WO2023097938A1 - 风力发电机组变桨轴承的寿命评估方法和装置 - Google Patents

风力发电机组变桨轴承的寿命评估方法和装置 Download PDF

Info

Publication number
WO2023097938A1
WO2023097938A1 PCT/CN2022/083377 CN2022083377W WO2023097938A1 WO 2023097938 A1 WO2023097938 A1 WO 2023097938A1 CN 2022083377 W CN2022083377 W CN 2022083377W WO 2023097938 A1 WO2023097938 A1 WO 2023097938A1
Authority
WO
WIPO (PCT)
Prior art keywords
pitch
estimated
life
driving torque
probability density
Prior art date
Application number
PCT/CN2022/083377
Other languages
English (en)
French (fr)
Inventor
张鹏飞
苏慧丽
韩炜
赵世培
Original Assignee
北京金风科创风电设备有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北京金风科创风电设备有限公司 filed Critical 北京金风科创风电设备有限公司
Priority to KR1020247002597A priority Critical patent/KR20240023651A/ko
Priority to AU2022401404A priority patent/AU2022401404A1/en
Priority to EP22899761.5A priority patent/EP4343587A1/en
Publication of WO2023097938A1 publication Critical patent/WO2023097938A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
    • F03D—WIND MOTORS
    • F03D15/00—Transmission of mechanical power
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
    • F03D—WIND MOTORS
    • F03D17/00—Monitoring or testing of wind motors, e.g. diagnostics
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
    • F03D—WIND MOTORS
    • F03D17/00—Monitoring or testing of wind motors, e.g. diagnostics
    • F03D17/005—Monitoring or testing of wind motors, e.g. diagnostics using computation methods, e.g. neural networks
    • F03D17/006—Estimation methods
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
    • F03D—WIND MOTORS
    • F03D17/00—Monitoring or testing of wind motors, e.g. diagnostics
    • F03D17/009—Monitoring or testing of wind motors, e.g. diagnostics characterised by the purpose
    • F03D17/011—Monitoring or testing of wind motors, e.g. diagnostics characterised by the purpose for monitoring mechanical loads or assessing fatigue; for monitoring structural integrity
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
    • F03D—WIND MOTORS
    • F03D17/00—Monitoring or testing of wind motors, e.g. diagnostics
    • F03D17/027—Monitoring or testing of wind motors, e.g. diagnostics characterised by the component being monitored or tested
    • F03D17/032—Bearings
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F16—ENGINEERING ELEMENTS AND UNITS; GENERAL MEASURES FOR PRODUCING AND MAINTAINING EFFECTIVE FUNCTIONING OF MACHINES OR INSTALLATIONS; THERMAL INSULATION IN GENERAL
    • F16C—SHAFTS; FLEXIBLE SHAFTS; ELEMENTS OR CRANKSHAFT MECHANISMS; ROTARY BODIES OTHER THAN GEARING ELEMENTS; BEARINGS
    • F16C19/00—Bearings with rolling contact, for exclusively rotary movement
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M13/00—Testing of machine parts
    • G01M13/04—Bearings
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
    • F03D—WIND MOTORS
    • F03D80/00—Details, components or accessories not provided for in groups F03D1/00 - F03D17/00
    • F03D80/70—Bearing or lubricating arrangements
    • F03D80/701—Pitch or yaw bearings
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E10/00—Energy generation through renewable energy sources
    • Y02E10/70—Wind energy
    • Y02E10/72—Wind turbines with rotation axis in wind direction

Definitions

  • the present disclosure relates to the field of wind power generation, and more particularly, to a method and device for evaluating the life of a pitch bearing of a wind power generating set.
  • the pitch bearing of the wind turbine is an important component connecting the blade and the hub. It is also used to transmit the load of the blade to the pitch system on the hub.
  • the pitch system is an important safety system of the wind turbine. If a failure occurs, there will be Potentially catastrophic. Therefore, during the operation of wind turbines, the life assessment of pitch bearings becomes particularly important and critical.
  • a life assessment method for a pitch bearing of a wind power generating set includes: obtaining the probability density of the pitch driving torque in M historical periods, where M is a positive integer; obtaining the The angular cumulative value of the pitch angle in each of the M historical periods; according to the pitch driving torque and its probability density in the M historical periods, and the angular cumulative value of the M historical periods, determine Equivalent load of the pitch bearing; from the equivalent load of the pitch bearing, the spent life of the pitch bearing is determined.
  • a life assessment device for a pitch bearing of a wind power generating set, the life assessment device comprising: a first acquisition unit configured to: acquire the probability density of the pitch driving torque in M historical periods , the M is a positive integer; the first acquisition unit is further configured to acquire the angular cumulative value of the pitch angle in each of the M historical periods; the equivalent unit is configured to: according to the variable The propeller driving torque and its probability density in the M historical periods, and the angular cumulative value of the M historical periods determine the equivalent load of the pitch bearing; the first calculation unit is configured to: according to the pitch The equivalent load of the bearing determines the spent life of the pitch bearing.
  • a computer-readable storage medium which, when the instructions in the computer-readable storage medium are executed by at least one processor, causes the at least one processor to perform the lifetime assessment method as described above .
  • a computer device comprising: at least one processor; at least one memory storing computer-executable instructions, wherein the computer-executable instructions, when executed by the at least one processor, cause The at least one processor executes the lifetime assessment method as described above.
  • This disclosure utilizes the operating data collected during the operation of the unit to realize the life evaluation of the pitch bearing, without additional data collection, and it is not necessary to configure additional data collection sensors, which can reduce product costs and save life evaluation costs. Time costs.
  • the present disclosure realizes the real-time online evaluation of the consumed life and the prediction of the remaining life in the future within the acceptable range of accuracy, and can perform predictive operation and maintenance and prediction of failure events on the premise of ensuring the safe operation of wind turbines. Thereby reducing the unplanned downtime of wind turbines and improving their economic benefits.
  • Fig. 1 is a flow chart illustrating a life assessment method of a pitch bearing of a wind power generating set according to an embodiment of the present disclosure.
  • Fig. 2 is a schematic flowchart illustrating an online life evaluation system for a pitch bearing according to an embodiment of the present disclosure.
  • Fig. 3 is a schematic flowchart illustrating a system for predicting the remaining life of a pitch bearing according to an embodiment of the present disclosure.
  • Fig. 4 is a block diagram illustrating a life evaluation device for a pitch bearing of a wind power generating set according to an embodiment of the present disclosure.
  • Fig. 5 is a block diagram illustrating a life evaluation device for a pitch bearing of a wind power generating set according to another embodiment of the present disclosure.
  • FIG. 6 is a block diagram illustrating a computer device according to an embodiment of the present disclosure.
  • first means “first”, “second” and “third” may be used herein to describe various members, components, regions, layers or sections, these members, components, regions, layers or sections should not be referred to as These terms are limited. On the contrary, these terms are only used to distinguish one element, component, region, layer or section from another element, component, region, layer or section.
  • a first member, a first component, a first region, a first layer, or a first portion referred to in examples described herein could also be termed a second member, a second component, or a first portion without departing from the teachings of the examples.
  • Fig. 1 is a flow chart illustrating a life assessment method of a pitch bearing of a wind power generating set according to an embodiment of the present disclosure.
  • the life assessment method can be realized by relying on the life assessment system, and the life assessment system can further include an online life assessment system for assessing the consumed life of the pitch bearing, and a remaining life prediction for assessing the estimated remaining life of the pitch bearing in the future system.
  • Fig. 2 is a schematic flowchart illustrating an online life evaluation system for a pitch bearing according to an embodiment of the present disclosure.
  • Fig. 3 is a schematic flowchart illustrating a system for predicting the remaining life of a pitch bearing according to an embodiment of the present disclosure.
  • step S101 the probability density of the pitch driving torque in M historical periods is obtained, where M is a positive integer. Since the pitch drive torque will affect the loss of the pitch bearing, this parameter is selected to realize the life evaluation of the pitch bearing. It should be noted that the pitch drive torque is the force that controls the blade rotation angle.
  • the cumulative running time of wind turbines often reaches several years. By dividing the cumulative running time into M historical periods according to a certain step size (for example, 10 minutes), the data can be processed separately for each historical period to ensure different The duration of the individual historical periods used in wind turbine assessments is consistent to ensure generalizability of the strategy.
  • the pitch driving torque usually changes within a certain interval, for example, the pitch driving torque of a wind turbine with a rated capacity of less than 5 MW is generally within the interval [-200KNm, 200KNm].
  • Multiple specific values can be selected from it, for example, according to the set step size, multiple specific pitch drive torques can be obtained, that is to say, the selected multiple pitch drive torques can form an arithmetic sequence, and then obtained in step S101 Probability density for each pitch drive torque for each historical period.
  • step S101 specifically includes: according to the operating data of M historical periods, determine the occurrence frequency and corresponding distribution parameters of the pitch driving torque in different pitch motion states (it should be understood that the correspondence here refers to the distribution parameters and The occurrence frequency corresponds to the pitch motion state), and the pitch motion state includes a positive state, a constant state, and a negative state; the probability density is determined according to the frequency and the corresponding distribution parameters.
  • the pitching motion state refers to the motion state that drives the blades to change pitch. For example, when the hydraulic rod that drives the blade root plate to rotate is gradually pushed out to the fully withdrawn state, the blade root plate drives the blades to gradually rotate to the feathering state.
  • the state of the pitching movement is the positive state; the opposite of the positive state is the negative state, that is, when the hydraulic rod that drives the blade root plate to rotate gradually returns to the hydraulic cylinder from the fully withdrawn state, the blade root plate drives the The blade rotates in the opposite direction gradually from the feathered state to the reversed pitch state.
  • the pitching motion state is in a negative state; when the hydraulic rod that drives the blade root plate to rotate remains stationary, the blade does not move, which can be regarded as pitching
  • the motion state is a constant state.
  • the operation data of the wind power generating set is collected by the data acquisition system as shown in FIG. step.
  • the operating data is specifically SCADA (Supervisory Control And Data Acquisition, that is, data acquisition and monitoring control system) data, including output power, impeller speed, generator torque, x-direction component of nacelle acceleration, y-direction component of nacelle acceleration, and pitch angle.
  • SCADA Supervisory Control And Data Acquisition
  • the pitch drive torque distribution evaluation module shown in Figure 2 can determine the distribution parameters of the pitch drive torque accordingly.
  • no additional data collection is required, and additional data collection sensors do not need to be configured, which can reduce product costs, save time and cost for life assessment, and improve economic performance.
  • three different pitching motion states can be obtained, and in each historical period, the pitching driving torque may appear in these three states at the same time, so it can be
  • the occurrence frequency and corresponding distribution parameters of the pitch driving torque in the three pitch motion states are determined, so as to describe the pitch driving torque more accurately.
  • the pitch movement usually only has the above three states, so the sum of the frequency of occurrence of the three is 1, and the frequency of occurrence is specifically:
  • f t,1 (L i ), f t,3 (L i ), and f t,2 (L i ) are the probability densities of pitch driving torque L i in three pitch motion states, respectively.
  • the probability density and frequency of occurrence are mainly represented by the letter f, here for distinction, the t in the subscript is placed in front of the number and behind the number respectively.
  • the probability density can be expressed by the mathematical expectation ⁇ and standard deviation ⁇ of the variable (here, the pitch driving torque L i ), that is, it can be expressed as:
  • the collected operation data of each historical period can be composed into a Multidimensional vector, denoted as running data column vector.
  • the output power Pwr, the impeller speed r, the generator torque T, the x-direction component Ax of the nacelle acceleration, the y-direction component Ay of the nacelle acceleration, and the pitch angle Pa can form a 6-dimensional vector [Pwr t , r t ,T t ,Ax t ,Ay t ,Pa t ], where the subscript t represents a certain historical period.
  • the calculation target that is, the frequency of occurrence and distribution parameters, can also form a multidimensional vector. For the previous example, it is
  • the above frequency of occurrence and corresponding distribution parameters may be determined by using the following equation.
  • the running data column vector is composed of multiple running data.
  • the number of rows of the correlation coefficient matrix may be eight. That is to say, the product of a row of coefficients in the correlation coefficient matrix a i,j and the first transfer function G 0 can be used as a weight to calculate the weighted sum of multiple operating data, so as to combine multiple operating data into one data. How many parameters are to be output in the end, and how many weighted sums are obtained.
  • the correlation coefficient matrix, the first correlation coefficient column vector, the second correlation coefficient column vector, the first transfer function, and the second transfer function are obtained through testing or training. It should be understood that the subscript i of each correlation coefficient here represents the serial number of the specific coefficient, and has nothing to do with the subscript i of the pitch driving torque.
  • step S102 the angle cumulative value of the pitch angle in each of the M historical periods is obtained.
  • the angle accumulation value is obtained by the pitching experience accumulation module shown in Fig. 2 .
  • Pitching is essentially that the blades turn through a certain angle under the action of pitching driving torque. Therefore, the accumulation of pitch driving torque in the dimension of pitch angle can fully reflect the load borne by the pitch bearing. By obtaining the above angle cumulative value, it can be used as the life evaluation basis of the pitch bearing.
  • the blades may be pitched positively and then pitched negatively, resulting in a cancellation of the pitch angle, so the angle cumulative value of the pitch angle can be calculated step by step according to the set frequency.
  • the following equation can be used:
  • the sampling frequency of the representative data is 1 Hz
  • ⁇ i is the i-th pitch angle in the historical period.
  • step S103 the equivalent load of the pitch bearing is determined according to the pitch driving torque and its probability density in M historical periods, and the angle accumulation value in M historical periods.
  • the equivalent load of the pitch bearing is obtained from the consumed life evaluation module of the pitch bearing as shown in Fig. 2 .
  • the equivalent load of the pitch bearing can be obtained by accumulating the pitch driving torque in the dimension of the pitch angle.
  • step S103 specifically includes: for each historical period, determine the product of the m-th power of each pitch driving torque, the probability density of the corresponding pitch driving torque, and the cumulative angle value, and determine the product of a plurality of pitch driving torques The sum of the products corresponding to the respective moments is summed to obtain the reference load in the corresponding historical period, where m is the material Wall coefficient of the pitch bearing; determine the average value of the reference load in multiple historical periods; determine the 1/m times of the average value of the reference load power, as the equivalent load on the pitch bearing.
  • the column of data below it is the probability density of the pitch driving torque L 1 in each historical period.
  • the column of data below it is the angular cumulative value of the pitch angle corresponding to the pitch driving torque L 1 in each historical period.
  • the difference is that, since the magnitude of the angle accumulation value is fixed within a historical period and will not change with the pitch driving torque, the angle accumulation value in each row (that is, each historical period) is the same.
  • step S104 the consumed life of the pitch bearing is determined according to the equivalent load of the pitch bearing.
  • the consumed life is obtained from the consumed life evaluation module of the pitch bearing as shown in Figure 2.
  • the ratio of the equivalent load of the pitch bearing to the design equivalent load can be determined first, and then the product of the ratio and the design life of the pitch bearing can be determined as the consumed life of the pitch bearing. It can be expressed as the following equation:
  • the life assessment method may further include: acquiring estimated wind resource parameters of multiple machine locations in the target future period, where the estimated wind resource parameters include estimated wind speed; parameters to determine the probability density of the pitch driving torque at multiple estimated wind speeds and the estimated angle accumulation value of the pitch angle in the target future time period; The probability density under the wind speed and the estimated angle accumulation value under multiple estimated wind speeds determine the estimated equivalent load of the pitch bearing in the target future period; according to the estimated equivalent load, determine the Estimated consumption life: Determine the estimated remaining life of the pitch bearing based on the design life, consumed life, and estimated consumed life of the pitch bearing.
  • the consumed life can be obtained through the pitch bearing online life evaluation system shown in Figure 2.
  • the remaining life although the future operating data of wind turbines cannot be obtained in advance, but The subsequent operation of the unit is affected by the wind resource conditions of the future wind farm.
  • the estimated equivalent load of the pitch bearing in the target future period can be estimated based on this, and then the consumption life estimation in the future target period can be completed , finally combining it with the design life and the consumed life, the estimated remaining life of the pitch bearing at the end of the target future period can be obtained.
  • the data acquisition system and wind resource statistical analysis module shown in Figure 3 can be used to obtain estimated wind resource parameters for the future target period. Therefore, additional data acquisition is also not required, and additional data acquisition sensors do not need to be configured, which can reduce product costs, save time and cost of life assessment, and improve economic performance.
  • the pitch bearing remaining life prediction system can use estimated wind resource parameters as input, and first estimate the pitch drive torque by the pitch drive torque distribution estimation module shown in Figure 3 (which can be compared with the pitch change in step S101 The probability density of the driving torque is the same) at multiple estimated wind speeds, and the estimated angle accumulation value of the pitch angle in the target future period is estimated by the pitching experience cumulative angle prediction module shown in Figure 3, and the estimation method can be Referring to step S101, it is realized by using a transfer function and a correlation coefficient.
  • the estimated wind speed v is the annual mean wind speed.
  • the estimated wind resource parameters also include turbulence intensity ti, wind shear ⁇ , and air density ⁇ .
  • each estimated wind speed v is likely to occur with different attribute values (turbulence intensity ti, wind shear ⁇ , etc.), the pitch driving torque distribution prediction module and the pitch experience cumulative angle prediction module
  • the input (namely the estimated wind resource parameter) is a single estimated wind speed v k and its attribute value, and the air density ⁇ is added to the estimated wind resource parameter as a constant attribute value, namely [v k , ti k , ⁇ k , ⁇ k ] is the input vector, where the subscript k represents the serial number of the estimated wind speed.
  • the division of the pitch motion state under a single estimated wind speed is consistent with the above, and the three pitch motion states are divided according to the positive, negative, and constant states of the blade pitch.
  • the output can still be the distribution parameters of the three pitch motion states and their corresponding occurrence frequencies, and then use the distribution parameters and occurrence frequencies of the pitch drive torque under the estimated wind speed to determine the probability of the pitch drive torque under the estimated wind speed density distribution.
  • the probability density of the pitch driving moment Li at a certain estimated wind speed v is:
  • the pitch driving torque at a single estimated wind speed may appear in three states at the same time, and the corresponding frequency of occurrence is f 1,v , 1-f 1,v -f 2,v , f 2,v , three states
  • the pitch driving torques below have their own probability densities, which are f v,1 (L i ), f v,3 (L i ), f v,2 (L i ), respectively. Similar to p t (L i ), since the probability density and frequency of occurrence are mainly represented by the letter f, here for distinction, the v in the subscript is placed in front of the number and behind the number respectively.
  • the pitch drive torque L i in the three pitch motion states can also obey the normal distribution, so the mathematical expectation ⁇ and The standard deviation ⁇ is represented, that is, the distribution parameter is The above frequency of occurrence and corresponding distribution parameters can be determined by using the following equation.
  • F 1 , F 0 are transfer functions
  • a m, n , b m , c n are correlation coefficients
  • the transfer functions and correlation coefficients can be obtained through simulation database training. Substituting the obtained frequency and distribution parameters into the expression of p v (L i ) above, p v (L i ) can be obtained.
  • Q 1 and Q 0 are transfer functions
  • a 1 , a 2 , a 3 , a 4 , b 1 , and c 1 are correlation coefficients
  • the transfer functions and correlation coefficients can be obtained through simulation database training.
  • step S103 After determining the distribution parameters of the pitch driving torque at multiple estimated wind speeds and the estimated angle accumulation value of the pitch angle in the target future time period, a method similar to the aforementioned step S103 can be used to generate the pitch as shown in Figure 3.
  • the estimated equivalent load and estimated consumed life are estimated by the bearing consumed life evaluation module, and finally the estimated remaining life is estimated by the pitch bearing remaining life prediction module shown in Figure 3 .
  • the steps of gradually determining the estimated equivalent load, estimated consumption life, and estimated remaining life include: determining the probability density of multiple estimated wind speeds; for each pitch driving torque at each estimated wind speed, Determine the probability density of the estimated wind speed, the m power of the pitch driving torque, the probability density of the pitch driving torque, and the product of the estimated angle accumulation value, and sum all the products to obtain the estimated reference load, where m is The material Wall coefficient of the pitch bearing; determine the 1/m power of the estimated reference load as the estimated equivalent load.
  • the pitch driving torque is divided into N, and the estimated equivalent load of the pitch bearing is:
  • f(v) is the probability density of the estimated wind speed, which is a Rayleigh distribution, which is only related to the annual average wind speed.
  • l cost is the consumed life, which is obtained from the aforementioned step S104
  • l pred is the estimated consumed life in the target future time period T, which satisfies:
  • the operating data collected during the operation of the wind turbine and the estimated wind resource parameters that can be predicted based on the current technology are used to realize the pitch bearing Life assessment does not require additional data collection, and there is no need to configure additional data collection sensors, which can reduce product costs and save time and cost for life assessment.
  • the present disclosure realizes the real-time online evaluation of the consumed life and the prediction of the remaining life in the future within the acceptable range of accuracy, and can perform predictive operation and maintenance and prediction of failure events on the premise of ensuring the safe operation of wind turbines. Thereby reducing the unplanned downtime of the wind power generating set and improving its economic benefits.
  • Fig. 4 is a block diagram illustrating a life evaluation device for a pitch bearing of a wind power generating set according to an embodiment of the present disclosure.
  • the life assessment device 400 for pitch bearings of wind power generators includes a first acquisition unit 401 , an equivalent unit 402 , and a first calculation unit 403 , corresponding to the online life assessment system for pitch bearings shown in FIG. 2 .
  • the first acquiring unit 401 can acquire the probability density of the pitch driving torque in M historical periods, where M is a positive integer. Since the pitch driving torque affects the loss of the pitch bearing, this parameter is chosen to realize the life estimation of the pitch bearing. At the same time, the cumulative operating time of wind turbines often reaches several years. By dividing the cumulative operating time into M historical periods according to a certain step size, data processing can be carried out for each historical period to ensure that different wind turbines are evaluated. The single historical periods used are of the same length to ensure generalizability of the strategy.
  • the pitch driving torque usually changes within a certain range, from which multiple specific values can be selected, for example, according to the set step size, to obtain multiple specific pitch driving torques, that is to say, the selected multiple
  • the pitch driving torque can form an arithmetic sequence, and then the probability density of each pitch driving torque in each historical period can be obtained.
  • the first acquisition unit 401 can specifically determine the occurrence frequency and corresponding distribution parameters of the pitch drive torque in different pitch motion states according to the operation data of M historical periods (it should be understood that the correspondence here refers to the distribution
  • the parameters correspond to the frequency of occurrence, and also correspond to the pitch motion state), the pitch motion state includes a positive state, a constant state, and a negative state; according to the frequency and the corresponding distribution parameters, the probability density is determined.
  • the operating data of the wind power generating set is collected by the data collection system shown in FIG. The operating data of the unit in M historical periods.
  • the operation data is specifically SCADA data, including output power, impeller speed, generator torque, x-direction component of nacelle acceleration, y-direction component of nacelle acceleration, and pitch angle, which can fully reflect the operating conditions of the unit, and these conditions are related to the existence of the pitch driving torque. Correlation, so the pitch drive torque distribution evaluation module shown in Figure 2 can determine the distribution parameters of the pitch drive torque accordingly.
  • SCADA data including output power, impeller speed, generator torque, x-direction component of nacelle acceleration, y-direction component of nacelle acceleration, and pitch angle, which can fully reflect the operating conditions of the unit, and these conditions are related to the existence of the pitch driving torque.
  • the pitch drive torque distribution evaluation module shown in Figure 2 can determine the distribution parameters of the pitch drive torque accordingly.
  • no additional data collection is required, and additional data collection sensors do not need to be configured, which can reduce product costs, save time and cost for life assessment, and improve economic performance.
  • three different pitching motion states can be obtained, and in each historical period, the pitching driving torque may appear in these three states at the same time, so it can be
  • the occurrence frequency and corresponding distribution parameters of the pitch driving torque in the three pitch motion states are determined, so as to describe the pitch driving torque more accurately.
  • the first acquisition unit 401 may first determine the product of the correlation coefficient matrix, the first transfer function, and the column vector of the operating data for each historical period to obtain the first column vector, wherein the number of rows of the correlation coefficient matrix is equal to the occurrence
  • the sum of frequency and the number of corresponding distribution parameters, the operation data column vector is composed of multiple operation data.
  • the weighted sum of is changed multiple times, as the output vector, the output vector includes the frequency of occurrence and the corresponding distribution parameters, and the data conversion is completed.
  • the correlation coefficient matrix, the first correlation coefficient column vector, the second correlation coefficient column vector, the first transfer function, and the second transfer function are obtained through testing or training.
  • the first obtaining unit 401 may also obtain the angle cumulative value of the pitch angle in each of the M historical time periods.
  • the angle accumulation value is obtained by the pitching experience accumulation module shown in Fig. 2 .
  • Pitching is essentially that the blades turn through a certain angle under the action of pitching driving torque. Therefore, the accumulation of pitch driving torque in the dimension of pitch angle can fully reflect the load borne by the pitch bearing.
  • By obtaining the above angle cumulative value it can be used as the life evaluation basis of the pitch bearing.
  • the equivalent unit 402 can determine the equivalent load of the pitch bearing according to the pitch driving torque and its probability density in M historical periods, and the angle accumulation value in M historical periods.
  • the equivalent load of the pitch bearing is obtained from the consumed life evaluation module of the pitch bearing as shown in Fig. 2 .
  • the equivalent load of the pitch bearing can be obtained by accumulating the pitch driving torque in the dimension of the pitch angle.
  • the equivalent unit 402 can specifically be executed as: for each historical period, determine the product of the m-th power of each pitch driving torque, the probability density of the corresponding pitch driving torque, and the cumulative value of the angle, and calculate The sum of the corresponding products of each pitch driving torque is obtained to obtain the reference load of the corresponding historical period, wherein, m is the material Wall coefficient of the pitch bearing; determine the average value of the reference load in multiple historical periods; determine the average value of the reference load 1/m power, as the equivalent load of the pitch bearing.
  • the first calculation unit 403 can determine the consumed life of the pitch bearing according to the equivalent load of the pitch bearing.
  • the consumed life is obtained from the consumed life evaluation module of the pitch bearing as shown in Figure 2.
  • the ratio of the equivalent load of the pitch bearing to the design equivalent load can be determined first, and then the product of the ratio and the design life of the pitch bearing can be determined as the consumed life of the pitch bearing.
  • Fig. 5 is a block diagram showing a life evaluation device of a pitch bearing of a wind power generating set according to another embodiment of the present disclosure.
  • the life evaluation device 500 of the pitch bearing of the wind power generating set includes a first acquisition unit 501, an equivalent unit 502, a first calculation unit 503, a second acquisition unit 504, a determination unit 505, an estimation unit 506, a second The calculation unit 507, wherein the first acquisition unit 501, the equivalent unit 502, and the first calculation unit 503 correspond to the pitch bearing online life evaluation system shown in Figure 2, and its execution actions are equivalent to the first acquisition unit 401, the equivalent The unit 402 and the first computing unit 403 are the same, and will not be repeated here.
  • the second acquisition unit 504 , the determination unit 505 , the estimation unit 506 , and the second calculation unit 507 correspond to the remaining life prediction system of the pitch bearing as shown in FIG. 3 .
  • the second acquiring unit 504 can acquire estimated wind resource parameters of multiple machine locations in the target future time period, where the estimated wind resource parameters include estimated wind speed.
  • the estimated wind speed is the annual mean wind speed.
  • Estimated wind resource parameters also include turbulence intensity, wind shear, and air density.
  • each estimated wind speed is likely to occur with different attribute values (turbulence intensity, wind shear, etc.)
  • the input of the pitch driving moment distribution prediction module and the pitch experience cumulative angle prediction module ( That is, the estimated wind resource parameter) is a single estimated wind speed and its attribute value
  • the air density is added to the estimated wind resource parameter as a constant attribute value.
  • the determination unit 505 can determine the probability density of the pitch driving torque at multiple estimated wind speeds and the estimated angle accumulation value of the pitch angle in the target future time period according to the estimated wind resource parameters.
  • the way of determining it can refer to the first acquisition unit 501, and realize it by using a transfer function and a correlation coefficient.
  • the estimation unit 506 can determine the pitch bearing in the target future time period according to the multiple estimated wind speeds, the pitch driving torque and its probability density under the multiple estimated wind speeds, and the estimated angle accumulation value under the multiple estimated wind speeds. estimated equivalent load.
  • the estimating unit 506 can specifically perform: determine the probability density of multiple estimated wind speeds; determine the probability density of the estimated wind speed, the The product of the power of m, the probability density of the pitch driving torque, and the cumulative value of the estimated angle is summed to obtain the estimated reference load, where m is the material Wall coefficient of the pitch bearing; determine the estimated reference load The 1/m power of the load is used as the estimated equivalent load.
  • the second calculation unit 507 can determine the estimated consumption life of the pitch bearing in the target future period according to the estimated equivalent load. For the determination method, refer to the first acquiring unit 503 .
  • the second calculation unit 507 may also determine the estimated remaining life of the pitch bearing according to the design life, consumed life, and estimated consumed life of the pitch bearing.
  • the consumed lifetime is obtained by the first calculation unit 503 .
  • the difference between the design life minus the consumed life and the estimated consumed life is the estimated remaining life.
  • the consumed life can be obtained through the pitch bearing online life evaluation system shown in Figure 2.
  • the remaining life although the future operating data of wind turbines cannot be obtained in advance, but The subsequent operation of the unit is affected by the wind resource conditions of the future wind farm.
  • the estimated equivalent load of the pitch bearing in the target future period can be estimated based on this, and then the consumption life estimation in the future target period can be completed , finally combining it with the design life and the consumed life, the estimated remaining life of the pitch bearing at the end of the target future period can be obtained.
  • the data acquisition system and wind resource statistical analysis module shown in Figure 3 can be used to obtain estimated wind resource parameters for the future target period. Therefore, additional data acquisition is also not required, and additional data acquisition sensors do not need to be configured, which can reduce product costs, save time and cost of life assessment, and improve economic performance.
  • the life assessment method of a pitch bearing of a wind power generating set may be written as a computer program and stored on a computer-readable storage medium.
  • the instructions corresponding to the computer program are executed by the processor, the above-mentioned method for evaluating the life of a pitch bearing of a wind power generating set can be realized.
  • Examples of computer readable 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 Disc storage, Hard Disk Drive (HDD), Solid State Drive ( SSD), memory cards (such as Multimedia Cards, Secure Digital (SD) or Extreme Digital (XD) cards), magnetic tapes, floppy disks, magneto-optical data storage devices, optical data storage devices, hard disks, solid-state disks, and any other device , said any other means configured to store in a non-transitory manner a computer program and any associated
  • the computer program and any associated data, data files and data structures are distributed over a networked computer system such that the computer program and any associated data, data files and data structures are processed by one or more processors or Computers store, access and execute in a distributed fashion.
  • FIG. 6 is a block diagram illustrating a computer device according to an embodiment of the present disclosure.
  • the computer device 600 includes at least one memory 601 and at least one processor 602, the at least one memory 601 stores a set of computer-executable instructions, when the set of computer-executable instructions is executed by the at least one processor 602, the execution A life assessment method for a pitch bearing of a wind power generating set according to an exemplary embodiment of the present disclosure.
  • the computer device 600 may be a PC computer, a tablet device, a personal digital assistant, a smart phone, or other devices capable of executing the set of instructions described above.
  • the computer device 600 is not necessarily a single electronic device, but may also be any assembly of devices or circuits capable of individually or jointly executing the above-mentioned instructions (or instruction sets).
  • the computer device 600 may also be part of an integrated control system or system manager, or may be configured as a portable electronic device that interfaces locally or remotely (eg, via wireless transmission).
  • processor 602 may include a central processing unit (CPU), a graphics processing unit (GPU), a programmable logic device, a special purpose processor system, a microcontroller, or a microprocessor.
  • a processor may also include analog processors, digital processors, microprocessors, multi-core processors, processor arrays, network processors, and the like.
  • the processor 602 can execute instructions or codes stored in the memory 601, wherein the memory 601 can also store data. Instructions and data may also be sent and received over the network via the network interface device, which may employ any known transmission protocol.
  • the memory 601 can be integrated with the processor 602, for example, RAM or flash memory is arranged in an integrated circuit microprocessor or the like. Additionally, memory 601 may comprise a separate device, such as an external disk drive, storage array, or any other storage device usable by the database system. Memory 601 and processor 602 may be operatively coupled, or may communicate with each other, such as through an I/O port, network connection, etc., such that processor 602 can read files stored in the memory.
  • the computer device 600 may also include a video display (such as a liquid crystal display) and a user interaction interface (such as a keyboard, mouse, touch input device, etc.). All components of computer device 600 may be connected to each other via a bus and/or network.
  • a video display such as a liquid crystal display
  • a user interaction interface such as a keyboard, mouse, touch input device, etc.
  • This disclosure utilizes the operating data that will be collected during the operation of the unit itself, and the estimated wind resource parameters that can be predicted based on the current technology, to realize the life evaluation of the pitch bearing. No additional data collection is required, and no additional configuration is required. Advanced data acquisition sensors can reduce product costs and save time and cost for life assessment.
  • the present disclosure realizes the real-time online evaluation of the consumed life and the prediction of the remaining life in the future within the acceptable range of accuracy, and can perform predictive operation and maintenance and prediction of failure events on the premise of ensuring the safe operation of wind turbines. Thereby reducing the unplanned downtime of wind turbines and improving their economic benefits.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Theoretical Computer Science (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Mechanical Engineering (AREA)
  • Combustion & Propulsion (AREA)
  • Sustainable Development (AREA)
  • Chemical & Material Sciences (AREA)
  • Sustainable Energy (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Operations Research (AREA)
  • Probability & Statistics with Applications (AREA)
  • Computing Systems (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Wind Motors (AREA)

Abstract

本公开提供一种风力发电机组变桨轴承的寿命评估方法和装置。该包括：获取变桨驱动力矩在M个历史时段的概率密度；获取M个历史时段中各个历史时段内变桨角度的角度累积值；根据变桨驱动力矩及其在M个历史时段的概率密度、M个历史时段的角度累积值，确定变桨轴承的等效载荷；根据变桨轴承的等效载荷，确定变桨轴承的已消耗寿命。

Description

风力发电机组变桨轴承的寿命评估方法和装置 技术领域
本公开涉及风力发电领域，更具体地，涉及一种风力发电机组变桨轴承的寿命评估方法和装置。
背景技术
风力发电机组变桨轴承是连接叶片与轮毂的重要组件，还用于将叶片的载荷传递至轮毂上的变桨系统，而变桨系统是风力发电机组的重要安全系统，若发生故障，则有可能带来毁灭性的灾难。因此，在风力发电机组运行过程中，对变桨轴承的寿命评估变得尤为重要且关键。
在相关技术中，通常需要安装载荷传感器，通过载荷传感器采集的变桨轴承的载荷，基于此对在役风力发电机组的变桨轴承进行寿命评估，但此方案安装周期时间长，资金耗费大，对风电场中的各台机组实现全覆盖需要付出较大的资金成本及时间成本。
发明内容
因此，利用目前能够直接获取到的参数来合理估算变桨轴承的载荷，对于如何在低成本下实现变桨轴承的在线寿命评估至关重要。
在一个总的方面，提供一种风力发电机组变桨轴承的寿命评估方法，所述寿命评估方法包括：获取变桨驱动力矩在M个历史时段的概率密度，所述M为正整数；获取所述M个历史时段中各个历史时段内变桨角度的角度累积值；根据所述变桨驱动力矩及其在所述M个历史时段的概率密度、所述M个历史时段的角度累积值，确定变桨轴承的等效载荷；根据所述变桨轴承的等效载荷，确定所述变桨轴承的已消耗寿命。
在另一总的方面，提供一种风力发电机组变桨轴承的寿命评估装置，所述寿命评估装置包括：第一获取单元，被配置为：获取变桨驱动力矩在M个历史时段的概率密度，所述M为正整数；所述第一获取单元还被配置为获取所述M个历史时段中各个历史时段内变桨角度的角度累积值；等效单元，被 配置为：根据所述变桨驱动力矩及其在所述M个历史时段的概率密度、所述M个历史时段的角度累积值，确定变桨轴承的等效载荷；第一计算单元，被配置为：根据所述变桨轴承的等效载荷，确定所述变桨轴承的已消耗寿命。
在另一总的方面，提供一种计算机可读存储介质，当所述计算机可读存储介质中的指令被至少一个处理器运行时，促使所述至少一个处理器执行如上所述的寿命评估方法。
在另一总的方面，提供一种计算机设备，包括：至少一个处理器；至少一个存储计算机可执行指令的存储器，其中，所述计算机可执行指令在被所述至少一个处理器运行时，促使所述至少一个处理器执行如上所述的寿命评估方法。
本公开利用机组运行过程中本身就会采集的运行数据来实现变桨轴承的寿命评估，不需要额外的数据采集，也就不必配置额外的数据采集传感器，可降低产品成本，并节约寿命评估的时间成本。本公开在精度可接受的范围内，实现了实时在线对已消耗寿命评估及未来的剩余寿命的预测，可以在保证风力发电机组安全运行的前提下，进行预测性运维及失效事件的预测，从而减少风力发电机组的非计划性停机时间，提高其经济效益。
应当理解的是，以上的一般描述和后文的细节描述仅是示例性和解释性的，并不能限制本公开。
附图说明
图1是示出根据本公开的实施例的风力发电机组变桨轴承的寿命评估方法的流程图。
图2是示出根据本公开的实施例的变桨轴承在线寿命评估系统的流程示意图。
图3是示出根据本公开的实施例的变桨轴承剩余寿命预测系统的流程示意图。
图4是示出根据本公开的一个实施例的风力发电机组变桨轴承的寿命评估装置的框图。
图5是示出根据本公开的另一个实施例的风力发电机组变桨轴承的寿命评估装置的框图。
图6是示出根据本公开的实施例的计算机设备的框图。
具体实施方式
提供下面的具体实施方式以帮助读者获得对在此描述的方法、设备和/或系统的全面理解。然而，在理解本申请的公开之后，在此描述的方法、设备和/或系统的各种改变、修改和等同物将是清楚的。例如，在此描述的操作的顺序仅是示例，并且不限于在此阐述的那些顺序，而是除了必须以特定的顺序发生的操作之外，可如在理解本申请的公开之后将是清楚的那样被改变。此外，为了更加清楚和简明，本领域已知的特征的描述可被省略。
在此描述的特征可以以不同的形式来实现，而不应被解释为限于在此描述的示例。相反，已提供在此描述的示例，以仅示出实现在此描述的方法、设备和/或系统的许多可行方式中的一些可行方式，所述许多可行方式在理解本申请的公开之后将是清楚的。
如在此使用的，术语“和/或”包括相关联的所列项中的任何一个以及任何两个或更多个的任何组合。
尽管在此可使用诸如“第一”、“第二”和“第三”的术语来描述各种构件、组件、区域、层或部分，但是这些构件、组件、区域、层或部分不应被这些术语所限制。相反，这些术语仅用于将一个构件、组件、区域、层或部分与另一构件、组件、区域、层或部分进行区分。因此，在不脱离示例的教导的情况下，在此描述的示例中所称的第一构件、第一组件、第一区域、第一层或第一部分也可被称为第二构件、第二组件、第二区域、第二层或第二部分。
在说明书中，当元件(诸如，层、区域或基底)被描述为“在”另一元件上、“连接到”或“结合到”另一元件时，该元件可直接“在”另一元件上、直接“连接到”或“结合到”另一元件，或者可存在介于其间的一个或多个其他元件。相反，当元件被描述为“直接在”另一元件上、“直接连接到”或“直接结合到”另一元件时，可不存在介于其间的其他元件。
在此使用的术语仅用于描述各种示例，并不将用于限制公开。除非上下文另外清楚地指示，否则单数形式也意在包括复数形式。术语“包含”、“包括”和“具有”说明存在叙述的特征、数量、操作、构件、元件和/或它们的组合，但不排除存在或添加一个或多个其他特征、数量、操作、构件、元件和/或它们的组合。
除非另有定义，否则在此使用的所有术语(包括技术术语和科学术语)具有与由本公开所属领域的普通技术人员在理解本公开之后通常理解的含义相同的含义。除非在此明确地如此定义，否则术语(诸如，在通用词典中定义的术语)应被解释为具有与它们在相关领域的上下文和本公开中的含义一致的含义，并且不应被理想化或过于形式化地解释。
此外，在示例的描述中，当认为公知的相关结构或功能的详细描述将引起对本公开的模糊解释时，将省略这样的详细描述。
图1是示出根据本公开的实施例的风力发电机组变桨轴承的寿命评估方法的流程图。寿命评估方法可依托于寿命评估系统来实现，寿命评估系统可进一步包括用于评估变桨轴承的已消耗寿命的在线寿命评估系统，以及用于评估变桨轴承未来预估剩余寿命的剩余寿命预测系统。图2是示出根据本公开的实施例的变桨轴承在线寿命评估系统的流程示意图。图3是示出根据本公开的实施例的变桨轴承剩余寿命预测系统的流程示意图。
参照图1，在步骤S101，获取变桨驱动力矩在M个历史时段的概率密度，其中，M为正整数。由于变桨驱动力矩会影响变桨轴承的损耗，因而选择该参数来实现变桨轴承的寿命评估，需要说明的是，变桨驱动力矩是控制叶片转动角度的力。同时，风力发电机组的累积运行时长常常达到数年，通过将累积运行时长按照一定的步长(例如10分钟)划分为M个历史时段，可针对每个历史时段分别进行数据处理，以保证不同风力发电机组评估时使用的单个历史时段的时长一致，以确保策略的普适性。
可选地，变桨驱动力矩通常在一定区间内变化，例如额定容量小于5MW的风力发电机组，其变桨驱动力矩一般在[-200KNm，200KNm]区间内。可从中选取多个具体数值，例如按照设定步长取值，得到多个具体的变桨驱动力矩，也就是说，选取的多个变桨驱动力矩可构成等差数列，然后在步骤S101获取每个历史时段下每个变桨驱动力矩的概率密度。
参照图2，步骤S101具体包括：根据M个历史时段的运行数据，确定变桨驱动力矩在不同变桨运动状态下的出现频率及对应的分布参数(应理解，这里的对应是指分布参数与出现频率相对应，同时也是与变桨运动状态相对应)，变桨运动状态包括正向状态、恒定状态、负向状态；根据频率及对应的分布参数，确定概率密度。需要说明的是，变桨运动状态指的是带动叶片变桨的运动状态，如，当带动叶片根盘转动的液压杆逐渐推出到全部退出状态 时，叶片根盘带动叶片逐渐转动到顺桨状态，此时变桨运动状态为正向状态；与正向状态相反的则是负向状态，即，当带动叶片根盘转动的液压杆从全部退出状态逐渐退回到液压缸时，叶片根盘带动叶片从顺桨状态逐渐反向转动到逆桨状态，此时变桨运动状态为负向状态；当带动叶片根盘转动的液压杆维持不动时，叶片不动，此时可以看做变桨运动状态为恒定状态。为了方便理解，简要说明各个结构的连接关系，液压杆可伸缩的安装在液压缸中，液压杆输出端通过液压杆轴承与叶片根盘上的变桨摇柄连接，液压杆的伸缩推动叶片根盘转动。根据本公开的实施例，风力发电机组的运行数据由如图2所示的数据采集系统采集得到，相应地，在执行步骤S101之前，还包括采集风力发电机组在M个历史时段的运行数据的步骤。运行数据具体是SCADA(Supervisory Control And Data Acquisition，即数据采集与监视控制系统)数据，包括输出功率、叶轮转速、发电机扭矩、机舱加速度x方向分量、机舱加速度y方向分量、桨矩角，能够充分反映机组运行情况，而这些情况与变桨驱动力矩存在关联，所以如图2所示的变桨驱动力矩分布评估模块可以据此确定变桨驱动力矩的分布参数。此外，由于这些是机组运行过程中本身就会采集的数据，不需要额外的数据采集，也就不必配置额外的数据采集传感器，可降低产品成本，并节约寿命评估的时间成本，提升经济性能。
依据叶片变桨的正向、负向、恒定三种状态，可以得到三种不同的变桨运动状态，而在每个历史时段，变桨驱动力矩都有可能同时出现这三种状态，因此可确定出变桨驱动力矩在三种变桨运动状态下的出现频率及对应的分布参数，从而更准确地描述变桨驱动力矩。
作为示例，对于一个历史时段，变桨运动通常只上述三种状态，所以三者的出现频率之和为1，出现频率具体为：
Figure PCTCN2022083377-appb-000001
基于此，可将变桨驱动力矩L i在某个历史时段t内的概率密度描述为：
p t(L i)＝f 1,t*f t,1(L i)+(1-f 1,t-f 2,t)*f t,3(L i)+f 2,t*f t,2(L i)
其中，f t,1(L i)、f t,3(L i)、f t,2(L i)分别为变桨驱动力矩L i在三种变桨运动状态下的概率密度。需说明的是，由于概率密度和出现频率都主要以字母f表 示，此处为做区分，下标中的t分别置于了数字前方和数字后方。以概率密度服从正态分布为例，则概率密度可通过变量(此处为变桨驱动力矩L i)的数学期望μ和标准差σ来表示，也就是可表示为：
Figure PCTCN2022083377-appb-000002
其中，
Figure PCTCN2022083377-appb-000003
为历史时段t内，变桨驱动力矩L i在三种变桨运动状态下的分布参数，下标f t,1、f t,2、f t,3分别代表正向、负向、恒定三种状态。也就是说，在该示例中，变桨驱动力矩L i的概率密度p t(L i)可利用f 1,t、f 2,t、
Figure PCTCN2022083377-appb-000004
这8个参数得到。
可选地，在根据运行数据确定变桨驱动力矩在不同变桨运动状态下的出现频率及对应的分布参数时，为便于进行数据计算，可以将采集到的每个历史时段的运行数据组成一个多维向量，记为运行数据列向量。例如对于前述的多个运行数据，输出功率Pwr、叶轮转速r、发电机扭矩T、机舱加速度x方向分量Ax、机舱加速度y方向分量Ay、桨矩角Pa，可组成6维向量[Pwr t,r t,T t,Ax t,Ay t,Pa t]，其中的下标t代表某个历史时段。计算目标，即出现频率及分布参数，也可以组成一个多维向量，对于前述示例，就是
Figure PCTCN2022083377-appb-000005
相应地，可利用如下等式确定上述出现频率及对应的分布参数。
Figure PCTCN2022083377-appb-000006
具体来说，就是先对每个历史时段，确定相关系数矩阵a i,j、第一传递函数G 0、运行数据列向量的乘积，得到第一列向量，其中，相关系数矩阵的行数等于出现频率及对应的分布参数的数量之和，运行数据列向量由多个运行数据构成。在该示例中，相关系数矩阵的行数可以为8。也就是说，可将相关系数矩阵a i,j中的一行系数和第一传递函数G 0的乘积作为权重，来计算多个 运行数据的加权和，从而将多个运行数据汇合为一个数据。最后要输出多少个参数，就得到多少个加权和。再确定第一列向量(由前述的多个加权和构成)与第一相关系数列向量b i之和，并与第二传递函数G 1相乘，得到第二列向量。接下来，确定第二列向量与第二相关系数列向量c i之和，相当于对前述的加权和进行多次变化，作为输出向量，输出向量就包括出现频率及对应的分布参数，完成了数据转换。根据本公开的实施例，相关系数矩阵、第一相关系数列向量、第二相关系数列向量、第一传递函数、第二传递函数通过测试或训练得到。应理解，此处各相关系数的下标i代表具体系数的序号，与变桨驱动力矩的下标i无关。
返回参照图1，在步骤S102，获取M个历史时段中各个历史时段内变桨角度的角度累积值。角度累积值由如图2所示的变桨经历累积模块获得。变桨实质上就是在变桨驱动力矩的作用下，叶片转过一定的角度。因此，变桨驱动力矩在变桨角度的维度上积累，就能够全面体现变桨轴承所承受的载荷。通过获取上述的角度累积值，可作为变桨轴承的寿命评估依据。
可选地，在一个历史时段内，叶片可能先正向变桨、再负向变桨，造成变桨角度的抵消，因此可以按照设定频率逐步统计变桨角度的角度累积值。作为示例，可采用如下等式：
Figure PCTCN2022083377-appb-000007
其中，i＝1,2,…，N，N＝600，表征数据的采样频率为1Hz，φ i为历史时段内第i次的变桨角度。
在步骤S103，根据变桨驱动力矩及其在M个历史时段的概率密度、M个历史时段的角度累积值，确定变桨轴承的等效载荷。变桨轴承的等效载荷由如图2所示的变桨轴承已消耗寿命评估模块得到。如前所述，通过将变桨驱动力矩在变桨角度的维度上进行积累，就能够得到变桨轴承的等效载荷。
可选地，步骤S103具体包括：对每个历史时段，确定每个变桨驱动力矩的m次幂、相应的变桨驱动力矩的概率密度、角度累积值的乘积，并对多个变桨驱动力矩各自对应的乘积求和，得到相应历史时段的参考载荷，其中，m是变桨轴承的材料沃尔系数；确定多个历史时段的参考载荷平均值；确定参考载荷平均值的1/m次幂，作为变桨轴承的等效载荷。
对于已经经历过M个历史时段的在役运行机组的变桨系统，N个变桨驱 动力矩的概率密度可以汇总得到如下的历程矩阵：
Figure PCTCN2022083377-appb-000008
其中，以变桨驱动力矩L 1为例，其下方的一列数据为变桨驱动力矩L 1在每个历史时段的概率密度。
M个历史时段的变桨角度的角度累积值可以汇总得到如下的历程矩阵：
Figure PCTCN2022083377-appb-000009
类似于上一个历程矩阵，以变桨驱动力矩L 1为例，其下方的一列数据为变桨驱动力矩L 1在每个历史时段对应的变桨角度的角度累积值。不同之处在于，由于在一个历史时段内，角度累积值的大小是固定的，不会随变桨驱动力矩变化，所有每一行(即每个历史时段)中的角度累积值相同。
变桨轴承的等效载荷为：
Figure PCTCN2022083377-appb-000010
在步骤S104，根据变桨轴承的等效载荷，确定变桨轴承的已消耗寿命。已消耗寿命由如图2所示的变桨轴承已消耗寿命评估模块得到。具体地，可先确定变桨轴承的等效载荷与设计等效载荷的比值，再确定该比值与变桨轴承的设计寿命的乘积，作为变桨轴承的已消耗寿命。可表示为如下等式：
Figure PCTCN2022083377-appb-000011
其中，
Figure PCTCN2022083377-appb-000012
为变桨轴承的设计等效载荷，lifetime design为变桨轴承的设计寿命。
可选地，根据本公开的实施例的寿命评估方法还可包括：获取目标未来 时段的多个机位点的预估风资源参数，预估风资源参数包括预估风速；根据预估风资源参数，确定变桨驱动力矩在多个预估风速下的概率密度及目标未来时段内变桨角度的预估角度积累值；根据多个预估风速、变桨驱动力矩及其在多个预估风速下的概率密度、多个预估风速下的预估角度积累值，确定变桨轴承在目标未来时段的预估等效载荷；根据预估等效载荷，确定变桨轴承在目标未来时段的预估消耗寿命；根据变桨轴承的设计寿命、已消耗寿命、预估消耗寿命，确定变桨轴承的预估剩余寿命。
对于已运行的风力发电机组的变桨轴承，通过如图2所示的变桨轴承在线寿命评估系统可以得到已消耗寿命，对于剩余寿命，虽无法提前获取风力发电机组在未来的运行数据，但机组后续的运行是受未来风电场的风资源条件的影响的。通过获取目标未来时段的多个机位点的预估风资源参数，就可基于此估算出变桨轴承在目标未来时段的预估等效载荷，进而完成在未来目标时段的消耗寿命的预估，最终将之与设计寿命、已消耗寿命相结合，即可得到变桨轴承在目标未来时段终点的预估剩余寿命。由于基于目前的风资源分析技术已经具备对未来几年的风资源参数进行预测的能力，可采用如图3所示的数据采集系统和风资源统计分析模块得到未来目标时段的预估风资源参数，因而同样不需要额外的数据采集，也就不必配置额外的数据采集传感器，可降低产品成本，并节约寿命评估的时间成本，提升经济性能。
具体地，变桨轴承剩余寿命预测系统可将预估风资源参数作为输入，先由如图3所示的变桨驱动力矩分布预估模块估算出变桨驱动力矩(可与步骤S101的变桨驱动力矩相同)在多个预估风速下的概率密度，并由如图3所示的变桨经历累积角度预测模块估算出目标未来时段内变桨角度的预估角度积累值，其估算方式可参考步骤S101，利用传递函数和相关系数实现。
作为示例，预估风速v为年平均风速。预估风资源参数还包括湍流强度ti、风切变α、空气密度ρ。对于一个风电场，每个预估风速v均有可能发生且带有不同的属性值(湍流强度ti、风切变α等)，变桨驱动力矩分布预估模块和变桨经历累积角度预测模块的输入(即预估风资源参数)为单个预估风速v k及其属性值，并将空气密度ρ作为恒定的属性值加入到预估风资源参数中，即[v k,ti k,α k,ρ k]为输入向量，这里下标k代表预估风速的序号。
对于变桨驱动力矩分布预估模块，单个预估风速下变桨运动状态的划分与前述一致，三种变桨运动状态根据叶片变桨的正向、负向、恒定三种状态 进行划分。输出依然可为三个变桨运动状态的分布参数及分别对应的出现频率，进而利用变桨驱动力矩在预估风速下的分布参数及出现频率，确定变桨驱动力矩在预估风速下的概率密度分布。对于变桨驱动力矩L i在某预估风速v下的概率密度为：
p v(L i)＝f 1,v*f v,1(L i)+(1-f 1,v-f 2,v)*f v,3(L i)+f 2,v*f v,2(L i)
单个预估风速下的变桨驱动力矩均有可能同时出现三种状态，其对应的出现频率为f 1,v、1-f 1,v-f 2,v、f 2,v，三种状态下的变桨驱动力矩均有各自的概率密度，分别为f v,1(L i)、f v,3(L i)、f v,2(L i)。类似p t(L i)，由于概率密度和出现频率都主要以字母f表示，此处为做区分，下标中的v分别置于了数字前方和数字后方。此外，参考历史时段t内，变桨驱动力矩L i在三种变桨运动状态下的概率密度，此处在预估风速v下，变桨驱动力矩L i在三种变桨运动状态下的概率密度f v,1(L i)、f v,3(L i)、f v,2(L i)亦可服从正态分布，因而也可通过变桨驱动力矩L i的数学期望μ和标准差σ来表示，即分布参数为
Figure PCTCN2022083377-appb-000013
可利用如下等式确定上述出现频率及对应的分布参数。
Figure PCTCN2022083377-appb-000014
其中，F 1,F 0为传递函数，a m,n,b m,c n为相关系数，传递函数及相关系数的获得可通过仿真的数据库训练获得。将得到的出现频率和分布参数代入前述p v(L i)的表达式，即可得到p v(L i)。
对于变桨经历累积角度预测模块，在对单个预估风速下的变桨角度的预估角度积累值进行预测时，其预测方程如下所示：
Figure PCTCN2022083377-appb-000015
其中，Q 1,Q 0为传递函数，a 1,a 2,a 3,a 4,b 1,c 1为相关系数，传递函数及相 关系数的获得可通过仿真的数据库训练获得。
确定变桨驱动力矩在多个预估风速下的分布参数及目标未来时段内变桨角度的预估角度积累值后，可采用与前述步骤S103类似的方式，由如图3所示的变桨轴承已消耗寿命评估模块估算出预估等效载荷和预估消耗寿命，最终由如图3所示的变桨轴承剩余寿命预测模块估算出预估剩余寿命。
可选地，逐步确定预估等效载荷、预估消耗寿命、预估剩余寿命的步骤包括：确定多个预估风速的概率密度；对每个预估风速下的每个变桨驱动力矩，确定预估风速的概率密度、变桨驱动力矩的m次幂、变桨驱动力矩的概率密度、预估角度积累值的乘积，并对全部乘积求和，得到预估参考载荷，其中，m是变桨轴承的材料沃尔系数；确定预估参考载荷的1/m次幂，作为预估等效载荷。
对于目标未来时段T的风力发电机组的变桨系统，将变桨驱动力矩划分为N个，变桨轴承的预估等效载荷为：
Figure PCTCN2022083377-appb-000016
其中，f(v)为预估风速的概率密度，为瑞利分布，仅与年平均风速有关。
则变桨轴承在目标未来时段T的预估剩余寿命为：
l remain＝lifetime design-l cost-l pred
其中l cost为已消耗寿命，由前述步骤S104得到，l pred为目标未来时段T内的预估消耗寿命，满足：
Figure PCTCN2022083377-appb-000017
根据本公开的实施例的风力发电机组变桨轴承的寿命评估方法，利用机组运行过程中本身就会采集的运行数据，以及基于目前技术可以预测到的预估风资源参数，来实现变桨轴承的寿命评估，不需要额外的数据采集，也就不必配置额外的数据采集传感器，可降低产品成本，并节约寿命评估的时间成本。本公开在精度可接受的范围内，实现了实时在线对已消耗寿命评估及未来的剩余寿命的预测，可以在保证风力发电机组安全运行的前提下，进行预测性运维及失效事件的预测，从而减少风力发电机组的非计划性停机时间， 提高其经济效益。
图4是示出根据本公开的一个实施例的风力发电机组变桨轴承的寿命评估装置的框图。
参照图4，风力发电机组变桨轴承的寿命评估装置400包括第一获取单元401、等效单元402、第一计算单元403，对应于如图2所示的变桨轴承在线寿命评估系统。
第一获取单元401可获取变桨驱动力矩在M个历史时段的概率密度，其中，M为正整数。由于变桨驱动力矩会影响变桨轴承的损耗，因而选择该参数来实现变桨轴承的寿命评估。同时，风力发电机组的累积运行时长常常达到数年，通过将累积运行时长按照一定的步长划分为M个历史时段，可针对每个历史时段分别进行数据处理，以保证不同风力发电机组评估时使用的单个历史时段的时长一致，以确保策略的普适性。
可选地，变桨驱动力矩通常在一定区间内变化，可从中选取多个具体数值，例如按照设定步长取值，得到多个具体的变桨驱动力矩，也就是说，选取的多个变桨驱动力矩可构成等差数列，然后获取每个历史时段下每个变桨驱动力矩的概率密度。
参照图2，第一获取单元401具体可根据M个历史时段的运行数据，确定变桨驱动力矩在不同变桨运动状态下的出现频率及对应的分布参数(应理解，这里的对应是指分布参数与出现频率相对应，同时也是与变桨运动状态相对应)，变桨运动状态包括正向状态、恒定状态、负向状态；根据频率及对应的分布参数，确定概率密度。根据本公开的实施例，风力发电机组的运行数据由如图2所示的数据采集系统采集得到，相应地，在确定出现频率及对应的分布参数之前，第一获取单元401还可采集风力发电机组在M个历史时段的运行数据。运行数据具体是SCADA数据，包括输出功率、叶轮转速、发电机扭矩、机舱加速度x方向分量、机舱加速度y方向分量、桨矩角，能够充分反映机组运行情况，而这些情况与变桨驱动力矩存在关联，所以如图2所示的变桨驱动力矩分布评估模块可以据此确定变桨驱动力矩的分布参数。此外，由于这些是机组运行过程中本身就会采集的数据，不需要额外的数据采集，也就不必配置额外的数据采集传感器，可降低产品成本，并节约寿命评估的时间成本，提升经济性能。
依据叶片变桨的正向、负向、恒定三种状态，可以得到三种不同的变桨 运动状态，而在每个历史时段，变桨驱动力矩都有可能同时出现这三种状态，因此可确定出变桨驱动力矩在三种变桨运动状态下的出现频率及对应的分布参数，从而更准确地描述变桨驱动力矩。
可选地，第一获取单元401可先对每个历史时段，确定相关系数矩阵、第一传递函数、运行数据列向量的乘积，得到第一列向量，其中，相关系数矩阵的行数等于出现频率及对应的分布参数的数量之和，运行数据列向量由多个运行数据构成。再确定第一列向量与第一相关系数列向量之和，并与第二传递函数相乘，得到第二列向量，确定第二列向量与第二相关系数列向量之和，相当于对前述的加权和进行多次变化，作为输出向量，输出向量就包括出现频率及对应的分布参数，完成了数据转换。根据本公开的实施例，相关系数矩阵、第一相关系数列向量、第二相关系数列向量、第一传递函数、第二传递函数通过测试或训练得到。
第一获取单元401还可获取M个历史时段中各个历史时段内变桨角度的角度累积值。角度累积值由如图2所示的变桨经历累积模块获得。变桨实质上就是在变桨驱动力矩的作用下，叶片转过一定的角度。因此，变桨驱动力矩在变桨角度的维度上积累，就能够全面体现变桨轴承所承受的载荷。通过获取上述的角度累积值，可作为变桨轴承的寿命评估依据。
等效单元402可根据变桨驱动力矩及其在M个历史时段的概率密度、M个历史时段的角度累积值，确定变桨轴承的等效载荷。变桨轴承的等效载荷由如图2所示的变桨轴承已消耗寿命评估模块得到。如前所述，通过将变桨驱动力矩在变桨角度的维度上进行积累，就能够得到变桨轴承的等效载荷。
可选地，等效单元402具体可执行为：对每个历史时段，确定每个变桨驱动力矩的m次幂、相应的变桨驱动力矩的概率密度、角度累积值的乘积，并对多个变桨驱动力矩各自对应的乘积求和，得到相应历史时段的参考载荷，其中，m是变桨轴承的材料沃尔系数；确定多个历史时段的参考载荷平均值；确定参考载荷平均值的1/m次幂，作为变桨轴承的等效载荷。
第一计算单元403可根据变桨轴承的等效载荷，确定变桨轴承的已消耗寿命。已消耗寿命由如图2所示的变桨轴承已消耗寿命评估模块得到。具体地，可先确定变桨轴承的等效载荷与设计等效载荷的比值，再确定该比值与变桨轴承的设计寿命的乘积，作为变桨轴承的已消耗寿命。
图5是示出根据本公开的另一个实施例的风力发电机组变桨轴承的寿命 评估装置的框图。
参照图5，风力发电机组变桨轴承的寿命评估装置500包括第一获取单元501、等效单元502、第一计算单元503、第二获取单元504、确定单元505、预估单元506、第二计算单元507，其中，第一获取单元501、等效单元502、第一计算单元503对应于如图2所示的变桨轴承在线寿命评估系统，其执行动作与第一获取单元401、等效单元402、第一计算单元403一致，在此不再赘述。第二获取单元504、确定单元505、预估单元506、第二计算单元507对应于如图3所示的变桨轴承剩余寿命预测系统。
第二获取单元504可获取目标未来时段的多个机位点的预估风资源参数，预估风资源参数包括预估风速。作为示例，预估风速为年平均风速。预估风资源参数还包括湍流强度、风切变、空气密度。对于一个风电场，每个预估风速均有可能发生且带有不同的属性值(湍流强度、风切变等)，变桨驱动力矩分布预估模块和变桨经历累积角度预测模块的输入(即预估风资源参数)为单个预估风速及其属性值，并将空气密度作为恒定的属性值加入到预估风资源参数中。
确定单元505可根据预估风资源参数，确定变桨驱动力矩在多个预估风速下的概率密度及目标未来时段内变桨角度的预估角度积累值。其确定方式可参考第一获取单元501，利用传递函数和相关系数实现。
预估单元506可根据多个预估风速、变桨驱动力矩及其在多个预估风速下的概率密度、多个预估风速下的预估角度积累值，确定变桨轴承在目标未来时段的预估等效载荷。
可选地，预估单元506具体可执行：确定多个预估风速的概率密度；对每个预估风速下的每个变桨驱动力矩，确定预估风速的概率密度、变桨驱动力矩的m次幂、变桨驱动力矩的概率密度、预估角度积累值的乘积，并对全部乘积求和，得到预估参考载荷，其中，m是变桨轴承的材料沃尔系数；确定预估参考载荷的1/m次幂，作为预估等效载荷。
第二计算单元507可根据预估等效载荷，确定变桨轴承在目标未来时段的预估消耗寿命。其确定方式可参考第一获取单元503。
第二计算单元507还可根据变桨轴承的设计寿命、已消耗寿命、预估消耗寿命，确定变桨轴承的预估剩余寿命。这里，已消耗寿命由第一计算单元503得到。具体地，设计寿命减去已消耗寿命、预估消耗寿命所得的差值， 即为预估剩余寿命。
对于已运行的风力发电机组的变桨轴承，通过如图2所示的变桨轴承在线寿命评估系统可以得到已消耗寿命，对于剩余寿命，虽无法提前获取风力发电机组在未来的运行数据，但机组后续的运行是受未来风电场的风资源条件的影响的。通过获取目标未来时段的多个机位点的预估风资源参数，就可基于此估算出变桨轴承在目标未来时段的预估等效载荷，进而完成在未来目标时段的消耗寿命的预估，最终将之与设计寿命、已消耗寿命相结合，即可得到变桨轴承在目标未来时段终点的预估剩余寿命。由于基于目前的风资源分析技术已经具备对未来几年的风资源参数进行预测的能力，可采用如图3所示的数据采集系统和风资源统计分析模块得到未来目标时段的预估风资源参数，因而同样不需要额外的数据采集，也就不必配置额外的数据采集传感器，可降低产品成本，并节约寿命评估的时间成本，提升经济性能。
根据本公开的实施例的风力发电机组变桨轴承的寿命评估方法可被编写为计算机程序并被存储在计算机可读存储介质上。当所述计算机程序对应的指令被处理器执行时，可实现如上所述的风力发电机组变桨轴承的寿命评估方法。计算机可读存储介质的示例包括：只读存储器(ROM)、随机存取可编程只读存储器(PROM)、电可擦除可编程只读存储器(EEPROM)、随机存取存储器(RAM)、动态随机存取存储器(DRAM)、静态随机存取存储器(SRAM)、闪存、非易失性存储器、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、蓝光或光盘存储器、硬盘驱动器(HDD)、固态硬盘(SSD)、卡式存储器(诸如，多媒体卡、安全数字(SD)卡或极速数字(XD)卡)、磁带、软盘、磁光数据存储装置、光学数据存储装置、硬盘、固态盘以及任何其他装置，所述任何其他装置被配置为以非暂时性方式存储计算机程序以及任何相关联的数据、数据文件和数据结构并将所述计算机程序以及任何相关联的数据、数据文件和数据结构提供给处理器或计算机使得处理器或计算机能执行所述计算机程序。在一个示例中，计算机程序以及任何相关联的数据、数据文件和数据结构分布在联网的计算机系统上，使得计算机程序以及任何相关联的数据、数据文件和数据结构通过一个或多个处理器或计算机以分布式方式存储、访问和执行。
图6是示出根据本公开的实施例的计算机设备的框图。
参照图6，计算机设备600包括至少一个存储器601和至少一个处理器602，所述至少一个存储器601中存储有计算机可执行指令集合，当计算机可执行指令集合被至少一个处理器602执行时，执行根据本公开的示例性实施例的风力发电机组变桨轴承的寿命评估方法。
作为示例，计算机设备600可以是PC计算机、平板装置、个人数字助理、智能手机、或其他能够执行上述指令集合的装置。这里，计算机设备600并非必须是单个的电子设备，还可以是任何能够单独或联合执行上述指令(或指令集)的装置或电路的集合体。计算机设备600还可以是集成控制系统或系统管理器的一部分，或者可被配置为与本地或远程(例如，经由无线传输)以接口互联的便携式电子设备。
在计算机设备600中，处理器602可包括中央处理器(CPU)、图形处理器(GPU)、可编程逻辑装置、专用处理器系统、微控制器或微处理器。作为示例而非限制，处理器还可包括模拟处理器、数字处理器、微处理器、多核处理器、处理器阵列、网络处理器等。
处理器602可运行存储在存储器601中的指令或代码，其中，存储器601还可以存储数据。指令和数据还可经由网络接口装置而通过网络被发送和接收，其中，网络接口装置可采用任何已知的传输协议。
存储器601可与处理器602集成为一体，例如，将RAM或闪存布置在集成电路微处理器等之内。此外，存储器601可包括独立的装置，诸如，外部盘驱动、存储阵列或任何数据库系统可使用的其他存储装置。存储器601和处理器602可在操作上进行耦合，或者可例如通过I/O端口、网络连接等互相通信，使得处理器602能够读取存储在存储器中的文件。
此外，计算机设备600还可包括视频显示器(诸如，液晶显示器)和用户交互接口(诸如，键盘、鼠标、触摸输入装置等)。计算机设备600的所有组件可经由总线和/或网络而彼此连接。
本公开利用机组运行过程中本身就会采集的运行数据，以及基于目前技术可以预测到的预估风资源参数，来实现变桨轴承的寿命评估，不需要额外的数据采集，也就不必配置额外的数据采集传感器，可降低产品成本，并节约寿命评估的时间成本。本公开在精度可接受的范围内，实现了实时在线对已消耗寿命评估及未来的剩余寿命的预测，可以在保证风力发电机组安全运行的前提下，进行预测性运维及失效事件的预测，从而减少风力发电机组的 非计划性停机时间，提高其经济效益。
以上对本公开的具体实施方式进行了详细描述，虽然已表示和描述了一些实施例，但本领域技术人员应该理解，在不脱离由权利要求及其等同物限定其范围的本公开的原理和精神的情况下，可对这些实施例进行修改和变型，这些修改和变型也应在本公开的权利要求的保护范围内。

Claims (20)

  1. 一种风力发电机组变桨轴承的寿命评估方法，其特征在于，所述寿命评估方法包括：
    获取变桨驱动力矩在M个历史时段的概率密度，所述M为正整数；
    获取所述M个历史时段中各个历史时段内变桨角度的角度累积值；
    根据所述变桨驱动力矩及其在所述M个历史时段的概率密度、所述M个历史时段的角度累积值，确定变桨轴承的等效载荷；
    根据所述变桨轴承的等效载荷，确定所述变桨轴承的已消耗寿命。
  2. 如权利要求1所述的寿命评估方法，其特征在于，所述获取变桨驱动力矩在M个历史时段的概率密度，包括：
    根据所述M个历史时段的运行数据，确定所述变桨驱动力矩在不同变桨运动状态下的出现频率及对应的分布参数，所述变桨运动状态包括正向状态、恒定状态、负向状态；
    根据所述频率及对应的分布参数，确定所述概率密度。
  3. 如权利要求2所述的寿命评估方法，其特征在于，所述根据所述M个历史时段的运行数据，确定所述变桨驱动力矩在不同变桨运动状态下的出现频率及对应的分布参数，包括：
    对每个历史时段，确定相关系数矩阵、第一传递函数、运行数据列向量的乘积，得到第一列向量，其中，所述相关系数矩阵的行数等于所述出现频率及对应的分布参数的数量之和，所述运行数据列向量由多个运行数据构成；
    确定所述第一列向量与第一相关系数列向量之和，并与第二传递函数相乘，得到第二列向量；
    确定所述第二列向量与第二相关系数列向量之和，作为输出向量，所述输出向量包括所述出现频率及对应的分布参数，
    其中，所述相关系数矩阵、所述第一相关系数列向量、所述第二相关系数列向量、所述第一传递函数、所述第二传递函数通过测试或训练得到。
  4. 如权利要求2所述的寿命评估方法，其特征在于，所述根据所述M个历史时段的运行数据，确定所述变桨驱动力矩在不同变桨运动状态下的出现频率及对应的分布参数之前，还包括：
    采集所述风力发电机组在所述M个历史时段的运行数据；其中，所述运 行数据包括输出功率、叶轮转速、发电机扭矩、机舱加速度x方向分量、机舱加速度y方向分量、桨矩角。
  5. 如权利要求1所述的寿命评估方法，其特征在于，所述变桨驱动力矩的数量为多个，其中，所述根据所述变桨驱动力矩及其在所述M个历史时段的概率密度、所述M个历史时段的角度累积值，确定变桨轴承的等效载荷，包括：
    对每个历史时段，确定每个所述变桨驱动力矩的m次幂、相应的所述变桨驱动力矩的概率密度、所述角度累积值的乘积，并对多个所述变桨驱动力矩各自对应的所述乘积求和，得到相应历史时段的参考载荷，其中，m是所述变桨轴承的材料沃尔系数；
    确定所述多个历史时段的参考载荷平均值；
    确定所述参考载荷平均值的1/m次幂，作为所述变桨轴承的等效载荷。
  6. 如权利要求5所述的寿命评估方法，其特征在于，多个所述变桨驱动力矩通过以下步骤得到：
    对变桨驱动力矩变化区间，按照设定步长进行取值，得到多个所述变桨驱动力矩。
  7. 如权利要求1至6中任一项所述的寿命评估方法，其特征在于，还包括：
    获取目标未来时段的多个机位点的预估风资源参数，所述预估风资源参数包括预估风速；
    根据所述预估风资源参数，确定所述变桨驱动力矩在多个预估风速下的概率密度及所述目标未来时段内变桨角度的预估角度积累值；
    根据所述多个预估风速、所述变桨驱动力矩及其在所述多个预估风速下的概率密度、所述多个预估风速下的预估角度积累值，确定所述变桨轴承在所述目标未来时段的预估等效载荷；
    根据所述预估等效载荷，确定所述变桨轴承在所述目标未来时段的预估消耗寿命；
    根据所述变桨轴承的设计寿命、所述已消耗寿命、所述预估消耗寿命，确定所述变桨轴承的预估剩余寿命。
  8. 如权利要求7所述的寿命评估方法，其特征在于，所述预估风资源参数还包括湍流强度、风切变、空气密度。
  9. 如权利要求7所述的寿命评估方法，其特征在于，所述根据所述多个预估风速、所述变桨驱动力矩及其在所述多个预估风速下的概率密度、所述多个预估风速下的预估角度积累值，确定所述变桨轴承在所述目标未来时段的预估等效载荷，包括：
    确定所述多个预估风速的概率密度；
    对每个所述预估风速下的每个所述变桨驱动力矩，确定所述预估风速的概率密度、所述变桨驱动力矩的m次幂、所述变桨驱动力矩的概率密度、所述预估角度积累值的乘积，并对全部所述乘积求和，得到预估参考载荷，其中，m是所述变桨轴承的材料沃尔系数；
    确定所述预估参考载荷的1/m次幂，作为所述预估等效载荷。
  10. 一种风力发电机组变桨轴承的寿命评估装置，其特征在于，所述寿命评估装置包括：
    第一获取单元，被配置为：获取变桨驱动力矩在M个历史时段的概率密度，所述M为正整数；
    所述第一获取单元还被配置为获取所述M个历史时段中各个历史时段内变桨角度的角度累积值；
    等效单元，被配置为：根据所述变桨驱动力矩及其在所述M个历史时段的概率密度、所述M个历史时段的角度累积值，确定变桨轴承的等效载荷；
    第一计算单元，被配置为：根据所述变桨轴承的等效载荷，确定所述变桨轴承的已消耗寿命。
  11. 如权利要求10所述的寿命评估装置，其特征在于，所述第一获取单元还被配置为：
    根据所述M个历史时段的运行数据，确定所述变桨驱动力矩在不同变桨运动状态下的出现频率及对应的分布参数，所述变桨运动状态包括正向状态、恒定状态、负向状态；
    根据所述频率及对应的分布参数，确定所述概率密度。
  12. 如权利要求11所述的寿命评估装置，其特征在于，所述第一获取单元还被配置为：
    对每个历史时段，确定相关系数矩阵、第一传递函数、运行数据列向量的乘积，得到第一列向量，其中，所述相关系数矩阵的行数等于所述出现频率及对应的分布参数的数量之和，所述运行数据列向量由多个运行数据构成；
    确定所述第一列向量与第一相关系数列向量之和，并与第二传递函数相乘，得到第二列向量；
    确定所述第二列向量与第二相关系数列向量之和，作为输出向量，所述输出向量包括所述出现频率及对应的分布参数，
    其中，所述相关系数矩阵、所述第一相关系数列向量、所述第二相关系数列向量、所述第一传递函数、所述第二传递函数通过测试或训练得到。
  13. 如权利要求11所述的寿命评估装置，其特征在于，所述第一获取单元还被配置为：
    采集所述风力发电机组在所述M个历史时段的运行数据；其中，所述运行数据包括输出功率、叶轮转速、发电机扭矩、机舱加速度x方向分量、机舱加速度y方向分量、桨矩角。
  14. 如权利要求10所述的寿命评估装置，其特征在于，所述变桨驱动力矩的数量为多个，所述等效单元还被配置为：
    对每个历史时段，确定每个所述变桨驱动力矩的m次幂、相应的所述变桨驱动力矩的概率密度、所述角度累积值的乘积，并对所述多个变桨驱动力矩各自对应的所述乘积求和，得到相应历史时段的参考载荷，其中，m是所述变桨轴承的材料沃尔系数；
    确定所述多个历史时段的参考载荷平均值；
    确定所述参考载荷平均值的1/m次幂，作为所述变桨轴承的等效载荷。
  15. 如权利要求14所述的寿命评估装置，其特征在于，多个所述变桨驱动力矩通过以下步骤得到：
    对变桨驱动力矩变化区间，按照设定步长进行取值，得到多个变桨驱动力矩。
  16. 如权利要求10至15中任一项所述的寿命评估装置，其特征在于，还包括：
    第二获取单元，被配置为：获取目标未来时段的多个机位点的预估风资源参数，所述预估风资源参数包括预估风速；
    确定单元，被配置为：根据所述预估风资源参数，确定所述变桨驱动力矩在多个预估风速下的概率密度及所述目标未来时段内变桨角度的预估角度积累值；
    预估单元，被配置为：根据所述多个预估风速、所述变桨驱动力矩及其 在所述多个预估风速下的概率密度、所述多个预估风速下的预估角度积累值，确定所述变桨轴承在所述目标未来时段的预估等效载荷；
    第二计算单元，被配置为：根据所述预估等效载荷，确定所述变桨轴承在所述目标未来时段的预估消耗寿命；
    所述第二计算单元还被配置为：根据所述变桨轴承的设计寿命、所述已消耗寿命、所述预估消耗寿命，确定所述变桨轴承的预估剩余寿命。
  17. 如权利要求16所述的寿命评估装置，其特征在于，所述预估风资源参数还包括湍流强度、风切变、空气密度。
  18. 如权利要求16所述的寿命评估装置，其特征在于，所述预估单元还被配置为：
    确定所述多个预估风速的概率密度；
    对每个所述预估风速下的每个所述变桨驱动力矩，确定所述预估风速的概率密度、所述变桨驱动力矩的m次幂、所述变桨驱动力矩的概率密度、所述预估角度积累值的乘积，并对全部所述乘积求和，得到预估参考载荷，其中，m是所述变桨轴承的材料沃尔系数；
    确定所述预估参考载荷的1/m次幂，作为所述预估等效载荷。
  19. 一种计算机可读存储介质，其特征在于，当所述计算机可读存储介质中的指令被至少一个处理器运行时，促使所述至少一个处理器执行如权利要求1到9中的任一权利要求所述的寿命评估方法。
  20. 一种计算机设备，其特征在于，包括：
    至少一个处理器；
    至少一个存储计算机可执行指令的存储器，
    其中，所述计算机可执行指令在被所述至少一个处理器运行时，促使所述至少一个处理器执行如权利要求1到9中的任一权利要求所述的寿命评估方法。
PCT/CN2022/083377 2021-11-30 2022-03-28 风力发电机组变桨轴承的寿命评估方法和装置 WO2023097938A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
KR1020247002597A KR20240023651A (ko) 2021-11-30 2022-03-28 풍력 터빈의 피치 베어링의 수명을 평가하는 방법 및 디바이스
AU2022401404A AU2022401404A1 (en) 2021-11-30 2022-03-28 Method and device for evaluating service life of pitch bearing of wind turbine
EP22899761.5A EP4343587A1 (en) 2021-11-30 2022-03-28 Method and device for evaluating service life of pitch bearing of wind turbine

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111438432.7A CN116204762A (zh) 2021-11-30 2021-11-30 风力发电机组变桨轴承的寿命评估方法和装置
CN202111438432.7 2021-11-30

Publications (1)

Publication Number Publication Date
WO2023097938A1 true WO2023097938A1 (zh) 2023-06-08

Family

ID=86508115

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/083377 WO2023097938A1 (zh) 2021-11-30 2022-03-28 风力发电机组变桨轴承的寿命评估方法和装置

Country Status (5)

Country Link
EP (1) EP4343587A1 (zh)
KR (1) KR20240023651A (zh)
CN (1) CN116204762A (zh)
AU (1) AU2022401404A1 (zh)
WO (1) WO2023097938A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100332153A1 (en) * 2007-12-14 2010-12-30 Reinder Hindrik Vegter Method of Determining Fatigue Life and Remaining Life
CN105134510A (zh) * 2015-09-18 2015-12-09 北京中恒博瑞数字电力科技有限公司 一种风力发电机组变桨系统的状态监测和故障诊断方法
CN107590356A (zh) * 2017-10-31 2018-01-16 国电联合动力技术有限公司 一种风力发电机组主轴轴承自动选型方法及存储设备
CN113623142A (zh) * 2021-09-15 2021-11-09 中国船舶重工集团海装风电股份有限公司 风力发电机组变桨轴承齿寿命评估方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100332153A1 (en) * 2007-12-14 2010-12-30 Reinder Hindrik Vegter Method of Determining Fatigue Life and Remaining Life
CN105134510A (zh) * 2015-09-18 2015-12-09 北京中恒博瑞数字电力科技有限公司 一种风力发电机组变桨系统的状态监测和故障诊断方法
CN107590356A (zh) * 2017-10-31 2018-01-16 国电联合动力技术有限公司 一种风力发电机组主轴轴承自动选型方法及存储设备
CN113623142A (zh) * 2021-09-15 2021-11-09 中国船舶重工集团海装风电股份有限公司 风力发电机组变桨轴承齿寿命评估方法

Also Published As

Publication number Publication date
KR20240023651A (ko) 2024-02-22
EP4343587A1 (en) 2024-03-27
AU2022401404A1 (en) 2024-02-01
CN116204762A (zh) 2023-06-02

Similar Documents

Publication Publication Date Title
EP3791060B1 (en) Wind turbine control method
Cai et al. Unsteady aerodynamics simulation of a full-scale horizontal axis wind turbine using CFD methodology
Zhang et al. Wind turbine fault detection based on SCADA data analysis using ANN
Castaignet et al. Full‐scale test of trailing edge flaps on a Vestas V27 wind turbine: active load reduction and system identification
EP3779184B1 (en) Yaw control method, device and system for wind turbine
US10883475B2 (en) Method for monitoring and assessing power performance changes of a wind turbine
EP3947961B1 (en) A method for estimating remaining useful life of components of an operational wind turbine
WO2023123685A1 (zh) 风力发电机组的寿命评估方法和装置
Campobasso et al. Rapid estimate of wind turbine energy loss due to blade leading edge delamination using artificial neural networks
CN110139983B (zh) 实证评估风力涡轮机发电机运行的方法和系统
WO2023097938A1 (zh) 风力发电机组变桨轴承的寿命评估方法和装置
CN112682276B (zh) 风机叶片覆冰状态预测方法、装置、介质和电子设备
WO2019230191A1 (ja) 風力発電システム
WO2023092915A1 (zh) 风电场的功率控制方法和装置
Slot et al. Importance of shear in site assessment of wind turbine fatigue loads
Astolfi et al. Data-driven methods for the analysis of wind turbine yaw control optimization
Gao et al. Real-time monitoring, fault prediction and health management for offshore wind turbine systems
Cavazzini et al. Machine learning-aided assessment of wind turbine energy losses due to blade leading edge damage
Peco Chacon et al. Support vector machine and K-fold cross-validation to detect false alarms in wind turbines
da Rosa et al. Wind Turbine Blade Mass Imbalance Detection Using Artificial Intelligence
Yang et al. Detection of wind turbine blade abnormalities through a deep learning model integrating VAE and neural ODE
CN116838547B (zh) 多转子风力发电机的监测方法及多转子风力发电机
Oliveira Schmidt et al. LSTM Neural Networks Using the SMOTE Algorithm for Wind Turbine Fault Prediction
CN116104710A (zh) 一种风力发电机故障检测方法及系统
CN116842672A (zh) 风力发电机组的主轴承的预定载荷确定方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22899761

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2022899761

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 18574154

Country of ref document: US

ENP Entry into the national phase

Ref document number: 2022899761

Country of ref document: EP

Effective date: 20231222

WWE Wipo information: entry into national phase

Ref document number: 2022401404

Country of ref document: AU

Ref document number: AU2022401404

Country of ref document: AU

ENP Entry into the national phase

Ref document number: 20247002597

Country of ref document: KR

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 1020247002597

Country of ref document: KR

ENP Entry into the national phase

Ref document number: 2022401404

Country of ref document: AU

Date of ref document: 20220328

Kind code of ref document: A

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: 112024001625

Country of ref document: BR