Wind turbine generator yaw anomaly detection, shutdown protection methods and wind turbine generator
By combining filtering of nacelle position data with internal variables of the unit, abnormal yaw states of wind turbines are identified and graded protection is implemented, solving the problem of insufficient real-time online protection in existing technologies and achieving safe operation and extended lifespan of the equipment.
Patent Information
- Application Number
- CN202611550246.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-10-08
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies for wind turbine yaw system fault detection lack real-time online protection, cannot identify various abnormal states during yaw operation, leading to equipment damage, and have high hardware configuration requirements, inflexible protection strategies, and inability to identify complex yaw abnormal states.
By filtering based on cabin position data acquisition, the yaw speed is determined. Combined with internal crew variables and yaw initiation commands, real-time identification and graded protection of multiple types of abnormal yaw states are achieved, including cabin skidding in the non-yaw phase, underspeed in the yaw phase, continuous anti-towing, and intermittent anti-towing. A lightweight algorithm is deployed in the main control system.
It enables accurate identification and timely protection against various types of yaw anomalies, reducing equipment wear and tear, extending equipment life, minimizing unnecessary downtime, and lowering hardware requirements. It is suitable for both existing and newly built units.
Smart Images

Figure CN123109972A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of online fault detection and safety protection technology for wind turbine yaw systems, and belongs to the field of data filtering algorithm combined with multi-condition timing discrimination and hierarchical protection and control technology of wind power main control system. In particular, it relates to a method for detecting abnormal yaw state of wind turbine generator, a method for protecting wind turbine generator from abnormal yaw shutdown, and a wind turbine generator. Background Technology
[0002] The existing technology for fault detection and safety protection of wind turbine yaw systems has the following deficiencies: 1. Some solutions are merely post-event offline data analysis solutions, processing historical SCADA datasets stored in wind farms. They require batch filtering of yaw state datasets and rely on vibration acceleration statistical features for anomaly identification, but cannot be embedded into the turbine's main control system to achieve real-time online closed-loop protection. They only output warnings to the maintenance terminal and cannot directly link the turbine's main control system to execute real-time protection actions such as stopping yaw or shutting down the turbine. Furthermore, the identification feature is vibration acceleration, without considering yaw speed direction or amplitude. They lack the ability to identify yaw anomalies strongly correlated with position and speed, such as underspeed, reverse towing, and nacelle slippage. They cannot protect the turbine in time when an anomaly occurs, as the abnormal load continues to act on equipment components, and warnings are only issued after component damage has already occurred, resulting in significant protection lag. 2. Some solutions are limited to the standby phase after yaw demand ends and yaw action stops, failing to identify faults during the yaw operation process. They cannot identify abnormal states such as under-yawing, continuous reverse towing, or intermittent reverse towing during yaw operation. The judgment criteria involve numerous environmental and braking parameters such as wind speed and direction, reverse wind load, and braking efficiency coefficient, resulting in complex calculations that rely on braking pressure sensors and high-precision wind measurement equipment, placing high demands on the unit's hardware configuration. Furthermore, they lack statistical logic for the duration and frequency of abnormalities during yaw action, making it impossible to identify repeated impact degradation during yaw. There are blind spots in identifying intermittent reverse towing and jamming / under-yawing faults caused by gusts in special terrain, leading to the continuous accumulation of hidden impact loads, accelerating fatigue aging of yaw drivetrain components, and hindering extended unit lifespan. 3. The shortcomings of fixed yaw speed threshold protection schemes are more prominent: relying solely on a single threshold for yaw speed amplitude for fault determination lacks joint judgment based on operating condition information such as yaw start / stop commands and yaw direction commands, resulting in incomplete coverage of yaw abnormal state identification, only able to identify simple yaw underspeed; the protection strategy is one-size-fits-all and cannot achieve graded protection; lacking dual constraints of timing duration and event trigger frequency, it is easily affected by sensor jumps and instantaneous signal disturbances, leading to false judgments and false shutdowns, while failing to identify recurring intermittent abnormalities, resulting in the continuous accumulation of alternating impact loads, accelerating the wear and aging of the yaw motor and reducer, and shortening the overall service life of the unit. Summary of the Invention
[0003] This application provides a method for detecting abnormal yaw states of a wind turbine generator set, a method for protecting a wind turbine generator set from abnormal yaw shutdown, and a wind turbine generator set. It can cover four types of abnormal yaw states: "standby slip + underspeed / continuous reverse towing / intermittent reverse towing during yaw operation". It also takes into account the needs of real-time identification of multiple types of abnormal yaw states, lightweight algorithm deployment, and active protection and life extension of the generator set.
[0004] According to one of the solutions in this application, a method for detecting abnormal yaw states of wind turbine generators is provided, including: Yaw speed is determined by filtering based on cabin position data acquisition; Synchronously acquire internal variables of the unit to align with the timestamp of the cabin azimuth data; In response to different yaw start commands, corresponding judgments are performed under different unit operating conditions, and judgments under other operating conditions are prohibited under one operating condition. The unit operating conditions include at least the yaw standby operating condition and the yaw execution operating condition.
[0005] In some embodiments, determining the yaw speed based on filtering of cabin position data acquisition includes: Based on the sampled data from the cabin, the cabin position angle value is obtained through angle conversion; The yaw speed is obtained by combining the cabin position angle value based on the low-pass filter with the position change rate.
[0006] In some embodiments, synchronously acquiring unit internal variables includes: synchronously reading unit internal variables from the master control real-time database within the same master control cycle; the variables are at least from generator speed, unit active power, yaw start command, and yaw direction command.
[0007] In some embodiments, the determination includes at least: In yaw standby mode and when the yaw direction command indicates no yaw, cabin skidding is determined during the non-yaw phase. Under yaw execution conditions and when the yaw direction command indicates yaw, the following criteria are used to identify continuous underspeed during the yaw phase, continuous anti-towing during the yaw phase, and intermittent anti-towing during the yaw phase.
[0008] In some embodiments, the discrimination further includes: Reverse direction event determination includes determining whether the crew has yawed and towed back based on whether the yaw speed and the left and right yaw direction commands meet the predetermined conditions. The single-event filtering rule states that if the number of consecutive reverse states is not less than a preset number of master control cycles, the reverse event is considered a valid event.
[0009] According to one of the solutions in this application, a method for yaw anomaly shutdown protection of a wind turbine generator is provided. Based on the determined yaw anomaly state, the method includes: In the first abnormal scenario, the current yaw command is latched by the first protection command, and the yaw action lockout is released after a preset delay; In the second abnormal scenario, if the cumulative number of alarms per day is not less than the threshold, the second protection command will lock the yaw control command and the yaw mechanical structure until the first reset condition is met. In the third abnormal scenario, the third protection command, upon identifying abnormal cabin slip characteristics in the non-yaw state, locks the yaw control command and the yaw mechanical structure until the second reset condition is met.
[0010] In some embodiments, the first abnormal scenario includes, under the condition of yaw execution and yaw direction command indicating yaw, the following are identified: continuous underspeed during yaw phase, continuous anti-towing during yaw phase, and intermittent anti-towing during yaw phase.
[0011] In some embodiments, the second abnormal scenario includes a single instance of underspeed during yaw execution when the yaw direction command indicates yaw.
[0012] In some embodiments, the third abnormal scenario includes the cabin skidding determination during the non-yawing phase when the yawing standby condition is in which the yawing direction command indicates no yawing.
[0013] According to one of the embodiments of this application, a wind turbine generator is provided, including a controller configured to perform the steps of the method described above.
[0014] The wind turbine generator set yaw anomaly detection method of various embodiments of this application determines the yaw speed based at least on the filtering of nacelle position data acquisition; synchronously acquires internal variables of the generator set to align with the timestamp of the nacelle azimuth data; responds to different yaw start commands, performs corresponding discrimination under different generator set operating conditions, and prohibits discrimination under other operating conditions under one operating condition. The generator set operating conditions include at least yaw standby operating condition and yaw execution operating condition. Thus, relying on the filtering and processing of nacelle position time sequence data, and combined with multi-dimensional operating signals of the generator set, a multi-type yaw anomaly state discrimination logic is established, and a differentiated hierarchical shutdown and protection strategy is executed. This solves the technical pain point that the traditional single yaw speed threshold detection scheme cannot identify multiple complex yaw anomaly states. Under the premise of ensuring the safe operation of the generator set, it reduces the alternating impact load of various components of the yaw system, suppresses abnormal equipment wear, and achieves the goal of extending the service life of the yaw system and the entire generator set.
[0015] It should be understood that the foregoing general description and the following detailed description are exemplary and illustrative only, and are not intended to limit the scope of this application. Attached Figure Description
[0016] To more clearly illustrate the technical solutions in this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0017] Figure 1 The illustration shows a schematic diagram of the implementation process of a wind turbine generator yaw anomaly detection method and related shutdown protection method according to an embodiment of this application. Detailed Implementation
[0018] Various embodiments and features of this application are described herein with reference to the accompanying drawings.
[0019] It should be understood that various modifications can be made to the embodiments described herein. Therefore, the above description should not be considered as limiting, but merely as an example of embodiments. Other modifications within the scope and spirit of this application will be apparent to those skilled in the art.
[0020] The accompanying drawings, which are included in and form part of this specification, illustrate embodiments of the present application and, together with the general description of the present application given above and the detailed description of the embodiments given below, serve to explain the principles of the present application.
[0021] These and other features of this application will become apparent from the following description of preferred forms of embodiments given as non-limiting examples, with reference to the accompanying drawings.
[0022] It should also be understood that although this application has been described with reference to some specific examples, those skilled in the art can certainly implement many other equivalent forms of this application.
[0023] The above and other aspects, features and advantages of this application will become more apparent when taken in conjunction with the accompanying drawings and in view of the following detailed description.
[0024] Specific embodiments of this application are described thereafter with reference to the accompanying drawings; however, it should be understood that the claimed embodiments are merely examples of this application, which can be implemented in various ways. Well-known and / or repeated functions and structures are not described in detail to avoid unnecessary or redundant details that could obscure the application. Therefore, the specific structural and functional details claimed herein are not intended to be limiting, but merely serve as the basis and representative basis for the claims to teach those skilled in the art to use this application in a variety of substantially any suitable detailed structures.
[0025] This specification may use the phrases “in one embodiment,” “in another embodiment,” “in yet another embodiment,” or “in other embodiments,” all of which may refer to one or more of the same or different embodiments according to this application.
[0026] The yaw system is the core actuator for wind turbine generators to achieve precise wind alignment and efficient wind energy capture. The reliability of the yaw system directly affects the wind energy capture efficiency of the generator, the fatigue load of transmission components such as the yaw motor, yaw reducer, and braking mechanism, as well as the overall operational safety of the generator. It is a key object for extending the life of equipment in wind farms with complex terrain.
[0027] With the development of the wind power industry, wind farms are increasingly being built in special terrain areas such as ridges, steep slopes, and cliffs, where turbulent winds and gusts are prominent. Simultaneously, the blade size and swept area of large-megawatt turbines are continuously increasing, significantly enhancing the aerodynamic reaction forces on the rotor. This makes the turbines highly susceptible to various yaw anomalies during yaw-facing operation, including under-yawing speed, continuous reverse drag, intermittent reverse drag, and nacelle slippage during non-yaw operations. These anomalies can cause yaw motor overload, excessive output current, accelerated insulation aging, and bearing wear. They can also generate impact loads on the yaw gear pair, drastically shortening the service life of yaw chain components, and in severe cases, directly causing yaw motor burnout, gear damage, and other equipment damage.
[0028] Existing solutions for wind turbine detection and protection either only perform offline post-event analysis and cannot provide online real-time protection, or only cover nacelle skidding during standby. Neither can fully cover the four types of yaw anomalies: "standby skidding + underspeed / continuous anti-tow / intermittent anti-tow during yaw operation." It is difficult to simultaneously meet the needs of real-time identification of multiple types of yaw anomalies, lightweight algorithm deployment, and active protection and life extension of the turbine.
[0029] Based on the background section described above, this application provides illustrative solutions to address the deficiencies in the prior art through embodiments, but these are not intended to limit the scope of patent protection claimed in this application.
[0030] To address the above issues, as one possible solution, refer to Figure 1 This illustration shows a schematic flowchart of an implementation process for a wind turbine generator yaw anomaly detection method according to an embodiment of this application (which also incorporates a wind turbine generator yaw anomaly shutdown protection method). The embodiment of this application provides a wind turbine generator yaw anomaly detection method, including: Yaw speed is determined by filtering based on cabin position data acquisition; Synchronously acquire internal variables of the unit to align with the timestamp of the cabin azimuth data; In response to different yaw start commands, corresponding judgments are performed under different unit operating conditions, and judgments under other operating conditions are prohibited under one operating condition. The unit operating conditions include at least the yaw standby operating condition and the yaw execution operating condition.
[0031] Regarding the foregoing content, this application aims to provide at least one method for detecting abnormal yaw states in wind turbine generators. This method can perform first-order low-pass filtering preprocessing on the raw nacelle position data to eliminate sampling outlier interference and calculate a smooth real-time yaw speed. By combining yaw start commands, left and right yaw direction commands, and other generator operation signals, and through a combination of duration thresholds and event trigger count thresholds, it can distinguish and identify four typical abnormal yaw states: nacelle slippage in the non-yaw phase, continuous underspeed in the yaw phase, continuous anti-towing in the yaw phase, and intermittent anti-towing in the yaw phase. It also fully covers both standby and yaw execution operating conditions. The algorithm only uses the nacelle position and main control yaw command signals, without the need for additional sensors such as wind speed and braking pressure, making the algorithm lightweight.
[0032] One of the inventive concepts of this application is to overcome the shortcomings of existing yaw protection technologies, such as the use of a single static threshold and the limited types of yaw anomaly identification. The technical solution of this application can integrate the algorithm module into the main control system of the wind turbine, using data collected by the nacelle position sensor and the main control operation command of the unit as input sources. The main technical process can be summarized as: nacelle position data acquisition and filtering and yaw speed calculation, synchronous acquisition of unit operation data, identification of multiple types of yaw anomaly states, and execution of graded shutdown and protection logic. The whole solution forms an integrated operation logic of "data acquisition - filtering calculation - multi-state identification - graded shutdown protection".
[0033] This document provides a necessary description of the wind turbine generator yaw anomaly detection method of this application by exemplarily describing steps S1 to S3 as some implementation methods, but it is not intended to be the sole limitation on the scope of protection of the technical solution of this application.
[0034] In some implementations, the embodiments of this application may be: determining the yaw speed based on the filtering of cabin position data, including: obtaining the cabin position angle value by angle conversion based on the sampled data from the cabin; and obtaining the yaw speed by combining the cabin position angle value with the position change rate based on the low-pass filtered value.
[0035] Step S1: Real-time acquisition, filtering, and yaw speed calculation of cabin position data.
[0036] As an example, this embodiment can read the sampled data uploaded by the cabin position sensor in real time, and output a cabin position angle value with a variation range of ±720° after angle conversion. A first-order low-pass filter is used to filter the cabin position angle value to eliminate velocity calculation disturbances caused by abnormal sampling outliers. The filtering calculation formula is as follows: , In the formula: where, This is the current sampled value. This is the current filter output value. This is the filtered output value from the previous moment. For filter coefficients (0 < <1), the execution cycle of the main control program can be 10ms. The initial value can be set to 0.01.
[0037] Based on the filtered cabin position, the rate of position change can be calculated with a control cycle of 10ms to obtain the real-time yaw speed. The calculation formula is as follows: , In the formula: where, For real-time yaw speed, This represents the current filtered cabin position angle value. This represents the filtered cabin position angle value from the previous moment. A negative yaw rate indicates a decrease in cabin position angle, representing a left yaw; a positive yaw rate indicates an increase in cabin position angle, representing a right yaw.
[0038] In some implementation schemes, the embodiments of this application may include: synchronously acquiring internal variables of the generator set, including: synchronously reading internal variables of the generator set from the real-time database of the master control system within the same master control cycle; the variables are at least from generator speed, generator set active power, yaw start command, and yaw direction command.
[0039] In this embodiment, step S2 can be added on top of step S1.
[0040] Step S2: Synchronous acquisition of unit operation data.
[0041] As an example, this embodiment can synchronously read internal unit variables from the main control real-time database within the same main control cycle, achieving alignment with the nacelle azimuth data timestamp. The variable list and collection requirements are as follows: Generator speed: collected by generator encoder, unit rpm; Unit active power: real-time output active power of converter, unit kW; Yaw start command: Boolean value, 1 = yaw enabled; 0 = yaw disabled; Left yaw direction command: Boolean value, 0 = left yaw stopped, 1 = left yaw started; Right yaw direction command: Boolean value, 0 = right yaw stopped, 1 = right yaw started.
[0042] In some implementations, the embodiments of this application may include: the discrimination at least includes: cabin skidding discrimination during the non-yawing phase when the yawing standby condition and the yawing direction command indicates no yawing; and continuous underspeed discrimination, continuous anti-towing discrimination, and intermittent anti-towing discrimination during the yawing phase when the yawing execution condition and the yawing direction command indicates yawing. It may also include: reverse direction event determination, including determining whether the crew has engaged in yawing anti-towing based on whether the yawing speed and the left and right yawing direction commands meet predetermined conditions; and a single event filtering rule, where a reverse event is considered a valid event if the reverse state continues for no less than a preset number of master control cycles.
[0043] In this embodiment, step S3 can be added on top of step S2.
[0044] Step S3: Identification of multiple types of yaw anomalies.
[0045] As illustrated, the algorithm in this embodiment can be executed cycle by cycle in units of a master control predetermined time (e.g., 10ms). The yaw start command divides the unit into two operating conditions: yaw start command = 0, entering yaw standby mode; yaw start command = 1, entering yaw execution mode. The discrimination branches under the two operating conditions are isolated from each other; the yaw standby mode does not execute the yaw execution mode criteria, and the yaw execution mode does not execute the cabin taxiing criteria. All timings are accumulated using a master control cycle counter, incrementing by 1 every interval (e.g., 10ms). In specific implementations, it can be set so that once an abnormal condition is not met, the corresponding counter is immediately reset to zero, without accumulating across intermittent states. The discrimination schemes involved in the various embodiments of the technical solution of this application are illustrated below.
[0046] Cabin skidding detection during non-yaw phase: Triggering conditions: Yaw start command = 0, yaw direction command = 0; Pre-condition constraints: When the unit is connected to the grid, the active power is greater than 5% of the rated power; eliminate slight angle drift under shutdown, disconnection from the grid, and windless conditions; Determine yaw speed on a cycle ,when If the absolute value of the yaw is greater than the first threshold, the slip counter starts to increment; if the yaw start command is 1, the slip counter is reset to zero. When the cumulative duration of the skid counter is greater than or equal to the time threshold T1, the cabin skid is determined to be abnormal.
[0047] Determining sustained underspeed during the yaw phase: Triggering conditions: Yaw start command = 1, left or right yaw direction command = 1; Pre-screening conditions: When the main control issues an active speed reduction command, the wind target position is reached, or the unit is in a start-stop transition condition, the underspeed counter is frozen and no fault judgment is made. Determine yaw speed on a cycle ,when If the absolute value is less than the second threshold, the underspeed counter starts to increment; if the condition is not met, the counter is reset to zero. When the cumulative duration of the underspeed counter is greater than or equal to the time threshold T2, the yaw is determined to be continuously underspeed abnormal.
[0048] Continuous anti-tow detection during yaw phase: Triggering conditions: Yaw start command = 1, left or right yaw direction command = 1; Direction determination: based on calculated yaw speed Compare with the yaw direction command; if the yaw speed... <0 and right yaw direction command = 1, or yaw speed If the value is >0 and the left yaw direction command is =1, then the crew is determined to be in a yaw reverse towing situation; The direction is reversed cycle by cycle; if the above conditions are met, the reverse drag timer starts to accumulate; if the direction returns to normal, the counter is reset to zero. When the cumulative duration of the reverse counter is greater than or equal to the time threshold T3, the yaw is determined to be continuously reverse-dragging abnormal.
[0049] Intermittent anti-towing detection during yaw phase: Triggering conditions: Yaw start command = 1, left or right yaw direction command = 1; A first-in-first-out (FIFO) sliding window is used, with a total window period T4 (e.g., initially set to 5 minutes). Each valid reverse event within the window is marked in the queue.
[0050] Reverse direction event determination: based on calculated yaw speed Compare with the yaw direction command; if the yaw speed... <0 and right yaw direction command = 1, or yaw speed If the value is greater than 0 and the left yaw direction command is 1, then the crew is determined to be in a yaw reverse towing situation.
[0051] Single event filtering rule: Only when the reverse state is consecutively greater than or equal to a preset number (e.g., 2) of master control cycles is it counted as a valid reverse event; instantaneous reverse spikes with only 1 cycle are directly discarded and not queued.
[0052] In the above-described discrimination method, the embodiment of this application can be implemented as follows: The following steps are performed every master control cycle: Determine if a valid reverse event is met; mark new events and enqueue them; if the queue is full, pop the oldest event from the front of the queue; count the total number of valid reverse events in the queue. When the number of valid reverse events in the queue exceeds the frequency threshold, an intermittent yaw anti-drag anomaly is determined. If the yaw start command = 1, the number of valid reverse events is reset to zero.
[0053] The wind turbine generator yaw anomaly detection method disclosed in the embodiments of this application, compared with the prior art, abandons the traditional single threshold judgment mode that relies solely on the yaw speed amplitude; and differs from the limited scenario of only handling standby skidding after yaw ends, the technical solution of this application combines yaw start command and yaw direction command to distinguish between the standby and yaw execution conditions of the generator, and combines duration threshold and abnormal trigger number threshold per unit time to identify four different yaw anomaly states with different mechanisms at once: non-yaw nacelle skidding, continuous underspeed, continuous anti-tow, and intermittent anti-tow. It fully covers all operating conditions, including standby and yaw execution, expands the coverage of yaw anomaly identification, promptly captures hidden impact anomalies, reduces damage to yaw transmission components from unknown loads, and extends the service life of the unit. On the one hand, the nacelle position data is preprocessed with a first-order low-pass filter to filter out sensor jump out-of-field interference, improve the stability of yaw speed calculation, and reduce misjudgments and false protection caused by signal disturbances. Thus, it only uses the nacelle position and main control command signals, without the need for additional sensor inputs such as wind speed and braking pressure. The algorithm is lightweight and easy to deploy in the main control of various existing and newly built units.
[0054] To address the above issues, as one possible solution, refer to Figure 1 This illustration shows a schematic flowchart of an implementation process for a wind turbine generator yaw anomaly shutdown protection method according to an embodiment of this application (the illustration also incorporates a wind turbine generator yaw anomaly state detection method). The embodiment of this application provides a wind turbine generator yaw anomaly shutdown protection method, which, based on the determined yaw anomaly state, includes: In the first abnormal scenario, the current yaw command is latched by the first protection command, and the yaw action lockout is released after a preset delay; In the second abnormal scenario, if the cumulative number of alarms per day is not less than the threshold, the second protection command will lock the yaw control command and the yaw mechanical structure until the first reset condition is met. In the third abnormal scenario, the third protection command, upon identifying abnormal cabin slip characteristics in the non-yaw state, locks the yaw control command and the yaw mechanical structure until the second reset condition is met.
[0055] In light of the foregoing, this application aims to provide at least one method for protecting wind turbine generators from yaw anomalies during shutdown. This method allows for the configuration of tiered shutdown and protection logic for different yaw anomaly states, distinguishing between alarm retry and fault shutdown, and identifying recurring degradation phenomena by statistically analyzing the frequency of abnormal triggers within a given period. While ensuring the safety of the entire generator by promptly handling dangerous faults, this method reduces unnecessary shutdowns, lowers the impact and alternating loads on transmission components such as the yaw motor and yaw reducer, suppresses fatigue wear of parts, and extends the service life of the yaw system and the entire generator.
[0056] This document provides a necessary description of the wind turbine generator yaw abnormal shutdown protection method of this application by exemplarily using step S4 as some implementation methods, and does not constitute the sole limitation on the scope of protection of the technical solution of this application.
[0057] Step S4: Execution of graded shutdown and protection logic.
[0058] As illustrated, this embodiment can, based on the refined yaw anomaly status labels identified in the output, such as those identified by the method described in the previous embodiment, match differentiated and hierarchical protection and handling strategies for different fault severity levels and failure mechanisms. The technical solution of this application aims at least to divide yaw anomaly protection into a three-level closed-loop handling logic: alarm retry fault-tolerant protection, frequency statistics-based escalation fault shutdown protection, and direct shutdown protection for major failures. Each logic has independent criteria, does not conflict with each other, and is progressively hierarchical, achieving precise protection control of "minor anomaly fault-tolerant retry, frequent anomaly downgrade shutdown, and immediate blocking of major risks."
[0059] In some implementations, the embodiments of this application may include: the first abnormal scenario includes the following when the yaw execution condition is underway and the yaw direction command indicates yaw, the yaw phase is continuously underspeeded, the yaw phase is continuously reverse-towed, and the yaw phase is intermittently reverse-towed.
[0060] As illustrated, the alarm retry fault-tolerant protection in this application embodiment is applicable to abnormal scenarios. This includes three types of minor, controllable anomalies identified only during the active yaw operation phase of the aircraft: continuous underspeed during yaw, continuous anti-tow during yaw, and intermittent anti-tow during yaw. These anomalies only represent abnormal yaw loads and power matching deviations, and do not indicate structural risks such as brake failure or cabin instability. They are capable of short-term fault tolerance and retrying to resume operation.
[0061] In this type of scenario, this embodiment can be implemented using instructions adapted to that scenario: Abnormal Trigger Judgment: When the system detects any of the above-mentioned abnormal yaw states in real time and continuously meets the judgment period, it immediately generates and sends a yaw underspeed warning alarm signal. The unit's main control system immediately latches the current yaw command, instantly terminates the ongoing clockwise / counterclockwise yaw rotation action, cuts off the yaw motor drive output, and stops the operation of the yaw reducer and transmission mechanism to avoid continuous abnormal operation from aggravating mechanical load impact. Delayed retry interlock: After the unit stops yaw, the built-in delay timer is started and the unit waits strictly for the preset interval T5 (initial value is set to 5 minutes). During the delay period, the main control system interlocks all manual and automatic yaw trigger commands, prohibits the unit from restarting the yaw action, and reserves time for the mechanical transmission structure load to be released and the electrical parameters to be reset. Retry and Count Archiving: After the interval time T5 ends, the yaw action interlock is automatically released, allowing the unit to re-initiate the yaw wind-fighting action according to the wind direction deviation and grid control requirements; at the same time, the main control system's built-in fault counter completes one valid yaw underspeed alarm count. This count is included in the daily alarm frequency statistics log. The data is stored in real time, without power loss, and without repeated accumulation, providing original data support for subsequent fault escalation judgment. This embodiment implements a protection technology solution at this level. Compared with the shortcomings of the existing technology, the effect is at least reflected in the ability to achieve fault-tolerant operation by stopping yaw retry for occasional yaw anomalies caused by instantaneous wind speed disturbances and electrical instantaneous fluctuations, thereby avoiding unnecessary unit shutdown caused by a single abnormal pulse and improving unit availability.
[0062] In some implementations, the embodiments of this application may include: the second abnormal scenario includes a single yaw underspeed determined when the yaw execution condition is underway and the yaw direction command indicates yaw.
[0063] As illustrated, the frequency statistics upgrade fault shutdown protection in this application embodiment can be applied to abnormal scenarios, including those where yaw underspeed alarms frequently recur within a single day's operating cycle. A single yaw underspeed alarm is a controllable, instantaneous anomaly, but high-frequency repetition indicates that core components such as the yaw motor, reducer, brake pads, and transmission gears have experienced latent faults such as fatigue degradation, excessive clearances, and power attenuation. Continuous operation will lead to secondary faults such as accelerated component wear, yaw jamming, and overload exceeding limits. Therefore, frequency statistics are needed to achieve upgraded fault protection.
[0064] In this type of scenario, this embodiment can be implemented using instructions adapted to that scenario: Frequency statistics interval: With a fixed statistical period (e.g., 00:00-24:00 on a natural single day), the main control system accumulates all valid yaw and underspeed alarms in real time on the same day, and removes invalid alarms generated by momentary false triggers and manual reset tests, and only counts the automatically triggered alarm events under normal operating conditions of the unit; Fault escalation criteria: preset frequency protection threshold; when the cumulative number of valid alarms per day is greater than or equal to the preset frequency protection threshold, the system determines that the yaw mechanism has a persistent and recurring abnormality, and the risk of component deterioration cannot be ignored, triggering the protection escalation logic; This embodiment implements a protection technology solution for shutdown execution and alarm at this level: through the adapted command, the original early warning alarm is immediately upgraded to a yaw underspeed fault shutdown signal, and the unit executes a safe shutdown procedure: firstly, it exits the grid connection state, retracts the propeller to a safe position, and then completely locks all control commands of the yaw system and locks the yaw mechanical structure; at the same time, the unit's background and remote monitoring platform pop-up alarms accurately push fault codes, daily alarm frequency, abnormal trigger time period, yaw operation parameters and other traceability information, and forcibly remind the operation and maintenance personnel to come to the site for investigation; This embodiment implements a manual reset mechanism for the protection technology at this level: Automatic reset is not supported for this fault shutdown. Maintenance personnel must manually reset the fault signal on-site after completing the yaw electrical circuit test, motor performance verification, mechanical transmission wear inspection, and brake mechanism debugging. Only then can the unit restart the yaw function and grid-connected operation, thus preventing the equipment from operating with defects.
[0065] In some implementations, the embodiments of this application may include: the third abnormal scenario includes the cabin skidding determination during the non-yawing phase when the yawing direction command indicates no yawing under yawing standby conditions.
[0066] As illustrated by the example, the direct fault shutdown protection in this application embodiment can be applied to abnormal scenarios including nacelle slip anomalies identified during the non-yaw stationary phase of the unit. This anomaly differs from load anomalies during operation, directly proving the complete failure of the yaw brake mechanism and the malfunction of the locking device. Under wind load, the nacelle experiences autonomous slippage, which is a high-risk safety fault, posing significant safety hazards such as nacelle deflection, transmission failure, tower load exceeding limits, and unit overturning, with no room for error tolerance or retry.
[0067] In this type of scenario, this embodiment can be implemented using instructions adapted to that scenario: Instantaneous fault determination: Once the system identifies abnormal cabin skidding characteristics in a non-yawing state, it will instantly trigger the highest level of fault protection and generate a non-yawing cabin skidding fault signal without delay or frequency accumulation. This embodiment implements an emergency shutdown interlocking protection technology solution at this level: the unit immediately executes the highest priority safety shutdown logic, completes the disconnection from the grid, emergency propeller retraction, and overall shutdown operations within seconds, and permanently locks the yaw control system and hydraulic brake system drive circuit, prohibiting the execution of any yaw unlocking, reset, or action commands, thus completely eliminating safety accidents caused by further nacelle slippage; This embodiment implements a mandatory maintenance and troubleshooting mechanism for the protection technology solution at this level: the fault is an irreversible major fault, the system locks all start-up permissions of the unit, and it cannot be reset and restarted remotely or locally; maintenance personnel must conduct a comprehensive on-site inspection of the yaw brake pad wear, hydraulic braking pressure, locking mechanism clips, displacement sensors, control circuits and other core components, and complete fault repair, performance verification and safety testing before the fault can be manually unlocked and the unit can be restored to operation.
[0068] The wind turbine generator yaw abnormality shutdown protection method disclosed in the embodiments of this application establishes a graded shutdown protection and handling mechanism compared with the prior art. For underspeed and reverse drag abnormalities during yaw operation, a fault-tolerant strategy of stopping yaw and delaying retry is first adopted to avoid frequent and unnecessary shutdowns. By statistically analyzing the alarm frequency within a period, recurring intermittent deterioration problems are identified, and the frequency is upgraded to fault shutdown when it exceeds the standard. For high-risk nacelle slip faults, the generator is shut down directly, taking into account both power generation operation and equipment safety, reducing repeated start-stop and mechanical shock, reducing fatigue wear of yaw motor and reducer, and extending equipment life.
[0069] The technical solution of this application also provides a wind turbine generator set yaw abnormal state detection device and a shutdown protection device, including one or more processing modules, configured to execute the wind turbine generator set yaw abnormal state detection device and shutdown protection method described above, and at least configured to execute the specific implementation of steps S1, S2, S3, and S4.
[0070] To address the aforementioned issues, as one solution, embodiments of this application provide a wind turbine generator, including a controller configured to perform the steps of the method described above.
[0071] Compared with the prior art described above, the beneficial effects of the wind turbine generator yaw anomaly detection method, the wind turbine generator yaw anomaly shutdown protection method, and the wind turbine generator itself are at least reflected in the following aspects: The technical solution of this application obtains a stable yaw speed by filtering the cabin position signal. Combined with the multi-dimensional operation instructions of the crew, it adopts a combination of "time sequence + count" criteria to achieve refined identification of four types of abnormal yaw states, making up for many technical defects in the existing technology, such as only being able to provide offline post-event warnings and incomplete coverage of operating conditions. The technical solution of this application enables accurate identification of high-risk yaw abnormal states of nacelle skidding and timely shutdown to avoid major safety accidents; for yaw underspeed, continuous reverse towing, and intermittent reverse towing, a graded shutdown protection strategy of fault-tolerant retry + frequency upgrade is adopted to reduce unnecessary failure shutdowns and ensure the unit's power generation time. The technical solution proposed in this application can identify intermittent and recurring yaw impact-related abnormal states, triggering alarm shutdowns before irreversible and severe damage to components occurs. The algorithm input is simple, requiring no new hardware and is compatible with the retrofitting of existing units. It reduces the alternating impact loads on the yaw motor, yaw gear, and braking mechanism, reduces component fatigue wear, effectively extends the service life of the yaw system and the entire equipment, and lowers the cost of replacing parts during the later stages of wind farm maintenance.
[0072] Based on the above-mentioned inventive concept, the wind turbine generator set yaw anomaly detection method, wind turbine generator set yaw anomaly shutdown protection method, and wind turbine generator set of various embodiments of this disclosure determine the yaw speed based at least on the filtering of the nacelle position data acquisition; synchronously acquire the unit's internal variables to align with the nacelle azimuth data timestamp; respond to different yaw start commands, perform corresponding discrimination under different unit operating conditions, and prohibit discrimination under other operating conditions under one operating condition. The unit operating conditions include at least the yaw standby operating condition and the yaw execution operating condition. Thus, relying on the nacelle position time sequence data filtering processing, combined with the multi-dimensional operating signals of the unit, multiple types of yaw anomaly state discrimination logic are established, and differentiated hierarchical shutdown and protection strategies are executed. This solves the technical pain point that the traditional single yaw speed threshold detection scheme cannot identify multiple complex yaw anomaly states. Under the premise of ensuring the safe operation of the unit, it reduces the alternating impact load of various components of the yaw system, suppresses abnormal equipment wear, and achieves the goal of extending the service life of the yaw system and the whole unit.
[0073] This application also provides a computer-readable storage medium storing computer-executable instructions thereon. When executed by a processor, the computer-executable instructions mainly implement the above-described method for detecting abnormal yaw states of a wind turbine generator set, including: Yaw speed is determined by filtering based on cabin position data acquisition; Synchronously acquire internal variables of the unit to align with the timestamp of the cabin azimuth data; In response to different yaw start commands, corresponding judgments are performed under different unit operating conditions, and judgments under other operating conditions are prohibited under one operating condition. The unit operating conditions include at least the yaw standby operating condition and the yaw execution operating condition.
[0074] This application also provides a computer-readable storage medium storing computer-executable instructions thereon. When executed by a processor, the computer-executable instructions mainly implement the wind turbine generator yaw anomaly shutdown protection method described above, which, based on the determined yaw anomaly state, includes: In the first abnormal scenario, the current yaw command is latched by the first protection command, and the yaw action lockout is released after a preset delay; In the second abnormal scenario, if the cumulative number of alarms per day is not less than the threshold, the second protection command will lock the yaw control command and the yaw mechanical structure until the first reset condition is met. In the third abnormal scenario, the third protection command, upon identifying abnormal cabin slip characteristics in the non-yaw state, locks the yaw control command and the yaw mechanical structure until the second reset condition is met.
[0075] The above embodiments are merely exemplary embodiments of this application and are not intended to limit this application. The scope of protection of this application is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to this application within its substance and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of this application.
Claims
1. Methods for detecting abnormal yaw conditions in wind turbine generators, including: Yaw speed is determined by filtering based on cabin position data acquisition; Synchronously acquire internal variables of the unit to align with the timestamp of the cabin azimuth data; In response to different yaw start commands, corresponding judgments are performed under different unit operating conditions, and judgments under other operating conditions are prohibited under one operating condition. The unit operating conditions include at least the yaw standby operating condition and the yaw execution operating condition.
2. The method of claim 1, wherein, Based on the filtering of cabin position data, the yaw speed is determined, including: Based on the sampled data from the cabin, the cabin position angle value is obtained through angle conversion; The yaw speed is obtained by combining the cabin position angle value based on the low-pass filter with the position change rate.
3. The method of claim 2, wherein, Synchronously acquire internal variables of the unit, including: synchronously read internal variables of the unit from the real-time database of the main control within the same main control cycle; the variables are at least from generator speed, unit active power, yaw start command, and yaw direction command.
4. The method of claim 3, wherein, The discrimination includes at least: In yaw standby mode and when the yaw direction command indicates no yaw, cabin skidding is determined during the non-yaw phase. Under yaw execution conditions and when the yaw direction command indicates yaw, the following criteria are used to identify continuous underspeed during the yaw phase, continuous anti-towing during the yaw phase, and intermittent anti-towing during the yaw phase.
5. The method of claim 4, wherein, The discrimination also includes: Reverse direction event determination includes determining whether the crew has yawed and towed back based on whether the yaw speed and the left and right yaw direction commands meet the predetermined conditions. The single-event filtering rule states that if the number of consecutive reverse states is not less than a preset number of master control cycles, the reverse event is considered a valid event.
6. A method for protecting a wind turbine generator from yaw anomalies during shutdown, wherein the method includes, based on the determined yaw anomaly state: In the first abnormal scenario, the current yaw command is latched by the first protection command, and the yaw action lockout is released after a preset delay; In the second abnormal scenario, if the cumulative number of alarms per day is not less than the threshold, the second protection command will lock the yaw control command and the yaw mechanical structure until the first reset condition is met. In the third abnormal scenario, the third protection command, upon identifying abnormal cabin slip characteristics in the non-yaw state, locks the yaw control command and the yaw mechanical structure until the second reset condition is met.
7. The method of claim 6, wherein, The first abnormal scenario includes the following when the yaw execution condition is underway and the yaw direction command indicates yaw, and the following are the yaw phases: continuous underspeed, continuous anti-towing, and intermittent anti-towing.
8. The method of claim 6, wherein, The second abnormal scenario includes a single instance of underspeed during yaw execution when the yaw direction command indicates yaw.
9. The method of claim 6, wherein, The third abnormal scenario includes the cabin skidding detection during the non-yawing phase when the yawing direction command indicates no yawing and the yawing standby condition is in effect.
10. A wind turbine generator, including a controller configured to perform the steps of the method as described in any one of claims 1 to 5; and / or The steps of the method as described in any one of claims 6 to 9.
