• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The invention relates to a system for diagnosing (1) a rotor unbalance of a wind turbine from acceleration data measured at a nacelle of the wind turbine which is supported by a tower. According to the invention, the diagnostic system (1) comprises: - a measuring device (7) equipped with a vibration sensor, three axes (X, Y, Z) which is adapted to measure corresponding acceleration data vibration phenomena occurring at the nacelle; and a processing system (8) for the acceleration data adapted to determine the imbalance of the rotor as a function of the acceleration data measured on at least two axes (X, Y, Z) at the nacelle. The invention also relates to a method for diagnosing the rotor imbalance of a wind turbine.
    公开号:FR3073496A1
    申请号:FR1760715
    申请日:2017-11-15
    公开日:2019-05-17
    发明作者:Jerome Imbert；Bruno Pinto；Kevin MICHEL
    申请人:Sereema；
    IPC主号:
    专利说明:

    System and method for diagnosing a rotor imbalance of a wind turbine
    The present invention enters the field of sustainable development, renewable energies and more particularly wind energy. Indeed, the invention relates to a system for diagnosing a rotor imbalance of a wind turbine.
    In the context of the development of renewable energies, many wind farms have established themselves quickly. The operators of these wind farms had to put in place maintenance and monitoring solutions for the operating condition of each of the wind turbines that make up their wind farm. Thus, the operating condition of each wind turbine is essential to optimize energy production and prevent any risk of damage.
    In this context, it is known that the imbalance -rotor of a wind turbine can be a source of information -quant to the wear of the blades, the physical state of the blades or their state of assembly.
    Technical solutions have therefore been developed to optimize the operation of a wind turbine.
    A first solution described in document US 2012/0183399 consists of a balancing system for the rotor and the blades of a wind turbine. For this purpose, the balancing system uses a measuring device which comprises a vibration sensor and a sensor - of the speed of rotation - of the rotor of the electromagnetic sensor type. Through in particular a method of measuring the imbalance with N different pitch configurations -of the blades followed, on the one hand, by a calculation step to estimate the necessary pitch correction of the blades, and on the other hand, d 'a step to correct the pitch of the blades. The balancing system corrects the rotor imbalance identified during the rotor rotation phases. It should be noted that the pitch of a blade corresponds to its angle of inclination relative to a reference zero at the level of the rotor hub.
    In order to correct the rotor imbalance, the balancing system uses the nacelle acceleration data measured by the vibration sensor, the blade speed data measured by the rotor speed sensor and the known pitch of the blades.
    Depending on the type of wind turbine, such a balancing system involves a modification of the wind turbine settings to make measurements with N pitch pitch configurations of the blades. Such changes in pitch configuration can result in lower energy production. In addition, the rotor imbalance is measured over short operating periods which are not very representative of the different operating conditions of a wind turbine. For example, depending on weather conditions to which a wind turbine is subjected, such as wind direction and force, temperature, rain, snow, frost, etc. In this context, measurements over a short period do not represent the variability of the vibrations that apply to the rotor. Thus, this process may require new interventions and increase the maintenance cost of the wind turbine.
    A second solution described by document WO 2016/169964 consists of a balancing system which uses a system for adjusting the pitch of each blade of the rotor and a device for measuring the rotor imbalance. More particularly, the measuring device comprises an angle detector adapted to detect the pitch of each blade, a sensor of the speed of rotation of the rotor and at least one vibration sensor of the nacelle. The balancing system determines the imbalance of the rotor during a few periods of rotation of the rotor, for example two or three, each rotation being chosen according to a predefined rotation speed of the rotor. Thus, the imbalance of the rotor is determined as a function of the vibration data of the nacelle measured by the vibration sensor and of the data of rotational speed of the rotor measured by the sensor of rotational speed of the rotor. In a second step, the balancing system corrects the pitch of one or more blades so as to correct the determined rotor imbalance. The balancing system thus repeats these steps until the rotor imbalance is eliminated. This balancing system also has the drawback of modifying the settings of the wind turbine to make measurements with N pitch configurations of the blades. Such changes in pitch configuration can result in lower energy production. In addition, the diagnosis is carried out over periods not representative of the operating conditions of a wind turbine.
    A third solution described by document WO 2009/129617 consists of a system for balancing the rotor imbalance of a wind turbine. For this purpose, the balancing system includes a measuring device equipped with a rotor speed sensor and three vibration sensors, which are formed by accelerometers. From time to time, the sensors are installed precisely at the level of the nacelle in order to measure vibrations along the axis of rotation of the rotor and along two axes parallel to each other and perpendicular to the axis of rotation of the rotor. .
    Once the measuring device is installed, a test battery is carried out so as to determine the two components of the rotor imbalance of a wind turbine, namely the aerodynamic imbalance of the rotor and the mass imbalance of the rotor.
    In order to determine the aerodynamic imbalance of the rotor, measurements of acceleration data are carried out during a period of operation of the wind turbine. the acceleration data which correspond to the torsional vibrations of the nacelle and to the axial vibrations are compared with reference data obtained from a reference rotation cycle. When the acceleration data exceeds a certain threshold it means that the rotor has an aerodynamic imbalance. To detect the source of the problem, visual inspection and / or realignment of the blades is performed.
    In addition, the mass imbalance is determined via measurements of acceleration data carried out during a period of operation of the wind turbine during which a mass is secured to a blade. Like the first two solutions cited in the state of the art, this third balancing system requires stopping the production of energy to determine the rotor imbalance and correct it. In addition, the rotor imbalance is also measured over a short period which is not very representative of the operating conditions of a wind turbine.
    In this context, the applicant has developed a system for diagnosing the imbalance of the alternative and innovative rotor. Through continuous and systematic acceleration data measurements and analyzes during significant periods of wind turbine operation, the diagnostic system provides reliable data without interrupting power production and without changing settings of the wind turbine.
    To this end, a first aspect of the invention relates to a system for diagnosing a rotor imbalance of a wind turbine from acceleration data measured at the level of a nacelle of the wind turbine which is supported by a tower, characterized in that it comprises:
    a measuring device equipped with a three-axis vibration sensor X, Y, Z which is suitable for measuring the vibrational phenomena occurring at the level of the nacelle; and
    - an acceleration data processing system adapted to determine the imbalance of the rotor as a function of the acceleration data measured on at least two axes X, Y, Z at the level of the nacelle.
    The diagnostic system of the invention makes it possible to determine a rotor imbalance through a vibration analysis which is based solely on measurements of acceleration data which correspond to the vibration phenomena occurring at the level of the nacelle.
    According to a first characteristic of the first aspect of the invention, the processing system comprises an analysis module adapted to analyze acceleration data measured along at least one axis X, ï, Z during an operating period Tf, the module analysis to identify vibrations V3P due to the passage of a blade in front of the tower along at least one axis X, V, Z. According to the invention the analysis module performs systematic analysis operations continuously measured vibrational phenomena and makes it possible to determine if the wind turbine is in a period of operation by identifying the vibrations V3P.
    According to a feature of the first characteristic of the first aspect of the invention, the analysis module identifies the V3P vibrations after an operation of refining the measured acceleration data. More particularly, the refining operation includes:
    a step of projecting the acceleration data, collected in a three-dimensional frame X '', X '', Z '' of the rotor;
    a step of forming a frequency spectrum of the acceleration data projected in the three-dimensional frame X '', Y '', Z '' of the rotor;
    - a step of oversampling the frequency spectrum; and
    a step of selection of peaks corresponding to the V3P vibrations which is carried out by the product of each peak of the frequency spectrum oversampled with the peak of at least the first five harmonics of said peak.
    According to a second characteristic of the first aspect of the invention, the processing system comprises a periodicity module which is adapted to determine and record a regular rotation period Tr of the rotor as a function of the vibrations V3P identified. To this end, the periodicity module scans the vibrations V3P identified during an operating period Tf so as to detect at least one vibration V3P whose frequency is close to a reference frequency 3P of the wind turbine. Thus, when the periodicity module detects at least one vibration V3P whose frequency is close to the value of the theoretical frequency 3P of the wind turbine, the periodicity module preselects the operating period Tf.
    In this context, when the periodicity module selects a threshold number NS of operating periods Tf which follow one another chronologically, the periodicity module records a regular rotation period Tr of the rotor.
    According to a third characteristic of the first aspect of the invention, the processing system comprises a frequency processing module suitable * for quantifying along at least two axes X '', ï '', Z '' of the vibrations V1P of the nacelle due to an imbalance of the rotor during a period of regular rotation Tr recorded.
    For these purposes, the frequency processing module quantifies vibrations V1P whose frequency is close to a reference IP frequency of the wind turbine.
    According to a fourth characteristic of the first aspect of the invention, the processing system comprises a calculation module adapted to determine the imbalance of the rotor as a function of the vibrations V1P along at least two axes X '', ï '', Z '' which have been quantified during a determined number Nd of regular rotation periods Tr recorded.
    Advantageously, the processing of the acceleration data measured over a plurality of stable periods of operation of the wind turbine, that is to say periods when the rotor rotates at regular speed, makes it possible to determine a rotor imbalance on a large number of operating periods and thus increase the accuracy of the diagnosis. Indeed, the processing of large number of operating periods makes it possible to reduce the impact of variations in the measured vibrational phenomena due to the external conditions to which the wind turbine is subjected.
    The diagnostic system according to the invention makes it possible to determine the rotor imbalance based solely on acceleration data which can be measured without stopping the production of energy and without intervening on the wind turbine and its settings.
    The calculation module determines, as a function of the vibrations V1P along at least two axes X '', Y '', Z '', an aerodynamic imbalance of the rotor and a mass imbalance of the rotor which form the rotor imbalance.
    Advantageously, the diagnostic system performs simultaneous processing of the V1P acceleration data along at least two axes making it possible to simultaneously determine the aerodynamic and mass imbalances of the rotor.
    According to a feature of the previous four characteristics of the first aspect of the invention, the analysis module, the periodicity module, the frequentxel processing module and the calculation module are algorithms adapted to be executed by at least one computer terminal.
    According to a fifth characteristic of the first aspect of the invention, the measuring device is arranged on an axis parallel to the longitudinal axis of the nacelle. This positioning of the measuring device makes it possible to project the acceleration data into the three-dimensional frame of the rotor.
    A second aspect of the invention relates to a method for diagnosing a rotor imbalance of a wind turbine from acceleration data measured at the level of a wind turbine nacelle which is supported by a tower, characterized in that it comprises :
    a step of measurement along three axes X, ï, Z of acceleration data corresponding to the vibrational phenomena occurring at the level of the nacelle; and
    a step of processing the acceleration data so as to determine the imbalance of the rotor as a function of the acceleration data measured on at least two axes X, ï, Z at the level of the nacelle.
    Other particularities and advantages will appear in the detailed description which follows, of a non-limiting exemplary embodiment of the invention, which is illustrated by FIGS. 1 to 9 placed in the appendix and in which:
    FIG. 1 is a schematic representation of a system for diagnosing the rotor imbalance of a wind turbine which is in accordance with an exemplary embodiment of
    The invention;
    Figure 2 is a representation of a raw vibration signal of acceleration data;
    FIG. 3 is a simplified representation of an oversampled frequency spectrum of acceleration data.
    - Figure 4 is a simplified representation of a transformed frequency spectrum;
    - Figure 5 is a representation of a histogram corresponding to a distribution of the quantized axial vibrations of the rotor for which a type adjustment
    Weibull has been realized;
    - Figure 6 is a representation of a h i s togr amme corresponding to a distribution of the quantized transverse vibrations of the rotor for which a Gaussian type adjustment has been made;
    - Figure 7 is a graphical representation of an axial correction factor;
    FIG. 8 is a graphical representation of a distribution curve of the aerodynamic imbalance of the rotor; and
    - Figure 9 is a schematic representation of a method of diagnosing the rotor imbalance of a wind turbine performed by the diagnostic system of Figure 1.
    The invention relates to a system 1 for diagnosing a rotor imbalance of a wind turbine.
    Generally speaking, a wind turbine consists of a tower which is anchored to the ground or to the seabed. The tower supports a nacelle at the level of which the rotor is formed. The rotor extends along a longitudinal axis which merges with its axis of rotation X and has a pitch angle relative to the longitudinal axis X 'of the nacelle. In general, the pitch angle of the axis of rotation X '' of the rotor is 5 °. Here, and in most cases, the rotor has three blades.
    As illustrated in FIGS. 1 and 9, the diagnostic system 1 makes it possible, through a diagnostic process, to determine the rotor imbalance from acceleration data measured at the level of the nacelle of the wind turbine. The acceleration data correspond to vibrational phenomena that the nacelle undergoes during the operation of the wind turbine. The origins of a vibrational phenomenon can be multiple and depend on weather conditions, disturbances induced by the rotation of the rotor, or even disturbances induced by the passage of a blade in front of the tower etc.
    As illustrated in FIG. 1, in order to measure the vibrations undergone by the nacelle, the diagnostic system 1 comprises a measurement device 7 which is equipped with a vibration sensor having three measurement axes X, Y ′, Z, Le vibration sensor makes it possible to measure acceleration data which correspond to vibrational phenomena. The measured acceleration data can be represented in the form of a raw vibration signal as illustrated in FIG. 2. Here, the vibration sensor is formed by an accelerometer with three measurement axes.
    X Y Z.
    The measuring device 7 is positioned on the nacelle so that one of the measuring axes X, Y, Z of the vibration sensor is positioned parallel to the longitudinal axis X '. Here, the measurement axis X is positioned parallel to the longitudinal axis X '. Preferably, the measuring device 7 is rigidly secured to a weather mast of 1 / wind turbine.
    Once positioned, the measurement device 7 is put into operation and performs a systematic and continuous measurement operation of the acceleration data, along the three measurement axes X, Y, Z.
    the measuring device 7 is adapted to transfer, the acceleration data measured at the level of the nacelle, to a processing system 8 which is adapted to determine the rotor imbalance from the acceleration data measured on · at least two axes (X '', T '', 2 '').
    In the example illustrated in FIG. 1, the processing system S comprises an analysis module 9 which receives the acceleration data transferred by the measuring device 7.
    In addition, the measuring device 7 may include a memory enabling it to store the data collected locally.
    Here, the analysis module 9 is an algorithm stored and executed by a computer terminal 10 which can be on board at the level of the wind turbine or be located at a distance from the wind turbine.
    In the case where the computer terminal 10 is located at a distance from the wind turbine, the measurement device 7 comprises means of transmission through a telecommunications network, such as a GSM or satellite network.
    In order to limit interventions at the nacelle of the wind turbine, the measuring device 7 is supplied with electrical energy by a battery or by the wind turbine.
    More specifically, the analysis module 9 is integrated into the diagnostic system 1, and is adapted to continuously analyze the acceleration data transmitted by the measurement device 7 so as to determine an operating period Tf of the wind turbine . Preferably, the operating period Tf is between 15 seconds and 90 seconds and preferably the operating period Tf is between 20 and 60 seconds.
    An operating period Tf is determined when the analysis module 9 identifies vibrations V3P of the nacelle, along at least one axis X, X, Z, which are induced by the passage of a blade in front of the tower.
    For this purpose, the analysis module 9 performs a process for refining the acceleration data which makes it possible to obtain a frequency spectrum as illustrated in FIG. 3. The frequency spectrum of FIG. 3 comprises, on the one hand , on the abscissa the frequency of the measured vibration, and on the other hand, on the ordinate, the level of acceleration of the vibratory signal which corresponds to the intensity of the measured vibration.
    The process of refining the measured acceleration data makes it possible to highlight, as a function of their acceleration level and their frequency, the vibrations V3P of the nacelle, which correspond to the vibrations induced by the passage of a blade in front the wind turbine tower.
    To this end, the refining process uses a succession of processing operations · so as to improve the readability of the frequency spectrum by leaving only the V3P vibrations corresponding to the passage of a blade in front of the tower. In the example in FIG. 4, only one peak remains, it corresponds to a vibration V3P of the nacelle.
    With this in mind, the refining process includes a first operation of projecting the acceleration data into a three-dimensional frame X ', Y', Z 'of : the nacelle. The gravity vector G which is considered to be relatively constant by nature and corresponds to the axis Z ′ of the nacelle.
    In addition, the gravity vector G is used as a reference to project, through a first rotation matrix, the acceleration data X, Y, Z into the three-dimensional frame Χ ', Υ', Ζ 'of the nacelle .
    Whatever the inclination of the measuring device 7, the projection of the acceleration data into the reference frame X ', Τ', Z 'makes it possible to match the axis Z' with the gravity vector
    G.
    refining involves operation
    process
    a second
    rotation
    a second
    landmark since
    landmark
    indicated
    rotor present
    landmark
    The two operations for projecting the acceleration data make it possible, on the one hand, to limit the positioning constraints of the measuring device 7, and on the other hand, to project the acceleration data in the three-dimensional frame X '', T '', Z '' of the rotor on which the forces of the rotor imbalance are reflected.
    The refining process includes an operation of forming a frequency spectrum of the acceleration data projected in the three-dimensional frame X '', T '', Z '' of the rotor. Typically, the frequency spectrum formation operation can be performed using a Fourier transform.
    In order to improve the resolution of the frequency spectrum, the refining process includes a step of oversampling the frequency spectrum. This oversampling step contributes to improving the precision of the identification of the peaks and / or signals corresponding to the V3P vibrations (illustrated in FIG. 3).
    The refining process includes a signal identification operation corresponding to the V3P vibrations of the nacelle. In order to perform this operation, the analysis module 9 generates a transformed frequency spectrum by producing the product of each peak and / or signal of the oversampled frequency spectrum with the peak and / or signal of at least the first five harmonics of said pie and / or signal.
    Indeed, only the vibrations induced by the passage of a blade in front of the tower excite the fundamental vibration frequency and at least five of its first harmonics on the X axis ''. Thus, the product of each peak and / or signal of the oversampled frequency spectrum with the peak and / or signal of at least the first five harmonics of the oversampled frequency spectrum allows only the V3P vibrations corresponding to the passage of a blade to remain. in front of the tower.
    The analysis module 9 identifies the fundamental of the vibrations
    V3P by a selection of significant peaks compared to the noise of the transformed frequency spectrum illustrated in FIG. 4. In practice, each significant peak corresponds to a V3P vibration and the selection of significant peaks can be carried out over a predetermined frequency range which depends on the data. wind turbine manufacturers. For example, the analysis module 9 can be configured to select significant peaks when they are in a frequency range between 0.1 Hz and 2 Hz.
    Through the refining process, the analysis module 9 makes it possible to determine a frequency 3P of the rotor by extracting the fundamental of the vibrations V3P from the acceleration data measured by the measuring device 7. The frequency 3P of the rotor corresponds to the frequency passage of a blade in front of the tower.
    As illustrated in FIG. 1, the processing system 8 comprises a periodicity module 11 which is adapted to record in a memory M of the computer terminal period Tr of regular rotation of the rotor as a function of the vibrations V3P identified. The memory M of the computer terminal 10 can be constituted by a hard disk or a flash memory.
    In the present example, the periodicity module 11 is an algorithm stored and executed by the computer terminal 10.
    Through a process for analyzing the vibrations V3P identified on a transformed frequency spectrum of successive operating periods Tf, the periodicity module 11 creates and records a period Tr in the memory M of the computer terminal 10.
    To this end, the analysis method carried out by the periodicity module 11 comprises an operation of scanning the vibrations V3P identified on the transformed frequency spectrum of an operating period Tf. During the scanning operation, the periodicity module 11 compares the frequency of the identified vibrations V3P with a reference frequency 3P of the wind turbine.
    Here, the reference frequency 3P of the wind turbine corresponds to a theoretical frequency 3P which depends on the type of wind turbine and relates to a known manufacturer data. The theoretical 3P frequency corresponds, for example, to the value of the 3P frequency during nominal operation of the rotor, where the forces caused by the imbalance of the rotor are exerted on the rotor as much as possible.
    When the periodicity module 11 detects on the transformed frequency spectrum of an operating period Tf, at least one vibration V3P whose frequency is close to the reference frequency 3P of the wind turbine, the periodicity module 11 performs an operation of preselection of said operating period Tf. In practice, said preselected operating period Tf is stored temporarily and / or permanently in a memory Mv of the periodicity module 11.
    The memory Mv of the periodicity module 11 can be constituted by a random access memory, a hard disk or even a flash memory.
    When the periodicity module 11 preselects a threshold number NS of operating period Tf which follow one another chronologically, the periodicity module 11 performs an assembly operation of the threshold number NS of successive operating periods Tf to create a regular rotation period Tr rotor.
    In the present example, the threshold number NS of preselected operating periods Tf necessary to create a regular rotation period Tr can be between 5 and 30 preselected operating periods Tf and preferably between 10 and 20 preselected operating periods Tf.
    Here, each regular rotation period Tr determined via a vibration analysis corresponds to a period of 2 to 15 minutes during which the rotor rotates at a regular speed.
    In this example, the periodicity module 11 records each period Tr in the memory M of the computer terminal 10.
    In addition, in the example of FIG. 1, the processing system 8 of the diagnostic system 1 comprises a frequency processing module 12 adapted to determine vibrations V1P of the nacelle which are induced by the imbalance of the rotor.
    In the present example, the frequency processing module 12 is an algorithm stored and executed by the computer terminal
    10.
    Through a frequency processing method of a regular rotation period Tr recorded in the memory M, the frequency processing module 12 determines vibrations V1P whose frequency is close to a reference frequency 1P of the wind turbine.
    The frequency 1P corresponds to the speed of rotation of the rotor. When the rotational speed of the rotor is stable, the frequency IP is also stable.
    Furthermore, in this example, the reference frequency IP of the wind turbine corresponds to a theoretical IP frequency which depends on the type of wind turbine and relates to a known manufacturer data. The theoretical IP frequency corresponds, for example, to the value of the IP frequency during nominal rotor operation during which the forces caused by the imbalance of the rotor are exerted on the rotor as much as possible.
    The frequency module 12 will quantify ie, over a regular rotation period Tr of the vibrations V1P whose value is between -20% and + 20% of the reference IP frequency, preferably the frequency processing module 12 quantifies VIP vibrations whose value is between 15% and + 15% of the reference IP frequency and preferably the frequency processing module 12 quantifies V1P vibrations whose value is between -10% and + 5% of the frequency Reference IP.
    For this purpose, the frequency processing module 12 performs a step of quantifying the acceleration level of the axial and transverse vibrations V1P around the reference frequency 1P, that is to say, the quantification of the vibrational intensity of the VIP axial vibrations, along the X '' axis, and transverse VIP vibrations, along the Y axis.
    In the example illustrated in FIG. 1, the processing system 8 of the diagnostic system 1 comprises a calculation module 13 adapted to determine the unbalanced rotor as a function of the axial VIP and transverse VIP vibrations of a determined number Nd of periods of regular Tr rotation recorded in memory M,
    Here, the calculation module 13 is an algorithm stored and executed by the computer terminal 10.
    In this example, the determined number Nd of regular rotation periods Tr is between 100 and 400, and preferably the determined number Nd of regular rotation periods Tr is between 200 and 300.
    In order to determine the rotor imbalance, the calculation module 13 performs statistical processing operations which make it possible to obtain a global value of the axial and transverse vibrations V1P for the determined number Nd of regular rotation periods Tr in the memory M. In this example, the calculation module 13 independently processes the global values of the axial and transverse vibrations V1P.
    In known manner, the rotor imbalance of a wind turbine has two components, an aerodynamic imbalance of the rotor and a mass imbalance of the rotor.
    It has been shown that the V1P axial vibrations are mainly the result of aerodynamic imbalance and mainly -induced by different aerodynamic behavior between each blade of the wind turbine. -V1P axial vibrations vary with the force of the wind. In addition, the aerodynamic imbalance of a rotor can be due to an error in the assembly of the blades such as a pitch angle shift of one or more blades, to a deformation of the blades which can be induced. by lightning strikes, differentiated wear of the blades, or a manufacturing difference between each blade.
    In the case of transverse vibrations V1P, they mainly result from a mass imbalance of the rotor and are mainly induced by the rotation of the rotor. The mass imbalance of the rotor can be due to a problem of cracks in a blade - which can favor an accumulation of water in this blade, or a deposit of material on the blades, for example a deposit of frost.
    FIG. 5 shows a distribution of the axial vibrations V1P for the determined number Nd of regular rotation periods Tr according to a first histogram 14. The histogram 14 presents on the abscissa axis the acceleration values of the axial vibrations V1P and on l axis of ordinates the proportion and / or the quantity of axial vibrations V1P. The distribution is carried out by the calculation module 13.
    In order to determine an overall value of the axial V1P vibrations, it is necessary to interpret the distribution of the axial V1P vibrations on the histogram 14.
    As previously indicated, when the rotor speed is constant, the specificity of the distribution of the axial VIP vibrations is due to different wind speeds which apply to the rotor and the blades. To interpret such a phenomenon, the calculation module 13 uses a statistical model such as Weibull's law. A global value of the axial VIP vibrations is determined through the calculation of parameter A of the law of
    Weibull which represents the characteristic value of the distribution of the values of the axial VIP vibrations.
    In the example illustrated in FIG. 5, the calculation module 13 adjusts a Weibull curve 16 to the distribution of the vibrations
    V1P axial so as to determine the parameter A of the adjusted Weibull curve.
    The figure 6 present a distribution of the vibration VIP 15 transversal for the number determined nd of periods of rotation. Tr regular. according to a second hi stogram 17.
    The axis histogram
    abscissa
    acceleration.
    present on
    Transversal VIP
    sure
    axis
    ordered the proportion and / or the quantity of transverse VIP vibrations. The distribution is carried out by the calculation module. 13.
    In order to determine an overall value of the transverse VIP vibrations, it is necessary to interpret the distribution of the transverse V1P vibrations on the histogram 17.
    As indicated previously, the transverse VlP vibrations vary mainly with the speed of rotation of the rotor.
    However, the regular rotation periods Tr selected are characterized by a constant speed of rotation of the rotor. By way of facts and as illustrated in FIG. 6, the distribution of the transverse VIP vibrations on the histogram presents a conventional profile of Gaussian distribution. In this context, a Gaussian function constitutes a relevant model for determining an overall value characteristic of the distribution of transverse VIP vibrations.
    Thus, the calculation module 13 adjusts a Gaussian curve 18 on the histogram 17 of distribution of the transverse V1P vibrations using a Gaussian function. In this example, the overall value of the transverse VlP vibrations corresponds to the mean value μ of the adjusted Gaussian curve 18.
    In addition, as has been mentioned, aerodynamic imbalance is the main source of VlP axial vibrations. In known manner, the mass imbalance of the rotor also constitutes an ancillary origin of the axial vibrations VlP. However, in proportion to the aerodynamic imbalance of the rotor, the mass imbalance of the rotor represents a smaller part of the intensity of the axial VlP vibrations,
    Conversely, the main origin of the VlP transverse vibrations is the mass imbalance of the rotor, the aerodynamic imbalance of the rotor constituting an ancillary origin of the VlP transverse vibrations. However, in proportion to the mass imbalance of the rotor, the aerodynamic imbalance of the rotor represents a smaller part of the intensity of the transverse VlP vibrations.
    In this context, for the same determined number Nd of regular rotation periods Tr, the calculation module 13 performs an operation of correcting the overall values of the axial VlP vibrations A and transverse μ relative to each other.
    Preferably, the operations for correcting the overall values of the axial vibrations A and transverse μ are independent ',
    To this end, the calculation module 13 can use a logistic function which makes it possible to calculate a correction coefficient for each global value A, μ.
    In the example of FIG. 7, is illustrated graphically, in the form of a curve of an axial correction coefficient of the global value A. Here, the axial correction coefficient is between 0 and 1. In practice, the more the axial correction coefficient has a value close to the value 1, the lower the share of transverse V1P vibrations in the intensity of axial vibrational phenomena.
    In order to obtain the correction coefficient for the overall value A, the calculation module 13 can use the logistic function below:
    axial correction factor (x) = where x = (overall value μ V1P transverse / overall value A V1P axial), and c = 0.25, k = l.1, x 0 = 2.5 can be used as coefficient values for a type d 'determined wind turbine.
    The calculation module 13 determines the aerodynamic imbalance of the rotor as a function of the product of the overall value A of the axial vibrations V1P and the axial correction coefficient.
    In the example of FIG. 8, the aerodynamic imbalance is represented graphically by a curve whose value varies between 0 and 1. The -value 0 corresponding to the perfect balance of the rotor, while the value 1 corresponds to an imbalance theoretical maximum that cannot be reached. In practice, when the diagnostic system 8 determines an aerodynamic imbalance which is greater than a value between 0.2 and 0.3, it. it is advisable to work on the wind turbine in question in order to avoid any damage to the blades or the rotor.
    In order to determine the aerodynamic imbalance of the rotor, the calculation module 13 uses a logistic function:
    aerodynamic imbalance diagnosis (x) = where x = overall value A axial V1P * axial correction factor (overall_transverse_VlP / overall_axial_axial_V'lP), and k = 0.9, x 0 = 2.75 which can be used as coefficient values for a type of wind turbine determined.
    Here, the axial and transverse values Vlp are expressed in mg (milig), the milig being a unit of acceleration.
    In the same way, the calculation module 13 determines a transverse correction coefficient of the global value μ for example using a logistic function of the same type as that described for the axial correction coefficient:
    f transverse corrective actor (x) = j ——- y
    WHERE X (global value A axial VlP / global value μ
    Transverse VIP), and c = 0.2, k = ll, x 0 = 1.75 which can be used as coefficient values for a given type of wind turbine.
    From a graphical point of view, the transverse correction factor presents a curve profile similar to that of the axial correction factor which is illustrated in FIG. 7.
    Similarly, the more the transverse correction coefficient has a value close to the value • of the axial VlP vibrations in the intensity
    1, the lower the share of transverse vibrational phenomena.
    The calculation module 13 determines the mass imbalance as a function of the product of the transverse correction coefficient and of the overall value of the transverse VlP vibrations. The mass imbalance of the rotor is determined using a logistics function of the same type as that used to determine the aerodynamic imbalance:
    mass imbalance diagnosis (x) where x = global value μ transverse VlP * transverse corrective factor Cvaleur_ <jlobale ^ axiale_VlP / global value_transverse_yiP), and k = 0.7, x 0 = 2.25 • which can be used as values of coefficients for a model determined wind turbine.
    The mass imbalance of the rotor can also be expressed on a scale of 0 to 1 according to an interpretation similar to the aerodynamic imbalance of the rotor, Namely, that the value 0 corresponds to the perfect balance of the rotor, while the value 1 corresponds to a theoretical maximum imbalance which cannot be reached.
    From a graphic point of view, the mass imbalance of the rotor presents a curve profile similar to that of the aerodynamic imbalance which is illustrated in FIG. 8.
    In practice, when the diagnostic system determines a mass imbalance which is greater than values between 0.2 and 0.3, it is also advisable to intervene on the wind turbine in question in order to avoid any damage to the level of blades and / or. rotor.
    The diagnostic system 1 operates continuously and makes it possible to transmit data continuously and in real time to the 15 operators of the wind farms.
    The diagnostic system 1 makes it possible to determine, through a diagnostic process, the aerodynamic and mass imbalance of the rotor without intervening in the normal operation of the wind turbine.
    Based on the information received by wind farm operators, they can optimize the management of their wind farms. The diagnostic system makes it possible in particular to guide the operators of wind farms in their procedures for maintaining the windscreen by identifying wind turbines that have a rotor imbalance.
    权利要求:
    Claims (10)
    [1" id="c-fr-0001]
    Resells! cation
    1. Diagnostic system (1) of a rotor imbalance of a
    wind turbine from measured acceleration data at level of a gondola of the wind turbine which is supported through a tower, characterized in what he includes:- a device for measure (7) equipped with a sensor of
    three-axis vibration (X, Y, Z) which is suitable for measuring acceleration data corresponding to the vibrational phenomena occurring at the level of the nacelle; and
    - a processing system (8) of acceleration data adapted to determine the imbalance of the rotor as a function of the acceleration data measured on at least two axes (X, Y, Z) at the level of the nacelle.
    [2" id="c-fr-0002]
    2. Diagnostic system
    d) according to claim 1, characterized in that the processing system comprises an analysis module (9) adapted to analyze acceleration data measured along, at least one axis (X,
    Y, Z) during an operating period (Tf), the analysis module (9) making it possible to identify vibrations (V3P) due to the passage of a blade in front of the tower along at least one axis (X, Y, Z).
    [3" id="c-fr-0003]
    3. Diagnostic system (1) according to claim 2, characterized in that the analysis module (9) identifies the vibrations (V3P) after 'a refining operation of the measured acceleration data, the refining operation comprising:
    a step of projecting the acceleration data collected in a three-dimensional coordinate system (X '', Y '', Zi ') of the rotor;
    a step of forming a frequency spectrum of the acceleration data projected in the three-dimensional frame (X '', Y '', Z '') of the rotor;
    a step of oversampling of the frequency spectrum;
    and a step of selecting peaks corresponding to the vibrations (V3P) which is carried out by the product of
    [4" id="c-fr-0004]
    4.
    [5" id="c-fr-0005]
    5.
    each peak of the spectrum the peak of at least peak.
    Diagnostic system and 3, characterized in (1) this frequency oversampled with five first harmonics of said according to one of claims 2 that the processing system (8) comprises a periodicity module (11) ci. irx-ULS st ^ ïérrod.e regular rotor according to identified.
    Diagnostic system characterized in vibration operation vibration frequency module of what the (1) according to which is adapted to rotation (Tr) vibrations (V3P) claim 4, periodicity module (11) scans (V3P) identified during '' one (Tf) so as to detect (V3P) whose frequency is (3P) reference wind turbine, at the period of at least one neighbor of when the periodicity (11) detects at least (V3P) whose frequency (3P) periodicity operation a vibration frequency is close to the theoretical · of the wind turbine, (11) (Tf).
    value of the module preselects the period of
    Diagnostic system (1) according to one of claims 4 and 5, characterized in that • when periodicity (11) selects a number chronologically, the module of which registers • a period of the threshold module (NS) of succeed rotation periodicity (11)
    System m
    characterized comprises a frequency processing module system (12) adapted to quantify along at least two axes (X '', Y '', 2 '') vibrations (VIP) of the nacelle due to an imbalance of the rotor during a regular rotation period (Tr) recorded.
    [6" id="c-fr-0006]
    8. Diagnostic system (1) according to claim 7, characterized in that the frequency processing module (12) quantifies vibrations (VIP) whose frequency is close to a reference frequency (IP) of the wind turbine.
    [7" id="c-fr-0007]
    9. diagnostic system (1) according to one of claims 7 and 8, characterized in that the treatment system (8) comprises a calculation module (13) adapted to determine the imbalance of the rotor as a function of the vibrations (VIP ) according to at least two axes (X '', Y '', Z r ') which have been quantified during a determined number (Nd) of regular rotation periods (Tr) recorded.
    Diagnostic system (1) according to claim 9, characterized in that the calculation module (13) determines.
    as a function of the vibrations (VIP) quantified along at least two axes (X, Y, Z), an aerodynamic imbalance of the rotor which form the rotor and an imbalance of rotor imbalance.
    mass of the
    Diagnostic system (1) according to one of claims and
    [8" id="c-fr-0008]
    10, characterized in that the analysis module (9d, the periodicity module (
    [9" id="c-fr-0009]
    11), the frequency processing module (12) and the calculation module (13) are algorithms adapted to be executed by at least one computer terminal (10).
    [10" id="c-fr-0010]
    12. Method for diagnosing a rotor imbalance of a wind turbine from acceleration data measured at the level of a wind turbine nacelle which is supported by a tower, characterized in that it comprises:
    - a measurement step along three axes (X, Y, Z) of acceleration data corresponding to the vibrational phenomena occurring at the level of the nacelle; and a step of processing the acceleration data so as to determine the imbalance of the rotor as a function of the acceleration data measured on at least two axes (X, Y, Z) at the level of the nacelle.
    类似技术:
    公开号 | 公开日 | 专利标题
    EP3710693B1|2022-01-26|System and method for diagnosing a rotor unbalance of a wind turbine
    EP2496921B1|2020-10-21|Method for detecting damage in at least one engine roller bearing
    JP2019515192A|2019-06-06|Estimation of yaw misalignment for wind turbines
    FR3024502A1|2016-02-05|
    FR2919661A1|2009-02-06|SYSTEM AND METHOD FOR DETECTING A BASIC EXCENTRIC OFFSET OF A ROTOR
    CA2943397C|2019-10-15|Method for assessing whether or not a measured value of a physical parameter of an aircraft engine is normal
    FR2949220A1|2011-02-25|METHOD AND SYSTEM FOR DETECTING THE INGESTION OF AN OBJECT BY AN AIRCRAFT TURBOJUSTER DURING A MISSION
    FR2692037A1|1993-12-10|Method for diagnosing an evolutionary process.
    FR2958911A1|2011-10-21|METHOD AND SYSTEM FOR MONITORING THE OIL LEVEL CONTAINED IN A RESERVOIR OF AN AIRCRAFT ENGINE
    JP2010084770A5|2014-05-15|
    EP3250974B1|2021-09-08|Method, system and computer program for learning phase of an acoustic or vibratory analysis of a machine
    WO2015086959A1|2015-06-18|Method and device for detecting mechanical faults in a variable speed rotating machine
    CA2964815A1|2016-04-28|Method and device for monitoring an aircraft engine vane wheel via balance position measurement
    FR2988130A1|2013-09-20|DEFECT DETECTION SYSTEM ON AN AIRCRAFT ENGINE BEARING WHEEL
    CA2907602A1|2014-09-25|Method for monitoring deformation of a rotating element via a monitoring device employing optical fibre, and wind turbine equipped with such a device
    EP1792278A1|2007-06-06|Method for detecting and tracking punctual targets, in an optoelectronic surveillance system
    EP3650687B1|2021-06-16|Method for determining an induction factor for a wind turbine provided with a laser remote detection sensor
    FR2966597A1|2012-04-27|METHOD FOR MONITORING A ROTATING ELEMENT BELONGING TO A MECHANICAL TRANSMISSION OF AN AEROGENERATOR
    EP3862560A1|2021-08-11|Method for determining the wind speed in the rotor plane of a wind turbine
    EP3862563A1|2021-08-11|Method for determining the direction of the wind by means of a laser remote detection sensor
    FR3076580A1|2019-07-12|ESTIMATION OF THE GAME BETWEEN A MAT AND THE FOUNDATIONS OF A WINDMILL
    EP3578929A1|2019-12-11|Method and system for detecting the angular position of the blades of an impeller of a turbine engine
    CA3076097A1|2020-09-18|Method for predicting wind speed in the rotor plane of a wind turbine equipped with a laser remote sensor
    WO2022002563A1|2022-01-06|Method for determining wind velocity components by means of a laser remote sensor and by means of a temporal coherence
    FR3110100A1|2021-11-19|Method and system for determining causes of defects observed in laser welding processes
    同族专利:
    公开号 | 公开日
    EP3710693B1|2022-01-26|
    WO2019097167A1|2019-05-23|
    FR3073496B1|2020-11-20|
    EP3710693A1|2020-09-23|
    US20200363282A1|2020-11-19|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    EP1719910A1|2004-02-27|2006-11-08|Mitsubishi Heavy Industries, Ltd.|Wind turbine generator, active vibration damping method for the same, and wind turbine tower|
    WO2009000787A2|2007-06-25|2008-12-31|Siemens Aktiengesellschaft|Monitoring of blade frequencies of a wind turbine|
    EP2472238A1|2010-12-29|2012-07-04|Siemens Aktiengesellschaft|Determination of a vibrational frequency of a wind turbine rotor blade with a sensor device being placed at a structural component being assigned to and/or being part of the rotor|
    US20120183399A1|2011-01-19|2012-07-19|Hamilton Sundstrand Corporation|Method and apparatus for balancing wind turbines|
    WO2016169963A1|2015-04-23|2016-10-27|Envision Energy Aps|Method of correcting rotor imbalance and wind turbine thereof|CN111691963A|2020-06-09|2020-09-22|安徽江淮汽车集团股份有限公司|Method and device for detecting dynamic balance of automobile fan|CA2778216A1|2008-04-24|2009-10-29|Composotech Structures Inc.|A method and system for determining an imbalance of a wind turbine rotor|
    EP3601789B1|2017-03-21|2020-11-18|Vestas Wind Systems A/S|System and method to manage torsional oscillation of a wind turbine tower|US11208986B2|2019-06-27|2021-12-28|Uptake Technologies, Inc.|Computer system and method for detecting irregular yaw activity at a wind turbine|
    US10975841B2|2019-08-02|2021-04-13|Uptake Technologies, Inc.|Computer system and method for detecting rotor imbalance at a wind turbine|
    法律状态:
    2019-05-17| PLSC| Publication of the preliminary search report|Effective date: 20190517 |
    2019-11-25| PLFP| Fee payment|Year of fee payment: 3 |
    2020-11-20| PLFP| Fee payment|Year of fee payment: 4 |
    2021-11-26| PLFP| Fee payment|Year of fee payment: 5 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1760715|2017-11-15|
    FR1760715A|FR3073496B1|2017-11-15|2017-11-15|SYSTEM AND METHOD FOR DIAGNOSING A ROTOR IMBALANCE OF A WIND TURBINE|FR1760715A| FR3073496B1|2017-11-15|2017-11-15|SYSTEM AND METHOD FOR DIAGNOSING A ROTOR IMBALANCE OF A WIND TURBINE|
    US16/764,767| US20200363282A1|2017-11-15|2018-11-15|System and method for diagnosing a rotor unbalance of a wind turbine|
    EP18819190.2A| EP3710693B1|2017-11-15|2018-11-15|System and method for diagnosing a rotor unbalance of a wind turbine|
    PCT/FR2018/052849| WO2019097167A1|2017-11-15|2018-11-15|System and method for diagnosing a rotor unbalance of a wind turbine|
    [返回顶部]