TWI547639B - Determination of Abnormal Degree of Wind Turbine Generator and Judgment of Abnormal Degree of Wind Turbine - Google Patents

Description

Method for determining abnormal degree of wind power generator and method for determining abnormal degree of wind power generator

The present invention relates to an abnormality degree determination system for a wind power generator and a method for determining an abnormal degree of a wind power generator, and more particularly to a technique for determining an abnormal degree of a wind power generator using lightning strike information.

For example, Japanese Laid-Open Patent Publication No. 2012-117446 (Patent Document 1) is known. In this method, it is described that a lightning strike detecting device capable of reliably determining a lightning strike and a lightning strike position can be reliably determined by a simple, low-cost, and highly reliable structure. Further, Japanese Laid-Open Patent Publication No. 2011-236884 (Patent Document 2) is known. In this publication, it is described that a reliable and simple method is provided, that is, an improved lightning strike detecting device for detecting a lightning strike on a wind turbine is considered.

[Advanced technical literature] [Patent Literature]

Patent Document 1: Japanese Laid-Open Patent Publication No. 2012-117446

Patent Document 2: Japanese Laid-Open Patent Publication No. 2011-236884

In the conventional technology, it is not described that the lightning strike detection is performed, and the degree of abnormality affected by the lightning strike is considered. Therefore, although the fact of lightning strikes can be understood, the degree of damage generated in the blades of wind turbines is unknown. The present invention considers the influence of a lightning strike and aims to "can discern whether an abnormality is generated".

In order to solve the above problems, the abnormality degree determination system for a wind power generator according to the present invention is characterized in that: a sensor for detecting a state of the wind power generator; and determining based on information measured by the sensor An abnormality determining unit for generating an abnormality in the wind power generator; and a lightning strike information reading unit for reading lightning strike information of the wind power generator or lightning strike information around the wind power generator; and reading by the lightning strike information reading unit The lightning strike information and the determination by the abnormality determination unit are used to determine the degree of abnormality determination unit.

Further, a method for determining an abnormal degree of a wind power generator for solving the above problem is characterized in that: a sensor for detecting a state of a wind power generator is provided, and based on information measured by the sensor, Whether the wind turbine generates an abnormality, and reads the lightning strike information of the wind power generator or the lightning strike information around the wind power generator, according to the read the lightning The degree of abnormality is determined by hitting the information and determining whether or not an abnormality has occurred.

According to the present invention, it is possible to discriminate whether or not an abnormality is caused by considering the influence of a lightning strike.

001‧‧‧Leaf Diagnostic System

002‧‧‧ sensor

003‧‧‧Lightning Information Reading Department

004‧‧‧Signal Processing Department

005‧‧‧Abnormality Department

006‧‧‧Abnormal Position Calculation Department

007‧‧‧Database

008‧‧‧Abnormal degree determination department

009‧‧‧Lightning Detection Department

010‧‧‧Diagnostic Results Joint Department

100‧‧‧ leaves

S01‧‧‧ input signal

S02‧‧‧ lightning strike information

S03‧‧‧ signal processing

S04‧‧‧Abnormal judgment

S05‧‧‧Abnormal position determination

S06‧‧‧Exception degree determination

S07‧‧‧ lightning strike judgment

S08‧‧‧Output Department

G01‧‧‧Abnormal degree

G02‧‧‧ lightning test results

G03‧‧‧Abnormal adjustment

601‧‧‧Database structure example

602‧‧‧Last lightning information

701‧‧‧Anomaly curve example

Figure 1: An example of a structural diagram of a blade diagnostic system.

Figure 2: An example of a system flow chart.

Fig. 3 is an example of the abnormality degree determining unit.

Figure 4: An example of a database structure.

Figure 5: An example of an abnormality curve.

Figure 6: Overall view of the wind turbine.

Hereinafter, embodiments will be described using the drawings. The following description is only an example, and the embodiment of the present invention is not limited to the following specific aspects.

[Example 1]

In the present embodiment, an example of a system for diagnosing a blade of a wind power generator and calculating an abnormal degree when a lightning strike occurs is described. The blade of a wind power generator refers to the wing used by the wind power generator. sheet.

Fig. 1 is an example of a blade diagnostic system of a wind power generator of the present embodiment. The blade diagnostic system 001 is composed of a sensor 002, a lightning strike information reading unit 003, a signal processing unit 004, an abnormality determining unit 005, an abnormal position calculating unit 006, and a database operating as a recording unit (below). It is abbreviated as DB) 007, an abnormality degree determining unit 008, a lightning strike detecting unit 009, and a diagnosis result merging unit 010. The sensor 002 detects information for abnormality determination by the abnormality determining unit 005, and calculation of the abnormal position of the abnormal position calculating unit 006 even when a plurality of sensors are used. Further, in the lightning strike information reading unit 003, lightning strike information of the wind turbine itself or its surroundings (for example, within 1 km) is acquired. Then, based on the information acquired from the sensor 002 and the lightning strike information reading unit 003, the abnormality degree determination unit 008 performs the abnormality degree determination. For the abnormality of the wind turbine blade, it is considered that the abnormality is formed stepwise. Examples of the abnormality of the wind turbine blade include cracking, peeling and cutting of the wind turbine blade material, and the like.

In the blade diagnostic system 001, the sensor 002 is used to feed data to the signal processing unit 004. The signal processing unit 004 performs estimation of the feature amount used by the abnormality determining unit 005 based on the information read by the sensor 002. As far as the sensor 002 is concerned, an acoustic sensor, vibration, AE (Acoustic Emission, for example, detecting above 20 kHz) sensor, strain sensor can be considered. For example, in the case where the sensor is an acoustic sensor, the signal processing in the signal processing unit 004 can be considered as Fourier transformation, Hilbert transformation, absolute value processing of waveforms. The data processed by the signal processing unit 004 determines the state of the device by the abnormality determining unit 005. There are normal or abnormalities in the state discrimination of the machine, and in the discrimination of normality and abnormality, for example, a statistical method called threshold determination and clustering (reference: introduction to statistics of the University of Tokyo). The data determined by the abnormality determining unit 005 is sent to the abnormal position calculating unit 006. Here, the position calculation of the abnormality on the blade is assumed. As for the abnormal position calculation method, for example, a method in which a place where the sensor is mounted is set as an abnormality, or a method in which a plurality of sensors are used to estimate a place can be considered. In the sensor 002, when the number of installations is one, the abnormal position calculation is not performed. In the sensor 002, when a plurality of sensors are mounted and the sensor 002 is an acoustic sensor, in order to estimate the abnormal position calculation, it is conceivable to perform position estimation based on "phase difference". . The data processed by the abnormality determining unit 005 and the abnormal position calculating unit 006 is sent to the abnormality determining unit 008.

Simultaneously with the information processing obtained by the above-described sensor 002, the lightning strike information reading unit 003 reads the information of the lightning strike. Specifically, in the lightning strike information reading unit 003, it is conceivable to use a lightning strike sensor attached to the diagnostic device. In addition, for example, it is also possible to obtain surrounding lightning strike information from a GPS or a weather forecast. In the measurement method of the lightning strike sensor, a so-called voltage measurement or temperature measurement is specifically mentioned as an example, and a method of reading "lightning strike information around the wind power generator installation place" is used. More specific examples, there are so-called GPS and get the wind turbine installation site from the website of the Meteorological Bureau. The method of lightning strike information around. The lightning strike detecting unit 009 determines whether or not a lightning strike has occurred based on the information acquired by the lightning strike information reading unit 003. In the determination method of the lightning strike detection unit 009, when the detection means of the lightning strike information reading unit 003 detects a certain physical quantity by a sensor such as a voltage or a temperature, the threshold value is determined in advance, and the GPS is used. And the weather website, etc., that is, not only limited to the wind turbine itself, but also in the case of using lightning information including the surrounding, the information on the lightning strike obtained by the weather bureau website is directly used. The detection result of the lightning strike detecting unit 009 is sent to the abnormality degree determining unit 008.

The abnormality degree determining unit 008 performs the abnormal degree determination using the abnormality degree curve accumulated by the DB007 and the past lightning strike information in addition to the data obtained by the lightning strike detecting unit 009. In this paper, the so-called abnormality curve is a line type indicating the relationship between the degree of abnormality and the abnormality index. The abnormality data determined by the abnormality degree determining unit 008 is sent to the diagnosis result merging unit 010. The so-called degree of abnormality refers to the degree of damage caused by lightning strikes on the wind turbine blade. It is the level that will immediately stop, even if it is not stopped immediately, it does not cause an abnormal level, so-called damage degree and operation possibility Sexuality is associated.

Hereinafter, an actual example will be described. Here, wind turbine blade diagnosis is selected as an example. In Fig. 2, a flowchart is shown. The structure diagram of Fig. 1 and the flowchart of Fig. 2 will be described. Further, in this example, the sensor 002 in Fig. 1 is formed by attaching one blade to the AE sensor, and installing three in total at three places (one for each location). First, the sensor 002 will "use the AE sensor to obtain the loss. The incoming signal (AE waveform) is sent to the signal processing unit 004. In the flowchart of Fig. 2, the AE waveform is taken in by the input signal S01. Next, the signal processing unit 004 (corresponding to the signal processing S03 of Fig. 2) executes the processing of the input signal. Here, more specifically, the absolute value of the AE waveform is calculated. Then, in this example, during the calculation of the AE waveform, the input signal is AD-converted to obtain its absolute value. The absolute value of the AE waveform is taken as the feature amount. The feature amount processed by the signal processing unit 004 is sent to the abnormality determining unit 005 (abnormality determination S04 of Fig. 2). The abnormality determining unit 005 performs abnormality determination for the feature amount calculated by the signal processing unit 004. In this example, the method of abnormality determination is to use clustering. This means that the normal data is pre-registered in the DB, and the difference between the normal data and the "data input by the sensor 002 and processed by the signal processing unit 004 and input to the abnormality determining unit 005" is compared. Methods. In this example, the standard deviation (σ) calculated as the difference between the "normal data and the statistical distance formed by the clustering between the data input by the abnormality determining unit 005" is 3 or more. It is considered abnormal. Further, in the present embodiment, the standard deviation (σ) calculated at this time is referred to as an abnormality index. The abnormality determining unit 005 performs normal and abnormal recognition, and sends the abnormality determination result to the abnormal position calculating unit 006 (second map, abnormal position determination S05). The abnormal position calculating unit 006 calculates the abnormal position of the wind turbine blade using the characteristics calculated by the sensor 002, the signal processing unit 004, and the abnormality determining unit 005. In this example, although three sensors (one for each one) are installed in each of the three blades, it is also considered to divide the blades of the wind turbine into the front end, the center, and the root. of 3 parts. When calculating the abnormal position, focusing on "detecting the peak waveform of the abnormality from the sensor near the abnormality", the first of the three sensors is calculated to indicate "representation". An abnormality occurs near the sensor of the abnormal peak waveform. When calculating, the time interval between the peaks detected by one sensor (the time width of the peak waveform) and the peak waveform detected by the remaining sensors is used. In addition, if a certain time width is exceeded (for example, "from 10 o'clock to 11 o'clock" as a unit of abnormality detection), a peak is generated. In this case, there may be occasionally at that time. The sensor in the portion of the width that is closest to the abnormal location produces the initial peak. Relative to such a situation, it is also considered to utilize information before the specific time width. In this way, it is possible to prevent the detection of "the first peak of the sensor that is accidentally in the portion of the time width that is not closest to the abnormal position". The abnormal position calculated by the abnormal position calculating unit 006 and the result of the abnormality determining unit 005 are sent to the abnormality degree determining unit 008 (second map, abnormality degree determining S06).

The abnormality degree determination unit 008 refers to the abnormality degree curve registered in the DB 007, and further calculates the abnormality degree based on the normal abnormality recognition result calculated by the abnormality determination unit 005. Further, at this time, lightning strike information read from the lightning strike information reading unit 003 (Fig. 2, lightning strike information S02) is used to determine whether or not there is a lightning strike (the determination is performed by the lightning strike determination S07 of Fig. 2) for the abnormality degree determination. Time. The method for determining the degree of abnormality is such that when the number of lightning strikes is one, the abnormality index is not adjusted, and when the number of lightning strikes is two, the abnormality index is corrected, and the degree of abnormality of the same abnormality index is changed. This is due to: in the case of wind turbine blades, when When it comes to the impact of lightning strikes, regardless of whether the sensor value changes, etc., it is necessary to consider the case of damage caused by past lightning strikes. As an example of the case where the abnormality index is 3.0 (σ), when the number of lightning strikes exists in the past, the abnormality index is raised by one point. The degree of abnormality calculated by the abnormality degree determining unit 008 is sent to the diagnosis result merging unit 010. It is conceivable to connect the diagnosis result merging unit 010 to the control-maintenance system to make the monitoring result available. For the method (method) to be used, for example, there is control for retracting operation (referring to an operation mode in which the rotation speed is slowed compared to a normal rotation speed), or a state notification to a maintenance person (for maintenance personnel) The wireless terminal that is provided has a notification that the windmill needs attention, etc., and the method of use is not limited to this.

Fig. 3 shows the flow of the abnormality degree determining unit 008. The "degree of abnormality G01 calculated by the abnormality determining unit 005", the "lightning strike information G02 obtained by the lightning strike detecting unit 009", and the "degree of abnormality adjustment G03 executed by the abnormality degree G01 and the lightning strike information G02" . In the abnormality degree adjustment G03, the abnormality index calculated by the abnormality determining unit 005 is corrected. As for the specific correction method, it can be considered that the improvement of the 1(σ) score is performed on the abnormality index according to whether there is a lightning strike. In the improvement of the abnormality index, a method of "whether or not there is a lightning strike and a relationship between the abnormality determination result determined by the abnormality determination unit 005" and accumulated in the database 007 can be considered. For the "abnormality determination result determined by the abnormality determination unit 005" and the "elevation value of the abnormality index", a method of analyzing and constructing by simulation and a method of using a theoretical value at the time of design can be considered. Lightning strike information G02 means: when lightning strikes the wind turbine In the case of blades, or lightning strikes around the wind turbine (such as within 1km radius). In the lightning strike information G02, the sensor used by the lightning strike information reading unit 003 determines whether or not the "light strikes the blade" and the "light strikes the information around the wind turbine". Then, for example, in the case of direct hitting (referred to as lightning strike), the evaluation of the abnormality index is further improved compared to the case of lightning strike around the wind turbine. In addition, there is a certain amount of adjustment in the past accumulation information for the presence of a lightning strike when the elevation of the abnormality indicator is performed. The output is executed after the correction of the blade abnormality index is performed.

Fig. 4 shows an example of the construction of the database 007. In the database (DB), abnormal indicators and abnormalities, and past lightning strike information are displayed. In the present embodiment, it is a table defined as an abnormality index-abnormality degree. The number of lightning strikes indicated in the table is the number of times the wind turbine blade is designed as an index, and does not indicate "the number of lightning strikes on the wind turbine blade before being read by the sensor 002". Therefore, in the past, the lightning strike information is additionally provided in the form of displaying "the number of times the lightning strike of the wind turbine blade is formed in the DB before the sensor 002 is read in". When the lightning strike is determined by the lightning strike determination S07, the digital portion of 602 is continuously updated every time the determination is made.

The example in Fig. 4 shows an example in which the lightning strike information is one lightning strike in the past when the abnormality index is "1", "2", or "3 or more". In addition, for the degree of abnormality, an example is shown in which the degree of abnormality is divided into three stages. For the degree of abnormality, it can also be assumed to be 5 steps. Other stages such as paragraphs that differ from the three stages. In the illustrated example, since the past lightning strike is once, the degree of abnormality with respect to the abnormality index is increased by one stage. When the abnormality index calculated by the abnormality determining unit 005 is 2 (σ), when one lightning strike has occurred in the past, the abnormality index is corrected to 3 (σ). In this case, it is assumed that the abnormality index is increased by one abnormality index every time the number of lightning strikes is exceeded, but the operation of changing the abnormality index corresponding to the installation location or the like can be considered.

Figure 5 shows an example of an abnormality curve. The so-called abnormality curve is a line type indicating the relationship between the degree of abnormality and the abnormality index. The following method can be used for the abnormality curve: it is produced based on the experimental results at the time of designing the wind turbine blade, or is analyzed and produced by simulation or the like, and "in the wind turbine blade, the wind turbine blade is executed. The method of making an exception by simulation.

In the example of FIG. 5, the degree of abnormality determined by the abnormality degree determining unit 008 is divided into an abnormality index and an abnormality degree in three stages. The degree of abnormality becomes 1, 2, and 3 in order from small to large. Here, 1 means mild (no need to stop the operation immediately, or no need to repair within 1 month), 2 means moderate (although there is no need to stop the operation immediately, but must be repaired within 1 month), 3 indicates heavy (immediate shutdown level). In response to the lightning strike information shown in FIG. 4, the abnormality index is corrected in the abnormality curve registered in advance in the DB, and the determination result sent to the diagnosis result merging unit 010 is formed.

Fig. 6 is a view showing a state in which the sensor 002 of the abnormality degree determination system is actually mounted on the blade 100 of the wind power generator. although In this case, the sensor 002 is mounted on the blade 100 of the wind power generator, but other system construction machines are installed at a place separated from the wind power generator, making it possible to perform the following situations: for example, the operator performs monitoring. The wind power generator has: a blade 100 of a wind power generator that is subjected to wind and rotates; and a nacelle 6 that supports the blade 100 to be rotatable through the hub 2 and supports the load of the blade 100; and supports the nacelle 6 to be rotatable (yaw) Rotation; also known as oscillating rotation. The rotation of the blades is transmitted to the speed increaser 4 through the main shaft 3, and after the speed increase by the speed increaser 4, the rotational energy is transmitted to the generator 5. The generator 5 is a member that performs the power generation operation by rotating the rotor by the rotational energy.

According to this embodiment, it is possible to know the degree of damage caused by the lightning strike and whether it can continue to operate. By knowing the degree of damage caused by a lightning strike, it is possible to prevent the wind turbine blade from being operated in the case of "using a visual inspection to determine that the wind turbine blade is not abnormal after the lightning strike, and the wind turbine is operated again" Sudden failure of wind turbine blade damage caused by lightning strikes. In addition, by knowing the degree of abnormality of the wind turbine blade, it is possible to formulate a maintenance plan and an operation plan of the wind turbine that has considered the degree of abnormality.

In the above embodiment, the wind turbine blade has been described. However, the present invention is not limited to the "damage of the wind turbine blade", and can be applied to a wind turbine generator, such as a nacelle or a tower. machine. However, since the blade is placed at the highest position in the wind turbine and the possibility of lightning strike is the highest, it is particularly effective for the blade.

001‧‧‧Leaf Diagnostic System

002‧‧‧ sensor

003‧‧‧Lightning Information Reading Department

004‧‧‧Signal Processing Department

005‧‧‧Abnormality Department

005‧‧‧Abnormal Position Calculation Department

007‧‧‧Database

008‧‧‧Abnormal degree determination department

009‧‧‧Lightning Detection Department

010‧‧‧Diagnostic Results Joint Department

100‧‧‧ leaves

Claims (12)

An abnormality degree determining system for a wind power generator, comprising: a sensor for detecting a state of the wind power generator; and an abnormality determining unit, determining that the wind is generated according to the information measured by the sensor Whether the motor generates an abnormality; and a lightning strike information reading unit for reading lightning strike information of the wind power generator or lightning strike information around the wind power generator; and an abnormality degree determining unit, according to the lightning strike read by the lightning strike information reading unit The information and the determination by the abnormality determining unit determine the degree of abnormality; and the database stores the past lightning strike information, and the abnormality degree determining unit determines the degree of abnormality by using the past lightning strike information stored in the database. The abnormality degree determination system for a wind power generator according to the first aspect of the invention is characterized in that the recording unit for recording the number of lightning strikes to the wind turbine or the number of lightning strikes around the wind turbine is provided, and the abnormality degree determination unit is also provided. The degree of abnormality is determined based on the number of lightning strikes recorded by the recording unit. The abnormality degree determining system for a wind power generator according to claim 1 or 2, wherein the sensor is provided in the plurality of wind turbines by using a plurality of information obtained by the plurality of sensors. Compare and specify the abnormal position. The difference of wind turbines as described in item 3 of the patent application scope The normality determination system, wherein the sensor is disposed at a front end portion of the blade of the wind power generator, a root portion, and at least three positions of the front end portion and an intermediate portion of the root portion. In the abnormality degree determination system of the wind power generator according to the first or second aspect of the invention, the abnormality degree determination unit performs the determination of the abnormality level based on the abnormality index obtained from the abnormality determination unit, and corresponds to the past. The number of lightning strikes is performed to perform the aforementioned increase in the degree of abnormality. The abnormality degree determination system for a wind power generator according to claim 5, wherein in the case where the lightning strike is directly applied to the wind turbine, the abnormality is further caused than when the lightning strikes the periphery of the wind turbine Increased in degree. A wind power generator comprising: an abnormality degree determining system for a wind power generator as recited in claim 1, 2, 3 or 4; and a blade of a wind power generator rotating under wind; and The blade is supported as a nacelle that is rotatable and supports the load of the blade; and the nacelle is supported as a tower that can be rotated and rotated. A method for determining an abnormal degree of a wind power generator, comprising: a sensor for detecting a state of a wind power generator, and determining whether the wind power generator is different according to information measured by the sensor Usually, the lightning strike information of the wind power generator or the lightning strike information around the wind power generator is read, and the degree of abnormality is determined based on the read lightning strike information and whether the abnormality is generated, and the abnormality degree is determined. The accumulated lightning strike information after the accumulation is executed. The method for determining an abnormal degree of a wind power generator according to claim 8, wherein the abnormality degree is determined based on a past lightning strike number of the wind power generator or a number of lightning strikes around the wind power generator. The method for determining an abnormal degree of a wind power generator according to claim 9, wherein the sensor is configured to compare a plurality of waveforms obtained by the plurality of sensors, and An abnormal position is specified based on the time difference of the waveform information. The method for determining the degree of abnormality of a wind power generator according to the eighth aspect, the ninth aspect, or the tenth aspect, wherein the abnormality level is determined based on the abnormality index, and the degree of abnormality is increased in accordance with the number of past lightning strikes. The method for determining the degree of abnormality of a wind power generator according to claim 11, wherein in the case where the lightning strike is directly applied to the wind turbine, the abnormality is further caused than when the lightning strikes the periphery of the wind turbine Increased in degree.