CN118088396A

CN118088396A - Abnormality detection method for pitch system, storage medium, and computing device

Info

Publication number: CN118088396A
Application number: CN202410449393.8A
Authority: CN
Inventors: 马磊; 王玲玲; 杜新
Original assignee: Jinfeng Technology Co ltd
Current assignee: Jinfeng Technology Co ltd
Priority date: 2024-04-15
Filing date: 2024-04-15
Publication date: 2024-05-28

Abstract

An anomaly detection method, a storage medium, and a computing device for a pitch system are disclosed. The abnormality detection method includes: acquiring operation data of each pitch motor of the pitch system during the pitch operation of the pitch system; accumulating and summing the operation data of each variable pitch motor at each sampling moment in a preset time period to obtain the data and value of each sampling moment; data consistency judgment is carried out on the data and the value of each sampling moment; determining that the pitch system is operating normally in response to the consistency of the data and the values at each sampling moment; and determining that an anomaly exists in the pitch system in response to the data and values at each sampling instant not having consistency. Therefore, the problems that the SCADA operation data sampling period is long, continuous data does not exist, effective monitoring cannot be achieved, and the problems that the operation state of a wind turbine generator set cannot be accurately analyzed in the prior art such as spectrum analysis, jump detection and amplitude detection can be achieved.

Description

Abnormality detection method for pitch system, storage medium, and computing device

Technical Field

The present disclosure relates generally to the field of wind power generation technology, and more particularly, to an anomaly detection method, a storage medium, and a computing device for a pitch system.

Background

With the progress of science and technology, the electromechanical equipment of modern enterprises is comprehensively and rapidly developed towards the directions of large-scale, continuous and automatic, and the objective requirements of reducing the production cost, improving the productivity, saving the energy, tightening the labor force, reducing the waste products and improving the product quality are met. Whether various machines and equipment run stably directly influences the economic benefit of enterprises, and some critical equipment even plays a role in determining the fates of the enterprises. Therefore, how to avoid equipment failure, especially catastrophic failure, has been a very important issue. Since the occurrence of an accident has long been unpredictable, people have to take the following two countermeasures.

Firstly, the equipment is overhauled after being thoroughly damaged. The method has great economic loss, because the equipment is maintained after being thoroughly damaged, expensive maintenance cost is often required, and the 'insufficient maintenance' is often caused, so that catastrophic damage is caused, the equipment is required to be replaced when the equipment is light, and casualties are caused when the equipment is heavy.

And secondly, periodically overhauling equipment. This method is somewhat planned and preventive, but has the disadvantage of being very blind, and if the equipment fails, the shut down can result in significant economic losses. This approach tends to result in "overstock", resulting in increased costs and even wasted devices.

Therefore, reasonable maintenance modes are predictive, namely, hidden dangers can be monitored in early stage of equipment failure, and forecast in advance so as to take measures timely and reasonably. The maintenance mode can timely and pointedly maintain the equipment, not only can ensure that the equipment is in a good running state, but also can fully utilize the service life of parts, and effectively avoid 'insufficient maintenance' and 'excessive maintenance'.

In terms of the wind power generation industry, the method for carrying out abnormal early warning on main devices of the wind generating set has great practical significance on safe operation of the wind generating set and maintenance cost reduction. In general, data collection is a prerequisite for implementing abnormal (fault) early warning, and is the basis for implementing abnormal (fault) early warning. However, existing data files and analysis methods in the wind power generation field have some disadvantages.

Firstly, although the sampling period of the fault file is 20ms, the original data and the data change trend can be obtained to the greatest extent, but the defect is that the B file for recording the running environment state information can be generated only when the wind generating set is in fault shutdown, and the time length is only 1 minute generally, so that the real-time monitoring and fault early warning of the wind generating set are inconvenient, and the 24-hour continuous monitoring of the wind generating set cannot be realized.

Second, continuous data acquisition, although capable of obtaining 20ms sampled data over a long period of time, is generally more suitable for data collection and fault analysis of a single wind turbine generator system. The data acquisition needs to be carried out on all the wind generating sets in the wind power plant by means of hardware equipment such as a PC (personal computer) or an industrial personal computer, so that a plurality of industrial personal computers are required to be configured, the cost is high, the time consumption is long, and the problem of data acquisition interruption easily occurs.

Third, while the transient data file may be derived by the SCADA system and is continuous data, the SCADA system samples for a longer period, typically 5 s-7 s, to record a data value. Because the sampling period is too long, the existing analysis methods (such as spectrum analysis, fluctuation detection, jump detection, peak detection, margin detection and the like) cannot accurately analyze the running state of the wind turbine generator system.

Fourthly, a PLC built-in algorithm developed in the PLC controller can output a detection result to the SCADA, but the built-in method firstly needs to upgrade a program of the controller, and data before upgrade cannot be analyzed; secondly, the detection of the parameter adjustment is inconvenient, and the data before the parameter adjustment cannot be analyzed; thirdly, the data only has fixed result output, and the database is inflexible.

Fig. 1 is a waveform diagram showing data fluctuations of B file and transient data file records. As shown in fig. 1, the left side view is a waveform of data fluctuation recorded in the B file, and the right side view is a waveform of data fluctuation recorded in the instantaneous data file. As can be seen from fig. 1, although the data recorded in the B file having the sampling period of 20ms greatly fluctuates, only one data value shown in the red box, i.e., the transient data file, can be acquired in the transient file having the sampling period of 7s, and it is difficult to restore the trend of the data.

Disclosure of Invention

Therefore, the embodiment of the disclosure provides an anomaly detection method, a storage medium and a computing device for a pitch system, which can perform anomaly diagnosis and early warning based on SCADA operation data, and on one hand solve the problems that the SCADA operation data has a long sampling period, no continuous data and cannot realize effective monitoring, and on the other hand, realize the problems that the operation state of a wind turbine generator set cannot be accurately analyzed by the existing spectrum analysis, jump detection, amplitude detection and the like.

In one general aspect, there is provided an abnormality detection method of a pitch system, the abnormality detection method including: acquiring operation data of each pitch motor of the pitch system during the pitch operation of the pitch system; accumulating and summing the operation data of each variable pitch motor at each sampling moment in a preset time period to obtain the data and value of each sampling moment; data consistency judgment is carried out on the data and the value of each sampling moment; determining that the pitch system is operating normally in response to the consistency of the data and the values at each sampling moment; and determining that an anomaly exists in the pitch system in response to the data and values at each sampling instant not having consistency.

Optionally, in response to the data and values at each sampling instant not having consistency, the step of determining that an anomaly exists in the pitch system comprises: whether an abnormality of the pitch system is caused by an abnormality of the pitch bearing or by an abnormality of a blade of the wind turbine generator set is determined based on the data and the value at each sampling instant.

Optionally, the anomaly detection method further includes: acquiring the rotating speed of an impeller of the wind generating set during the pitch operation of the pitch system; calculating an impeller rotation period based on the impeller rotation speed; responsive to the impeller rotation period not temporally coincident with the sampling period, the step of cumulatively summing operational data of the respective pitch motors for each sampling instant over a predetermined period of time is performed.

Optionally, the step of performing data consistency judgment on the data and the value at each sampling time includes: determining whether the data and the value of each sampling moment are smaller than a first preset threshold value or not; and determining that the data and the value at each sampling moment have consistency in response to the data and the value at each sampling moment being smaller than a first preset threshold.

Optionally, the step of performing data consistency judgment on the data and the value at each sampling time further includes: responding to the fact that the data sum value of at least one sampling moment is larger than or equal to the first preset threshold value, calculating the difference value of the data sum value of each sampling moment and the data sum value of the previous sampling moment, and obtaining the difference value of the data sum value of each sampling moment, wherein each sampling moment comprises all sampling moments except the first sampling moment in a preset time period; and carrying out data consistency judgment on the data and the value of each sampling time based on the difference value of the data and the value of each sampling time.

Optionally, the step of performing data consistency judgment on the data and the value at each sampling time includes: calculating the difference value between the data and the value of each sampling moment and the data and the value of the previous sampling moment to obtain the difference value of the data and the value of each sampling moment, wherein each sampling moment comprises all sampling moments except the first sampling moment in a preset time period; and carrying out data consistency judgment on the data and the value of each sampling time based on the difference value of the data and the value of each sampling time.

Optionally, the step of performing data consistency judgment on the data and the value at each sampling time based on the difference value of the data and the value at each sampling time includes: for each sampling instant, determining whether an absolute value of a difference of the data and the value at the respective sampling instant is less than a second preset threshold, and accumulating the first count value in response to the absolute value of the difference of the data and the value at the respective sampling instant being less than the second preset threshold; in response to the first count value being greater than the first count threshold, it is determined that the data and values at each sampling instant have consistency.

Optionally, the step of performing data consistency judgment on the data and the value at each sampling time based on the difference value of the data and the value at each sampling time includes: for each sampling instant, determining whether an absolute value of a difference of the data and the value at the respective sampling instant is greater than or equal to a second preset threshold, and accumulating the second count value in response to the absolute value of the difference of the data and the value at the respective sampling instant being greater than or equal to the second preset threshold; in response to the second count value being greater than the second count threshold, it is determined that the data and values at each sampling instant do not have consistency.

Optionally, in response to the data and values at each sampling instant not having consistency, the step of determining that an anomaly exists in the pitch system comprises: in response to the second count value being greater than the second count threshold, it is determined that the abnormality of the pitch system is caused by a pitch bearing abnormality.

Optionally, the step of performing data consistency judgment on the data and the value at each sampling time based on the difference value of the data and the value at each sampling time includes: for each sampling instant, determining whether a sign of a difference between the data and the value at the respective sampling instant and a second preset threshold is different from a sign of a difference between the data and the value at the previous sampling instant and the second preset threshold, and accumulating a third count value in response to the sign of a difference between the data and the value at the respective sampling instant and the second preset threshold being different from the sign of a difference between the data and the value at the previous sampling instant and the second preset threshold; in response to the third count value being greater than the third count threshold, it is determined that the data and values at each sampling instant do not have consistency.

Optionally, in response to the data and values at each sampling instant not having consistency, the step of determining that an anomaly exists in the pitch system comprises: determining whether the respective sampling instants to which the third count value is added are regularly distributed in time in response to the data and the value at each sampling instant not having consistency; determining that the abnormality of the pitch system is caused by a blade abnormality of the wind turbine generator system in response to the respective sampling instants accumulating the third count values being regularly distributed over time; in response to the respective sampling instants accumulating the third count values being irregularly distributed in time, it is determined that the abnormality of the pitch system is caused by a pitch bearing abnormality.

Optionally, the operation data of each pitch motor is SCADA operation data and includes at least one of a torque value, a current value, a given speed value, and an actual speed value of each pitch motor.

In another general aspect, there is provided a storage medium storing a computer program which, when executed by a processor, implements the anomaly detection method of a pitch system as described above.

In another general aspect, there is provided a computing device, the computing device comprising: a processor; and a memory storing a computer program which, when executed by the processor, implements the abnormality detection method of the pitch system as described above.

According to the abnormality detection method, the storage medium and the computing equipment of the variable pitch system, abnormality diagnosis and early warning based on SCADA operation data can be achieved, the problem that the sampling period of the SCADA operation data is too long is effectively solved, correct analysis of the operation data of the wind turbine generator set, which cannot be achieved through existing spectrum analysis, jump detection, amplitude detection and the like, is achieved, and the application value of the operation data is improved.

On the other hand, according to the abnormality detection method, the storage medium and the computing equipment of the pitch system, which are disclosed by the embodiment of the disclosure, the PLC program is not required to be updated and optimized, and the detection parameters are not required to be adjusted repeatedly, so that the method and the device are particularly suitable for SCADA transient data files generated in historical dates, and the effectiveness and the reliability of abnormality detection are greatly improved.

In addition, according to the abnormality detection method, the storage medium and the computing device of the pitch system, which are disclosed by the embodiment of the invention, the detection of frequency spectrum and jump amplitude is not involved, so that the influence of torque value fluctuation is small, and the probability of false detection can be effectively reduced. Meanwhile, the problem that the number of files B is small (generated only when the machine is stopped) can be solved, and 24-hour monitoring of the running state of the variable pitch system is realized; and the SCADA program and the communication protocol between the wind generating set and the central monitoring are not required to be modified, so that the implementation is convenient, and the development workload is small.

Drawings

The foregoing and other objects and features of embodiments of the present disclosure will become more apparent from the following description taken in conjunction with the accompanying drawings in which the embodiments are shown, in which:

FIG. 1 is a waveform diagram showing data fluctuations of a B file and a transient data file record;

fig. 2 is a diagram showing waveforms and vector exploded views of a standard three-phase alternating current signal;

FIG. 3 shows a blade force analysis diagram of a circular region in which the blades rotate;

FIG. 4 is a graph showing torque curves of three pitch motors when a wind turbine is pitched;

FIG. 5 is a graph showing torque curves of three pitch motors when a wind turbine is pitched;

FIG. 6 is a graph illustrating torque curves of three pitch motors during operation and shutdown of a wind turbine;

FIG. 7 is a graph illustrating a torque curve of a blade anomaly time varying pitch system;

fig. 8 is a diagram showing a change curve of three-phase data;

FIG. 9 is a flowchart illustrating an anomaly detection method of a pitch system according to an embodiment of the present disclosure;

fig. 10 is a graph showing current values of three pitch motors when a pitch bearing is abnormal;

fig. 11 is a waveform diagram showing data and values at each sampling timing;

fig. 12 is a block diagram illustrating a computing device according to an embodiment of the present disclosure.

Detailed Description

The following detailed description is provided to assist the reader in obtaining a thorough understanding of the methods, apparatus, and/or systems described herein. However, various changes, modifications, and equivalents of the methods, apparatus, and/or systems described herein will be apparent after an understanding of the present disclosure. For example, the order of operations described herein is merely an example and is not limited to those set forth herein, but may be altered as will be apparent after an understanding of the disclosure of the application, except for operations that must occur in a specific order. Furthermore, descriptions of features known in the art may be omitted for clarity and conciseness.

Next, the implementation principle of the abnormality detection method of the pitch system according to the embodiment of the present disclosure will be described first.

Fig. 2 is a diagram showing waveforms and vector exploded views of a standard three-phase ac signal.

Referring to fig. 2, when the three-phase load is balanced, i.e., the three-phase load resistance is as large, the three-phase current is as large. Let the resistor be R, then the three phase instantaneous voltages are usin α, usin (α+2/3π), usin (α+4/3π), respectively. u is the maximum voltage, α is any electrical angle, which can be understood to be any time, pi is the arc angle, 2/3 pi is 120 degrees, and sin is a sine function.

The three-phase currents are respectively as follows: usin α/R, usin (. Alpha. +2/3π)/R, usin (. Alpha. +4/3π)/R.

The sum of the currents is usin α/R+ usin (α+2/3π)/R+ usin (α+4/3π)/R, and can be further deduced as follows.

The final result of the operation is equal to 0 ampere, i.e., the sum of three phase currents is 0.

Table 1 is a table showing data (e.g., current, etc.) for a standard three-phase power. As shown in the "AB difference" column of table 1, the maximum absolute value of the difference is 1.73, and thus accurate state detection cannot be achieved by merely judging the difference of two-phase data. However, as can be seen from table 1, the sum of the three-phase data is 0 at any time. Therefore, the method and the device utilize the principle to research the operation mechanism and the operation characteristics of the wind generating set, develop an abnormality diagnosis and early warning method based on SCADA operation data, and effectively solve the problems that the sampling time is long, continuous data is not available and effective monitoring cannot be realized in the prior art.

TABLE 1

The operational characteristics of the wind power generation set are specifically analyzed as follows.

FIG. 3 shows a blade force analysis diagram of a wind turbine. As shown in fig. 3, the mass 204 is the equivalent mass m of the blade 107 as it rotates when mounted on the hub; the first partial mass 205 is the partial mass of the mass 204 in the direction perpendicular to the blade 107; the second partial mass 206 is a partial mass of the mass 204 in a direction parallel to the rotation axis 304. The mass 204, the first sub-mass 205 and the second sub-mass 206 are all located in a vertical plane formed by the three blades 107, and the angle formed by the directions in which the mass 204 and the second sub-mass 206 are located is a (degrees).

FIG. 4 shows a blade force analysis of a circular area where the blade rotates. As shown in fig. 4, the third partial mass 302 is the partial mass of the first partial mass 205 in a direction tangential to the circle in which the blade 107 rotates, and the fourth partial mass 303 is the partial mass of the first partial mass 205 in a direction perpendicular to the blade axis (i.e., the axis of rotation 305 in fig. 4). As can be seen from fig. 4, the first partial mass 205 is perpendicular to the blade flank surface but not perpendicular to the rotation axis 304; the third sub-mass 302 is perpendicular to the rotation axis 304.

Further, as can be seen from fig. 3, the first partial mass 205 is equal to the mass 204 multiplied by sina, assuming that the position vertically upwards is 0 degrees, the azimuth angle of the blade 107 at the 0 degree position measured by the azimuth angle sensor 112 is set to a1 (degrees), and as can be seen from fig. 3, a1 is 0; the azimuth angle of the blade 107 on the right half of fig. 3 is set to a2 (degrees), a2 being a1+120 (degrees); the azimuth angle of the blade 107 on the left half of fig. 3 is set to a3 (degrees), and a3 is a1+240 (degrees). In each of the blades 107 shown in fig. 3, a1=0, so a2=120, a3=240; the angle a=180-120=60 degrees between the direction in which the mass 204 is located and the direction in which the second partial mass 206 is located can be calculated.

On the other hand, as can be seen from fig. 4, the third partial mass 302 is equal in size to the first partial mass 205 multiplied by sinb, and b is the blade angle (i.e. pitch angle) measured by the encoder, i.e. the angle between the blade side and the vertical plane 301 formed by the three blades 107 (as shown in fig. 4), the vertical plane shown in fig. 4 being the same plane as the plane in fig. 3.

Assuming that the initial position of the blade 107 is a 90 degree position, when the blade 107 is located on the right half plane (i.e., the right half plane of the plane in fig. 3) of the vertical plane 301 formed by the three blades 107 and the pitch direction is 0 degree, the additional moment of inertia I2 of the blade 107 acts as a resistance to the pitch; when the blade 107 is located in the right half of the vertical plane 301 and the pitch direction is 180 degrees, the additional moment of inertia I2 of the blade 107 contributes to the pitch.

When the blade 107 is located on the left half plane of the vertical plane 301 (i.e. the left half plane of the plane in fig. 3), and the pitch direction is 0 degrees, the additional moment of inertia I2 of the blade 107 acts as a booster to the pitch; when the blade 107 is located in the left half plane of the vertical plane 301 and the pitch direction is 180 degrees pitch, the additional moment of inertia I2 of the blade 107 acts as a resistance to pitch.

From this, the magnitude of the additional moment of inertia of the blade 107 can be derived as:

I2 ＝ R²×m×sina×sinb×p×(a-180)/|(a-180)| (1)

where p represents the pitch direction of the blade 107, p= +1 when the blade 107 is pitched to the 0 degree position, and p= -1 when the blade 107 is pitched to the 180 degree position.

Meanwhile, the magnitude N of the torque caused by the gravity of the blade 107 can be found as:

N ＝ m×sina×sinb×p×(a-180)/|(a-180)| (2)

As can be seen from formulas (1) and (2), when the nacelle direction of the wind turbine generator system has a deviation angle a with the wind direction, the three blades of the wind turbine generator system are different in lift force, and meanwhile, the blades have flanges and chord parts, so that the stress of the three blades is unbalanced, and the load of the wind turbine generator system is increased. When the wind direction forms an acute angle with the rotation direction of the blade, the right half part faces the wind generating set, and the resistance is high and the lift force is low; when the wind direction and the rotation direction of the blades form an obtuse angle, the left half part of the blade faces the wind generating set, the resistance is high, and the lift force is small, wherein the lift force refers to the power of the wind driven blades rotating in the rotation direction of the impeller; the resistance refers to the load force applied by the blade when pitching.

Fig. 5 is a diagram showing torque curves of three pitch motors when a wind turbine is pitching. Curves 501, 502, 503 are torque curves of three pitch motors, respectively. As can be seen from fig. 5, under the normal pitch condition, the torque of the three pitch motors all show sinusoidal fluctuation, and the phase difference is 120 degrees, which indicates that the torque value of the pitch motor shows periodic variation along with the rotation of the impeller.

FIG. 6 is a graph illustrating torque curves for three pitch motors during operation and shutdown of a wind turbine. Curves 601, 602, 603 are torque curves of three pitch motors, respectively. As can be seen from fig. 6, under the normal pitch condition, the torque of the three pitch motors all presents sinusoidal fluctuation, and the sinusoidal fluctuation period of the wind turbine generator system gradually changes along with the decrease of the rotation speed of the impeller.

Based on fig. 2 to 6, the feasibility basis of the present disclosure can be further derived. Specifically, when the pitch system of the wind generating set adjusts the pitch, the speed is divided into positive and negative directions, and the directions of the pitch motors are respectively turned. A positive speed indicates the pitch system pitch (90 degrees pitch) and a positive speed indicates the pitch system pitch (0 degrees pitch), so the torque and current values of the wind turbine generator and the pitch drive are also positive and negative.

Fig. 7 is a diagram showing a torque curve of the blade-abnormal-time pitch system. As can be seen from fig. 7, the torque 703 is significantly higher than the other two axes due to the greater resistance of the blade and the pitch motor, and the fluctuation period is (822-355) 20 ms=9340 ms=9.34 s, which is 5-7 seconds greater than the data sampling period of the SCADA system. Thus, in SCADA operational data, this phenomenon can be effectively identified by the abnormality detection method of the pitch system according to the embodiment of the present disclosure.

Further, in fig. 7, although it seems that abnormality recognition can be performed by the difference of three torques, since the torque values of three pitch motors are sinusoidal and the phase difference is 120 degrees, the difference of any two torque values may be relatively large under normal conditions (see the analysis in table 1). Further, fig. 8 is a diagram showing a change curve of three-phase data. As shown in fig. 8, if the data sampling point is located exactly at time t1, t2 or t3, the difference between the two phase data will be relatively large, which will cause a great interference to the anomaly detection, and even cause erroneous detection.

The advantages and feasibility of the anomaly detection method of the pitch system according to the embodiments of the present disclosure are also reflected in: because the loads of the variable pitch motors are different, the current values and the torque values are different, and the loads have a larger relationship with wind speed, wind direction, gear oil temperature and the like, the differences are compared independently, and the abnormality cannot be identified effectively.

An abnormality detection method of a pitch system according to an embodiment of the present disclosure is described in detail below.

Fig. 9 is a flowchart illustrating an anomaly detection method of a pitch system according to an embodiment of the present disclosure. According to embodiments of the present disclosure, the anomaly detection method of the pitch system may be performed by a SCADA industrial personal computer, a data analysis platform, or other computing device disposed in a wind farm control center. However, the present disclosure is not limited thereto, and the abnormality detection method of the pitch system may be performed by an upper computer disposed outside the wind farm, or may even be performed by a master control of a single wind turbine generator set, as long as the execution device is capable of acquiring SCADA operation data.

Referring to fig. 9, in step S901, operation data of respective pitch motors of a pitch system of a wind turbine generator system may be acquired during a pitch operation performed by the pitch system. As described above, the operation data of each of the pitch motors may be SCADA operation data, and include at least one of a torque value, a current value, a given speed value, and an actual speed value of each of the pitch motors. In this way, according to the embodiments of the present disclosure, abnormality detection may be performed using various operation data of the pitch motor, thereby improving usability of an abnormality detection method of the pitch system.

According to embodiments of the present disclosure, the angle values of the individual blades may also be obtained, and it may be determined whether the wind turbine is in a pitch state (including wind turbine start-up, power-limited pitch, rotational speed-limited pitch, etc.) based on the angle values, i.e., whether it is during a pitch operation. For example, when the angle value is not 0, it may indicate that the wind turbine is in a pitched state. Alternatively, if the pitch system has a 0-speed maintaining function (i.e. the torque of the pitch motor is not 0 when the pitch is not changed), it may be determined whether the wind generating set is in a pitch state by the torque value of the pitch motor. However, the disclosure is not limited thereto, and whether the generator set is in a pitch state may be determined by various other existing methods. When it is determined that the wind turbine generator set is in a pitch state, an abnormality detection method of a pitch system according to an embodiment of the present disclosure may be performed.

In step S902, the operation data of the respective pitch motors at each sampling time within the predetermined period may be summed together to obtain a data sum value at each sampling time. For example, at each sampling instant, the torque values (or current value, given speed value, actual speed value) of the three pitch motors may be accumulated to obtain the data and value for each sampling instant. The duration of the predetermined period of time may be, for example, 5 minutes, but the present disclosure is not limited thereto. The duration of the predetermined period of time may be greater than or less than 5 minutes.

According to embodiments of the present disclosure, the rotational speed of the impeller of the wind turbine may also be obtained during the pitch operation performed by the pitch system, and the impeller rotation period may be calculated based on the rotational speed of the impeller. Thereafter, when it is determined that the impeller rotation period is not temporally coincident with the sampling period, step S902 may be performed. That is, if the rotation period of the impeller and the sampling period substantially coincide in time (i.e., the cycle start-stop points substantially coincide in time), the abnormality detection method of the pitch system is ended, and abnormality detection is temporarily not performed. This is because if the impeller rotation period coincides substantially in time with the sampling period, meaning that at two times the data sampling time, three blades coincide with the last azimuth position, the data of the two samples substantially coincide, and the referenceable meaning of the data becomes smaller. Therefore, by executing consistency judgment of the rotation period and the sampling period of the impeller, the reliability of operation data can be ensured, and the probability of false detection is further reduced.

In step S903, data consistency determination may be performed for the data and the value at each sampling time.

In step S904, it may be determined that the pitch system is operating properly in response to the data and values at each sampling instant having consistency. In step S905, it may be determined that an abnormality exists in the pitch system in response to the data and the value at each sampling instant not having consistency. Further, in step S905, it may also be determined whether the abnormality of the pitch system is caused by a pitch bearing abnormality or by a blade abnormality of the wind turbine generator system based on the data and the value of each sampling time. In this way, accurate detection of pitch bearing anomalies (e.g., pitch bearing stuck, etc.) and blade anomalies (e.g., blade breakage, etc.) can be further achieved, thereby improving the usability of the anomaly detection method of the pitch system.

Next, the data consistency determination will be described in detail.

According to an embodiment of the present disclosure, in step S903, it may be determined whether the data and the value at each sampling instant are less than a first preset threshold. And if the data and the value of each sampling time are smaller than the first preset threshold value, determining that the data and the value of each sampling time have consistency. Here, by judging whether or not the data and the value at each sampling timing are smaller than the first preset threshold, it can be determined whether or not the data and the value at each sampling timing are close to 0. For example, the first preset threshold may be set to 0.5, but the present disclosure is not limited thereto. The first preset threshold value can be arbitrarily set by a person skilled in the art according to actual needs, as long as it is ensured that it is possible to determine whether the data and the value at each sampling instant are close to 0. As shown in table 1, the sum of the three-phase data is 0 at any time. If the data sum value is continuously close to 0, the operation data of each variable pitch motor is in accordance with the characteristic of the three-phase alternating current curve. Therefore, by determining whether the data sum value at each sampling time is close to 0, data consistency determination can be quickly performed, and the method execution speed can be improved.

If the data sum value of at least one sampling moment is larger than or equal to a first preset threshold value, the difference value between the data sum value of each sampling moment and the data sum value of the previous sampling moment can be further calculated to obtain the difference value of the data sum value of each sampling moment, and then the data consistency judgment is carried out on the data sum value of each sampling moment based on the difference value of the data sum value of each sampling moment. Here, each sampling instant includes all sampling instants within a predetermined period except the first sampling instant. According to the present embodiment, the data consistency determination can be further performed using the difference between the data and the value at each sampling timing, thereby improving the usability of the data and ensuring the reliability of the data consistency determination.

Alternatively, it may not be determined whether the data and value at each sampling timing are close to 0, but the data consistency determination may be made directly using the difference of the data and value at each sampling timing. That is, in step S903, the difference between the data and the value at each sampling time and the data and the value at the previous sampling time may be calculated, to obtain the difference between the data and the value at each sampling time; and then, based on the difference value of the data and the value of each sampling time, carrying out data consistency judgment on the data and the value of each sampling time. As described above, each sampling instant includes all sampling instants within the predetermined period except the first sampling instant. Thus, repeated data consistency judgment can be avoided, and the execution speed of the method is improved.

According to one embodiment of the present disclosure, the step of performing data consistency judgment on the data and the value of each sampling time based on the difference value of the data and the value of each sampling time may include: for each sampling instant, determining whether an absolute value of a difference of the data and the value at the respective sampling instant is less than a second preset threshold, and accumulating the first count value in response to the absolute value of the difference of the data and the value at the respective sampling instant being less than the second preset threshold; in response to the first count value being greater than the first count threshold, it is determined that the data and values at each sampling instant have consistency. Here, the first count value may be reset to zero when the abnormality detection method of the pitch system is started. The second preset threshold may take a value in the range of 5 to 10, but the present disclosure is not limited thereto. Further, the first count threshold may be set to an integer value greater than or equal to 10, but the present disclosure is not limited thereto. For example, the first count threshold may also be set to an integer value greater than or equal to 5. In practice, the setting of the first count threshold needs to be considered from two aspects. In one aspect, the first count threshold may be set to a larger value so that the accuracy of anomaly identification may be higher. On the other hand, the first count threshold cannot be set too large, otherwise it would lead to missed detection. As can be seen from equation (2), since the deviation of the wind direction may have a certain influence on the center line of the torque value, when judging the data consistency, it can be judged whether the difference between the data and the value at the previous and subsequent sampling times is close (i.e., whether the change of the value is stable), thereby realizing the data consistency judgment. Thus, accuracy and reliability of data consistency judgment can be improved.

Table 2 is a table showing the difference between the data and the values at partial sampling times corresponding to the torque curves of the three pitch motors at the time of pitching of the wind turbine shown in fig. 6.

TABLE 2

/>

As shown in table 2, the difference between the data and the value at each sampling instant is at most-4.6, and the absolute value is less than 5. Thus, it can be determined that the data and the value at each sampling instant have consistency.

Optionally, the step of performing data consistency determination on the data and the value at each sampling time based on the difference between the data and the value at each sampling time includes: for each sampling instant, determining whether an absolute value of a difference of the data and the value at the respective sampling instant is greater than or equal to a second preset threshold, and accumulating the second count value in response to the absolute value of the difference of the data and the value at the respective sampling instant being greater than or equal to the second preset threshold; in response to the second count value being greater than the second count threshold, it is determined that the data and values at each sampling instant do not have consistency. Here, the second count value may be reset to zero when the abnormality detection method of the pitch system is started. As described above, the second preset threshold value may take a value in the range of 5 to 10, but the present disclosure is not limited thereto. Further, the second count threshold may be set to be the same as the first count threshold or may be set to be different from the first count threshold. Thus, accuracy and reliability of data consistency judgment can be improved.

Upon determining that the data and values at each sampling instant do not have consistency as described above, in step S905, it may be determined that the abnormality of the pitch system is caused by a pitch bearing abnormality in response to the second count value being greater than the second count threshold. In this way, accurate detection of pitch bearing anomalies (e.g., pitch bearing jamming, etc.) can be further achieved, thereby improving the usability of the anomaly detection method of the pitch system.

Table 3 is a table showing the difference between the data and the values at partial sampling times corresponding to the torque curves of the three pitch motors at the time of the stop of the wind turbine shown in fig. 7.

TABLE 3 Table 3

/>

As shown in table 3, the absolute value of the difference between the data and the value is greater than the second preset threshold a plurality of times, resulting in the second count value being greater than the second count threshold, so it can be determined that the data and the value at each sampling instant do not have consistency, i.e., that there is an abnormality in the pitch system.

According to another embodiment of the present disclosure, the step of performing data consistency judgment on the data and the value of each sampling time based on the difference value of the data and the value of each sampling time may include: for each sampling instant, determining whether a sign of a difference between the data and the value at the respective sampling instant and a second preset threshold is different from a sign of a difference between the data and the value at the previous sampling instant and the second preset threshold, and accumulating a third count value in response to the sign of a difference between the data and the value at the respective sampling instant and the second preset threshold being different from the sign of a difference between the data and the value at the previous sampling instant and the second preset threshold; in response to the third count value being greater than the third count threshold, it is determined that the data and values at each sampling instant do not have consistency. Here, the third count value may be reset to zero when the abnormality detection method of the pitch system is started. The third count threshold may be set to an integer value greater than or equal to 10, but the disclosure is not limited thereto. For example, the third count threshold may also be set to an integer value greater than or equal to 5. In practice, the setting of the third counting threshold also needs to be considered from two aspects. In one aspect, the third count threshold may be set to a larger value so that the accuracy of anomaly identification may be higher. On the other hand, the third counting threshold cannot be set too large, otherwise it would lead to missed detection. According to an embodiment of the present disclosure, if the difference between the data and the value at the sampling time is greater than or less than the second preset threshold (i.e., the difference between the data and the value and the second preset threshold is a positive value for a while and a negative value for a while), it is indicated that the blade of the wind turbine may be abnormal. The specific physical phenomenon can be shown in fig. 7. Therefore, by further judging whether the difference value of the data and the value at each sampling time is consistent with the positive and negative variation of the difference value of the second preset threshold value, the usability of the data can be improved.

Specifically, when it is determined that the data and the value at each sampling timing do not have consistency as described above, in step S905, it is determined whether the respective sampling timings at which the third count value is accumulated are regularly distributed in time in response to the data and the value at each sampling timing not having consistency; determining that the abnormality of the pitch system is caused by a blade abnormality of the wind turbine generator system in response to the respective sampling instants accumulating the third count values being regularly distributed over time; in response to the respective sampling instants accumulating the third count values being irregularly distributed in time, it is determined that the abnormality of the pitch system is caused by a pitch bearing abnormality.

The principle of determining whether an abnormality of the pitch system is caused by an abnormality of the pitch bearing or by an abnormality of the blades of the wind turbine is as follows. Firstly, the acting force of the blades on the variable-pitch motor is changed along with the rotation of the impeller and is related to the azimuth angle of the blades, so that the change period is relatively longer and the change period and the rotation speed of the impeller are more regular. Therefore, if the positive and negative changes of the difference between the data and the value and the second preset threshold value are relatively regular, as shown in fig. 7, it is possible to further judge that the blade is abnormal. Secondly, due to the ball structure of the pitch bearing, when the pitch bearing is abnormal, the current of the pitch motor is intermittently influenced. For pitch drives: the load is increasing and reacts to the main flux, i.e. the armature, in which case the input current is also increasing to counteract the load-induced reaction, so that the total flux remains substantially unchanged. When the pitch bearing is abnormal, the load of the pitch bearing is neglected and reduced in the rotating process of the blade, so that the current can be greatly fluctuated, and the phenomenon has no relevance with the rotating speed of the impeller. Therefore, if the positive and negative changes of the difference between the data and the value and the second preset threshold value are irregular, the abnormal of the pitch bearing can be further judged. Thus, the accurate detection of the abnormal of the variable-pitch bearing and the abnormal of the blade can be further realized, and the usability of the abnormal detection method of the variable-pitch system is improved.

Fig. 10 is a graph showing current values of three pitch motors when the pitch bearing is abnormal. In fig. 10, the abscissa indicates a time value, and the 0 time indicates a time when the wind turbine generator system triggers a fault shutdown, and the ordinate indicates a relevant variable value. Curve 1001 is the operation data (current) of the pitch motor 1, curve 1002 is the operation data (current) of the pitch motor 2, and curve 1003 is the operation data (current) of the pitch motor 3.

The operation data of the three pitch motors in fig. 10 are accumulated and summed according to each sampling time, and the points are taken at equal intervals, so that a waveform diagram of the data and the values of each sampling time shown in fig. 11 can be obtained. As can be seen from fig. 11, the fluctuation of the data and the values is irregular, and this feature can be used as a basis for distinguishing the abnormal types of the pitch system.

Fig. 12 is a block diagram illustrating a computing device according to an embodiment of the present disclosure. As described above, the computing device may be a SCADA industrial personal computer disposed in a wind farm control center, but the disclosure is not limited thereto.

Referring to fig. 12, a computing device 1200 according to an embodiment of the disclosure includes a processor 1210 and a memory 1220. Processor 1210 may include, but is not limited to, a Central Processing Unit (CPU), a Digital Signal Processor (DSP), a microcomputer, a Field Programmable Gate Array (FPGA), a system on a chip (SoC), a microprocessor, an Application Specific Integrated Circuit (ASIC), or the like. Processor 1210 may obtain SCADA operation data for the target wind turbine. The memory 1220 may store computer programs to be executed by the processor 1210. Memory 1220 may include high-speed random access memory and/or non-volatile computer-readable storage media. When processor 1210 executes the computer program stored in memory 1220, an anomaly detection method for a pitch system as described above may be implemented.

An abnormality detection method of a pitch system according to an embodiment of the present disclosure may be written as a computer program and stored on a storage medium. The anomaly detection method of the pitch system as described above may be implemented when the computer program is executed by a processor. Examples of storage media include: read-only memory (ROM), random-access programmable read-only memory (PROM), electrically erasable programmable read-only memory (EEPROM), random-access memory (RAM), dynamic random-access memory (DRAM), static random-access memory (SRAM), flash memory, nonvolatile memory, CD-ROM, CD-R, CD + R, CD-RW, CD+RW, DVD-ROM, DVD-R, DVD + R, DVD-RW, DVD+RW, DVD-RAM, BD-ROM, BD-R, BD-R LTH, BD-RE, blu-ray or optical disk storage, hard Disk Drives (HDD), solid State Disks (SSD), card-type memories (such as multimedia cards, secure Digital (SD) cards or ultra-fast digital (XD) cards), magnetic tapes, floppy disks, magneto-optical data storage devices, hard disks, solid state disks, and any other devices configured to store computer programs and any associated data, data files and data structures in a non-transitory manner and to provide the computer programs and any associated data, data files and data structures to a processor or computer to enable the processor or computer to execute the programs. In one example, the computer program and any associated data, data files, and data structures are distributed across networked computer systems such that the computer program and any associated data, data files, and data structures are stored, accessed, and executed in a distributed manner by one or more processors or computers.

In addition, according to the abnormality detection method, the storage medium and the computing device of the pitch system, diagnosis and analysis of bearing abnormality and blade abnormality of the pitch system can be realized by performing abnormality detection based on SCADA operation data, and the detection accuracy is high and is not influenced by factors such as wind speed, wind direction, environmental temperature and the like.

In addition, according to the abnormality detection method, the storage medium and the computing equipment of the pitch system, the PLC program is not required to be updated and optimized, detection parameters are not required to be adjusted repeatedly, the method and the device are particularly suitable for SCADA transient data files generated in historical dates, and the effectiveness and the reliability of abnormality detection are greatly improved.

In addition, according to the abnormality detection method, the storage medium, and the computing device of the pitch system of the embodiment of the present disclosure, when actually applied, it is not necessary to convert the operation data according to the wind direction value, but the characteristics of the root operation data are processed, so that the accuracy of abnormality detection can be improved. This is because the wind direction is changed instantaneously, and although the blade stress is related to the wind direction, if wind direction data is added at the time of calculation, the data becomes disordered and a large disturbance is generated.

Although a few embodiments of the present disclosure have been shown and described, it would be appreciated by those skilled in the art that changes may be made in these embodiments without departing from the principles and spirit of the disclosure, the scope of which is defined in the claims and their equivalents.

Claims

1. An anomaly detection method for a pitch system, the anomaly detection method comprising:

acquiring operation data of each pitch motor of the pitch system during the pitch operation of the pitch system;

accumulating and summing the operation data of each variable pitch motor at each sampling moment in a preset time period to obtain the data and value of each sampling moment;

data consistency judgment is carried out on the data and the value of each sampling moment;

Determining that the pitch system is operating normally in response to the consistency of the data and the values at each sampling moment; and

And determining that the pitch system is abnormal in response to the data and the value at each sampling moment not having consistency.

2. The anomaly detection method of claim 1, wherein determining that an anomaly exists in the pitch system in response to the data and values at each sampling instant not having consistency comprises:

whether an abnormality of the pitch system is caused by an abnormality of the pitch bearing or by an abnormality of a blade of the wind turbine generator set is determined based on the data and the value at each sampling instant.

3. The abnormality detection method according to claim 1, characterized in that the abnormality detection method further comprises:

acquiring the rotating speed of an impeller of the wind generating set during the pitch operation of the pitch system;

calculating an impeller rotation period based on the impeller rotation speed;

responsive to the impeller rotation period not temporally coincident with the sampling period, the step of cumulatively summing operational data of the respective pitch motors for each sampling instant over a predetermined period of time is performed.

4. The abnormality detection method according to claim 1, wherein the step of making a data consistency judgment for the data and the value at each sampling timing includes:

determining whether the data and the value of each sampling moment are smaller than a first preset threshold value or not;

and determining that the data and the value at each sampling moment have consistency in response to the data and the value at each sampling moment being smaller than a first preset threshold.

5. The anomaly detection method of claim 4, wherein the step of performing data consistency determination for the data and the value at each sampling time further comprises:

responding to the fact that the data sum value of at least one sampling moment is larger than or equal to the first preset threshold value, calculating the difference value of the data sum value of each sampling moment and the data sum value of the previous sampling moment, and obtaining the difference value of the data sum value of each sampling moment, wherein each sampling moment comprises all sampling moments except the first sampling moment in a preset time period;

and carrying out data consistency judgment on the data and the value of each sampling time based on the difference value of the data and the value of each sampling time.

6. The abnormality detection method according to claim 1, wherein the step of making a data consistency judgment for the data and the value at each sampling timing includes:

Calculating the difference value between the data and the value of each sampling moment and the data and the value of the previous sampling moment to obtain the difference value of the data and the value of each sampling moment, wherein each sampling moment comprises all sampling moments except the first sampling moment in a preset time period;

7. The abnormality detection method according to claim 5 or 6, characterized in that the step of making a data consistency judgment for the data and the value at each sampling time based on the difference between the data and the value at each sampling time includes:

for each sampling instant, determining whether an absolute value of a difference of the data and the value at the respective sampling instant is less than a second preset threshold, and accumulating the first count value in response to the absolute value of the difference of the data and the value at the respective sampling instant being less than the second preset threshold;

in response to the first count value being greater than the first count threshold, it is determined that the data and values at each sampling instant have consistency.

8. The abnormality detection method according to claim 5 or 6, characterized in that the step of making a data consistency judgment for the data and the value at each sampling time based on the difference between the data and the value at each sampling time includes:

For each sampling instant, determining whether an absolute value of a difference of the data and the value at the respective sampling instant is greater than or equal to a second preset threshold, and accumulating the second count value in response to the absolute value of the difference of the data and the value at the respective sampling instant being greater than or equal to the second preset threshold;

in response to the second count value being greater than the second count threshold, it is determined that the data and values at each sampling instant do not have consistency.

9. The anomaly detection method of claim 8, wherein determining that an anomaly exists in the pitch system in response to the data and values at each sampling instant not having consistency comprises:

in response to the second count value being greater than the second count threshold, it is determined that the abnormality of the pitch system is caused by a pitch bearing abnormality.

10. The abnormality detection method according to claim 5 or 6, characterized in that the step of making a data consistency judgment for the data and the value at each sampling time based on the difference between the data and the value at each sampling time includes:

For each sampling instant, determining whether a sign of a difference between the data and the value at the respective sampling instant and a second preset threshold is different from a sign of a difference between the data and the value at the previous sampling instant and the second preset threshold, and accumulating a third count value in response to the sign of a difference between the data and the value at the respective sampling instant and the second preset threshold being different from the sign of a difference between the data and the value at the previous sampling instant and the second preset threshold;

In response to the third count value being greater than the third count threshold, it is determined that the data and values at each sampling instant do not have consistency.

11. The anomaly detection method of claim 10, wherein determining that an anomaly exists in the pitch system in response to the data and values at each sampling instant not having consistency comprises:

Determining whether the respective sampling instants to which the third count value is added are regularly distributed in time in response to the data and the value at each sampling instant not having consistency;

determining that the abnormality of the pitch system is caused by a blade abnormality of the wind turbine generator system in response to the respective sampling instants accumulating the third count values being regularly distributed over time;

in response to the respective sampling instants accumulating the third count values being irregularly distributed in time, it is determined that the abnormality of the pitch system is caused by a pitch bearing abnormality.

12. The anomaly detection method of claim 1, wherein the operational data of each of the pitch motors is SCADA operational data and includes at least one of a torque value, a current value, a given speed value, and an actual speed value of each of the pitch motors.

13. A storage medium storing a computer program, characterized in that the abnormality detection method of a pitch system according to any one of claims 1 to 12 is implemented when the computer program is executed by a processor.

14. A computing device, the computing device comprising:

A processor; and

A memory storing a computer program which, when executed by a processor, implements the abnormality detection method of a pitch system according to any one of claims 1 to 13.