind power data provided by an embodiment of the present disclosure. Figure 1 As shown, the method mainly comprises the following steps:

[0054] S101, selecting target data during normal operation from historical wind power data corresponding to the wind turbine generator set;

[0055] Optionally, in step S101, selecting target data during normal operation from historical wind power data corresponding to the wind turbine generator set includes:

[0056] The characteristic values ​​of a plurality of characteristics in the historical wind power data at corresponding moments are selected to form the target data.

[0057] In the specific implementation, it is considered that during the normal operation of the wind turbine, the data generated are mainly concentrated in a relatively stable area. For wind power data, they are mainly distributed near the power curve. Therefore, collecting stable wind power data to establish a normal data structure model can not only identify abnormal data based on the distribution characteristics of normal data, but also reduce the impact of real-time data fluctuations on detection. Then, the target data during normal operation can be selected from the historical wind power data corresponding to the wind turbine.

[0058] For example, select a historical data set D from a data acquisition and monitoring control system, where D = {d i}, i = 1, ..., t, d i is the sampling data of the wind turbine at time i. i =(d i1 ,…,d in ), d ij is the feature value of the jth feature of WT at time i. Then, a normal data set X is selected according to formulas (1)-(3).

[0059]

[0060] D1={d i |0.95p r <d ip <1.05p r} (2)

[0061] D2={d i |v curve -Δv<d iv <v curve+Δv} (3)

[0062] Where w is the number of normal data in X. p and d v are power and wind speed respectively. ip and d iv are the power and wind speed at time i respectively. r and v r are rated power and wind speed respectively. curve is the wind speed corresponding to the power curve. Δv is the range of normal data wind speed.

[0063] S102, obtaining a completely random felling tree according to the target data, and establishing and initializing a normal data structure model of the wind turbine generator according to the completely random felling tree;

[0064] In a specific implementation, after obtaining the target data, if the target data is stable normal data, a completely random felling tree can be obtained according to the target data, and a normal data structure model of the wind turbine generator set can be established and initialized according to the completely random felling tree.

[0065] S103, collecting the data to be detected and updating the normal data structure model to obtain a target detection model;

[0066] During specific implementation, taking into account that the normal data structure model is established based on historical data, the model has a strong dependence on the initial data and is not sufficiently extensible in the data stream, the data to be detected can be collected to update the normal data structure model, to obtain a target detection model, and to learn the structure of normal wind power data in the data stream, thereby reducing dependence on the initial normal data set and enhancing the extensibility of the model in the data stream.

[0067] S104, inputting the data to be detected into the target detection model to obtain a detection result.

[0068] After obtaining the target detection model, the data to be detected can be input into the target detection model, and the target detection model identifies whether the state of each data in the data to be detected is normal, thereby obtaining the detection result. The specific framework diagram of the real-time detection method for abnormal wind power data is as follows: Figure 2 shown.

[0069] The real-time detection method for abnormal wind power data provided in this embodiment selects data during normal operation to establish a stable normal data structure model, learns the structure of normal data to update the model, and then uses the obtained target detection model to detect the data to be detected, thereby improving detection efficiency, operational convenience and detection accuracy.

[0070] On the basis of the above embodiment, the step of obtaining a completely random felling tree according to the target data, and establishing and initializing a normal data structure model of the wind turbine according to the completely random felling tree includes:

[0071] Complete segmentation of all the target data to obtain the completely random felled tree;

[0072] The left node in the completely random cut tree is set to 0, and the right node is set to 1, to obtain the position vectors of all nodes in the completely random cut tree;

[0073] Calculate the depth of each layer of nodes in the completely random felling tree;

[0074] Calculate the index of each node according to the position vector and depth corresponding to each node;

[0075] The normal data structure model is established and initialized according to all the indexes.

[0076] For example, the steps of establishing and initializing the normal data structure model may be as follows:

[0077] Step 1: According to The probability of randomly selecting a feature in the target data X, where l i =max x∈S X i -min x∈S X i is the difference between the maximum and minimum eigenvalues ​​of the i-th feature.

[0078] Step 2. Select X i,ave =(minX i +maxX i ) / 2 is used as the boundary for splitting the data.

[0079] Step 3. X left ={X|X i ≤X i,ave},X right =X\X left .

[0080] Step 4. In X left and X right Steps 1-3 are recursively executed until all data are separated and a Complete Random Cut Tree (CRCT) is obtained.

[0081] Step 5: In the completely random felling tree, the left child node of each node is set to 0, and the right child node is set to 1, and the position vector f(x i ).

[0082] Step 6: Set the depth of each node in the completely random felling tree. i , the depth of the completely random felling tree increases by 1 for each additional layer of nodes.

[0083] Step 7: Calculate the index of each node based on its position vector and its depth in the completely random cut tree.

[0084] Step 8: The normal data structure model (NDSM) is obtained as the completely random felled tree with node indexes.

[0085] Among them, step 1 considers the influence of different feature magnitudes and selects different features by non-uniform probability. In order to avoid the branches leaning to one side due to uneven data distribution in the binary tree, the mean of the selected features is used as the boundary for segmenting the data in step 2. In step 4, in order to avoid the accumulation of data at the same node causing mutual occlusion between data, steps 1-3 are repeated until all data are completely divided. In step 7, the node index is used to locate each data point in the normal data structure model.

[0086] On the basis of the above embodiment, the step of collecting the data to be detected, updating the normal data structure model, and obtaining the target detection model includes:

[0087] Selecting a brother node corresponding to the data to be detected from the normal data structure model;

[0088] Calculating the feature difference between the data to be detected and the brother node;

[0089] The normal data structure model is updated according to the feature difference to obtain the target detection model.

[0090] In specific implementation, considering that the quality of the model depends on the selection of normal data, selecting more comprehensive data when building the model will inevitably increase the data cost. A model update strategy can be set to enable the normal data structure model to learn new normal data structures over time and reduce dependence on initial data selection.

[0091] The specific steps are as follows:

[0092] Step 1: If x t If it is abnormal data, its index is returned and it is deleted from NDSM according to the index.

[0093] Step 2: If x t If it is normal data, then find x in NDSM tThe sibling node of .

[0094] Step 3: Traverse x t and x q All features of , and calculate ΔD i =|x t，i -x q，i |, where x t，i and x q，i They are x t and x q The i-th eigenvalue of .

[0095] Step 4: If Then x t Retained, d i is the minimum resolution of the ith feature of WT, and the maximum amount of data that can be expanded does not exceed

[0096] Step 5: When the amount of data in NDSM reaches the maximum value and the newly collected data x t If the condition in (4) is met, the data with the largest depth in NDSM will be deleted.

[0097] Step 6: If x t If the condition in step 4 is not met, delete x t .

[0098] Step 7: Return the updated normal data structure model. The key is to evaluate the similarity of the data (steps 2-4). In a binary tree, if two data x t and x q If the difference between the two data is less than a certain value, there is a high probability that the two data are under the same parent node. Therefore, we can find the two most similar data in a data set based on whether the two data are under the same parent node. In terms of data distribution, the impact of these two data on other data is basically the same, and one node can replace another node. When the difference between the i-th feature of two sibling nodes is less than a certain value, x is replaced. t and x q One of them is judged as redundant data.

[0099] Furthermore, the step of inputting the data to be detected into the target detection model to obtain the detection result includes:

[0100] Sampling the data to be detected and inserting the sampled data into the normal data structure model and calculating the complexity change of the target detection model;

[0101] Whether all the data in the data to be detected are abnormal is determined according to the complexity change and the change threshold, and the detection result is output.

[0102] In the specific implementation, considering the model update, the data in the target detection model will change over time. In order to better adapt to this dynamic change, a dynamic abnormal data real-time detection process is proposed.

[0103] First, the current moment data x t The samples are taken and inserted into the normal data structure model.

[0104] Then, insert x t The position vector of the node in the target detection model changes when t=t. In order to calculate the change of the position vector, we define the model complexity DISP at time t: t , which is used to quantify the sum of the position vectors of all nodes in the target detection model, is calculated as follows:

[0105]

[0106] Among them, W t is the amount of data in the target detection model at time t.

[0107] According to DISP t-1 and DISP t , insert x t The position vector change caused by this is the complexity change of the model ΔDISP t We can get:

[0108] ΔDISP t =DISP-DISP t-1 (5)

[0109] Finally, x is determined based on the distribution of normal data in the target detection model at time t-1. t Is it abnormal? The judgment conditions are given in (6).

[0110]

[0111] Q t =Q t-1,mean +λQ t-1,std (7)

[0112]

[0113]

[0114] Among them, Q t is the threshold value obtained by the normal data structure model at time t-1. λ is the weight coefficient, usually between 0 and 2. ΔDISP t-1，i is the change in model complexity caused by deleting the i-th node at time t-1 in the target detection model. t-1，mean and Qt-1，std is ΔDISP t-1，i The mean and standard deviation of .

[0115] It should be noted that the choice of λ will have an impact on the model update. As λ increases, more data will be retained in the normal data structure model, and the detection time will also increase.

[0116] Optionally, after the step of inputting the data to be detected into the target detection model to obtain the detection result, the method further includes:

[0117] Eliminate abnormal data in the data to be detected according to the detection result;

[0118] The characteristic difference between the data in the normal data structure model that is most similar to the remaining data is calculated, and it is determined whether to keep the remaining data in the normal data structure model according to the characteristic difference and the characteristic threshold.

[0119] In specific implementation, after obtaining the detection result, abnormal data in the data to be detected can be eliminated according to the detection result, and then the feature difference of the data most similar to the remaining data in the normal data structure model is calculated, and it is determined whether to retain the remaining data in the normal data structure model based on the feature difference and the feature threshold.

[0120] The overall process of the embodiment of the present disclosure is specifically as follows: first, a normal data set X that can reflect the overall data structure of the wind turbine is obtained in the data acquisition and monitoring control system, and a normal data structure model is initialized with X.

[0121] Secondly, the sample data x t Insert the normal data structure model and calculate the insertion x t Model complexity DISP of normal data structure model before and after t-1 and DISP t , obtained by x t The complexity change caused by ΔDISP t Then according to ΔDISP and threshold Q t Judge x t Is it abnormal?

[0122] Finally, for abnormal data, delete it from its normal data structure model; for normal data, calculate x t In the normal data structure model, t The difference D between the i-th feature of the most similar data nearest Then according to D nearest and d i Judge x t Whether to keep in normal data structure model.

[0123] In order to further illustrate the real-time detection method for abnormal wind power data in the embodiment of the present disclosure, the detection performance can be evaluated by selecting precision, recall, and F1 as evaluation indicators. These indicators are defined as follows:

[0124] precision=TP / (TP+FP) (10)

[0125] recall=TP / (TP+FN) (11)

[0126] F1=(2×precision×recall) / (precision+recall) (12)

[0127] Among them, TP, FP, FN and TN are the number of detected true positive, false positive, false negative and true negative data respectively.

[0128] In addition, in order to evaluate the degree of discreteness between the power curve and the data in the normal data structure model, an additional evaluation indicator, the root mean square error (RMSE), can be used. The calculation formula is:

[0129]

[0130] Among them, RMSE i is the RMSE of the i-th wind speed interval (the selection of wind speed interval refers to the standard power curve provided by the manufacturer). N i is the amount of data in the ith interval. i is the power on the power curve of the ith interval. i，j is the j-th power data in the i-th interval.

[0131] It should be noted that the smaller the RMSE, the lower the data dispersion and the more stable the data structure.

[0132] Corresponding to the above method embodiment, see Figure 3 The embodiment of the present disclosure also provides a real-time detection system 30 for abnormal wind power data, including:

[0133] A selection module 301 is used to select target data during normal operation from historical wind power data corresponding to the wind turbine generator set;

[0134] Establishing module 302, for obtaining a completely random felling tree according to the target data, and establishing and initializing a normal data structure model of the wind turbine according to the completely random felling tree;

[0135] The acquisition module 303 is used to acquire the data to be detected and update the normal data structure model to obtain a target detection model;

[0136] The detection module 304 is used to input the data to be detected into the target detection model to obtain a detection result.

[0137] Figure 3 The system shown can correspondingly execute the contents of the above method embodiment. For the parts not described in detail in this embodiment, refer to the contents recorded in the above method embodiment and will not be repeated here.

[0138] See also Figure 4 The embodiment of the present disclosure further provides an electronic device 40, which includes: at least one processor and a memory connected to the at least one processor. The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the real-time detection method for abnormal wind power data in the aforementioned method embodiment.

[0139] The embodiment of the present disclosure further provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the computer to execute the real-time detection method for abnormal wind power data in the aforementioned method embodiment.

[0140] The embodiment of the present disclosure also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the real-time detection method for abnormal wind power data in the aforementioned method embodiment.

[0141] Reference below Figure 4 , which shows a schematic diagram of the structure of an electronic device 40 suitable for implementing the embodiment of the present disclosure. The electronic device in the embodiment of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0142] like Figure 4As shown, the electronic device 40 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 401, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 402 or a program loaded from a storage device 408 to a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the electronic device 40 are also stored. The processing device 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0143] Typically, the following devices may be connected to the I / O interface 405: input devices 406 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 407 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 408 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 409. The communication device 409 may allow the electronic device 40 to communicate wirelessly or wired with other devices to exchange data. Although the electronic device 40 with various devices is shown in the figure, it should be understood that it is not required to implement or have all the devices shown. More or fewer devices may be implemented or have instead.

[0144] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 409, or installed from the storage device 408, or installed from the ROM 402. When the computer program is executed by the processing device 401, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0145] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0146] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0147] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device can execute the relevant steps of the method embodiment.

[0148] Alternatively, the computer-readable medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device can perform the relevant steps of the method embodiment.

[0149] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0150] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0151] The units involved in the embodiments described in the present disclosure may be implemented by software or by hardware.

[0152] It should be understood that various parts of the present disclosure may be implemented in hardware, software, firmware, or a combination thereof.

[0153] The above is only a specific implementation of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present disclosure should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A real-time detection method for abnormal wind power data, characterized in that: include: Select target data during normal operation from historical wind power data corresponding to the wind turbine generator set; Obtaining a completely random felling tree according to the target data, and establishing and initializing a normal data structure model of the wind turbine generator set according to the completely random felling tree; The step of obtaining a completely random felling tree according to the target data, and establishing and initializing a normal data structure model of the wind turbine generator according to the completely random felling tree comprises: Complete segmentation of all the target data to obtain the completely random felled tree; The left node in the completely random cut tree is set to 0, and the right node is set to 1, to obtain the position vectors of all nodes in the completely random cut tree; Calculate the depth of each layer of nodes in the completely random felling tree; Calculate the index of each node according to the position vector and depth corresponding to each node; Establish and initialize the normal data structure model according to all the indexes; Collecting the data to be detected to update the normal data structure model to obtain a target detection model; The step of collecting the data to be detected and updating the normal data structure model to obtain the target detection model includes: Selecting a brother node corresponding to the data to be detected from the normal data structure model; Calculating the feature difference between the data to be detected and the brother node; Update the normal data structure model according to the feature difference to obtain the target detection model; Inputting the data to be detected into the target detection model to obtain a detection result; The step of inputting the data to be detected into the target detection model to obtain a detection result includes: Sampling the data to be detected and inserting the sampled data into the normal data structure model and calculating the complexity change of the target detection model; Whether all the data in the data to be detected are abnormal is determined according to the complexity change and the change threshold, and the detection result is output.

2. The method according to claim 1, characterized in that The step of selecting target data during normal operation from the historical wind power data corresponding to the wind turbine generator set includes: The characteristic values ​​of a plurality of characteristics in the historical wind power data at corresponding moments are selected to form the target data.

3. The method according to claim 1, characterized in that After the step of inputting the data to be detected into the target detection model to obtain the detection result, the method further includes: Eliminate abnormal data in the data to be detected according to the detection result; The characteristic difference between the data in the normal data structure model that is most similar to the remaining data is calculated, and it is determined whether to keep the remaining data in the normal data structure model according to the characteristic difference and the characteristic threshold.

4. A real-time detection system for abnormal wind power data, characterized in that: include: A selection module is used to select target data during normal operation from historical wind power data corresponding to the wind turbine; An establishment module is used to obtain a completely random felling tree according to the target data, and to establish and initialize a normal data structure model of the wind turbine according to the completely random felling tree; The specific process of establishing the module includes: Complete segmentation of all the target data to obtain the completely random felled tree; The left node in the completely random cut tree is set to 0, and the right node is set to 1, to obtain the position vectors of all nodes in the completely random cut tree; Calculate the depth of each layer of nodes in the completely random felling tree; Calculate the index of each node according to the position vector and depth corresponding to each node; Establish and initialize the normal data structure model according to all the indexes; An acquisition module, used for acquiring the data to be detected, updating the normal data structure model, and obtaining a target detection model; The specific process of the acquisition module includes: Selecting a brother node corresponding to the data to be detected from the normal data structure model; Calculating the feature difference between the data to be detected and the brother node; Update the normal data structure model according to the feature difference to obtain the target detection model; A detection module, used to input the data to be detected into the target detection model to obtain a detection result; The specific process of the detection module includes: Sampling the data to be detected and inserting the sampled data into the normal data structure model and calculating the complexity change of the target detection model; Whether all the data in the data to be detected are abnormal is determined according to the complexity change and the change threshold, and the detection result is output.

5. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the real-time detection method for abnormal wind power data according to any one of claims 1 to 3.

6. A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable the computer to execute the method for real-time detection of abnormal wind power data according to any one of claims 1 to 3.

Images