Wind turbine and method and device for detecting faults thereof

Abstract

Provided are a wind turbine and a fault detection method and device thereof. The fault detection method of the wind turbine comprises: acquiring predetermined operation data of the wind turbine; taking a preset time period as a sliding window, and sequentially calculating a sliding average value and a sliding standard deviation of the predetermined operation data at each time; calculating a sliding cumulative value in the sliding window at each time based on the predetermined operation data and the sliding average value at each time; and determining that the wind turbine has a fault when the sliding cumulative value and the sliding standard deviation satisfy a preset fault condition. The fault detection method of the wind turbine can quickly identify the abnormal divergence of the conversion torque, avoid the risk of tower collapse of the wind turbine caused by long-time divergence of the torque, and has low CPU occupancy and is easy to implement in engineering.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of wind power generation. More particularly, the present disclosure relates to a wind turbine and a fault detection method and device thereof. BACKGROUND

[0002] A wind turbine generates electricity by converting wind energy into electrical energy, and the process of generating electricity requires a converter and a generator to convert kinetic energy into electrical energy. The generator and the converter control the torque of the wind turbine, and the abnormal protection of the torque loop is very important. In the case of abnormal converter and generator, torque divergence will occur, and low speed high torque will also cause torque divergence, resulting in large range fluctuation of wind turbine torque, large vibration, and serious impact on the safety of the wind turbine.

[0003] The existing fan protection strategy cannot identify the phenomenon of torque divergence. The fan running in the power generation state will have torque divergence for a long time, and the acceleration fault will not be protected to shut down. Even if the acceleration exceeds the vibration threshold, the acceleration fault will be reported, but the wind turbine has been in serious torque divergence for a long time, and the safety of the wind turbine cannot be protected. SUMMARY

[0004] An exemplary embodiment of the present disclosure provides a wind turbine and a fault detection method and device thereof to detect the fault of the wind turbine in time.

[0005] According to an exemplary embodiment of the present disclosure, a fault detection method of a wind turbine is provided, comprising: acquiring predetermined operation data of the wind turbine; taking a preset time period as a sliding window, and sequentially calculating a sliding average value and a sliding standard deviation of the predetermined operation data at each time; calculating a sliding cumulative value in the sliding window at each time based on the predetermined operation data and the sliding average value at each time; and determining that the wind turbine has a fault when the sliding cumulative value and the sliding standard deviation meet a preset fault condition.

[0006] Optionally, the predetermined operation data can include a converter feedback torque of the wind turbine.

[0007] Optionally, the step of sequentially calculating the sliding average value and the sliding standard deviation of the predetermined operation data at each time can include: calculating an average value of the converter feedback torque in the preset time period before each time as the sliding average value at each time; and calculating a standard deviation of the converter feedback torque in the preset time period before each time as the sliding standard deviation at each time.

[0008] Optionally, the step of calculating the sliding cumulative value in the sliding window of each time based on the predetermined operating data and the sliding average value of each time can comprise: calculating, for each time, a difference value between the converter feedback torque and the sliding average value; comparing the difference value of each time with a torque minimum change threshold value respectively; determining a state value of the torque change of each time based on the comparison result; and calculating the sliding cumulative value in the sliding window of each time based on the state value of the torque change of each time.

[0009] Optionally, the step of determining the state value of the torque change of each time based on the comparison result can comprise: determining the state value of the torque change as a first state value when the difference value is greater than a torque maximum change threshold value; determining the state value of the torque change as a second state value when the difference value is less than a torque minimum change threshold value; and determining the state value of the torque change as a third state value when the difference value is neither greater than the torque maximum change threshold value nor less than the torque minimum change threshold value.

[0010] Optionally, the step of calculating the sliding cumulative value in the sliding window of each time based on the state value of the torque change of each time can comprise: calculating, for each state value of the torque change in the sliding window of each time, a product of two state values of the torque change in sequence; and recording a number of times that the product meets a preset condition as the sliding cumulative value in the sliding window of each time.

[0011] Optionally, when the first state value is 1, the second state value is -1, and the third state value is 0, the preset condition can be that the product is less than 0.

[0012] Optionally, the preset fault condition can comprise: a number of times that, in the sliding window, the time at which the sliding cumulative value is greater than a preset cumulative threshold value and the sliding standard deviation is greater than a preset standard deviation threshold value appears continuously is greater than a first threshold value.

[0013] Optionally, the fault detection method can further comprise: performing a fault reset operation when the sliding cumulative value and the sliding standard deviation in the sliding window meet a preset fault reset condition.

[0014] Optionally, the preset fault reset condition can comprise: a number of times that, in the sliding window, the time at which the sliding cumulative value is less than or equal to a preset cumulative threshold value and the sliding standard deviation is less than or equal to a preset standard deviation threshold value appears continuously is greater than a second threshold value.

[0015] Optionally, the fault detection method can further comprise: performing at least one of a power limiting operation, a pitch angle limiting operation, and a shutdown operation on the wind turbine when it is determined that a fault is detected.

[0016] According to an example embodiment of the present disclosure, a fault detection device of a wind turbine is provided, comprising: a data acquisition unit configured to acquire predetermined operation data of the wind turbine; a data calculation unit configured to calculate, in a preset time period as a sliding window, a sliding average value and a sliding standard deviation of the predetermined operation data at each time point in sequence; an accumulated value calculation unit configured to calculate a sliding accumulated value in the sliding window at each time point based on the predetermined operation data and the sliding average value at each time point; and a fault determination unit configured to determine that the wind turbine has a fault when the sliding accumulated value and the sliding standard deviation satisfy a preset fault condition.

[0017] Optionally, the predetermined operation data can include a converter feedback torque of the wind turbine.

[0018] Optionally, the data calculation unit can be configured to calculate, as the sliding average value at each time point, an average value of the converter feedback torque in the preset time period before each time point; and calculate, as the sliding standard deviation at each time point, a standard deviation of the converter feedback torque in the preset time period before each time point.

[0019] Optionally, the accumulated value calculation unit can be configured to calculate, for each time point, a difference value between the converter feedback torque and the sliding average value; compare the difference value at each time point with a torque minimum change threshold value respectively; determine a state value of torque change at each time point based on a comparison result; and calculate the sliding accumulated value in the sliding window at each time point based on the state value of torque change at each time point.

[0020] Optionally, the accumulated value calculation unit can be configured to determine the state value of torque change as a first state value when the difference value is greater than a torque maximum change threshold value; determine the state value of torque change as a second state value when the difference value is less than the torque minimum change threshold value; and determine the state value of torque change as a third state value when the difference value is neither greater than the torque maximum change threshold value nor less than the torque minimum change threshold value.

[0021] Optionally, the accumulated value calculation unit can be configured to calculate, for each state value of torque change in the sliding window at each time point, a product of two adjacent state values of torque change in sequence; and record a number of times that the product satisfies a preset condition as the sliding accumulated value in the sliding window at each time point.

[0022] Optionally, when the first state value is 1, the second state value is -1, and the third state value is 0, the preset condition can be that the product is less than 0.

[0023] Optionally, the preset fault condition can include that, in the sliding window, a number of times that a time point at which the sliding accumulated value is greater than a preset accumulated threshold value and the sliding standard deviation is greater than a preset standard deviation threshold value appears continuously is greater than a first threshold value.

[0024] Optionally, the fault detection apparatus can further comprise a fault reset unit configured to perform a fault reset operation when the sliding cumulative value and the sliding standard deviation in the sliding window satisfy a preset fault reset condition.

[0025] Optionally, the preset fault reset condition can comprise that the number of times of continuous occurrence of the moment when the sliding cumulative value is less than or equal to a preset cumulative threshold value and the sliding standard deviation is less than or equal to a preset standard deviation threshold value in the sliding window is greater than a second threshold value.

[0026] Optionally, the fault detection apparatus can further comprise an execution unit configured to perform at least one of power limiting, pitch angle limiting and shutdown operation on the wind turbine when it is determined that a fault is detected.

[0027] According to an example embodiment of the present disclosure, a wind turbine is provided, comprising: an impeller installed on a rotating main shaft; a generator directly connected to the rotating main shaft; a converter connected to a power grid after processing the electric energy output by the generator; and a fault detection apparatus. The fault detection apparatus comprises: a data acquisition unit configured to acquire predetermined operating data of the wind turbine; a data calculation unit configured to calculate, in a preset time period as a sliding window, a sliding average value and a sliding standard deviation of the predetermined operating data at each moment in sequence; a cumulative value calculation unit configured to calculate a sliding cumulative value in the sliding window at each moment based on the predetermined operating data and the sliding average value at each moment; and a fault determination unit configured to determine that the wind turbine has a fault when the sliding cumulative value and the sliding standard deviation satisfy a preset fault condition.

[0028] According to an example embodiment of the present disclosure, a computer readable storage medium having a computer program stored thereon is provided, when the computer program is executed by a processor, a fault detection method of a wind turbine according to an example embodiment of the present disclosure is implemented.

[0029] According to an example embodiment of the present disclosure, a computing apparatus is provided, comprising: at least one processor; at least one memory having a computer program stored thereon, when the computer program is executed by the at least one processor, a fault detection method of a wind turbine according to an example embodiment of the present disclosure is implemented.

[0030] According to an example embodiment of the present disclosure, a computer program product is provided, instructions in the computer program product can be executed by a processor of a computer device to complete a fault detection method of a wind turbine according to an example embodiment of the present disclosure.

[0031] The wind turbine and the fault detection method and device thereof according to the example embodiments of the present disclosure can quickly identify the abnormal divergence of the converter torque, avoid the risk of tower collapse of the wind turbine due to long-time divergence of the torque, by first acquiring predetermined operation data of the wind turbine, and taking a preset time period as a sliding window, sequentially calculating a sliding average value and a sliding standard deviation of the predetermined operation data at each time point, then calculating a sliding cumulative value in the sliding window at each time point based on the predetermined operation data and the sliding average value at each time point, and determining that the wind turbine has a fault when the sliding cumulative value and the sliding standard deviation meet a preset fault condition. In addition, the CPU occupancy is low when the fault detection method of the wind turbine according to the example embodiments of the present disclosure is implemented, and it is easy to implement in engineering.

[0032] Additional aspects and / or advantages of the general inventive concept will be set forth in part in the description which follows, and in part will be obvious from the description, or can be learned by practice of the general inventive concept. BRIEF DESCRIPTION OF DRAWINGS

[0033] The above and other objects and features of the example embodiments of the present disclosure will become more apparent from the following description of the example embodiments of the present disclosure given in conjunction with the accompanying drawings, in which:

[0034] Figure 1 A flowchart showing a fault detection method of a wind turbine according to an example embodiment of the present disclosure;

[0035] Figure 2 An example flowchart showing data acquisition and processing before fault detection of a wind turbine according to an example embodiment of the present disclosure;

[0036] Figure 3 An example flowchart showing a fault detection method of a wind turbine according to an example embodiment of the present disclosure;

[0037] Figure 4 A logic diagram showing a fault shutdown according to an example embodiment of the present disclosure;

[0038] Figure 5 A block diagram showing a fault detection device of a wind turbine according to an example embodiment of the present disclosure;

[0039] Figure 6 A system diagram showing a wind turbine according to an example embodiment of the present disclosure; and

[0040] Figure 7 A schematic diagram showing a computing device according to an example embodiment of the present disclosure. DETAILED DESCRIPTION

[0041] Reference will now be made in detail embodiments of the present disclosure, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to like elements throughout. The embodiments will be explained by referring to the drawings in order to explain the present disclosure.

[0042] Figure 1 A flow chart of a fault detection method of a wind turbine according to an exemplary embodiment of the present disclosure is shown.

[0043] Referring to Figure 1 In step S101, predetermined operation data of the wind turbine is acquired.

[0044] In an exemplary embodiment of the present disclosure, the predetermined operation data can include converter feedback torque of the wind turbine.

[0045] Specifically, real-time data and initialization parameter information of each wind turbine in the wind farm can be acquired. The real-time data of the wind turbine can include, for example, but not limited to, working state of the wind turbine, converter feedback torque, feedback power and the like. The initialization parameter information of the wind turbine can include, for example, but not limited to, threshold value of standard deviation of site torque fluctuation of the wind turbine, fault delay trigger time, fault delay reset time, minimum torque crossing times and the like.

[0046] In step S102, a preset time period is taken as a sliding window, and the sliding average value and the sliding standard deviation of the predetermined operation data at each time are calculated in turn.

[0047] As an example, the preset time period can be 300 ms. Here, the preset time period can also be other time length, which is not limited in the present disclosure. The 300 ms sliding average value and the 300 ms interactive standard deviation can be calculated for the converter feedback torque.

[0048] In an exemplary embodiment of the present disclosure, when the sliding average value and the sliding standard deviation of the predetermined operation data at each time are calculated in turn, the average value of the converter feedback torque within the preset time period before each time can be calculated as the sliding average value at each time, and the standard deviation of the converter feedback torque within the preset time period before each time can be calculated as the sliding standard deviation at each time. That is, the average value of the converter feedback torque within 300 ms before each time can be calculated as the sliding average value at each time, and the standard deviation of the converter feedback torque within 300 ms before each time can be calculated as the sliding standard deviation at each time.

[0049] In step S103, the sliding cumulative value in the sliding window at each time is calculated based on the predetermined operation data and the sliding average value at each time.

[0050] In the exemplary embodiments of the present disclosure, when the sliding cumulative value in the sliding window of each time instant is calculated based on the predetermined operating data and the sliding average value of each time instant, the difference between the converter feedback torque and the sliding average value can be calculated first for each time instant, and then the difference of each time instant is compared with the torque minimum change threshold value respectively, the state value of the torque change of each time instant is determined based on the comparison result, and the sliding cumulative value in the sliding window of each time instant is calculated based on the state value of the torque change of each time instant.

[0051] In the exemplary embodiments of the present disclosure, when the state value of the torque change of each time instant is determined based on the comparison result, the state value of the torque change can be determined as a first state value when the difference is greater than the torque maximum change threshold value, as a second state value when the difference is less than the torque minimum change threshold value, and as a third state value when the difference is neither greater than the torque maximum change threshold value nor less than the torque minimum change threshold value.

[0052] For example, the first state value can be 1, the second state value can be -1, and the third state value can be 0.

[0053] In the exemplary embodiments of the present disclosure, when the sliding cumulative value in the sliding window of each time instant is calculated based on the state value of the torque change of each time instant, the product of the state values of the torque change of the two adjacent time instants in the sliding window of each time instant can be calculated in sequence first, and then the number of times that the product of the state values of the torque change of the two adjacent time instants meets a preset condition is recorded as the sliding cumulative value in the sliding window of each time instant.

[0054] In the exemplary embodiments of the present disclosure, when the first state value is 1, the second state value is -1, and the third state value is 0, the preset condition can be that the product is less than 0.

[0055] For example, for each time instant, the difference between the converter feedback torque and the sliding average value is calculated, the difference is compared with the torque minimum change threshold value to determine the three states of the torque change. Here, the state values of the three states can be, for example, but not limited to, 1, 0, and -1. If the product of the state values of the two adjacent time instants is less than 0, the number of times of the torque change crossing is increased by 1; if the product of the state values of the two adjacent time instants is not less than 0, the number of times of the torque change crossing is increased by 0. The number of times of crossing within 300 ms is taken as the sliding cumulative value in the sliding window of the time instant. Other judgment methods such as fast Fourier transform can also be used to determine the number of times of the torque crossing, which is not limited in the present disclosure.

[0056] In step S104, when the sliding cumulative value and the sliding standard deviation meet the preset fault condition, it is determined that the wind turbine has a fault.

[0057] In the example embodiments of the present disclosure, the preset fault condition can include that, in the sliding window, the number of times that the sliding cumulative value is greater than the preset cumulative threshold value and the sliding standard deviation is greater than the preset standard deviation threshold value continuously occurs at the same time is greater than a first threshold value. Based on the preset fault condition, the occurring torque divergence anomaly of the converter can be quickly identified, and the risk of tower collapse of the wind turbine caused by long-time divergence of the torque can be avoided.

[0058] In the example embodiments of the present disclosure, when the sliding cumulative value and the sliding standard deviation in the sliding window satisfy the preset fault reset condition, a fault reset operation can be performed.

[0059] In the example embodiments of the present disclosure, the preset fault reset condition can include that, in the sliding window, the number of times that the sliding cumulative value is less than or equal to the preset cumulative threshold value and the sliding standard deviation is less than or equal to the preset standard deviation threshold value continuously occurs at the same time is greater than a second threshold value.

[0060] As an example, when the number of times of passing through is greater than the torque passing through number setting value, and the standard deviation of the torque is greater than the torque standard deviation setting value, if the duration is greater than a preset time, the torque divergence fault is triggered. When the number of times of passing through is not greater than the torque passing through number setting value, or the standard deviation of the torque is not greater than the torque standard deviation setting value, if the duration is greater than a preset time, the torque divergence fault is reset.

[0061] In the example embodiments of the present disclosure, when it is determined that a fault is detected, at least one of the power limiting, the pitch angle limiting, and the shutdown operation can be performed on the wind turbine.

[0062] As an example, after the torque divergence fault is triggered, if the current power is greater than a torque divergence minimum power trigger setting threshold value, the wind turbine is protected by shutdown.

[0063] In addition, when the fault detection method of the wind turbine according to the example embodiments of the present disclosure is run, the CPU occupancy rate is low, and it is easy to implement engineering.

[0064] Figure 2 An example flowchart of data collection and processing before fault detection of the wind turbine according to the example embodiments of the present disclosure is shown. Figure 3 An example flowchart of the fault detection method of the wind turbine according to the example embodiments of the present disclosure is shown. Figure 4 A fault shutdown logic diagram according to the example embodiments of the present disclosure is shown.

[0065] As Figure 2As shown, firstly, real-time data (e.g., unit operating status, converter feedback torque, feedback power) and initialization parameter information (e.g., wind turbine position torque fluctuation standard deviation threshold, fault delay trigger time T1, fault delay reset time T2, minimum torque crossover count N, etc.) are acquired for each unit in the wind farm. Then, a 300ms moving average (Torque_Mean) and a 300ms interaction standard deviation (Torque_Std) can be calculated for the converter feedback torque (Torque_Con).

[0066] like Figure 3 As shown, firstly, the difference between the converter feedback torque (Torque_Con) and the 300ms sliding average value (Torque_Mean) is calculated, and this difference is compared with the maximum torque change threshold (PA_rTorqueDiffErr) and the minimum torque change threshold (-PA_rTorqueDiffErr) to determine the three states of torque change (1, 0, -1). The determination is based on multiplying the torque change states of two consecutive torques; a value less than 0 is 1, otherwise it is 0. A 300ms sliding cumulative value is calculated for the number of torque change crossovers, recording the number of crossovers within 300ms. When the number of crossovers exceeds the parameter torque crossover count setting (PA_crossNum), and the standard deviation of the torque is greater than the parameter torque standard deviation setting (PA_TorqueStd), if the duration is T1, a torque divergence fault is triggered. When the number of crossovers is not greater than the parameter torque crossover count setting (PA_crossNum), or the standard deviation of the torque is not greater than the parameter torque standard deviation setting (PA_TorqueStd), if the duration is T2, the torque divergence fault is reset.

[0067] like Figure 4 As shown, after a fault is triggered, if the current power is greater than the minimum power trigger threshold for torque divergence (Pa_PowerMin), the unit will perform a shutdown protection.

[0068] Furthermore, according to exemplary embodiments of the present disclosure, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed, implements a fault detection method for a wind turbine according to exemplary embodiments of the present disclosure.

[0069] In the exemplary embodiment of the present disclosure, the computer readable storage medium can carry one or more programs, which when executed can implement the following steps: obtaining predetermined operation data of the wind turbine; taking a preset time period as a sliding window, sequentially calculating a sliding average value and a sliding standard deviation of the predetermined operation data at each time; calculating a sliding cumulative value in the sliding window at each time based on the predetermined operation data and the sliding average value at each time; and determining that the wind turbine has a fault when the sliding cumulative value and the sliding standard deviation satisfy a preset fault condition.

[0070] The computer readable storage medium may, for example, be, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In an embodiment of the present disclosure, the computer readable storage medium can be any tangible medium that contains or stores a computer program that can be used by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium contained in the computer program can be transmitted by any suitable medium, including but not limited to: a wire, an optical fiber, an RF (radio frequency), etc., or any suitable combination of the foregoing. The computer readable storage medium can be contained in any device; it can also exist separately without being assembled into the device.

[0071] In addition, according to the exemplary embodiment of the present disclosure, a computer program product is also provided, in which the instructions can be executed by the processor of the computer device to complete the method for detecting the fault of the wind turbine according to the exemplary embodiment of the present disclosure.

[0072] The above has been described in conjunction with Figures 1 to 4 The method for detecting the fault of the wind turbine according to the exemplary embodiment of the present disclosure is described. In the following, the wind turbine, the fault detection device of the wind turbine and the units thereof according to the exemplary embodiment of the present disclosure will be described with reference to Figure 5 and FIG.

[0073] Figure 5 A block diagram of the fault detection device of the wind turbine according to the exemplary embodiment of the present disclosure is shown.

[0074] With reference to Figure 5 , the fault detection device of the wind turbine comprises a data acquisition unit 51, a data calculation unit 52, a cumulative value calculation unit 53 and a fault determination unit 54.

[0075] The data acquisition unit 51 is configured to acquire predetermined operation data of the wind turbine generator.

[0076] In the example embodiment of the present disclosure, the predetermined operation data includes converter feedback torque of the wind turbine generator.

[0077] The data calculation unit 52 is configured to calculate, as a sliding average value and a sliding standard deviation of the predetermined operation data at each time point, a sliding window with a preset time period.

[0078] In the example embodiment of the present disclosure, the data calculation unit 52 can be configured to calculate, as the sliding average value at each time point, an average value of the converter feedback torque within the preset time period before each time point, and calculate, as the sliding standard deviation at each time point, a standard deviation of the converter feedback torque within the preset time period before each time point.

[0079] The cumulative value calculation unit 53 is configured to calculate, based on the predetermined operation data and the sliding average value at each time point, a sliding cumulative value in the sliding window at each time point.

[0080] In the example embodiment of the present disclosure, the cumulative value calculation unit 53 can be configured to, for each time point, calculate a difference value between the converter feedback torque and the sliding average value, compare the difference value at each time point with a torque minimum change threshold value respectively, determine a state value of torque change at each time point based on the comparison result, and calculate the sliding cumulative value in the sliding window at each time point based on the state value of torque change at each time point.

[0081] In the example embodiment of the present disclosure, the cumulative value calculation unit 53 can be configured to determine the state value of torque change as a first state value when the difference value is greater than a torque maximum change threshold value, determine the state value of torque change as a second state value when the difference value is less than a torque minimum change threshold value, and determine the state value of torque change as a third state value when the difference value is neither greater than the torque maximum change threshold value nor less than the torque minimum change threshold value.

[0082] In the example embodiment of the present disclosure, the cumulative value calculation unit 53 can be configured to, for each state value of torque change in the sliding window at each time point, calculate, in sequence, a product of two state values of torque change in front and behind, and record a number of times that the product meets a preset condition as the sliding cumulative value in the sliding window at each time point.

[0083] In the example embodiment of the present disclosure, when the first state value is 1, the second state value is -1, and the third state value is 0, the preset condition can be that the product is less than 0.

[0084] The fault determination unit 54 is configured to determine that the wind turbine is in fault when the sliding cumulative value and the sliding standard deviation satisfy a preset fault condition.

[0085] In the exemplary embodiments of the present disclosure, the preset fault condition can include that the number of times that the sliding cumulative value is greater than the preset cumulative threshold value and the sliding standard deviation is greater than the preset standard deviation threshold value in the sliding window continuously occurs for more than a first threshold value.

[0086] In the exemplary embodiments of the present disclosure, the fault detection device can further include a fault reset unit (not shown) configured to perform a fault reset operation when the sliding cumulative value and the sliding standard deviation in the sliding window satisfy a preset fault reset condition.

[0087] In the exemplary embodiments of the present disclosure, the preset fault reset condition can include that the number of times that the sliding cumulative value is less than or equal to the preset cumulative threshold value and the sliding standard deviation is less than or equal to the preset standard deviation threshold value in the sliding window continuously occurs for more than a second threshold value.

[0088] In the exemplary embodiments of the present disclosure, the fault detection device can further include an execution unit (not shown) configured to perform at least one of power limiting, pitch angle limiting, and shutdown operation on the wind turbine when it is determined that the fault is detected.

[0089] Figure 6 A system schematic diagram of a wind turbine is provided for the exemplary embodiments of the present disclosure. Here, the wind turbine is a permanent magnet direct drive type. The wind turbine 600 includes an impeller 602 mounted on a rotating main shaft, and the pitch angle of the blades in the impeller 602 is controlled by a pitch signal. A generator 604 is directly connected to the rotating main shaft, and the electrical energy output by the generator 604 is output after passing through a converter 610 and then connected to a power grid after passing through a transformer 606. A control device 620 can collect the pitch angle of the impeller 602, and control the impeller speed and the generator speed by setting the value of the blade pitch angle; the control device 620 can also collect and control the converter torque through the converter 610. A fault detection device 630 identifies the torque divergence fault of the wind turbine by detecting the converter torque. The control device 620 and the fault detection device 630 can be set as a whole, or can be set separately.

[0090] Figure 6 A wind turbine provided by an embodiment of the present application is shown, which includes a fault detection device 630 including a data acquisition unit 51, a data calculation unit 52, a cumulative value calculation unit 53, and a fault determination unit 54 as shown in Figure 5

[0091] In an embodiment, Figure 6 ​The computer program stored in the fault detection device 630 shown is executed by the fault detection device 630 to implement the fault detection method of the wind turbine according to the example embodiments of the present disclosure.

[0092] The wind turbine and the fault detection method thereof according to the example embodiments of the present disclosure have been described above in combination with Figure 5 and Figure 6 The wind turbine and the fault detection device thereof according to the example embodiments of the present disclosure have been described. Next, the computing device according to the example embodiments of the present disclosure will be described in combination with Figure 7 The computing device according to the example embodiments of the present disclosure will be described.

[0093] Figure 7 A schematic diagram of the computing device according to the example embodiments of the present disclosure is shown.

[0094] Referring to Figure 7 , the computing device 7 according to the example embodiments of the present disclosure includes a memory 71 and a processor 72, and the memory 71 stores a computer program which, when executed by the processor 72, implements the fault detection method of the wind turbine according to the example embodiments of the present disclosure.

[0095] In the example embodiments of the present disclosure, when the computer program is executed by the processor 72, the following steps can be implemented: obtaining predetermined operation data of the wind turbine; taking a preset time period as a sliding window, and sequentially calculating a sliding average value and a sliding standard deviation of the predetermined operation data at each time; calculating a sliding cumulative value in the sliding window at each time based on the predetermined operation data and the sliding average value at each time; and determining that the wind turbine has a fault when the sliding cumulative value and the sliding standard deviation satisfy a preset fault condition.

[0096] The wind turbine and the fault detection method, device thereof according to the example embodiments of the present disclosure have been described above in combination with Figures 1 to 7 However, it should be understood that: Figure 5 The fault detection device of the wind turbine and the units thereof shown in Figure 7 The computing device shown in

[0097] The wind turbine and the fault detection method and device thereof according to the example embodiments of the present disclosure, by first acquiring predetermined operation data of the wind turbine, and taking a preset time period as a sliding window, sequentially calculating a sliding average value and a sliding standard deviation of the predetermined operation data at each time point, then calculating a sliding cumulative value in the sliding window at each time point based on the predetermined operation data and the sliding average value at each time point, when the sliding cumulative value and the sliding standard deviation satisfy a preset fault condition, determining that the wind turbine has a fault, thereby quickly identifying the abnormal divergence of the conversion torque that occurs, avoiding the risk of tower collapse of the unit due to long-time divergence of the torque. In addition, when the fault detection method of the wind turbine according to the example embodiments of the present disclosure is running, the CPU occupancy is low, and it is easy to implement engineering.

[0098] Although the present disclosure has been particularly shown and described with reference to example embodiments thereof, it will be understood by those skilled in the art that various changes in form and details can be made therein without departing from the spirit and scope of the present disclosure as defined by the appended claims.

Claims

1. A method for detecting a fault of a wind turbine, comprising: obtaining predetermined operating data of the wind turbine; calculating a sliding average value and a sliding standard deviation of the predetermined operating data at each time point in a sliding window with a preset time period; calculating a sliding cumulative value in the sliding window at each time point based on the predetermined operating data and the sliding average value at each time point; determining that the wind turbine has a fault when the sliding cumulative value and the sliding standard deviation satisfy a preset fault condition.

2. The fault detection method of claim 1, wherein, The predetermined operating data comprises a converter feedback torque of the wind turbine.

3. The fault detection method of claim 2, wherein, The step of calculating the sliding average value and the sliding standard deviation of the predetermined operating data at each time point in sequence comprises: respectively calculating an average value of the converter feedback torque in the preset time period before each time point as the sliding average value at each time point; respectively calculating a standard deviation of the converter feedback torque in the preset time period before each time point as the sliding standard deviation at each time point.

4. The fault detection method of claim 2, wherein, The step of calculating the sliding cumulative value in the sliding window at each time point based on the predetermined operating data and the sliding average value at each time point comprises: calculating a difference value between the converter feedback torque and the sliding average value for each time point; respectively comparing the difference value at each time point with a torque minimum change threshold value; determining a state value of torque change at each time point based on a comparison result; calculating the sliding cumulative value in the sliding window at each time point based on the state value of torque change at each time point.

5. The fault detection method of claim 4, wherein, The step of determining the state value of torque change at each time point based on the comparison result comprises: determining the state value of torque change as a first state value when the difference value is greater than a torque maximum change threshold value; determining the state value of torque change as a second state value when the difference value is less than a torque minimum change threshold value; determining the state value of torque change as a third state value when the difference value is neither greater than the torque maximum change threshold value nor less than the torque minimum change threshold value.

6. The fault detection method of claim 5, wherein, The step of calculating the sliding cumulative value in the sliding window at each time point based on the state value of torque change at each time point comprises: for each state value of torque change in the sliding window at each time point, calculating a product of two state values of torque change in sequence; recording a number of times that the product satisfies a preset condition as the sliding cumulative value in the sliding window at each time point.

7. The fault detection method of claim 6, wherein, When the first state value is 1, the second state value is -1, and the third state value is 0, the preset condition is that the product is less than 0.

8. The fault detection method of claim 1, wherein, The preset fault condition comprises that, in the sliding window, a number of times that a time point at which the sliding cumulative value is greater than a preset cumulative threshold value and the sliding standard deviation is greater than a preset standard deviation threshold value appears continuously is greater than a first threshold value. 9.The method of claim 1, further comprising: performing a fault reset operation when the sliding cumulative value and the sliding standard deviation in the sliding window satisfy a preset fault reset condition.

10. The fault detection method of claim 9, wherein, The preset fault reset condition comprises that, in the sliding window, a number of times that a time point at which the sliding cumulative value is less than or equal to the preset cumulative threshold value and the sliding standard deviation is less than or equal to the preset standard deviation threshold value appears continuously is greater than a second threshold value. 11.The method of claim 1, further comprising: When it is determined that a fault is detected, at least one of power limiting, pitch angle limiting, and shutdown operation is performed on the wind turbine.

12. A fault detection device of a wind turbine, comprising: a data acquisition unit configured to acquire predetermined operating data of the wind turbine; a data calculation unit configured to calculate, in a preset time period as a sliding window, a sliding average value and a sliding standard deviation of the predetermined operating data at each time point in sequence; a cumulative value calculation unit configured to calculate, based on the predetermined operating data and the sliding average value at each time point, a sliding cumulative value in the sliding window at each time point; and a fault determination unit configured to determine that the wind turbine has a fault when the sliding cumulative value and the sliding standard deviation satisfy a preset fault condition. The predetermined operating data comprises a converter feedback torque of the wind turbine.

13. The fault detection apparatus of claim 12, wherein, The data calculation unit is configured to:

14. The fault detection apparatus of claim 13, wherein, calculate, as the sliding average value at each time point, an average value of the converter feedback torque in the preset time period before each time point, respectively; and calculate, as the sliding standard deviation at each time point, a standard deviation of the converter feedback torque in the preset time period before each time point, respectively. The cumulative value calculation unit is configured to:

15. The fault detection apparatus of claim 13, wherein, calculate, for each time point, a difference value between the converter feedback torque and the sliding average value; compare, for each time point, the difference value with a torque minimum change threshold value, respectively; determine, based on a comparison result, a state value of torque change at each time point; and calculate, based on the state value of torque change at each time point, the sliding cumulative value in the sliding window at each time point. The cumulative value calculation unit is configured to:

16. The fault detection apparatus of claim 15, wherein, determine the state value of torque change as a first state value when the difference value is greater than a torque maximum change threshold value; determine the state value of torque change as a second state value when the difference value is less than the torque minimum change threshold value; and determine the state value of torque change as a third state value when the difference value is neither greater than the torque maximum change threshold value nor less than the torque minimum change threshold value. The cumulative value calculation unit is configured to:

17. The fault detection apparatus of claim 16, wherein, calculate, for each state value of torque change in the sliding window at each time point, a product of two state values of torque change in sequence; and record a number of times that the product satisfies a preset condition as the sliding cumulative value in the sliding window at each time point. When the first state value is 1, the second state value is -1, and the third state value is 0, the preset condition is that the product is less than 0.

18. The fault detection apparatus of claim 17, wherein, The preset fault condition comprises that, in the sliding window, a number of times that a time point at which the sliding cumulative value is greater than a preset cumulative threshold value and the sliding standard deviation is greater than a preset standard deviation threshold value appears continuously is greater than a first threshold value.

19. The fault detection apparatus of claim 12, wherein, 20. The fault detection device of claim 12, further comprising: a fault reset unit configured to perform a fault reset operation when the sliding cumulative value and the sliding standard deviation in the sliding window satisfy a preset fault reset condition. The preset fault reset condition comprises that, in the sliding window, a number of times that a time point at which the sliding cumulative value is less than or equal to the preset cumulative threshold value and the sliding standard deviation is less than or equal to the preset standard deviation threshold value appears continuously is greater than a second threshold value.

21. The fault detection apparatus of claim 20, wherein, 22. The fault detection device of claim 12, further comprising: ​ The execution unit is configured to perform at least one of power limiting, pitch angle limiting and shutdown operation on the wind turbine when it is determined that the fault is detected.

23. A computer readable storage medium storing a computer program, wherein, The computer program, when executed by a processor, implements the wind turbine fault detection method of any one of claims 1-11.

24. A computing device comprising: at least one processor; at least one memory storing a computer program which, when executed by the at least one processor, implements the wind turbine fault detection method of any one of claims 1-11.

25. A wind turbine generator comprising: comprising: a rotor mounted on a rotating main shaft; a generator directly connected to the rotating main shaft; a converter connected to a power grid after processing the power output by the generator; and a fault detection device comprising: a data acquisition unit configured to acquire predetermined operating data of the wind turbine; a data calculation unit configured to calculate, in a preset time period as a sliding window, a sliding average value and a sliding standard deviation of the predetermined operating data at each time point in sequence; a cumulative value calculation unit configured to calculate a sliding cumulative value in the sliding window at each time point based on the predetermined operating data and the sliding average value at each time point; and a fault determination unit configured to determine that the wind turbine has a fault when the sliding cumulative value and the sliding standard deviation satisfy a preset fault condition. ​

Images