Fan bearing temperature prediction and early warning system based on artificial intelligence
The AI-based wind turbine bearing temperature prediction and early warning system utilizes mechanistic models and machine learning methods for real-time prediction and status evaluation, solving the problem of low accuracy in wind turbine bearing temperature fault prediction. This enables early fault detection and effective maintenance, improving the reliability and stability of equipment operation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING WEIZHIXINYE TECH CO LTD
- Filing Date
- 2023-05-29
- Publication Date
- 2026-07-03
AI Technical Summary
Existing technologies lack accurate prediction of wind turbine bearing temperature faults, leading to significant impacts on wind turbine shutdowns and power generation. Effective prediction and early warning methods are also lacking.
An AI-based wind turbine bearing temperature prediction and early warning system is adopted. The system collects relevant parameters and historical data through a data acquisition module, establishes a bearing temperature model using a mechanism model and machine learning methods, and performs real-time prediction and status evaluation in conjunction with an anomaly discrimination model to guide the maintenance of systems that may affect bearing temperature.
It enables real-time prediction and fault warning of wind turbine bearing temperature, allows for the development of reasonable maintenance plans, avoids destructive failures, and improves the reliability and stability of equipment operation.
Smart Images
Figure CN116816708B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based wind turbine bearing temperature prediction and early warning system. Background Technology
[0002] Statistical analysis of historical wind turbine fault records reveals that bearing temperature failure is a frequent type of fault. Once a bearing temperature alarm occurs, it can severely impact the wind turbine, leading to shutdowns and ultimately affecting the overall power generation. Therefore, predicting future bearing temperature failures can help understand future trends and the urgency of potential alarms, providing guidance for on-site engineers to implement effective and reasonable protective measures. However, current predictions for bearing temperature failures are not very accurate, thus offering limited practical guidance. Summary of the Invention
[0003] In view of this, the purpose of this application is to propose an artificial intelligence-based wind turbine bearing temperature prediction and early warning system, which can specifically solve the existing problems.
[0004] To achieve the above objectives, this application also proposes an artificial intelligence-based wind turbine bearing temperature prediction and early warning system, comprising:
[0005] The data acquisition module acquires historical and real-time data on relevant parameters causing the temperature rise of the fan bearing and the temperature of the bearing cooling medium. The relevant parameters include at least: fan y-axis vibration, fan x-axis vibration, fan bearing temperature, front bearing temperature at the drive end, middle bearing temperature at the drive end, rear bearing temperature at the drive end, stator coil temperature, fan oil filter differential pressure, drive end lubricating oil pressure, average wind speed, fan external temperature, fan speed, gearbox oil temperature, gearbox cooling water temperature, gearbox lubricating oil temperature, and converter actual torque.
[0006] The real-time prediction module establishes a bearing temperature model based on the mechanism model and machine learning methods, and inputs the historical data and real-time data into the bearing temperature model to make real-time predictions of the wind turbine bearing temperature.
[0007] The status evaluation module establishes an anomaly detection model based on artificial intelligence. The predicted wind turbine bearing temperature is input into the anomaly detection model to obtain a real-time evaluation result of the wind turbine bearing status. When the real-time evaluation result indicates a fault, a fault prediction is made.
[0008] The maintenance guidance module guides the maintenance of systems that may affect bearing temperature based on the fault type of the real-time evaluation results. These systems include at least the following: cooling water system, fan, gearbox, and converter.
[0009] Furthermore, the historical and real-time data are obtained through the SCADA system.
[0010] Furthermore, the establishment of the bearing temperature model based on the mechanistic model and machine learning methods includes:
[0011] Based on 3D modeling software, the physical, geometric, mechanical, and material properties of wind turbine bearings are defined, thereby enabling multi-physics, multi-scale, and multi-probability simulation reconstruction of the mechanism model in 3D virtual space.
[0012] Based on machine learning, the relevant parameters are deeply mined, and the relevant parameters are fed back into the mechanism model to optimize the mechanism model;
[0013] The dynamic principle of wind turbine bearings is analyzed, and the corresponding parametric equations are constructed by combining Hertzian contact theory and deformation mechanism.
[0014] Based on the parametric equations and the optimized mechanism model, a bearing temperature model with multiple degrees of freedom is constructed.
[0015] Furthermore, the historical and real-time data are input into the bearing temperature model, including:
[0016] The historical data is normalized; the historical data is divided into training and testing datasets according to a certain ratio; the bearing temperature model is trained using the testing dataset; the real-time data is input into the trained bearing temperature model; or,
[0017] The bearing temperature model uses a multilayer perceptron (MLP). The MLP consists of three layers: an input layer, a hidden layer, and an output layer. The input layer takes an n-dimensional vector as input, which represents n neurons. The neurons in the hidden layer are derived from the input layer. If the input layer is represented by a vector x, the output of the hidden layer is f(w1x+b1), where w1 is the weight and b1 is the bias. The f function is the sigmoid activation function.
[0018] Furthermore, the establishment of the artificial intelligence-based anomaly detection model includes:
[0019] Anomaly detection models can be constructed using regression or neural network models, or...
[0020] The anomaly detection model includes an LSTM model, which is composed of multiple model units, each of which includes a forget gate, an input gate, and an output gate. In the LSTM model, the first model unit generates output data based on the input data at time 1, and the m-th model unit generates output data based on the output dataset of the previous unit and the input data at time m.
[0021] Furthermore, the step of inputting the predicted wind turbine bearing temperature into the anomaly discrimination model to obtain a real-time evaluation result of the wind turbine bearing condition includes:
[0022] Based on the predicted wind turbine bearing temperature and the anomaly discrimination model, a real-time evaluation result on the wind turbine bearing status is generated. The anomaly discrimination model includes comparing the predicted sample and the preset reference value, and judging whether the predicted wind turbine bearing temperature meets the preset anomaly judgment conditions. If not, a normal prediction analysis result is output; if yes, an abnormal prediction analysis result is output.
[0023] Furthermore, when the fault type is fan blockage, check whether the fan filter is clogged with dust or other debris;
[0024] When the fault type is cooling water system abnormality, check whether the water flow at the inlet and outlet of the cooling water system is normal.
[0025] When the fault type is gearbox malfunction, check whether the gearbox operating speed is within the normal range.
[0026] Furthermore, when the fault type is converter malfunction, check and handle one of the following converter faults: power and torque mismatch, rectification not ready, driver board connection error, low converter temperature, bus charging timeout, main circuit breaker closing timeout, fan overheating, UPS alarm, circuit breaker status error, or main filter abnormal disconnection.
[0027] In summary, the advantages of this application and the user experience it brings are as follows:
[0028] The wind turbine bearing temperature prediction and early warning system based on artificial intelligence proposed in this application can reasonably formulate maintenance plans, avoid the occurrence of destructive failures, achieve the system design goal of early detection of minor accidents, and improve the reliability and stability of equipment operation. Attached Figure Description
[0029] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the various drawings denote the same or similar parts or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings depict only some embodiments disclosed in this application and should not be construed as limiting the scope of this application.
[0030] Figure 1 A schematic diagram of an artificial intelligence-based wind turbine bearing temperature prediction and early warning system according to an embodiment of this application is shown.
[0031] Figure 2 The diagram illustrates the specific implementation method of the data acquisition module.
[0032] Figure 3 A schematic diagram illustrating a specific implementation method of the state evaluation module according to an embodiment of this application is shown.
[0033] Figure 4 A schematic diagram illustrating a specific implementation method of the maintenance guidance module according to an embodiment of this application is shown.
[0034] Figure 5 A schematic diagram of the structure of an electronic device provided in one embodiment of this application is shown.
[0035] Figure 6 A schematic diagram of a storage medium provided in one embodiment of this application is shown. Detailed Implementation
[0036] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.
[0037] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.
[0038] The wind turbine bearing temperature prediction and early warning system based on artificial intelligence of this application establishes a bearing temperature model by using relevant parameters that cause bearing temperature rise and historical data of bearing cooling medium temperature, and uses mechanism model and machine learning methods to predict the wind turbine bearing temperature in real time. It also uses an artificial intelligence-based anomaly discrimination model to evaluate the bearing condition in real time, predict faults in advance, and guide the maintenance of cooling water system or other systems that may affect bearing temperature.
[0039] Correspondingly, such as Figure 1 As shown in the figure, the application embodiment provides an artificial intelligence-based wind turbine bearing temperature prediction and early warning system, which includes:
[0040] The data acquisition module acquires historical and real-time data on relevant parameters causing the temperature rise of the fan bearing and the temperature of the bearing cooling medium. The relevant parameters include at least: fan y-axis vibration, fan x-axis vibration, fan bearing temperature, front bearing temperature at the drive end, middle bearing temperature at the drive end, rear bearing temperature at the drive end, stator coil temperature, fan oil filter differential pressure, drive end lubricating oil pressure, average wind speed, fan external temperature, fan speed, gearbox oil temperature, gearbox cooling water temperature, gearbox lubricating oil temperature, and converter actual torque.
[0041] The real-time prediction module establishes a bearing temperature model based on the mechanism model and machine learning methods, and inputs the historical data and real-time data into the bearing temperature model to make real-time predictions of the wind turbine bearing temperature.
[0042] The status evaluation module establishes an anomaly detection model based on artificial intelligence. The predicted wind turbine bearing temperature is input into the anomaly detection model to obtain a real-time evaluation result of the wind turbine bearing status. When the real-time evaluation result indicates a fault, a fault prediction is made.
[0043] The maintenance guidance module guides the maintenance of systems that may affect bearing temperature based on the fault type of the real-time evaluation results. These systems include at least the following: cooling water system, fan, gearbox, and converter.
[0044] The following details the specific implementation methods and technical aspects of each module:
[0045] The data acquisition module includes the following steps:
[0046] SCADA systems are computer-based production process control and scheduling automation systems. They can monitor and control on-site operating equipment and are currently most widely used and technologically mature in power systems. Therefore, the data acquisition module can directly access all parameters related to the wind turbine's operating status from the SCADA system. Considering the temporal nature of wind turbine bearing temperature and related status parameters, this invention not only selects current data for these status parameters but also incorporates historical data to enhance the accuracy of prediction results.
[0047] Real-time prediction module, such as Figure 2 As shown, it includes the following steps:
[0048] S21. Establish a bearing temperature model based on the mechanism model and machine learning methods, including:
[0049] Step 1: Define the physical, geometric, mechanical, and material properties of the wind turbine bearing using 3D modeling software, and then perform multi-physics, multi-scale, and multi-probability simulation reconstruction of the mechanism model in 3D virtual space;
[0050] Step 2: Based on machine learning, perform in-depth mining of the relevant parameters and feed the relevant parameters back into the mechanism model to optimize the mechanism model;
[0051] Step 3: Analyze the dynamic principle of the wind turbine bearing, combine Hertzian contact theory and deformation mechanism to construct the corresponding parametric equations;
[0052] Step 4: Based on the parametric equations and the optimized mechanism model, construct a bearing temperature model with multiple degrees of freedom.
[0053] S22. Input the historical data and real-time data into the bearing temperature model, including:
[0054] The historical data is normalized; the historical data is divided into training and testing datasets according to a certain ratio; the bearing temperature model is trained using the testing dataset; the real-time data is input into the trained bearing temperature model; or,
[0055] The bearing temperature model uses a multilayer perceptron (MLP). The MLP consists of three layers: an input layer, a hidden layer, and an output layer. The input layer takes an n-dimensional vector as input, which represents n neurons. The neurons in the hidden layer are derived from the input layer. If the input layer is represented by a vector x, the output of the hidden layer is f(w1x+b1), where w1 is the weight and b1 is the bias. The f function is the sigmoid activation function.
[0056] S23. Real-time prediction of fan bearing temperature, including:
[0057] The bearing temperature model outputs a normalized temperature as the predicted wind turbine bearing temperature.
[0058] Status evaluation module, such as Figure 3 As shown, it includes the following steps:
[0059] S31. Establish an anomaly detection model based on artificial intelligence, including:
[0060] Anomaly detection models can be constructed using regression or neural network models, or...
[0061] The anomaly detection model includes an LSTM model, which is composed of multiple model units, each of which includes a forget gate, an input gate, and an output gate. In the LSTM model, the first model unit generates output data based on the input data at time 1, and the m-th model unit generates output data based on the output dataset of the previous unit and the input data at time m.
[0062] S32. Input the predicted wind turbine bearing temperature into the anomaly discrimination model to obtain the real-time evaluation results of the wind turbine bearing condition, including:
[0063] Based on the predicted wind turbine bearing temperature and the anomaly discrimination model, a real-time evaluation result on the wind turbine bearing status is generated. The anomaly discrimination model includes comparing the predicted sample and the preset reference value, and judging whether the predicted wind turbine bearing temperature meets the preset anomaly judgment conditions. If not, a normal prediction analysis result is output; if yes, an abnormal prediction analysis result is output.
[0064] Specifically, sampling parameters can be calculated from the predicted samples; the sampling parameters are compared with the preset reference value to determine whether the sampling parameters exceed the preset reference value; when the sampling parameters exceed a predetermined threshold, abnormal prediction analysis results are output.
[0065] S33. When the real-time evaluation result indicates a fault, a fault prediction is made.
[0066] When the real-time evaluation result is abnormal, the fault type is determined by comparing the fan bearing temperature with a preset temperature-fault lookup table, and then announced or alerted using sound, light, or electrical methods. When the real-time evaluation result is normal operation, the system returns to the data acquisition module to continue acquiring historical and real-time data.
[0067] Instructions for troubleshooting modules, such as Figure 4 As shown, it includes:
[0068] S41. When the fault type is fan blockage, check whether the fan filter is blocked by dust or other debris.
[0069] S42. When the fault type is cooling water system abnormality, check whether the water flow at the inlet and outlet of the cooling water system is normal.
[0070] S43. When the fault type is gearbox abnormality, check whether the gearbox operating speed is within the normal range;
[0071] S44. When the fault type is converter abnormal, check and handle one of the following converter faults: power and torque mismatch, rectification not ready, driver board connection error, low converter temperature, bus charging timeout, main circuit breaker closing timeout, fan overheating, UPS alarm, circuit breaker status error, main filter abnormal disconnection.
[0072] Please refer to Figure 5 This illustrates a schematic diagram of an electronic device provided by some embodiments of this application. For example... Figure 5 As shown, the electronic device 20 includes: a processor 200, a memory 201, a bus 202, and a communication interface 203. The processor 200, the communication interface 203, and the memory 201 are connected via the bus 202. The memory 201 stores a computer program that can run on the processor 200. When the processor 200 runs the computer program, it executes the artificial intelligence-based wind turbine bearing temperature prediction and early warning system provided in any of the foregoing embodiments of this application.
[0073] The memory 201 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 203 (which can be wired or wireless), such as the Internet, wide area network, local area network, or metropolitan area network.
[0074] Bus 202 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The memory 201 is used to store programs. After receiving an execution instruction, the processor 200 executes the program. The artificial intelligence-based wind turbine bearing temperature prediction and early warning system disclosed in any of the foregoing embodiments of this application can be applied to the processor 200, or implemented by the processor 200.
[0075] The processor 200 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 200 or by instructions in software form. The processor 200 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules may reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 201. The processor 200 reads the information in memory 201 and, in conjunction with its hardware, completes the steps of the above method.
[0076] The electronic device provided in this application embodiment and the fan bearing temperature prediction and early warning system based on artificial intelligence provided in this application embodiment are based on the same inventive concept and have the same beneficial effects as the methods they adopt, operate or implement.
[0077] This application also provides a computer-readable storage medium corresponding to the artificial intelligence-based wind turbine bearing temperature prediction and early warning system provided in the foregoing embodiments. Please refer to... Figure 6 The computer-readable storage medium shown is an optical disc 30, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it executes the artificial intelligence-based wind turbine bearing temperature prediction and early warning system provided in any of the foregoing embodiments.
[0078] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical and magnetic storage media, which will not be elaborated here.
[0079] The computer-readable storage medium provided in the above embodiments of this application and the artificial intelligence-based wind turbine bearing temperature prediction and early warning system provided in the embodiments of this application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application stored therein.
[0080] It should be noted that:
[0081] The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. The required structure for constructing such systems is apparent from the above description. Furthermore, this application is not directed to any particular programming language. It should be understood that the content of this application described herein can be implemented using various programming languages, and the above description of specific languages is for the purpose of disclosing the best mode of implementation of this application.
[0082] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this application may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.
[0083] Similarly, it should be understood that, in order to simplify this application and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of this application, various features of this application are sometimes grouped together into a single embodiment, figure, or description thereof. However, this method of disclosure should not be construed as reflecting an intention that the claimed application requires more features than are expressly recited in each claim. Rather, as reflected in the following claims, inventive aspects lie in fewer than all features of a single foregoing disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of this application.
[0084] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.
[0085] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features but not others included in other embodiments, combinations of features from different embodiments are intended to be within the scope of this application and form different embodiments. For example, in the following claims, any of the claimed embodiments can be used in any combination.
[0086] The various component embodiments of this application can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some or all of the components in the virtual machine creation system according to the embodiments of this application. This application can also be implemented as a device or system program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such an implementation of this application can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.
[0087] It should be noted that the above embodiments are illustrative of this application and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. This application can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several systems, several of these systems may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.
[0088] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various variations or substitutions within the technical scope disclosed in this application, and these should all be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
Claims
1. An artificial intelligence-based fan bearing temperature prediction and early warning system, characterized in that, include: The data acquisition module acquires historical and real-time data on relevant parameters causing the temperature rise of the fan bearing and the temperature of the bearing cooling medium. The relevant parameters include at least: fan y-axis vibration, fan x-axis vibration, fan bearing temperature, front bearing temperature at the drive end, middle bearing temperature at the drive end, rear bearing temperature at the drive end, stator coil temperature, fan oil filter differential pressure, drive end lubricating oil pressure, average wind speed, fan external temperature, fan speed, gearbox oil temperature, gearbox cooling water temperature, gearbox lubricating oil temperature, and converter actual torque. The real-time prediction module establishes a bearing temperature model based on the mechanism model and machine learning methods, and inputs the historical data and real-time data into the bearing temperature model to make real-time predictions of the wind turbine bearing temperature. The status evaluation module establishes an anomaly detection model based on artificial intelligence. The predicted wind turbine bearing temperature is input into the anomaly detection model to obtain a real-time evaluation result of the wind turbine bearing status. When the real-time evaluation result indicates a fault, a fault prediction is made. The maintenance guidance module guides the maintenance of systems that may affect bearing temperature based on the fault type of the real-time evaluation results. These systems include at least the following: cooling water system, fan, gearbox, and converter. The establishment of the bearing temperature model based on the mechanism model and machine learning method includes: Based on 3D modeling software, the physical, geometric, mechanical, and material properties of wind turbine bearings are defined, thereby enabling multi-physics, multi-scale, and multi-probability simulation reconstruction of the mechanism model in 3D virtual space. Based on machine learning, the relevant parameters are deeply mined, and the relevant parameters are fed back into the mechanism model to optimize the mechanism model; The dynamic principle of wind turbine bearings is analyzed, and the corresponding parametric equations are constructed by combining Hertzian contact theory and deformation mechanism. Based on the parametric equations and the optimized mechanism model, a bearing temperature model with multiple degrees of freedom is constructed.
2. The system according to claim 1, characterized in that, The historical and real-time data are obtained through the SCADA system.
3. The system according to claim 1, characterized in that, Inputting the historical and real-time data into the bearing temperature model includes: The historical data is normalized and then divided into training and testing datasets according to a set ratio. The bearing temperature model is trained using the testing dataset, and the real-time data is input into the trained bearing temperature model. The bearing temperature model employs a multilayer perceptron (MLP). The MLP consists of three layers: an input layer, a hidden layer, and an output layer. The input layer receives an n-dimensional vector, representing n neurons. The neurons in the hidden layer are derived from the input layer. If the input layer is represented by a vector x, the output of the hidden layer is: ;in, It's weight. It is a bias, the stated The function is the sigmoid activation function.
4. The system according to claim 3, characterized in that, The establishment of an anomaly detection model based on artificial intelligence includes: Anomaly detection models can be constructed using regression models or neural network models. In the case where a neural network model is used to construct the anomaly detection model, the anomaly detection model includes an LSTM model, which is composed of multiple model units, each of which includes a forget gate, an input gate, and an output gate; wherein, the first model unit in the LSTM model generates output data based on the input data at time 1, and the m-th model unit generates output data based on the output dataset of the previous unit and the input data at time m.
5. The system according to claim 4, characterized in that, The process of inputting the predicted wind turbine bearing temperature into the anomaly discrimination model to obtain real-time evaluation results of the wind turbine bearing condition includes: Based on the predicted wind turbine bearing temperature and the anomaly discrimination model, a real-time evaluation result on the wind turbine bearing status is generated. The anomaly discrimination model includes comparing the predicted sample and the preset reference value, and judging whether the predicted wind turbine bearing temperature meets the preset anomaly judgment conditions. If not, a normal prediction analysis result is output; if yes, an abnormal prediction analysis result is output.
6. The system according to claim 5, characterized in that, When the fault type is fan blockage, check whether the fan filter is clogged with dust or other debris; When the fault type is cooling water system abnormality, check whether the water flow at the inlet and outlet of the cooling water system is normal. When the fault type is gearbox malfunction, check whether the gearbox operating speed is within the normal range.
7. The system according to claim 5 or 6, characterized in that, When the fault type is converter malfunction, check and handle one of the following converter faults: power and torque mismatch, rectification not ready, driver board connection error, low converter temperature, bus charging timeout, main circuit breaker closing timeout, fan overheating, UPS alarm, circuit breaker status error, or main filter abnormal disconnection.