Fault detection method for gearbox of wind generating set and related device
The observation data of wind turbine gearboxes are screened and preprocessed by the IPSO-LSTM network model, which solves the problem of difficulty in capturing the fault characteristics of wind turbine gearboxes, realizes accurate prediction of wind turbine gearbox faults, and improves the reliability and accuracy of fault detection.
Patent Information
- Application Number
- CN202510893304.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-30
- Publication Date
- 2025-10-17
AI Technical Summary
Existing technologies find it difficult to accurately capture the subtle fault characteristics of wind turbine gearboxes in the initial fault stage, resulting in difficulties and uncertainty in fault prediction, affecting the operating efficiency and power generation of the unit, and data-driven methods lack universal applicability to wind turbines.
The IPSO-LSTM network model is used to screen and preprocess the observation data of the wind turbine gearbox. The feature data is screened by combining the maximum information coefficient. The IPSO-LSTM network model is then used to perform fault diagnosis. The grid, environment, and equipment status data are used to improve the prediction accuracy.
It achieves accurate prediction of wind turbine gearbox failures, reduces unit downtime and losses, and improves the reliability and accuracy of fault detection.
Smart Images
Figure CN120804973A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application belongs to the technical field of wind power generation, and relates to a wind turbine gearbox fault detection method and a related device. BACKGROUND
[0002] The energy structure in China is currently dominated by coal, and the transformation of new energy generation is the only way to achieve "zero-carbon power". China is rich in wind power resources, and under the background of carbon neutrality, wind power will become the focus of future new energy development.
[0003] A wind turbine is a complex electromechanical system composed of a rotor, a bearing, a main shaft, a gearbox, a generator, a tower, an yawing system, a sensor, etc. The transmission system includes key components such as the gearbox and the main bearing. Although the failure rate of the transmission system of the wind turbine is not the highest, it is the longest downtime of the unit, causing much greater losses. Fault repair requires a lot of manpower and resources, and will affect the power generation and operating efficiency of the wind turbine. The wind turbine transmission system in the initial failure stage shows weak fault characteristic components that are not easy to capture, which increases the difficulty and uncertainty of fault prediction. Therefore, an accurate fault detection method is particularly important, as it can accurately capture fault characteristic components, determine the fault state of the transmission system, establish a transmission system fault warning mechanism, and avoid major failures.
[0004] Fault detection mainly includes methods based on physical theory and data-driven methods. However, wind turbines are very complex, and it is difficult to derive a mathematical model, which is not universal. With the improvement of the SCADA system and the growth of data transmission, storage, and computing power, data-driven methods are applied to fault diagnosis of wind turbines. Yang et al. focused on the vibration frequency of the fault signal, extracted the fault signal features of the equipment based on spectral kurtosis, and monitored and diagnosed the faults of the wind turbine gearbox. Cambron et al. established a temperature difference index based on real-time operating data collected by the SCADA monitoring system platform to achieve condition monitoring of key equipment. Wang et al. proposed a deep extreme learning machine method based on deep convolutional neural networks and whale optimization algorithm optimization, which performed well in rotating machinery fault diagnosis. SUMMARY
[0005] The purpose of the present application is to overcome the shortcomings of the prior art and provide a wind turbine gearbox fault detection method and related device that can accurately predict wind turbine gearbox faults.
[0006] To achieve the above purpose, the present application discloses a wind turbine gearbox fault detection method, comprising:
[0007] Obtain observation data of wind turbine gearbox;
[0008] Performing information screening on the observation data of the wind turbine gearbox to obtain screened information;
[0009] The screened information is input into the trained IPSO-LSTM network model to determine whether a fault occurs in the wind turbine gearbox.
[0010] The wind turbine gearbox fault detection method of the present invention is further improved in that:
[0011] Furthermore, the observation data of the gearbox of the wind turbine generator set includes power grid data, environmental data and equipment status data, wherein the power grid data includes the active power, reactive power, power grid frequency, line voltage and line current of the power grid; the environmental data includes wind speed, wind direction angle, air pressure and temperature; the equipment status data includes gearbox oil temperature, gearbox front and rear axle temperature, gearbox oil pressure and main hydraulic pressure.
[0012] Furthermore, the process of filtering the observation data of the wind turbine gearbox to obtain the filtered information is as follows:
[0013] The maximum information coefficient is used to screen the observation data of the wind turbine gearbox to obtain target data and characteristic data.
[0014] Furthermore, before the information screening of the observation data of the wind turbine gearbox is performed, the following steps are further included:
[0015] The observed data of the wind turbine gearbox are used to remove abnormal values and supplement missing values.
[0016] Furthermore, it also includes:
[0017] The IPSO-LSTM network model is evaluated by the mean absolute error, root mean square error, and symmetric mean absolute percentage error to determine the prediction performance of the IPSO-LSTM network model. When the prediction performance of the IPSO-LSTM network model is insufficient, the IPSO-LSTM network model is retrained.
[0018] Furthermore, it also includes:
[0019] When the prediction performance of the IPSO-LSTM network model is insufficient, a warning signal or an alarm signal is issued.
[0020] The present invention discloses a wind turbine gearbox fault detection system, comprising:
[0021] An acquisition module is configured to acquire observation data of a wind turbine gearbox.
[0022] A screening module is configured to perform information screening on the observation data of the wind turbine gearbox to obtain screened information.
[0023] A judgment module is configured to input the screened information into the trained IPSO-LSTM network model to determine whether the wind turbine gearbox has a fault.
[0024] The wind turbine gearbox fault detection system has the following further improvements:
[0025] Further, the observation data of the wind turbine gearbox includes power grid data, environmental data and equipment state data, wherein the power grid data includes active power, reactive power, power grid frequency, line voltage and line current of the power grid, the environmental data includes wind speed, wind direction angle, air pressure and temperature, and the equipment state data includes gear box oil temperature, gear box front and rear shaft temperature, gear box oil pressure and main hydraulic pressure.
[0026] The application discloses a computer device, which comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the wind turbine gearbox fault detection method when executing the computer program.
[0027] The application discloses a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the wind turbine gearbox fault detection method when executed by a processor.
[0028] The application has the following advantages:
[0029] The wind turbine gearbox fault detection method and related device of the application input the screened information into the trained IPSO-LSTM network model to determine whether the wind turbine gearbox has a fault, optimize the hidden layer parameters in the LSTM by using IPSO, make the hidden layer parameters more suitable for the prediction of such data, and improve the prediction accuracy.
[0030] Further, MIC is selected for information screening to avoid the problem of large workload of grid division caused by large amount of data.
[0031] Further, in terms of data acquisition, power grid data, environmental data and equipment state data are selected to improve the reliability and accuracy of the judgment, considering the multi-source nature of the data. BRIEF DESCRIPTION OF DRAWINGS
[0032] The accompanying drawings, which form a part of this specification, are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification. The drawings illustrate exemplary embodiments of the application and, together with the description, serve to explain the application. In the drawings:
[0033] Figure 1 is a flow chart of the method of the present application;
[0034] Figure 2 is a flow chart of the IPSO-LSTM model. DETAILED DESCRIPTION
[0035] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.
[0036] In the description of the present application, it should be understood that the terms "comprising" and "including" indicate the presence of described features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or sets thereof.
[0037] It should also be understood that the terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification and the appended claims of the present application, unless otherwise clearly indicated by the context, the singular forms "a", "an" and "the" are intended to include the plural forms.
[0038] It should be further understood that the term "and / or" used in the specification and the appended claims of the present application means any combination of one or more of the associated listed items and all possible combinations thereof, and includes these combinations, for example, A and / or B can represent three cases of A alone, A and B together, and B alone. In addition, the character " / " in the present application generally represents an "or" relationship between the front and rear associated objects.
[0039] It should be understood that although the terms first, second, third, etc. may be used in the embodiments of the present application to describe the preset ranges, etc., these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from each other. For example, the first preset range can also be referred to as the second preset range, and similarly, the second preset range can also be referred to as the first preset range without departing from the scope of the embodiments of the present application.
[0040] Depending on the context, the word "if" as used herein can be interpreted to mean "when" or "while" or "in response to determining" or "in response to detecting." Similarly, the phrase "if it is determined" or "if [a stated condition or event] is detected" can be interpreted to mean "upon determining" or "in response to determining" or "upon detecting [the stated condition or event]" or "in response to detecting [the stated condition or event]."
[0041] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application but not all the embodiments. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in the present application fall within the scope of protection of the present application.
[0042] Various structural schematic diagrams according to the disclosed embodiments of the present application are shown in the drawings. These diagrams are not drawn to scale, in which some details are exaggerated for the purpose of clarity and some details can be omitted. The shapes of various regions, layers and the relative size and position relationship therebetween shown in the drawings are only exemplary, and in actuality, there can be deviations due to manufacturing tolerances or technical limitations, and regions / layers with different shapes, sizes and relative positions can be additionally designed by those skilled in the art according to actual needs.
[0043] Embodiment one
[0044] With reference to Figure 1 and Figure 2 , the wind turbine gearbox fault detection method described in the present application comprises the following steps:
[0045] 1) Selecting part of the data monitored by the SCADA system before and after the gearbox fails, including power grid data, environmental data and equipment state data, wherein the power grid data includes active power, reactive power, power grid frequency, line voltage and line current, the environmental data includes wind speed, wind direction angle, air pressure and temperature, and the equipment state data includes gearbox oil temperature, gearbox front and rear shaft temperature, gearbox oil pressure and main hydraulic pressure;
[0046] 2) Preprocessing the collected data, i.e. eliminating outliers and supplementing missing values;
[0047] 2.1) Remove the wind speed data that is not between the cut-in wind speed and the cut-out wind speed of the wind turbine;
[0048] 2.2) Remove the data with power less than zero;
[0049] 2.3) Remove the data in the shutdown state;
[0050] 2.4) Fill in the missing values using linear interpolation. For the yth missing data, the method of filling in the missing data using linear interpolation is:
[0051]
[0052] where m < y < m + l, m is the relative position of the missing data, and l is the time interval of the missing data.
[0053] 3) Further verify the cleaned data using probability density distribution;
[0054] 4) Divide the data into training data and test data, where the training data is the data when the device is in a healthy state, and the test data is the data when the device is in any state;
[0055] 5) The change of wind power is closely related to the performance of the generator, which can reflect the health status and failure development trend of the wind turbine, so the active power, i.e. wind power, is selected as the target data;
[0056] 6) For the training data, the maximal information coefficient (MIC) is used to filter the information of the target data and other data, and the MIC is used to measure the correlation between two feature variables. The specific process of step 6) is:
[0057] 61) Divide the data into grids, where the method of determining the optimal grid division by simplifying part of the grid division is used, and the specific process is:
[0058] 611) Fix the division of the X-axis and perform different divisions of the Y-axis. In all possible grid divisions, the division with the largest mutual information is the optimal division of the Y-axis;
[0059] 612) Fix the division of the Y-axis and perform different divisions of the X-axis. In all possible grid divisions, the division with the largest mutual information is the optimal division of the X-axis;
[0060] 62) Calculate the mutual information value under different grid resolutions, and the calculation process of the mutual information value is:
[0061] 621) Let the variable X be discrete, and take values X1, X2, …, Xn. Let the variable Y be discrete, and take values Y1, Y2, …, Ym. Let the joint distribution of X and Y be P(X,Y). The mutual information value is calculated as: nThe information entropy H(X) is:
[0062]
[0063] where P(X1), P(X2), …, P(Xn) are the probability distribution of X. n
[0064] 622) Let the joint probability of two discrete random variables X and Y be P(X, Y), and the joint entropy be:
[0065]
[0066] 623) For discrete random variables X and Y, the conditional entropy is:
[0067]
[0068] where H(Y|X) is the remaining unknown information when X is known.
[0069] The relationship between joint entropy and conditional entropy is:
[0070] H(X, Y) = H(X) + H(Y|X) = H(Y) + H(X|Y)
[0071] The mutual information I(X, Y) between X and Y is:
[0072]
[0073] 63) The normalized mutual information value is standardized, and the standardization formula is:
[0074]
[0075] where I(X; Y) st is the normalized mutual information value, and |X| and |Y| are the grid numbers of X and Y, respectively.
[0076] 64) By the optimal X and Y axis partition, the feature matrix composed of the mutual information coefficient is obtained, and the largest element AMIC is selected from it.
[0077] 7) Take the target data and the data obtained after screening as input, establish the IPSO-LSTM model, and the flow chart of the IPSO-LSTM network model is as Figure 2 shown, and the specific steps are:
[0078] 71) Convert the input data into matrix form and initialize the IPSO algorithm parameters.
[0079] 72) Define fitness; use the mean square error of the predicted value of the LSTM network as the particle fitness value fit.
[0080]
[0081] 73) Take the position information of the particles as the parameters of the LSTM network, and construct multiple LSTM networks.
[0082] 74) Train all the networks to obtain the adaptive value of each particle, update the individual extreme value and the group extreme value.
[0083] 75) Update the particle speed and position information with nonlinear inertia weight iteration according to the individual extreme value and the group extreme value.
[0084] 76) If the condition is met or the maximum number of iterations is reached, go to step 77), otherwise go to step 73).
[0085] 77) Obtain the optimized parameters, increase the number of iterations to 100, and retrain the LSTM network.
[0086] 78) Predict through the trained IPSO-LSTM network.
[0087] 8) Select the mean absolute error, root mean square error and symmetric average absolute percentage error as the three main error evaluation indexes to evaluate the prediction model, and each evaluation index is:
[0088]
[0089] 9) Calculate the root mean square error between the predicted value and the true value to judge the size of the residual error;
[0090] 10) Use EWMA to set the warning threshold and alarm threshold to judge the trend of the residual error, so as to judge the occurrence and development of the fault. EWMA is a statistical method that can smooth the data set, and its calculation amount will decrease over time. In addition, EWMA is also sensitive to small changes. The calculation method of standard EWMA is:
[0091] V t = βRe t + (1-β) V t-1
[0092] Where Re t is the residual error at time t, and β represents the weight coefficient of the EWMA control chart for historical data, belonging to (0, 1].
[0093] 11) For test data, all parameter settings are the same as training data, repeat steps 7) to 9), if the obtained residual value is greater than the set threshold, a warning or alarm is issued, otherwise it is normal operation.
[0094] Example two
[0095] The wind turbine gearbox fault detection system comprises:
[0096] An acquisition module is configured to acquire observation data of a wind turbine gearbox.
[0097] A screening module is configured to perform information screening on the observation data of the wind turbine gearbox to obtain screened information.
[0098] A judgment module is configured to input the screened information into the trained IPSO-LSTM network model to determine whether the wind turbine gearbox has a fault.
[0099] In this embodiment, the observation data of the wind turbine gearbox includes power grid data, environmental data and equipment state data, wherein the power grid data includes active power, reactive power, power grid frequency, line voltage and line current, the environmental data includes wind speed, wind direction angle, air pressure and temperature, and the equipment state data includes gearbox oil temperature, gearbox front and rear shaft temperature, gearbox oil pressure and main hydraulic pressure.
[0100] The division of the modules in the embodiments of the present application is illustrative, and is merely a logical function division. In actual implementation, another division manner can be used. In addition, each function module in each embodiment of the present application can be integrated in one processor, or can be physically separated, or two or more modules can be integrated in one module. The integrated module can be realized in the form of hardware or in the form of a software function module.
[0101] Embodiment three
[0102] The computer device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the wind turbine gearbox fault detection method are implemented, for example, comprising: obtaining observation data of a wind turbine gearbox; performing information screening on the observation data of the wind turbine gearbox to obtain screened information; and inputting the screened information into a trained IPSO-LSTM network model to determine whether the wind turbine gearbox has failed. The memory can include an internal memory, such as a high-speed random memory, and can also include a non-volatile memory, such as at least one disk memory. The processor, network interface, and memory are connected to each other through an internal bus, which can be an industry standard architecture bus, a peripheral component interconnect standard bus, an extended industry standard architecture bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The memory is used to store programs, and specifically, the programs can include program codes, and the program codes include computer operation instructions. The memory can include an internal memory and a non-volatile memory, and provide instructions and data to the processor.
[0103] Embodiment four
[0104] A computer readable storage medium stores a computer program. When the computer program is executed by a processor, the steps of the wind turbine gearbox fault detection method are implemented, for example, comprising: obtaining observation data of a wind turbine gearbox; performing information screening on the observation data of the wind turbine gearbox to obtain screened information; and inputting the screened information into a trained IPSO-LSTM network model to determine whether the wind turbine gearbox has failed. Specifically, the computer readable storage medium includes but is not limited to, for example, a volatile memory and / or a non-volatile memory. The volatile memory can include a random access memory (RAM) and / or a cache memory, etc. The non-volatile memory can include a read-only memory (ROM), a hard disk, a flash memory, an optical disk, a magnetic disk, etc.
[0105] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROMs, optical storage, etc.) containing computer usable program code.
[0106] The present application is described in reference to the flow diagrams and / or block diagrams of the methods, devices (systems), and computer program products according to embodiments of the present application. It will be understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow diagrams and / or block diagrams block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks.
[0107] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the flow diagram and / or block diagram block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks.
[0108] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the flow diagram and / or block diagram block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks.
[0109] Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the present application be limited only by the scope of the claims, including any amendments thereof, and can include any adaptations or variations of the specific embodiments discussed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the application being indicated by the following claims.
[0110] It is to be understood that the application is not limited to the precise construction described in the specification and shown in the drawings, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the application. The scope of the application is limited only by the claims appended hereto.
[0111] The above description is only the preferred embodiment of the present application, not any limitation to the present application, any simple modification, change and equivalent structure change according to the technical essence of the present application to the above embodiment are still within the protection scope of the present application technical solution.
Claims
1. A method for detecting a gearbox fault in a wind turbine generator set, characterized in that: include: Obtain observation data of wind turbine gearbox; Performing information screening on the observation data of the wind turbine gearbox to obtain screened information; The screened information is input into the trained IPSO-LSTM network model to determine whether a fault occurs in the wind turbine gearbox.
2. The wind turbine generator set gearbox fault detection method according to claim 1, characterized in that: The observation data of the gearbox of the wind turbine generator set includes power grid data, environmental data and equipment status data, among which the power grid data includes the active power, reactive power, power grid frequency, line voltage and line current of the power grid; the environmental data includes wind speed, wind direction angle, air pressure and temperature; the equipment status data includes gearbox oil temperature, gearbox front and rear axle temperature, gearbox oil pressure and main hydraulic pressure.
3. The wind turbine generator set gearbox fault detection method according to claim 1, characterized in that: The process of filtering the observation data of the wind turbine gearbox to obtain the filtered information is as follows: The maximum information coefficient is used to screen the observation data of the wind turbine gearbox to obtain target data and characteristic data.
4. The wind turbine generator set gearbox fault detection method according to claim 1, characterized in that: Before the information screening of the observation data of the wind turbine gearbox is performed, the method further includes: The observed data of the wind turbine gearbox are used to remove abnormal values and supplement missing values.
5. The wind turbine generator set gearbox fault detection method according to claim 1, characterized in that: Also includes: The IPSO-LSTM network model is evaluated by the mean absolute error, root mean square error, and symmetric mean absolute percentage error to determine the prediction performance of the IPSO-LSTM network model. When the prediction performance of the IPSO-LSTM network model is insufficient, the IPSO-LSTM network model is retrained.
6. The wind turbine generator set gearbox fault detection method according to claim 5, characterized in that: Also includes: When the prediction performance of the IPSO-LSTM network model is insufficient, a warning signal or an alarm signal is issued.
7. A wind turbine gearbox fault detection system, characterized in that: include: An acquisition module, used for acquiring observation data of the gearbox of the wind turbine generator set; A screening module, configured to screen the observation data of the wind turbine gearbox to obtain screened information; The judgment module is used to input the screened information into the trained IPSO-LSTM network model to determine whether the wind turbine gearbox has a fault.
8. The wind turbine generator gearbox fault detection system according to claim 7, characterized in that: The observation data of the gearbox of the wind turbine generator set includes power grid data, environmental data and equipment status data, among which the power grid data includes the active power, reactive power, power grid frequency, line voltage and line current of the power grid; the environmental data includes wind speed, wind direction angle, air pressure and temperature; the equipment status data includes gearbox oil temperature, gearbox front and rear axle temperature, gearbox oil pressure and main hydraulic pressure.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the wind turbine gearbox fault detection method according to any one of claims 1 to 6 are implemented.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the wind turbine gearbox fault detection method according to any one of claims 1 to 6 are implemented.